article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
new imaging capabilities are driving revolutionary advances in science and technology , making it possible to image biological systems in three dimensions with stunning detail . for materials systems that can not be labeled using fluorescent markers , and
that are opaque to visible light , shorter wavelength light has advantages . as nanosystems relevant to the semiconductor industry
become increasingly complex , there is a critical need for microscopy techniques that can distinguish between different materials .
this capability is realized using spectromicroscopy techniques in the extreme ultraviolet ( euv ) and x - ray range , which take advantage of element - specific absorption contrast . however ,
established techniques such as near edge x - ray absorption fine structure ( nexafs ) @xcite , x - ray absorption near edge structure ( xanes ) @xcite , and x - ray fluorescence @xcite spectromicroscopies can only record images at one wavelength at a time : either a monochromator or spectrometer must be scanned in order to collect a full series of images .
in contrast , ptychographical information multiplexing @xcite ( pim ) , which is an extension to ptychographical coherent diffractive imaging ( cdi ) , enables the collection of images at multiple wavelengths simultaneously .
when combined with euv and x - ray wavelengths , this technique can be used to produce images with elemental contrast without the need for wavelength scanning .
ptychographic cdi @xcite has achieved great success due to its high fidelity and robustness in situations where other cdi methods fail . in a ptychographical microscope
, the object is scanned area - by - area to record a diffraction pattern at each scan position , ensuring overlap between adjacent scan positions .
the diffraction patterns and scan locations are then fed into an iterative algorithm , to retrieve the phase of the diffracted fields . by back - propagating the diffraction field to the sample position ,
complex - valued images of the object are obtained with quantitative amplitude and phase information .
scanning with overlap is a simple , yet elegant way of introducing redundant information into the diffraction - based measurements , distinguishing ptychography from other computational imaging techniques .
in addition to removing experimental restrictions required for other types of cdi , such as sample isolation , recent breakthroughs demonstrate that the information redundancy in ptychography also enables the reconstruction of state mixtures , i.e. the decomposition of mutually incoherent modes @xcite . thibault and
menzel @xcite demonstrated the reconstruction of five dominant spatial modes in a partially coherent x - ray beam used for ptychography illumination .
shortly thereafter , batey et al .
@xcite simultaneously illuminated a sample with three separate single - color visible lasers with blue , green and red wavelengths , and successfully recovered the sample s response for each of these three colors using the pim technique .
cdi requires light sources with a high degree of spatial coherence . to address this challenge at short wavelengths , coherent light sources based on tabletop high harmonic generation ( hhg ) and large - scale free electron lasers ( fels ) are under rapid development .
high harmonic generation @xcite is an extreme nonlinear process that coherently upconverts infrared , visible , or uv light to extreme ultraviolet ( euv ) and x - ray wavelengths .
hhg is unique as a light source because the generation conditions can be adjusted so that the light emerges either as a coherent supercontinuum ( corresponding to an isolated attosecond burst in time ) , or as a comb of harmonics , with periodicity both in space and time . under the right conditions , the harmonics are spectrally narrow ( @xmath0 ) @xcite and also tunable by changing the driving laser wavelength or chirp . since its demonstration in 2007 , cdi using hhg sources has become a successful imaging modality , enabling high resolution image reconstruction both in transmission @xcite and more recently in reflection @xcite . despite these successes ,
hhg cdi contains untapped potential because the spectral structure of hhg has been underutilized .
indeed , recent work used broadband hhg beams as a way to make more efficient use of available photons @xcite , by assuming that the object looks identical at all constituent wavelengths @xcite . in this work ,
we extend pim @xcite to utilize multiple high harmonics as the illumination source , in an important spectral range offering element - specific contrast due to adjacent absorption edges . by relaxing the previous assumption of a uniform spectral response
, we retrieve a wealth of information about the sample , including the spectrally - dependent amplitude and phase that encode the sample composition and topography . to achieve this
, we illuminate the sample with several harmonic orders simultaneously , and employ the pim algorithm to reconstruct independent images at each wavelength .
thus , in addition to high - contrast , high - spatial - resolution imaging systems , we can now achieve intrinsic element - specific contrast at multiple wavelengths simultaneously .
information redundancy in ptychography not only allows the sample , but also the illumination to be reconstructed .
here we aslo demonstrate in situ imaging of a multicolor euv beam .
finally , we note that the combination of a comb of coherent harmonics and the pim algorithm is the most efficient use of hhg euv radiation for imaging to date because there is no energy loss from any multilayer mirrors or monochromatizing optics .
the experimental setup is shown in fig .
1 . phase matched high harmonics near 30 nm were generated by focusing ultrashort pulses from a ti : sapphire amplifier ( kmlabs dragon , 790 nm central wavelength , 5 khz repetition rate , 1.5 mj pulse energy , 22 fs pulse width ) into a hollow waveguide ( 150 inner diameter , 5 cm long ) filled with argon at 35 torr .
in contrast to previous experiments , no euv multilayer mirrors were used to select a single harmonic .
instead , an ellipsoidal mirror refocused the all the harmonics onto the sample , with an estimated beam diameter of 10 based on a knife - edge measurement .
due to geometrical constraints of our imaging chamber , we added a pair of gold mirrors placed at an angle of 45 to steer the beam .
the sample ( titanium features patterned on silicon @xcite ) was placed at a 50.5 angle of incidence .
diffraction from the sample , which is an incoherent superposition of scattered light from all four harmonics , was measured by an euv - sensitive ccd ( andor ) detector placed 55 mm away from the sample , and was positioned normal to the specular reflection of the beam . the ptychographical scan consisted of a 28 @xmath1 11 rectangular grid ( with added random offsets up to 20% of the step size to prevent periodic artifacts in the reconstruction ) .
we chose a scanning step size of 1 to ensure enough information redundancy for the ptychographical dataset .
the equivalent numerical aperture ( na ) of the collected diffraction is 0.086 , providing a half - pitch resolution near 170 nm .
the exposure time was set to 1.5 seconds to avoid saturation of the ccd .
we accumulated three exposures at each position to increase the signal - to - noise ratio , resulting in a total exposure time of 23 minutes . in a chamber designed specifically for this experiment ,
the gold steering mirrors , which have a combined reflectivity of 2% , could be removed and the required exposure time would then be reduced to 30 seconds using the hhg source described here . , or effective na ( scale bar 0.02 ) .
( c ) spectral intensity .
dashed - line : line cut along the magenta line in ( b ) .
solid line : higher - resolution estimate of the spectrum by deconvolving the modulus square of the remapped diffraction in ( b ) . ] to precisely determine the wavelengths of the harmonics , we placed a two - dimensional ( 2d ) , @xmath2 nm period nano - pillar grating adjacent to the sample in the same plane .
this grating acted as a low - resolution spectrometer .
diffraction from this grating illuminated by the euv harmonic comb is shown in fig .
we can clearly see the first - order diffraction peaks corresponding to spatial frequency @xmath3 for different harmonics .
we remapped @xcite the diffraction pattern to a uniform grid of spatial frequencies normalized by @xmath4 , with the result shown in fig .
2(b ) . the spectral intensity , calculated as the modulus square of the diffraction amplitude along the magenta line in fig .
2(b ) , is shown as a dashed line in fig .
2(c ) , where the wavelength axis is calculated as the product of and the normalized spatial frequency .
the peaks are located at 29.1 , 31.5 , 34.6 , and 38.1 nm , and correspond to harmonic orders 27 , 25 , 23 , and 21 of the laser driving wavelength centered at 790 nm . the spectrum obtained in this way
are blurred due to the fact that the hhg beam was not focused onto the detector ( see dashed line in fig .
2(c ) ) . to obtain spectral intensities more representative of the actual hhg spectrum
, we performed a 2d deconvolution of the remapped diffraction intensity using the lucy - richardson algorithm , with the zero - order diffraction taken to be the point spread function .
the results of this analysis give an upper bound to the harmonic widths of @xmath5 ( see solid line in fig .
we ignore the weak harmonic at 41 nm .
the four - harmonic spectrum spans a width @xmath6 of 27% .
we use a modified version of the pim algorithm @xcite to reconstruct the response of the sample to the different hhg wavelengths present in the illumination .
the modification is that we do not constrain the spectral weights during iterations , and let the relative scale of the probe and the sample float freely , while their product is invariant .
this way , the algorithm will also work for applications where spectral weight information is missing , with the understanding that for each color , there is an undetermined relative scale factor between the reconstructed probe and the reconstructed object .
the initial guess for the probe beams is obtained from a knife - edge focus measurement @xcite .
the entire reconstruction process consisted of @xmath7 1300 iterations . during the final @xmath7 100 iterations
, we implemented the position refinement algorithm @xcite for each color .
an illustration of how multicolor diffraction patterns are decomposed using pim is shown in fig .
we select the scan position centered around the crosshair mark in the inset of fig .
3(a ) , and display the measured diffraction intensity ( after remapping @xcite ) .
figure 3(b - e ) show the four quasi - monochromatic components , as determined by the pim algorithm , that sum incoherently to form the pattern shown in fig .
figure 3(f ) shows the intensity profiles of fig .
3(a - e ) along the dashed line .
the lower contrast of the diffraction fringes in the mixed diffraction pattern in fig .
3(a ) relative to that of its coherent components is clear .
the different colors have different scattering angles for the same spatial frequencies and different spatial distribution of illumination fields , causing blurring in the mixed diffraction pattern .
we reconstructed the complex amplitude of the reflection from the sample , with the moduli shown in fig .
4(a - d ) and phases shown in fig .
4(e - h ) . as a reference , an sem image of the sample
is shown in fig .
4(m ) . excluding the reconstruction for the 21st harmonic ,
the main features of the sample are reconstructed well .
there are several types of artifacts present in the pim reconstructions , such as double - lining near edges and noise on the silicon substrate . compared with our previous reconstruction with single color illumination @xcite ,
the multicolor reconstruction has degraded quality .
the probable reason is that while the signal - to - noise ratio for the sum of all four colors is about the same as for the single - color case , after decomposition into four colors , each color has much lower signal - to - noise ratio leading to lower image quality .
this is mainly a technical limitation due to the limited dynamic range of the detector used in this experiment , and should not be considered to be a fundamental limitation to the technique .
the image reconstruction quality is especially low for the 21st harmonic , which not only shows a low spectral weight in the measured spectrum in fig .
2(c ) but also has a lower reflectivity from titanium : the reflectivity at 38.1 nm is only 11% of that at 29.1 nm @xcite .
ignoring other weak harmonics present in the spectrum may also lead to reconstruction errors . reconstructed moduli and phases of the hhg beam at four different wavelengths are shown in fig .
4(i - l ) , with fig .
4(n ) showing the different color beams on the same plane to provide a direct view of the entire hhg beam . the outline of each beam in fig .
4(n ) corresponds to the @xmath8 intensity level . in terms of the hhg probe beam reconstruction
, we can clearly see astigmatism in fig .
4(i - l ) , with the wavefront converging horizontally and diverging vertically .
this extracted wavefront curvature agrees with our observations during the experiment : using a knife - edge measurement we found that the x - focus is after the y - focus by roughly 400 along the beam axis , and we positioned the sample at the midpoint of the x- and y - foci .
are normalized such that @xmath9 for silicon for all wavelengths ; while in ( b ) the phases for silicon are set to zero for all wavelengths .
( c ) and ( d ) show the comparison of modulus @xmath10 and phase difference @xmath11 from theory , the single color experiment in previous work , and the multicolor experiment in this work . ]
to verify the spectral contrast obtained in the pim experiments , we performed a comparison against both the previous single - color work @xcite as well as theoretical calculations .
for the region marked by a white rectangle in fig .
4(a ) , we performed histogram analysis of the modulus ratio of the titanium region versus that of the silicon region @xmath12 , and also the phase difference between the two regions , @xmath11 , as shown in fig .
5(a ) and fig .
to obtain the theoretical values of these two quantities , we used the results from a previous characterization of the same sample @xcite : the thickness of the oxidation layers on top of the titanium and the silicon regions are assumed to be 2.92 nm and 3.01 nm thick , respectively . from an atomic force microscopy measurement , we characterized the roughness of the titanium region and the silicon region to be 0.91 nm and 0.23 nm respectively , and the height difference @xmath13 to be 32.7 nm . we found evidence of up to 1 nm of carbon buildup in samples maintained in the same environment as the one used in this work , using auger electron spectroscopy . here
we assume a thickness between 0 and 1 nm of the carbon contamination , which is a source of uncertainty in the theoretical reflectivities of the sample surfaces . with this model of the sample , and
tabulated values of indices of refractions and reflectivities @xcite , we calculated @xmath10 and @xmath11 ( the total phase results from both the complex fresnel reflection coefficient and the height difference @xcite ) as a function of wavelength .
the comparison between these calculations and the results from the previous single color experiment @xcite along with the current multicolor experiment is shown in fig .
5(c ) and fig .
both of these comparisons demonstrate good agreement between the multicolor and single - color experimental results and theoretical simulations .
as illustrated above , the combination of high harmonic combs and ptychographic information multiplexed diffraction imaging allows the amplitude and phase response of a sample to be recovered at multiple wavelengths simultaneously .
each of the four illuminating probes , one for each phase - matched harmonic , is reconstructed separately , over a spectral range corresponding to @xmath14 .
no wavelength - scanning or separation hardware was used , making the experimental setup simpler than other techniques while at the same time enabling more efficient use of available photons .
the combination of a comb of coherent harmonics with the pim algorithm is the most efficient use of hhg euv radiation for imaging to date because there is no energy loss from any multilayer mirrors or monochromatizing optics .
the limitations to this new spectromicroscopy technique are not currently well understood , including how many colors , @xmath15 , can be reconstructed simultaneously , or what spectral resolving power , @xmath16 , can be achieved .
the limit for @xmath15 can be estimated by simply considering the ratio between the number of knowns and unknowns in the phase retrieval problem . if a point on the sample is illuminated with @xmath17 overlapping ptychography scan positions , i.e. it is independently measured @xmath17 times , and the diffraction pattern is sampled at an oversampling @xcite ratio @xmath18 , then there is @xmath19-fold information redundancy for that point . beyond @xmath20 ,
no further information is gained , so for @xmath21 the information redundancy is simply @xmath22 .
each color needs two - fold redundancy for phase retrieval , so an upper bound for @xmath15 is @xmath23 for @xmath24 , and @xmath25 for @xmath21 .
it is reasonable to believe that there is a point beyond which increasing @xmath17 ( increasing overlap between scan positions ) fails to result in increased information redundancy i.e. adjacent scans are no longer independent .
regarding spectral resolution , the detection of scattered light at high numerical aperture can help separate adjacent colors via diffraction . in object space
, this translates to the fact that in order to discriminate between adjacent colors , the respective fields of view must differ in image size ( number of pixels ) . with this in mind
the spectral resolution is limited to approximately @xmath26 , where @xmath27 is the maximum distance between scan positions , and na is the numerical aperture .
thus an increase in either the na or the maximum distance between scan positions can result in increased spectral resolution .
for the experiment described here , the estimated upper bound for @xmath28 is @xmath7100 , much greater than the 4 colors reconstructed here .
the spectral resolving power is estimated to be 80 , which is higher than the necessary resolving power of 15 necessary to distinguish adjacent harmonics .
looking forward , the spatial resolution can easily reach below 50 nm by placing the detector closer to the sample for a higher numerical aperture @xcite .
various element - specific absorption edges in the euv / x - ray spectrum range at higher photon energies than studied here provide opportunities for high - contrast spectral imaging .
broadband and narrowband shorter wavelength hhg sources are being developed that are bright ( i.e. phase matched ) and that can operate in the water window range @xcite for biological imaging , reaching photon energies up to 1.6 kev @xcite .
it is also possible to extend this technique to multicolor sources other than hhg , such as synchrotrons and free electron lasers .
the hhg source used in this work has a favorable spectral structure with discrete spectral lines , each of which has a narrow bandwidth .
the question of how to address more continuous , broadband spectra with this technique will be addressed in future work .
each harmonic has a transform limited pulse duration on the order of @xmath29 femtoseconds ( 30 nm wavelength with a spectral bandwidth of @xmath30 and a gaussian profile corresponds to pulse duration of 5 femtoseconds ) . by making use of the ultrashort pulse duration of the hhg source in a pump - probe geometry
, this technique can be used for spectral imaging of ultrafast charge , spin and phonon dynamics in functioning nanosystems @xcite .
this work was performed at jila .
we thank dr .
ming - chang chen for his help on the ellipsoidal mirror .
we also gratefully acknowledge support from the defense advanced research projects agency ( darpa ) pulse program , the nsf engineering center in euv science and technology , the semiconductor research corporation ( 2013-oj-2443 ) , the national science foundation cosi igert 0801680 , the national science foundation graduate research fellowship program , ford foundation fellowship program , and the katherine burr blodgett fellowship program .
the current address of b.z .
is kla - tencor corporation , one technology drive , milpitas , ca 95035 , usa .
the current address of m.h.s .
is slac national accelerator laboratory , menlo park , ca 94025 , usa . | we demonstrate a new scheme of spectromicroscopy in the extreme ultraviolet ( euv ) spectral range , where the spectral response of the sample at different wavelengths is imaged simultaneously .
it is enabled by applying ptychographical information multiplexing ( pim ) to a tabletop euv source based on high harmonic generation , where four spectrally narrow harmonics near 30 nm form a spectral comb structure .
extending pim from previously demonstrated visible wavelengths to the euv / x - ray wavelengths promises much higher spatial resolution and more powerful spectral contrast mechanism , making pim an attractive spectromicroscopy method in both the microscopy and the spectroscopy aspects . besides the sample , the multicolor euv beam is also imaged in situ , making our method a powerful beam characterization technique .
no hardware is used to separate or narrow down the wavelengths , leading to efficient use of the euv radiation
. 99 h. ade and b. hsiao , `` x - ray linear dichroism microscopy , '' science * 262 * , 14271429 ( 1993 ) .
h. ade , j. kirz , s. l. hulbert , e. d. johnson , e. anderson , and d. kern , `` x - ray spectromicroscopy with a zone plate generated microprobe , '' appl .
. lett . * 56 * , 18411843 ( 1990 ) .
p. horowitz and j. a. howell , `` a scanning x - ray microscope using synchrotron radiation , '' science * 178 * , 608611 ( 1972 ) .
d. j. batey , d. claus , and j. m. rodenburg , `` information multiplexing in ptychography , '' ultramicroscopy * 138 * , 1321 ( 2014 ) .
h. m. l. faulkner and j. m. rodenburg , `` movable aperture lensless transmission microscopy : a novel phase retrieval algorithm , '' phys .
rev . lett . * 93 * , 023903 ( 2004 ) .
j. m. rodenburg , a. c. hurst , a. g. cullis , b. r. dobson , f. pfeiffer , o. bunk , c. david , k. jefimovs , and i. johnson , `` hard - x - ray lensless imaging of extended objects , '' phys .
rev .
lett . * 98 * , 034801 ( 2007 ) .
p. thibault , m. dierolf , a. menzel , o. bunk , c. david , and f. pfeiffer , `` high - resolution scanning x - ray diffraction microscopy , '' science * 321 * , 379382 ( 2008 ) .
m. d. seaberg , b. zhang , d. f. gardner , e. r. shanblatt , m. m. murnane , h. c. kapteyn , and d. e. adams , `` tabletop nanometer extreme ultraviolet imaging in an extended reflection mode using coherent fresnel ptychography , '' optica * 1 * , 3944 ( 2014 ) . b. zhang , d. f. gardner , m. d. seaberg , e. r. shanblatt , h. c. kapteyn , m. m. murnane , and d. e. adams , `` high contrast 3d imaging of surfaces near the wavelength limit using tabletop euv ptychography , '' ultramicroscopy * 158 * , 98104 ( 2015 ) .
p. thibault and a. menzel , `` reconstructing state mixtures from diffraction measurements , '' nature * 494 * , 6871 ( 2013 ) .
r. karl , c. bevis , r. lopez - rios , j. reichanadter , d. gardner , c. porter , e. shanblatt , m. tanksalvala , g. f. mancini , m. murnane , h. kapteyn , and d. adams , `` spatial , spectral , and polarization multiplexed ptychography , '' opt .
express * 23 * , 3025030258 ( 2015 ) .
a. rundquist , c. g. durfee iii , z. chang , c. herne , s. backus , m. m. murnane , and h. c. kapteyn , `` phase - matched generation of coherent soft x - rays , '' science * 280 * , 14121415 ( 1998 ) .
r. a. bartels , s. backus , e. zeek , l. misoguti , g. vdovin , i. p. christov , m. m. murnane , and h. c. kapteyn , `` shaped - pulse optimization of coherent emission of high - harmonic soft x - rays , '' nature * 406 * , 164166 ( 2000 ) .
m .-
c . chen , p. arpin , t. popmintchev , m. gerrity , b. zhang , m. d. seaberg , d. popmintchev , m. m. murnane , and h. c. kapteyn , `` bright , coherent , ultrafast soft x - ray harmonics spanning the water window from a tabletop light source , '' phys .
rev .
lett .
* 105 * , 173901 ( 2010 ) .
t. popmintchev , m .- c .
chen , d. popmintchev , p. arpin , s. brown , s. aliauskas , g. andriukaitis , t. baliunas , o. d. mcke , a. pugzlys , a. baltuka , b. shim , s. e. schrauth , a. gaeta , c. hernndez - garca , l. plaja , a. becker , a. jaron - becker , m. m. murnane , and h. c. kapteyn , `` bright coherent ultrahigh harmonics in the kev x - ray regime from mid - infrared femtosecond lasers , '' science * 336 * , 12871291 ( 2012 ) .
r. a. bartels , a. paul , m. m. murnane , h. c. kapteyn , s. backus , y. liu , and d. t. attwood , `` absolute determination of the wavelength and spectrum of an extreme - ultraviolet beam by a young s double - slit measurement , '' opt .
lett .
* 27 * , 707709 ( 2002 ) .
d. popmintchev , c. hernndez - garca , f. dollar , c. a. mancuso , j. a. prez - hernndez , m .- c .
chen , a. hankla , x. gao , b. shim , a. l. gaeta , m. tarazkar , d. a. romanov , r. j. levis , j. a. gaffney , m. foord , s. b. libby , a. jaron - becker , a. becker , l. plaja , m. m. murnane , h. c. kapteyn , and t. popmintchev , `` ultraviolet surprise : efficient soft x - ray high - harmonic generation in multiply ionized plasmas , '' science * 350 * , 12251231 ( 2015 )
. t. fan , p. grychtol , r. knut , c. hernndez - garca , d. d. hickstein , d. zusin , c. gentry , f. j. dollar , c. a. mancuso , c. w. hogle , o. kfir , d. legut , k. carva , j. l. ellis , k. m. dorney , c. chen , o. g. shpyrko , e. e. fullerton , o. cohen , p. m. oppeneer , d. b. miloevi , a. becker , a. a. jaro - becker , t. popmintchev , m. m. murnane , and h. c. kapteyn , `` bright circularly polarized soft x - ray high harmonics for x - ray magnetic circular dichroism , '' proc .
natl . acad .
sci .
u.s.a . *
112 * , 1420614211 ( 2015 ) .
r. l. sandberg , a. paul , d. a. raymondson , s. hdrich , d. m. gaudiosi , j. holtsnider , r. i. tobey , o. cohen , m. m. murnane , h. c. kapteyn , c. song , j. miao , y. liu , and f. salmassi , `` lensless eiffractive imaging using tabletop coherent high - harmonic soft - x - ray beams , '' phys .
rev .
lett . * 99 * , 098103 ( 2007 ) . m. d. seaberg , d. e. adams , e. l. townsend , d. a. raymondson , w. f. schlotter , y. liu , c. s. menoni , l. rong , c .- c .
chen , j. miao , h. c. kapteyn , and m. m. murnane , `` ultrahigh 22 nm resolution coherent diffractive imaging using a desktop 13 nm high harmonic source , '' opt .
express * 19 * , 2247022479 ( 2011 ) .
b. zhang , m. d. seaberg , d. e. adams , d. f. gardner , e. r. shanblatt , j. m. shaw , w. chao , e. m. gullikson , f. salmassi , h. c. kapteyn , and m. m. murnane , `` full field tabletop euv coherent diffractive imaging in a transmission geometry , '' opt .
express * 21 * , 2197021980 ( 2013 ) .
d. f. gardner , b. zhang , m. d. seaberg , l. s. martin , d. e. adams , f. salmassi , e. m. gullikson , h. c. kapteyn , and m. m. murnane , `` high numerical aperture reflection mode coherent diffraction microscopy using off - axis apertured illumination , '' opt . express * 20 * , 1905019059 ( 2012 ) .
b. chen , r. a. dilanian , s. teichmann , b. abbey , a. g. peele , g. j. williams , p. hannaford , l. van dao , h. m. quiney , and k. a. nugent , `` multiple wavelength diffractive imaging , '' phys .
rev .
a * 79 * , 023809 ( 2009 ) .
b. abbey , l. w. whitehead , h. m. quiney , d. j. vine , g. a. cadenazzi , c. a. henderson , k. a. nugent , e. balaur , c. t. putkunz , a. g. peele , g. j. williams , and i. mcnulty , `` lensless imaging using broadband x - ray sources , '' nat .
photonics * 5 * , 420424 ( 2011 ) .
s. witte , v. t. tenner , d. we noom , and k. se eikema , `` lensless diffractive imaging with ultra - broadband table - top sources : from infrared to extreme - ultraviolet wavelengths , '' light sci
. appl .
* 3 * , e163 ( 2014 ) .
f. zhang , i. peterson , j. vila - comamala , a. diaz , f. berenguer , r. bean , b. chen , a. menzel , i. k. robinson , and j. m. rodenburg , `` translation position determination in ptychographic coherent diffraction imaging , '' opt .
express * 21 * , 1359213606 ( 2013 ) .
b. l. henke , e. m. gullikson , and j. c. davis , `` x - ray interactions : photoabsorption , scattering , transmission , and reflection at e = 50 - 30,000 ev , z = 1 - 92 , '' at .
data nucl .
data tables * 54 * , 181342 ( 1993 ) .
r. h. t. bates , `` fourier phase problems are uniquely uolvable in more than one dimension .
i : underlying theory , '' optik * 61 * , 247262 ( 1982 ) .
k. m. hoogeboom - pot , j. n. hernandez - charpak , x. gu , t. d. frazer , e. h. anderson , w. chao , r. w. falcone , r. yang , m. m. murnane , h. c. kapteyn , and d. nardi , `` a new regime of nanoscale thermal transport : collective diffusion increases dissipation efficiency , '' proc .
natl .
acad .
sci .
u.s.a . *
112 * , 48464851 ( 2015 ) .
e. turgut , c. la - o vorakiat , j. m. shaw , p. grychtol , h. t. nembach , d. rudolf , r. adam , m. aeschlimann , c. m. schneider , t. j. silva , m. m. murnane , h. c. kapteyn , and s. mathias , `` controlling the competition between optically induced ultrafast spin - flip scattering and spin transport in magnetic multilayers , '' phys .
rev .
lett . * 110 * , 197201 ( 2013 ) . |
with the remarkable development of the cooling technique for two - component atomic fermi gases and magnetic - field feshbach resonance @xcite which can change both the strength and sign of the atomic interaction characterized by the scattering length @xmath0 @xcite , experiments have explored the crossover region between the strongly attractive and repulsive interaction . for the spin mixture of the fermionic atoms , there would be dimers of fermionic atom pairs in different form dependent on the attractive or repulsive interaction between atoms .
on the side of repulsive interaction ( bec side ) , there exists molecule which is a short - range fermionic atom pairs @xcite .
in several recent experiments @xcite , the evidence for bose - einstein condensates of diatomic molecules has been given .
on the side of attractive interaction ( bcs side ) , one expects that the dimer is in a state of long - range fermionic pairs analogously to the electronic cooper pairs .
the magnetic - field feshbach resonance gives us an important opportunity to investigate the pairing phenomena in the ultracold fermi gases especially the bcs - bec crossover which has been discussed in a lot of theoretical investigations @xcite and experimental researches jin - fermion , mit - fermion , grimm , salomon , thomas , grimm1,thomas - c2,cheng , thomas - heat such as condensate fraction @xcite , collective excitations @xcite , pairing gap @xcite and the recent measurement of heat capacity @xcite . in the bcs - bec crossover ,
a particular interesting problem is the properties of the ultracold gases at the location of the feshbach resonance where the scattering length between atoms is divergent . at or extremely close to the location of the feshbach resonance ,
the absolute value of the scattering length @xmath0 can be comparable to or even much larger than the average distance @xmath1 between particles .
in this situation , the ultracold gas can be described in the unitarity limit carlson , heiselberg , ho1,ho , ghem where there is a universal behavior for the system .
the feshbach resonant technique has given us the experimental means to investigate the ultracold gas with @xmath2 .
on resonance , the fraction of dimers in zero - momentum state is investigated experimentally in @xcite after the conversion of pairs of atoms into bound molecules .
the experiment by mit group @xcite shows that there is a high fraction of dimers of fermionic atoms extremely close to the feshbach resonance . on resonance , the frequency of the radial breathing mode is given in @xcite by observing the collective oscillations of the system . in the present work , based on a mixture of the atomic and dimeric gases in the unitarity limit @xmath3
, a very simple theory is developed to explain the recent experimental results in @xcite and thomas at the location of the feshbach resonance .
in the present experiments on bcs - bec crossover , an equal mixture of fermionic atoms in two different spin states are confined in a harmonic trap to realize atomic collisions . in thermal equilibrium ,
the system comprises fermionic atoms and dimers , and its dynamic equilibrium is characterized by the fact that the gibbs free energy @xmath4 of the system is a minimum . assuming that @xmath5 and @xmath6 being the chemical potential of the fermi gases and @xmath7 being the chemical potential of the dimeric gas
, the minimum of the gibbs free energy means that @xmath8 the chemical potential of the dimeric gas is @xmath9 with @xmath10 being the dimeric energy and @xmath11 being the contribution to the chemical potential due to the thermal equilibrium of the dimeric gas . assuming that @xmath12 is the total density distribution of the fermi gases without the presence of dimers and @xmath13 is the fraction of the dimers immersed in the ultracold fermi gases , the density distribution of the dimeric gas @xmath14 and fermi gas @xmath15 ( and @xmath16 ) in a spin state are respectively given by @xmath17 and @xmath18 . for the unitarity limit @xmath3
, there is a _ universality hypothesis _ ( see @xcite ) _ _ _ _ that although the scattering length @xmath0 will play an important role , it will not appear in the final result of a physical quantity such as chemical potential because it can be regarded as infinity . in this case , the length scale @xmath19 rather than @xmath0 will appear in the final result of a physical quantity which means a universal behavior for a system . by using the universality hypothesis and local density approximation , at zero temperature
, the chemical potential @xmath5 ( @xmath20 ) of the fermi gas is given by @xmath21 where @xmath22 is the external potential of the fermionic atoms . without @xmath23 ,
the above expression gives the chemical potential of an ideal fermi gas in the local density approximation .
the parameter @xmath23 shows the role of the extremely large scattering length @xmath0 in the unitarity limit .
the parameter @xmath23 was firstly measured in @xcite with a careful experimental investigation of the strongly interacting degenerate two - component fermi gases near the feshbach resonance .
@xmath23 has been also calculated with different theoretical methods @xcite .
the scattering length @xmath24 ( @xmath25 ) @xcite of the dimers is comparable to the scattering length @xmath0 between fermionic atoms , thus the dimeric gas will also show a universal behavior on resonance . in the unitarity limit , for the bosonic dimer gas , the scattering length @xmath24 should not appear in the final result of a physical quantity too . according to the universality hypothesis
, we assume here that the bosonic dimer gas shows an analogous universal behavior with the fermi gas ( see also @xcite ) . based on this assumption
, one has the chemical potential for the dimeric gas at zero temperature @xmath26 the parameter @xmath27 in @xmath7 is also due to the large scattering length @xmath24 between dimers and it can be calculated in the unitarity limit . near the feshbach resonance , the dimeric energy @xmath28 .
thus , on resonance , @xmath10 can be omitted in the above expression . from the dynamic equilibrium condition @xmath29 , one gets the following general equation to determine the fraction of dimers at zero temperature and feshbach resonance : @xmath30 where @xmath31 .
we show the relation between the fraction of dimers @xmath13 and different value of @xmath32 in fig . 1 . in the presence of the dimers ,
the chemical potential of the system is obviously lower than the case without the dimers .
thus , the mixture gas of the atoms and dimers is a stable state of the system . for @xmath33 , there are three important characteristics for the dimeric gas : \(i ) _ the above result for the fraction of the dimeric gas is a general result once the system is in the unitarity limit . _
\(ii ) _ there is a quite high fraction of the dimers as shown in fig.1 .
_ in the present experiments , the fraction of dimers is determined by the molecules in zero - momentum state after a fast magnetic field transfer ( the magnetic field is swept below the feshbach resonant magnetic field @xmath34 so that the scattering length between fermionic atoms becomes positive ) to create bound molecules from the dimers . after the dimer - molecule conversion , in @xcite the maximum fraction of the molecules in zero - momentum state is observed to be @xmath35 which agrees with our theoretical result . from this experimental result , @xmath32 is estimated to be @xmath36 .
\(iii ) _ the dimer comprising two fermionic atoms is quite stable .
_ at zero temperature , the dimeric gas is immersed in the degenerate fermi gases . due to
pauli blocking comes from the degenerate fermi gases , any dimer can not be dissociated into two fermionic atoms once the equilibrium is attained so that @xmath37 .
in fact , the stability of the dimeric gas is consistent with the experiment @xcite that extremely close to the feshbach resonance there is no obvious decreasing of the molecules in zero - momentum state after the dimer - molecule conversion process even after @xmath38 hold time of the final magnetic field . due to the stability and pauli blocking
, the dimer - molecule conversion will always convert the fermionic atom pairs in a same dimer into bound molecules .
thus at zero temperature , all the dimers will be converted into molecules with zero - momentum state during the dimer - molecule conversion process .
this means that the fraction of dimers in thermal equilibrium investigated here can be used to explain the experimental result of the fraction of zero - momentum molecules after the dimer - molecule conversion . which can be calculated in the unitarity limit .
the high fraction of the dimers agrees with the experiment @xcite where the maximum fraction of the molecules in zero - momentum state is observed to be @xmath35 after the dimers are converted into molecules .
the dashed line shows the effect of the existence of the dimeric gas on the width of gases confined in the optical trap .
we see that the existence of dimers has an obvious effect of decreasing the width of the gases in the trap . ] in a recent experiment @xcite , the frequency of a radial breathing mode is observed extremely close to the feshbach resonance and found to agree well with the theoretical predication based on hydrodynamic theory in the unitarity limit @xcite . combining with the experimental result in @xcite ( also shown in the present work ) that there is a high fraction of molecules in zero - momentum state after the dimer - molecule conversion process ,
we see that the dimeric gas should play a dominant role in determining the frequency of collective oscillations due to its high fraction . for the radial breathing mode
, the agreement of the experiment with theoretical result in the unitarity limit shows that it is reasonable to describe the bosonic dimer gas by the unitarity hypothesis which is used to give eq .
( [ chemical - dimer ] ) .
in fact , an analogous form of eq .
( chemical - dimer ) is used to calculate the frequency of the radial breathing mode in @xcite , whose result agrees well with the experiment in @xcite .
different from the theoretical model in @xcite , however , in the present work on resonance , the dimeric gas should play a dominant role in the frequency of the collective oscillations at zero temperature , rather than the fermi gases . on resonance ,
the mixture of the fermionic atoms and dimers will play a role in the width of the gases confined in the harmonic trap . from the dynamic equilibrium @xmath37 ,
the width of the fermi gas is equal to that of the dimeric gas .
assume that @xmath39 is the width of the mixture gases for the atoms and dimers , while @xmath40 is the width of the gases without considering the existence of the dimers .
after a simple calculation , one has the ratio @xmath41 . the dashed line in fig.1 shows @xmath42 as a function of @xmath32 .
we see that the existence of dimers in dynamic equilibrium will decrease the width of the gases .
in summary , the role of the dimers of fermionic atoms is discussed extremely close to the feshbach resonance where both the fermionic atoms and dimers show a universal behavior based on the universality hypothesis .
the dynamic equilibrium of the mixed atom - dimer gas is considered for the case of a harmonic trap through the analysis of the gibbs free energy .
based on this simple theoretical model , it is found that the high fraction of the stable dimeric gas in the present work agrees well with the experiment mit - fermion where there is a high fraction of molecules in zero - momentum state after the conversion of the dimers into molecules . extremely close to the feshbach resonance , the dimeric gas plays a dominant role for the frequency of the radial breathing mode observed in @xcite .
on resonance , the stability of the dimeric gas would be useful in the creation of molecular bec on the bec side and condensate of atomic cooper pairs on the bcs side . crossing the feshbach resonant magnetic field @xmath34 slowly from above ( or below ) would contribute to the creation of molecular bec ( or condensate of atomic cooper pairs ) because there is a formation of a stable dimeric gas when @xmath34 is swept cross slowly .
for example , in @xcite , the molecular bec is created after the magnetic field is swept slowly from above to below @xmath34 . c. a. regal , m. greiner , and d. s. jin , phys .
92 ( 2004 ) 083201 ; c. a. regal , c. ticknor , j. l. bohn , and d. s. jin , nature 424 ( 2003 ) 47 ; j. cubizolles , t. bourdel , s. j. j. m. f. kokkelmans , g. v. shlyapnikov , and c. salomon , phys .
91 ( 2003 ) 240401 ; k. e. strecker , g. b. partridge , and r. g. hulet , phys .
91 ( 2003 ) 080406 ; s. jochim , m. bartenstein , a. altmeyer , g. hendl , c. chin , j. h. denschlag , and r. grimm , phys . rev .
91 ( 2003 ) 240402 . | at the location of a magnetic - field feshbach resonance , a mixture gas of fermionic atoms and dimers of fermionic atom pairs is investigated in the unitarity limit where the absolute value of the scattering length is much larger than the mean distance between atoms .
the dynamic equilibrium of the mixture gases is characterized by the minimum of the gibbs free energy . for the fermionic atoms and dimers with divergent scattering length
, it is found that the fraction of the dimers based on a very simple theory agrees with the high fraction of zero - momentum molecules observed in a recent experiment ( m. w. zwierlein et al , phys .
rev .
lett .
92 , 120403 ( 2004 ) ) .
the dimeric gas can be also used to interpret the frequency of the radial breathing mode observed in the experiment by j. kinast et al ( phys
. rev .
lett .
92 , 150402 ( 2004 ) ) .
pacs numbers : 03.75.ss , 05.30.fk , 05.30.jp , 03.75.hh keywords : fermi gas , dimeric gas , feshbach resonance , ultracold |
our understanding of light hadron spectroscopy has been challenged in recent years by the discovery of several exotic states that can not be easily accommodated into the quark model picture @xcite .
one of them is the resonance @xmath6 @xcite ( or @xmath3 , as we will refer to it from now on ) .
the @xmath3 was first observed by the babar collaboration @xcite with mass @xmath7 mev and width @xmath8 mev @xcite in the @xmath9 reaction , and also found by bes in @xmath10 decay with @xmath11 mev and @xmath12 mev @xcite .
the belle collaboration has performed the most precise measurements up to now of the reactions @xmath13 and @xmath2 finding @xmath14 mev and @xmath15 mev @xcite .
the obtained width is larger than in previous measurements but the error is large .
the same feature has been found in a combined fit to both babar and belle data on @xmath13 and @xmath2 , yielding @xmath16 mev and @xmath17 mev @xcite .
these experimental findings have triggered a significant theoretical activity aimed at unraveling the nature and properties of this resonance .
it has been interpreted as a tetraquark @xcite , with a mass of @xmath18 gev @xcite or @xmath19 gev @xcite calculated using qcd sum rules .
it has also been identified with the lightest hybrid @xmath20 state @xcite with mass in the range 2.1 - 2.2 gev @xcite and a width of 100 - 150 mev @xcite .
conventional @xmath21 states in @xmath22 or @xmath23 configurations have been considered as their masses are expected to be compatible with that of the @xmath3 .
the width of the @xmath22 state is estimated to be @xmath24-@xmath25 mev @xcite while the @xmath23 is disfavored due to the rather large ( @xmath26 mev @xcite ) predicted width . the large width obtained in ref .
@xcite also makes the interpretation of the @xmath3 as a dynamically generated excited state of the @xmath27 meson , within the multichannel generalization of the resonance - spectrum expansion model @xcite , quite unlikely but further improvements of the model might change this conclusion @xcite .
reference @xcite studies the three - body @xmath28 scattering with two - body interactions taken from unitarized chiral perturbation theory @xcite and a resonance with 2170 mev of mass but a width of only 20 mev is generated . in ref .
@xcite we obtained a good description of the @xmath29 scattering data in the threshold region ( @xmath30 gev ) using chiral lagrangians coupled to vector mesons , supporting the conclusion that the @xmath3 has a large mesonic @xmath31 component . in this contest , it is relevant to establish whether there is an isovector companion of the isoscalar @xmath3 .
such an investigation will help constraining theoretical models and their parameters , leading to a better understanding of meson properties in the energy region around 2 gev .
in particular , the faddeev - type calculation of ref .
@xcite that obtains the @xmath3 as a dynamically generated state finds no resonance in the isovector @xmath32 s - wave channel .
experimentally , this isovector resonance could show up in @xmath33 , as suggested in a recent theoretical study of this process @xcite .
it could also be observed in the @xmath34 reaction because the @xmath1 couples strongly to @xmath35 @xcite .
one should stress that the calculations of refs .
@xcite do not take into account @xmath0 final state interactions ( fsi ) which , resonant or not , could be large and have a sizable impact on the predicted cross sections .
in this article we apply the formalism of ref .
@xcite to the s - wave @xmath32 scattering and discuss the possible presence of an isovector @xmath36 dynamically generated resonance around the @xmath0 threshold for parameters that satisfactorily describe the isoscalar @xmath31 channel .
fsi corrections to @xmath33 are also studied . the formalism for @xmath0 scattering
is developed in sec .
[ sec : ff ] followed by the derivation of the scattering amplitudes .
section [ sec : fb ] contains the results and discussions thereof .
our concluding remarks are given in sec .
[ sec : conclu ] .
in order to obtain the @xmath32 amplitude we follow closely our previous paper @xcite , replacing the isoscalar @xmath37 by the isovector @xmath1 .
first , the scattering of the @xmath27 resonance with an s - wave neutral pair of the lightest pseudoscalar mesons in isospin @xmath38 is investigated .
two types of meson pairs are then possible , namely , @xmath39 and @xmath40 ( already a pure @xmath38 state . ) the following channels result @xmath41 both @xmath0 s- and d- waves contribute to the @xmath42 channel but since we are interested in the threshold region around 2 gev , d - wave terms can be neglected .
they are suppressed by powers of @xmath43 , where @xmath44 is the three momentum of the @xmath32 pair in the center of mass ( cm ) , and @xmath45 is the number of @xmath0 d - wave initial and final states involved .
moreover , as both @xmath27 and @xmath1 are very close to the @xmath46 threshold , the amplitude at tree level is dominated by diagram fig .
[ fig : pole0]a .
the main reason is that the propagator of the kaon intermediate state is almost on - shell . ,
@xmath47 and @xmath48 vertices in the @xmath49 lagrangian of eq .
( [ vvff ] ) .
see ref .
@xcite for a detailed analysis where all the other tree - level diagrams originating from the same set of lagrangians employed are discussed and shown to be suppressed compared to fig .
[ fig : pole0]a . ]
we also include the local term of fig .
[ fig : pole0]b because the off - shell part of the four - pseudoscalar - meson vertex can cancel the kaon propagator generating local terms .
therefore one has to consider simultaneously the sum of amplitudes from both diagrams as any splitting would depend on field parameterization
. the required vertices can be obtained from the lowest order su(3 ) chiral lagrangian @xcite @xmath50 with @xmath51 the pion weak decay constant in the chiral limit , that we approximate to @xmath52 mev .
the octet of the lightest pseudoscalar fields are included in @xmath53 as @xmath54 assuming minimal coupling , the covariant derivative @xmath55 incorporates the lightest octet of vector resonances @xmath56 as external fields : @xmath57 \ , , { \nonumber}\\ v_\mu&=\left ( \begin{array}{ccc } \frac{\rho^0}{\sqrt{2}}+\frac{1}{\sqrt{2}}\omega & \rho^+ & { k^*}^+\\ \rho^- & -\frac{\rho^0}{\sqrt{2}}+\frac{1}{\sqrt{2}}\omega & { k^*}^0\\ { k^*}^- & { { \overline{k}^ * } } ^0 & \phi \end{array } \right ) \ , , \label{matv}\end{aligned}\ ] ] where @xmath58 is a coupling constant .
we have assumed ideal mixing , so that @xmath59 and @xmath60 , with @xmath61 and @xmath62 being the @xmath63 octet and singlet vector states . as a result , the following lagrangians involving vector and pseudoscalar mesons arise from eq . :
@xmath64 where @xmath65 and @xmath66 and @xmath67 are the pion and kaon masses . to construct the @xmath68 amplitude we cast @xmath69 as @xmath70 with @xmath71 and @xmath72 the kaon four - momenta .
the global minus sign appears because we identify @xmath73 to be consistent with the convention adopted in the chiral lagrangians eq .
( [ uu3 ] ) .
we denote the amplitudes for the reaction channels @xmath74 as @xmath75 , @xmath76 , @xmath77 and @xmath78 , from top to bottom .
these amplitudes were calculated in ref .
@xcite for the diagrams of fig .
[ fig : pole0 ] assuming isospin symmetry .
the result for diagram fig .
[ fig : pole0]a is @xmath79 \right .
{ \nonumber}\\ & \left .
+ 6 ( u_b-2m_k^2)\bigl[d(q+k_2)+d(q - k'_1)\bigr ] \right\ } \,,{\nonumber}\\ t_{nn}^{(a)}&=t^{(a)}_{cc } \,,{\nonumber}\\ t_{nc}^{(a)}&=\frac{1}{2}t^{(a)}_{cc } \ , , { \nonumber}\\ t_{cn}^{(a)}&=t_{nc}^{(a ) } = \frac{1}{2}t^{(a)}_{cc } \ , , \label{tcn}\end{aligned}\ ] ] where @xmath80 ( @xmath81 ) is the polarization four - vector of the initial ( final ) @xmath27 meson , @xmath82 , @xmath83 and @xmath84 .
the kaon propagator @xmath85 is given by @xmath86 with @xmath87 .
for the contact term ( diagram of fig .
[ fig : pole0]b ) the result is @xmath88 taking into account eq .
one finds for the @xmath68 channel of eq .
( [ channels ] ) @xmath89 where @xmath90 ( and analogously for @xmath78 , @xmath77 and @xmath76 ) . therefore , @xmath91+k_1\cdot k'_2 \bigl[d(q+k_2)+d(q - k'_1)\bigr ] \bigr\ } \ , .
\label{ti1}\end{aligned}\ ] ] proceeding in the same way , the @xmath92 and @xmath93 amplitudes are found to be @xmath94 \left [ d(q+k_1)+d(q+k_2 ) \right]\ , , { \nonumber}\\ t^{i=1}_{21 } & = -\frac{2 g^2}{\sqrt{6 } f^2}{\epsilon}\cdot { \epsilon } ' \left [ 3 k_1\cdot k_2 + m_\pi^2 \right ] \left [ d(q - k'_1)+d(q - k'_2 ) \right]\ , , \label{extra.t}\end{aligned}\ ] ] where @xmath71 , @xmath72 ( @xmath95 , @xmath96 ) are the four - momenta of the initial ( final ) pseudoscalars .
the gell - mann - okubo mass relation @xmath97 has been used to simplify the final expressions .
notice that there is no local term due to a cancellation between the contact term from fig .
[ fig : pole0]b and the local part from fig .
[ fig : pole0]a
. finally , @xmath98 is absent at tree level because there are no @xmath99 or @xmath100 vertexes with only @xmath101 and @xmath102 mesons .
because of the absence of the contact terms for @xmath103 , @xmath104 and @xmath105 there is no need to further consider these processes in order to obtain the @xmath0 interaction kernel .
it can be obtained directly from @xmath106 , as it is explicitly worked out below .
next , we consider initial- and final - state re - scattering of the pseudoscalar mesons in @xmath38 and s - wave from the diagrams in fig .
[ fig : pole0 ] , as shown in fig .
[ fig : itera ] for the nonlocal part of the interaction .
the re - scattering chains , made of @xmath46 and @xmath107 pairs , contain the poles of the initial and final @xmath1 resonances @xcite .
below , the residue at the @xmath1 double pole will be identified as the @xmath108 interaction kernel @xmath109 .
we follow refs.@xcite , where the @xmath38 s - wave meson - meson scattering was studied with @xmath46 and @xmath110 coupled channels , and the @xmath1 resonance was dynamically generated from the meson - meson self - interactions .
this conclusion is also shared with other approaches like refs.@xcite .
the @xmath38 s - wave meson - meson amplitudes @xmath111 fulfill the bethe - salpeter equation in coupled channels @xcite @xmath112 where the indices @xmath113 denote the @xmath46 and @xmath107 channels .
the @xmath114-matrix is given in terms of the on - shell part of the @xmath38 s - wave meson - meson amplitudes at tree level @xmath115 and the @xmath46 and @xmath107 unitary scalar loop functions , @xmath116 and @xmath117 in this order .
given in eqs .
( [ gff ] ) and ( [ g.cut ] ) , obtained from a dispersion relation and with cut - off regularization , respectively , are also applicable here after the appropriate replacement of masses . ]
notice that the @xmath115 factorize in eq .
they are calculated from @xmath118 , eq .
( [ vvff ] ) , with the resulting expressions @xcite @xmath119 with @xmath120 being the invariant mass squared of the meson pair . in presence of re - scattering of the initial and final two - body hadronic states ,
the dressed amplitudes @xmath121 can be cast as @xmath122 the first ( last ) term in parentheses accounts for the initial ( final ) state interactions between the pair of pseudoscalar mesons in i=1 and s - wave . for its derivation and other applications see refs .
the @xmath123 part , which contains the @xmath27 interaction with the pseudoscalar pair projected into s - wave , consists of two terms , @xmath124 .
the first one is a local term , present only in the @xmath68 channel as shown above . from eq .
( [ ti1 ] ) @xmath125 where only the leading non relativistic contribution to @xmath126 has been kept ; this approximation is justified for small @xmath27 ( and @xmath1 ) velocities in the @xmath0 cm frame . the second term @xmath127 is given by the triangular loop diagrams depicted in fig . [ fig : triangle ] with only kaons in the internal lines . for the @xmath128 vertices
we take only the on - shell amplitudes of eq .
( [ v22.eq ] ) .
the off - shell parts are proportional to the inverse of kaon propagators and cancel with them in the calculation of the loop , giving rise to amplitudes that do not correspond anymore to the dominant triangular kaon - loop but to other topologies @xcite . nonetheless one should bare in mind that some of these sub - leading contributions may alter the contact term , fixed above from the tree level amplitudes . we obtain @xmath129 where @xmath130 \ , , \label{ms}\end{aligned}\ ] ] with @xmath131 \ , ; \label{c}\end{aligned}\ ] ] @xmath120 ( @xmath132 ) stand for the invariant mass squared of the initial ( final ) pseudoscalar - meson pair . inside the integral
we take @xmath133 , which holds at the @xmath1 double pole .
we account for the s - wave projection by averaging over @xmath134 with @xmath135 the relative angle between @xmath27 incoming ( @xmath136 ) and outgoing ( @xmath137 ) momenta in the cm frame . in terms of this angle @xmath138 . as for the contact term we approximate @xmath126 .
further details on the derivation of the triangular - loop amplitude can be found in ref .
@xcite . given in eq .
( [ c ] ) , although more compact , coincides with eq . ( 2.24 ) of ref . @xcite . ]
altogether , @xmath139 substituting the previous expression in eq .
( [ ifsi ] ) and using eq .
( [ bs ] ) one finds that @xmath140 [ \delta_{1j } - g_1(k'^2 ) t_{j1}(k'^2 ) ] - t_{i1}(k^2 ) t_{j1}(k'^2 ) l_s \right\ } \ , .
\label{eq.ms}\ ] ] now we proceed to extract the @xmath0 interaction kernel . for this purpose
we notice that the scattering amplitude @xmath141 contains the @xmath1 resonance pole with residue @xmath142 where @xmath143 denotes the @xmath1 pole position .
therefore , @xmath144\ , .
\label{vff}\end{aligned}\ ] ] the @xmath145 factor appears because @xmath146 contains two extra @xmath147 couplings that should be removed in order to isolate the @xmath1 resonances .
finally , the @xmath0 s - wave scattering amplitude is @xmath148 for a general derivation of this equation , analogous to eq . , based on the n / d method see refs .
@xcite . using dispersion relations , the @xmath0 loop function , @xmath149 ,
is found to be @xcite @xmath150\biggr\}\ , , \label{gff}\end{aligned}\ ] ] with @xmath151 . while the renormalization scale @xmath152 is fixed the to value of the @xmath135 meson mass , @xmath153 mev
, the subtraction constant @xmath154 has to be fitted to data @xcite .
the loop - function can also be regularized with a three - momentum cut - off @xmath155 @xcite , @xmath156 with @xmath157 .. ] it is instructive to compare the real part of the @xmath149 functions that result from the two methods . for this
we fix @xmath158 gev , corresponding to the pole mass obtained in ref .
the comparison is presented in fig .
[ fig : gs ] . on the left panel , eq .
is evaluated varying the subtraction constant @xmath154 from @xmath159 to @xmath160 in steps of @xmath161 starting from the top while on the right one , eq .
is plotted for @xmath155 between 0.8 and 1.2 gev ( around the typical hadronic scale @xmath162 ) in steps of 0.1 gev from top to bottom .
we observe a significant overlap between both functions in the threshold region ( @xmath30 gev ) for values of @xmath154 between @xmath163 and @xmath164 .
this interval contains indeed the @xmath154 values obtained in ref .
@xcite by fitting the @xmath29 cross section .
this coincidence is interpreted as an indication that the @xmath3 resonance is to a large extent dynamically generated .
now , we investigate this possibility for the @xmath38 s - wave @xmath0 scattering .
in this investigation , we consider two possibilities for the @xmath1 properties ( pole position and residue ) , as they depend on the adopted approach . in the first one ,
the bethe - salpeter ( bs ) equation for meson - meson scattering was solved using cut - off regularization for the loop function @xcite . in the second case , the n / d method was used with the meson - meson loop function obtained with a dispersion relation @xcite .
it additionally includes the s - channel exchanges of tree - level scalar resonances , corresponding to a flavor singlet of mass close to 1 gev and a higher octet of mass around 1.4 gev .
pole position obtained with the n / d approach is almost identical to the one obtained with the inverse amplitude method @xcite . ] in both studies the @xmath46 and @xmath107 coupled channels were considered for @xmath38 .
the @xmath1 properties extracted in these references are listed in table [ a0proper ] .
.@xmath1 properties , pole position @xmath166 and residue @xmath167 , used as input .
[ a0proper ] [ cols="^,^,^",options="header " , ] [ bb]@xmath168 [ tc]@xmath169 [ gev ] we calculate @xmath168 for the four possible combinations of the parameter sets in tables [ a0proper ] and [ f0proper ] . as mentioned above
, some of the discarded contributions to the triangle loop could modify the local term in eq . .
for this reason , we first exclude the local contribution and concentrate on the more robust triangular topology . the @xmath168 dependence on the @xmath32 invariant mass is shown in fig .
[ t2_nonl ] .
all the curves show a prominent enhancement below the @xmath170 threshold ( 2.01 gev ) that hints at the presence of a dynamically generated resonance located quite close but above the @xmath171 threshold ( 1.7 gev ) . for fit 2 ,
the peak is narrower and has a maximum at a higher @xmath169 but it is 2.5 times weaker than for fit 1 ( notice the different scales in the plots ) . [ bb]@xmath168 [ tc]@xmath169 [ gev ] let us now take into account the local term in the kernel as given in eq . .
for the sake of consistency the @xmath46 unitarity scalar loop function , @xmath172 , is evaluated making use of the same regularization procedure employed in generating the @xmath1 resonance from refs .
@xcite . hence ,
when the bs set is used , @xmath173 is computed using a cut off regularization with @xmath174 gev @xcite while , when the n / d parameters are considered , @xmath173 is obtained from a dispersion relation with the renormalization scale fixed at the @xmath135 mass , @xmath175 gev , and a subtraction constant of @xmath176 @xcite .
the new results are shown in fig [ t2 ] . in the bs case , for both fits 1 , 2 , the enhancements observed before in fig .
[ t2_nonl ] are flatten away by the presence of the local term .
this agrees with the results of ref .
@xcite , that also makes use of the meson - meson amplitudes from ref .
@xcite , where no isovector @xmath42 resonance was generated .
remarkably , when the n / d set is employed the resonance peak is still clearly seen , and at a higher invariant mass with respect to fig .
[ t2_nonl ] , but with a @xmath168 smaller by almost a factor two .
considerable differences between bs and n / d results are also observed above @xmath177 gev : while the bs curve goes fast to zero , the n / d one remains nearly flat at least up to @xmath178 gev .
the main difference between the two choices has to do with the actual value of the coupling squared @xmath179 , particularly for its imaginary part . in this way , if the bs @xcite @xmath1 pole position in table [ a0proper ] were used with the couplings of the n / d @xcite pole one would obtain also @xmath0 broad peaks similar to those shown by the dashed lines in fig .
[ t2_nonl ] . in ref .
@xcite it was found that the fits to babar @xcite and belle @xcite data in the region of the @xmath3 resonance were stable against variation of the contact term in the @xmath31 kernel .
now there is more sensitivity because the @xmath1 pole positions ( table [ a0proper ] ) are not so close to the @xmath46 threshold as the @xmath37 ones ( table [ f0proper ] ) .
for this reason , the three point function @xmath180 , eq . , is smaller than in the @xmath37 case so that interferences with smaller contributions are more relevant . for the n / d @xcite @xmath1 pole position
the local term amounts at around a 20% of the leading @xmath180 contribution .
however , for the bs @xcite pole the corrections from the local term increase significantly with energy above 2 gev .
one should notice that @xmath181 in eq . is larger by around a factor 4 for the bs pole than for the n / d one .
due to the uncertainties in the pole position and couplings of the @xmath1 resonance as well as the local term in @xmath182 , eq . , we can not arrive to a definite conclusion on the existence of an isovector companion to the @xmath3 in the @xmath0 system .
nevertheless , we can state that if the @xmath1 properties are close to those predicted by the n / d study of ref .
@xcite the present model predicts a resonance behavior of dynamical origin in the @xmath32 scattering around 1.8 - 2 gev .
s - wave amplitude does not depend on the precise value of the subtraction constant @xmath154 as far as it has a natural value @xmath183 . ]
[ bb]@xmath109 [ tc]@xmath169 [ gev ] in fig .
[ k ] we show real and imaginary parts of the interaction potential @xmath109 for fit 2 and both bs and n / d sets . in the region of @xmath184-@xmath185 , where @xmath168 has a peak in the n / d case , the imaginary parts corresponding to bs and n / d are quite similar .
instead , the real part for the n / d choice is positive ( attractive ) in the hole energy range of interest and larger than the bs real part , which even turns negative ( repulsive ) at @xmath186 gev .
this explains the large differences observed in @xmath168 .
one should stress that @xmath109 has an imaginary part due to a number of reasons : the finite @xmath1 width , responsible for the imaginary part of the @xmath1 pole position , the fact that @xmath179 is complex , and also the imaginary part of @xmath187 .
actually , @xmath182 should be interpreted as an optical potential .
width , one has to evaluate @xmath182 at the @xmath1 pole position with positive imaginary part so that @xmath188 ^ 2 + i \epsilon$ ] , in agreement with eq . . instead
, in @xmath149 , @xmath166 should appear with a negative imaginary part to guarantee that , in the zero - width limit , the sign of the imaginary part is the same dictated by the @xmath189 prescription of eq . .
such analytical extrapolations in the masses of external particles are discussed in refs .
@xcite . ]
so far , the @xmath1 pole position has been used as a complex value for the @xmath166 mass .
it is instructive to calculate the amplitude squared taking instead a convolution over the @xmath1 mass distribution determined by its width , so that only real masses appear now in @xmath149 , which has then its cut along the real axis above threshold , as required by two - body unitarity with real masses .
namely , we calculate @xmath190 ^ 2 + { { \rm im}\,}(m_{a_0})^2 } |t_{\phi a_0}(s , m_{a_0},w)|^2 \ , , \label{conv}\ ] ] with @xmath191 defined as @xmath192 and the normalization @xmath193 ^ 2 + { { \rm im}\,}(m_{a_0})^2 } \,.\ ] ] @xmath194 and @xmath195 are the real and ( positive ) imaginary part of the @xmath1 pole position .
the integration interval around the maximum of the distribution , characterized by @xmath196 , should be enough to cover the region where the @xmath1 strength is concentrated . in fig .
[ t2conv ] we compare the results obtained in this way with those obtained from eq . at a fixed complex @xmath166 .
this is done for fit 2 , both bs and n / d parameters and using @xmath197 .
only small differences arise in the hight of the peak so that one can conclude that the two approaches produce the same qualitative features , as one would expect based on physical reasons .
[ tc]@xmath169 [ gev ] the findings described above have direct implications for the @xmath4 reaction with the @xmath110 invariant mass in the @xmath1 mass region .
state with the physical @xmath102 particle , neglecting @xmath200 mixing .
this is also done in refs .
@xcite , from where the meson - meson scattering amplitudes in the @xmath1 channel have been obtained .
new studies indicate that the @xmath1 coupling to @xmath201 is very small @xcite . ]
this process has been investigated in ref .
@xcite where the presence of the @xmath1 is properly taken into account by replacing the lowest order @xmath202 tree level vertex from @xmath203 eq . by the unitarized amplitude of ref .
however , the corrections due to @xmath0 re - scattering ( fsi ) were not included .
here we consider the impact of these fsi on the total @xmath4 cross section using the previously derived @xmath0 amplitude . under the assumption that the @xmath4 reaction is dominated by the @xmath0 channel
, the cross section after fsi can be cast as @xcite @xmath204 we take @xmath205 from ref .
@xcite ( fig . 5 ) , which was obtained by integrating the @xmath206 invariant mass @xmath207 in the @xmath1 region ( 850 - 1100 mev ) so that our assumption of @xmath208 dominance is justified .
the results are shown in fig .
[ fig : cross ] for the different parameter sets .
we find considerable fsi corrections .
in particular , for fit 1 the reduction of the cross section is large , even a factor five at some energies . with the bs choice
, the cross section does not exhibit any structure and is smoother than the one without fsi . instead , for the n
/ d set a peak ( quite prominent for fit 2 ) is observed at @xmath209 gev .
these results clearly show the interest of measuring experimentally the @xmath210 invariant mass distribution so as to confirm the existence of this new isovector @xmath36 resonance that would be observed as a clear peak in data . the existence of this resonance is favored by our results since it appears when the @xmath1 properties from the later and more complete n / d @xcite calculation are adopted .
[ bb]@xmath211 [ nb ] [ tc]@xmath169 [ gev ]
we have studied the @xmath38 s - wave @xmath32 dynamics around threshold paying special attention to the possible dynamical generation of an isovector @xmath36 scalar resonance . following the approach of ref .
@xcite , where the related isoscalar s - wave @xmath31 scattering was investigated , we first considered the scattering of the @xmath27 resonance with a pair of light pseudoscalar mesons at tree level using chiral lagrangians coupled to vector mesons by minimal coupling .
the re - scattering of the two pseudoscalars in @xmath38 and s - wave generates dynamically the @xmath1 .
we have used the information about this state ( pole position and residue in the @xmath212 channel ) from two different studies of meson - meson scattering in coupled channels to determine the @xmath32 scattering potential without introducing new extra free parameters .
afterwards the full amplitude is obtained by resummation of the @xmath32 unitarity loops .
the parameter @xmath213 , characterizing @xmath214 scattering at threshold , and the @xmath208 subtraction constant @xmath154 are obtained from two different fits to @xmath215 babar @xcite and belle @xcite data .
we find that if the physical @xmath216 properties correspond to those extracted with the n / d method in ref .
@xcite ( see table [ a0proper ] ) , the present model predicts a resonance of dynamical origin around 1.8 - 2 gev .
a broader resonance is also generated when the @xmath1 pole position and couplings are taken from the bs study of ref .
@xcite if the strength of the local term in the @xmath32 interaction kernel is reduced .
furthermore , we have determined the @xmath32 final state interactions that strongly modify the cross section for the reaction @xmath199 when the @xmath110 invariant mass is in the @xmath1 region .
if the @xmath1 properties from the n / d method are taken , a strong clearly visible peak around 2.03 gev is observed , signaling the presence of the dynamically generated isovector @xmath42 resonance . for the @xmath1 bs pole of ref .
@xcite no peak is generated but a strong reduction of the cross - section takes place .
the present results further support the idea that a study of the @xmath217 reaction , which should be accessible at present @xmath218 factories @xcite , may provide novel relevant information about hadronic structure and interactions in the 2 gev region .
we thank mauro napsuciale and carlos vaquera - araujo for sending us their results corresponding to the dotted lines in fig . [
fig : cross ] , and eulogio oset for useful discussions .
this work has been partially funded by the mec grant fpa2007 - 6277 and fundacin sneca grant 11871/pi/09 .
we also acknowledge the financial support from the bmbf grant 06bn411 , eu - research infrastructure integrating activity `` study of strongly interacting matter '' ( hadronphysics2 , grant n. 227431 ) under the seventh framework program of eu and the consolider - ingenio 2010 program cpan ( csd2007 - 00042 ) .
s. l. zhu , int .
j. mod .
e * 17 * ( 2008 ) 283 . c. amsler
_ et al . _
[ particle data group ] , phys .
b * 667 * ( 2008 ) 1 .
b. aubert _ et al . _ [ babar collaboration ] , phys .
d * 74 * ( 2006 ) 091103 .
b. aubert _ et al .
_ [ babar collaboration ] , phys .
d * 76 * ( 2007 ) 012008 .
m. ablikim , _ et al . _
[ bes collaboration ] , phys .
* 100 * ( 2008 ) 102003 .
c. p. shen _ et al . _ [ belle collaboration ] , phys .
d * 80 * ( 2009 ) 031101 .
c. p. shen and c. z. yuan , arxiv:0911.1591 [ hep - ex ] .
z. g. wang , nucl .
a * 791 * ( 2007 ) 106 .
chen , x. liu , a. hosaka and s .- l .
zhu , phys .
d * 78 * ( 2008 ) 034012 .
n. v. drenska , r. faccini and a. d. polosa , phys .
b * 669 * ( 2008 ) 160 .
ding and m .- l .
yan , phys .
b * 650 * ( 2007 ) 390 .
n. isgur and j. e. paton , phys .
d * 31 * ( 1985 ) 2910 ; n. isgur , r. kokoski and j. e. paton , phys .
* 54 * ( 1985 ) 869 .
t. barnes , f. e. close and e. s. swanson , phys . rev .
d * 52 * ( 1995 ) 5242 .
ding and m .- l .
yan , phys .
b * 657 * ( 2007 ) 49 .
t. barnes , n. black and p. r. page , phys .
d * 68 * ( 2003 ) 054014 .
s. coito , g. rupp and e. van beveren , phys .
d * 80 * , 094011 ( 2009 ) .
e. van beveren and g. rupp , int
. j. theor .
group theory nonlinear opt .
* 11 * ( 2006 ) 179 .
e. van beveren and g. rupp , ann .
* 324 * ( 2009 ) 1620 .
a. martinez torres , k. p. khemchandani , l. s. geng , m. napsuciale and e. oset , phys .
d * 78 * ( 2008 ) 074031 .
j. a. oller and e. oset , nucl .
phys . a * 620 * ( 1997 ) 438 ; ( e)-ibid .
a * 652 * ( 1999 ) 407 .
l. roca , e. oset and j. singh , phys .
d * 72 * ( 2005 ) 014002 .
l. alvarez - ruso , j. a. oller and j. m. alarcn , phys .
d * 80 * ( 2009 ) 054011 .
c. a. vaquera - araujo and m. napsuciale , phys .
b * 681 * ( 2009 ) 434 .
s. gomez - avila , m. napsuciale and e. oset , phys .
d * 79 * ( 2009 ) 034018 .
j. gasser and h. leutwyler , nucl .
b * 250 * ( 1985 ) 465 .
j. d. weinstein and n. isgur , phys .
* 48 * ( 1982 ) 659 ; phys .
d * 41 * ( 1990 ) 2236 .
g. janssen , b. c. pearce , k. holinde and j. speth , phys .
d * 52 * ( 1995 ) 2690 .
j. a. oller and e. oset , phys .
d * 60 * ( 1999 ) 074023 .
j. a. oller , nucl .
a * 727 * ( 2003 ) 353 .
j. a. oller and e. oset , nucl .
a * 629 * ( 1998 ) 739 .
j. a. oller , phys .
d * 71 * ( 2005 ) 054030 .
j. a. oller and u. g. meissner , phys .
b * 500 * ( 2001 ) 263 ; phys .
d * 64 * ( 2001 ) 014006 ; j. a. oller , phys .
b * 477 * ( 2000 ) 187 .
j. a. oller , e. oset and j. r. pelaez , phys .
d * 59 * ( 1999 ) 074001 ; [ e]-ibid .
d * 60 * ( 1999 ) 099906 ; d * 75 * ( 2007 ) 099903 . m. albaladejo and j. a. oller , phys . rev .
* 101 * ( 2008 ) 252002 .
g. bargon , introduction to dispersion techniques in field theory " , w. a. benjamin , inc , new york , amsterdam , 1965 .
m. albaladejo , j. a. oller and l. roca , forthcoming ; zhi - hui guo , j. a. oller and j. prades , to appear soon .
we thank m. albaladejo and z. h. guo for providing us the coupling of the @xmath1 to @xmath201 before publication . | we have studied the @xmath0 s - wave scattering at threshold energies employing chiral lagrangians coupled to vector mesons by minimal coupling . the interaction is described without new free parameters by considering the scalar isovector @xmath1 resonance as dynamically generated in coupled channels , and demanding that the recently measured @xmath2 cross section is reproduced . for some realistic choices of the parameters ,
the presence of a dynamically generated isovector companion of the @xmath3 is revealed .
we have also investigated the corrections to the @xmath4 reaction cross section that arise from @xmath0 re - scattering in the final state .
they are typically large and modify substantially the cross section . for a suitable choice of parameters , the presence of the resonance would manifest itself as a clear peak at @xmath5 gev in @xmath4 . |
antiferromagnetically interacting spin systems with a low dimensional structure often show a ground state unexpected from a classical picture .
this is due to a quantum spin fluctuation , which destabilizes the n@xmath7el state and tends to bring the system to the non - magnetic state such as the haldane state or the singlet dime state .
the title compounds ( ch@xmath0)@xmath1chnh@xmath0-cucl@xmath0 and ( ch@xmath0)@xmath1chnh@xmath0-cubr@xmath0 abbreviated as ipa - cucl@xmath0 and ipa - cubr@xmath0 are low - dimensional spin - gap systems with an isomorphic crystal structure @xcite . the origin of the spin gap is quite different for these two compounds . in cl - system , which has recently been revealed by an inelastic neutron experiment to be a spin ladder@xcite , the most dominant coupling on the neighboring two spins with @xmath8=@xmath9 on the rung leads a formation of a ferromagnetic dimer .
these fake @xmath8=1 spins interact weakly with antiferromagnetic bonds along a ladder , so that the system is well considered as the haldane system , that is , @xmath8=1 antiferromagnetic heisenberg chain @xcite .
the spin excitation gap in this system is reported to be around 10k @xcite .
while in br - system , the dominant interaction between the two neighboring spins is antiferromagnetic , so that at low temperatures , they form a non magnetic dimer , which interacts weakly with other dimers .
the spin excitation gap is reported to be around 80 - 100k @xcite .
though these two cl and br systems look quite different , one should notice that the underlying physics is the same in that the two have a gapped ground state .
the subject of this study is to investigate what will happen when these two non magnetic systems with quite different spin - spin couplings are mixed . in a classical picture
, one should note that a uniform antiferromagnetic ordering is instable against bond randomness , and that a spin - glass state is favorable .
this is because ferromagnetic bonds embedded randomly in antiferromagnetic bonds randomly reverse the staggered phase factor of the antiferromagnetism , and hence destabilize it . in the quantum mechanical picture , however , the effect of randomness is not self - evident . disorder usually breaks the order , but sometimes it brings order to disorder@xcite . for example
, it is well known that doped holes of only three percent destroy the n@xmath7el order in high-@xmath10 cuprates @xcite .
this is explained in terms of the frustration effect . in spin peierls systems , on the contrary
, an extremely small fraction of non magnetic impurities causes the antiferromagnetic order by destructing singlet dimers @xcite . in the present system , where there is neither a frustration effect nor a destruction of a dimer
, we expect a novel - type impurity effect .
so far , investigations on macroscopic quantities such as uniform magnetization and specific heat on ipa - cu(cl@xmath11br@xmath3)@xmath0 have shown that there exists a phase transition in the limited range of the mixing ratio @xmath4=0.44 - 0.87 @xcite . from a uniaxial anisotropy in
the magnetization observed below a certain temperature , they claim that an antiferromagnetic - like ordering takes place at low temperatures .
the dependence of the transition temperature on the mixing ratio indicates that there exist the two quantum critical points , and that both the two are of the first order . as a theoretical approach
, nakamura has studied the @xmath8=@xmath9 ferromagnetic and antiferromagnetic random alternating heisenberg quantum spin chain by a quantum monte - carlo method to show that a uniform antiferromagnetic state is stable in a certain region of @xmath4 , and that there exists a quantum phase transition at the edge of the region @xcite .
he also claims that the quantum effect is indispensable to the occurrence of the uniform antiferromagnetic ordering . on the other hand ,
fisher and his colleagues had worked on the lattice boson system by introducing a randomness to the onsite chemical potential , and predicted the appearance of the new ground state , bose - glass phase at the absolute zero @xcite .
it has been well known that the field - induced or thermally - excited triplets in spin gap systems including the present compound can be treated as nearly - free bosons @xcite , and that their theory is believed to be applicable to disordered spin gap systems .
in fact , their prediction is supported by recent experiments on a solid solution of other two spin gap systems tlcucl@xmath0 and kcucl@xmath0 , on which a possibility of the existence of the bose - glass phase has been reported@xcite . in this paper , we report by a microscopic probe of @xmath12sr an extensive study on the spin state in the two samples with mixing ratios @xmath4=0.80 and 0.95 , which are in and out of the ordering region @xcite .
we expect that this experiment will be of help for understanding the quantum effect of disorder in spin gap systems .
single crystals of ( ch@xmath0)@xmath1chnh@xmath0cu(cl@xmath2br@xmath3)@xmath0 with @xmath4=0.85 and 0.95 have been grown by an evaporation method from isopropylalcohol solution of ( ch@xmath0)@xmath1chnh@xmath13hx and cux@xmath1 with x = cl , br.@xcite a mixing ratio @xmath4 in crystals is determined by the icp method .
the outer shape of crystals is a rectangular - solid .
following manaka _
, we refer the orthogonal three planes on the outer shape of crystals as a , b , and c - plane @xcite . before the muon experiments
, macroscopic quantities of the single crystals were measured to confirm their quality and homogeneity .
the specific heat of the sample @xmath4=0.85 measured by the thermal relaxation method shows a cusp at @xmath14@xmath1512k , which agrees with manaka s report .
the uniform magnetization also shows a significant anisotropy below this temperature , suggesting an existence of a phase transition .
no spurious curie - term is observed at low temperatures down to 2k in either the two samples .
these observations are quantitatively consistent with manaka s results @xcite .
measurements of @xmath12sr have been carried out at riken - ral muon facility in u. k. using a spin - polarized pulsed surface - muon ( @xmath16 ) with a momentum of 27mev/@xmath17 .
the two single crystals with @xmath4=0.85 and one with @xmath4=0.95 were used for muon measurements .
they are attached by an apiezon n grease on the silver plate with a purity of four nines , connected at the bottom of the cryostat .
a typical size of crystals are 10@xmath182@xmath1820mm@xmath19 .
the incident muon beam is perpendicular to c - plane of the crystals .
the sample temperature was controlled by oxford @xmath19he cryostat in the range of 0.3 - 14k .
the typical temperature stability during measurements is @xmath2010mk above 1.5k , and @xmath202mk below 1.5k .
figure 1 shows the profile of zf ( zero - field ) muon rotation spectra of the sample @xmath4=0.85 .
a clear rotation is observed at low temperatures .
a short time region within 1@xmath12sec of relaxation curves is fit to the function @xmath21 , where @xmath22 , @xmath23 , and @xmath24 are fitting parameters , corresponding to an inhomogeneity of the local field , a mean value of the local field , and a phase constant .
the temperature dependence of @xmath23 is shown at fig .
2 , where a scaling function @xmath25 is fitted to the data to obtain the transition temperature @xmath5=11.65(@xmath200.05)k and the critical exponent @xmath6=0.33(@xmath200.02 ) . the local field of the muon site at 0.33k
is estimated to be @xmath26=@xmath27=360 oe . when the rotating frequency becomes close to the pulse width of the muon beam 70nsec / fwhm ,
the oscillating amplitude is appreciably decreased at low temperatures . assuming the shape of the muon pulse as to be gaussian , we have calculated the convolution and found that the amplitude should decrease to be 65% , which is comparable to the observed value of 50% .
this reduction is due to the effect of a finite pulse width of the muon burst rather than the change in the electronic spin state .
the relaxation rate @xmath22 is around 5 - 7@xmath12sec@xmath28 in the ordered state , and shows a slight increase below 6k .
the distribution of the internal field is @xmath29=@xmath30=130 - 160oe . in fig .
3 , we show relaxation curves under various longitudinal fields ( lf ) up to 10oe . in paramagnetic state at 14k , the asymmetry change in the entire zf curve is approximately 8% .
the early part of curves within a few micro seconds shows the gaussian decay , indicating that the muon relaxation is caused by nuclear spins , which create a quasi static field around muon sites .
this is confirmed by the observation that the weak lf of 100oe completely suppresses the relaxation . when the temperature is lowered to 11.8k , which is higher than @xmath5 only by 0.15k , no sign of critical slowing down is observed , that is , the relaxation curve is identical to those in the paramagnetic state as seen in fig .
1 . below @xmath5 , as shown in the lower panel of fig .
3 , the initial asymmetry is drastically reduced by the large internal field due to the magnetic ordering . a very slow component of relaxation persists at the lowest temperature 0.33k .
this relaxation is considered to be caused by a quasi static nuclear spin fluctuation , because it is easily suppressed by a weak lf of 10oe .
the zf - asymmetry change due to this nuclear spin component is 4% , which is almost a half of that in paramagnetic state at 14k . , where @xmath22 , @xmath23 , @xmath24 are fitting constants . ]
next , we proceed to the results on the sample with @xmath4=0.95 , which shows no anomalies both in the specific heat and the macroscopic susceptibility at low temperatures down to 2k .
figure 4 shows typical muon relaxation curves under lf up to 3900oe . in 10k
, entire muon relaxation is driven by the quasi static nuclear spin fluctuation , because an initial part of the relaxation decays with gaussian - like function , and hence the relaxation is completely suppressed by a weak lf of a few tens oe . in 0.33k ,
the relaxation is divided into the two components .
one is the quasi static nuclear spin part , which has the gaussian form , and is suppressed by a weak lf of 20oe .
the other is a dynamically - fluctuating electronic part , which has the lorentzian form , and survives relatively high a field up to 3900oe .
the relaxation rate @xmath22 was obtained by fitting the exponential function to the relaxation curves under lf above 20oe .
the fraction of the latter components is around one half of the entire asymmetry , indicating that , roughly speaking , a half of the muons are not affected by any electron spins .
( open square ) and relaxation rate @xmath22 ( solid square ) .
the curve is the scaling function @xmath31 , where @xmath5 and @xmath6 is the n@xmath7el temperature and the critical exponent .
the two parameters are obtained by the least square method to be @xmath5=11.65k and @xmath6=0.33(@xmath200.02 ) ( _ inset _ ) , which is close the value for 3d - ising model . a fitting line with the exponent of 3d - heisenberg model @xmath6=0.367 is also shown for comparison . ] the external field ( lf ) dependence of @xmath22 shown at fig .
5 directly maps the fourier component of the fluctuating field produced by 3@xmath32-spins .
assuming that the spin fluctuation of 3@xmath32-spins has a classical lorentzian spectrum , we fitted the function @xmath33 to the observed data to obtain the mean fluctuation time and the fluctuation amplitude to be @xmath34=0.44@xmath12sec and @xmath29=55 oe . the temperature dependence of @xmath22 under lf of 100oe , which is high enough for a complete suppression of the effect of nuclear spins , is shown in fig .
a significant increase that looks to be diverging at the zero temperature is observed below 5k .
here we mention that in both the two samples with and without a static order , approximately half of the muons are relaxed by the local field produced by 3@xmath32-spins .
this means that the other half only feel the local field produced by nuclear spins , fluctuation of which is quasi static , or by the 3@xmath32-spins fluctuating with frequency out of the muon time window .
we will discuss this point in the following section .
= 0.85 at paramagnetic state ( upper panel , 14k ) and ordered state ( lower panel , 0.33k ) under various longitudinal fields ( lf ) up to 100oe . ]
the muon spin rotation observed at low temperatures @xmath35=11.65k indicates the existence of the long - range and static magnetic order in the sample of @xmath4=0.85 . judging from the single fourier component observed in the rotational profile ,
the spin structure is considered to be simple antiferromagnetic .
this result microscopically demonstrates that the solid solution of the two spin gap systems has a magnetic ground state .
the critical exponent of the order parameter @xmath6=0.33 agrees with the theoretical value of 3d - ising model @xcite .
the appearance of the three dimensional exponent is explained as that the phase transition in this low dimensional spin system is set off by weak inter - ladder interactions , the path of which runs three - dimensionally in the system .
the ising - type universality class is consistent with the existence of the effective uniaxial anisotropy @xmath36 due to the formation of ferromagnetic dimers in ipa - cucl@xmath0 , reported recently by manaka@xcite .
next , we discuss the volume fraction of the magnetic order . as stated above , half of the muons are not affected by the local field produced by 3@xmath32-spins in both the two samples .
this observation indicates the two possibilities .
one is that there are the two muon stopping sites .
the first site is , as in many other compounds , located near the negative ions of cl@xmath37 or br@xmath37 , which are close to cu-3@xmath32 spins , and bears a large local field . actually , if we assume that the ordered moment is around 1 @xmath38@xcite , and that its interaction with muon is of a classical dipole - dipole type , the distance between the muon site and the cu atom is calculated from the observed local field of 360 oe to be 0.3 - 0.37 nm , which is comparable with that between cl and cu atoms 0.22 - 0.23 nm , supporting our idea .
the second site may be located , for example , around or inside the anion molecule ( ch@xmath0)@xmath1chnh@xmath0 , which is far from 3@xmath32-spins , and its local field should be very small . by considering this second site , we can understand why the half of muons are not affected by the magnetic field produced by 3@xmath32-spins even though the volume fraction of the magnetic order is nearly unity .
the other possibility is that half of the muons only experience electronic moments fluctuating outside the muon time window .
we emphasize here that the possibility of the phase separation in samples is excluded , because the sample quality is assured both by the macroscopic susceptibility containing no spurious curie - term , and by the sharpness in the phase transition observed by the specific heat and by the present muon experiment ( fig .
2 ) . furthermore , if half the volume of the sample @xmath4=0.85 is contaminated by other concentration which do not shows magnetic order ( @xmath390.87 ) , the half of the contaminating region is expected to show an appreciable relaxation due to the dynamical spin fluctuation as is observed in the sample @xmath4=0.95 .
this contradicts to the observed fact .
the profile of the relaxation curves in the paramagnetic state does not change from one at high temperatures until the temperature reaches at @xmath5 .
this means that the temperature region where the critical slowing down of 3@xmath32-spins , if exists , starts and its characteristic frequency gets into the muon time window ( @xmath40sec@xmath28 ) must be extremely narrow , that is , @xmath410.1k . on the other hand ,
the sample of @xmath4=0.95 shows a diverging behavior of @xmath22 toward the absolute zero , indicating that a critical slowing down of the spin fluctuation takes place in rather a wide range of temperature , though there is no long range order at low temperature down to 0.33k .
this sample of @xmath4=0.95 has been indicated to be gapped by the results of macroscopic measurements of manaka s group @xcite and also of the present study .
nevertheless , the present muon observation clearly demonstrates that the system has the magnetic instability in the ground state .
we can immediately identify the dominant q - part of this instability to be antiferromagnetic like , because the uniform susceptibility vanishes at low temperatures . in regard to the origin of the magnetic instability , we propose the two possibilities .
one is the precursor effect to the bose - glass phase at the absolute zero @xcite . in solid - solution of the two spin - gap systems ( tl , k)cucl@xmath0 , an anomalous enhancement of @xmath22 as well as nmr-@xmath42
have been reported and the possibility of the bose - glass phase is suggested.@xcite the other possible origin is the fluctuation effect around the quantum critical point at @xmath4=0.87 , which divides the area into the antiferromagnetic and paramagnetic regions . in the samples
@xmath4=0.95 , though the static order does not occur , the antiferromagnetic component of @xmath43 may be highly enhanced .
this is analogous to the case of high-@xmath10 cuprates of the superconducting phase , where a pronounced antiferromagnetic fluctuation in 3@xmath32-spins persists , even though there is no magnetic order at finite temperatures .
@xcite finally , we do not deny a possibility of the long - range order in this sample at a finite temperature much lower than the present study .
the experiment in the temperature region of dilution - refrigerator is now on the progress .
by an intensive @xmath12sr study on the solid solution of two spin - gap systems ( ch@xmath0)@xmath1chnh@xmath0-cu(cl@xmath2br@xmath3 ) ( @xmath4=0.85 , 0.95 ) , we have confirmed that the sample @xmath4=0.85 shows a long range magnetic order at @xmath5=11.65k .
the local field at muon site is 360oe at the lowest temperature .
the critical exponent estimated from the temperature dependence of the local field is @xmath6=0.33 , which agrees with the theoretical value of 3d - ising model . in the sample of @xmath4=0.95 , which does not show any long range order , a significant critical slowing down of the electron spin fluctuation toward the absolute zero was observed . the characteristic correlation time and fluctuation amplitude of the 3@xmath32-spin fluctuation at 0.33k
was obtained as 0.44@xmath12sec and 160oe . in the sample @xmath4=0.95 under longitudinal field ( lf ) of 100oe .
the inset shows the profile of spectra at each temperatures .
each spectrum other than that for 0.33k is shifted vertically by 1% .
curves are fitted exponential functions from which @xmath22 was obtained . ]
the authors would like to show special thanks to h. manaka for kind and important discussions on the experimental results and the sample syntheses .
they also thank t. masuda and a. zheludev for a valuable discussion , and k. noda for a kind assistance in measurements of the specific heat .
this work has been supported by the toray science foundation , saneyoshi scholarship foundation , kurata memorial hitachi science and technology foundation , the joint programs and grant - in - aid for the jsps , and the grant - in - aid for scientific research on priority area from ministry of education , culture , sports , science , and technology of japan .
the br - system , which is isomorphic with cl - system , had been reported to be an alternating chain from the temperature dependence of the macroscopic susceptibility.@xcite .
no determinate reports to judge whether or not it is have been made until now . according to a recent report by t. masuda _ et al .
_ , @xmath44-axis is perpendicular to c - plane and @xmath45-axis along the ladder is nearly perpendicular to b - plane . t. masuda , a. zheludev , h. manaka , _ et al . private communication . _
our preliminary results of @xmath46h - nmr on ipa-(cu@xmath2br@xmath3)@xmath0 suggests that the ordered moment is around 1 @xmath38 .
k. kanada , t. saito , a. oosawa and t. goto , _ presented in qubs2006 : an icm2006 satellite conference ( tokai , 2006 ) . _ r. j. birgeneau , d. r. gabbe , h. p. jenssen , m. a. kastner , p. j. picone , t. r. thurston , g. shirane , y. endoh , m. sato , k. yamada , y. hidaka , m. oda , y. enomoto , m. suzuki and t. murakami , phys . rev . *
38 * , 6614 ( 1988 ) . | muon - spin - rotation and relaxation studies have been performed on ( ch@xmath0)@xmath1chnh@xmath0cu(cl@xmath2br@xmath3)@xmath0 with @xmath4=0.85 and 0.95 , which are solid solutions of the two isomorphic spin - gap systems ( ch@xmath0)@xmath1chnh@xmath0cucl@xmath0 and ( ch@xmath0)@xmath1chnh@xmath0cubr@xmath0 with different spin gaps .
the sample with @xmath4=0.85 showed a clear muon spin rotation under zero - field below @xmath5=11.65k , indicating the existence of a long - range antiferromagnetic order .
a critical exponent of the hyperfine field was obtained to be @xmath6=0.33 , which agrees with 3d - ising model . in the other sample with @xmath4=0.95 , an anomalous enhancement of the muon spin relaxation
was observed at very low temperatures indicating a critical slowing down due to a magnetic instability of the ground state . |
a new generation of experiments have recently been completed , or are presently under way or in their planning phase to study the effects of parity - violating ( pv ) interactions in @xmath1@xmath1 elastic scattering @xcite , @xmath0@xmath1 radiative capture @xcite and deuteron electro - disintegration @xcite at low energies .
there is also considerable interest in determining the extent to which hadronic weak interactions can affect the longitudinal asymmetry measured by the sample collaboration in quasi - elastic scattering of polarized electrons off the deuteron @xcite , and therefore influence the extraction from these data ( and those on the proton @xcite ) of the nucleon s strange magnetic and axial - vector form factors at four - momentum transfers squared of @xmath13 and @xmath14 ( gev / c)@xmath15 .
the present is the third in a series of papers dealing with the theoretical investigation of pv interaction effects in two - nucleon systems .
the first @xcite was devoted to @xmath16@xmath1 elastic scattering , and presented a calculation of the longitudinal asymmetry induced by pv interactions in the lab - energy range 0350 mev .
the second @xcite provided a rather cursory account of a study of the pv asymmetries in @xmath12@xmath1 radiative capture at thermal neutron energies and in deuteron electro - disintegration at quasi - elastic kinematics .
this work further extends that of ref .
@xcite by investigating the neutron spin rotation at zero energy and the longitudinal asymmetry in @xmath12@xmath1 elastic scattering at lab energies between 0 and 350 mev , and the photon - helicity dependence of the @xmath3 cross - section from threshold up to energies of 20 mev .
it also provides a thorough analysis of the results already presented in ref .
@xcite .
we adopt the pv potential developed by desplanques _
et al . _
@xcite over twenty years ago , the so - called ddh model . in the @xmath0@xmath1 sector
, it is conveniently parameterized in terms of @xmath6- , @xmath7- , and @xmath8-meson exchanges . in ref .
@xcite the pion and vector - meson weak coupling constants were estimated within a quark model approach incorporating symmetry techniques like su(6)@xmath17 and current algebra requirements . due to limitations inherent to such an analysis , however , the coupling constants so determined had rather wide ranges of allowed values .
our prime motivations are to develop a systematic and consistent framework for studying pv observables in the few - nucleon systems , where accurate microscopic calculations are feasible , and to use available and forthcoming experimental data on these observables to constrain the strengths of the short- and long - range parts of the two - nucleon weak interaction . indeed , in ref .
@xcite we showed , for the case of the longitudinal asymmetry measured in @xmath16@xmath1 elastic scattering , how available experimental data provide strong constraints on allowable combinations of @xmath7- and @xmath8-meson weak coupling constants .
the remainder of the present paper is organized as follows . in sec .
[ sec : pot ] the pv potential as well as the parity - conserving strong - interaction potentials used in this work are briefly discussed , while in sec .
[ sec : cnt ] the model for the nuclear electro - weak currents is described , including the electromagnetic two - body terms induced by the presence of pv interactions . in sec .
[ sec : app ] a self - consistent treatment of the @xmath0@xmath1 bound- and scattering - state problems in both configuration and momentum spaces is provided , patterned after that of ref .
@xcite , and in sec .
[ sec : obs ] explicit expressions are derived for the longitudinal asymmetry and spin rotation in @xmath12@xmath1 elastic scattering , the photon asymmetry in @xmath12@xmath1 radiative capture , and the asymmetries in deuteron photo- and electro - disintegration . in sec .
[ sec : cal ] the techniques used to calculate the pv observables are briefly reviewed , while in sec . [ sec : res ] a fairly detailed analysis of the results is offered . finally , sec .
[ sec : cons ] contains some concluding remarks .
the parity - conserving ( pc ) , strong - interaction potentials used in the present work are the argonne @xmath5 ( av18 ) @xcite , nijmegen i ( nijm - i ) @xcite , and cd - bonn ( bonn ) @xcite models .
they were discussed in ref .
@xcite in connection with the calculation of the longitudinal asymmetry in @xmath18@xmath1 elastic scattering . here
, we briefly summarize a few salient points .
the av18 and nijm - i potentials were fitted to the nijmegen database of 1992 @xcite , consisting of 1787 @xmath1@xmath1 and 2514 @xmath0@xmath1 scattering data , and both produced @xmath19 per datum close to one .
the latest version of the charge - dependent bonn potential , however , has been fit to the 1999 database , consisting of 2932 @xmath1@xmath1 and 3058 @xmath0@xmath1 data , for which it gives @xmath19 per datum of 1.01 and 1.02 , respectively @xcite . the substantial increase in the number of @xmath1@xmath1 data
is due to the development of novel experimental techniques
internally polarized gas targets and stored , cooled beams . indeed ,
using this technology , iucf has produced a large number of @xmath1@xmath1 spin - correlation parameters of very high precision , see for example ref .
it is worth noting that the av18 potential , as an example , fits the post-1992 and both pre- and post-1992 @xmath1@xmath1 ( @xmath0@xmath1 ) data with @xmath19 s of 1.74 ( 1.02 ) and 1.35 ( 1.07 ) , respectively @xcite .
therefore , while the quality of their fits ( to the @xmath1@xmath1 data ) has deteriorated somewhat in regard to the extended 1999-database , the av18 and nijm - i models can still be considered realistic . as already mentioned in sec .
[ sec : intro ] , the form of the parity - violating ( pv ) weak - interaction potential was derived in ref .
@xcite the ddh model . in the isospin space of the @xmath0@xmath1 pair it is expressed as @xmath20 where @xmath21=0 or 1 .
the diagonal and off - diagonal terms are then obtained as @xmath22_{+ } \bigg ] \
, \label{eq : ddh1}\end{aligned}\ ] ] @xmath23_{+ } \ , \label{eq : ddh2}\end{aligned}\ ] ] and @xmath24=@xmath25 . in the equations above the relative position and momentum
are defined as @xmath26 and @xmath27 , respectively , @xmath28_{+}$ ] denotes the anticommutator , and @xmath29 , @xmath30 , @xmath31 , and @xmath32 are the proton , pion , @xmath7-meson , @xmath8-meson masses , respectively .
the yukawa function @xmath33 , suitably modified by the inclusion of monopole form factors , is given by @xmath34 \bigg\ } \>\>,\ ] ] where @xmath35 .
note that @xmath36 denotes @xmath37 , and that the terms proportional to @xmath36 in eqs .
( [ eq : ddh1 ] ) and ( [ eq : ddh2 ] ) are usually written in the form of a commutator , since @xmath38_{- } = m_\alpha y^\prime ( m_\alpha r)\ , \hat{\bf r } \ .\ ] ] finally , the values for the strong - interaction @xmath6-meson pseudoscalar coupling constant @xmath39 , and @xmath7- and @xmath8-meson vector and tensor coupling constants @xmath40 and @xmath41 , as well as for the cutoff parameters @xmath42 , are taken from the bonn model @xcite , and are listed in table [ tb : ddh ] .
the weak - interaction vector - meson coupling constants @xmath43 and @xmath44 correspond to the following combinations of ddh parameters @xmath45 where @xmath46=0,1 .
the values for these and for @xmath47 , @xmath48 , and @xmath49 are listed in table [ tb : ddh ] .
note that we have taken the linear combination of @xmath7- and @xmath8-meson weak coupling constants corresponding to @xmath18@xmath1 elastic scattering from the earlier analysis @xcite of these experiments .
the remaining couplings are the `` best value '' estimates , suggested in ref .
@xcite . in order to study the sensitivity of the calculated pv observables to variations in the weak coupling constants , we also consider in the present work a @xmath50 corresponding to the original `` best value '' estimates for these from ref .
@xcite , see table [ tb : ddhb ] .
the electromagnetic current and charge operators , respectively @xmath51 and @xmath7 , are expanded into a sum of one- and two - body terms @xmath52 and similarly for @xmath53 .
the one - body terms have the standard expressions obtained from a non - relativistic reduction of the covariant single - nucleon current @xcite .
the two - body charge operators are those derived in ref .
@xcite ; they only enter in the calculation of the asymmetry in the deuteron electro - disintegration at quasi - elastic kinematics , and will not be discussed further here .
the two - body currents have terms associated with the parity - conserving ( pc ) and parity - violating ( pv ) components of the interaction , respectively @xmath54 and @xmath55 .
the operators @xmath56 were derived explicitly in ref .
@xcite , and a complete listing of those relative to the argonne @xmath5 ( av18 ) interaction @xcite has been given most recently in ref .
only the two - body currents associated with @xmath6- and @xmath7-exchange are retained in the case of the bonn ( bonn ) @xcite and nijmegen - i ( nijm - i ) @xcite interactions .
in addition to these , the purely transverse two - body currents associated with the excitation of @xmath57-isobars and the @xmath58 and @xmath59 mechanisms are included in all calculations .
again explicit expressions for these operators can be found in ref .
note , however , that the @xmath57-isobar degrees of freedom are treated in perturbation theory rather than with the transition - correlation - operator method @xcite , and that only effects due to single @xmath57 excitation are considered , according to eqs .
( 2.15 ) and ( 3.4 ) in ref .
@xcite . before moving on to a discussion of the pv currents , we briefly review , for later reference , the question of conservation of the electromagnetic current for the case of the av18 . as pointed out in ref .
@xcite , the currents from its @xmath60 part ( specifically , its isospin - dependent central , spin - spin , and tensor components ) are strictly conserved . in a one - boson - exchange model , which the av18 is not , these interaction components arise from @xmath6- and @xmath7-exchange . the currents from the av18 momentum - dependent ( @xmath1-dependent ) components the spin - orbit , @xmath61 , and quadratic - spin - orbit terms are also included .
@xcite and later papers , the currents from the spin - orbit term were derived by generalizing the procedure used to obtain the @xmath60 currents .
it was assumed that the isospin - independent ( isospin - dependent ) central and spin - orbit interactions were due to @xmath62 and @xmath8 exchanges ( @xmath7 exchange ) , and the associated two - body currents were constructed by considering corresponding @xmath63@xmath64-pair diagrams involving these meson exchanges .
the currents from the @xmath61 and quadratic - spin - orbit interactions were obtained , instead , by minimal substitution @xcite .
the currents from the @xmath1-dependent interactions are strictly not conserved , as one can easily surmise by considering their commutator with the charge density operator .
for example , in the case of the isospin - dependent @xmath61 and @xmath65 interactions , this commutator requires the presence of currents with the isospin structure @xmath66 , which can not be generated by minimal substitution @xcite .
we will return to this issue in sec .
[ sec : r_np ] . here , we only want to emphasize that the currents from the @xmath1-dependent terms in the av18 are short - ranged .
their contributions to isovector observables , such as , for example , the magnetic form factors of the trinucleons @xcite , are found to be numerically much smaller than those due to the leading @xmath60 currents .
these currents also lead to small , although non - negligible , corrections to isoscalar observables , such as the deuteron magnetic moment and @xmath67 structure function @xcite .
however , in the case of the pv asymmetry in the @xmath2@xmath1 radiative capture at thermal energies under consideration in the present study , they will turn out to play an important role ( see sec .
[ sec : r_np ] ) .
the ddh pv interaction @xcite is parameterized in terms of @xmath6- , @xmath7- , and @xmath8-meson exchanges .
the meson - nucleon phenomenological lagrangian densities have been given most recently in ref .
we adopt here the notation and conventions of that work , except that we use pseudo - vector coupling for the @xmath6@xmath64@xmath64 interaction lagrangian , i.e. @xmath68 with @xmath69=@xmath70 .
the resulting @xmath10@xmath6@xmath64@xmath64 coupling is given by @xmath71 and the @xmath10@xmath6@xmath64@xmath64 current is then obtained from the feynman amplitude in fig . [
fig : feyn ] , panel ( a ) .
the complete pv @xmath6-exchange current is derived from a non - relativistic reduction of both amplitudes in fig .
[ fig : feyn ] , and to leading order reads : @xmath72 \left ( \bbox{\sigma}_i \cdot { \bf k}_i - \bbox{\sigma}_j \cdot { \bf k}_j \right ) \bigg ] \ , \end{aligned}\ ] ] where @xmath73 is defined as @xmath74 @xmath75=@xmath76 is the fractional momentum delivered to nucleon @xmath77 ( with these definitions @xmath78=@xmath79 ) , and @xmath80 is a short - range cutoff . in the limit of point - like couplings ,
the current above is identical to that listed in eqs .
( a7a ) and ( a9a ) of ref .
@xmath64@xmath63-pair terms arising from the photon coupling to the nucleon line containing a pv @xmath6@xmath64@xmath64 vertex do not contribute to leading order .
lastly , we note that @xmath81 satisfies current conservation with the pv @xmath6-exchange interaction , given in momentum space by @xmath82 since @xmath83&+ & \left [ v^{\rm pc}_{\pi}({\bf k}_i,{\bf k}_j-{\bf q } ) \ , , \ , p_j \right ] \nonumber \\ & = & -\frac{f_\pi\ , h_\pi}{\sqrt{2}\ , m_\pi } \left ( \bbox{\tau}_i \cdot \bbox{\tau}_j - \tau_{z , i } \tau_{z , j } \right ) \left [ v_\pi(k_i ) \left ( \bbox{\sigma}_i+\bbox{\sigma}_j \right ) \cdot { \bf k}_i + i \rightleftharpoons j \right ] \ , \end{aligned}\ ] ] which is easily seen to be the same as @xmath84 . here
@xmath85 denotes the isospin projection operator @xmath86 in the present work , the pv currents induced by @xmath7- and @xmath8-meson exchanges have been neglected , since , due their short - range character , the associated contributions are expected to be tiny .
we note , however , that in eq .
( a5 ) of ref .
@xcite a @xmath10@xmath7@xmath64@xmath64 contact term , originating from gauging the @xmath7@xmath64@xmath64 tensor coupling , has been ignored .
it is given by @xmath87 and leads , in leading order , to an additional term in eq .
( a7c ) of ref .
@xcite of the form @xmath88 + i \rightleftharpoons j \ , \ ] ] with @xmath89 defined similarly as in eq .
( [ eq : vpl ] ) . the term above , when combined with that having the same structure in the second line of eq .
( a7c ) , generates a contribution proportional to @xmath90 which , in view of the large value of the @xmath7@xmath64@xmath64 tensor coupling constant ( @xmath91=6.6 ) , is expected to be dominant in the pv @xmath7-exchange current . finally , there is a pv one - body current originating from the nucleon s anapole moment .
it can be derived , for example , by considering pion - loop diagrams where one of the vertices involves a pv @xmath6@xmath64@xmath64 coupling ; it has the structure , to leading order , given by @xmath92 \bbox{\sigma}_i { \rm e}^ { { \rm i } { \bf q } \cdot { \bf r}_i } \ , \ ] ] where @xmath93 is the four - momentum transfer , @xmath93=@xmath94 , and the isoscalar and isovector anapole form factors are normalized as @xmath95 with @xmath96=1.6 and @xmath97=0.4 from a calculation of pion - loop contributions @xcite .
more recent estimates of the nucleon anapole form factors predict @xcite somewhat different values for @xmath98 .
a complete treatment would require estimates of short - distance contributions @xcite and electroweak radiative corrections .
thus , in view of the uncertainties in the quantitative estimate of these effects , we will continue to use the values above in the present study ( see also sec . [
sec : r_dee ] ) . note that @xmath99 vanishes for real - photon transitions .
in the standard model the vector part of the neutral weak current , @xmath100 , is related to the isoscalar ( @xmath101 ) and isovector ( @xmath102 ) components of the electromagnetic current , denoted respectively as @xmath103 and @xmath104 , via @xmath105 where @xmath106 is the weinberg angle , and therefore the associated one- and two - body weak charge and current operators are easily obtained from those given in sec [ sec : cnt ] . the axial charge and
current operators too have one- and two - body terms .
only the axial current @xmath107 is needed in the present work .
the one- and two - body operators are essentially those listed in ref .
@xcite , except for obvious changes in the isospin structure having to do with the fact that we are dealing here with neutral rather than charge - raising / lowering weak currents , and for the inclusion of nucleon and @xmath108 axial form factors the parameterization adopted for these is given in ref .
note that in ref .
@xcite the relativistic corrections in @xmath109 and two - body axial currents were neglected , in line with the expectation , confirmed in the present study , that the associated contributions were small .
finally , the neutral weak currents given above are at tree - level , electro - weak radiative corrections as well as strange - quark contributions to the vector and axial - vector currents @xcite have been ignored .
these effects have been taken into account in recent calculations of the longitudinal asymmetry in @xmath110 at quasi - elastic kinematics @xcite , however , they will not be discussed further here .
in this section we discuss the @xmath0@xmath1 scattering- and bound - state problems in the presence of a potential @xmath111 given by @xmath112 where @xmath113 and @xmath50 denote the parity - conserving ( pc ) and parity - violating ( pv ) components induced by the strong ( including electromagnetic ) and weak interactions , respectively .
the formalism and notation are similar to those developed in ref .
@xcite .
the lippmann - schwinger equation for the @xmath64@xmath64 scattering state @xmath114 , where @xmath115 is the relative momentum , @xmath101 , @xmath116 , @xmath46 , and @xmath117=0 specify the pair spin , spin - projection , isospin , and isospin - projection states ( note that the label @xmath117=0 is unnecessary , since @xmath111 is diagonal in @xmath117 ) , can be written as @xcite @xmath118 where @xmath119 is the free hamiltonian , and @xmath120 are the eigenstates of @xmath119 , namely plane waves , @xmath121 \chi^s_{m_s } \eta^t_{m_t=0 } \nonumber \\ & = & 4\pi\sqrt{2 } \sum_{jm_jl } { \rm i}^l\ , \epsilon_{lst } \,\ , j_l(pr ) [ z_{lsm_s}^{jm_j}(\hat{\bf p})]^ * \ , { \cal y}_{lsj}^{m_j}(\hat{\bf r})\eta^t_0 \>\>. \label{eq : pwe}\end{aligned}\ ] ] here @xmath122 denotes the regular spherical bessel function , and the following definitions have been introduced : @xmath123 @xmath124 \>\>.\ ] ] the factor @xmath125 ensures that the plane waves are properly antisymmetrized .
the @xmath46-matrix corresponding to the potential @xmath111 is defined as @xcite @xmath126 insertion of the plane - wave states @xmath127 into the right - hand - side of the lippmann - schwinger equation leads to @xmath128 from which the partial wave expansion of the scattering state is easily obtained by first noting that the potential , and hence the @xmath46-matrix , can be expanded as @xmath129^*\ , v^j_{l^\prime s^\prime t^\prime , lst } ( p^\prime;p ) \>\ > , \label{eq : tmae}\end{aligned}\ ] ] with @xmath130 after insertion of the corresponding expansion for the @xmath46-matrix into eq .
( [ eq : lsa ] ) and a number of standard manipulations , the scattering - state wave function can be written as @xmath131^ * \nonumber \\ & & \frac{w^j_{l^\prime s^\prime t^\prime , lst}(r;p)}{r } { \cal y}_{l^\prime s^\prime j}^{m_j}(\hat{\bf r})\ , \eta^{t^\prime}_0 \>\ > , \label{eq : psipw}\end{aligned}\ ] ] with @xmath132 where the label @xmath133 ( @xmath134 ) stands for the set of quantum numbers @xmath135 ( @xmath136 ) .
the ( complex ) radial wave function @xmath137 behaves in the asymptotic region @xmath138 as @xmath139 \>\ > , \label{eq : asy}\ ] ] where the on - shell ( @xmath140 ) @xmath101-matrix has been introduced , @xmath141 and the functions @xmath142 are defined in terms of the regular and irregular ( @xmath143 ) spherical bessel functions as @xmath144 the coupled - channel schrdinger equations for the radial wave functions @xmath137 read : @xmath145 w^j_{\alpha^\prime,\alpha}(r;p ) + \sum_{\beta } r \ , v^j_{\alpha^\prime,\beta}(r)\ , \frac{1}{r } w^j_{\beta,\alpha}(r;p ) = 0 \>\ > , \label{eq : cschr}\ ] ] with @xmath146 where , because of time reversal invariance , the matrix @xmath147 can be shown to be real and symmetric ( this is the reason for the somewhat unconventional phase factor in eq .
( [ eq : vpe ] ) ; in order to maintain symmetry for both the @xmath113- and @xmath50-matrices , and hence the @xmath101-matrix , the states used here differ by a factor @xmath148 from those usually used in nucleon - nucleon scattering analyses ) .
the asymptotic behavior of the @xmath137 s is given in eq .
( [ eq : asy ] ) , while explicit expressions for the radial functions @xmath149 can be found in ref .
@xcite those associated with @xmath113 are well known .
there are two coupled channels for @xmath150=0 , and four coupled channels for @xmath151 .
the situation is summarized in table [ tb : chan ] . again because of the invariance under time - inversion transformations of @xmath152 , the @xmath101-matrix is symmetric ( apart from also being unitary ) , and can therefore be written as @xcite @xmath153 where @xmath154 is a real orthogonal matrix , and @xmath155 is a diagonal matrix of the form @xmath156 here @xmath157 is the ( real ) phase - shift in channel @xmath133 , which is function of the energy @xmath158 with @xmath159 . the mixing matrix @xmath154 can be written as @xmath160 where @xmath161 is the @xmath162 or @xmath163 orthogonal matrix , that includes the coupling between channels @xmath77 and @xmath164 only , for example @xmath165 \simeq 1 + \epsilon^j_{13 } \left [ \begin{array}{cccc } 0 & 0 & 1 & 0 \\ 0 & 0 & 0 & 0 \\ -1 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 \end{array } \right ] \>\>. $ ] 0.7 cm note that no coupling is allowed between channels 3 and 4 in the notation of table [ tb : chan ] , and hence @xmath166 .
thus , for @xmath150=0 there are two phase - shifts and a mixing angle , while for @xmath151 there are four phase - shifts and five mixing angles . of course ,
since @xmath167 , the mixing angles @xmath168 induced by @xmath50 are @xmath169 , a fact already exploited in the last expression above for @xmath154 . given the channel ordering in table [ tb : chan ] ,
table [ tb : mixing ] specifies which of the channel mixings are induced by @xmath113 and which by @xmath50 .
the reality of the potential matrix elements @xmath170 makes it possible to construct real solutions of the schrdinger equation ( [ eq : cschr ] ) .
the problem is reduced to determining the relation between these solutions and the complex @xmath137 s functions . using eq .
( [ eq : sdiag ] ) and @xmath171 , the @xmath137 s can be expressed in the asymptotic region as @xmath172 } { pr } \ , u_{\beta \alpha } \>\>.\end{aligned}\ ] ] the expression above is real apart from the @xmath173 . to eliminate this factor ,
the following linear combinations of the @xmath137 s are introduced @xmath174 \>\ > , \label{eq : uasy}\end{aligned}\ ] ] and the @xmath175 s are then the sought real solutions of eq .
( [ eq : cschr ] ) . the asymptotic behavior of the @xmath175 s can now be read off from eq .
( [ eq : uasy ] ) once the @xmath154-matrices above have been constructed .
the latter can be written , up to linear terms in the smallmixing angles induced by @xmath50 , as @xmath176 \quad j=0 \>\ > , $ ] 0.7 cm @xmath177 \quad j \geq 1 \>\>. $ ] 0.7 cm inverting the first line of eq .
( [ eq : uasy ] ) , @xmath178 and inserting the resulting expressions into eq . ( [ eq : cschr ] ) leads to the schrdinger equations satisfied by the ( real ) functions @xmath175 .
they are identical to those of eq .
( [ eq : cschr ] ) , but for the @xmath137 s being replaced by the @xmath175 s .
these equations are then solved by standard numerical techniques .
note that : i ) @xmath179 , since the diagonal matrix elements of @xmath50 vanish because of parity selection rules ; ii ) terms of the type @xmath180 involving the product of a pv potential matrix element with a @xmath50-induced wave function are neglected .
finally , the physical amplitude for @xmath0@xmath1 elastic scattering from an initial state with spin projections @xmath181 , @xmath182 to a final state with spin projections @xmath183 , @xmath184 is given by @xmath185 where the amplitude @xmath186 is related to the @xmath46-matrix defined in eq .
( [ eq : tmas ] ) via @xmath187 and the factor @xmath188 comes from the clebsch - gordan coefficients combining the neutron and proton states to total initial ( final ) isospin @xmath46 ( @xmath189 ) .
note that the direction of the initial momentum @xmath115 has been taken to define the spin quantization axis ( the @xmath190-axis ) , @xmath191 is the angle between @xmath192 and @xmath193 , the direction of the final momentum , and the energy @xmath194 ( @xmath195 ) . using the expansion of the @xmath46-matrix , eq .
( [ eq : tmae ] ) with @xmath196 replaced by @xmath197 , and the relation between the @xmath101- and @xmath46-matrices , eq . ( [ eq : sma ] ) , the amplitude induced by @xmath152 can be expressed as @xmath198 where again @xmath199 . in order to consider the pc momentum - space bonn ( bonn ) @xcite and nijmegen ( nijm - i ) @xcite potentials , it is useful to formulate the @xmath0@xmath1 scattering problem in @xmath1-space .
one way to accomplish this is to solve for the @xmath200-matrix @xcite @xmath201 where @xmath202 denotes a principal - value integration , and the @xmath1-space matrix elements of the potential are defined in eq .
( [ eq : vpep ] ) .
the integral equations ( [ eq : kma ] ) are discretized , and the resulting systems of linear equations are solved by direct numerical inversion .
the principal - value integration is eliminated by a standard subtraction technique @xcite .
once the @xmath200-matrices in the various channels have been determined , the corresponding ( on - shell ) @xmath101-matrices are obtained from @xmath203^{-1 } \left [ 1 - 2{\rm i } \
, \mu p\ , k^j(p;p ) \right ] \ ,
\label{eq : skma}\ ] ] and from these the amplitudes @xmath204 , eq .
( [ eq : am ] ) , are constructed .
some of the studies of pv effects in the @xmath0@xmath1 system of interest here , specifically those relative to the @xmath2@xmath1 radiative capture , @xmath205 photo - disintegration at threshold and @xmath206@xmath0@xmath1 electro - disintegration in quasi - elastic kinematics are more conveniently carried out in @xmath207-space , and therefore require @xmath207-space wave functions . to this end ,
one first re - writes eq .
( [ eq : psip ] ) in a compact notation as @xmath208 where the matrices @xmath209_{\alpha^\prime,\alpha}\equiv w^j_{\alpha^\prime,\alpha}(r;p)$ ] and @xmath210_{\alpha^\prime,\alpha}\equiv \delta_{\alpha^\prime,\alpha } j_{l^\prime}(pr)$ ] have been introduced for ease of presentation .
then , by making use of the following relation between the off - shell @xmath46- and @xmath200-matrices @xmath211 which on - shell leads to @xmath212^{-1 } k^j(p;p ) \ , \ ] ] one can simply express the @xmath213 matrix of solutions in terms of the previously determined @xmath200-matrix as @xmath214\nonumber \\ & \times & \bigg [ 1 + 2{\rm i}\ , \mu p\ , k^j(p;p)\bigg]^{-1 } \ .
\label{eq : psirk}\end{aligned}\ ] ] the bessel transforms above are carried out numerically by gaussian integration over a uniform @xmath1-grid extending up to momenta @xmath215 125 @xmath216 .
the computer programs have been successfully tested by comparing , for the pc argonne @xmath5 ( av18 ) @xcite and pv ddh @xcite potentials , @xmath207-space wave functions as obtained from eq .
( [ eq : psirk ] ) and by direct solution of the schrdinger equations , eq .
( [ eq : cschr ] ) .
the deuteron state has @xmath150=1 and its normalized wave function is written in @xmath207-space as @xmath217 it has pc components with @xmath218@xmath101@xmath46=010 and 210 , the standard @xmath219s@xmath220 and @xmath219d@xmath220 waves ( however , note again the unconventional phase factor @xmath148 , which makes the sign of the d - wave opposite to that of the s - wave ) , and pv components with @xmath218@xmath101@xmath46=100 and 111 , the @xmath221p@xmath220 and @xmath219p@xmath220 waves ( which are real functions because of the phase choice above ) .
the radial functions are determined by solving the schrdinger equation ( [ eq : cschr ] ) in @xmath150=1 channel with the boundary conditions @xmath222 in the limit @xmath223 and @xmath224\ , \\
u_{100}(r ) & \propto & \frac{{\rm e}^{-\kappa r}}{r } \bigg ( 1 + \frac{1}{\kappa r}\bigg)\ , \end{aligned}\ ] ] in the asymptotic region .
the asymptotic behavior of @xmath225 is identical to that of @xmath226 above , and the constant @xmath227 denotes the combination @xmath228 , where @xmath229 is the deuteron binding energy ( 2.225 mev ) . in @xmath1-space
the deuteron wave function is obtained from solutions of the homogeneous integral equations @xmath230 and from these @xmath231 figure [ fig : deut ] displays the functions @xmath232 obtained with the pc av18 @xcite ( bonn @xcite ) and pv ddh @xcite potentials ( the values for the coupling constants and cutoff parameters in the ddh potential are those listed in table [ tb : ddh ] ) .
the pc @xmath219s@xmath220 and @xmath219d@xmath220 components are not very sensitive to the input pc potential .
for example , most of the difference between the av18 and bonn @xmath219d@xmath220 waves is due to non - localities present in the one - pion - exchange ( ope ) part of the bonn potential .
in fact , it has been known for over two decades @xcite , and recently re - emphasized by forest @xcite , that the local and non - local ope interactions are related to each other by a unitary transformation . therefore the differences between local and non - local ope can not be of any consequence for the prediction of observables , such as binding energies and electromagnetic form factors , provided , of course , that three - body interactions and/or two - body currents generated by the unitary transformation are also included
see ref .
@xcite for a recent demonstration of this fact within the context of a calculation of the deuteron structure function @xmath233 and tensor observable @xmath234 based on the local av18 and non - local bonn potentials and associated ( unitarily consistent ) electromagnetic currents .
this point was also stressed in ref .
@xcite .
the pv @xmath219p@xmath220 component is , in magnitude , much larger than the @xmath221p@xmath220 .
this is easily understood , since the long - range @xmath6-exchange term in the ddh potential is non - vanishing only for transitions in which @xmath235=1 , and therefore does not contribute in the @xmath221p@xmath220 channel . in this channel , however , the ddh @xmath7- and @xmath8-exchange terms play a role . note that , because of the short - range character of the associated dynamics , the av18 and bonn @xmath221p@xmath220 waves show considerably more model - dependence than the corresponding @xmath219p@xmath220 waves . finally , in fig .
[ fig : deut ] the pv @xmath219p@xmath220 wave obtained with the av18 and a truncated ddh potential , retaining only the short - range @xmath7- and @xmath8-exchanges , is also shown .
the comparison between the @xmath219p@xmath220 waves corresponding to the full and truncated ddh potentials demonstrates that this channel is indeed dominated by the @xmath6-exchange term in the ddh .
in this section we give explicit expressions for parity - violating ( pv ) observables in the @xmath0@xmath1 system , including the longitudinal asymmetry and spin rotation in @xmath2@xmath1 elastic scattering , the photon asymmetry in @xmath2@xmath1 radiative capture , and the asymmetries in deuteron photo - disintegration @xmath236 in the threshold region and electro - disintegration @xmath237 in quasi - elastic kinematics .
the differential cross section for scattering of a neutron with initial polarization @xmath181 is given by @xmath238 and the longitudinal asymmetry is defined as @xmath239 where @xmath240 denote the initial polarizations @xmath241 .
the total asymmetry @xmath242 , integrated over the solid angle , then reads @xmath243 where @xmath244 is the spin - averaged differential cross section .
the optical theorem allows @xmath242 to be simply expressed as @xmath245 m_{11t^\prime,11t}(e,0 ) + \frac{(-)^{t+t^\prime}}{2 } \sum_{s\ , s^\prime } \left[1 - ( -)^{s+s^\prime } \right ] m_{s^\prime 0 t^\prime , s0t}(e,0 ) \bigg ] \nonumber \\ & \times & \frac{1 } { { \rm i m } \sum_{sm_st } m_{sm_st , sm_st}(e,0 ) } \>\ > ,
\label{eq : asy_t}\end{aligned}\ ] ] where in the equation above use has been made of the symmetry property @xmath246 it is clear that the numerator of @xmath242 would vanish in the absence of pv interactions , since @xmath113 , in contrast to @xmath50 , can not change the total spin @xmath101 or isospin @xmath46 of the @xmath0@xmath1 pair . in particular , the long - range part of @xmath50 due to pion exchange can only contribute to the first term in the numerator of eq .
( [ eq : asy_t ] ) , since it is diagonal in @xmath101 , but non - vanishing for transitions @xmath247=1 .
the transmission of a low - energy neutron beam through matter is described in terms of an index of refraction . a heuristic argument , outlined in ref .
@xcite , and the more rigorous although less transparent derivation presented in ref .
@xcite show that a neutron with spin projection @xmath248 , after traversing a slab of width @xmath249 of matter , is described by an asymptotic wave function given by @xmath250 where @xmath115=@xmath251 is the initial relative momentum ( assumed along the @xmath190-axis ) , and the index of refraction @xmath252 is related to the density @xmath7 of scattering centers in matter and the forward scattering amplitude .
for the specific case under consideration here neutron scattering from hydrogen this relation reads : @xmath253 thus a neutron , initially polarized in the @xmath254-direction , @xmath255 having traversed a slab of matter , is described in the asymptotic region by a wave function given by @xmath256/\sqrt{2 } \ .\ ] ] in the absence of @xmath50 , the difference @xmath257 vanishes , since it is proportional to the sum over @xmath21 in the numerator of eq .
( [ eq : asy_t ] ) , while @xmath258 where @xmath62 is the spin - averaged cross section introduced above , and hence there will be an attenuation in the beam flux proportional to exp(@xmath259 ) .
the real part of @xmath260 instead generates an un - observable phase factor .
however , if pv interactions are present , then the real part of the ( now non - vanishing ) difference @xmath257 leads to a rotation of the neutron polarization by an angle @xmath261 explicitly given by @xcite @xmath262_{\theta=0 } \>\>. \label{eq : nphi}\ ] ] in the center - of - mass ( cm ) frame , the radiative transition amplitude between an initial continuum state with neutron and proton in spin - projection states @xmath181 and @xmath182 , respectively , and in relative momentum @xmath115 , and a final deuteron state in spin - projection state @xmath263 , recoiling with momentum @xmath264 , is given by @xmath265 where @xmath78 is the momentum of the emitted photon and @xmath266 , @xmath267 , are the spherical components of its polarization vector , and @xmath268 is the nuclear electromagnetic current operator .
note that @xmath269 has been taken along the @xmath190-axis , the spin - quantization axis .
the initial @xmath0@xmath1 continuum state , satisfying outgoing - wave boundary conditions , is related to that constructed in sec .
[ sec : pwest ] via @xmath270 where in the first line the factor @xmath271 is from a clebsch - gordan coefficient combining the neutron and proton states to total isospin @xmath46 , and in the second line @xmath272 and the states @xmath273 have wave functions @xmath274 with @xmath133=@xmath218@xmath101@xmath46 and similarly for @xmath134 .
the quantum numbers @xmath133 and @xmath134 characterize the incoming and outgoing waves , respectively .
the cm differential cross - section for capture of a neutron with spin projection @xmath181 is then written as @xmath275 where @xmath191 is the angle between @xmath276 and @xmath277 and @xmath278 here @xmath133 is the fine - structure constant , @xmath263 is the deuteron mass , @xmath111 is the relative velocity , @xmath279 , and the photon energy @xmath280 is given by @xmath281 \simeq |e_d|+\frac{p^2}{2\mu } \ .\ ] ] the photon asymmetry @xmath282 is defined as in eq .
( [ eq : athe ] ) with @xmath283 replaced by @xmath284 . by expanding the matrix elements of the current operator in terms of reduced matrix elements ( rmes ) of electric ( @xmath285 ) and magnetic ( @xmath286 ) multipole operators as @xcite @xmath287 \ , \label{eq : mult}\end{aligned}\ ] ] with @xmath288 and @xmath289=@xmath285 or @xmath286 , one finds , by retaining the @xmath221s@xmath290 and @xmath219s@xmath220 channels in the sum over @xmath133 in eq .
( [ eq : psia ] ) , the only relevant incoming waves in the energy regime of interest here ( fractions of ev ) , @xmath291 where @xmath292 is given by @xmath293 + { \rm re}\,\big[e_1^*(^3{\rm s}_1 ) m_1(^3{\rm s}_1)\big ] } { |m_1(^1{\rm s}_0)|^2 + |e_1(^1{\rm s}_0)|^2 + |m_1(^3{\rm s}_1)|^2 + |e_1(^3{\rm s}_1)|^2 } \ , \ ] ] and in eq .
( [ eq : mult ] ) the @xmath294 are standard rotation matrices @xcite .
the @xmath295 and @xmath296 rmes should carry a superscript @xmath297 ; it has been dropped for ease of presentantion .
several comments are now in order .
firstly , the photon asymmetry has the expected dependence on @xmath298 , since @xmath299 .
note that the contributions of higher order multipole operators with @xmath300=2 have been ignored in the equation above . secondly , because of the definition of the states in eq .
( [ eq : psiap ] ) , a generic rme @xmath301 is expressed as @xmath302 namely as a sum over the contributions of outgoing channels @xmath134 corresponding to an incoming channel @xmath133 ; for example , @xmath303 thirdly , in the absence of parity - violating interactions , the only surviving rmes are the @xmath304 , @xmath305 , and @xmath306 , and therefore the parameter @xmath292 vanishes .
furthermore , the @xmath305 rme also vanishes due to orthogonality of the initial and final states ( in the limit in which isoscalar two - body currents are neglected ) , while the @xmath306 rme is suppressed in the energy regime of interest here .
thus , the standard result for the spin - averaged radiative capture cross - section , integrated over the solid angle , follows : @xmath307
lastly , when pv interactions are present , the analysis is more delicate , since then , in addition to ad - mixing small opposite - parity components into the wave functions corresponding to @xmath113 , these interactions also induce two - body terms in the electromagnetic current operator , as discussed in sec .
[ sec : cnt ] .
thus , @xmath51=@xmath308 , where @xmath309 includes the convection and spin - magnetization currents of single nucleons as well as the two - body currents associated with @xmath113 , while @xmath310 includes those terms generated by @xmath50 .
the multipole operators can then be written as @xmath311=@xmath312 , and those constructed from @xmath310 have un - natural parities , namely @xmath313 for @xmath314 and @xmath315 for @xmath316 .
therefore , for example , the rmes @xmath317 and @xmath318 connect the pc @xmath221s@xmath290 state to , respectively , the pc and pv components of the deuteron . a straight - forward analysis
then shows that , up to linear terms in effects induced by @xmath50 either in the wave functions or currents , the parameter @xmath292 is given by @xmath319 \bigg ] } { |m_1(^1{\rm s}_0,^1{\rm s}_0)|^2 } \ , \ ] ] where again terms containing the rmes @xmath305 and @xmath306 have been neglected . in the expression above ,
the rme @xmath320 has also been neglected , since transitions induced by the isoscalar electric dipole operator are strongly suppressed @xcite .
thus , the only relevant transitions are those connecting the @xmath219p@xmath220 pv @xmath0@xmath1 state to the pc deuteron component and the @xmath219s@xmath220 and @xmath219d@xmath220 pc @xmath0@xmath1 states to the @xmath219p@xmath220 pv deuteron component .
the relevant matrix element in the photo - disintegration of a deuteron initially at rest in the laboratory is @xmath321 in the notation of sec .
[ sec : npr ] above . here
@xmath322 represents an @xmath0@xmath1 scattering state with total momentum @xmath78 and relative momentum @xmath115 , satisfying incoming wave boundary conditions .
its ( internal ) wave function has the same partial wave expansion given in eq .
( [ eq : psipw ] ) , except for the replacement @xmath323^*$ ] .
the cross section for absorption of a photon of helicity @xmath324 , summed over the final states and averaged over the initial spin projections of the deuteron , reads : @xmath325 where @xmath326 is the energy of the final state , @xmath327 and @xmath328 note that the factor 1/2 above is introduced to avoid double - counting the final states , and that the dependence upon the boundary condition of the continuum wave functions , the superscript @xmath329 , has been reinserted in the rmes of the electric and magnetic multipoles , namely @xmath330 and @xmath289=@xmath285 or @xmath286 .
the resulting pv asymmetry , defined as @xmath331 , is expressed as @xmath332 } { \sum_{j\alpha } \sum_{l \geq 1 } \epsilon_\alpha \ , \left[\ , \mid\ ! m_l^{(-)}(j,\alpha)\
! \mid^2 + \mid \ !
e_l^{(-)}(j,\alpha ) \ !
\mid^2 \ , \right ] } \label{eq : pg}\ ] ] and therefore vanishes unless ( i ) the initial and/or final states do not have definite parity ( as is the case here because of the presence of pv @xmath64@xmath64 interactions ) and/or ( ii ) the electric and magnetic multipole operators have unnatural parities @xmath315 and @xmath313 , respectively , because of two - body pv electromagnetic currents associated with pv @xmath64@xmath64 interactions @xcite . it is easily shown that in the inverse process @xmath333 the expression for the photon circular polarization parameter is identical to that given above , but for the rmes @xmath334 and @xmath335 being replaced by the corresponding @xmath336 and @xmath337 , defined in the previous section .
indeed , by making use of the transformation properties of the states and electric and magnetic multipole operators under time reversal @xmath338 , @xmath339 one finds the following relation for the rmes : @xmath340 and similarly for the @xmath286 s .
hence , the circular polarizations measured in the direct and inverse processes are the same .
the longitudinal asymmetry in the inclusive scattering of polarized electrons off a nuclear target results from the interference of amplitudes associated with photon and @xmath341 exchanges as well as from the presence of parity - violating components in the nucleon - nucleon interaction . for completeness
, we summarize below the relevant formulae .
the initial and final electron ( nucleus ) four - momenta are labeled by @xmath342 and @xmath343 ( @xmath344 and @xmath345 ) , respectively , while the four - momentum transfer @xmath346 is defined as @xmath347 .
the amplitudes for the @xmath10- and @xmath11-exchange processes are then given by @xcite @xmath348 where @xmath349 is the fermi constant for muon decay , @xmath350 and @xmath351 are the standard model values for the neutral - current couplings to the electron given in terms of the weinberg angle @xmath106 , @xmath352 and @xmath353 are the initial and final electron spinors , and @xmath354 and @xmath355 denote matrix elements of the electromagnetic and weak neutral currents , i.e. @xmath356 and similarly for @xmath355 . here
@xmath357 and @xmath358 represent the initial deuteron state and final @xmath0@xmath1 scattering state with incoming - wave boundary conditions ( the @xmath329 solution ) , respectively .
note that in the amplitude @xmath359 the @xmath93 dependence of the @xmath341 propagator has been ignored , since @xmath360 . the parity - violating asymmetry @xmath361 is given by the ratio of the difference over the sum of the inclusive cross sections @xmath362 for incident electrons with helicities @xmath363 .
it depends on the three - momentum and energy transfers , @xmath280 and @xmath8 , and scattering angle , @xmath364 , of the electron and conveniently expressed @xmath365 standard manipulations then lead to the following expression for the asymmetry in the extreme relativistic limit for the electron @xcite @xmath366 where the @xmath111 s are defined in terms of electron kinematical variables , @xmath367 the @xmath368 s are the nuclear electro - weak response functions , which depend on @xmath280 and @xmath8 , to be defined below . to this end , it is first convenient to separate the weak current @xmath369 into its vector @xmath100 and axial - vector @xmath370 components , and to write correspondingly @xmath371 the response functions can then be expressed as @xmath372 \ , \label{eq : rl } \\
r_t^{\gamma,{\rm a}}(q,\omega ) & = & \overline{\sum_i }
\sum_f \delta(\omega+e_d - e_f ) \ , { \rm re } \left [ { \bf j}^\gamma_{t , fi}({\bf q } ) \cdot { \bf j}^{{\rm a } * } _ { t , fi}({\bf q } ) \right ] \ , \label{eq : rt } \\ r_{t^\prime}^{\gamma,{\rm b}}(q,\omega)&=&\overline{\sum_i } \sum_f \delta(\omega+e_d - e_f ) \ , { \rm i m } \left [ { \bf j}^\gamma_{fi}({\bf q } ) \times { \bf j}^{{\rm b } * } _ { fi}({\bf q } ) \right]_z \ , \label{eq : rtpp } \end{aligned}\ ] ] where @xmath373 is the ground - state energy of the deuteron ( assumed at rest in the laboratory ) , @xmath326 is the energy of the final scattering state , and in eqs . (
[ eq : rl ] ) and ( [ eq : rt ] ) [ eq .
( [ eq : rtpp ] ) ] the superscript @xmath374 ( @xmath375 ) is either @xmath10 or @xmath376 ( @xmath10 or 5 ) .
note that there is a sum over the final states and an average over the initial spin projections of the deuteron . in the expressions above for the @xmath368 s
, it has been assumed that the three - momentum transfer @xmath78 is along the @xmath190-axis , which defines the spin quantization axis for the nuclear states .
the asymmetry induced by hadronic weak interactions , @xmath377 , is easily seen to be proportional to the interference of electric and magnetic multipole contributions as in eq .
( [ eq : pg ] ) , indeed @xmath378 namely @xmath379 is related , of course for @xmath8=@xmath280 , to the difference of helicity - dependent photo - disintegration cross sections .
similar considerations to those in sec .
[ sec : gdnp ] allow one to conclude that this response would vanish in the absence of pv @xmath64@xmath64 interactions ( note , however , that in the present case there is , in addition to two - body pv currents , also a one - body pv term originating from radiative electro - weak corrections , the anapole current @xcite ) .
in this section we briefly review the techniques used to calculate the pv observables in the @xmath0@xmath1 system these are similar to those discussed most recently in ref .
@xcite . the deuteron wave function in eq .
( [ eq : dw ] ) is written , for each spatial configuration @xmath380 , as a vector in the spin - isospin space of the two nucleons , @xmath381 where @xmath382 and @xmath383 are the components of @xmath384 in this basis . the scattering wave function in eq .
( [ eq : psipw ] ) is first approximated by retaining pc and pv interaction effects in all channels up to a certain pre - selected @xmath385 and by using spherical bessel function for channels with @xmath386 , and is then expanded , for any given @xmath380 , in the same basis @xmath387 defined above .
the radial functions @xmath388 are obtained with the methods discussed in sec .
[ sec : p - space ] .
matrix elements of the electromagnetic ( and neutral - weak ) current operators are written schematically as @xmath389 the spin - isospin algebra is performed exactly with techniques similar to those developed in ref .
@xcite , while the @xmath380-space integrations are carried out efficiently by gaussian quadratures .
note that no multipole expansion of the transition operators is required .
extensive and independent tests of the computer programs have been completed successfully .
in this section we present results for the longitudinal asymmetry and spin rotation in @xmath2@xmath1 elastic scattering , the photon asymmetry in the @xmath2@xmath1 radiative capture at thermal energies , and the asymmetries in the threshold photo - disintegration and quasi - elastic electro - disintegration of the deuteron . to provide an estimate for the model dependence of these results , we consider several different high - quality interactions fit to strong - interaction data , the argonne @xmath5 ( av18 ) @xcite , bonn ( bonn ) @xcite and nijmegen - i ( nijm - i ) @xcite interactions .
we adopt the standard ddh @xcite one - boson exchange model of the parity - violating ( pv ) interaction , and solve the schrdinger equation for the scattering state and deuteron bound state with the methods discussed in sec .
[ sec : app ] .
the values for the meson - nucleon coupling constants and cutoff parameters in the ddh model are those listed in table [ tb : ddh ] .
note that we have taken the linear combination of @xmath7- and @xmath8-meson weak coupling constants corresponding to @xmath18@xmath1 elastic scattering from an earlier analysis @xcite of these experiments .
the remaining couplings are the ddh `` best value '' estimates . as in the earlier analysis of @xmath390 scattering , the cutoff values in the meson - exchange interaction are taken from the bonn potential .
it is also useful to introduce here some of the notation adopted in the following subsections to denote variations on the ddh model defined above .
the pv interaction denoted as ddhb corresponds to a ddh model with @xmath6 , @xmath7 and @xmath8 weak coupling constants as specified by the `` best value '' set of ref .
@xcite , while the pv interaction denoted as ddh@xmath6 includes only the @xmath6-exchange term in the ddh model with the `` best value '' for the weak @xmath6@xmath64@xmath64 coupling constant .
the remaining pion and vector - meson strong interaction coupling constants and short - range cutoff parameters are as given in table [ tb : ddhb ] for the ddhb model and are the same as in table [ tb : ddh ] for the ddh@xmath6 model . finally
, while the short - range contributions to the pv interaction should not be viewed as resulting solely from the exchange of single mesons , the six parameters of the ddh model are still useful in characterizing all the low - energy pv mixings . for example , two - pion exchange could play a role @xcite , however we assume that its effects can be included , at least at low energy , through the present combination of pion- and short - range terms . in this section we present results for the longitudinal asymmetry and neutron spin rotation in @xmath2@xmath1 elastic scattering .
figures [ fig : e0][fig : e2a ] show the mixing parameters @xmath391 induced by av18 model in combination with the ddh , ddhb , and ddh@xmath6 interactions .
only those @xmath391 induced by the pv interaction are displayed , namely for @xmath150=0 @xmath392 and for @xmath393 @xmath391 with @xmath300=1,2 and @xmath29=3,4 in the notation of tables [ tb : chan ] and [ tb : mixing ] .
the definitions adopted for the phase - shifts and mixing parameters are those introduced in sec .
[ sec : phase ] . up to linear terms in @xmath50 , the @xmath394 and @xmath395 values
are not affected by weak interactions , and are determined solely by the strong interaction .
they are identical to those listed in ref .
@xcite , but for two differences .
firstly , the blatt - biedenharn parameterization is used here for the @xmath101-matrix @xcite rather than the bar - phase parameterization of it @xcite employed in ref .
@xcite . secondly , because of the phase choice in the potential components ( see eq .
( [ eq : vpe ] ) and comment below it ) , the mixing parameters @xmath395 have opposite sign relative to those listed in ref . @xcite .
the coupling between channels with the same pair isospin @xmath46 is induced by the short - range part of the ddh interaction , associated with vector - meson exchanges ; its long - range component , due to pion exchange , vanishes in this case . as a result , the mixing parameters in figs .
[ fig : e0 ] , [ fig : e1a ] , and [ fig : e2a ] , calculated with the ddh and ddhb models , are rather different , reflecting the large differences in the values for the some of the strong and weak coupling constants and short - range cutoffs between these two models , see tables [ tb : ddh ] and [ tb : ddhb ] .
the mixing parameters between channels with @xmath396=1 , figs . [ fig : e1b ] and [ fig : e2b ] , in which the pion - exchange term is present , are still rather sensitive to the short - range behavior of the pv interaction , as reflected again by the differences in the ddh and ddhb predictions .
however , this sensitivity is much reduced for the more peripheral waves , such as the @xmath219p@xmath397-@xmath219d@xmath397 and @xmath219f@xmath397-@xmath219d@xmath397 channels .
figures [ fig : e0ab][fig : e1aab ] are meant to illustrate the sentitivity of the mixing angles to the input strong - interaction potential , which can be quite large , particularly in channels , such as the @xmath219d@xmath220-@xmath221p@xmath220 . the total longitudinal asymmetry , defined in eq . ( [ eq : asye ] ) , is shown in fig .
[ fig : anp ] for a number of combinations of strong- and weak - interaction potentials .
the asymmetries were calculated by retaining in the partial wave expansion for the amplitude , eq .
( [ eq : am ] ) , all channels with @xmath150 up to @xmath385=6 .
there is very little sensitivity to the input strong - interaction potential . as also remarked in ref .
@xcite , this reduced sensitivity is undoutbly a consequence of the fact that present potentials are fitted to extended @xmath1@xmath1 and @xmath1@xmath0 databases with high accuracy .
figure [ fig : anpj ] shows that the total asymmetries obtained by including only the @xmath150=0 and 1 channels ( @xmath221s@xmath290-@xmath219p@xmath290 , @xmath219s@xmath220-@xmath219p@xmath220 , @xmath219d@xmath220-@xmath219p@xmath220 , @xmath219s@xmath220-@xmath221p@xmath220 , and @xmath219d@xmath220-@xmath221p@xmath220 ) and , in addition , the @xmath150=2 channels , and finally all @xmath150 channels up to @xmath385=6 . in the energy range ( 0200 ) mev the asymmetry is dominated by the @xmath150=02 contributions . for completeness , we present in figs . [
fig : snpa ] and [ fig : anpa ] results for the angular distributions of the ( pc ) spin - averaged differential cross section and ( pv ) longitudinal asymmetry at center - of - mass energies of 20 mev and 100 mev .
the asymmetry @xmath398 is defined in eq .
( [ eq : athe ] ) . the predictions for the neutron spin rotation per unit length , @xmath399 with @xmath261 defined in eq .
( [ eq : nphi ] ) , are listed in table [ tb : nspin ] in the limit of vanishing incident neutron energy .
the density of liquid hydrogen is taken as @xmath7=@xmath400 atoms-@xmath401 . while results corresponding to different input strong interactions are within @xmath402 10% of each other , the calculated values show significant sensitivity to the short - range bahavior of the pv interaction , columns labeled ddh and ddhb .
it is worth re - emphasizing that the longitudinal asymmetry in @xmath18@xmath1 elastic scattering predicted by the ddhb model is at variance with that observed experimentally @xcite .
the short - range cutoff parameters and combinations of @xmath7- and @xmath8-meson pv coupling constants in @xmath403=11 , respectively @xmath404 and @xmath405 , were constrained , in the ddh model , to reproduce this ( measured ) asymmetry @xcite .
an additional difference between the ddh and ddhb models is in the values adopted for the ( pc ) @xmath7-meson tensor coupling to the nucleon , 6.1 in the ddh ( from the bonn interaction ) and 3.7 in the ddhb ( consistent with estimates from vector - meson dominance ) .
hence , the ddhb results are not realistic .
comparison between the ddh and ddh@xmath6 predictions , however , indicates that the neutron spin rotation is sensitive to the long - range part of @xmath50 , and therefore a measurement of this observable would be useful in constraining the pv @xmath6@xmath64@xmath64 coupling constant .
finally , there is a sign difference between the present results and those reported in ref .
it is not due to the different strong interaction potential used in that calculation .
indeed , with the paris potential @xcite in combination with the ddhb model we obtain @xmath399=@xmath406 rad-@xmath407 , the same magnitude but opposite sign than given in ref .
@xcite .
in order to understand this discrepancy , we have carried out a calculation of the neutron spin rotation , which ignores strong - interaction effects .
it is equivalent to a first - order ( in @xmath50 ) perturbative estimate of this observable , and the corresponding results , listed in the last row of table [ tb : nspin ] ( row labeled `` plane waves '' ) , demonstrate that strong - interaction distorsion effects are crucial , in fact they are responsible for flipping the sign of @xmath261 .
this is in contradiction with the statement reported in the first paragraph after eq .
( 6 ) of ref .
@xcite : avishai and grange claim that the `` plane - wave '' prediction with the ddhb model is @xmath408 rad-@xmath407 , namely it has the same sign as in their full calculation .
the sign difference between the predictions obtained by either including or neglecting strong - interaction distorsion effects can easily be understood . for simplicity , consider the ddh@xmath6 model , in which case the relevant matrix element contributing to @xmath261 is @xmath409p@xmath410@xmath411@xmath412s@xmath413 , connecting the continuum @xmath46=0 @xmath219s@xmath220 and @xmath46=1 @xmath219p@xmath220 channels .
the essential difference between the un - distorted and distorted @xmath219s@xmath220 wave functions is the presence of a node in the latter , thus ensuring its orthogonality to the deuteron @xmath219s@xmath220 component .
it is this node that causes the sign flip .
the pv asymmetry @xmath292 in the @xmath221h(@xmath2,@xmath10)@xmath15h reaction at thermal neutron energies is calculated for the av18 , bonn and nijm - i interactions .
the asymmetry is expected to be constant for low - energy neutrons up to energies well beyond the 115 mev averaged in the experiment currently running at the lansce facility @xcite .
each strong interaction model has associated two - body currents . for the av18
we consider the currents from the momentum - independent terms the @xmath6- and @xmath7-exchange currents from its @xmath60 part as well as from the momentum - dependent terms , as reviewed in sec .
[ sec : cnt ] .
further discussion of the av18 currents is given below .
for the bonn and nijm - i interactions , we retain only the @xmath6- and @xmath7-exchange currents with cutoff parameters taken from the bonn model ( @xmath80=1.72 gev and @xmath414=1.31 gev ) , while we neglect contributions from other meson exchanges . in all calculations , however , the currents associated with the @xmath57 excitation and @xmath8@xmath6@xmath10 transition have been included .
the total cross section @xmath415 is due to the well - known @xmath416 transition connecting the pc @xmath221s@xmath290 @xmath0@xmath1 state to the pc deuteron state .
the calculated values for each model are given in table [ tb : np_x ] , both for one - body ( impulse ) currents alone and for the one- and two - body currents .
in each case the largest two - body contribution , approximately two - thirds of the total , comes from the currents associated with pion exchange .
the total cross section is in good agreement with experimental results , which are variously quoted as 334.2(0.5 ) mb @xcite or 332.6(0.7 ) mb @xcite . it would be possible to adjust , for example , the transition magnetic moment @xmath417 of the @xmath57-excitation current to precisely fit one of these values , here we simply choose a @xmath417 of 3 n.m . ,
which is consistent with an analysis of @xmath10-@xmath64 data at resonance . as discussed sec .
[ sec : npr ] , the pv asymmetry arises from an interference between the @xmath416 term above and the @xmath9 transition , connecting the @xmath219p@xmath220 pv @xmath0@xmath1 state to the pc deuteron state and the @xmath219s@xmath220 pc @xmath0@xmath1 state to the @xmath219p@xmath220 pv deuteron state .
the @xmath9 transitions proceeding through the pv @xmath221p@xmath220 @xmath0@xmath1 or deuteron states are suppressed , because of an isospin selection rule forbidding isoscalar electric - dipole transitions and also because of spin - state orthogonality . in principle
, there is a relativistic correction to the electric dipole operator , associated with the definition of the center - of - energy @xcite .
however , its contribution in transitions proceeding through the @xmath221p@xmath220 channel vanishes too , since the associated operator is diagonal in the pair spin .
the calculated asymmetries are listed in table [ tb : np_x ] .
the results are consistent with earlier @xcite and more recent @xcite estimates , and are in agreement with each other at the few - per - cent level , which is also the magnitude of the contributions from the short - range terms . in particular , they show that this observable is very sensitive to the weak pv @xmath6@xmath64@xmath64 coupling constant , while it is essentially unaffected by short - range contributions ( in this context , see also fig .
[ fig : deut ] ) .
the @xmath9 transition has been calculated in the long - wavelength approximation ( lwa ) with the siegert form of the @xmath9 operator ( see , for example , eq .
( 4.5 ) of ref .
@xcite ) , thus eliminating many of the model dependencies and leaving only simple ( long - range ) matrix elements . in the notation of sec .
[ sec : npr ] , the associated reduced matrix elements ( rmes ) are explicitly given by @xmath418 \nonumber \\
& -&w(r;^3\!{\rm p}_1 ) \left [ u(r;^3\!{\rm s}_1 ) -\frac{1}{\sqrt{2 } } u(r;^3\!{\rm d}_1 ) \right ] \bigg ] \ , \end{aligned}\ ] ] where the @xmath419 s and @xmath352 s denote the @xmath0@xmath1 continuum and deuteron radial wave functions defined , respectively , as in secs .
[ sec : pwest ] and [ sec : deut ] ( only the outgoing channel quantum numbers are displayed for the @xmath419 s ) .
corrections beyond the lwa terms in @xmath9 transitions have been found to be quite small . for completeness
, we also give the well known expression for the @xmath416 rme , as calculated to leading order in @xmath280 and in the limit in which only one - body currents are retained , @xmath420 where the combination @xmath421=4.706 n.m .
is the nucleon isovector magnetic moment .
we have also calculated the @xmath9 contributions with the full current density operator @xmath422 , namely by evaluating matrix elements of @xmath423 where @xmath424 are standard vector spherical harmonics . to the extent that retardation corrections beyond the lwa of the @xmath9 operator are negligible @xcite
, this should produce identical results _ provided _ the current is exactly conserved . in order to satisfy current conservation , currents from both the strong ( pc ) and weak ( pv ) interactions
are required , as discussed in sec .
[ sec : cnt ] . in the following
we keep only the @xmath6-exchange term in the ddh interaction ( with their `` best guess '' for the weak @xmath6@xmath64@xmath64 coupling constant ) , and use the av18 strong - interaction model . as reviewed in sec .
[ sec : cnt ] , the pc two - body currents constructed from the @xmath60 part of the av18 interaction ( the @xmath6- and @xmath7-exchange currents ) exactly satisfy current conservation with it .
the same holds true for the pv @xmath6-exchange currents derived from the ddh interaction in sec .
[ sec : cntpv ] .
however , the pc two - body currents originating from the isospin- and momentum - dependent terms of the av18 are strictly not conserved ( see below ) .
the associated contributions , while generally quite small , play here a crucial role because of the large cancellation between the ( pc ) @xmath60 currents from the av18 and the ( pv ) @xmath6-currents from the ddh .
this point is illustrated in table [ tb : cntre ] .
note that the pc currents from @xmath57-excitation and @xmath8@xmath6@xmath10 transition are transverse and therefore do not affect the @xmath9 matrix element .
however , they slightly reduce the pv asymmetry , since their contributions increase the @xmath416 matrix element by @xmath425% .
they are not listed in table [ tb : cntre ] .
the asymmetry is given by the sum of the two columns in table [ tb : cntre ] , namely @xmath426 ( last row ) .
this value should be compared to @xmath427 , obtained with the siegert form of the @xmath9 operator for the same interactions ( and currents for the @xmath416 matrix element ) .
as already mentioned , we have explicitly verified that retardation corrections in the @xmath9 operator are too small to account for the difference .
thus the latter is to be ascribed to the lack of current conservation , originating from the isospin- and momentum - dependent terms of the av18 . to substantiate this claim ,
we have carried out a calculation based on a @xmath428 reduction @xcite of the av18 ( denoted as av8 ) , constrained to reproduce the binding energy of the deuteron and the isoscalar combinations of the s- and p - wave phase shifts ( note , however , that we do include in the av8 the electromagnetic terms from the av18 , omitted in ref .
@xcite ) . for the av8 model , the @xmath6- and @xmath7-exchange currents from the isospin - dependent central , spin - spin , and tensor interaction components
are constructed as for the av18 , and therefore are exactly conserved .
however , the currents from the isospin - independent interaction , @xmath429 , are derived by minimal substitution , @xmath430 where @xmath431 and @xmath432 are the electric charge and vector potential , respectively , and @xmath85 is the proton projection operator .
the linear terms in @xmath432 are written as @xmath433 , and the resulting spin - orbit current density or , rather , its fourier transform reads @xmath434 in the case of the isospin - dependent terms , after symmetrizing @xmath435_+/2 $ ] , one obtains @xmath436 where @xmath437 while minimal substitution ensures that the current is indeed conserved for the isospin - independent interaction , i.e. @xmath438 \ , \label{eq : j_con}\ ] ] this prescription does not lead to a conserved current for the isospin - dependent one , since the commutator above generates an isovector term of the type @xmath439_+ \ .\ ] ] physically , this corresponds to the fact that isospin - dependent interactions are associated with the exchange of charged particles , which an electromagnetic field can couple to .
one can enforce current conservation by introducing an additional term @xcite , which in the case of @xmath440 is taken as @xmath441_+ \ .\ ] ] the results obtained for the total cross section and pv asymmetry with the av8 and pion - only ddh interactions and associated ( exactly conserved ) currents are listed in table [ tb : cnte1 ] .
a few comments are in order .
firstly , the @xmath416 cross section in impulse approximation is @xmath215 30% smaller than predicted with the av18 interaction .
this is due to the fact that the @xmath0@xmath1 singlet scattering length obtained with the av8 ( truncated ) model is 19.74 fm , and so is about 15% smaller in magnitude than its physical value , 23.75 fm , reproduced by the av18 within less than 0.1% @xcite .
secondly , the enhancement of the @xmath416 cross section in impulse approximation due to ( pc ) two - body currents , 9.3% , is essentially consistent with that predicted with the av18 .
lastly , the pv asymmetry obtained with the full currents is close to that calculated with the siegert form of the @xmath9 operator .
the remaning @xmath425% difference is due to numerical inaccuracies as well as additional corrections from retardation terms and higher order multipoles . both of these effects
are included in the full - current calculation .
note the crucial role played by the spin - orbit currents constructed above .
the photo - disintegration cross sections calculated with the av18 and bonn models from threshold to 20 mev photon energies are in excellent agreement with data @xcite@xcite , see fig .
[ fig : xdg ] .
the model dependence between the av18 and bonn results is negligible . in the calculations the final @xmath0@xmath1 states include interaction effects in all channels up to @xmath150=5 and spherical bessel functions for @xmath442 , as discussed in the next section . in the energy regime of interest here ,
the ( total ) cross section is dominated by the contributions of @xmath9 transitions connecting the deuteron to the @xmath0@xmath1 triplet p - waves .
the siegert form is used for the @xmath9 operator .
because of the way the calculations are carried out ( see sec . [ sec : cal ] ) , it is conveniently implemented by making use of the following identity for the current density operator @xmath422 , or rather its fourier transform @xmath268 , @xmath443 \ , \label{eq : j_s}\end{aligned}\ ] ] where in the first line the volume integral of @xmath422 has been re - expressed in terms of the divergence of the current , ignoring vanishing surface contributions , and in the second line use has been made of the continuity equation . here
@xmath444 is the charge density operator . in evaluating the matrix elements in eq .
( [ eq : jmin ] ) the commutator term reduces to @xmath445 \rightarrow { \rm i } q \ , \int{\rm d}{\bf x } \ , { \bf x } \ , \rho({\bf x } ) \simeq { \rm i } q \ , \sum_i p_i \ , { \bf r}_i \ , \ ] ] where @xmath85 is the proton projection operator introduced earlier , and relativisitc corrections to @xmath444 , such as those associated with spin - orbit and pion - exchange contributions @xcite , have been neglected .
we have also calculated the photo - disintegration cross section by using the expression given in eq .
( [ eq : e1op ] ) for the @xmath9 operator , or equivalently by calculating matrix elements of the current @xmath268 without resorting to the identity in eq .
( [ eq : j_s ] ) .
the results obtained by including only the one - body terms and both the one- and two - body terms in @xmath268 are compared with those obtained in the siegert - based calculation ( as well as with data ) in fig .
[ fig : xdgc ] . the same conclusions as in the previous section remain valid here . had the current been exactly conserved , then the siegert - based and full @xmath268 calculations would have produced identical results .
the small differences in the case of the av18 model , as an example , are to be ascribed to missing isovector currents associated with its momentum - dependent interaction components ( see previous section ) .
the pv photon polarization parameter @xmath446 , obtained with various combinations of pc and pv interactions , is displayed in fig .
[ fig : adg ] , while its value at a photon energy @xmath215 1.3 kev above breakup threshold is listed in table [ tb : tadg ] .
all results presented below use the current operator in the form given on the right - hand - side of eq .
( [ eq : j_s ] ) .
note that , as discussed in sec .
[ sec : gdnp ] , the parameters @xmath446 for the direct @xmath3 and inverse @xmath333 processes are the same .
in the threshold region , a few kev above breakup , the expression for @xmath446 reduces to @xmath447 } { |m_1(^1{\rm s}_0)|^2 } \ , \ ] ] where , in the notation of the previous section , the @xmath448 rme is defined as in eq .
( [ eq : m1def ] ) and similarly for @xmath449 . in this energy region ,
the only relevant channel in the final @xmath0@xmath1 state has @xmath150=0 , see discussion at the end of sec .
[ sec : npr ] . note that the combination of rmes occurring in @xmath446 is different from that in @xmath292 , the photon angular asymmetry parameter measured in @xmath12@xmath1 radiative capture .
indeed , in contrast to @xmath292 , the photon polarization parameter is almost entirely determined by the short - range part of the ddh interaction , mediated by vector - meson exchanges ( and having isoscalar and isotensor character @xcite ) , see table [ tb : tadg ] and fig .
[ fig : adg ] .
this is easily understood , since in the @xmath221s@xmath290-@xmath219p@xmath290 channel the pion - exchange component of the ddh interaction vanishes .
furthermore , the @xmath9 transition connecting the @xmath221s@xmath290 @xmath0@xmath1 continuum state to the pv @xmath219p@xmath220 component of the deuteron , which is predominantly induced by the pion - exchange interaction , is strongly suppressed , to leading order , by spin - state othogonality .
higher order corrections , associated with retardation effects and relativistic contribution to the electric dipole operator , were estimated in ref .
@xcite and were found to be of the order of a few % of the leading result arising from vector - meson exchanges .
some of these corrections are retained in the present study . the predictions in table [ tb : tadg ] and in fig .
[ fig : adg ] display great sensitivity both to the strengths of the pv vector - meson couplings to the nucleon and to differences in the short - range behavior of the strong - interaction potentials , thus reinforcing the conclusion that these short - ranged meson couplings are not in themselves physical observables , rather the parity - violating mixings are the physically relevant parameters .
note that the @xmath215 5% decrease in @xmath446 values between the rows labelled `` impulse '' and `` full '' is due to the corresponding 5% enhancement of the @xmath416 transition connecting the pc @xmath221s@xmath290 and deuteron states , due to two - body terms in the electromagnetic current included in the `` full '' calculation . the results in table [ tb : tadg ] are consistent both in sign and order of magnitude with those of ealier studies @xcite , remaining numerical differences are to be ascribed to different strong- and weak - interaction potentials adopted in these earlier works .
indeed , we have explicitely verified that by using the pc av18 potential and the cabibbo model for the pv potential @xcite we obtain @xmath446 values close to those reported in ref .
@xcite . however , our results seem to be at variance with those of ref .
@xcite at photon energies a few mev above above the breakup threshold .
in particular , table ii in that paper suggests that at 10 and 20 mev the dominant contribution to @xmath446 is from the pv pion - exchange interaction and that @xmath446 has the values @xmath450 and @xmath451 , respectively .
this is in contrast to what reported in fig .
[ fig : adg ] of the present work , curves labelled av18+ddhb and av18+ddh@xmath6 .
there is a two - order of magnitude difference between the values referred to above and those obtained here .
it is not obvious whether this is due to the use in ref .
@xcite of the hamada - johnston @xcite pc potential a pv potential `` close '' to our model ddhb is adopted .
the results in table [ tb : tadg ] are consistent with the latest experimental determination , @xmath446=@xmath452 @xcite , but about two orders of magnitude smaller than an earlier measurement @xcite .
figure [ fig : adgj ] shows the photon - polarization parameter obtained by including pv admixtures in the @xmath0@xmath1 continuum wave functions of all channels with @xmath453 and @xmath385=0 , 1 , 2 and 5 . in the energy range
explored so far , @xmath454 is essentially given by the contributions of the @xmath150=0 and 1 channels .
finally , fig .
[ fig : adg2 ] illustrates the effects of two - body terms in the electromagnetic current , written as in the right - hand - side of eq .
( [ eq : j_s ] ) .
the associated contributions are of the order of a few % relative to those from one - body terms . in this section
we present results for the asymmetries @xmath377 and @xmath455 obtained by including one- and two - body terms in the electromagnetic and neutral - weak currents .
note , however , that only the pv two - body terms associated with @xmath6-exchange in the ddh interaction are considered in the present calculations ( in addition , of course , to the pc terms discussed in sec .
[ sec : cnt ] ) .
the pv currents from @xmath7- and @xmath8-exchange have been neglected , since they are expected to play a minor role due to their short - range character .
one should also observe that at the higher momentum transfers of interest here , 100300 mev / c , relevant for the sample experiments @xcite , it is not possible to include the contributions of electric multipole operators through the siegert theorem , these must be calculated explicitly from the full current .
the @xmath455 contribution was recently studied in ref .
@xcite , where it was shown that two - body terms in the nuclear electromagnetic and weak neutral currents only produce ( 12)% corrections to the asymmetry due to the corresponding single - nucleon currents . the present study a short account of which has been published in ref .
@xcite investigates the asymmetry originating from hadronic weak interactions .
it updates and sharpens earlier predictions obtained in refs .
@xcite for example , these calculations did not include the effects of two - body currents induced by pv interactions .
the present calculation proceeds as discussed in sec .
[ sec : cal ] .
we have used the av18 or bonn models ( and associated currents ) in combination with the full ddh interaction ( with coupling and cutoff values as given in table [ tb : ddh ] ) .
the final state , labeled by the relative momentum @xmath115 , pair spin and @xmath190-projection @xmath456 , and pair isospin @xmath46 ( @xmath117=0 ) , is expanded in partial waves ; pc and pv interaction effects are retained in all partial waves with @xmath457 , while spherical bessel functions are employed for @xmath458 . in the quasi - elastic regime of interest here , it has been found that interaction effects are negligible for @xmath458 . in figs .
[ fig : xsiii ] and [ fig : asiii ] we show , respectively , the inclusive cross section and the asymmetries @xmath455 and @xmath377 , obtained with the av18 and ddh interactions , for one of the two sample kinematics , corresponding to a three - momentum transfer range between 176 mev and 206 mev at the low and high ends of the spectrum in the scattered electron energy @xmath459 , the four - momentum transfer @xmath460 at the top of the quasi - elastic peak is @xmath215 0.039 gev@xmath15 . the rise in the cross section at the high end of the @xmath459 spectrum the threshold region is due to the @xmath416 transition connecting the deuteron to the ( quasi - bound ) @xmath0@xmath1 @xmath221s@xmath290 state .
note that , because of the well - known destructive interference between the one - body current contributions originating from the deuteron s- and d - wave components , two - body current contributions are relatively large in this threshold region .
however , they only amount to a @xmath215 5% correction in the quasi - elastic peak region . in fig .
[ fig : asiii ] the asymmetry @xmath455 labeled ( 1 + 2)-body@xmath455 is defined in eq .
( [ eq : agz])includes , in addition to one - body , two - body terms in the electromagnetic and neutral weak currents ( in both the vector and axial - vector components of the latter ) .
these two - body contributions are negligible over the whole @xmath459 spectrum . however
, the ( pc and pv ) two - body electromagnetic currents play a relatively more significant role in the asymmetry @xmath461 , eq .
( [ eq : agg ] ) . in fig .
[ fig : asddh ] we display separately , for the asymmetry @xmath461 , the contributions originating from i ) the presence in the wave functions of opposite - parity components induced by the ddh interaction ( solid curve ) and ii ) the anapole current and the pv two - body current associated with @xmath6-exchange ( dashed curve ) .
the latter are positive and fairly constant as function of @xmath459 , while the former exhibit a pronounced dependence upon @xmath459 .
note that , up to linear terms in the effects induced by pv interactions , the asymmetry @xmath377 is obtained as the sum of these two contributions .
the bonn model leads to predictions for the inclusive cross section and asymmetries , that are very close to those obtained with the av18 , as shown for @xmath455 and @xmath461 in fig .
[ fig : asbonn ] . thus the strong - interaction model dependence is negligible for these observables . in fig .
[ fig : as ] we present results for the asymmetries corresponding to a four - momentum transfer @xmath460 at the top of the quasi - elastic peak of about 0.094 gev@xmath15 , the three - momentum transfer values span the range ( 266327 ) mev over the @xmath459 spectrum shown .
the calculations are based on the av18 model and include one- and two - body currents .
the asymmetry from @xmath10-@xmath11 interference scales with @xmath93 and therefore is , in magnitude , about a factor of two larger than calculated in fig .
[ fig : asiii ] where @xmath4620.039 gev@xmath15 .
the contributions to @xmath461 exhibit , as functions of @xmath459 , a behavior qualitatively similar to that obtained at the lower @xmath460 value , see fig .
[ fig : asddh ] . in fig .
[ fig : aspai ] we compare results for the contribution to @xmath461 due to the presence in the wave function of opposite - parity components induced by the full ddh and a truncated version of it , including only the pion - exchange term .
note that , up to linear terms in the effects produced by pv interactions , the other contribution to @xmath461 , namely that originating from ( pv ) one- and two - body currents , remains the same as in fig .
[ fig : asddh ] , since as mentioned earlier only the ( pv ) two - body currents associated with pion - exchange are considered in the present work .
figure [ fig : aspai ] shows that the asymmetry @xmath461 is dominated by the long - range pion - exchange contribution .
hence , @xmath461 will scale essentially linearly with the pv @xmath6@xmath64@xmath64 coupling constant .
finally , fig .
[ fig : asv6 ] is meant to illustrate the sensitivity of the asymmetry @xmath377 to those pc two - body currents derived from the momentum - dependent interaction components of the av18 model ( i.e. , the spin - orbit , @xmath61 , and quadratic spin - orbit terms ) , see sec .
[ sec : cnt ] and [ sec : r_np ] .
while these currents play a crucial role in the photon asymmetry in the @xmath2@xmath1 radiative capture at thermal neutron energy , they give negligible contributions to the present observable at quasi - elastic kinematics .
these results demonstrate that , in the kinematics of the sample experiments @xcite , the asymmetry from @xmath10-@xmath11 interference is dominated by one - body currents , and that it is two - orders of magnitude larger than that associated with the pv hadronic weak interaction .
hence even the largest estimates of the weak @xmath6@xmath64@xmath64 coupling constant will not affect extractions of single - nucleon matrix elements .
these conclusions corroborate those of the authors of ref .
@xcite , who have carried out a similar study of the impact of hadronic weak interaction on quasi - elastic electro - deuteron scattering .
a systematic study of parity - violating ( pv ) observables in the @xmath0@xmath1 system , including the asymmetries in @xmath12@xmath1 radiative capture and @xmath3 photo - disintegration , the spin rotation and longitudinal asymmetry in @xmath12@xmath1 elastic scattering , and the asymmetry in electro - disintegration of the deuteron by polarized electrons at quasi - elastic kinematics , has been carried out by using a variety of latest - generation , strong - interaction potentials in combination with the ddh model of the pv hadronic weak interaction .
we find that the model dependence of the @xmath12@xmath1-capture asymmetry is quite small , at a level similar to the expected contributions of the short - range parts of the interaction .
this process is in fact dominated by the long - range interaction components associated with pion exchange .
a measurement of the @xmath12@xmath1-capture asymmetry is then a clean probe of that physics .
similarly , we find that the asymmetry in the @xmath4 reaction at quasi - elastic kinematics is a very clean probe of the electro - weak properties of individual nucleons .
the processes associated with two nucleons , including pv admixtures in the deuteron and scattering wave functions and electromagnetic two - body currents induced by hadronic weak interactions , play a very small role at the values of momentum transfers explored so far @xcite .
we also find that the neutron spin rotation is sensitive to both the pion and vector - meson pv couplings to the nucleon , while exhibiting a modest model dependence , at the level of 5 - 10% , due to the input strong - interaction potential adopted in the calculation .
thus a measurement of this observable @xcite , when combined with measurements of the asymmetries in @xmath12@xmath1 radiative capture @xcite and @xmath16@xmath1 elastic scattering @xcite , could provide useful constraints for some of these pv amplitudes .
the asymmetry in the deuteron disintegration by circularly polarized photons from threshold up to 20 mev energies is dominated by the short - range components of the ddh interaction .
however , it also displays enhanced sensitivity to the short - range behavior in the strong - interaction potentials .
indeed , predictions for the asymmetry at threshold differ by almost a factor of two , depending on whether the argonne @xmath5 or bonn 2000 interactions are used in the calculations .
therefore , this observable can not provide an unambiguous value of short - range weak meson nucleon couplings ; however , they would be valuable in placing constraints on the hadronic weak mixing angles .
finally , the issue of electromagnetic current conservation in the presence of parity - conserving ( pc ) and pv potetials has been carefully investigated .
in particular , in the case of the @xmath463 and @xmath3 processes dramatic cancellations occur betweeen the contributions associated with the two - body currents induced , respectively , by the pc and pv potentials .
the authors wish to thank l.e .
marcucci and m. viviani for interesting discussions and illuminating correspondence , and g. hale for making available to them experimental data sets of the deuteron photo - disintegration .
the work of j.c . and m.p .
was supported by the u.s .
department of energy under contract w-7405-eng-36 , and the work of r.s . was supported by doe contract de - ac05 - 84er40150 under which the southeastern universities research association ( sura ) operates the thomas jefferson national accelerator facility . finally , some of the calculations were made possible by grants of computing time from the national energy research supercomputer center .
berdoz _ et al .
_ , phys .
lett . * 87 * , 272301 ( 2001 ) .
snow _ et al .
_ , nucl .
instrum . and methods phys
. res . a * 440 * 729 ( 2000 ) .
b. wojtsekhowski and w.t.h .
van oers ( spokespersons ) , jlab letter - of - intent 00 - 002 .
r. hasty _ et al . _ , science * 290 * , 2117 ( 2000 ) .
spayde _ et al .
_ , phys .
lett . * 84 * , 1106 ( 2000 ) .
j. carlson , r. schiavilla , v.r .
brown , and b.f .
gibson , phys .
c * 65 * , 035502 ( 2002 ) .
r. schiavilla , j. carlson , and m. paris , phys .
c * 67 * , 032501 ( 2003 ) .
b. desplanques , j.f .
donoghue , and b.r .
holstein , ann .
( n.y . ) * 124 * , 449 ( 1980 ) .
wiringa , v.g.j .
stoks , and r. schiavilla , phys .
c * 51 * , 38 ( 1995 ) .
stoks , r.a.m .
klomp , c.p.f .
terheggen , and j.j .
de swart , phys .
c * 49 * , 2950 ( 1994 ) .
r. machleidt , phys .
c * 63 * , 024001 ( 2001 ) .
bergervoet _ et al .
_ , phys . rev .
c * 41 * , 1435 ( 1990 ) .
stoks , r.a.m .
klomp , m.c.m .
rentmeester , and j.j .
de swart , phys .
c * 48 * , 792 ( 1993 ) .
przewoski _ et al .
_ , phys .
c * 58 * , 1897 ( 1998 ) .
m. viviani , r. schiavilla , and a. kievsky , phys .
c * 54 * , 534 ( 1996 ) .
r. schiavilla , v.r .
pandharipande , and d.o .
riska , phys .
c * 41 * , 309 ( 1990 ) .
r. schiavilla , v.r .
pandharipande , and d.o .
riska , phys .
c * 40 * , 2294 ( 1989 ) .
r. schiavilla , r.b .
wiringa , v.r .
pandharipande , and j. carlson , phys .
c * 45 * , 2628 ( 1992 ) .
j. carlson , d.o .
riska , r. schiavilla , and r.b .
wiringa , phys .
c * 42 * , 830 ( 1990 ) .
marcucci , d.o .
riska , and r. schiavilla , phys .
c * 58 * , 3069 ( 1998 ) .
r. schiavilla and d.o .
riska , phys .
c * 43 * , 437 ( 1991 ) . w.c .
haxton , c .- p
. liu , and m.j .
ramsey - musolf , phys .
c * 65 * , 045502 ( 2002 ) .
haxton , e.m .
henley , and m.j .
musolf , phys . rev .
lett . * 63 * , 949 ( 1989 ) .
musolf and b.r .
holstein , phys .
d * 43 * , 2956 ( 1991 ) .
riska , nucl .
a * 678 * , 79 ( 2000 ) .
maekawa and u. van kolck , phys .
* b478 * , 73 ( 2000 ) .
zhu , s.j .
puglia , b.r .
holstein , m.j .
ramsey - musolf , phys .
d * 62 * , 033008 ( 2000 ) .
marcucci , r. schiavilla , m. viviani , a. kievsky , s. rosati , and j.f .
beacom , phys .
c * 63 * , 015801 ( 2000 ) .
marcucci , r. schiavilla , s. rosati , a. kievsky , and m. viviani , phys .
c * 66 * , 054003 ( 2002 ) .
musolf , t.w .
donnelly , j. dubach , s.j .
pollock , s. kowalski , and e.j .
beise , phys . rep . * 239 * , 1 ( 1994 ) .
l. diaconescu , r. schiavilla , and u. van kolck , phys .
c * 63 * , 044007 ( 2001 ) .
ito _ et al .
_ , in preparation .
goldberger and k.m .
watson , _ collision theory _
( wiley , new york , 1964 ) . w. glckle , _ the quantum mechanical few - body problem _
( springer - verlag , berlin , 1983 ) .
j.l . friar , ann .
( n.y . ) * 104 * , 380 ( 1977 ) .
forest , phys .
c * 61 * , 034007 ( 2000 ) .
r. schiavilla , nucl .
a * 689 * , 84c ( 2001 ) .
e. fermi , _ nuclear physics _ ( the university of chicago press , chicago , 1950 ) . f.c .
michel , phys .
rev . * 133 * , b329 ( 1964 ) .
l. stodolsky , phys .
* 50b * , 352 ( 1974 ) ; phys . lett . * 96b * , 127 ( 1980 ) .
y. avishai and p. grange , j. phys .
g * 10 * , l263 ( 1984 ) .
edmonds , _ angular momentum in quantum mechanics _ ( princeton university press , princeton , 1957 ) .
m. viviani , a. kievsky , l.e .
marcucci , s. rosati , and r. schiavilla , phys .
c * 61 * , 064001 ( 2000 ) .
nollett , r.b .
wiringa , and r. schiavilla , phys .
c * 63 * , 024003 ( 2001 ) .
musolf and t.w .
donnelly , nucl .
* a546 * , 509 ( 1992 ) .
blatt and l.c .
biedenharn , phys .
rev . * 86 * , 399 ( 1952 ) .
stapp , t.j .
ypsilantis , and n. metropolis , phys .
rev . * 105 * , 302 ( 1957 ) .
m. lacombe , b. loiseau , j .-
richard , r. vinh mau , p. pireraud , and r. de toureil , phys .
* c * 21 , 861 ( 1980 ) .
d. bowman , private communication .
cox , s.a.r .
wynchank , and c.h .
collie , nucl .
74 * , 497 ( 1965 ) .
mughabghab , m. divadeenam , and n.e .
holden , _ neutron cross sections from neutron resonance parameters and thermal cross sections _ ( academic press , london , 1981 ) , http://isotopes.lbl.gov/ngdata/sig.htm .
b. desplanques , nucl .
* a242 * , 423 ( 1975 ) ; _ ibidem _ * a335 * , 147 ( 1980 ) .
mckellar , nucl .
* a254 * , 147 ( 1975 ) . b. desplanques , phys . lett .
* b512 * , 305 ( 2001 ) .
hyun , t .- s .
park , and d .-
min , phys .
* b516 * , 321 ( 2001 ) .
wiringa and s.c .
pieper , phys .
lett . * 89 * , 182501 ( 2002 ) .
marcucci , a. kievsky , s. rosati , r. schiavilla , and m. viviani , private communication .
_ , phys .
rev . * 80 * , 211 ( 1950 ) .
snell , e.c .
barker , and r.l .
sternberg , phys .
80 * , 637 ( 1950 ) .
colgate , phys .
* 83 * , 1262 ( 1951 ) .
carver _ et al .
_ , nature * 167 * , 154 ( 1951 )
. y. birenbaum , s. kahane , and r. moreh , phys .
c * 32 * , 1825 ( 1985 ) .
r. moreh , t.j .
kennett , and w.v .
prestwich , phys .
c * 39 * , 1247 ( 1989 ) .
a. de graeve _ et al .
_ , phys . rev .
c * 45 * , 860 ( 1992 ) .
danilov , phys .
lett . * 18 * , 40 ( 1965 ) .
j.l . friar and b.h.j .
mckellar , phys .
* 123b * , 284 ( 1983 ) .
e. hadjimichael and e. fischbach , phys .
d * 3 * , 755 ( 1971 ) .
lassey and b.h.j .
mckellar , phys .
c * 11 * , 349 ( 1975 ) and * 12 * , 721 ( e ) ( 1975 ) .
craver , e. fischbach , y.e .
kim , and a. tubis , phys .
d * 13 * , 1376 ( 1976 ) . h.c .
lee , phys .
lett . * 41 * , 843 ( 1978 ) .
henley , ann .
* 19 * , 367 ( 1969 )
. t. oka , phys .
d * 27 * , 523 ( 1983 ) .
t. hamada and i.d .
johnston , nucl .
phys . * 34 * , 382 ( 1962 ) .
knyazkov _ et al . _ , nucl .
* a417 * , 209 ( 1984 ) . v.m .
lobashov _ et al .
_ , nucl .
* a197 * , 241 ( 1972 ) .
beck and r.d .
mckeown , ann .
nucl . part .
* 51 * , 189 ( 2001 ) .
p. hwang and e.m .
henley , ann .
( n.y . ) * 129 * , 47 ( 1980 ) . w .-
p. hwang , e.m .
henley , and g.a .
miller , ann .
( n.y . ) * 137 * , 378 ( 1981 ) .
liu , g. prezeau , and m.j .
ramsey - musolf , phys .
c * 67 * , 035502 ( 2003 ) .
snow , private communication .
cdddddd & @xmath464 & @xmath41 & @xmath465 & @xmath466 & @xmath467 & @xmath42 ( gev / c ) + @xmath6 & 13.9 & & & 4.56 & & 1.72 + @xmath7 & 0.84 & 6.1 & 16.4 & 2.77 & 13.7 & 1.31 + @xmath8 & 20 . & 0 . & 3.23 & 1.94 & & 1.50 cddddddd & @xmath464 & @xmath41 & @xmath465 & @xmath466 & @xmath467 & @xmath42 ( gev / c ) + @xmath6 & 13.9 & & & 4.56 & & 2.4 + @xmath7 & 0.84 & 3.7 & 11.4 & 0.19 & 9.5 & 2.4 + @xmath8 & 20 . & 0 . &
1.90 & 1.14 & & 2.4 cdddd & & + & & av8 ( pc ) currents & ddh@xmath6 ( pv ) currents & total + impulse & 226.4 & 17.7 & & 17.7 + @xmath468 & 239.2 & 57.9 & 51.3 & 6.60 + @xmath469 & 241.7 & 50.3 & 51.1 & + 0.790 + @xmath471 & 241.7 & 57.0 & 51.1 & 5.89 + @xmath472 & 247.4 & 56.3 & 50.5 & 5.82 + siegert @xmath9 & & & & 5.76 + | we investigate parity - violating observables in the @xmath0@xmath1 system , including the longitudinal asymmetry and neutron - spin rotation in @xmath2@xmath1 elastic scattering , the photon asymmetry in @xmath2@xmath1 radiative capture , and the asymmetries in deuteron photo - disintegration @xmath3 in the threshold region and electro - disintegration @xmath4 in quasi - elastic kinematics . to have an estimate of the model dependence for the various predictions , a number of different , latest - generation strong - interaction potentials
argonne @xmath5 , bonn 2000 , and nijmegen i are used in combination with a weak - interaction potential consisting of @xmath6- , @xmath7- , and @xmath8-meson exchanges the model known as ddh .
the complete bound and scattering problems in the presence of parity - conserving , including electromagnetic , and parity - violating potentials is solved in both configuration and momentum space .
the issue of electromagnetic current conservation is examined carefully .
we find large cancellations between the asymmetries induced by the parity - violating interactions and those arising from the associated pion - exchange currents . in the @xmath2@xmath1 capture ,
the model dependence is nevertheless quite small , because of constraints arising through the siegert evaluation of the relevant @xmath9 matrix elements . in
quasi - elastic electron scattering these processes are found to be insignificant compared to the asymmetry produced by @xmath10-@xmath11 interference on individual nucleons .
these two experiments , then , provide clean probes of different aspects of weak - interaction physics associated with parity violation in the @xmath0@xmath1 system . finally , we find that the neutron spin rotation in @xmath12@xmath1 elastic scattering and asymmetry in deuteron disintegration by circularly - polarized photons exhibit significant sensitivity both to the values used for the weak vector - meson couplings in the ddh model and to the input strong - interaction potential adopted in the calculation . this reinforces the conclusion that these short - ranged meson couplings are not in themselves physical observables , rather the parity - violating mixings are the physically relevant parameters . |
consider the following four fundamental assumptions : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ \(1 ) a hilbert state - vector describes the real condition and modes of being of a physical entity , and not our beliefs about them ; \(2 ) the reduction of a state - vector during a measurement is an objective process of the change of state , and not a subjective process of the change of belief ; \(3 ) the experimenter s consciousness does not play any causal role in a measurement process ; \(4 ) a microscopic entity , like an electron , is a non - spatial entity which can only be spatialized when interacting with a measurement apparatus , or when incorporated into a macroscopic object .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ as far as we know , only two interpretations of quantum mechanics take the above four assumptions seriously : kastner s possibilist transactional interpretation ( pti ) @xcite and aerts hidden - measurement interpretation ( hmi ) @xcite .
curiously , the initial versions of these two interpretations were `` synchronously '' reported for the first time in 1985 @xcite and both were subsequently published in a physics journal in 1986 @xcite .
to say it all , the same synchronicity also holds for the so - called objective collapse theories , a first version of which was proposed by g. ghirardi , a. rimini and t. weber @xcite . however , and as far as we know , the latter only agree with the first three of the above four assumptions . in any case , in the present article we shall only be interested in establishing a parallel between pti and hmi , and more specifically to that implementation of the latter called the extended bloch representation ( ebr ) @xcite . in brief , the possibilist transactional interpretation is a time - symmetric re - interpretation of standard quantum mechanics ( sqm ) , originally proposed by john cramer @xcite and then further elaborated by ruth e. kastner @xcite , adding to it the hypothesis that transactions do not form in our @xmath1-dimensional euclidean space , but in a pre - empirical space ( an idea that , as far as we know , is not supported by cramer @xcite ) .
more precisely , pti ( as well as cramer s non - possibilist original version of the interpretation ) transposes to the quantum domain the idea of wheeler and feynman of considering also advanced waves ( travelling backward in time ) in their interpretation of classical electrodynamics .
this is done by observing that when taking the non - relativistic limit of dirac or klein - gordon equations , one is left not with a single equation , but with two distinct equations : the usual schroedinger equation , describing the evolution of hilbert space ket - vectors , and its hermitian conjugate , describing the evolution of the ( dual ) bra - linear forms .
contrary to sqm , both equations are considered to be physically relevant in pti .
more precisely , kets are associated with retarded offer waves of possibilities ( owp ) , evolving forward in time , which are emitted by those entities that , in a given context , play the role of sources ( think of an atom in an exited state ) , whereas bras are associated to advanced confirmation waves of possibilities ( cwp ) , evolving backward in time , emitted by quantum entities playing the role of absorbers , in response to the owp received from a source .
this a - temporal offer - confirmation dialogue between a source and its potential absorbers gives rise to a so - called _ incipient transaction _ , which is the possibility for an _ actualized transaction _ , a process that selects a specific outcome , in the form of an actual transfer of a given quantum of energy and momentum , detectable in our ordinary spatiotemporal theatre .
more precisely , let @xmath2 be a source encountering a given number @xmath3 of absorbers ( detectors ) , labeled by the properties @xmath4 , @xmath5 . according to the transactional ontology
, the retarded owp , @xmath6 , emitted by @xmath2 , stimulates the emission of advanced cwp , @xmath7 , @xmath5 , generated during the ( potential ) absorption of the owp by the different detectors .
this produces a feedback process between the source and the absorbers that , according to the transactional narrative , unfolds according to certain hypotheses .
forward and backward in time owp and cwp evolutions between the source and the absorbers are assumed to produce equivalent effects ( using a classical analogy of a wave actually propagating in space , one would say that the advanced wave travels across the same medium of the retarded wave , thus experiencing the same attenuation by it when propagating from the absorber to the source ) ; the amplitudes of the cwp are assumed to be directly proportional to the incoming owp ( a proportionality precisely expressed by the scalar product ) ; and @xmath2 is assumed to emit a possibly infinite collection of owp , simultaneously , in accordance with the superposition principle and the fact that it is in the very nature of possibilities to be all available at once ( every owp emitted by @xmath2 is assumed to be echoed back from all absorbers , in accordance with the fact that in experiments a source is usually confronted not with a single absorber , but with a vast collection of equivalent absorbers , which can all simultaneously generate a convenient cwp ) .
it is not the purpose of the present article to assess the plausibility of the above physical description and the associated assumptions .
certainly , until proven to the contrary , it is a possible description .
however , we observe that it is insufficient to derive the born rule . indeed , the above account only describes a process leading to a weighted set of competing possible transactions , called incipient transactions .
more precisely , these possible transactions can be associated with projection operators @xmath8 ( obtained by matching the possible final owp and cwp components ) , and the above `` dialogue '' between the source and the absorbers describes a deterministic transition from an initial pure state @xmath9 to a fully reduced mixed state : @xmath10 this , however , does nt say how a potential transaction can become an actual transaction ( and produce the so - called _ handshake _ ) .
in other terms , we have an explicit physical story in pti , about how an incipient transaction forms , but there is no explicit physical story describing the realization of a particular transaction ( outcome ) , as opposed to the many competing `` incipient '' ones .
this is why in pti an additional assumption is considered , which is the following : when there is an incipient transaction between a source @xmath2 and some absorbers @xmath4 , @xmath5 , as described by the above mixed state , the source - absorbers system can only select and actualize one of them , through a _ weighted symmetry breaking _ ( wsb ) process ; and the probability for the actualization of the @xmath11-th incipient transaction is precisely equal to @xmath12 , i.e. , is given by the born rule .
more specifically , regarding the use of the notion of symmetry breaking in the pti , kastner writes @xcite ( emphasis is ours ) : `` [ ... ] spontaneous symmetry breaking can be consistently extended to pti s account of the realized notion of one particular transaction out of several , or even many , incipient ones .
the mathematics describing the situation provides us with a set of possible states of the system , but only one of those can be realized .
the new feature appearing in the pti account is that this set of possible outcomes is weighted by the square of the probability amplitude for that outcome .
so , the proposed interpretation extends the basic principle of spontaneous symmetry breaking over a set of possible states : they certainly can not all be realized ( just as in the case of classical symmetry breaking ) , so _
the natural interpretation of the weight of a possible state is as a physical propensity , corresponding to an objective probability of the actualization of the state in question_. if we like , we can call this a `` weighted symmetry breaking '' or wsb . ''
the problem in the above description is not the idea that potential transactions would be actualized by means of a symmetry breaking process , but the assertion that such symmetry breaking would be weighted by physical propensities precisely given by the square of the probability amplitudes .
this is clearly a circular statement : if we want to explain the origin of the quantum mechanical born rule , then we can not use in our explanations the born rule itself , even if we use the term ` propensity ' instead of ` probability ' to describe the statistical weights .
we can see here a similarity between pti and _ decoherence theory _
@xcite , where the transition ( [ reduced ] ) is not described as the result of an interaction between owp and cwp , but as the effect of an immersion of the physical entity in question in a heat bath , causing the off - diagonal elements of the initial state @xmath9 , in the measurement s basis , to rapidly vanish , as the superposition gets `` diluted into the environment . ''
but in decoherence theory , as well as in pti , one still needs to explain how the pre - measurement state of the entity is ultimately collapsed , i.e. , one has to explain why the different possible transitions @xmath13 happen with exactly the probabilities @xmath14 , @xmath15 , when the same measurement is repeated many times . in other terms
, one needs to explain , _ in a non - circular way _ , why the experimental outcomes relative frequencies are precisely the probabilities predicted by the born rule .
what we are emphasizing is that pti , like decoherence theory , do not offer a possible solution to the measurement problem , as they are unable to derive the born rule .
they just provide an explanation for the transition of the initial vector ( or ray ) state to a pre - collapse fully reduced density matrix state .
in decoherence theory the latter needs to be interpreted as a classical mixture of states , if one wants to establish a connection with the born rule .
this however would be an _ ad hoc _ assumption , not defensible from the theory itself , which is why many adherents to decoherence theory are compelled to complete their scheme by resorting to a many - worlds picture ( although this does nt solve the problem of the derivation of the born rule @xcite ) .
on the other hand , in pti an `` incipient transaction state '' is not assumed to correspond to a classical mixture of actualized transactions , but to an objective condition of unstable equilibrium giving rise to a symmetry breaking process , finally producing a single outcome . considering the pti narrative ,
it is certainly natural , and in a sense plausible , to assume the existence of such a symmetry breaking , triggered by some uncontrollable fluctuations .
however , what remains unexplained is why such process would happen exactly with the weights predicted by the born rule . in other terms , if it is reasonable to assume , in the ambit of a realistic description of a quantum measurement , that the born rule describes a physical process akin to a symmetry breaking , one still needs to explain why exactly the born weights would result from it , and not others .
as we are now going to explain , if pti is unable to derive the born rule , this is because it only describes the first element of a bipartite process , when measurements are non - degenerate , or of a tripartite process , when measurements are degenerate .
but before doing so , we need to stress an important point .
when considering a symmetry breaking process , what one is implicitly assuming is that _ hidden interactions _
are at work .
let us explain in which sense they are hidden , using a simple example .
imagine a person pushing on a cylindrical stick vertically planted in the ground , along its vertical axis . by doing so ,
s / he can cause the stick to flex in an a priori unpredictable direction , breaking in this way the initial rotational symmetry and creating a direction that was nt present prior to the `` pushing experiment . '' when the stick flexes in a given direction , it does so because it is pushed in that direction .
in other terms , this happens because a specific interaction between the hand and the stick is actualized , which produces a specific outcome .
of course , when the person vertically pushes on the stick , s / he can not control all the infinitesimal fluctuations that will ultimately produce a hand - stick interaction slightly deviating from the vertical line .
this means that prior to the pushing all these deviating hand - stick interactions are only potential , as only one will in the end be actualized , in a way that can not be predicted in advance . for each flexing direction
, we have a certain number of possible and equivalent hand - stick interactions that are all able to cause the stick to bend in that direction .
thus , if we have the same probability of obtaining the different possible directions , this means that we have the same number of hand - stick interactions that are available to be actualized for each of them .
so , as the stick example illustrates , a symmetry breaking process is a process of actualization of a hidden interaction .
prior to the process , these interactions remain ` hidden ' as they are not yet actual interactions and one can not predict in advance which one will be actualized , the process of actualization being not under the control of the experimenter ( here the person pushing on the stick ) . as we will explain , the interactions possibly intervening in a quantum measurement are hidden precisely for these reasons , and in fact also for an additional one , but for the moment let us reflect on a key aspect of this description . in the stick example , as we said , one expects to obtain the same probability for each direction .
however , it is easy to imagine experimental situations where there would be a bias towards certain directions ; these are situations that we could conveniently call , in accordance with kastner , of wsb . situations of this kind could happen because the stick would not be planted in the ground in a perfectly vertical way , or because the ` way the person pushes onto it ' may favor certain directions in a systematic way .
but whatever the reasons for having certain directions being selected more frequently than others , we can certainly explain these differences by considering that the number of hidden hand - stick interactions available to be actualized is not anymore the same for each direction , and that the observed experimental probabilities precisely express this difference in the relative number of potential interactions associated with each outcome - direction .
if we want to describe the quantum probabilities as a wsb process , as suggested by kastner , we thus need to show that the born rule can be derived as a process governed by hidden - measurement interactions , i.e. , as a process where a predetermined number of hidden interactions can be associated to each possible outcome , so that the relative numbers of these interactions precisely correspond to the born rule of probabilistic assignment .
what is interesting , and remarkable , is that such hidden - interaction mechanism is actually built in the quantum formalism , which means that when we derive the born rule using the hmi and the associated ebr , we follow a methodological route very similar to that of the pti . indeed , in the same way as the pti does not posit new mathematical structures , but simply postulate a physical referent for those mathematical entities that ( according to the pti perspective ) remained uninterpreted in the standard formalism ( the advanced states appearing in the hilbert space inner products , in the complex conjugate schroedinger equation , and in the born rule ) , in the same way the hidden measurement program is implemented by interpreting the density matrix states also as a description of pure states ( and not just as classical statistical mixtures of states ) , and the hyper - surfaces generated by the outcome - states , within the generalized bloch representation , as a description of the hidden measurement interactions available in that specific measurement context .
we are now going to describe the unfolding of a measurement process according to the hmi .
we will do so by using the specific terminology of the pti .
this of course is not necessary , and alternative narratives can be also considered .
what however needs to be part of all possible narratives , is the idea that there are elements of reality ( or ` beables ' , to use bell s designation ) describable as ` hidden measurement - interactions ' , which are responsible for producing one of the outcomes ( [ reduced2 ] ) , in accordance with the born rule .
we start by explaining how states can generally be represented in the generalized bloch sphere .
let @xmath16 be a density matrix in a @xmath3-dimensional hilbert space @xmath17 .
we introduce a determination of the generators of @xmath18 ( the special unitary group of degree @xmath3 ) , which are traceless and orthogonal self - adjoint matrices @xmath19 , @xmath20 , that can be chosen to be normalized as @xmath21 .
then , one can show that there is a unique @xmath22-dimensional real vector @xmath23 such that @xmath24 can be written @xcite : @xmath25 where we have defined @xmath26 .
in other terms , we can represent states by means of real vectors in a generalized unit bloch sphere @xmath27 .
for @xmath28 , we have only 3 generators and they correspond to the well known pauli matrices , whereas for @xmath29 we have 8 generators that can be chosen to correspond to the so - called gell - mann matrices .
the @xmath28 case is special , as the @xmath1-dimensional vectors @xmath23 representing the states completely fill the @xmath1-dimensional unit bloch sphere . instead
, for @xmath30 , the @xmath22-dimensional generalized bloch sphere is not completely filled with states , but contains a convex region filled with states , inscribed in the latter , whose shape and orientation depend on the specific choice of generators . also , it is easy to show that ray - states ( one - dimensional projection operators , or pure states , in the standard quantum terminology ) are always represented by unit vectors ( @xmath31 ) , whereas density matrix states are described by vectors that are internal to the sphere ( @xmath32 )
. we will not enter into the mathematical details of this generalized bloch representation , for which we refer the interested reader to @xcite and the references cited therein .
what is important to mention here , for the purpose of our discussion , is that the if @xmath33 is a basis of @xmath17 , for instance the basis associated with the spectral family of a given observable , then to each projection @xmath34 we can associate a @xmath22-dimensional unit vectors @xmath35 in the generalized bloch sphere , such that @xmath36 , and it is possible to show that @xmath37 for @xmath38 .
this means that the @xmath3 vectors @xmath35 , representative of the outcome - states @xmath8 , form a @xmath39-dimensional simplex @xmath40 , inscribed in the generalized @xmath22-dimensional bloch sphere @xmath27 . for simplicity
, we just consider here the non - degenerate situation with three outcomes @xmath41 ) , the general situation being a straightforward generalization of it .
according to the pti narrative , incipient transaction are established through owp - cwp encounters in a pre - spatiotemporal ( pst ) realm , which here we can identify with the @xmath42-dimensional generalized bloch sphere in which the simplex generated by the three outcome - states ( an equilateral triangle ) is inscribed .
it can then be shown that the establishment of an incipient transaction , corresponding to transition ( [ reduced ] ) , corresponds to a movement of the abstract point particle representative of the initial state @xmath43 , from position @xmath23 at the surface of the sphere , to position @xmath44 on the @xmath45-dimensional simplex @xmath46 , along a path orthogonal to the latter ( see fig .
[ triangolo ] ) , with @xmath44 the vector representative of the fully reduced mixed state , i.e. , @xmath47 . in other terms
, the ( deterministic ) establishment of an incipient transaction corresponds to the abstract point particle plunging into the sphere and reaching a specific position on the measurement simplex . at this point
, we can observe that the on - simplex vector @xmath44 gives rise to a partitioning of @xmath46 into three distinct triangular subregions @xmath48 , @xmath49 and @xmath50 , as illustrated in fig .
[ triangolo ] .
subregion @xmath48 has vertexes @xmath51 , @xmath52 and @xmath44 , subregion @xmath49 has vertexes @xmath53 , @xmath52 and @xmath44 , and subregion @xmath50 has vertexes @xmath53 , @xmath53 and @xmath44 .
according to the ebr , these subregions of @xmath46 are precisely the referents for the hidden interactions responsible for the indeterministic collapse of the reduced state into one of the available outcomes .
more precisely , we can consider each point @xmath54 as the representative of a potential interaction , able to deterministically produce an outcome . which of these interactions is associated with which outcome precisely depends on the partitioning produced by the abstract point particle .
more precisely , all the hidden interactions belonging to @xmath4 are responsible for the outcome @xmath8 , represented by the unit vector @xmath55 , @xmath56 . in other terms , when considering the structure of the generalized bloch sphere , inherited by the geometry of the hilbert space , we see that its interpretation can be extended to provide an explicit description of the wsb mechanism that was only postulated ( but not explained ) in the pti .
it is important to emphasize that we are not resorting here to some _ ad hoc _ mathematical structures to describe the indeterministic collapse , which would be absent from quantum theory : we are just interpreting structures already appearing in the theory , which traditionally have not received any specific interpretation .
this way of proceeding should be particularly appealing to the proponents of ti and pti , as this is precisely how they also proceeded , when postulating a physical referent for the advanced states appearing in the hilbert space inner products .
now , is our interpretation of the points in the simplex , as referents for the interactions producing the different outcomes , sufficient to derive the born rule , as a wsb process ?
the answer is affirmative . indeed , if all points in @xmath46 correspond to potential interactions that can be equally actualized during the mesurement , and if those belonging to subregion @xmath4 are precisely those responsible for the outcome @xmath55 , then the probability @xmath57 for the transition ( [ reduced2 ] ) has to be given by the ratio : @xmath58 where @xmath59 denotes the lebesgue measure , and @xmath60 . by a straightforward calculation , it is then possible to show that the r.h.s . of ( [ born ] ) is exactly equal to @xmath61 , and that this remains true in the general situation of a measurement having an arbitrary number @xmath3 of outcomes @xcite . in other terms
, the ebr of quantum mechanics provides a natural explanation of quantum measurements as processes resulting from situations of unstable equilibrium , where hidden measurement - interactions mutually compete in the actualization a specific outcome . which one of the available interactions will each time be actualized depends of course on the fluctuations that are inevitably part of the measurement context , but the more interactions are available to produce a given outcome and the more probable such outcome will be .
this is precisely expressed by the ratio ( [ born ] ) , which is perfectly equivalent to the born rule , and this is also the missing description of the wsb process postulated in the pti .
a few remarks are in order .
first of all , we refer the interested reader to @xcite , to take due knowledge of all the mathematical aspects of the ebr . in these references , a mechanical - like picture is also given , where the measurement simplex is described as if it was an elastic and disintegrable hyper - membrane , which by contracting is able to bring the abstract point particle to one of the vertices of the simplex , in accordance with the projection postulate .
more precisely , one can think of the potentiality region of the simplex as a uniform elastic substance which can attract the point particle , causing it to stick onto it , at point @xmath44 .
the instability of the substance will then produce its disintegration , starting from an unpredictable point , belonging to one of the three subregions @xmath48 , @xmath49 or @xmath50 , delineated by the line segments connecting the particle s position to the vertex points . in this description
one has to think of these line segments as `` tension lines '' altering the functioning of the membrane , in the sense of making it less easy to disintegrate along them .
the disintegration can then only propagate inside the specific subregion where it starts , causing the two associated anchor points to tear away , producing the detachment of the membrane , which being elastic then contracts towards the only remaining anchor point , drawing to that position also the point particle attached to it , corresponding to the outcome of the measurement ( see fig . [ collapse ] ) .
considering the pti account , one can of course interpret this abstract membrane s collapse as a description of the absorption of the entity s owp by one of the @xmath3 available detectors .
but of course , the details of how this process truly unfolds , behind the spatiotemporal scenes , to produce an actualized transaction , or more simply an outcome state , are not so important . what is important is that the hmi provides a very general mechanism for the description of a measurement , which can be used to derive the born rule when the set of states is hilbertian ( or even more general rules of probabilistic assignment when the set of states is non - hilbertian ;
see for instance @xcite ) and is able to explain the indeterministic collapse of a fully reduced density matrix state to an outcome state as an objective wsb process .
the description of more general measurement situations , when certain outcomes are experimentally undistinguishable , is also possible within the hmi and the associated ebr . for this ,
one simply has to `` fuse '' together the subregions associated with the same degenerate outcome .
this will alter the collapse of the membrane and the point particle will not anymore necessarily be drawn to a vertex point of the simplex , but can also be drawn to one of its sub - simplexes .
a final purification - like process then needs to be considered in this case , to bring the collapsed state back to the surface of the sphere , in accordance with the the lders - von neumann projection formula ( see @xcite for the details ) .
it is interesting to note that in the same way an incipient transaction is considered in pti to be a real physical condition of interaction between a source and the absorbers , in the ebr a density matrix state is also considered to be a a condition truly describing the reality of the measured entity during the measurement process .
therefore , both the pti and the ebr accounts suggest that the notion of pure state ( intended here as a notion describing an objective , and not subjective , condition of the entity under consideration ) should be extended to also include density matrices . by the way , the latter would not only be the missing states without which it is impossible to properly describe a measurement , but also those without which it is impossible to understand , without paradoxes , the very notion of entanglement ( see the analysis in @xcite ) .
we have previously mentioned that the hidden interactions responsible for a wsb process in a quantum measurement are `` more hidden '' than those intervening in an experiment like the one with the cylindrical stick planted in the ground .
this is so because these interactions are not only potentials , at the start of the measurement , but are also _ non - spatial _ , or more generally non - spatiotemporal , exactly as in the pti account incipient transactions are also considered to be dynamical events taking place in a pre - spatiotemporal ( pst ) realm . in that respect , it is worth observing that the passage from cramer s `` standard '' ti to kastner s `` extended '' pti was operated by the latter author because of difficulties the ti narrative faced , like the interpretation of multiparticle offer and confirmation waves , the possibility of causal loop leading either to inconsistency or inconclusive quantitative predictions ( see @xcite , pp- 199 - 200 , and @xcite ) , and of course the nature of the wsb process leading to an actualized transaction @xcite . to solve these difficulties kastner
proposed that offer and confirmation waves should not be considered as spatial waves living in the three - dimensional euclidean space , but as dynamically efficacious possibilities forming a higher - dimensional space , indicated as `` pre - spacetime '' ( see however @xcite , for cramer s new approach regarding these dimensionality questions , which according to him would make the retreat to higher - dimensions unnecessary ) .
it is important to emphasize that the hypothesis of a non - spatial realm describing the reality of quantum entities and their interactions is a inescapable hypothesis if one considers that : ( 1 ) quantum measurements are not just discovery processes , i.e. , observations of properties that are already actual prior to the observation , but also , in part , creation processes , provoking a real change of the measured entities ; ( 2 ) the states of quantum entities are descriptions of their objective modes of being .
this means that before a measurement of , say , the position of an electron prepared in a superposition state ( with resepct to the position observable ) , one can not affirm that the electron was in some unknown place prior to its localization . as a consequence , an electron ( and in general all microscopic physical entities )
are to be considered fundamentally non - spatial ( and only in that sense non - local ) entities .
this means that the wave function describing an electron can not be interpreted as a physical wave that would be present in space , but as a description of the availability of the electron in being spatialized in the different possible regions ( with the born rule predicting with certainty the different localization probabilities , which therefore are to be interpreted as elements of reality in the einstein - podolsky - rosen sense ) .
and the same also holds of course for the other potential properties of a microscopic quantum entity , like ( linear and angular ) momentum .
again , an electron will generally not possess a well - defined momentum before its measurement , as the latter will only be partly created by the measurement itself .
what we are stressing here is that the notion of _ non - spatiality _ , i.e. , of a pre - empirical pst layer of our reality , as kastner calls it , is in fact implicit in any realistic approach to quantum mechanics that would take seriously its formalism ( see for instance the discussion in @xcite ) .
this means that the fourth assumption mentioned in the beginning of this note actually follows from the first two , and this is the reason why the key notion of non - spatiality was introduced by one of us as early as the late eighties @xcite and discussed since then in a number of works , in what has been called the _ creation - discovery view _ ; see @xcite and the references cited therein , and the discussions in @xcite .
to conclude , our attempt in the present note was to frame part of the pti ontology in the ambit of the hmi ontology , of the associated creation - discovery view and ebr . regarding the latter , for what concerns the description of the initial preparatory stage of a measurement , during which the abstract point particle plunges into the generalized bloch sphere to reach the measurement simplex , we emphasize that the ebr does not prescribe any specific physical mechanism for producing such transition .
therefore , one can adopt here the pti narrative , or that of decoherence theory , or any other physical account able to explain how a vector state , in a measurement context , can deterministically transform into a fully reduced mixed state , with the latter being interpreted not as a mixture , but as the real pre - collapse condition of the measured entity in that specific experimental context .
but then , a wsb will have to take place , and for its description one can not avoid introducing the key notion of hidden measurement - interactions , if one wants to explain not only qualitatively but also quantitatively the emergence of the quantum ( born ) weights .
again , the ebr is agnostic regarding the physical origin of these hidden interactions : all we know for the time being is that they naturally emerge from the quantum formalism , when state vectors are `` squared '' and represented in the generalized bloch sphere .
the hmi certainly reinforce the idea that quantum measurements should be interpreted as wsb processes , similarly to experimental situations occurring in classical physics , like in our example of the flexing of a stick .
indeed , the quantum behavior of an entity would not be so much a consequence of its microscopic or macroscopic nature , but of the way one decides to actively experiment with it , by means of specific protocols .
this is why it is possible to replicate all sorts of quantum effects also by acting on macroscopic entities working at room temperature @xcite , and that the quantum formalism has been so successfully applied in the past decade to also model human cognition processes ( see @xcite and the references cited therein ) .
but , as we said already , the dramatic difference between microscopic and macroscopic entities is that only the latter can be attached a permanent presence in our spatiotemporal theater , whereas the former are only `` dragged '' into it because of their interaction with a detection system or , to use the pti jargon , because a pre - spatiotemporal incipient transaction is actualized , giving rise to an observable phenomena in the laboratory .
aerts , d. : the entity and modern physics : the creation discovery view of reality . in : e. castellani
interpreting bodies : classical and quantum objects in modern physics , pp .
princeton , princeton unversity press ( 1998 ) aerts , d. : the stuff the world is made of : physics and reality . in : aerts d. , broekaert , j. and mathijs , e. ( eds . )
einstein meets magritte : an interdisciplinary reflection , pp . 129183 .
dordrecht , springer netherlands ( 1999 ) aerts , d. : a possible explanation for the probabilities of quantum mechanics and a macroscopical situation that violates bell inequalities . in : mittelstaedt ,
p. and stachow , e. w. ( eds . ) recent developments in quantum logic ; grundlagen der exacten naturwissenschaften ; vol.6 , wissenschaftverlag , pp . 235251 .
mannheim , bibliographisches
institut ( 1985 ) ghirardi , g. c. , rimini , a. , and weber , t. : a model for a unified quantum description of macroscopic and microscopic systems . in : accardi ,
l. et al ( eds . ) quantum probability and applications .
springer , berlin ( 1985 ) aerts , d. and sassoli de bianchi , m. : a possible solution to the second entanglement paradox . to appear in : aerts , d. , de ronde , c. , freytes , h. and giuntini , r. ( eds . ) quantum mechanics and quantum information . physical , philosophical and logical approaches .
world scientific publishing company , singapore ( 2016 ) .
arxiv:1502.06249 [ quant - ph ] aerts , d. and sassoli de bianchi , m. : the gtr - model : a universal framework for quantum - like measurements . to appear in : aerts , d. , de ronde , c. , freytes , h. and giuntini , r. ( eds . ) quantum mechanics and quantum information . physical , philosophical and logical approaches .
world scientific publishing company , singapore ( 2016 ) .
arxiv:1512.00880 [ quant - ph ] aerts , d. : an attempt to imagine parts of the reality of the micro - world .
pp . 325 ,
in : mizerski , j. et al ( eds . ) problems in quantum physics ii ; gdansk 89 .
world scientific publishing company , singapore ( 1990 ) | the purpose of the present note is twofold .
firstly , we highlight the similarities between the ontologies of kastner s possibilist transactional interpretation ( pti ) of quantum mechanics an extension of cramer s transactional interpretation and the authors hidden - measurement interpretation ( hmi ) .
secondly , we observe that although a weighted symmetry breaking ( wsb ) process was proposed in the pti , to explain the actualization of incipient transactions , no specific mechanism was actually provided to explain why the weights of such symmetry breaking are precisely those given by the born rule . in other terms , pti , similarly to decoherence theory , does nt explain a quantum measurement in a complete way , but just the transition from a pure state to a fully reduced density matrix state . on the other hand , the recently derived extended bloch representation ( ebr ) a specific implementation the hmi precisely provides such missing piece of explanation , i.e. , a qualitative description of the wsb as a process of actualization of hidden measurement - interactions and , more importantly , a quantitative prediction of the values of the associated weights that is compatible with the born rule of probabilistic assignment .
therefore , from the pti viewpoint , the ebr provide the missing link for a complete description of a quantum measurement .
however , ebr is in a sense more general than pti , as it does not rely on the specific notion of transaction , and therefore remains compatible with other physical mechanisms that could be at the origin of the measurement - interactions .
+ * keywords * possibilist transactional interpretation @xmath0 hidden - measurement interpretation @xmath0 extended bloch representation @xmath0 weighted symmetry breaking @xmath0 born rule @xmath0 non - spatiality |
in this work we look for lifshitz tails behaviour of the interband light absorption coefficient ( ilac ) defined in equation ( eqnl4 ) .
the standard definition of the ilac involves considering a pair of operators of the form @xmath0 , with @xmath1 the laplacian on either @xmath2 , in the discrete case or on @xmath3 in the continuous case , and taking a random potential @xmath4 . restricting these operators
@xmath5 to boxes @xmath6 gives operators with discrete spectra so that in any finite region of energy these operators have only finitely many eigenvalues .
using this fact one can define the quantity @xmath7 where @xmath8 are the eigen functions of the operators @xmath9 restricted to the box @xmath6 , corresponding to the eigenvalues @xmath10 respectively .
the limit of the above quantity , when it exits , gives the ilac .
we consider a correlation measure ( mentioned also in @xcite ) @xmath11 and identify the ilac as the distribution function of a marginal of the measure @xmath11 in a diagonal direction .
this identification enables us to prove theorems on the lifshitz tails behaviour of the ilac more easily since it involves only comparing the marginal of @xmath11 with the density of states of either of the operators @xmath5 .
we also do not need to approximate to define the ilac , but can obtain the function directly . in the next section
, we present an abstract version of the correlation measure @xmath11 and the density of states @xmath12 for a pair of random covariant operators and obtain relations between the two .
we start with a definition of a random family of self adjoint operators which are covariant under a group action .
[ hyp1 ] 1 .
@xmath13 is a ( separable , complex ) hilbert space , @xmath14 a probability space .
2 . there is a locally compact abelian group @xmath15 and @xmath16 is a group of unitary operators on @xmath13 , i.e. the @xmath17 are unitary and @xmath18 , @xmath19 , @xmath20 3 .
there is a discrete subgroup @xmath21 of @xmath15 and an orthogonal projection @xmath22 on @xmath13 such that @xmath23 , @xmath24 are orthogonal partitions of unity on @xmath13 .
we set @xmath25 4 .
@xmath26 is a group of probability preserving transformations on @xmath27 .
a family @xmath28 of self - adjoint operators on @xmath13 is called measurable if the family @xmath29 is measurable it is known ( see @xcite , @xcite and section 2.4 of @xcite ) that a family of _ bounded _ self - adjoint operators is measurable iff it s weakly measurable .
moreover , if @xmath30 is a measurable family of self - adjoint operators then for any bounded measurable function @xmath31 the operator family @xmath32 is weakly measurable .
( also in @xcite , @xcite , section 2.4 @xcite ) . finally , the product of weakly measurable families is weakly measurable ( see @xcite ) .
a weakly measurable family @xmath33 of bounded operators is called _ covariant _ ( with respect to @xmath34 ) if @xmath35 also , a measurable family @xmath33 of self adjoint operators is called _ covariant _ ( with respect to @xmath34 ) if @xmath35 if @xmath33 is a covariant family of self - adjoint operators and @xmath31 is a bounded measurable function , then the family @xmath32 is covariant ( also in @xcite , @xcite ) .
moreover , if both @xmath33 and @xmath36 are covariant families of bounded operators , then @xmath37 is a covariant family .
we denote by @xmath38 the trace norm of a trace class operator @xmath39 .
[ propl2 ] let @xmath33 and @xmath36 be covariant families of bounded operators and assume that @xmath40 and @xmath41 are trace class and @xmath42 then : @xmath43 @xmath44 in the last step we used the fact that @xmath45 is a partition of unity also .
now we take expectations of either side of the above equation and obtain @xmath46 [ corl1 ] 1 .
if @xmath47 are covariant families of bounded operators satisfying the condition ( [ eqnl5 ] ) then : @xmath48 2 .
if @xmath49 are covariant families of bounded , positive ( i.e. @xmath50 ) operators satisfying the conditions ( [ eqnl5 ] ) then @xmath51 the first assertion is clear as we can apply the proposition to the covariant families @xmath52 and @xmath53 . for the second claim
we observe that @xmath54 with a @xmath55 .
@xmath53 as a function of the covariant family @xmath36 is covariant as well . moreover , since @xmath33 is positive ( and the hilbert space is complex ) , @xmath33 is self - adjoint and so is @xmath53 . by part ( i ) of the corollary we have : @xmath56 [ hypl2 ]
let @xmath57 be family of self - adjoint operators , which are bounded below , on a hilbert space @xmath58 .
let @xmath59 be the ( projection - valued ) spectral measure of @xmath57 such that for any bounded borel set @xmath60 , the operators @xmath61 are trace class for a.e .
@xmath62 and form a * covariant * family of operators . for operators
@xmath57 satisfying the above hypothesis , it is clear that for any finite @xmath63 , the spectral measure @xmath64 ) = e_{_{h_\omega } } ( [ c , x ] ) $ ] , with @xmath65 finite and smaller than the infimum of the spectrum of @xmath57 .
therefore the hypothesis implies that for any finite @xmath63 , the operators @xmath66 ) , e_{_{h_\omega } } ( ( -\infty , x ] ) p$ ] are trace class .
therefore we can now define the density of states for such operators .
[ defl0 ] let @xmath57 be a family of self adjoint operators satisfying hypothesis [ hypl2 ] .
then the _ density of states _ of this family is defined to be the unique @xmath67-finite measure @xmath12 associated with the monotone right continuous function @xmath68 , @xmath69 ) p ) \right),\ ] ] via @xmath70 ) = f(b ) - f(a ) , ~ a , b \in { \mathbb{r}}. $ ] thus for any bounded borel set @xmath60 , @xmath71 agrees with the right hand side of the above relation with @xmath60 replacing @xmath72 $ ] . in the above framework
we define another measure that is used to define the interband light absorption coefficient ( ilac ) . to do this we need a pair @xmath5 of self - adjoint operators as in the hypothesis [ hypl2 ] and consider the associated projection valued measures @xmath73 .
we then define the density of states of these operators by , @xmath74 consider the semi algebra @xmath75 of subsets of @xmath76 where @xmath77 : a , b \in
{ \mathbb r}\ } \cup \{(a , \infty ) : a \in { \mathbb r}\ } \cup \{(-\infty , a ] : a \in { \mathbb r}\}.\ ] ] we define the correlation measure @xmath11 on @xmath75 as @xmath78 where @xmath11 is set to be @xmath79 if either @xmath60 or @xmath39 is an unbounded element of @xmath80 . this set function takes values in @xmath81 $ ] if @xmath22 is trace class and in @xmath82 $ ] , if @xmath83 ) } $ ] are trace class only for bounded intervals @xmath84 $ ] , in view of proposition [ propl1 ] .
we set @xmath85 it is a simple exercise to see that this @xmath11 is well defined on @xmath75 and via standard measure theory extends as a @xmath67-finite measure to the whole borel @xmath67-algebra of @xmath76 .. using the hypothesis 2 , and proposition [ propl2 ] we see that the following is valid .
[ propl1 ] consider the operators @xmath5 satisfying hypothesis [ hypl2 ] and let @xmath86 and @xmath11 be as in equation ( [ eqnl3 ] ) .
then for any @xmath87 bounded , 1 .
@xmath88 2 .
@xmath89 3 .
the following inequalities are valid @xmath90 since the subsets @xmath91 are bounded the operators @xmath92 are covariant trace class operators satisfying the inequality ( [ eqnl5 ] )
. therefore the result follows by an application of proposition [ propl2 ] and corollary [ corl1 ] .
we collect the arguments about @xmath11 in a proposition .
[ propl3 ] consider a pair of covariant operators @xmath5 satisfying the hypothesis [ hypl2 ] and consider the correlation measures @xmath11 extended to the borel @xmath67-algebra on @xmath76 from that given by equation ( [ eqnl3 ] )
. then the following are valid . 1 . if @xmath22 is trace class , then @xmath11 is a probability measure on @xmath76 , with support in the closure of @xmath93 .
2 . if @xmath22 is not trace class but , @xmath94 ) p$ ] is trace class , for bounded intervals @xmath95 $ ] , then @xmath11 is a positive @xmath67-finite measure on @xmath76 , with support in the closure of @xmath93 .
[ reml1 ] typically the first case occurs for operators on @xmath2 and the second case occurs in @xmath3 .
we take the transformation @xmath96 on @xmath97 given by @xmath98 using this @xmath96 we define the interband light absorption coefficient ( ilac ) @xmath60 as the distribution function , @xmath99\right ) , ~ \mathrm{where } ~ \nu(b ) = \rho \circ t^{-1 } ( b\times { \mathbb r})\ ] ] in the above equation the factor @xmath100 is because of the normalisation we used for @xmath96 , so that this definition of ilac agrees with the standard one in the case of finite box operators .
we also note that since the operators @xmath5 are assumed to be bounded below @xmath101 . in the case
when @xmath102 in hypothesis [ hyp1 ] is ergodic with respect to the action of @xmath15 on @xmath27 , then , the spectra @xmath103 of covariant families of operators @xmath5 are almost everywhere constant sets . in such a case
we can talk about the infimum of spectra of @xmath5 without reference to @xmath62 .
in this context we have the following theorem . [ thml1 ]
suppose @xmath5 are a pair of random families of self - adjoint operators satisfying hypothesis [ hyp1 ] .
assume further that @xmath102 is ergodic with respect to the action of @xmath15 on @xmath27 . 1 .
. then @xmath105 .
2 . let @xmath106 .
then @xmath107 .
we shall prove the first case , the other proof is similar ( where one has to use the fact that @xmath108 respectively for the other case and work it out ) .
let @xmath109 to be the infima of the spectra @xmath110 of @xmath111 .
we consider the closure of the cartesian product @xmath112 of the spectra of @xmath5 , which is the support of the measure @xmath11 .
therefore if we denotes points of @xmath113 by @xmath114 , so that @xmath115 , then the possible values of @xmath116 have a lower bound @xmath117 , so @xmath118 implies @xmath119 , ( see figure 2 ) .
this immediately implies the inclusions ( the first inclusion is clear and the second one uses the above ) : @xmath120 this then would lead to the inequalities that @xmath121 \times { \mathbb{r } } ) \\ & = \rho \left(\{(\lambda_1 , \lambda_2 ) : e_-+e_+ \leq \lambda_1 + \lambda_2 \leq e_-+e_++a\ } \right ) \\ & \leq \rho \left ( ( e_- , e_-+2a ) \times ( e_+ , e_+ + 2a ) \right ) \\ & \leq min
\ { \rho \left ( ( e_- , e_-+2a ) \times { \mathbb r}\right ) , \rho \left ( ( e_+ , e_++2a ) \times { \mathbb r}\right ) \ } \\ & \leq min \ { n_-\left ( ( e_- -2a , e_-+ 2a ) \right),n_+\left ( ( e_+ -2a , e_++ 2a ) \right ) \},\end{aligned}\ ] ] where the last inequality comes from proposition [ propl1](3 ) and enlarging the intervals slightly , which only increases the bound since @xmath86 are measures . if the density of states @xmath86 have lifshitz tails behaviour @xmath122 as @xmath123 goes to zero , for an appropriate @xmath124 , at @xmath125 respectively , then we have @xmath126 for @xmath127 for some @xmath124 , so , using the above inequalities , @xmath128 in the case when the density of states @xmath86 have lifshitz tails behaviour at other internal band edges , the same behaviour is valid for ilac under some conditions .
suppose the spectra of @xmath5 consist of bands @xmath129 $ ] .
then the product of the spectra is @xmath130 \times [ a_j^- , b_j^-]$ ] .
let us denote @xmath131 \times [ a_j^- , b_j^-]$ ] .
then , the measure @xmath11 is supported on the set @xmath132 .
we index the pairs @xmath133 by @xmath134 and use @xmath135 to denote a rectangle forming part of @xmath113 henceforth .
so we have @xmath136 .
the central point in the proof of theorem [ thml1 ] is that if @xmath137 is a corner of the rectangle formed by the lowest bands of the spectra of @xmath5 , then the strip @xmath138 intersected with the support of @xmath11 is a triangle of side length @xmath139 , ( see figure 2 ) , hence its @xmath11 measure is smaller than that of the square with the corner @xmath137 and side length @xmath140 , as can be seen in the figure 2 .
as we see in figure 1 , there may be some rectangles in the support of @xmath11 , with this property .
those rectangles in figure 1 , where this is not true are marked by @xmath141 and the solid lines are those lines @xmath142 for which this feature is valid and the dashed lines are those for which this is not true . in the definition below the sets @xmath143 and we denote the coordinates of @xmath97 by @xmath114 .
[ defl1 ] let the support of @xmath11 be @xmath136 , with @xmath144 \times [ a_j^- , b_j^- ] , ~ \beta = ( ij)$ ]
. then we call a corner @xmath137 of a rectangle @xmath135 * good * , if the intersection of the line @xmath145 with @xmath113 consists of finitely many points and all of them are corners of rectangles forming @xmath113 .
given a corner @xmath137 in @xmath113 we shall denote by @xmath146 the set of corners that lie on the line @xmath145 .
[ thml2 ] let spectra of @xmath5 be as in theorem [ thml1 ] and let @xmath113 be the support of the measure @xmath11 given in equation [ eqnl3 ] .
let @xmath60 , as given in equation ( [ eqnl4 ] ) be the corresponding ilac . if @xmath137 is a good corner in @xmath113 .
denote the elements of @xmath146 by @xmath147 .
then we have @xmath148 firstly we note that if we take a rectangle , @xmath135 , then only the lower - left and the top - right corners are candidates of being _ good _ corners , since for the other two corners , the line @xmath149 that contains the said corner will pass through the rectangle and hence has infinitely many points
. we will prove the theorem for a good corner @xmath137 which is a lower left corner of a rectangle , the proof for the case of a top - right good corner is similar . in this case
we see immediately that if @xmath137 is a good corner in @xmath113 , then the intersection of the strip @xmath150 with @xmath113 is contained in finitely many rectangles @xmath135 forming @xmath113 .
further @xmath151 is contained in a square of side length @xmath140 contained in @xmath135 and having one corner common with a corner of @xmath135 . given a good corner @xmath137 and the associated strip @xmath152 , let @xmath153 denote the corner of rectangle @xmath154 that has nonempty intersection with it .
( note that this corner satisfies @xmath155 ) .
then whenever @xmath137 is a good corner we have the inequality , with @xmath156 ranging over a finite set , @xmath157\times [ d_\gamma , d_\gamma + 2a ] .
\end{aligned}\ ] ] this inequality implies immediately that : @xmath158 where in the last inequality we enlarged the sets using the fact that @xmath86 are measures .
this shows that at ilac has the same continuity property as the density of states at the band edges . in the theorem below we
identify good corners for a simple case of spectra having two bands .
[ thml3 ] consider a pair of self adjoint operators @xmath5 as in theorem [ thml1 ] .
suppose a.e .
@xmath62 , the spectra of @xmath111 are given by @xmath159 $ ] and @xmath160 $ ] , respectively , where @xmath161 are listed in the increasing order .
then the corners @xmath162 are _ good _ whenever @xmath163 satisfy , @xmath164 in the case @xmath165 , even the corners @xmath166 are good .
this is direct verification to see that the diagonal lines @xmath149 passing through the respective corners do not intersect any other rectangle . in the latter case
when the spectra are the same , we have @xmath167 and @xmath168 , hence the stated result . in the symmetric case @xmath169 , however , the rectangles @xmath170 if @xmath171 . in this case
the above assumption still ensures that the lower - left and top - right corners of the rectangles are good .
consider @xmath172 and the discrete laplacian @xmath173 .
consider real valued i.i.d random variables @xmath174 with common distribution @xmath175 .
let @xmath176 denote the operator of multiplication by the sequence @xmath177 .
consider the operators @xmath178 taking @xmath179 , it is known that operators @xmath180 are covariant . the projection @xmath22 is taken to be the projection @xmath181 onto the subspace generated by the vector @xmath182 , which is an element of the standard basis for @xmath172 .
then the density of states in these models are given by @xmath183 and the correlation measure @xmath11 is given by @xmath184 and is a probability measure as per proposition [ propl3 ] ( 1 ) , since @xmath22 is trace class in this case .
in this model the density of states of @xmath5 are shown to have lifshitz tails behaviour at the bottom of the spectra @xcite , under the condition that @xmath175 satisfies @xmath185 , where @xmath123 is the infimum of the support of @xmath175 . in the case when the support of @xmath175 has two closed intervals @xmath186 \cup [ a_2 , b_2]$],(@xmath187 arranged in an increasing order so that @xmath188 for all @xmath189 ) and such that @xmath190 , simon @xcite proves the lifshitz tails behaviour at the internal band edges , if @xmath175 satisfies @xmath191 and @xmath192 for all @xmath189 .
when @xmath193 $ ] and @xmath194 $ ] are disjoint , lifshitz tails behaviour at the band edges is also shown for the associated density of states @xmath12 .
that is at any of the band edges one has @xmath195 as @xmath196 .
an application of theorems [ thml1 ] , [ thml2 ] shows that the results are true for the ilac @xmath60 , namely [ thml4 ] consider the anderson models as above on @xmath2 .
if @xmath197 \cup [ a_2^\pm , b_2^\pm]$ ] , are @xmath198 .
then , for some @xmath199 , * ( external band edge case ) for @xmath200 , one has @xmath201 * ( internal band edge case ) if the gap between the intervals @xmath202 $ ] and @xmath203 $ ] is large enough , then @xmath204 for any @xmath205 .
we gave a simple example of a discrete model , however there are many more , those with periodic backgrounds @xcite , those which are unbounded @xcite and so on .
we refer to the review @xcite for the numerous cases where the lifshitz tails for the density of states is proved and for which our theorem applies to the ilac .
let us start by stating a theorem which is essentially a very weak version of the uncertainty principle .
we start with a couple of lemmas .
we recall the definition of the trace ideal @xmath208 to be those bounded operators @xmath60 with the property @xmath209 is trace class . recall that elements of @xmath210 are called trace class operators .
let @xmath217 be an operator of multiplication by a function @xmath218 on @xmath3 on its natural domain and such that @xmath217 is bounded with respect to @xmath219 having relative bound smaller than @xmath220 .
this means the operator @xmath221 is bounded .
then @xmath222 is also self adjoint ( kato - rellich theorem ) on the domain of @xmath219 and its spectrum is also bounded below .
writing @xmath223 , we see that @xmath224 is also bounded .
let @xmath22 denote the operator of multiplication by the indicator function @xmath225 of a bounded region @xmath226 on @xmath3 .
let @xmath227 denote the spectral measure of a bounded borel set @xmath60 , with respect to the ( projection valued ) spectral measure of @xmath228 .
then , [ thmla1 ] consider @xmath3 and the operator @xmath229 .
let @xmath217 be an operator of multiplication by a function @xmath218 , such that @xmath217 is relatively bounded w.r.t . @xmath219 with relative bound @xmath230 and consider @xmath231 .
suppose either 1 .
@xmath232 , then @xmath233 and @xmath234 are hilbert - schmidt , so @xmath235 is trace class for any bounded borel set @xmath60 .
2 . suppose @xmath236 and suppose @xmath217 is bounded or @xmath237 are relatively bounded w.r.t @xmath219
. then @xmath233 and @xmath238 are trace class .
\(1 ) writing @xmath239 , we see that since all the factors are bounded , it is enough to show that @xmath240 is hilbert - schmidt
. the operator @xmath241 is multiplication by @xmath242 after taking fourier transforms and hence is in @xmath243 .
therefore an application of lemma [ leml1 ] , shows that the product is hilbert - schmidt .
\(2 ) we will prove that @xmath244 , the proof for @xmath238 is similar . by taking a compactly supported smooth function @xmath245 which is value @xmath220 on the closure of @xmath60
, we have @xmath246 .
we will therefore show that @xmath247 is trace class for any compactly supported smooth function @xmath245 .
we also note that the function @xmath248 is again a function of the same type as @xmath245 .
first consider @xmath254 .
we write this product as @xmath255 and consider ( recalling @xmath211 ) , @xmath256 the product of the first two factors is in @xmath257 ( since @xmath258 is integrable ) , by lemma [ leml1 ] , the next two factors form a bounded operator ( which can be seen by taking fourier transforms ) .
the final two factors form a bounded operator as argued before the lemma .
therefore the entire product is in @xmath257 .
since @xmath259 is bounded also , we get that @xmath260 .
now assume that @xmath261 , and show that @xmath262 .
we write , @xmath263 , then @xmath264\psi(h ) + ( x^2+i)^{-1}(h+i)^{-1}(x^2+i)^{-n}\psi(h).\nonumber\end{aligned}\ ] ] using theorem 2.8 ( 2.5b ) in @xcite , ( which says @xmath265 with @xmath266 ) , so using the induction hypothesis and the already proved fact that @xmath267 , the last term is seen to be in @xmath268 .
so we concentrate on the first term .
@xmath269\psi(h ) \\ & = ( x^2+i)^{-1}(h+i)^{-1}[h,(x^2+i)^{-n}](h+i)^{-1}\psi(h ) \nonumber \\ & = ( x^2+i)^{-1}(h+i)^{-1}[h_0,(x^2+i)^{-n}](h+i)^{-1}\psi(h ) \nonumber \\ & = ( x^2+i)^{-1}(h+i)^{-1 } \times\nonumber \\ & \left(-4ni\sum_{j=1}^d p_j x_j(x^2+i)^{-1 } -2d(x^2+i)^{-1 } + 4d(n+1 ) x^2(x^2+i)^{-2 } \right)\nonumber \\ & \times ( x^2+i)^{-n}\psi_1(h ) \nonumber \\ \end{aligned}\ ] ] where we set @xmath270 , where @xmath271 . @xmath272 and using lemma [ leml1 ] , lemma [ leml2 ] below
, we see that this expression is in @xmath257 .
induction hypothesis gives @xmath273 . therefore combining these two facts
we see that @xmath274 .
consider the case when @xmath278 is bounded for each @xmath279 .
then writing the expression using commutators @xmath280(h+i)^{-1 } \\ & = ( h_0 + 1)^\half p_j(h+i)^{-1 } + ( h_0 + 1)^\half(h+i)^{-1 } ( \frac{\partial}{\partial x_j}v)(h+i)^{-1}.\end{aligned}\ ] ] the boundedness of the first term was seen before since @xmath281 and @xmath282 are bounded .
second term is bounded by the assumption on @xmath217 and the boundedness of @xmath283 .
now consider the case when @xmath217 is bounded , then taking @xmath284 in the domain of @xmath219 , we have @xmath285 where @xmath65 is a positive constant such that @xmath286 is a positive operator ( which is possible since @xmath228 is bounded below ) . since @xmath286 is positive it has a unique square root , so using the boundedness of @xmath217 and the above inequality , we obtain , for some finite @xmath287 , @xmath288 taking @xmath289 , for a set of @xmath290 coming from @xmath291 , we see that @xmath292 @xmath293 independent of @xmath290 .
this shows that @xmath294 has a bounded inverse and that its inverse @xmath295 and @xmath294 are both bounded ( since @xmath296 bounded implies @xmath297 is also bounded ) . therefore writing @xmath298 we see that the left hand side is bounded .
consider @xmath299 , @xmath300 , @xmath301 , i.i.d random variables with distribution @xmath175 having compact support .
let @xmath6 denote the unit cube centred at @xmath302 and @xmath303 denote the unit cube centred at the point @xmath304 .
let @xmath305 , where @xmath306 is the operator of multiplication by the indicator function of @xmath60 . then taking @xmath307 we see that , since @xmath176 is bounded for each @xmath62 , the conditions of theorem [ thmla1 ] are satisfied . further taking @xmath308 and @xmath309 , @xmath310 , @xmath311 , in hypothesis [ hyp1 ] , the spectral projections @xmath312 are covariant families in the sense of definition [ hyp1 ] .
the theorem [ thmla1 ] shows that @xmath313 is trace class whenever @xmath314 is a bounded interval .
hence we can define the density of states and the ilac as in equations ( [ eqnl00 ] ) and ( [ eqnl3 ] ) , by taking @xmath22 to be multiplication by @xmath315
. therefore the theorems [ thml1 ] , [ thml2 ] are valid in this case .
* @xmath316 , @xmath317 i.i.d.random variables whose distribution has compact support and @xmath318 a nice function with @xmath319 summable .
* an addition of a periodic background potential @xmath320 to the random potential above .
* addition of magnetic fields .
let us remark that in the above examples we can even replace the laplacian @xmath321 with a real polynomial function @xmath322 of @xmath323 and the results go through , if for some @xmath324 , the polynomial satisfies @xmath325 m.s .
atoyan , e.m .
kazaryan , h.a .
sarkisyan : _ interband light absorption in parabolic quantum dot in the presence of electrical and magnetic fields _ , physica e : low - dimensional systems and nanostructures vol 31 , 83 - 85 ( 2006 ) . w. kirsch and b. metzger : _ the integrated density of states for random schr dinger operators _ , _ spectral theory and mathematical physics : a festschrift in honor of barry simon s 60th birthday _ , proc .
sympos . pure math .
, vol 76 649696 ( 2007 ) .
w. kirsch and l.a.pastur : _ the large - time asymptotics of some wiener integrals and the interband light absorption coefficient in the deep fluctuation spectrum _ ,
, vol 132 , 365382 ( 1990 ) .
b. khoruzenko , w. kirsch and l.a.pastur : _ the interband light absorption coefficient in the weak disorder regime : an asymptotically exactly solvable model _ ,
j. phys . a : math .
vol 27 , 2527 - 2543 ( 1994 ) .
klopp , frederic : _ lifshitz tails for random perturbations of periodic schrdinger operators _ , spectral and inverse spectral theory ( goa , 2000 ) , proc .
indian acad .
vol 112 , no .
1 , 147162 ( 2002 ) . | in this paper we consider the interband light absorption coefficient for various models . we show that at the lower and upper edges of the spectrum the lifshitz tails behaviour of the density of states implies similar behaviour for the ilac at appropriate energies .
the lifshitz tails property is also exhibited at some points corresponding to the internal band edges of the density of states . |
plutonium - based materials have been extensively studied due to their interesting physical behaviors of the 5__f _ _ states and have always attracted particular attention because of their importance in nuclear fuel cycle @xcite .
recently , experimental reports @xcite on the strategies of storage of pu - based waste illustrated that metallic plutonium surface easily oxidizes to pu@xmath0o@xmath2 and puo@xmath0 in surrounding of air and moisture .
previous calculations @xcite using the hybrid density functional of ( heyd , scuseria , and enzerhof ) hse have established trends of the electronic properties of actinide dioxides ano@xmath0 ( an = th@xmath1es ) . along the series , mott insulators ( @xmath3@xmath4@xmath3 )
are obtained prior to puo@xmath0 , whereas above amo@xmath0 , charge - transfer insulators ( o 2@xmath5an 5@xmath3 ) are observed .
puo@xmath0 and amo@xmath0 lie in a crossover from localized to delocalized _ f _ electron character with increasing _ z_. they have strong 5@xmath3 - 2@xmath6 orbital energy degeneracy , which can lead to unexpected orbital mixing .
recent theoretical work performed by petit _
et al _ @xcite demonstrated that the dioxide is the most stable oxide for the actinides from np onward .
all these works indicate that investigations of puo@xmath0 have particular meaning in actinide compounds .
at ambient condition , puo@xmath0 crystallizes in a fluorite structure with space group _ fm@xmath7m_. and at 39 gpa , puo@xmath0 undergoes a phase transition to an orthorhombic structure of cotunnite type with space group _ pnma _ @xcite . in our previous work @xcite , we have systematically investigated the structural , electronic , mechanical , and optical properties of fluorite puo@xmath0 within the lda+__u _ _ and gga+__u _ _ formalisms . we found that comparing with the experimental data and the theoretical results , the accuracy of our atomic - structure prediction for antiferromagnetic ( afm ) puo@xmath0 is quite satisfactory by tuning the effective hubbard parameter _
u _ in a range of 3 - 4 ev within the lda / gga+__u _ _ approaches .
subsequently , jomard _ et al _ @xcite confirmed our results and further reported optical and thermodynamic properties of puo@xmath0 .
in 2008 , yin and savrasov @xcite successfully obtained the phonon dispersions of both uo@xmath0 and puo@xmath0 by employing the lda+dmft scheme . in 2009 , minamoto _ et al . _
@xcite investigated the thermodynamic properties of puo@xmath0 based on their calculated phonon dispersion within the pure gga scheme .
however , although many efforts have been performed on puo@xmath0 , little is known on its theoretical tensile strength @xcite .
the ideal strength of materials is the stress that is required to force deformation or fracture at the elastic instability .
although the strength of a real crystal can be changed by the existing cracks , dislocations , grain boundaries , and other microstructural features , its theoretical value can never be raised , i.e. , the theoretical strength sets an upper bound on the attainable stress . in this study , we focus our sight on the structural , electronic , mechanical , and magnetic features of puo@xmath0 under tension . the stress - strain relationships and the ideal tensile strengths are obtained by performing a first - principles computational tensile test ( fpctt ) @xcite . under tension , the bonding nature and electronic occupation characters are systematically studied .
the first - principles density - functional theory ( dft ) calculations on the basis of the frozen - core projected augmented wave method of blchl @xcite are performed within the vienna _ ab initio _ simulation package ( vasp ) @xcite , where the lda @xcite is employed to describe electron exchange and correlation . for the plane - wave set , a cutoff energy of 500 ev is used .
k_-point mesh in the full wedge of the brillouin zone ( bz ) is sampled by 9@xmath89@xmath99 grids according to the monkhorst - pack @xcite for the fluorite unit cell and all atoms are fully relaxed until the hellmann - feynman forces become less than 0.001 ev / .
the plutonium @xmath10 and the oxygen 2__s__@xmath112__p__@xmath12 electrons are treated as valence electrons .
the strong on - site coulomb repulsion among the localized pu 5__f _ _ electrons is described by using the lda+__u _ _ formalism formulated by dudarev _ et al . _ @xcite . in this paper the coulomb energy ( @xmath13 ) and the exchange energy ( @xmath14 ) are set to be constants : @xmath13=4.75 ev and @xmath14=0.75 ev .
these values of @xmath13 and @xmath14 are the same as those in our previous study of plutonium oxides @xcite . using these parameters
, the lda+__u _ _ gives @xmath15=5.362 , which is very close to the experimental value of 5.398 @xcite . and
our results reproduce all the features included in our previous work @xcite .
in particular , we recover the main conclusion that although the pure lda fail to depict the electronic structure , especially the insulating nature and the occupied - state character of puo@xmath0 , by tuning the effective hubbard parameter in a reasonable range , the lda+__u _ _ approaches can prominently improve upon the pure lda calculations and , thus , can provide a satisfactory qualitative electronic structure description comparable with the photoemission experiments @xcite .
spin - polarized calculations are performed and we find that the afm spin alignment is the most stable configuration among nonmagnetic , ferromagnetic ( fm ) , and afm configurations . the total - energy difference ( @xmath16 per formula unit at respective optimum geometries ) within the lda+__u _ _ is calculated to be 0.705 ev . in the fpctt , the stress - strain relationship and the ideal tensile strength are calculated by deforming puo@xmath0 crystal to failure .
the anisotropy of the tensile strength is tested by pulling the initial fluorite structure along the [ 001 ] , [ 100 ] , [ 110 ] , and [ 111 ] directions . as shown in fig .
[ tensile1 ] , three geometric structures are constructed to investigate the tensile strengths in the four typical crystallographic directions : [ tensile1](a ) shows a general fluorite structure of puo@xmath0 with four pu and eight o atoms ; [ tensile1](b ) a body - centered tetrahedral ( bct ) unitcell with two pu and four o ; and [ tensile1](c ) a orthorhombic unitcell with six pu and twelve o. in fpctt , the tensile stress is calculated according to the nielsen - martin scheme @xcite @xmath17 where @xmath18 is the strain tensor ( @xmath19=1,2,3 ) and @xmath20 is the volume at the given tensile strain .
tensile processes along the [ 001 ] , [ 100 ] , [ 110 ] , and [ 111 ] directions are implemented by increasing the lattice constants of these three orientations step by step . at each step ,
the structure is fully relaxed until all other five stress components vanish except that in the tensile direction .
or [ 100 ] , ( b ) [ 110 ] , and ( c ) [ 111 ] orientations . blue atoms are plutonium atoms while white atoms are oxygen atoms .
the afm order is indicated by white arrows . ] the calculated total energy , stress , and spin moments as functions of uniaxial tensile strain for afm puo@xmath0 in the [ 001 ] , [ 100 ] , [ 110 ] , and [ 111 ] directions are shown in fig .
[ tensile2 ] .
evolutions of the lattice parameters with strain in all four tensile processes are presented in fig .
[ tensile3 ]
. clearly , all four energy - strain curves increase with increasing tensile strain , but one can easily find the inflexions by performing differentiations .
actually , at strains of 0.36 , 0.36 , 0.22 , and 0.18 , the stresses reach maxima of 81.2 , 80.5 , 28.3 , and 16.8 gpa for pulling in the [ 001 ] , [ 100 ] , [ 110 ] , and [ 111 ] directions , respectively .
these results clearly indicate that the [ 001 ] direction is the strongest tensile direction and [ 111 ] the weakest . in the [ 001 ] , [ 100 ] , [ 110 ] , and [ 111 ] directions .
] in the ( a ) [ 001 ] , ( b ) [ 100 ] , ( c ) [ 110 ] , and ( d ) [ 111 ] directions . ] in fact , there are eight pu@xmath1o bonds per formula unit for fluorite puo@xmath0 .
the angle of all eight bonds with respect to the pulling direction is 45@xmath21 in [ 001 ] direction . for pulling direction of [ 100 ] ,
the bonding structure is same with that of [ 001 ] direction .
so their pressure behaviors of total energy , stress , and spin moments are almost same before strain of 0.40 .
the special behaviors of total energy , stress , and spin moments over @xmath22=0.40 for pulling in the [ 100 ] direction have tight relations with the abrupt behavior of its crystal structure .
in fact , over @xmath22=0.40 the face - centered tetragonal ( fct ) structure will transit into an orthorhombic structure , as indicated by fig .
[ tensile3](b ) .
the lattice parameter of [ 001 ] direction is shortened and [ 010 ] direction elongated by tensile deformation . however , in [ 110 ] direction only four bonds make an angle of 45@xmath21 with the pulling direction . four other bonds are vertical to the pulling direction . in [ 111 ] direction ,
two bonds are parallel to the pulling direction and six others make an angle of about 19.5@xmath21 with the pulling direction .
it is easy to understand that the bonds vertical to the pulling direction have no contributions on the tensile strength and the bonds parallel to the pulling direction are easy to fracture under tensile deformation .
therefore , the fact that the tensile strength along the [ 001 ] or [ 001 ] direction is stronger than that along [ 110 ] and [ 111 ] directions is understandable .
besides , we note that the stress in [ 110 ] direction experiences an abrupt decrease process after strain up to 0.24 .
this is due to the fact that the corresponding four pu@xmath1o bonds ( make an angle of 45@xmath21 with the pulling direction ) have been pulled to fracture .
the fracture behaviors have been clarified by plotting valence electron charge density maps ( not shown ) . under the same strain , the abrupt increase of spin moment
can be clearly seen [ fig .
[ tensile2](c ) ] .
while the spin moments in [ 110 ] and [ 111 ] directions only increase from 4.13 to 4.23 and 4.33 @xmath23 , respectively , the spin moments in [ 001 ] and [ 100 ] directions are increased up to about 5.27 and 5.11 @xmath23 , respectively , at the end of tensile deformation .
in addition , the evolutions of the lattice parameters with strain in fig . [ tensile3 ] clearly show that along with the increase of the lattice parameter in the pulling direction , other two lattice parameters vertical to the pulling direction are decreased smoothly for tensile strains along [ 001 ] , [ 110 ] , and [ 111 ] directions . in [ 001 ]
direction , the evolutions of the lattice parameters along [ 100 ] and [ 010 ] directions are absolutely same due to the structural symmetry .
for all these three tensile deformations , no structural transition has been observed in our present fpctt study . the structural transition over @xmath22=0.40 for [ 001 ]
direction tensile deformation is mainly due to the magnetic structure . in ( 01@xmath24 )
plane under strains of ( a ) 0.0 and ( b ) 0.5 along [ 001 ] direction tensile strain .
the contour lines are drawn from 0.0 to 1.5 at 0.1 e / @xmath25 intervals . ]
o bonds in ( 01@xmath24 ) plane , ( b ) correlated minimum values of charge density along the bonds , and ( c)number of electrons transfer from each pu to o atom on the tensile strain along the [ 001 ] direction . ] in order to investigate carefully the physical nature of puo@xmath0 under tension , in the following we systematically study the electronic structures of its afm phase within lda+__u _ _ formalism for pulling along the [ 001 ] direction .
the valence charge density maps under strains of 0.00 and 0.50 along [ 001 ] direction tensile deformation are plotted in fig .
[ charge1 ] .
clearly , the interatomic distances in the [ 011 ] direction are elongated and that in the [ 100 ] direction are shortened under tension . under strain of 0.50
, fractures of the pu@xmath1o bonds can be seen . these results are understandable based on our foregoing statements for its pressure behaviors of total energy , stress , and lattice parameters . for explicitly indicating the ionic / covalent character of puo@xmath0 under tension , we further plot in fig .
[ charge2 ] the evolutions of pu@xmath1o bond length in ( 01@xmath24 ) plane , correlated minimum values of charge density along the pu@xmath1o bond , and number of electrons transfer from each pu to o atom under tensile deformation along the [ 001 ] direction . the electron transfer analysis is performed according to the bader analysis @xcite and similar ionic / covalent character investigations for thorium dioxide and hydrides have been conducted in our previous works @xcite . obviously , as indicated by fig .
[ charge2](a ) , the pu@xmath1o bond length in ( 01@xmath24 ) plane is elongated by tensile stress in [ 001 ] direction .
the initial minimum value of charge density ( 0.53 @xmath26/@xmath25 ) along the pu@xmath1o bond , comparable to that along the np@xmath1o bonds included in our previous study of npo@xmath0 @xcite , is prominently larger than that along the th@xmath1o bonds ( 0.45 e / @xmath25 ) in tho@xmath0 @xcite .
this indicates that the pu@xmath1o and np@xmath1o bonds have stronger covalency than the th@xmath1o bonds . under tension , this minimum value of charge density for pu@xmath1o bond decreases near linearly to 0.26 e / @xmath25 at the end of tensile deformation , which explicitly illustrates a decreasing behavior of covalent feature for pu@xmath1o bond .
on the other hand , the number of electrons transfer from each pu to o atom under tensile deformation also decrease with strain .
electrons of each atom are localized to their dominated region and thus reduces the ionicity for puo@xmath0 .
overall speaking , the strong mixed ionic / covalent character of pu@xmath1o bond is weakened by tensile strain . direction .
( b ) total dos and pdos under strains up to 0.00 , 0.32 , and 0.50 .
the fermi energy level is set at zero . ]
figure [ dos](a ) shows the dependence of the insulating band gap @xmath27 on the tensile strain along the [ 001 ] direction and [ dos](b ) presents the total dos as well as the projected dos for the pu 5@xmath3 and o 2@xmath6 orbitals under strains up to 0.00 , 0.32 , and 0.50 .
interestingly , the band gap varies smoothly under strain from 0.00 to 0.12 and then becomes to decrease near linearly from 1.66 ev to zero under strain of 0.12 to 0.32 .
this explicitly indicates that along the [ 001 ] direction tensile deformation puo@xmath0 will occur an insulator - to - metal transition after tensile stress exceeds about 79 gpa .
figure [ dos](b ) further illustrates this conclusion . the dos at strain of 0.50 shows that the electrons of pu and o atoms are separated from hybridization feature to exhibit one - atom character . and
the main contribution for the metallic property is from both pu 5@xmath3 and o 2@xmath6 orbitals .
in conclusion , we have studied the structural , mechanical , magnetic , and electronic properties of puo@xmath0 under tension by first - principles dft within the lda+__u _ _ formalism .
our calculated ideal tensile strengths are 81.2 , 80.5 , 28.3 , and 16.8 gpa , respectively , for pulling in the [ 001 ] , [ 100 ] , [ 110 ] , and [ 111 ] directions .
the [ 001 ] and [ 100 ] directions are prominently stronger than other two directions since that more pu@xmath1o bonds participate in the pulling process .
while the spin moments in [ 110 ] and [ 111 ] directions only increase a little , the spin moments in [ 001 ] and [ 100 ] directions increase up to about 5.3 and 5.1 @xmath23 , respectively . through charge and density of states analysis along the [ 001 ] direction
, we conclude that the initial strong mixed ionic / covalent character of pu@xmath1o bond is weakened by tensile strain and puo@xmath0 will occur an insulator - to - metal transition after tensile stress exceeds about 79 gpa . the main contribution for the metallic property
is from both pu 5@xmath3 and o 2@xmath6 orbitals at high strain domain .
this work was supported by nsfc under grant no .
51071032 , by the national basic security research program of china , and by the foundations for development of science and technology of china academy of engineering physics under grant no . | we perform a first - principles computational tensile test on puo@xmath0 based on density - functional theory within local density approximation ( lda)+__u _ _ formalism to investigate its structural , mechanical , magnetic , and intrinsic bonding properties in the four representative directions : [ 001 ] , [ 100 ] , [ 110 ] , and [ 111 ] .
the stress - strain relations show that the ideal tensile strengths in the four directions are 81.2 , 80.5 , 28.3 , and 16.8 gpa at strains of 0.36 , 0.36 , 0.22 , and 0.18 , respectively .
the [ 001 ] and [ 100 ] directions are prominently stronger than other two directions since that more pu@xmath1o bonds participate in the pulling process . through charge and density of states analysis along the [ 001 ] direction
, we find that the strong mixed ionic / covalent character of pu@xmath1o bond is weakened by tensile strain and puo@xmath0 will exhibit an insulator - to - metal transition after tensile stress exceeds about 79 gpa . |
the radio luminosity function ( rlf ) seeks to derive from observed samples and surveys of radio sources , their space density per unit co - moving volume and how this changes with source luminosity .
it is an essential input to both gravitational lensing studies which probe cosmic geometry ( e.g. kochanek 1996 ) and modelling of the clustering of radio sources ( magliocchetti et al .
the rlf derived at low radio frequencies is important for understanding the content of high frequency surveys and hence deriving jet beaming parameters ( e.g. jackson & wall 1999 ) . the shape and evolution of the rlf provide important constraints on the nature of radio activity in massive galaxies and its cosmic evolution .
longair ( 1966 ) was one of the first to attempt to determine the evolution of the radio source population . at that time
there were very few radio source redshifts known and the main constraint upon his models came from the low frequency source counts down to @xmath7 jy .
longair found that the data were best - fit by models where the most powerful radio sources undergo greater cosmic evolution ( in comoving space density @xmath1 ) than less - powerful sources .
the subsequent acquisition of a substantial fraction of the redshifts for the 3crr complete sample of laing , riley & longair ( 1983 ) enabled the evolution to be better constrained . using the @xmath8 test ( see section [ vvmax ] )
they concluded that the most powerful radio sources undergo substantial evolution with similar evolution for both powerful radio galaxies and quasars .
unfortunately , the tight correlation between radio luminosity and redshift in this sample meant that they were unable to make any progress on the form of the evolving rlf .
wall , pearson & longair ( 1980 ) suggested that complete samples a factor of @xmath9 times fainter than the 3crr sample would be necessary to differentiate between possible models of the rlf .
the most comprehensive study of the radio luminosity function prior to this work was performed by dunlop & peacock ( 1990 ; hereafter dp90 ) .
using several complete samples selected at 2.7 ghz with lower flux - limits ranging from 2.0 jy to 0.1 jy , they considered the flat- and steep - spectrum populations separately and derived the rlf for each population .
their constraints on the rlf of flat - spectrum sources were somewhat tighter than those derived by peacock ( 1985 ) , but the major breakthrough of their paper concerned the constraints they were able to place on the rlf of steep - spectrum radio sources .
the strong positive evolution in @xmath1 with @xmath0 out to @xmath4 inferred by longair ( 1966 ) , wall et al .
( 1980 ) and laing et al .
( 1983 ) was mapped out in some detail , and more controversially their results indicated a decline in the co - moving number density of both populations beyond a redshift of @xmath10 the so - called ` redshift cut off ' .
dp90 concluded that for powerful sources a model of pure luminosity evolution ( ple evolution only of the break luminosity with redshift , analogous to that of boyle , shanks & peterson 1988 for the quasar optical luminosity function ) fitted the data well ( but only for @xmath11 ) .
a similar model which incorporates negative density evolution at high redshift ( the lde model ) was found to work well for both @xmath11 and @xmath12 .
a recent update on the dp90 work can be found in dunlop ( 1998 ) , and a critical re - evaluation of the evidence for a redshift cut off in the flat - spectrum population can be found in jarvis & rawlings ( 2000 ) .
derivations of the luminosity functions of luminous agn selected at all frequencies ( radio , optical and x - ray ) have shown a broken power - law form with a steeper slope at the high luminosity end than at the low luminosity end ( e.g. dp90 , boyle et al .
1988 , page et al .
these studies were also able to adequately describe the positive evolution of the luminosity function from @xmath13 to @xmath14 by assuming pure luminosity evolution .
more recent studies with different datasets have shown some deviations from ple ( e.g. goldschmidt & miller 1998 , miyaji , hasinger & schmidt 2000 ) .
the physical rationale behind ple is that the agn have long lifetimes ( comparable with the hubble time ) and they decline in luminosity from @xmath15 to @xmath13 .
however , at least in the case of the high luminosity radio population , we know such a model is not viable because the ages of frii radio sources in the surveys are well - constrained to be @xmath16 a few @xmath17 yr ( e.g. blundell , rawlings & willott 1999 ) .
furthermore , for x - ray and optically - selected quasars , accretion at the rates required for their high luminosities ( over a hubble time ) would lead to much more massive black holes than are observed in the local universe ( e.g. cavaliere & padovani 1989 ) .
thus , although ple models do fit the gross features of the data , the physical basis of these models is not clear .
the radio source population is composed of several seemingly different types of object .
one division between types is concerned with radio structure which falls into two distinct classes : fri , or ` twin - jet ' , sources ; and frii , or ` classical double ' , sources ( fanaroff & riley 1974 ) . at low luminosities
@xmath18 , the vast majority of sources have an fri radio structure ( apart from the very low luminosity starbursts ) , whilst at higher luminosities the majority of sources have frii structure .
a second division is concerned with the optical properties of the population . at @xmath19 the fraction of objects in low frequency selected samples with observed broad - line nuclei changes rapidly from @xmath20 at higher radio luminosities to @xmath21 at lower luminosities ( willott et al .
it is not yet clear whether this represents a fundamental change in central engine properties because , for example , it could be due to the opening angle of an obscuring torus increasing with decreasing luminosity ( see willott et al .
however , it is clear that at least some frii sources lack even indirect evidence for an active quasar nucleus because , as pointed out by hine & longair ( 1979 ) and laing et al .
( 1994 ) , they have weak or absent emission lines . indeed it is plausible that the majority of frii radio galaxies ( per unit comoving volume ) show passive elliptical - galaxy spectra ( rixon , wall & benn 1991 ) and are part of the low
luminosity population considered here . in this paper
we have chosen to investigate the rlf in terms of a dual - population model where the less radio - luminous population is composed of fris and friis with weak / absent emission lines , and the more radio - luminous population of strong - line frii radio galaxies and quasars .
our motivation for doing this is that the presence / absence of a quasar nucleus , as indicated by emission line strength , is a distinction intimately connected to the properties of the central engine , whereas radio structure is strongly influenced by the larger - scale environment ( e.g. kaiser & alexander 1997 ) .
the critical radio luminosity [ @xmath22 which crudely divides the populations is also very close to the break in the rlf ( e.g. from dp90 ) , whereas the luminosity of the fri / frii divide is clearly well below the break .
tentative evidence for an increase in the fraction of quasars with redshift from @xmath13 to @xmath23 can then be explained if this low
luminosity population evolves less rapidly than the high luminosity population ( willott et al .
the @xmath1 of fri sources is already known to evolve less - strongly with @xmath0 than that of the radio - luminous friis ( e.g. urry & padovani 1995 ) , but here we will assume that this is because they are a sub - set of a larger population of low luminosity radio sources , including friis , which evolve less strongly .
we will show that differential evolution between the two populations mimics ple to a certain extent , but with a more plausible physical basis .
padovani & urry ( 1992 ) , urry & padovani ( 1995 ) and jackson & wall ( 1999 ) have adopted a different approach to determining the rlf from high frequency complete samples . instead of separating the flat- and steep - spectrum sources and deriving the rlf for each population , they use models built around unified schemes and relativistic beaming to determine the relative fractions of various classes of objects in the samples .
their models assume that bl lac objects are the beamed versions of fri and ( in the case of jackson & wall ) weak - line frii radio galaxies , and that flat - spectrum quasars are favourably oriented frii radio galaxies .
however , both these analyses required the rlf of unbeamed objects to be determined independently , i.e from low frequency samples . because they used only the 3crr sample to determine the luminosity distribution of radio sources , the tight @xmath24 correlation meant a degeneracy between luminosity and redshift effects
current models of this type are thus most useful for constraining the values of various beaming parameters and how these affect the population mix in high frequency selected samples , and have not provided new information on the evolution of steep - spectrum sources .
in this paper we present a new investigation of the steep - spectrum rlf which combines the bright 3crr sample with the new , fainter 6ce and 7crs redshift surveys ( based on the 6c and 7c radio surveys , respectively ) .
note that an investigation of the rlf of steep - spectrum quasars ( based on the 3crr and 7crs samples ) has been presented by willott et al .
1998 ( w98 ) . as the basis for an investigation of the steep - spectrum rlf , the 3crr/6ce/7crs dataset improves on that used by dp90 in four different ways . * * spectroscopic completeness .
* many of the sources in the dp90 samples do not have spectroscopic redshifts [ including , at the time of dp90 , @xmath25 of the sources in their faintest sample the parkes selected regions ( psr ) ] , and redshifts had to be estimated from the infrared hubble diagram ( the @xmath26 relation ) .
dunlop ( 1998 ) reports that further spectroscopy of these sources has shown their redshifts tended to be overestimated ( due to the positive correlation between radio luminosity and @xmath27-band luminosity , eales et al .
1997 ) , which he contends strengthens the dp90 evidence for a redshift cut off in high frequency selected samples .
however , as we discuss later ( section 5.1 ) , the assignment of systematically high redshifts to many sources may in fact have caused dp90 to overestimate the strength of any redshift cut off .
in contrast to the psr , the 6ce and 7crs have virtually complete spectroscopic redshifts .
* * coverage of the radio luminosity versus @xmath0 plane . *
the flux - limit of the psr is @xmath28 jy . for steep - spectrum sources with @xmath29 ,
this corresponds to @xmath30 jy , and for @xmath31 to @xmath32 jy , so the 7c redshift survey ( 7crs ) with a flux - limit of @xmath33 jy is a factor @xmath34 times deeper at a given @xmath0 than the psr depending on spectral index .
because of the correlations between @xmath35 and radio luminosity and @xmath0 ( e.g. blundell et al . 1999 ) this makes the 7crs sensitive to lower radio luminosity objects at the higher redshifts .
note that because the psr covers an area 3.5 times larger than the 7crs , the two samples actually contain similar numbers of high redshift objects .
* * radio completeness .
* dp90 found the source counts in the psr at 0.1 jy to be a factor of two lower than expected from the source counts at other frequencies .
they suggested that this is due to incompleteness in the original survey near its flux - limit , although this is by no means certain ( j.v . wall , priv .
comm . ) . although they attempted to correct for this ,
there remains some possibility of bias : methods based on the @xmath36 statistic are particularly sensitive to such problems , as objects close to the flux - limit have @xmath37 , and losing such objects would bias the statistic towards a spuriously low value .
the 6ce sample and the 7crs are based on surveys which are complete at the flux - densities of interest .
as riley ( 1989 ) discusses , the number of large angular size sources omitted from surveys such as these due to surface brightness effects is likely to be negligible .
* * benefits of low frequency selection*. the ultimate goal of our programme is to model the evolution in the rlf of steep - spectrum sources . as detailed in blundell
et al . ( 1999 ) , the physics of frii radio sources means that the mapping between radio luminosity and one of the key physical variables , the bulk power in the jets , is much closer if the radio luminosity is evaluated at low rest - frame frequencies provided , of course , these lie above the synchrotron self - absorption frequency of a given source .
modelling of high frequency luminosity is much more problematic , both because lobe emission at these frequencies is highly sensitive to source age and redshift ( because of inverse - compton scattering of lobe electrons by the microwave background ) , and because doppler - boosted cores and hotspots become much more important at high rest - frame frequencies . in section 2
we discuss the data available to constrain the steep - spectrum rlf .
section 3 describes the model fitting procedures and results . in section 4
the evolution of the radio source population is tested via the @xmath8 statistic .
section 5 compares our findings with those of previous rlf determinations . in section 6
we summarise our results and indicate the future work necessary to resolve the outstanding uncertainties .
the convention for the radio spectral index , @xmath38 , is that @xmath39 , where @xmath40 is the flux - density at frequency @xmath41 .
we assume throughout that @xmath42 and @xmath43 .
all results are presented for both flat ( @xmath44 ) and open ( @xmath45 ) cosmologies .
the radio luminosity function ( rlf ) is defined as the number of radio sources per unit co - moving volume per unit ( base 10 ) logarithm of luminosity , @xmath46 . to determine the rlf
, we will make use of three different types of radio data : complete samples with spectroscopic redshifts ; source counts ; and a low redshift sample to fix the local rlf .
= 0.9 a complete sample consists of every radio source in a certain area on the sky brighter than a specified flux - limit at the selection frequency .
if all these sources are identified optically with a galaxy or quasar and redshifts are determined , then the distribution of points on the @xmath47 plane , together with knowledge of the sample flux - limits and sky areas , in principle allow the determination of the rlf . the major uncertainty in such a determination is that at least some regions of the @xmath47 plane will be very sparsely populated , and at these @xmath48 and @xmath0 the rlf measurement will be prone to small number statistical uncertainties .
these typically arise because the identification of and/or redshift acquisition of faint radio sources can be very time - consuming , and due to the limited nature of time available at large telescopes , complete samples up until now have generally been limited to @xmath49 sources
. some regions of the @xmath47 plane are prone to a more fundamental problem : objects of a given @xmath48 may be sufficiently rare that there is simply too little observable comoving volume at the redshift of interest to obtain large samples .
our dataset of redshift surveys combines three low frequency complete samples with different flux - limits , which together provide unparalleled coverage of the @xmath50 plane ( see fig .
[ fig : allpz ] ) .
the faintest of the three complete samples used is the new 7c redshift survey ( 7crs ) .
this sample contains every source with a low frequency flux - density @xmath51 jy in three regions of sky covering a total of 0.022 sr .
the regions 7c - i and 7c - ii will be described in blundell et al .
( in prep . ) and 7c - iii in lacy et al .
( 1999a ) and references therein . for 7c - i and 7c - ii ,
multi - frequency radio data including high resolution radio maps from the vla will be presented in blundell et al . from measurements at several frequencies ,
the radio spectra have been fitted and hence radio luminosities at rest - frame 151 mhz determined .
spectral indices at various frequencies were also determined from these fits . for 7c - iii , the availability of 38 mhz data from the 8c survey in the same region has allowed the low frequency spectral indices to be calculated and hence rest - frame 151 mhz luminosities ( lacy et al .
1999a ) . due to the low frequency selection of the sample
, it is expected to contain few flat - spectrum ( @xmath52 ) sources . from inspection of the vla maps and radio spectra , sources which only have flux - densities above the sample limit due to their doppler - boosted cores can be identified . out of the total of 130 sources
, only one source fulfills these criteria the quasar 5c7.230 .
therefore this source is excluded from the analysis presented in this paper .
note however that there are five other quasars which have @xmath53 , the traditional separation value between flat- and steep - spectrum quasars .
all of these have @xmath54 and are compact ( @xmath55 arcsec ) , but have extended emission on _ both _ sides of the core , indicating that doppler - boosting is not the cause of this emission .
they all have sufficient extended flux to remain in the sample and as such they are not excluded on orientation bias grounds .
these sources are referred to as core - jets sources ( cjss ; see blundell et al . in prep . for more details ) .
the one other exclusion is 3c200 , because it is already included in the 3crr sample , leaving a total of 128 sources .
all 7c redshift survey sources have been reliably identified with an optical / near - infrared counterpart . for 7c - i and 7c - ii ,
spectroscopy has been attempted on all sources and secure redshifts obtained for @xmath56% of the sample ( willott et al .
, in prep . ) .
seven sources do not show any emission or absorption lines in their spectra .
for these objects , we have obtained multi - colour optical and near - infrared photometry ( in the @xmath57 , @xmath58 , @xmath59 , @xmath60 and @xmath27 bands ) to attempt to constrain their redshifts .
all of these objects have spectral energy distributions consistent with evolved stellar populations at redshifts in the range @xmath61 ( willott , rawlings & blundell 2000 ) .
these derived redshifts are generally consistent with those from the @xmath26 diagram , particularly given the increased scatter in the relationship at faint magnitudes ( eales et al .
note that radio galaxies in the redshift range @xmath62 are traditionally difficult to obtain redshifts for because there are no bright emission lines in the optical wavelength region .
the correlation between emission line and radio luminosities ( baum & heckman 1989 ; rawlings & saunders 1991 ; willott et al . 1999 ) exacerbates this problem for faint radio samples .
we believe that this is the primary reason why we have been unable to detect emission lines in these 7 sources .
it is extremely unlikely that any of these objects actually lie at @xmath63 .
details of the imaging and spectroscopy in the 7c - iii region are given in lacy et al .
( 1999a , 1999b , 2000 ) .
reasonably secure spectroscopic redshifts have been obtained for 81% of the sources and uncertain redshifts for a further 13% .
near - infrared magnitudes recently obtained support these uncertain redshifts .
only three sources have no redshift information .
near - infrared and optical imaging shows that two of them have seds typical of @xmath64 galaxies as is the case for those without spectroscopic redshifts in 7c - i and 7c - ii .
the other object ( 7c1748 + 6703 ) has a break between @xmath59 and @xmath27 that is suggestive of a redshift @xmath65 .
the existence of significant flux in its optical spectrum down to 6000 and a @xmath27-magnitude of 18.9 , suggest @xmath66 .
therefore we adopt @xmath67 in this paper .
the bright complete sample used is the 3crr sample of laing , riley & longair ( 1983 , hereafter lrl ) with complete redshift information for all 173 radio sources .
the flux - limit is 10.9 jy at 178 mhz , which translates to 12.4 jy at 151 mhz assuming a typical spectral index of 0.8 .
blundell et al .
( in prep . ) review the current status of this sample and presents the results of spectral fitting of multi - frequency radio data .
3c231 ( m82 ) is excluded because the radio emission from this source is due to a starburst and not an agn .
the two flat - spectrum quasars , 3c345 and 3c454.3 , are excluded because only doppler - boosting of their core fluxes raises their total fluxes above the flux - limit . to include objects such as these
would cause an overestimate in the number density of agn , since they only get into the samples because their jet axes are very close to our line - of - sight .
note that high frequency selected samples contain a much higher proportion of objects such as these , showing that our choice of low
frequency selection virtually eliminates these favourably oriented sources [ there are only three such sources ( @xmath68% ) in our complete samples ] .
the 6ce complete sample we use is a revision of the 6c sample of eales ( 1985 ) ( see rawlings , eales & lacy 2000 for details ) .
the flux - limits of this sample are @xmath69 jy and the sky area covered is 0.103 sr . only 3 of the 59 sources in this sample do not have redshifts determined from spectroscopy .
one object is occluded by a bright star and is excluded from further consideration ( without bias ) .
one object is faint in the near infrared ( @xmath70 ) , so we take it as a galaxy at @xmath71 in this paper .
the other is relatively bright in the near - ir but very faint in the optical .
its red colour suggests a galaxy with a redshift in the range @xmath72 and we assume @xmath73 here . none of the 6ce sources are clearly promoted into the sample by doppler - boosted core emission ( c.f 5c7.230 , 3c345 and 3c454.3 ) , so all 58 sources are included in our analysis .
full details of the sample , including optical spectra , are given in rawlings et al .
( 2000 ) .
the three complete samples with known redshift distributions contain a total of 356 sources ( after the exclusions mentioned above ) .
however , the number of radio sources at 151 mhz as a function of flux density , also known as the source counts , has been determined for thousands of sources from the much larger sky area 6c and 7c surveys ( hales , baldwin & warner 1988 and mcgilchrist et al .
1990 , respectively ) .
the 7c source counts go as faint as 0.1 jy a factor of five fainter than the 7crs . at these flux - densities
the source counts are our only constraint on the rlf . at the bright end
( @xmath74 jy ) , the source counts are obtained by binning the sources in the 3crr sample of lrl in 5 flux bins .
the three sources excluded in section [ compsamp ] were also excluded for the calculation of the source counts .
the 3crr source counts are converted from 178 to 151 mhz by assuming a spectral index of 0.8 .
figure [ fig : polysc ] shows the binned differential source counts from the 3crr , 6c and 7c surveys .
the counts are normalised to the differential source counts for a uniform distribution in a euclidean universe , such that @xmath75 , where @xmath76 and @xmath77 are the lower and upper flux limits of the bin .
these points were fitted by a third - order polynomial using a least - squares fit in logarithm space , i.e. @xmath78 the coefficients of the fit are @xmath79 and @xmath80 .
a higher order polynomial was not required to provide a decent fit to the data .
the error on this curve was estimated using the errors from the source count bins . at @xmath81
jy the fractional errors on the bins are approximately constant with flux - density at the level of 0.06 . at @xmath82 jy the fractional errors are given by a power - law with slope 0.49 such that the fractional error at 10 jy is 0.19 , for example .
= 0.47 it is well known that there is a flattening of the source counts slope at 1.4-ghz flux - densities @xmath83 mjy .
this is attributed to the counts becoming dominated by low
redshift star - forming galaxies ( condon 1984 ; windhorst et al .
although there is only one such starburst galaxy in our complete samples down to 0.5 jy ( 3c231 ) , the source count data goes as low as 0.1 jy , so it is essential to investigate the contribution of starburst galaxies to the source counts at these low flux - densities . to estimate the starburst contribution we used the @xmath84 m luminosity function of saunders et al .
( 1990 ) scaled by the mean of the ratio of iras @xmath84 m to nvss 1.4 ghz flux - densities ( cotton & condon 1998 ) .
cosmic evolution is accounted for by the ple model adopted by saunders et al . to translate this 1.4 ghz luminosity function to 151 mhz
, a spectral index of 0.7 was assumed for the starburst population .
the differential source counts at 151 mhz were then calculated from this luminosity function .
it is found that the starburst contribution at 0.1 jy is only 2% of the total source counts .
this is within the uncertainty in the source counts at 0.1 jy , so it can be safely neglected .
the radio luminosity function at low redshift ( @xmath85 ) is fairly well - determined . this therefore provides a boundary condition for determination of the evolving rlf .
unfortunately , the low space density of powerful radio sources and small cosmological volume out to @xmath86 means that the local rlf ( lrlf ) is only well - constrained at radio luminosities below @xmath87 w hz@xmath88 sr@xmath88 .
hence the lrlf will only constrain the faint end of the rlf .
another problem is that at very low luminosities ( @xmath89 w hz@xmath88sr@xmath88 ) the lrlf is dominated by the star - forming spiral / irregular galaxies which dominate the sub - mjy ( at 1.4 ghz ) source counts .
since we do not want to include this population in our rlf determination , it is essential that the lrlf of the agn population only is used .
a determination of the local radio luminosity function which treats agn and star - forming galaxies as two distinct populations is given by cotton & condon ( 1998 ) .
by cross - matching ugc galaxies ( nilson 1973 ) with the nvss catalogue ( condon et al .
1998 ) they derived luminosity functions at 1.4 ghz for each population .
the agn lrlf at 151 mhz was derived from this by assuming a radio spectral index of 0.8 .
this determination of the lrlf by cotton & condon is well - approximated by a power - law ( up to @xmath87w hz@xmath88 sr@xmath88 ) with an index of @xmath90 . from this model
the lrlf was determined in 10 bins in the range @xmath91 w hz@xmath88 sr@xmath92 .
the error on each bin was assumed to be 0.1 dex .
to find best - fit parameters for various models of the evolving rlf , a maximum likelihood method is adopted .
this method is similar to that used in w98 to determine the quasar rlf and we refer readers to that paper for further details .
the aim of the maximum likelihood method is to minimise the value of @xmath93 , which is defined as @xmath94+\ ] ] @xmath95 where @xmath96 is the differential co - moving volume element , @xmath97 is the sky area available from the samples for these values of @xmath48 and @xmath0 and @xmath46 is the model distribution being tested . in the first term of this equation ,
the sum is over all the @xmath98 sources in the combined sample .
the second term is simply the integrand of the model being tested and should give @xmath99 for good fits .
the only difference here is that the source counts and local radio luminosity function provide additional constraints for the fitting process
. the source counts and lrlf are one - dimensional functions and to estimate how well they fit the model , the value of @xmath100 is evaluated thus : @xmath101 where @xmath102 is the value of the data in the @xmath103th bin , similarly @xmath104 and @xmath105 are the model value and data error in the @xmath103th bin , respectively .
these are all determined in logarithm space for both the source counts and the lrlf .
the sum is over the @xmath98 bins in each case . in order to combine the constraints from all three types of data , we follow kochanek ( 1996 ) .
@xmath100 is related to the likelihood by @xmath106 ( likelihood ) , i.e. the same form as @xmath93 .
therefore the function that is now minimised is @xmath107 where @xmath108 is a constant which normalises @xmath93 so that equal statistical weight is given to all three types of data .
the value of @xmath108 theoretically comes from the dropped terms from equation [ eqn : qlike ] .
however , these are not possible to calculate directly , so instead one estimates the value of @xmath108 so that the first term in equation [ eqn : qlike ] equals @xmath109 and @xmath110 .
errors on the best - fit parameters are 68% confidence levels , as discussed in boyle et al .
( 1988 ) .
the goodness - of - fit of models are estimated using the 2-d kolmogorov - smirnov ( ks ) test ( peacock 1983 ) , as in w98 .
note that it is expected that the ks probabilities will be lower than for the quasar rlf fits , because in general the probability decreases with increased sample size in this test .
simulations of the @xmath50 plane are used to double check the goodness - of - fit . in addition , the predicted redshift distributions for the complete samples are derived from the models and compared with the actual distribution .
this is particularly useful in testing for the presence of a redshift cut off at high@xmath0 .
as explained in sec .
[ sec : intro ] , the rlf is modelled here as a combination of two populations which are allowed to have different rlf shapes and evolutionary properties . the low
luminosity radio source population ( composed of fris and low
excitation / weak emission line friis ) is modelled with a schechter function , @xmath111 where @xmath112 is the source number density as a function of radio luminosity @xmath48 , @xmath113 is a normalisation term , @xmath114 is the break luminosity and @xmath115 is the power - law slope .
the evolution function of @xmath116 is modelled as @xmath117 up to a maximum redshift @xmath118 beyond which there is no further evolution .
hence there are five free parameters to be fixed for the low luminosity population . for the high luminosity population a similar form is adopted , except in this case the exponential part of the schechter function is at low luminosity and the power - law region is at high luminosity , i.e. @xmath119 where the subscript h refers to the high luminosity population .
note that it is expected that @xmath120 , so that at this luminosity the decline in one population is compensated by the rise of the other , since no discontinuity or bi - modality in the rlf has yet been observed ( e.g. dp90 ) .
these forms for the luminosity functions of the two populations were chosen because the steep exponentials ensured a fairly small overlap in @xmath121 between the two populations . however , we currently have little information about the relative population fractions in the overlap region , so this is just a convenient method of separating the populations .
we note that the low - luminosity rlf of our model has a similar shape to the elliptical galaxy @xmath27-band luminosity ( and therefore mass ) function ( e.g. gardner et al .
1997 ) .
the high luminosity population is assumed to undergo density evolution of a similar form to that of the quasar rlf in w98 .
the @xmath0-distribution @xmath122 is given by a gaussian in redshift ( model a here ) , with a variant on this distribution being a one - tailed gaussian from zero redshift to the peak redshift and then a constant density to high redshift ( model b also known as the ` no cut off ' model since there is no redshift cut off ) .
note that the gaussian model a is a convenient approximation to the shape of the evolution of luminous optically - selected quasars ( e.g. warren , hewett & osmer 1994 ) , and to the evolution inferred for flat - spectrum radio sources by shaver et al .
( 1999 ) but see jarvis & rawlings ( 2000 ) for a contrary view .
there is no physical reason why one would expect the evolution to be symmetric in redshift . in w98
we showed that the gaussian redshift distribution from @xmath13 to the peak is very similar to the form @xmath123 , so we do not include models with this form of evolution here .
in w98 it was found that both these redshift distributions ( models a and b in this paper ) provided good fits because the number of quasars in the samples at high redshift was fairly small .
the inclusion of the 6ce and 7c - iii samples and the 7crs radio galaxies gives a much larger sample of @xmath124 objects here than for the quasar rlf .
hence , the high redshift evolution can be better determined .
therefore , an additional redshift distribution is tested here , which has a one - tailed gaussian rise to the peak redshift and then a one - tailed gaussian decline at higher redshifts which is allowed to have a different width from the rise .
this allows more shallow high redshift declines in @xmath125 ( such as that inferred by dp90 ) than the symmetric model a. this model ( c ) directly fits the strength of any decline in the co - moving density beyond the peak redshift @xmath126 , since the value of the high redshift gaussian width @xmath127 is free to be fit from 0 for an abrupt cut off to a large value for a very gradual decline at high redshift .
hence there are a total of six free parameters in the high luminosity population model with this form of redshift evolution .
to summarise our adopted rlf models , below we give all the equations needed to reproduce the rlf @xmath46 .
the values of the best - fit parameters are given in table [ tab : rlffit ] .
@xmath128 where @xmath129 @xmath130 @xmath131 the high luminosity evolution function @xmath122 has three different forms depending upon the model and the redshift : @xmath132 for model a at all redshifts and models b and c at @xmath133 ; @xmath134 for model b at @xmath135 ; @xmath136 for model c at @xmath135 .
[ cols="^,^,^,^,^,^,^,^,^,^,^,^,^",options="header " , ] we now briefly consider whether it is obvious that any of these three models provides the best - fit to the data and hence whether we can usefully constrain the strength of any high@xmath0 decline in the co - moving space density @xmath1 of radio sources .
we defer a full statistical analysis of this question to a future paper ( jarvis et al . , 2000 )
in which data from the 3crr and 6ce complete samples is combined with the results of targeted searches for @xmath137 radio galaxies ( the 6c * sample of blundell et al .
1998 ) . figure [ fig : zdistrlf ] shows the redshift - dependence of the high luminosity rlf for these three models with their free parameters fixed at the best - fit values .
all the models provide good fits to both the source counts and local rlf data , with reduced @xmath100 values of order one .
table [ tab : rlfgof ] shows the goodness - of - fit ( ks test probabilities ) of the models for the two cosmologies considered . in both cosmologies
the ks test shows that models b and c are marginally favoured over model a the symmetric decline model .
the extra parameter in model c ( @xmath127 the slope of the decline in number density at high redshift ) enables the strength of any high@xmath0 decline to be fitted directly .
the fact that the best - fitting values of @xmath127 are greater than those of @xmath138 but less than infinity indicates a moderate high@xmath0 decline fits the data best ( see fig . [
fig : zdistrlf ] ) .
we conclude , however , that we can not rule out with any level of confidence a constant space density up to @xmath139 .
= 0.95 simulations of the @xmath50 plane for the three models are shown in figure [ fig : mcrlf ] along with the actual @xmath50 data for the complete samples ( for @xmath140 ) .
statements made in this section concerning comparison of real and simulated data apply also to an @xmath141 cosmology .
the first thing to notice is that all the models reproduce the @xmath50 data reasonably well .
sources with emission line luminosities @xmath142 } / { \rm w } ) \geq 35.1 $ ] ( see willott et al . 1999 , 2000 for discussion of this adopted value ) are shown as filled symbols whereas those with lower luminosities are open . note that because the radio galaxies in the combined sample span a large range of redshift , balmer lines are often not observed and a classification scheme such as that adopted by laing et al .
( 1994 ) to distinguish between low and high ionisation sources can not be used .
therefore we make a distinction between the two populations simply on the basis of emission line luminosity . for the model simulations
, the high and low luminosity populations are similarly separated into filled and open symbols .
the transition between populations at @xmath143 w hz@xmath88 sr@xmath144 is well - reproduced by the models indicating that the two populations can approximately be separated according to the strength of their narrow emission lines . as seen in willott et al .
( 1999 ) , this quantity is likely to be closely related to the accretion rate of the source .
the scatter in the emission line radio correlation is not taken account of in the rlf models , which explains why the filled and open symbols overlap in the real data , but do not for the simulations .
this scatter is due to effects such as the range of radio source ages and environments which give different radio luminosities for a given jet power ( see willott et al .
1999 ; blundell et al .
1999 ) . at high redshifts ( @xmath145 ) the very few numbers of objects in the samples means that small number statistics become important .
for a comparison with simulated data , the situation is even worse because both the data and the simulations have independent poisson errors .
therefore , the simulations shown in this section should not be used to distinguish between best - fitting models ( a direct integration of the rlf over redshift is preferable , see sec .
[ sec : zdist ] ) , but these simulations could give an indication of a poor fit .
in any case , the simulations look very similar , even at @xmath145 , for all the three models tested .
we now consider the redshift distributions for the 3crr , 6ce and 7crs complete samples .
as these are obtained by integrating the rlf , the only source of small number statistics come from the data they are compared with . figure [ fig : nzmod ] shows histograms of the number of sources per bin of width @xmath146 for the 3crr , 6ce and 7crs samples separated into two populations depending upon emission line luminosities as before . for @xmath141 , the high / low luminosity divide is at @xmath142 } / { \rm w } ) = 35.4 $ ] .
also plotted are the model c rlf predictions for each population for the three samples .
= 0.42 = 0.42 = 0.42 for the 3crr sample , the model fits very well , the main exception being the low luminosity objects at @xmath147 , where there are several more sources observed than predicted .
this was also seen in the simulations and attributed to scatter in the emission line radio correlation . with only 58 sources in the 6ce sample ,
the effects of small number statistics are clear with an irregular observed distribution . however ,
in general the model redshift distribution is a fairly good approximation to the data .
there is something of an excess of low
luminosity sources predicted at low
redshifts ( @xmath148 ) , but this can be accounted for by the scatter , since there are several high luminosity objects observed at these redshifts .
for the 7crs sample , again we find a fairly good fit .
the main differences here are a small deficit predicted at @xmath149 and an excess at @xmath150 .
the redshift distributions at high redshift ( @xmath124 ) will be discussed later in this section .
figure [ fig : nzcumall ] plots the cumulative redshift distributions of the two samples for all three models .
the similarity of these three curves shows that the models all predict similar redshift distributions of sources .
the one - dimensional ks test was used to test the difference between the observed and model redshift distributions . for the 3crr sample
the significances are 0.011 ( 0.03 ) , 0.007 ( 0.03 ) and 0.007 ( 0.02 ) for models a , b and c , respectively ( @xmath141 values in brackets ) .
the values for the 6ce sample are 0.40 ( 0.32 ) , 0.15 ( 0.10 ) , 0.16 ( 0.18 ) and for the 7crs sample 0.09 ( 0.002 ) , 0.04 ( 0.0004 ) and 0.09 ( 0.001 ) .
the lowest values of ks significance are for the 7crs redshift distribution for the case of @xmath141 .
it is fairly easy to speculate on the cause of this low probability : the model rlfs appear to be too high at @xmath151 and @xmath152 and too low at @xmath153 and @xmath154 ( @xmath140 , for @xmath155 these luminosities are @xmath156 and @xmath157 ) .
this is almost certainly a limitation of our modelling procedure since these luminosities lie close to the break luminosities @xmath114 and @xmath158 which bracket the range over which the two populations overlap . away from these break luminosities the rates of exponential decline in luminosity of each population are fixed parameters in our model , and so their combination lacks the flexibility to fit perfectly all data within the overlap region .
this situation could be improved by using additional free parameters , for example parameters controlling the rates of exponential decline , or by use of free - form fits ( c.f .
dp90 ) . to compare the numbers of high redshift objects in the samples with the models , fig .
[ fig : nzdun ] plots the un - normalized cumulative redshift distributions at @xmath159 on a logarithmic scale . for the 3crr sample ,
model b ( that with no decline in @xmath1 at high redshifts ) predicts four sources at @xmath124 ( @xmath140 ; for @xmath155 the model predicts three sources ) whereas there is only one observed .
using poisson statistics the probability of observing one or less objects given a mean of four is 0.1 .
therefore the observations are marginally incompatible with the model . however , we note that the lack of @xmath124 objects in the 3crr sample could also be due to a steepening of the rlf at @xmath160 , due to a maximum radio luminosity of agn , perhaps related to the maximum black hole mass in galaxies at these redshifts .
this could be included in rlf models , but we do not do this in this paper , since there is little constraining data at such high luminosities and we wish to keep the numbers of free parameters in our models to a minimum .
note that the new equatorial sample of powerful radio sources defined by best , rttgering & lehnert ( 1999 ) which has selection criteria producing a sample very similar to the 3crr sample does indeed contain four sources at @xmath124 , so the lack of such sources in the 3crr sample may well be due to small number statistics .
the evolution of the rlf for such high luminosity sources is discussed in more detail in jarvis et al .
( 2000 ) . moving on to consider the 6ce and 7crs redshift distributions ( fig .
[ fig : nzdun ] ) , we find that all three models are consistent with the high@xmath0 data to within @xmath161 uncertainties .
there is no evidence here for a decline in the co - moving space density of radio sources at high redshift .
= 0.85 the main benefit of the addition of the 7crs complete sample to the existing 3crr and 6ce samples , is that it enables the evolution in the luminosity band @xmath162 ( @xmath140 ) to be determined over the wide redshift range @xmath163 to @xmath164 ( see fig .
[ fig : allpz ] ) .
dp90 found stronger evidence for a redshift cut off in the steep - spectrum population by considering only the sources within a similar luminosity band ( accounting for the difference in selection frequency ) .
therefore we now repeat our analysis using only the sources within this luminosity band ( note that for @xmath141 , the band chosen is that corresponding to equivalent source flux at @xmath165 , i.e. @xmath166 ) .
the bottom two panels of fig .
[ fig : nzdun ] repeat the 7crs cumulative redshift distributions , but now including only those sources in these luminosity bands . for @xmath140 ,
the no cut off model b now appears inconsistent with the @xmath167 data .
the model predicts four sources at @xmath167 whereas only one is observed .
this gives the same probability as we found with the @xmath124 3crr sources , i.e. a probability of 0.1 that the model and data are consistent .
for the case of @xmath141 , only two sources are predicted at @xmath167 by the model , so this is certainly consistent with the observation of one source .
thus we find very marginal evidence ( 90% confidence level ) that the numbers of objects in the @xmath162 luminosity band falls slightly below the no cut off model prediction .
there is however , one systematic effect which might reduce even this significance .
the expected numbers of high redshift sources were calculated from the models assuming a spectral index between 151 mhz and @xmath168 mhz of @xmath169 . however , if high-@xmath0 sources were systematically significantly steeper than @xmath169 , then the expected number of high-@xmath0 sources would be smaller .
this is because a source of a given luminosity on the flux - limit with @xmath170 , would clearly fall below the flux - limit if @xmath35 were greater than this ( see figure 1 in blundell et al .
1999 ) . our studies of complete samples ( blundell et al .
1999 ) do not reveal any intrinsic correlation between redshift and @xmath35 evaluated at 151mhz ( @xmath171 ) , however the correlation between luminosity and @xmath171 will mean that any high - redshift sources ( which must be high luminosity ) will tend to have steeper spectral indices .
the consequences of this effect are discussed in detail in section 3 of jarvis & rawlings ( 2000 ) and we defer interested readers to that paper .
re - calculation assuming @xmath172 at high
redshifts for model b gives an expectation of only two sources at @xmath167 , thereby removing any discrepancy between this model and the data .
finally , we note that the chosen luminosity band of @xmath173 excludes the four highest redshift objects in the combined sample ( see fig .
[ fig : allpz ] ) which have luminosities just above this limit . extending
the luminosity range considered to @xmath174 would give a similar result to those found for the entire samples .
in this section the @xmath8 test is applied to the complete sample data .
this test calculates the ratio of the volume @xmath175 in the universe enclosed by each source of redshift @xmath0 to the maximum volume @xmath176 , corresponding to the maximum redshift @xmath177 a source of this luminosity would have if its flux - density lies above one or more of the survey flux - limits .
if there is no cosmic evolution then sources would have values of @xmath8 randomly distributed between 0 and 1 ( rowan - robinson 1968 ; schmidt 1968 ) . hence by calculating the values of @xmath8 for all the sources in a sample and averaging , one can determine the mean evolution of the sample .
a value of @xmath178 greater than 0.5 corresponds to positive redshift evolution and less than 0.5 indicates negative evolution .
the distribution of @xmath178 expected in the case of no evolution is , for a large sample size @xmath98 , a gaussian centred on 0.5 with @xmath179 .
avni & bahcall ( 1980 ) devised a method for combining samples with various flux - limits and sky areas for the application of a generalised @xmath8 test .
this version combines several samples such that they can be treated as a single sample with a variable sky area depending upon the flux - limit .
the new test variables are the volume enclosed by sources @xmath180 and the volume available in the new sample @xmath181 .
note also that the fact that the 6ce complete sample has an upper as well as a lower flux - limit means that the volume available for a source in the 6ce sample is given by @xmath182 . to calculate values of @xmath181
one must make some assumption about the shapes of the radio spectra to determine values of @xmath177 . for this test
it is assumed that radio spectra are not significantly curved over the region of interest , and spectral indices at a rest - frame frequency of 151 mhz are used .
the known strong positive evolution from @xmath13 to @xmath183 would completely mask any change in the evolution at higher redshifts in the standard @xmath8 test .
therefore a banded version of the test is used here ( e.g. osmer & smith 1980 ) .
this modification calculates the mean @xmath184 restricting the analysis to @xmath185 at several different values of @xmath186 .
hence all the evolution below @xmath186 is masked out of the analysis .
the statistic is now @xmath187 where @xmath188 is the volume enclosed at redshift @xmath186 .
= 0.43 fig .
[ fig : vmax ] shows this banded @xmath8 test for the complete samples . in the banded test ,
the adjacent points are not statistically independent and the error bars are much larger than the typical dispersion of points .
note that at low
redshifts ( @xmath189 ) the strong positive evolution of the radio source population is clearly seen with the values of @xmath190 significantly above the 0.5 no - evolution line . at @xmath191
the values approach the 0.5 line where they remain ( within the errors ) out to the redshift limit of this sample ( @xmath192 ) .
the positive bump at @xmath193 is marginally significant , but at this point there are very few sources left in the sample , so no firm conclusions can be drawn from this . thus there is no evidence from the @xmath8 test for a high@xmath0 decline ( or increase ) in the comoving space density of radio sources .
= 0.43 the @xmath8 test can also be used to study the evolution of the low and high luminosity populations separately .
the complete samples were separated into two groups , depending upon their emission line luminosities as described in sec .
[ simul ] .
these groups are supposed to signify the division between objects in the low and high luminosity populations in the rlf .
the banded @xmath8 test was performed for each group and the results shown in fig .
[ fig : vmaxpop ] .
both populations undergo positive evolution at low
redshift ( @xmath148 ) with the evolution stronger ( and at a higher statistical significance ) for the high luminosity sources . at @xmath194
most sources in the complete samples belong to the high luminosity population , so its @xmath8 values are essentially the same as those in fig . [
fig : vmax ] for the whole dataset .
there are hints of continuing positive evolution in the regime @xmath195 for the low luminosity sources ( this is more apparent for @xmath140 than @xmath141 ) .
this evolution of the low
luminosity population is consistent with that derived in the model rlfs , where the positive evolution ceases at @xmath196 . in fig .
[ fig : vmaxlum ] we show the results of separating the sample into luminosity bands and repeating the @xmath8 test .
the @xmath197 evolution appears to depend strongly on luminosity
. the luminosity separation does not provide any evidence for negative evolution at high redshifts ( i.e. a redshift cut
off ) in either of the two high luminosity bands .
therefore we conclude from this section that the @xmath8 test shows positive evolution at @xmath198 which depends strongly on luminosity , but is inconclusive at higher redshifts .
in fig . [ fig : dprlf ] , our model - c rlf is plotted with the steep - spectrum ple model of dp90 at various redshifts ( @xmath199 only ) . the dp90 lde model which incorporates negative luminosity evolution at high redshift
is not shown here , since it is almost identical to the ple model in the redshift range @xmath200 ( which is to be expected as these regimes are constrained by data ) . to account for the fact that dp90 evaluated the rlf at high frequency ( 2.7 ghz ) ,
their luminosity scale has been transformed to 151 mhz assuming a global spectral index of 0.8 .
the kinks in the dp90 model at @xmath143 w hz@xmath88 sr@xmath201 are due to the use of a seven term power - law expansion for the non - evolving low
luminosity rlf component : there are no such powerful sources at low
redshift in their samples ( or indeed in the 3crr sample ) and hence no constraint from the data , so this feature is a numerical artefact . as described in sec . [
sec : zdist ] our modelling procedure is also imperfect : the bumps in our estimations of the rlf around the breaks in the high and low
luminosity populations may be artefacts of the restricted number of free parameters in our models .
= 0.48 in general , the dp90 ple model and model c agree reasonably well . at low luminosities
the ple model has a steeper slope of 0.69 than the slope of model c ( 0.54 ) .
again , this slope is only well constrained at low redshifts and the cause of a differing slope is probably due to the use here of a more recent determination of the local radio luminosity function .
the break luminosity appears at exactly the same value for the two models .
note how our use of two populations with differential evolution mimics the evolution of the break luminosity in the ple case .
since we know that radio sources are short - lived with respect to the hubble time , pure luminosity evolution has no physical meaning and there is clearly not one population of radio sources which fades over cosmic time .
in contrast , the two population model has some physical meaning and mimics ple by having only _ density _
evolution for both populations .
the high luminosity power - law slopes for the dp90 ple model and our model c are very similar . however , at @xmath165 there is a clear excess of high luminosity ( @xmath202 ) sources in the dp90 model . at @xmath203
the dp90 rlf is a factor of 2.1 higher than that of our model c. at @xmath204 and @xmath203 this factor has increased to 2.4 .
the effect of this is shown by a comparison of the expected high
redshift sources in the 6ce sample for rlf models with no redshift cut off in dunlop ( 1998 ) and in this paper ( fig .
[ fig : nzdun ] ) .
the plot by dunlop shows a predicted excess at @xmath124 over the number observed and this excess continues beyond the last data point at @xmath205 .
dunlop takes this as further evidence for the redshift cut off .
however , in fig .
[ fig : nzdun ] ( second panel from top ) we see that the 6ce redshift distribution is consistent with our model b at all redshifts
. how can these two models , both of which claim to be no cut off models , have such different predictions for the expected numbers of high redshift sources ?
the answer is that all no cut off models simply take the rlf at its peak at @xmath183 and ` freeze ' the shape and normalization up to high redshifts ( at least to @xmath206 ) .
therefore the dp90 models naturally predict more of these luminous sources at high redshifts because they have more of them at @xmath207 .
the next question is why do dp90 find a higher density of powerful sources at @xmath208 than found by our study ?
the answer is probably due to the fact that a very high fraction of the most distant sources in dp90 did not have reliable redshifts : inspection of fig .
9 of dp90 shows that only a quarter of their @xmath124 steep - spectrum psr sources had spectroscopic redshifts with the remainder estimated from the @xmath26 diagram .
dunlop ( 1998 ) reports that further spectroscopy has shown the redshifts in dp90 to be typically over - estimated ( most likely due to the positive correlation between radio luminosity and @xmath27-band luminosity observed by eales et al .
if these sources are actually at systematically lower redshifts then this explains why dp90 observe an excess of @xmath124 sources . extrapolating this high source density up to @xmath209
then predicts many sources in no cut off models .
note that because few of their redshift estimates were actually in the region of supposed decline ( at @xmath167 ) the redshift over - estimation has the opposite effect to the naive expectation that this new information strengthens claims for the redshift cut off . thus we have an explanation as to why dp90 find stronger evidence for a steep - spectrum redshift cut off than we find with the 3crr , 6ce and 7crs datasets ( having virtually complete spectroscopic redshifts ) .
dp90 argued that their @xmath8 test on the steep - spectrum population showed clear evidence for a redshift cut
off beyond @xmath165 .
however we do not find a similar behaviour in our low
frequency selected samples , with apparently no evolution between @xmath165 and @xmath210 .
inspection of figs .
12 and 13 of dp90 show that their evidence of negative density evolution at @xmath211 is highly tentative : their data are only about 1@xmath212 away from the null - hypothesis of no evolution .
the banded @xmath8 test suffers significantly from large statistical errors when only a few sources are left in the samples at the highest redshifts , and this tends to limit its usefulness in practice .
combined with the uncertainty in the dp90 redshift estimates , it is difficult to see clear evidence for a redshift cut off in their @xmath8 tests .
= 0.48 in line with unified schemes ( e.g. antonucci 1993 ) , the high luminosity part of the rlf derived here consists of radio galaxies and quasars which are identical except for the angle our line - of - sight makes with the jet axis . in willott
et al . ( 2000 ) we showed that the fraction of quasars in the high luminosity population is 0.4 .
therefore one would expect a very similar rlf to that of the quasars in w98 , with an increase in number density of @xmath213 .
[ fig : qrlfall ] plots the rlf derived for all radio sources in this paper and the quasar rlf from w98 multiplied by a factor of 2.5 . at high
redshifts the two rlfs appear very similar ( there are very few low redshift / luminosity quasars in the complete samples see willott et al .
2000 for a discussion of this ) .
the high luminosity power - law slope for model c is 2.3 , compared with 1.9 for the quasar rlf .
the peak for the high luminosity schechter function is at approximately the same luminosity ( @xmath214 ) as the break required by the source counts for the quasar rlf .
the quasar rlf of w98 had a flat slope below this break , whereas the models of this paper have very rapid declines with decreasing luminosity .
if the models described here are correct , then the quasar rlf should also have this schechter function form .
a survey at fainter flux - limits would probe the break region at high redshift and hopefully constrain the number of quasars ( or blrgs ) at lower radio luminosities .
a sample of quasars with a flux - limit of @xmath215 jy has already been defined by riley et al .
( 1999 ) , however incompleteness arising from the bright optical limit of this sample ( as discussed by w98 ) will complicate attempts to make quantitative estimates of the shape of the rlf at and below the break luminosity .
using low frequency selected radio samples with virtually complete redshift information , we have investigated the form of the radio luminosity function ( rlf ) and its evolution .
our results are generally in good agreement with previous work by dp90 ( fig.[fig : dprlf ] ) and w98 ( fig .
[ fig : qrlfall ] ) , and most discrepancies can be explained by limitations of the modelling methods forced on this and previous studies by the sparse sampling of the @xmath47 plane afforded by available redshift surveys of radio sources . for the reasons outlined in sec .
[ sec : intro ] , most notably our virtually complete spectroscopic redshifts , we believe our estimates of the steep - spectrum rlf to be the most accurate yet obtained .
we find that a simple dual - population model for the rlf fits the data well , requiring differential density evolution ( with @xmath0 ) for the two populations .
this behaviour mimics the effects of a pure luminosity evolution ( ple ) model but with a more plausible physical basis .
luminosity population can be associated with radio galaxies with weak emission lines , and includes sources with both fri and frii radio structures .
the comoving space density @xmath1 of the low
luminosity population rises by about one dex between @xmath2 and @xmath3 but can not yet be meaningfully constrained at higher redshifts .
the high luminosity population can be associated with radio galaxies and quasars with strong emission lines , and consists almost exclusively of sources with frii radio structure .
the comoving space density @xmath1 of this population rises by nearly three dex between @xmath2 and @xmath216 but can not yet be meaningfully constrained at higher redshifts .
our results on the rlf mirror the situation seen in x - ray and optically - selected samples of agn .
low luminosity objects exhibit a gradual [ @xmath217 from @xmath2 to @xmath3 ] rise in @xmath1 with @xmath0 which matches the rise seen in the rate of global star formation ( e.g. boyle & terlevich 1998 ; franceschini et al .
1999 ) , and also the rise in the galaxy merger rate ( e.g. le fevre et al . 2000 ) .
the @xmath1 of high luminosity objects rises much more dramatically [ @xmath219 from @xmath2 to @xmath220 .
the similarities between the cosmic evolution of radio sources and these different types of objects have also been pointed out by wall ( 1998 ) and dunlop ( 1998 ) .
we have re - investigated the question of whether there is direct evidence for a high redshift decline in the comoving space density @xmath1 of steep - spectrum radio sources . applying the @xmath8 test to the 3c/6ce/7crs dataset
provides no evidence for any such decline , which is a different result from that obtained by dp90 .
a fundamental limitation is that both our complete samples and those previously studied by dp90 include only a few sources with @xmath63 ( e.g. just 12 in 3c/6ce/7crs ) , so small number statistics are a huge problem in assessing the evidence for or against a redshift cut off .
this is considered in detail in jarvis et al .
( 2000 ) .
= 0.48 = 0.9 what would be the best way of clearing up this uncertainty regarding the redshift cut off ? larger samples are clearly needed , but in the case of the bright samples like 3crr , there is insufficient observable universe to improve much on the data already available .
significant progress on fainter radio samples is however achievable .
the choice of the flux - density level at which to concentrate efforts for the next large redshift survey is clearly crucial .
we believe that the vital factor is being able to efficiently probe objects at or just above the break in the high luminosity rlf at high redshift .
the reason for this is that integration of @xmath221 over the rlfs plotted in fig .
[ fig : allrlf ] shows that about @xmath222 of the luminosity density of the radio population at @xmath223 lies in the narrow luminosity range @xmath224 .
a future large redshift survey should be targeted at probing this population , which is seen only to @xmath3 in the 7crs .
[ fig : pzend ] shows flux - limits in the @xmath225 plane for various samples with this luminosity range shaded .
the plot illustrates that @xmath226 jy is therefore the natural choice for a new large redshift survey .
fainter surveys would be less useful as a probe of the total luminosity density at high redshift : inspection of fig .
[ fig : pzend ] shows that they would become completely dominated by the low
luminosity population , to investigate the likely content of a @xmath227 jy redshift survey , we have used models a , b and c rlfs to predict redshift distributions for a hypothetical 1000-source survey ( fig . [
fig : nz0_1 ] ) .
note that all models assume that the evolution of the low
luminosity population freezes at @xmath3 .
the first thing to notice is that the low
luminosity population dominates the total number of predicted sources in the case of @xmath199 , whereas in the @xmath141 case the two populations become approximately equal .
this same result can be seen in the source counts plots of fig .
[ fig : scects ] , and arises because the low
luminosity rlf is currently constrained only at low
redshift , and the ratio of high to low
redshift available volume is much larger in a low@xmath228 universe .
thus , although a large fraction of the sources will be at @xmath229 , we predict that in a low@xmath228 universe in which evolution of the low
luminosity population freezes at @xmath3 , of order 50 % of the sources will be in the high luminosity population and lie at high redshift , irrespective of the strength of the redshift cut off
. this should make determination of the evolution in the luminosity density a rather straightforward process .
there should be no longer be a problem of small number statistics , at least at @xmath230 . adopting model c , 7% ( 5% ) of a hypothetical 1000-source redshift survey would lie at @xmath231 for @xmath140 ( or @xmath141 ) , whereas the same model predicts 3% ( 2% ) for a survey like the 7crs . the pronounced high redshift tail in the case of ` no - cut off ' model b in fig .
[ fig : nz0_1 ] should be easily refutable with a 1000-source redshift survey at @xmath227 jy . for the symmetric - decline model a ( and @xmath140 ) one would expect about 35 @xmath231 sources in such a sample , whereas for model b we would expect 130 .
indeed , dunlop ( 1998 ) has already claimed to have ruled out the no - cut off case using indirect constraints on the redshift distribution of a lbds ( leiden - berkeley deep survey ) sample flux - limited at @xmath232 mjy , or @xmath233 mjy .
we note from fig .
[ fig : pzend ] that such a sample will be overwhelmingly dominated by the low luminosity population , and hence is too deep to provide optimal constraint on the evolution of the luminosity density of the rlf .
although , as we shall argue below , indirect evidence favours a gentle decline in @xmath1 at high redshift , dunlop s argument in favour of a redshift cut
off is not water tight .
it is currently predicated on an extrapolation between two samples , the psr and the lbds sample , neither of which is close to spectroscopic completeness , and which are separated by two orders of magnitude in flux - density ; at high redshift the psr and lbds sources lie on either side of the range in @xmath121 which is key to delineating the evolution of the luminosity density of the radio source population with redshift .
an alternative approach to obtaining complete redshift information for large samples is to refine the sample selection criteria to exclude the large numbers of low redshift sources .
this is being attempted by several groups using spectral index and/or linear size criteria ( e.g. bremer et al . 1998 ; blundell et al .
1998 ; rawlings et al . 1998
; de breuck et al . 2000 )
. however , the numbers of high@xmath0 sources which are missed in these samples are uncertain , so they will always struggle to prove the existence of a redshift cut off , although they can place a lower limit on the space density ( jarvis et al .
2000 ) .
another alternative approach is to study the evolution of flat - spectrum radio sources .
jarvis & rawlings ( 2000 ) have recently critically re - evaluated evidence in this area ( peacock 1985 ; dp90 ; shaver et al .
1996 ; shaver et al . 1999 ) and concluded that constant-@xmath1 models for the most luminous flat - spectrum sources can be ruled out at the @xmath234 level .
they favour a decline in @xmath1 for these objects by a factor @xmath235 between @xmath236 and @xmath139 , i.e. behaviour intermediate between models b and c considered in this paper .
however , they also point out that it is not yet clear how this decline should be interpreted in the context of the steep - spectrum population .
* we have used three low - frequency selected surveys with essentially complete spectroscopic redshift information , giving unprecedented coverage of the radio luminosity versus @xmath0 plane , to derive the most accurate measurement of the steep - spectrum rlf yet made . *
we find that a simple dual - population model for the rlf fits the data well , requiring differential density evolution ( with @xmath0 ) for the two populations .
* the low
luminosity population can be associated with radio galaxies with weak emission lines , and includes sources with both fri and frii radio structures ; its comoving space density @xmath1 rises by about one dex between @xmath2 and @xmath3 but can not yet be meaningfully constrained at higher redshifts .
the high luminosity population can be associated with radio galaxies and quasars with strong emission lines , and consists almost exclusively of sources with frii radio structure ; its @xmath1 rises by nearly three dex between @xmath237 and @xmath4 .
thanks to gary hill , julia riley and david rossitter for their important contributions to the 7c redshift survey . thanks also to matt jarvis and jasper wall for interesting discussions .
we thank the referee jim dunlop for a helpful referees report .
we acknowledge the use of the ukirt , the wht and the nrao vla which have together made this project possible .
the united kingdom infrared telescope is operated by the joint astronomy centre on behalf of the u.k . particle physics and astronomy research council .
the william herschel telescope is operated on the island of la palma by the isaac newton group in the spanish observatorio del roque de los muchachos of the instituto de astrofisica de canarias .
the very large array is operated by the national radio astronomy observatory which is a facility of the national science foundation operated under cooperative agreement by associated universities , inc .
this research has made use of the nasa / ipac extra - galactic database , which is operated by the jet propulsion laboratory , caltech , under contract with the national aeronautics and space administration .
cjw thanks pparc for support .
antonucci r.r.j . , 1993 ,
araa , 31 , 473 avni y. , bahcall j.n . , 1980 , apj , 235 , 694 baum s.a . ,
heckman t.m . , 1989 , apj , 336 , 702 best p.n . , rttgering h.j.a . ,
lehnert m.d . , 1999 ,
mnras , 310 , 223 blundell k.m . ,
rawlings s. , 1999 , nature , 399 , 330 blundell k.m . , rawlings s. ,
eales s.a .
, taylor g.b . ,
bradley a.d . , 1998 , mnras , 295 , 265 blundell k.m . , rawlings s. , willott c.j . , 1999 ,
aj , 117 , 677 boyle b.j . , terlevich r.j . , 1998 ,
mnras , 293l , 49 boyle b.j . , shanks t. , peterson b.a
. , 1988 , mnras , 235 , 935 bremer m.n . , rengelink r. , saunders r. , rttgering h.j.a . , miley g.k . ,
snellen i.a.g . , 1998 , in observational cosmology with the new radio surveys , eds .
bremer et al .
, 165 , kluwer cavaliere a. , padovani p. , 1989 ,
apj , 340 , l5 condon j.j . , 1984 ,
apj , 284 , 44 condon j.j . , cotton w.d . ,
greisen e.w . ,
. , perley r.a . ,
taylor g.b . ,
broderick j.j . , 1998 ,
aj , 115 , 1693 cotton w.d . , condon j.j . , 1998 , in observational cosmology with the new radio surveys , eds
bremer et al . , 45 , kluwer de breuck c. , van breugel w. , rttgering h.j.a . , miley g. , 2000 , a&as , 143 , 303 dunlop j.s . , 1998 , in observational cosmology with the new radio surveys , eds .
bremer et al . , 157 , kluwer dunlop j.s . , peacock j.a . , 1990 ,
mnras , 247 , 19 ( dp90 ) eales s.a .
, 1985 , mnras , 217 , 149 eales s.a . , rawlings s. , law - green j.d.b .
, cotter g. , lacy m. , 1997 , mnras , 291 , 593 fanaroff b.l .
, riley j.m . , 1974 , mnras , 167 , 31p franceschini a. , hasinger g. , miyaji t. , malquori d. , 1999 , mnras , 310l , 5 gardner j.p . , sharples r.m .
, frenk c.s . ,
carrasco b.e . , 1997 , apj , 480 , l99 goldschmidt p. , miller l. , 1998 , mnras , 293 , 107 hales s.e.g .
, baldwin j.e . , warner p.j . , 1988 , mnras , 234 , 919 hine r.g . , longair m.s . , 1979 , mnras , 188
, 111 jackson c.a . , wall j.v . , 1999 , mnras , 304 , 160 jarvis m.j . , rawlings s. , 2000 , mnras , in press , astro - ph/0006081 jarvis m.j . , rawlings s. , willott c.j . ,
blundell k.m .
, eales s.a . , lacy m. , 2000 , mnras , submitted kaiser , c.r . , alexander p. , 1997 ,
mnras , 286 , 215 kochanek c.s . , 1996 , apj , 473 , 595 lacy m. , rawlings s. , hill g.j . , bunker a.j . ,
ridgway s.e .
, stern d. , 1999a , mnras , 308 , 1096 lacy m. , kaiser m.e . ,
hill g.j . , rawlings s. , leyshon g. , 1999b , mnras , 308 , 1087 lacy m. , bunker a.j . ,
ridgway s.e . , 2000 ,
aj , 120 , 68 laing r.a . , riley j.m . , longair m.s . , 1983 , mnras , 204 , 151 ( lrl ) laing r.a . , jenkins c.r
, wall j.v . , unger s.w . , 1994 , in proceedings of the stromlo centenary symposium : physics of active galactic nuclei , ed
. g. bicknell , 201 , springer - verlag , berlin le fvre o. , abraham r. , lilly s.j . ,
ellis r.s . , brinchmann j. , schade d. , tresse l. , colless m. , crampton d. , glazebrook k. , hammer f. , broadhurst t. , 2000 , mnras , 311 , 565 longair m.s . , 1966 , mnras , 133 , 421 magliocchetti m. , maddox s.j . , lahav o. , wall j.v . , 1998 , mnras , 300 , 257 mcgilchrist m.m .
, baldwin j.e . , riley j.m . , titterington d.j . ,
waldram e.m . ,
warner p.j .
, 1990 , mnras , 246 , 110 miyaji t. , hasinger g. , schmidt m. , 2000 , a&a , 353 , 25 nilson p. , 1973
, uppsala general catalogue of galaxies , uppsala astronomical observatory osmer p.s . ,
smith m.g . , 1980 , apjs , 42 , 333 padovani p. , urry c.m .
, 1992 , apj , 387 , 449 page m.j . ,
carrera f.j . , hasinger g. , mason k.o . ,
mcmahon , r.g .
, mittaz j.p.d . ,
barcons x. , carballo r. , gonzalez - serrano i. , perez - fournon i. , 1996 , mnras , 281 , 576 peacock j.a . , 1983 , mnras , 202 , 615 peacock j.a . , 1985 , mnras , 217 , 601 rawlings s. , saunders r. , 1991 , nature , 349 , 138 rawlings s. , blundell k.m . , lacy m. , willott c.j .
, eales s.a . , 1998 , in observational cosmology with the new radio surveys , eds .
bremer et al . , 171 , kluwer rawlings s. , eales s.a . , lacy m. , 2000 , mnras , in press rees m.j .
, setti g. , 1968 , nature , 219 , 127 riley j.m
. , 1989 , mnras , 238 , 1055 riley j.m . , rawlings s. , mcmahon r.g . , blundell k.m . ,
miller p. , lacy m. , 1999 , mnras , 307 , 293 rixon g.t .
, wall j.v . , benn c.r . , 1991 , mnras , 251 , 243 rowan - robinson m. , 1968 , mnras , 138 , 445 saunders w. , rowan - robinson m. , lawrence a. , efstathiou g. , kaiser n. , ellis r.s .
, frenk c.s . , 1990 ,
mnras , 242 , 318 schmidt m. , 1968 , apj , 151 , 393 shaver p.a . , wall j.v . , kellermann k.i . ,
jackson c. , hawkins m.r.s .
, 1996 , nature , 384 , 439 shaver p.a .
, hook i.m . , jackson c. , wall j.v . , kellermann k.i
. , 1999 , in highly redshifted radio lines , eds .
carilli et al . , asp conf .
series vol .
156 , 163 .
urry c.m . , padovani p. , 1995 ,
pasp , 107 , 803 urry c.m . ,
padovani p. , stickel m. , 1991 , apj , 382 , 501 wall j.v .
, 1983 , in the origin and evolution of galaxies , eds .
jones and j.e .
jones , 295 , reidel wall j.v . , 1998 , in observational cosmology with the new radio surveys , eds .
bremer et al . , 129 , kluwer wall j.v . ,
pearson t.j .
, longair m.s . , 1980 , mnras , 193 , 683 warren s.j . , hewett p.c . ,
osmer p.s . , 1994 ,
apj , 421 , 412 willott c.j . , rawlings s. , blundell k.m . , lacy m. , 1998 , mnras , 300 , 625 ( w98 ) willott c.j . , rawlings s. , blundell k.m . , lacy m. , 1999 , mnras , 309 , 1017 willott c.j . , rawlings s. , blundell k.m . , lacy m. , 2000 , mnras , 316 , 449 willott c.j . , rawlings s. , blundell k.m . , 2000 , mnras , in press windhorst r.a . , miley g.k . ,
owen f.n . , kron r.g . ,
koo d.c . , 1985 ,
apj , 289 , 494 | we measure the radio luminosity function ( rlf ) of steep - spectrum radio sources using three redshift surveys of flux - limited samples selected at low ( 151 & 178 mhz ) radio frequency , low
frequency source counts and the local rlf .
the redshift surveys used are the new 7c redshift survey ( 7crs ) and the brighter 3crr and 6ce surveys totalling 356 sources with virtually complete redshift @xmath0 information .
this yields unprecedented coverage of the radio luminosity versus @xmath0 plane for steep - spectrum sources , and hence the most accurate measurements of the steep - spectrum rlf yet made .
we find that a simple dual - population model for the rlf fits the data well , requiring differential density evolution ( with @xmath0 ) for the two populations . the low
luminosity population can be associated with radio galaxies with weak emission lines , and includes sources with both fri and frii radio structures ; its comoving space density @xmath1 rises by about one dex between @xmath2 and @xmath3 but can not yet be meaningfully constrained at higher redshifts .
the high luminosity population can be associated with radio galaxies and quasars with strong emission lines , and consists almost exclusively of sources with frii radio structure ; its @xmath1 rises by nearly three dex between @xmath2 and @xmath4 .
these results mirror the situation seen in x - ray and optically - selected samples of agn where : ( i ) low luminosity objects exhibit a gradual rise in @xmath1 with @xmath0 which crudely matches the rises seen in the rates of global star formation and galaxy mergers ; and ( ii ) the density of high luminosity objects rises much more dramatically .
the integrated radio luminosity density of the combination of the two populations is controlled by the value of @xmath1 at the low
luminosity end of the rlf of the high luminosity population , a quantity which has been directly measured at @xmath3 by the 7crs .
we argue that robust determination of this quantity at higher redshifts requires a new redshift survey based on a large ( @xmath5 source ) sample about five times fainter than the 7crs .
epsf.sty radio continuum:@xmath6 galaxies galaxies:@xmath6active quasars:@xmath6general galaxies:@xmath6evolution |
fig.1 . structure of oxygen - deficient layered oxides , e.g. , manganites lnbamn@xmath0o@xmath8 ( a ) ; geometry of nuclear inelastic scattering of synchrotron radiation ( b ) ; phonon density of states of fe in unoriented powder of tbbafe@xmath0o@xmath8 ( c ) .
the @xmath163dependence of the @xmath86scaled coefficients of the legendre series expansion of the mdf , eq .
[ 01 ] ( a ) and contour polar plot of the isotropic ( @xmath164 ) site @xmath86scaled spectra asymmetry @xmath87 in function of the strength of preferred orientation @xmath41 and incidence angle @xmath58 ( b ) .
vibrational anisotropy @xmath96 in function of three variables @xmath134 , eq .
[ 15 ] , is shown as three slices ( contour plots ) of @xmath165 . in each @xmath166matched slice ,
the oriented samples of manganites lnbamn@xmath0o@xmath8 and lnbamn@xmath0o@xmath8 are shown by ln symbols and oxygen index as spots with coordinates @xmath167 obtained from rietveld refinement of parameters of preferred orientation .
mossbauer spectra in oriented samples of @xmath6fe - doped manganites .
two panels present the oxygen - poor and oxygen - rich series ( `` o@xmath8 '' and `` o@xmath9 '' ) . the spectra are fitted either with one or with two asymmetric doublets .
meltdowns of mn@xmath168/mn@xmath169 and mn@xmath169/mn@xmath170charge order in o5 and o6 series , respectively , are seen as single - site doublets .
the occurrence of two sites for fe in la and sm members of the lnbamn@xmath0o@xmath8 series is interpreted as remainder of unmolten charge order between mn@xmath168 and mn@xmath169 .
the asymmetries @xmath36 were fixed to be a single parameter of both doublets .
this is because each crystallite contain both these sites and their @xmath95 direction and sign coincide according to structure@xcite .
the angular dependence of the doublet asymmetry ( a ) as given by the eq .
( 17 ) and the mssbauer spectrum measured under the magic angle @xmath149 between the direction from source to detector and the normal to the gdbamn@xmath147fe@xmath148o@xmath9 sample absorber .
the distance between source and the sample center was @xmath171 4 cm and the illuminated area of the sample was 2@xmath1722 cm ( @xmath173 ) .
the angles at the absorber surface were thus varied between @xmath174 @xmath175 and @xmath176 .
this range between @xmath78 = 45@xmath177 and @xmath178 is shown in ( a ) by bold line .
a small residual asymmetry might have been resulted from the averaging in this range .
the fitted value of @xmath179r@xmath180 = 0.0045 is 5.5 times smaller than the normal - incidence value of @xmath181r@xmath180 in fig.4 . | while synthesizing the single crystals of novel materials is not always feasible , orienting the layered polycrystals becomes an attractive method in the studies of angular dependencies of inelastic scattering of x - rays or neutrons .
putting in use the rietveld analysis of layered structures in novel manganites and cuprates we develop the studies of their anisotropic properties with oriented powders instead of single crystals .
densities of phonon states ( dos ) and atomic thermal displacememts ( atd ) are anisotropic in the a - site ordered manganites lnbamn@xmath0o@xmath1 of both @xmath2 and @xmath3 series ( ln = y , la , sm , gd ) .
we establish the angular dependence of dos on textures of arbitrary strengths , link the textures observed by x - ray and @xmath4ray techniques , and solve the problem of disentanglement of goldanskii - karyagin effect ( gke ) and texture in mssbauer spectra .
two states of matter traditionally used in structure analysis are single crystals and polycrystalline powder .
when synthesizing the single crystal is a stiff task the random polycrystalline materials or even polycrystals oriented on a surface conformably to their persistent crystal habitus can be employed . via the method developed initially by rietveld@xcite
the refinement of texture , or preferred orientation , is a conventional procedure alongside with the refinement of atomic structure parameters .
randomization of the powders favors the perfect refinements of atomic parameters , therefore , any residual texture is usually not in line with the best sample preparation for x - ray profile analysis
. there appears , however , a class of problems , in which the well - oriented powders of platy or acicular crystallites can replace the unavailable single crystals . to them belong the studies in anisotropic properties of materials , in electric , magnetic properties , and in lattice dynamics .
the texture descriptions and determination of the preferred orientation parameters are thus becoming the issues of self - sustained interest .
recently , orienting powders to make the samples with varied degree of crystallite alignment was suggested@xcite to be useful in vibrational spectroscopy of anisotropic materials .
when the polycrystalline material is a ferromagnet or a superconductor it can be thoroughly subjected to texturing in an external magnetic field .
the vibrational properties in novel manganites and cuprates are anisotropic owing to their layered structure .
substitution of fe into @xmath5 jahn - teller cation site perturbs the electronic system at small doping rate , and modifies the existing charge and/or orbital order .
however , due to strong electron - phonon interactions the system keeps its attractiveness for high - resolution spectroscopies based on nuclear @xmath4resonance .
the fe - doped samples allowed us to probe the anisotropy of the dopant vibrations due to gke in @xmath6fe mssbauer spectroscopy@xcite .
the same doping procedure is extremely valuable in the synchrotron radiation vibrational spectroscopy using the nuclear inelastic scattering ( nis ) @xcite . in this work ,
we establish the relationships between the parameters of preferred orientation , the anisotropic phonon dos , and the mssbauer line intensities . using these relationships
we propose the novel technique of nis on the oriented powder samples .
[ ptb ] f1.eps previous works dealt with the texture manifestations in mssbauer spectra@xcite . for a unpolarized mssbauer source there occur two parameters of texture , which can be determined from spectra .
these parameters define the so - called `` minimum texture function''(mtf ) .
our approach is to employ the march - dollase@xcite function ( mdf ) implemented in programs for diffraction profile analysis@xcite .
we establish the relation between rietveld and mssbauer textures in terms of mdf and mtf , using the example of quadrupole doublet spectra for the @xmath6fe nuclei occupying @xmath7 of mn sites in paramagnetic state of novel manganites lnbamn@xmath0o@xmath8 @xcite and lnbamn@xmath0o@xmath9@xcite .
the symmetries are tetragonal , except a monoclinic member ybamn@xmath0o@xmath9 , and measurements were made with azimuthal rotation .
our texture function @xmath10 was uniaxial , @xmath11invariant ( fig.1 ) . in the proposed method ,
the dos derivation is based on two nis patterns and a rietveld pattern . owing to relationship between mtf and mdf , the mssbauer spectra and rietveld analysis become mutually complementary techniques .
indeed , the transmission ( mssbauer ) and reflection ( bragg or nis ) data may diverge if there occur some in - depth variations of texture .
the @xmath12resonance wavelength ( 0.86 ) is close to x - ray mo k@xmath13(0.7 ) , but the main component of nis radiation collected by the avalanche photodiode detector ( apd ) is fe @xmath14 ( 1.94 )@xcite , that is close to x - ray wavelength of cu @xmath14(1.54 ) . in the nis spectrum of an anisotropic crystal ,
the phonon dos is weighted by squared projection of the phonon polarization vectors to the wave vector of the x - ray quantum@xcite .
let the incident beam be launched under the angle @xmath15 with respect to preferred axis ( @xmath16-axis ) of a plate- or a needle - like crystallite .
the projected dos for this crystallite is : @xmath17 to introduce the averaging of the dos over the ensemble of aligned crystallites we must integrate these two terms with the volume of crystallites @xmath18 whose z - axis lies within the cone shell element @xmath19 : @xmath20 here @xmath21 and the orientation distribution function ( odf ) is normalized to unity .
the polar odf @xmath10 is to replace @xmath22 via the coordinate transform from the frame of the beam to the frame of the rotation stage .
the ratio of angular elements @xmath23 is @xmath24 @xmath25 and the jacobian of this transform is @xmath24 @xmath26 . using @xmath27 we obtain for the uniaxial symmetry @xmath28 powder averaging @xmath29 @xmath30 can be expressed via legendre function of the first kind @xmath31 : @xmath32 d(\cos\theta)\label{08}%\ ] ] prior measuring
the nis spectra mssbauer spectroscopy and rietveld analysis provide the results in terms of mtf and mdf , respectively . due to the axial symmetry , the texture function
@xmath10 would have only even terms in the legendre expansion series . @xmath33
the observable quantities , @xmath34 , and asymmetries in mssbauer spectra ( @xmath35 or @xmath36 , as denoted below ) are fully determined by the integrals of @xmath10 , i.e. , ( [ 08 ] ) and @xmath37 , therefore , the function @xmath10 consistent with experiment is not unique .
the simplest function @xmath10 conforming the eq .
( [ 08 ] ) , is the so - called mtf @xcite , composed of a linear combination of @xmath38 and @xmath39 terms , i.e. , the parabolic function of @xmath40 .
rietveld analysis@xcite specifies the preferred orientation with two fit parameters @xmath41 and @xmath42 of the mdf @xcite:@xmath43 \label{01}%\ ] ] [ ptb ] f2.eps the angle @xmath44 describes the misfit of the crystallite orientation with respect to the axis of the uniaxial texture ( fig.1 , b ) , and @xmath42 is the fraction of the oriented phase .
the march variable @xmath41 expresses the strength of the preferred orientation . in a sample made by pressing a layer of plates ( needles ) of the initial thickness @xmath45 @xmath46 down to the thickness @xmath47 @xmath48 ,
the degree of compression is @xmath49@xcite .
the function @xmath50 conserves the scattering matter , therefore , it has a clearer physical sense than @xmath10 originally used by rietveld @xmath51@xcite .
mdf is thus a true normalized angular distribution , equally good to the textures made of crystallites with either platy or acicular habitus@xcite .
ericsson and wppling have previously studied in chlorite@xcite the effect of texturing induced by compression of various degrees @xmath52 on the mssbauer line area asymmetry .
they have pointed out that their model `` represent an extreme case '' and that less randomness were observed in chlorite samples than predicted by their one - parametric compression model .
nagy @xcite has introduced a second parameter taking into account the particle shape and assumed that the texturing behavior of the particles under compression is shape - dependent . for the platy chlorite flakes nagy obtained the aspect ratio ( cylindrical @xmath53 ) as large as 0.6 @xcite .
the march - dollase function implemented in the fullprof program@xcite is also two - parametric , however , the meaning of the second parameter @xmath42 is unrelated to the method of particle alignment .
the unaligned random phase is just added with the fraction of @xmath54 @xmath42 . we show below that the mssbauer line area asymmetry depends for small asymmetries only on the product @xmath55 .
difference of physical meanings between @xmath42 and @xmath41 manifests itself only in strongly asymmetric spectra .
measurements with the rotated sample stage inclined to the incident beam ( fig.1 , b ) provide us with two components @xmath56 , @xmath57 of uniaxial dos in terms of the angle @xmath58 and the parameters @xmath41 and @xmath42 .
using mdf ( eq.[01 ] ) , we obtain @xmath59 \cos^{2}\omega+\frac{v(r,\phi)}{2}\sin^{2}\omega\end{aligned}\ ] ] via substitution @xmath60 , @xmath61 into eq.([05 ] ) . both @xmath62 and @xmath63 are imaginary for @xmath64 , however , @xmath65 is real for @xmath66 .
the ranges @xmath64 , and @xmath67 correspond to platy and stalky habits , respectively .
now from ( [ 17 ] ) we can express mtf through @xmath41 and @xmath42:@xmath68 it is shown in fig.2 ( a ) that the legendre series coefficients of the mdf are ( all , except @xmath69 ) returning to zero at @xmath70 and that the mdf is well approximated by @xmath71 in some vicinity of random polycrystal @xmath72 , because @xmath73 @xmath74 for @xmath75 around @xmath70 .
[ ptb ] f3.eps we are ready for determination of both @xmath56 and @xmath57 . from eq.([03 ] ) a couple of measurements of dos @xmath76 and @xmath77 at the angles @xmath78 and @xmath79 leads to : @xmath80 ( \cos^{2}\omega_{1}-\cos^{2}\omega_{2})}\\ g_{x}(e ) & = \overline{g}(e)-\frac{\delta g_{12}(e)}{2}\frac{(\cos^{2}% \omega_{1}+\cos^{2}\omega_{2})}{(\cos^{2}\omega_{1}-\cos^{2}\omega_{2})}%\end{aligned}\ ] ] @xmath81 , @xmath82 .
now we turn to the asymmetry of mssbauer spectra caused by texture with parameters @xmath41 and @xmath42 .
first , the intensity ratio for a single crystallite is to be examined depending on the orientation of the wave vector of the incident x - ray quantum with respect to the axes of the electric field gradient ( efg ) tensor at the site wherein the @xmath6fe nucleus is located .
the @xmath15-dependent clebsch - gordan coefficients determine the doublet line intensity ratio @xmath35@xcite .
we found that the quantity , which is proportional to the amount of the oriented phase @xmath42 is the deviation of relative line area from @xmath83 : @xmath84 powder averaging resolves , using eq.([05 ] ) , into @xmath85 \sin^{2}\omega\label{16}%\ ] ] eq.([16 ] ) remains true in @xmath86scaled form , i.e. with @xmath87 in place of @xmath36 , and @xmath65 in place of @xmath88 . from fig .
2(b ) starting with @xmath89 one finds the mdf variable @xmath41 at any angle @xmath58 except magic angle @xmath90 .
[ ptb ] f4.eps when the measurement is done at @xmath91 the texture does not manifest itself in the spectra . in a simple normal incidence transmission mossbauer spectrum @xmath92 , the expression for intensity ratio was formulated@xcite ,
@xmath93 however , the solutions were yet found either for texture effects , or for gke , separately only . in the proposed synchrotron experiments on anisotropic powders , both are crucial , therefore , the combined effects of texture and gke are of our interest . in ( [ 10 ] ) @xmath94 is the squared wave vector times the difference of mean - square vibrational displacements along @xmath95 , and in perpendicular direction , @xmath96 , @xmath97@xmath98 .
here again , it is more convenient to work with @xmath99 .
the replacement of @xmath50 with the mtf , eq .
[ 09 ] , makes our expression for @xmath36 integrable and expressible via two simple sigmoid functions @xmath100 and @xmath101 : @xmath102 } { 1 - 20\phi u(r)\sigma _ { 0}(\alpha)}\label{13}%\ ] ] with @xmath103 , @xmath104 , and @xmath105 here @xmath106 @xmath107 is the kummer confluent hypergeometric function .
the range of variation for both sigmoid functions @xmath108 and @xmath101 is between @xmath109 and @xmath110 ; for unoriented powder @xmath111 @xmath112 , and @xmath113 @xmath108 is the exact solution .
the accuracy of the solution ( [ 13 ] ) is better than 1% only in the narrow range of slight textures ( @xmath114 and @xmath115 ) , in which the mtf is a perfect approximation for mdf ( eq . [ 01 ] ) . in the same range ,
the @xmath116 differs from @xmath108 merely by shift of sigmoid flexpoint : @xmath117 the approximation ( [ 14 ] ) is as good as ( [ 13 ] ) for @xmath118 .
the parameters @xmath41 and @xmath42 enter to @xmath119 in equivalent form , however , become nonequivalent if the @xmath120approximation ( [ 14 ] ) is extended to a broader range of @xmath41 .
we extend it by taking into account the narrowing sigmoid for stronger textures : @xmath121 .
the accuracy of this approximation is better than 1% in the broad range of textures , @xmath122 .
the @xmath41 and @xmath42 dependencies of @xmath123 and @xmath124 can be well fitted with @xmath125 and @xmath126 .
the polynomials @xmath127 and @xmath128 could be expressed as dot products of coefficients vectors @xmath129 with the vector @xmath130 @xcite . in practice ,
it is crucial to find the vibrational anisotropy @xmath94 starting from @xmath36 , therefore , @xmath94 is given using the function inverse of @xmath131 : @xmath132 the function of @xmath133 variables @xmath134 is visualized via several 2d maps ( slices , fig.3 ) .
total range , in which @xmath135 is defined , is @xmath136 , and the slices are shown for @xmath137 , @xmath138 , and @xmath139 .
linear slope @xmath140 is exact near @xmath141 , but its discrepancy reaches 10% already at @xmath142 .
the parabolic model @xmath143 is better than @xmath144 in the range @xmath145 that covers the asymmetries observed in our spectra ( fig.4 ) .
we have checked that the rotation of the sample around an axis perpendicular to the @xmath146vector of the incident @xmath4quantum induces a difference of the doublet asymmetry . the same as in fig.4 absorber of gdbamn@xmath147fe@xmath148o@xmath9
was installed under the magic angle @xmath149 and the symmetric doublet was obtained .
a very small residual asymmetry can be explained by the divergence of the incident beam ( fig.5 ) .
[ ptb ] f5.eps both oxygen - saturated and oxygen - depleted families of layered manganites show negative @xmath36 , although their left - to - right area ratio changes the sign between `` o@xmath8 '' and `` o@xmath9 '' series .
this is because the main axis of efg @xmath150 is perpendicular to layers and @xmath151 in the pyramid feo@xmath8 , but @xmath152 in the feo@xmath9 octahedron compressed along @xmath153axis .
since the sign of @xmath95 changes from `` o@xmath8 '' to `` o@xmath9 '' series the doublet lines swap their positions .
@xmath36 remains negative according to its definition in eq.([11 ] ) .
therefore , in both panels of fig.4 , the stronger line is @xmath154 , and the weaker line is @xmath155 . in this respect ,
the layered manganites are quite similar to layered cuprates , wherein the ionic point charge model prescribes the same orientation and sign of @xmath95@xcite .
the values of @xmath41 and @xmath42 for our samples@xcite refined from x - ray diffraction patterns@xcite are shown by spots in fig.3 , at slices of the same @xmath36 as shown in fig.4 .
the value of @xmath94 increases with increasing size of ln within either `` o@xmath8 '' or `` o@xmath9 '' series , however , the vibrational anisotropy is positive in pyramid feo@xmath8 ( @xmath156 ) , but negative in the compressed octahedron feo@xmath9 ( @xmath157 ) .
this result is verifiable via refinement of the factors of anisotropic thermal displacements ( atd ) of mn from neutron diffraction data , however , till now such the data were refined with @xmath158models@xcite . on the other hand , in a similar bilayered structure of ybafecuo@xmath8 , the atd of fe shaped in feo@xmath8 pyramid as prolate cigar was found@xcite . also , the oblate pancake is not unexpected shape of atd in the oblate octahedron of lnbamn@xmath0o@xmath9 . in conclusion ,
we have shown how the information contained in the x - ray diffraction full - profile patterns of the aligned powders can be used for finding the direction - projected components of the phonon dos .
this information is sufficient , in principle , to solve the problem of determination of two dos components from at least two experiments either performed on two samples with different and non - zero degrees of alignment or conducted on one aligned - powder sample but under at least two different angles @xmath58 .
the same information is contained in assymetry of mssbauer line intensities , however , here the texture effect is entangled with gke . therefore , the useful for dos or for other properties information can be disentangled from mssbauer spectra in only cases when either gke or odf are characterized by some separate experiments .
one possibility is to determine the gke separately from the atd refinement using neutron diffraction full - profile analysis superior in the accuracy of @xmath159 compared to x - ray rietveld analysis .
another possibility is to employ the texture march - dollase parameters .
the latter possibility was realized in this work for two cases when the sign and orientation of uniaxial @xmath95 is known , namely , in lnbamn@xmath0o@xmath8 ( @xmath151 ) and in lnbamn@xmath0o@xmath9 ( @xmath152 ) .
our method is well applicable for these tetragonal structures , in which the axis of preferred orientation coincides with the direction perpendicular to layers and with the principal axis of electric field gradient .
more complicated textured systems , such as non - uniaxial or magnetic textures were beyond the scope of the present work .
it must be finally emphasized that the scope of possible applications of the present work is limited by the close correspondence between the distribution of march - dollase , which is theoretically substantiated only in pressed samples@xcite , and the real odf @xmath10 , which can be obtained experimentally with texture goniometers using several analytical methods , best of which combine the odf calculation with the rietveld structure refinement@xcite .
our approach is also good for samples containing the randomly oriented phase , because we employed the extended ( biparametric ) mdf @xmath50 .
this work was supported by asahi glass foundation and rfbr - jsps ( grant 07 - 02 - 91201 ) .
99 h.m .
rietveld , j. appl . cryst . * 2 , * 65 - 71 ( 1969 ) .
a. i. rykov , europhys .
lett . * 85 * , 16003 , p1-p6 ( 2009 ) .
a. rykov , v. caignaert , and b. raveau , j. solid .
st .
chem . * 109 * , 295 - 306 ( 1994 ) .
m. seto , y. yoda , s. kikuta , x.w .
zhang and m. ando , phys .
rev . lett.*74 * , 3828 - 3831 ( 1995 ) .
h .- d .
pfannes and u. gonser , appl .
phys .
* 1 * , 93 - 102 ( 1973 )
. h .- d .
pfannes and h. fisher , appl .
phys .
* 13 * , 317 - 325 ( 1977 ) .
j .- m .
greneche and f. varret , j. phys . *
c15 * , 5333 - 5344 ( 1982 ) .
w.a .
dollase , j. appl . cryst . * 19 * , 267 - 272 ( 1986 ) . r.a .
young and d.b .
wiles , adv.x - ray anal.*24*,1 - 23(1981 ) .
j. rodriguez - carvajal .
physica * b 192 * , 55 - 69 ( 1993 ) ; http://www-llb.cea.fr/fullweb/powder.htm .
f. millange , v. caignaert , b. domengs , b. raveau , and e. suard , chem . mater . * 10 * , 1974 - 1983 ( 1998 ) .
t. nakajima , h. kageyama , y. ueda , j. phys .
chem .
solids , * 63 * , 913 - 916 ( 2002 ) . t. nakajima , h. kageyama , h. yoshizawa , k. ohoyama , and y. ueda , j. phys .
. jpn . * 72 * , 3237 - 3242 ( 2003 ) .
a.j .
williams , j.p .
attfield , and s.a.t .
redfern , phys . rev . * b * * 72 * , 184426 - 1 - 13 ( 2005 ) .
v.g .
kohn , a.i .
chumakov , r. rffer , phys .
rev .
b. * 58 * , 8437 - 8444 ( 1998 ) .
t. ericsson and r. wppling , j. phys .
colloque * c6 * , 719 - 723 ( 1976 ) . d.l .
nagy , appl .
phys .
* 17 * , 269 - 274 ( 1978 ) .
best fits were obtained for the following coefficient vectors : a@xmath160=(4.34,-12.3 , 12.5,-5.35 , 0.84 ) , a@xmath0=(-5.27 , 16.1,-17.9 , 8.52,-1.47 ) , a@xmath161=(2.35,-7.85 , 9.47,-4.89 , 0.903 ) , b@xmath160=(11.5,-24.4 , 19.7,-8.22 , 1.33 ) , b@xmath0=(-7.79 , 24.2,-26.4 , 11.8,-1.77 ) , b@xmath162=(8.44,-29.6 , 36.5,-18.6 , 3.20 ) .
a.i .
rykov , a. ducouret , n. nguyen , v. caignaert , f. studer and b. raveau , hyperfine interact .
* 77 * ( 1993 ) 277 . since the structure of ybamn@xmath0o@xmath9 was monoclinic , the refinement of texture parameters was not applicable .
a.w .
mombr , k. prassides , c. christides , r.erwin , m. pissas , c. mitros , and d. niarchos , j. phys . :
cond . mat . *
10 * , 1247 - 1258 ( 1998 ) .
l. lutterotti , d. chateigner , s. ferrari , j. ricote , thin solid films * 450 * ( 2004 ) 34 - 41 . |
the intergalactic medium ( igm ) contains the majority of the universe s baryons .
the best available igm tracer is the intervening @xmath20 absorption forest observed in quasar spectra @xcite . contributing a lesser but still significant amount of absorption are intervening metal lines that trace the enrichment history of the igm ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
since metal absorption is often found far from the sites of star formation , analytic models and simulations strongly suggest that metals must be transported out from galaxies into the igm via large - scale galactic outflows ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the observation and interpretation of igm metal - line absorbers provides critical insights into the physics of galaxy formation .
first , galactic outflows that enrich the igm concurrently regulate the growth of galaxies by removing gas from star - forming regions .
second , the census of igm metals provides a unique and independent constraint on the integrated amount of star formation in our universe ; indeed , the metal census within galaxies falls significantly short of the expectations from cosmic star formation , which is known as the `` missing metals problem '' ( e.g. * ? ? ?
* ; * ? ? ?
finally , the amount and physical state of the metals in the igm alters how gas cools ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , thereby impacting the efficiency by which galaxies receive gas from the igm ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
hence studying igm metal absorbers provides unique insights into the cycle of baryons between galaxies and their surrounding gas . despite the importance of igm metal absorbers , the chasm between observations and a theory of metal lines remains wide .
many basic questions concerning igm chemical enrichment remain poorly understood . answering such questions
is now becoming feasible owing to the recent installation of the _ cosmic origins spectrograph _ ( cos ) on the hubble space telescope ( hst ) , whose central mission is to improve our knowledge of the low - redshift igm .
key questions include : @xmath21 _ what is the metallicity of the igm ?
_ at redshifts of @xmath22 measurements range between @xmath23 @xcite , while by @xmath1 a value of @xmath24 is typically quoted @xcite .
while this suggests an increasing metallicity with time , the interpretation is complicated by an evolution in how metals trace the igm baryon distribution .
owing to cosmic expansion , a given column density @xmath25 system traces a higher overdensity at lower redshifts @xcite ; a similar trend , modulated by evolving photo - ionisation rates , also exists for metals .
one consequence of these effects is that summing the column densities of metal lines to find an integrated ion density ( @xmath26 ) does not necessarily translate into a global metallicity ( @xmath27 ) .
@xcite argued that the non - trivial ionisation correction from @xmath3 to total carbon abundance evolved with redshift , resulting in a relatively flat @xmath28 that masks a significant increase in the metallicity of the igm .
these measurements have now been made down to @xmath29 ( e.g. * ? ? ?
* ; * ? ? ?
* ) , and cos should provide significantly more insights into the relationship between metallicity and density over most of cosmic time .
@xmath21 _ what temperature is the enriched igm gas ?
_ high - ionisation lines commonly observed in the igm such as @xmath3 and @xmath2 can arise both from photo - ionisation and collisional ionisation .
the low densities and hard quasar - dominated metagalactic flux can produce such high ionisation states , as can hotter gas in the warm - hot intergalactic medium ( whim ) of gas shock - heated during large - scale structure formation .
there is vigorous debate as to whether observed high - ionisation lines trace collisionally - ionised hot metals at @xmath30k ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) or low - density photo - ionised gas at @xmath31k ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* hereafter od09 ) .
the increased sensitivity of cos enables multiple ions to be detected in many systems , allowing independent constraints on the nature of high - ionisation systems .
@xmath21 _ how homogeneously are metals distributed in the igm ? _ if metals are well - mixed throughout the igm , metal lines will cool more baryons , possibly leading to increased gas accretion onto galaxies and hence an increase in star formation @xcite .
if metals are poorly mixed as observations suggest @xcite , then a census of @xmath2 can not straightforwardly be converted to a census of baryons ( e.g. * ? ? ?
* ; * ? ? ?
* ) since @xmath2 only traces the baryons in over - enriched regions ( od09 ; * ?
the analysis of observed metal - line systems often yields inconclusive answers , with complex absorbers suggesting multi - phase gas components ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . the higher signal - to - noise provided by cos will be critical for disentangling these effects .
a number of reasons exist for our less complete knowledge of metal absorbers compared to that of the @xmath20 forest .
observationally , metal lines are weaker and , therefore , harder to detect . moreover ,
many key ionisation states are buried within the @xmath20 forest or occur at x - ray wavelengths where current sensitivities are limited ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
theoretically , igm enrichment depends sensitively on the transport mechanism out of galaxies , which is poorly constrained , while the @xmath20 forest is less affected ( kollmeier et al .
2003 ; od09 ; * ? ? ?
* ; * ? ? ?
furthermore , converting metal ions into metallicities typically requires a detailed knowledge of the photo - ionisation rates at energies poorly constrained by the data or , if collisionally ionised , is highly sensitive to gas temperatures ( e.g. * ? ? ? * ) and hence detailed cooling rates ( e.g. * ? ? ?
* ) . however , there are many more metal transitions observable compared to hydrogen transitions , providing an opportunity to synthesise a wider suite of observations into a single coherent framework to determine the physical state of igm metals .
hydrodynamic simulations that evolve the metallicity and ionisation state of the igm self - consistently along with the enriching galaxy population offer a way to develop such a framework within a modern hierarchical context . in this paper ,
we employ state - of - the - art cosmological hydrodynamic simulations to study the @xmath32 igm as traced by high ionisation metal lines .
our strategy involves examining what can be learned by observations targeting quasar absorption line spectra randomly sampling the igm , which represents a volume - weighted measure .
we focus on high ionisation lines including @xmath2 , @xmath3 , @xmath4 , @xmath5 , @xmath13 , and @xmath14 .
low ionisation lines ( e.g. @xmath33 , @xmath34 , @xmath35 ) are not addressed as they trace higher density gas where self - shielding renders the assumption of a uniform ionisation background insufficient .
we briefly comment on @xmath36 and @xmath37 predictions .
this paper directly follows dav et al.(@xcite , paper i in this series , hereafter d10 ) , which discussed the @xmath20 forest and the physical conditions of baryons in the igm . the next paper in this series will study correlations between galaxies and absorbers ( kollmeier et al . in prep . , paper iii ) .
while we will show predictions for observable quantities with cos , we do not intend these to be used for direct comparisons with data .
we are more interested in showing how plausible variations in the physics manifest themselves in forefront observations , particularly in terms of answering some of the questions outlined above .
for instance , we will discuss how integrated column densities trace the underlying igm metallicity , how various ions observable with cos trace the temperature and density of the enriched igm baryons , and how absorber alignment statistics can constrain the homogeneity of metals in the igm .
we will do all this for several prescriptions of galactic outflows , to better understand how cos observations can constrain this important physical process of galaxy formation .
a description of our cosmological hydrodynamic simulations and our artificial spectra follows in 2 .
3 concentrates on the evolving physical conditions of metals , subdivided by phase between @xmath38 .
we simulate cos metal - line absorber observations in 4 and relate them to their physical conditions in 5 , demonstrating that only a fraction of igm metals are probed via uv resonance absorption lines .
we spend 6 exploring physically plausible model variations both to demonstrate the inherent uncertainties in interpreting metal - line observations and to distill what can be learned using relatively straight - forward statistics in a sample size similar to the one that cos will observe .
7 discusses the successes and failures of our simulations and compares them to simulations by other groups .
we conclude in 8 . for readers who want a briefer overview of the results
, we suggest first reading the conclusions section and following the sections listed after each of the main findings for more detail .
we utilise our modified version @xcite of the gadget-2 n - body + smoothed particle hydrodynamic ( sph ) code @xcite to evolve a series of cosmological simulations to @xmath1 .
the details of the simulations can be found in 2 of d10 ; they have also been used and described in @xcite .
briefly , our cosmology agrees with the wmap-7 constraints @xcite and uses the following values : @xmath39 , @xmath40 , @xmath41 , a primordial power spectrum index @xmath42 , an amplitude of the mass fluctuations scaled to @xmath43 , and @xmath44 .
we briefly recap the implementations of galactic winds in these simulations ; for a more extensive description see @xcite .
the basic technical implementation is the one described in @xcite and incorporated into gadget-2 . in a given timestep , a particle in a star - forming region has @xmath45 of being ejected , where @xmath46 is the mass - loading factor , sfr the star - formation rate , and @xmath47 the particle mass .
the ejection velocity , @xmath48 , is oriented according to @xmath49 , where @xmath50 is the particle velocity and @xmath51 is the gradient of the gravitational potential , resulting in winds propagating perpendicular to galactic disks . to allow the particle to escape the ism of its host ,
hydrodynamic forces are turned off until one of two criteria are met : either @xmath52 years have passed or ( more often ) the particle flows out to a density of less than 10% of the star formation critical density . in our constant wind ( cw )
model , we employ an ejection velocity of @xmath53 and a mass loading factor of @xmath54 for all galaxies .
the slow wind model has @xmath55 and @xmath56 .
both of these wind models are similar to @xcite , except for their assumed velocity ; they use @xmath57 .
the momentum - conserved wind model , inspired by the @xcite models of radiation pressure - driven winds , has @xmath58 where @xmath59 is the luminosity in units of the eddington luminosity required to expel gas from a galaxy potential , and @xmath60 where @xmath61 and @xmath62 is the galaxy s internal velocity dispersion calculated by a group finder run during the evolution of the simulation ( see * ? ? ?
the metallicity of an ejected wind particle is the metallicity of the sph particle at the time of ejection .
sph particles are continually self - enriched while they are eligible for star formation . in this scheme , both ejected winds and newly formed stars have the metallicity of the sph particle at the time of the ejection or formation event ; wind particles are essentially ejected ism sph particles . note that only particles above the critical star - formation density threshold are eligible to form stars or to become launched as a wind .
cooling , which is never suppressed , invariably allows the launched winds to cool and maintain a temperature of @xmath63 k while decoupled .
once the hydrodynamic forces are restored , which always occurs deep inside haloes given the two criteria listed above , the sph particle interacts normally . in practice ,
re - coupled wind particles show a bimodal behaviour : ( i ) they shock heat to temperatures above the peak cooling efficiencies for metal - enriched gas ( @xmath64 k ) , or ( ii ) they shock heat to a lower temperature where metal - line cooling is efficient and then rapidly cool to @xmath10 k. the most important determining factor appears to be @xmath48 , which we will show when we consider igm enrichment patterns among the different wind models in [ sec : metphys ] . our naming convention for our simulations continues the precedent we established in our recent papers : r[boxsize]n[number of particles per side][wind model ] , where the initial letter `` r '' indicates the particular choice of cosmology above .
the main simulations that we explore are evolved in a cubic periodic volume that is @xmath65 on a side : r48n384nw ( nw : no winds ) , r48n384cw ( cw : constant winds ) , r48n384sw ( sw : slow winds ) , and r48n384vzw ( vzw : momentum - conserved winds ) . these simulations were also used in @xcite , d10 , and @xcite .
the main simulations are listed in table [ table : sims ] .
the gas and dark matter particle masses are @xmath66 and @xmath67 , respectively .
lcccccccc name@xmath68 & @xmath69 & @xmath70 & @xmath71 & @xmath72 & @xmath73 & @xmath74 cooling@xmath75 & @xmath76 & @xmath77 + + r48n384nw & 48 & 384 & & & 0.0 & cie & 0.9 & 0.213 + r48n384cw & 48 & 384 & 680 & 2 & 0.96 & cie & 0.9 & 0.045 + r48n384sw & 48 & 384 & 340 & 2 & 0.24 & cie & 0.8 & 0.124 + r48n384vzw & 48 & 384 & @xmath78 & @xmath79 & 0.45 - 0.56 & cie & 0.67 & 0.097 + r16n128cw & 16 & 128 & 680 & 2 & 0.96 & cie & 0.9 & 0.040 + r16n128cw - pi & 16 & 128 & 680 & 2 & 0.96 & pi & 0.9 & 0.037 + r16n128vzw & 16 & 128 & @xmath78 & @xmath79 & 0.41 - 0.54 & cie & 0.67 & 0.096 + r16n128vzw - pi & 16 & 128 & @xmath78 & @xmath79 & 0.41 - 0.56 & pi & 0.67 & 0.088 + + [ table : sims ] four other simulations listed in table [ table : sims ] are evolved in a @xmath80 box with @xmath81 particles , the same resolution as the main simulations .
we use these to test the effects of metal - line cooling with photo - ionisation on metal - line observations , because @xcite and @xcite demonstrate that this could significantly alter observations of @xmath2 versus using collisionally ionised metal - line cooling rates . despite their smaller volume
, we will demonstrate in [ sec : zcool ] that these simulations results remain converged compared to the 48 @xmath82 simulations , likely because absorbers are mainly associated with sub-@xmath83 galaxies that are sufficiently sampled in both size volumes .
we add the suffix pi to those simulation names evolved using the photo - ionised cooling rates of @xcite .
spectra are generated as described in 2.3 of d10 .
briefly , 70 continuous lines of sight are created from @xmath38 for each model using our spectral generation code specexbin @xcite .
we generate mock spectra using only the strongest transition for an ion ( e.g. the 1031.93 line for @xmath2 , the 1548.19 line for @xmath3 ) and convolve the spectra with the cos line spread function ( lsf ) .
we add gaussian random noise with a signal to noise ratio s / n=30 per 6 @xmath84 pixel to simulate the highest quality data obtainable using cos .
we employ the continuum provided directly by the simulation , and do not attempt to mimic the observational procedure of fitting a continuum , since at low redshifts the sparseness of absorption features usually results in a robust and reliable continuum fit .
we fit absorbers in the spectra using autovp @xcite , obtaining a column density @xmath85 , line width @xmath86 , and redshift @xmath87 for each absorber .
we consider each metal transition separately and do not combine them into a single spectrum with all the transitions .
this assumes that , in observations , each transition can be reliably identified and fit without contamination from other lines , which is typically true at low redshifts .
the metagalactic flux amplitude is adjusted to match the observed evolution of the @xmath20 forest mean flux decrement .
we multiply the @xmath20 optical depth by the same values of @xmath88 that we calculated in d10 and presented in table [ table : sims ] of that paper . in practice , we renormalise our cloudy @xcite ionisation tables so that the magnitude of the ionisation background is multiplied by @xmath89 , which effectively shifts the ionisation fractions of all ions by @xmath89 in density under the assumption of an optically thin gas .
the vzw normalised ( * ? ? ?
* hm2001 ) @xmath90 background ( i.e. @xmath91 multiplied by @xmath92 ) is shown in figure [ fig : ionbkgd ] in black .
three other backgrounds are also displayed , which will be described in
[ sec : ionbkgdvar ] when we explore the effects of variations in the ionisation background .
flux decrement ( d10 ) .
three other backgrounds generated using cuba @xcite with varying slopes for quasar spectra above the lyman limit are also displayed ( hm2005 ) .
energies are listed in units of rydbergs as well as electron volts ( ev , top x - axis ) .
below are photo - ionisation cross - sections for various ions .
coloured regions show the magnitude of the cross - section from the ionisation state below to the specified ionisation state , while black regions show the cross - section from the specified state to the state above .
for these panels , levels of @xmath93 are indicated by large ticks on the @xmath94-axis and @xmath95 are indicated by small ticks , with the @xmath94-axis starting at zero in all cases . ]
we also display the photo - ionisation cross - sections of various ions used in cloudy @xcite on a linear scale in the bottom panels of figure [ fig : ionbkgd ] .
the coloured regions show the ionisation cross - section of the next lowest ionisation state to the specified ionisation state , while the black shading indicates the cross - section from that state to the next highest state .
for example , the magenta shading indicates the cross - section for @xmath96 to @xmath4 , and the black shading indicates the cross - section for @xmath4 to @xmath97 . with the exception of @xmath4 , all cases shown are lithium - like ions and display a trend of increasing ionisation energy and declining cross - section with higher atomic number .
metal ion cross - sections are complex , which is exemplified by the @xmath96 and @xmath4 cross - sections peaking above 10 ryd even though @xmath96 has the lowest ionisation energy ( 2.5 ryd ) of all the species we consider .
the multiple offset wedges , two in the case of lithium - like ions , correspond to different subshells in the ground state .
we follow the production of four heavy elements ( c , o , si , & fe ) from three sources : type ii supernovae ( sne ) , type ia sne , and agb stars as explained in @xcite .
the details of the enrichment scheme are in 2.1 of d10 .
cosmic oxygen and silicon production is dominated by type ii sne , while carbon production is augmented significantly by agb stars but still primarily arises from sne @xcite .
we do not follow neon and magnesium directly when generating @xmath5 and @xmath13 statistics but instead use oxygen as their proxy , which is a reasonable approximation given that all three elements arise primarily from type ii sne . the resulting abundances are [ ne / o]=0.16 and [ mg / o]=-0.21 for @xcite sne yields when using @xcite solar abundances , which we scale to throughout this paper whenever quoting metallicities
. we assume the same type ii sne origin for nitrogen by using [ n / o]=-0.40 , but note that this does not include secondary nitrogen production in agb stars , which is likely significant .
the solar metal mass fraction by weight , @xmath98 , is 0.0122 and the mass fractions of carbon , nitrogen , oxygen , neon , magnesium , and silicon are 0.00218 , 0.00062 , 0.00541 , 0.00103 , 0.00061 , and 0.00067 respectively .
we also do not consider depletion of these elements onto dust grains , although recent results suggest that there may be dust in the igm @xcite that contains as much as 50% of the mass of intergalactic metals .
the density - temperature phase space diagrams of metals residing in the igm are displayed in the left panels of figure [ fig : zphasespaces ] for our preferred vzw model at @xmath19 , 1 , and 0 .
the diagrams are coloured by metallicity while the brightness indicates the fraction of baryons .
on the right side are depicted the @xmath1 phase space diagrams for the other three wind models .
histograms along the x- and y - axes depict the relative fraction of metal mass at a given density and temperature , respectively .
the metals populate the cosmic phase diagram differently than the baryons ( see e.g. d10 ) with the density and temperature dependence of metallicity linked to the type of outflows that put them there , which we will discuss in [ sec : windphys ] .
here we focus on generic trends seen among all the wind models , contrasting them with the no - wind case .
following d10 , we divide the gas component into four phases , demarcated by the solid lines in the figures : * diffuse ( @xmath99 ) , * whim ( @xmath100 ) , * hot halo ( @xmath101 ) , * condensed ( @xmath102 ) , where @xmath103 k represents the division between the `` cool '' and `` hot '' phases , and @xmath104 represents the division between bound ( halo ) gas and unbound ( igm ) gas . these definitions are designed to provide a physically - motivated demarcation between cosmic baryon phases .
@xmath103 k is near the peak in the helium and metal cooling curve that causes a minimum in the temperature distribution of baryons , while the density threshold separates truly intergalactic baryons , i.e. those not associated with gravitationally collapsed structures , from baryons that lie predominantly within the virial radii of galaxy halos .
following @xcite , we set @xmath105 where @xmath106 @xmath104 evolves from a value of @xmath107 at high redshift to @xmath108 by @xmath1 . as in d10
, we adopt equation [ eqn : deltath ] to be 1/3rd the @xcite mean enclosed density , which corresponds to the density at @xmath109 for objects just undergoing collapse .
the condensed phase does _ not _ include star - forming ism gas , which has @xmath110 and lies to the right of the high - density edge of each diagram .
the dashed vertical line indicates the mean overdensity of the universe ( @xmath111 ) .
the percentages listed provide a fractional accounting of the total metals in all gas and star particles associated with each phase ; we also show the percentage of metals in galaxies in the lower right corner , and additionally divide this phase into stars and and star - forming ( ism ) gas . examining the evolutionary trends in the vzw model
, we see that the fraction of metals in the diffuse igm drops rapidly with time , from 18% to 3% from @xmath38 .
in contrast , the fraction in the whim phase is quite small ( @xmath112 ) and evolves downwards more slowly .
recall from d10 that the mass fraction of the whim in this model increases from @xmath113 during this time , while the fraction in the diffuse gas drops from @xmath114 .
hence overall , the fraction of metals in the igm ( i.e. diffuse+whim ) drops , a trend that is accentuated in the diffuse phase by a drop in the baryonic mass fraction in this phase , while for the whim the trend is countered by an increasing mass fraction .
the decrease in the igm metal fraction is balanced by an increase in the metal fraction in galaxies , from half at @xmath115 to almost @xmath116 at @xmath1 .
hence , from @xmath38 , fractionally the metals move from low - density regions to higher density regions .
we emphasise , however , that galaxies continue to enrich the igm up to the present day and that the metallicity of the igm does not actually decline ; only the _ fraction _ of metals in the igm declines .
we return to this point in [ sec : outsidein ] . by @xmath1
, we see that all the wind models have most of their metals within galaxies . in the no - wind case ,
only 3% of all metals are outside of galaxies , while in the cw case it is 37% .
the momentum - conserved and slow wind models are intermediate , with 1020% of their metals outside of galaxies .
still , these @xmath1 numbers are not large ; in all the wind models more metals are outside of galaxies at high-@xmath87 .
this means that the solution to the `` missing metals '' problem @xcite at early epochs depends more on the accounting of metals outside of galaxies than today .
examining the temperature histograms along the @xmath94-axis ( including only metals outside galaxies ) , we see that the gas - phase cosmic metals in all the models at all epochs have a maximum at @xmath10 k. this is an artificial temperature floor , since we do not include radiative cooling below @xmath117 k ; however , this also is nearly the equilibrium temperature where cooling equals photo - heating
. we will show in [ sec : zcool ] that when we examine models including metal - line cooling below @xmath117 k from @xcite , metals cluster around @xmath63 k , although with more variable temperatures extending above and below owing to different equilibrium temperatures dependent on density and metallicity .
the central result is that most cosmic metals are not in hot gas , either in the whim or hot halo gas . at @xmath1 , the igm metallicity distribution for all wind models
has a bimodal temperature distribution , with cool metals mostly at @xmath10 k and hot metals mainly residing at intracluster medium ( icm ) temperatures ( @xmath118 k ) .
few metals reside in between @xmath119 k and @xmath120 k ( except in the cw case ) because metal - line cooling is very efficient there , causing metal enriched gas to rapidly transition through these temperatures ( od09 ) . in [ sec : mixing ] , we state that the bimodality is stronger in our simulations compared to simulations that explicitly smooth metals , because metals confined to individual sph particles cool more rapidly over this intermediate temperature range .
the density histograms along the bottom of figure [ fig : zphasespaces ] for the vzw model ( left panels ) show a peak in metal fraction at @xmath121 at all redshifts .
this corresponds to an increasingly higher overdensity towards lower redshifts , reflecting the result from @xcite that the peak overdensity of metals increases with time .
this is qualitatively true in the other wind models as well ( not shown ) , although the actual distribution of metals in overdensity depends strongly on the particular wind model as we discuss in [ sec : windphys ] .
figure [ fig : zevol ] shows the mass fraction ( left panels ) and mass - weighted mean metallicity ( right panels ) of six baryonic phases ( stars - top , ism-2nd , diffuse-3rd , whim-4th , hot halo-5th , & condensed - bottom panels ) for our four wind models , from @xmath122 . at early epochs ,
star - forming gas holds the majority of metals .
as time passes , metals become progressively locked up into stars , so that by @xmath1 stars contain the majority of cosmic metals .
outflows suppress the metal content in stars and the ism somewhat but do not alter the overall trends .
the mean stellar and ism metallicities increase slowly from @xmath122 , but even at early epochs it is already close to solar , and by today it exceeds solar .
this is consistent with the early , rapid enrichment in galaxies typical in these models ( e.g. * ? ? ?
* ; * ? ? ?
* ) and is relevant for the igm because it shows that the material expelled from galaxies has near - solar metallicities at all epochs .
the metallicity of the igm depends then on how far the metals are driven , and how they mix with the unenriched ambient gas .
the metal fraction of the igm phases ( diffuse+whim ) generally drops from @xmath123 in the wind models .
this reflects the increasing difficulty that metals have in escaping from the denser regions where they are formed to less dense regions , as depicted in figure [ fig : zphasespaces ] . while the fraction of metals in the igm phases is generally not large , it is still sufficient to reproduce the observations of metal - line absorbers across cosmic time ( e.g. * ? ? ?
* ; * ? ? ?
* ) . for the whim and hot halo phases , the @xmath104 threshold that divides these two phases , first introduced in d10 , makes a stark difference in the whim metal content compared to previous whim definitions without such a density threshold .
the whim , which contains 24% of @xmath1 vzw baryons , holds only 1.2% of the @xmath1 metals and remains steady at [ z / h]@xmath124 since @xmath125 .
the whim forms through shock - heating owing to the gravitational collapse of large - scale structures @xcite while metal - enriching galactic outflows contribute fractionally less to the whim at late times @xcite .
the metal content of the hot halo phase steadily increases , and by @xmath1 it holds more metals than the combined igm phases ( cf .
5.2% vs. 3.9% for vzw ) .
the @xmath94-axis histograms for the vzw model clearly show the growing importance of hot halo metals from @xmath38 .
the @xmath1 panel indicates that most of these metals are at @xmath126 k and hence are in the intragroup or intracluster medium ( icm ) . in all the wind models , the hot halo metallicity reaches 0.1 @xmath98 at @xmath1 , but it also shows a density gradient such that the portion that would emit most strongly in x - rays has [ z / h]@xmath127 ( see figure [ fig : zphasespaces ] ) .
this is comparable to the measured metallicity of the icm ( e.g. * ? ? ?
* ; * ? ? ?
* ) and reconfirms that these simulations can generally reproduce the observed icm metallicities @xcite . subdivided by phase ( 1st - stars , 2nd - ism , 3rd - diffuse , 4th - whim , 5th - hot halo , & 6th - condensed )
four wind models are shown : solid blue- no - wind , dashed green- momentum - conserved winds , dotted red- constant winds , and long - dashed orange- slow winds . ]
more broadly , the fraction of metals in the igm increases at very early epochs as winds enrich diffuse cosmic gas @xcite , but then at later epochs the metals do not escape their haloes so easily or else re - accrete back into haloes . hence , the metal fraction in the diffuse phase peaks at @xmath128 in these wind models ( figure [ fig : zevol ] ) . meanwhile , the metallicity in all the phases increases , but slowly ; the most rapid increase is in the condensed phase gas , since this halo gas continues to be enriched even to late times by outflows .
these trends , while generally true , are in certain aspects quite sensitive to the wind model , as we will discuss next .
the value of studying enrichment patterns lies in their high sensitivity to galactic winds , upon which the process of galaxy formation heavily depends ( e.g. * ? ? ?
* ; * ? ? ?
we now contrast our vzw simulation to the three other wind models .
the no - wind model produces nearly twice as many stars and metals as any wind model , and is incapable of enriching the diffuse igm appreciably at any redshift , leaving it mostly pristine even by @xmath1 ( figure [ fig : zphasespaces ] , upper right ) . the necessity to enrich the diffuse igm to levels of at least [ z / h ] @xmath129 at @xmath130 ( e.g. * ? ? ?
* ; * ? ? ?
* ) is the main empirical requirement for large - scale galactic outflows .
the diffuse phase metallicity of the nw model remains below the y - axis in figure [ fig : zevol ] , only reaching [ z / h ] @xmath131 at @xmath125 and @xmath132 by @xmath1 .
this model does , however , achieve a hot halo metallicity comparable to our other feedback models , indicating that the enrichment of group and cluster haloes at least in part owes to a mechanism other than galactic outflows , such as dynamical stripping .
@xcite showed that their no - wind model reached observationally reasonable [ fe / h ] abundances via delayed type ia sne from intragroup / intracluster stars , but [ o / h ] abundances were too low compared to observations and the baryonic fraction in stars was too high .
overall , the no - wind model strongly fails to match basic observations of both galaxies ( e.g. * ? ? ? * ) and the igm .
the constant wind ( cw ) model has a high wind speed of @xmath133 emanating from every galaxy , which pushes shock - heated metals into voids .
this leads to a whim metal mass fraction more than 10 times higher than any other model ( figure [ fig : zevol ] , middle panel ) .
these metals remain hot , because adiabatic cooling from hubble expansion is not sufficient to return this gas to the photo - ionised igm equation of state ( eos ; * ? ? ?
* ; * ? ? ?
* ) within a hubble time . the excess heating is also reflected in the phase space diagram for this wind model ( figure [ fig : zphasespaces ] , lower right ) , which has a high - metallicity peak in underdense regions at @xmath134 k. the fast winds also increase the whim baryon percentage from 23 to 33% ( compared to the nw model ) , whereas the vzw and sw models increase the whim baryonic fraction by one tenth as much , a 1% increase ( d10 ) .
as @xcite show , a similar model with somewhat lower wind speeds ( @xmath57 ) produces @xmath3 absorbers at @xmath135 that are are too wide compared to the observations , indicating that there is too much igm heating at this early epoch . for the four models : momentum - conserved winds ( vzw ) , no winds ( nw ) , constant winds ( cw ) , and slow winds ( sw ) .
] the slow wind ( sw ) model with @xmath136 does not overly heat the early igm , and in some sense represents a hybrid of the vzw model ( similar average wind velocities ) and the cw model ( same mass loading ) . like cw , this model creates a concentration of shock - heated metals ( @xmath137 k ) in voids , but it is at lower temperatures and is much less prominent owing to the sw model s lower wind speeds . in this
model low - mass galaxies eject winds at high velocities relative to their escape velocity and high - mass galaxies , typically in dense environments , eject winds at less than the escape velocity .
hence , the sw model can not populate the icm with metals and , therefore , the metal phase space diagram in this regime mimics that of the no - wind model . in the vzw model , winds from small galaxies have lower velocities .
hence they deposit metals at higher densities where metal - line cooling is efficient and can return these metals to @xmath10 k. in hot gas , the vzw model produces a strong metallicity - density gradient by @xmath1 owing to its high outflow velocities from massive galaxies .
winds with @xmath138 speeds emanating from these galaxies rarely escape the halo owing to the intervening halo gas , but they do shock heat to temperatures high enough that the cooling times are long , and hence the metals do not re - accrete onto the galaxies .
figure [ fig : rhoz ] shows the mean metallicity - density relation in our four wind models from @xmath139 .
the no - wind model fails to provide widespread igm enrichment , but even dynamical processes can deposit a smattering of metals at the cosmic mean density by @xmath29 .
all the wind models show increased metallicities at all the plotted overdensities relative to the no - wind case .
the constant wind model results in a curious upturn in the mean metallicity at void densities ( @xmath140 ) , which is a generic feature of strong feedback models ( e.g. * ? ? ? * their figure 24 ) .
unfortunately this is very difficult to probe observationally using uv resonance lines owing to the low densities , as we will discuss in [ sec : absphyscond ] .
the slow wind model has a less pronounced upturn , and it shows a less widespread metal distribution particularly at early epochs owing to its lower wind speed .
the vzw model shows an interesting trend with overdensity , and the metallicity in filaments ( @xmath141 ) is nearly time - invariant .
continual enrichment from galaxies pumps metals into these regions , but this is approximately balanced by gravitational effects that draw the metals back closer to galaxies and by the accretion of more pristine gas . at higher overdensities , the mean metallicity increases with time , while at sub - mean densities , metals slowly diffuse deeper into voids @xcite .
the differences between the models offer an opportunity to probe outflows if the metallicity - density relation and its evolution can be reliably measured , which is a central goal of igm enrichment observations .
overall , the evolution of metal content by phase shows much more dependence on feedback than our corresponding study of the abundance of baryons by phase ( d10 ) .
while all the wind models significantly suppress the global star formation rate and hence metal production , the distribution of metals in the igm shows clear differences .
the velocity of the outflow and its motion through the ambient gas determines to what temperature the gas is shock heated and to what overdensity it reaches .
gas in sufficiently dense regions cools quickly , particularly since it is enriched .
hence , the outflow parameters critically determine the enrichment history of baryons outside of galaxies , providing an opportunity to constrain such parameters based on observations of igm metal lines .
a generic feature of our igm enrichment models is that they have a very early epoch where metals are distributed outwards into the igm , but over most of cosmic time the newly produced metals became confined to ever decreasing regions around the galaxies that produced them .
we call this outside - in " igm enrichment , and it has consequences not just for how metals enrich the igm , but for how wind material re - accretes onto galaxies @xcite . we have already seen trends exemplifying outside - in enrichment above .
the metal mass fraction in the igm decreases steadily since @xmath135 , and the peak overdensity of metals increases with time .
@xcite showed that the median distance reached by vzw outflows ( with a wide dispersion ) is @xmath142 physical kpc , implying widespread enrichment ( in a comoving sense ) early on and more confined enrichment at later epochs .
the evolution of the vzw metallicity - density relationship in figure [ fig : rhoz ] ( top panel ) shows that metals are in place at igm overdensities at @xmath143 and that the metallicity is mostly invariant and actually declines slightly at @xmath144 from @xmath38 .
newly produced metals are mostly confined to their own haloes and fall back onto their parent galaxies primarily from these overdensities ; however winds continue to enrich igm densities until @xmath1 , especially the winds from lower mass galaxies , often even reaching void densities .
additionally , some enriched whim material also falls into galaxy haloes and eventually onto the central galaxy .
another view of outside - in enrichment is provided by the age of @xmath1 metals as a function of overdensity , shown in figure [ fig : rhoageomega ] , top panel .
we define the metal age as the time since a wind particle has most recently been launched , which is a good indicator of the age for metals in the igm as these baryons are almost entirely enriched by winds .
the median age of wind - blown metals is shown by the solid lines , and the 16 and 84% cuts are indicated by the dashed lines for the vzw model .
figure [ fig : rhoageomega ] shows directly that lower overdensities are enriched earlier , particularly in the vzw model . for this model ,
the typical age of metals in voids is @xmath145 gyr , reaching a minimum of @xmath146 gyr at overdensities of a few thousand .
the constant-@xmath46 models show less dependence on age with overdensity in the igm , though at higher overdensities the metal ages are again significantly younger .
the evolution of the metallicity - density relation of the cw model ( figure [ fig : rhoz ] ) shows a significantly increasing metallicity within voids with time .
most metals were already in place at higher redshift in gas that continues to expand into voids by @xmath1 , but galaxies continue to enrich void overdensities below @xmath147 .
the same trends are true for the sw model , but they are shifted to higher overdensities and younger ages owing to its weaker winds . while the median age trends with overdensity are statistically significant and indicative of the history of metal enrichment , the larger scatter in all models at most densities prohibits a direct mapping of an overdensity to an age for a given wind model .
@xcite find a similar age - density anti - correlation using their definition of mean enrichment redshift .
our definition of age is a lower limit compared to theirs , because we measure the most recent wind launch as opposed to the redshift - weighted enrichment age .
both methods will find similar ages for wind particles that rapidly enrich and launch once , as is typical for metals that escape galaxies at high-@xmath87 , but we will find significantly younger ages for winds that recycle back onto galaxies and have multiple wind launches and enrichment epochs . as a function of density ( solid lines ) with the 16th and 84th percentage cuts ( dashed lines , vzw only ) for the momentum - conserved winds ( vzw , green ) , constant winds ( cw , red ) , and slow winds ( sw , orange ) .
lower overdensities are enriched earlier in all wind models in what we term `` outside - in '' enrichment .
the bottom panel shows cosmic ion densities of vzw metal species as a function of density , as well as , @xmath25 ( purple lines ) and total metal density ( grey lines ) .
the two panels can be compared to show that in general higher ionisation species generally trace lower overdensities , which have older metals . ] to relate the igm metal age to observables , we plot in the lower panel of figure [ fig : rhoageomega ] histograms of the @xmath1 vzw cosmic ion densities ( in 0.1 dex bins of overdensity ) for the ions that we follow and for @xmath25 . in general , higher ionisation metal species trace older igm metals occupying lower overdensities in the vzw model .
for example , observed @xmath5 and @xmath2 absorbers are likely to be tracing fairly old ( @xmath148 gyr ) metals , while @xmath3 and @xmath4 more often trace younger metals , although with a large age scatter .
every metal species shows a peak corresponding to the overdensity where the species is primarily photo - ionised .
a second peak at higher overdensities for @xmath14 and @xmath13 corresponds to collisionally ionised metals inside haloes .
@xmath5 also shows an extended shoulder toward halo densities corresponding to collisionally ionised metals .
this figure foreshadows our analysis in [ sec : absphyscond ] showing that species including @xmath3 , @xmath2 , and @xmath5 are primarily photo - ionised in our simulations .
interestingly , these three species exceed the cosmic density of @xmath25 at a range of igm densities .
the idea that the igm is enriched in an outside - in manner has a long history @xcite . in general
, these models postulated that the igm underwent widespread early enrichment typically during the reionisation epoch from some putative powerful source such as population iii stars , when halo potential wells were still small .
our model , while similar in character , is fundamentally different in that it does not invoke very early enrichment from mysterious sources .
instead , the outside - in pattern arises from the dynamics of galactic outflows from the epoch of early galaxies ( @xmath149 ) until today .
the source of igm enrichment is the observed population of high-@xmath87 star - forming galaxies that our simulations naturally reproduce @xcite .
energetically moderate outflows from these galaxies enrich a small , but rapidly growing fraction of the igm , in agreement with observed constraints from high-@xmath87 metal - lines ( e.g. * ? ? ?
these early outflows end up enriching a greater comoving volume than lower redshift outflows , which travel similar physical distances in our model @xcite .
in contrast , @xcite showed that enriching the igm entirely prior to @xmath150 would violate observations of the @xmath3 enrichment at those epochs .
we present metal - line predictions for cos observations in this section along with comparisons to other pre - cos data .
we consider statistics assuming a s / n=30 per @xmath151 pixel in our simulated spectra , convolved with the cos line spread function ( lsf ) assuming the fuv g130 m lsf at 1450 , just as we did in d10 .
we model the lsf as a central peak with a @xmath152 gaussian width of @xmath153 and substantial non - gaussian wings , keeping this fixed for all wavelengths to avoid introducing artificial evolution from a varying lsf .
we obtain line statistics by fitting voigt profiles to individual components using autovp @xcite .
each component has a fitted rest equivalent width ( @xmath154 ) , a column density ( @xmath85 ) , and linewidths ( @xmath86-parameters ) .
an absorption system is defined as all components that can be grouped together if they lie within 100 @xmath84 of another component .
systems are more easily comparable with observations , because data quality does not always allow the easy and consistent identification of components . for example
, it can be easier to divide a system into more components if the s / n and resolution are higher .
we sum up system column densities to calculate cosmic ion densities ( @xmath155 ) .
we discuss the equivalent width distribution of systems versus components for @xmath2 , where observations for both have been quantified .
we present results in this section for our four main simulations , plus a vzw model with turbulent broadening added after the completion of the simulation ( od09 ) , motivated by the observed @xmath2 linewidths .
cumulative equivalent width distributions ( ewds ) , presented in figure [ fig : ew ] , show the frequency of systems greater than or equal to the @xmath154 indicated on the @xmath156-axis .
they offer a simple and direct statistic to test theoretical models .
we show ewds of four metal lines : 1032 , 1548 , 1394 ( all between @xmath157 ) , and 770 ( at @xmath158 ) .
we show predictions for our four wind models and a fifth model ( vzw - turb in cyan ) of turbulent @xmath86-parameters ( @xmath159 ) added after the completion of the simulation using specexbin @xcite .
the ewds with no winds ( blue solid lines ) exhibit far fewer absorbers than models with winds , and significantly underestimate the observations of @xmath2 , @xmath3 , and @xmath4 . in general , the cw model ( red dotted lines ) produce fewer absorbers than the other wind models because this wind model ( i ) produces less metals overall owing to its greater suppression of star formation and ( ii ) injects more of its metals into the low - density whim where they are unobservable using these transitions .
compared to vzw , the sw model creates fewer @xmath3 and @xmath4 absorbers , which we will show in [ sec : absphyscond ] trace the condensed phase within haloes that are enriched to higher levels by vzw winds . at the same time there
exist more weak @xmath2 and @xmath5 lines in the sw model , because these arise from the diffuse phase that is enriched to higher levels via sw winds ( cf . vzw and sw @xmath1 metallicity - density relations in figure [ fig : rhoz ] ) .
we discuss turbulence as a model variation in
[ sec : turb ] , but we present it here because this is part of our favoured model for @xmath2 observations .
we also show the effect on @xmath3 , @xmath4 , and @xmath5 , although we argue later that this model may not be applicable to absorbers tracing higher densities .
as in od09 , adding @xmath160 ( see equations [ eqn : bturb1 ] & [ eqn : bturb2 ] in [ sec : turb ] ) increases the number of absorbers above 100 m , allowing us to fit the observed @xmath2 ewds , and it also blends some weaker components into stronger ones , thus reducing their numbers ( cf . vzw and vzw - turb model in the top panel of figure [ fig : ew ] ) . @xmath3 and @xmath4
also show a dramatic increase in the number of @xmath161 m components if we apply this same density - dependent model .
the simulations presented here produce statistically identical @xmath2 results to od09 for the vzw model
. however , the vzw model appears to over - produce the observed @xmath3 , while the cw model provides a better fit to this data @xcite and @xmath4 , which can be found at http://www.ucolick.org/@xmath162xavier/hstciv/ and http://www.ucolick.org/@xmath162xavier/hstsiiv/ ] .
we predict that cos will make many @xmath5 detections , especially between 10 - 20 m . in general ,
the lithium - like ions are harder to detect for higher atomic numbers corresponding to higher ionisation species , because both their oscillator strengths and wavelengths decline , leading to smaller equivalent widths for a given column density .
column densities provide a more direct accounting of the amount of metals present than equivalent widths , but they are more difficult to obtain from observations as cos often does not resolve individual metal - line absorbers .
nevertheless , autovp fits column densities @xmath85 and linewidths @xmath86 for every component .
given that we have applied cos instrumental characteristics ( albeit with a perhaps optimistic s / n ratio ) , these predictions should be representative of the level of information obtainable from ( the very best ) cos spectra .
we emphasise that these are distributions of inferred column densities obtained by voigt - profile fitting components and summing them together into systems , and not necessarily distributions of true column densities of physically identified , unblended components .
we display the differential column density distributions ( cdds ) of systems in figure [ fig : cdd ] . as in d10
, we plot @xmath163 multiplied by @xmath164 to improve the readability and to provide a dimensionless number that indicates the relative number of lines per unit redshift for a given @xmath165 .
again our four basic wind models are plotted , along with turbulent broadening added to the vzw model .
we show eight species over @xmath166 for a redshift range that falls within the cos far - uv ( fuv ) channel : @xmath2 , @xmath3 , @xmath4 , & @xmath36 all at @xmath157 , @xmath5 at @xmath158 , and @xmath37 , @xmath13 , & @xmath14 at @xmath167 .
the qualitative differences between the wind models and the data for the column density distributions are similar to those obtained for the equivalent widths as we discussed in the previous section . without winds ,
very few igm metal lines are present , and the predictions fall well short of available observations . the @xmath2 cdd is consistent with the data @xcite , @xmath3 , @xmath4 , and @xmath36 . ] for the vzw and vzw - turb models , but the same models produce too many @xmath3 absorbers , especially above @xmath168 , compared to the observations @xcite .
this disagreement is statistically significant , and further data from cos should provide much better constraints on weak @xmath3 absorbers .
@xmath4 predicts more strong lines and less weak lines than those observed @xcite , though the discrepancy is not as large as with @xmath3 , and the @xmath169 redshift range of this data set is larger and centred at much higher redshift than in our simulations ( @xmath170 vs. 0.25 ) . despite the different redshifts for @xmath4 , our simulated @xmath4 cdds remain statistically identical with @xmath171 and @xmath172 over the column density range of detected absorbers ( not shown ) .
we show @xmath36 cdds but caution that we do not follow the unique chemodynamic origin of nitrogen from secondary enrichment sources ( agb stars ) .
hence , the type ii sne origin assumed here for nitrogen likely represents a lower limit , as nitrogen is copiously produced in agb ejecta ( e.g. * ? ? ?
our models underestimate the @xcite statistics for higher @xmath36 columns , which appears to support this interpretation . for @xmath5 ( upper right panel of figure [ fig : cdd ] ) , pushing below @xmath173 is rewarding because the cdd rises sharply . at column densities @xmath174 ,
the line counts are predicted to be well under one per unit redshift , but at @xmath175 there should be @xmath176 per unit redshift . hence high s / n is essential to detect significant numbers of these absorbers .
note that our estimates are probably optimistic because we assume detection based on only the stronger component , but in actuality detection of the weaker 780 component is usually required for a robust line identification .
we predict @xmath37 should be generously abundant in the spectra of @xmath177 quasars probed by cos over a wide range of column densities .
we show the 630 @xmath37 statistics , because it is likely representative of a large number of lower ionisation species with far - uv transitions that should contribute significantly to intervening absorption in higher redshift quasars probed by cos .
@xmath37 should be among the strongest owing to a large oscillator strength and oxygen being the most abundant metal .
@xmath13 and @xmath14 absorbers are very rare in our sight lines and definitely require winds to be present at all .
we find only 2 @xmath13 and 6 @xmath14 absorbers over @xmath178 with @xmath179 m in the vzw model .
we argue in [ sec : absphase ] that we may not be resolving the conductive interfaces from which very high - ionisation lines such as @xmath5 and @xmath14 may arise , and hence that these absorbers may be more common than our current models predict .
looking at the vzw - turb model , in theory the total column density should be conserved ( unlike @xmath154 ) when adding turbulence in the way that we do using specexbin . but as one can see , in practice the column densities are increased when turbulence is added .
this occurs because for strong saturated lines , which preferentially receive the largest turbulent contributions , autovp tends to fit a higher column density given only a single line with no doublet information .
also , weaker lines may become blended or harder to detect .
hence , there is a noticeable increase in the predicted line frequencies especially for @xmath3 at @xmath180 and @xmath4 at @xmath173 , which correspond to the column densities where these transitions fall off the linear curve of growth and begin to become saturated without turbulent broadening .
cos spectra will be able to use the two doublet transitions to perform a curve of growth analysis and obtain more accurate @xmath85s for saturated components .
including such information in autovp may mitigate the increase seen at high-@xmath85 ; we leave such an analysis for a more careful comparison to real data .
integrating over a cdd or summing the column densities of lines provides a global view of the amount of cosmic metal absorption . to obtain the cosmic density of an ion
, we sum up all the absorber column densities between @xmath181 using @xmath182 where @xmath183 is the hubble constant , @xmath184 is the speed of light , @xmath185 is the critical density of the universe , and @xmath186 is the atomic weight of the given species .
@xmath187 is the pathlength in a @xmath188cdm universe where @xmath189 .
we evaluate equation [ equ : omega ] from the voigt - profile fitted column densities , just as is done for observations .
we choose the column density range both for historical reasons @xcite , and because it quantifies lower - column metal lines that are expected to arise in the igm as opposed to those within the galactic ism .
we bin the @xmath26s in bins of @xmath166 and plot the points with error bars in figure [ fig : omega ] for @xmath2 , @xmath3 , @xmath4 , and @xmath5 . for simplicity
, we use the same cos fuv instrument parameters for the entire @xmath32 range even though most redshifts fall outside of the cos wavelength range .
we show the same four wind models as before , and also the vzw model with turbulent broadening .
we also plot @xmath27 computed by summing up the total mass density of all the sph particles outside of galaxies ( @xmath190 ) for the appropriate metal species ( continuous lines ) .
@xmath26s usually do not change by much if we adjust the column density bounds from @xmath191 to @xmath192 or from @xmath193 to any higher value , because the vast majority of an ion s cosmic density resides in @xmath194 components in s / n=30 spectra .
the most dramatic exception is @xmath3 at @xmath195 , which can increase by a factor of more than two when including lines above @xmath196 .
models with outflows produce roughly constant @xmath197 and @xmath28 from @xmath38 , and agree with observations where available . while @xmath2 agrees with the data in all three observables
thus far explored , @xmath3 only agrees for @xmath28 because too many weak and too few strong absorbers exist compared to the current data @xcite . in general , the non - evolution in these species reflects the non - evolution in the total mass densities of metals outside of galaxies ( @xmath27s , lines in figure [ fig : omega ] ) , but the ion densities are shifted lower by a factor of @xmath198 .
the majority of the universe s metals are produced after @xmath19 , but this is offset by more metals remaining nearer to galaxies and being reincorporated into stars , i.e. outside - in " enrichment .
note that the ratio @xmath199 can not be treated as a global ionisation correction for an ion ( e.g. c/@xmath3 or o/@xmath2 ) , because we will show in [ sec : omegacorr ] that @xmath26 calculated using equation [ equ : omega ] does not recover the total cosmic density of that ion .
in contrast to the other ions shown , @xmath200 shows a steady decline from @xmath201 for the vzw model .
the photo - ionised @xmath5 component is within the detection limits of cos at higher redshift , and becomes harder to detect at low redshifts at wavelengths only accessible using the _ space telescope imaging spectrograph _ ( stis ) and the _ far uv spectroscopic explorer _ ( fuse ) .
observing @xmath5 at wavelengths between 1100 - 1700 with cos could be ideal to trace the peak in @xmath200 . the turbulent vzw model predicts higher @xmath155s than the vzw model even though the total column density theoretically should be preserved . as explained previously , the broader profiles with turbulent broadening are fitted more accurately by autovp than unbroadened thin saturated profiles .
this is especially true if the cosmic density of an ion is dominated by a few strong absorbers . in summary
, outflows are generically required to enrich the igm as observed , and there is a mild sensitivity in observations of @xmath154 , @xmath85 , and @xmath26 to the particular form of galactic outflows employed . while momentum - conserved winds plus turbulence best matches the @xmath2 data ( as it was in part designed to do ) , it overproduces @xmath3 and @xmath4 absorption .
@xmath5 shows a particularly steep column density distribution owing to a preponderance of weak photo - ionised lines , motivating deep observations to detect this interesting igm tracer .
the cosmic metal density is generally constant from @xmath202 , in contrast to its marked rise from @xmath203 @xcite , and this is broadly reflected in the constancy of @xmath26 for most ions over this redshift range .
component linewidths ( @xmath86-parameters ) are also fit with autovp , but are almost always dominated by the cos lsf ( the instrumental @xmath204 ) ; therefore , linewidths are often much wider than they are in stis data .
we show the linewidth histograms for the simulations with winds in figure [ fig : bparam ] along with the vzw - turb model , which uses the turbulent broadening model motivated by od09 to match the observed @xmath86-parameters of stis @xmath2 absorbers , but now applied to all ions .
the histograms apply to all @xmath2 , @xmath3 , and @xmath4 absorbers with @xmath205 m between @xmath157 , and @xmath5 absorbers with @xmath179 m at @xmath158 . without turbulent broadening ,
the linewidths of the three wind models are nearly identical .
the cos lsf makes it difficult to determine temperatures from thermal broadening in our case since we show in the next section that the majority of our absorbers are near @xmath10 k , corresponding to linewidths much smaller than the lsf .
we do not show the nw model owing to its low number statistics . adding turbulence using the heuristic prescription of od09
dramatically alters the @xmath86-parameter histograms for not just @xmath2 , to which this prescription was calibrated , but even more so for @xmath3 and @xmath4 .
autovp fits a maximum @xmath206 , which means that the majority of @xmath3 and @xmath4 absorbers in the 45 - 50 @xmath84 bin may even be wider . unlike thermal broadening ,
turbulent broadening is independent of the atomic species , and the broader linewidths for @xmath3 and @xmath4 are a result of the od09 turbulence model , which adds wider @xmath159 to higher density gas .
we argue in [ sec : turb ] that such wide absorbers likely represent a break - down of the turbulence model , because while there are few published observed linewidths of these species , thinner line profiles are apparent in the data ( e.g. * ? ? ?
@xmath5 absorbers are not as sensitive to the turbulence model , because they arise from lower densities where @xmath159 is small .
the choice to group components into systems or treat them individually is an important consideration when comparing simulations to observations .
k. cooksey ( private communication ) advises us that it is more proper to compare grouped systems to the data of @xcite , because they group components into systems if they lie within a few tens of @xmath84 .
@xcite groups metal - line components together into systems if their associated @xmath20 lines lie within @xmath207 of each other ( c. danforth , private communication ) .
@xcite publish both component and system ewds , which we show in figure [ fig : ewcompsys ] .
the vzw model indicates how important it is to compare the appropriate component and system ewds .
the vzw system ewd would agree with the component ewd of @xcite , but the appropriate comparison , the vzw component ewd , under - predicts the strongest observed @xmath2 components .
the shortfall of strong @xmath154 components was part of the motivation for od09 to attempt the vzw - turb model ( the other motivation being mismatched linewidths ) .
the vzw - turb makes less of a difference between the two types of ewds , because turbulent broadening increases the component @xmath154s so that the strongest component of a system becomes more dominant relative to weaker components ; thus the strongest component more likely sets the system @xmath154 resulting in less differences between the two methods .
the trend of this model s component and system ewds matches that of @xcite , although we appear to consistently over - estimate component and system frequencies at @xmath208 m .
the comparison to @xcite is more difficult because their search relies on the identification of the associated @xmath25 absorption , but their measurements appear consistent with @xcite systems .
we also tested the choice of grouping components within 250 @xmath84 instead of 100 @xmath84 , but this makes almost no difference and gives statistically identical results .
if components lie close to each other , they are almost always within 100 @xmath84 of another component .
we now link the observed metal - line absorber characteristics with their physical conditions just as we did for @xmath20 absorbers in 5 of d10 .
we concentrate on ion mass density - weighted physical parameters , mainly hydrogen density and temperature , using the same method described in od09 .
this weighting procedure sums the ion mass density , and hence absorption , of every sph particle contributing at the line centre of an identified absorber , accounting for peculiar velocities and temperature broadening .
this procedure is similar to the optical depth - weighting of @xcite with the only difference being that they sum physical quantities from each pixel across a line profile whereas we use the value at the line centroid .
@xcite find that the difference between these two methods is small , so the methods should be comparable .
we consider all absorbers where the strongest line transition has @xmath209 m in our 70 s / n=30 cos sight lines .
figure [ fig : zionphase ] displays total metals ( top panels ) and selected ions ( middle and bottom panels ) in density - temperature phase space at @xmath90 ( left ) and @xmath210 ( right ) .
the grey pixelated shading is the fractional @xmath155 summed from sph particles binned in @xmath211 dex pixels .
the temperature bimodality of metals described in
[ sec : rhotcond ] is obvious at both redshifts , with the metals pushing to lower densities and higher temperatures with time .
we show the phase space occupied by @xmath3 and @xmath2 absorption at @xmath90 as the grey shading in the middle left and bottom left panels , respectively .
these ions trace gas in a limited range in phase space , particularly weighted towards the cooler gas .
overlaid using coloured open circles are the ion mass density - weighted hydrogen densities and temperatures of simulated absorbers over a pathlength of @xmath212 ( 70 sight lines over @xmath213 ) .
colour and circle size both scale with the absorber column density as indicated by the legend above each panel .
our identified absorbers overlap with the darkest shading , which provides a consistency check that our ion weighting accurately captures the physical conditions of the absorbers .
the two sets of histograms along the @xmath156-axis ( for density ) and @xmath94-axis ( for temperature ) directly compare the sph - summed quantities ( solid grey histograms ) and the absorber - summed quantities ( open red histograms ) ; the amplitudes of these histograms can be directly compared .
for example , the mismatch of @xmath3 density histograms in denser halo gas indicates that summing the halo @xmath3 absorbers ( red ) underestimates the true @xmath28 arising from this phase ( grey ) .
the temperature histograms show that most of this missing absorption is between @xmath214 k. the same is true for @xmath4 probing even higher densities ( not shown ) .
the main cause for this is that saturated absorbers tracing halo gas have their column densities underestimated without turbulent broadening , which is apparent in the differences between the vzw and vzw - turb cdds in figure [ fig : cdd ] .
the mismatch is less severe but still noticeable for @xmath2 . at @xmath90 ,
both @xmath3 and @xmath2 predominantly trace the cooler phases of gas at temperatures closer to @xmath117 k than @xmath119 k. @xmath36 ( not shown ) traces the same temperatures , but at intermediate overdensities between @xmath3 and @xmath2 . for @xmath2 ,
this is in marked contrast to some previous work @xcite , all of whom find most of their @xmath2 at @xmath215 k at the same redshifts .
we believe it is the interplay of our momentum - conserved winds with the cooling rates in the @xmath216 k regime that results in our differing conclusion .
we will show in [ sec : zcool ] that using the photo - ionised metal - line cooling rates of @xcite does not fundamentally change the phase space traced by @xmath2 absorbers in our vzw model .
hence , we reiterate the conclusion by @xcite that , for a physically well - constrained outflow model , @xmath2 primarily traces diffuse , photo - ionised metals .
we show two higher ionisation species , @xmath5 and @xmath14 , at @xmath210 , because this is the redshift where cos is able to observe these lines .
the @xmath14 doublet at 499.3,521.1 falls into the cos sensitivity range at @xmath217 , but we show simulated absorbers and phase space centred at @xmath210 because these characteristics are not strongly evolving .
@xmath14 has a larger fraction of mass arising from collisional ionisation at densities corresponding mostly to halo gas .
surprisingly , most simulated @xmath5 absorbers are predicted to be photo - ionised , tracing overdensities of around @xmath218 .
however , the strongest simulated absorber is at @xmath219 tracing @xmath220 k gas , and is likely more representative of the three @xmath5 absorbers @xcite already observed , since it has a similar column density and hence probably a similar environment .
@xcite find a 0.25 @xmath221 galaxy at an impact parameter of 73 @xmath222 at nearly the same redshift as the @xmath223 @xmath5 absorber along the he0226 - 4110 sight line , suggesting that this absorption originates in collisionally ionised halo gas .
they further suggest that the origin is a conductive interface between cool clouds moving through a hot medium , based on the alignment of @xmath5 with lower ionisation species , something that would not be captured in our simulations .
@xcite note the existence of a 0.08 @xmath221 galaxy at 110 kpc at @xmath224 from the @xmath5 absorber at @xmath225 in the pks0405 - 123 sight line , and further argue this is likely tracing @xmath226 k gas either in a hot halo or nearby whim .
we emphasise that our model predicts numerous weaker @xmath5 absorbers tracing the diffuse photo - ionised component , as a separate and probably as yet undetected population .
these absorbers are mainly below @xmath227 ( cf .
@xmath228 , @xmath229 , @xmath230 for the discovered systems mentioned above ) and should arise outside haloes .
we can not distinguish any other obvious observational characteristics between the two types of @xmath5 absorbers , as the simulated @xmath86-parameters at cos resolution do not straightforwardly translate to temperature .
alignment with @xmath25 absorbers tracing overdensities of @xmath145 , i.e. @xmath231 at @xmath232 , is common for the photo - ionised @xmath5 absorbers .
upcoming deeper cos observations might be able to detect this population .
the true frequency of hot @xmath5 absorbers may exceed our predictions given that our simulations do not adequately resolve conductive interfaces within haloes .
@xcite estimates a higher frequency below @xmath233 , although their estimate is based on one absorber and is highly uncertain .
additionally , we predict a mass density of @xmath200 at @xmath210 almost @xmath234 times higher than at @xmath90 , which is suggestive of an even greater frequency of such absorbers at high redshift ; the increase is primarily from the increased neon traced in the photo - ionised component . given their distinct phase spaces ( hot halo and dense whim vs. diffuse ) , the potential for absorber bimodality is greater than for @xmath2 , but identifying it may require direct observation of an absorber s environment .
we will return to this discussion in a future paper discussing the galaxy - absorber connection .
it may initially be surprising that high ionisation species such as @xmath5 , @xmath13 , and @xmath14 can be excited via photo - ionisation given their high ionisation potentials , but the existence of such absorbers is straightforwardly traceable to the shape of the @xcite background .
we demonstrate this in figure [ fig : lox ] , where we show the line observability index ( lox , * ? ? ?
* ) the predicted equivalent width of metals lines associated with @xmath25 if the igm is uniformly painted with a metallicity of @xmath235 and with a uniform temperature of @xmath236 k. although these assumptions are overly simplistic , they are representative and serve to illustrate the main point . at @xmath210 , @xmath5 peaks near @xmath237 with equivalent widths as high as 8 m .
our simulated photo - ionised @xmath5 absorbers are stronger than this owing to the inhomogeneous enrichment , which results in higher metallicities for the detected absorbers , and furthermore arise at slightly higher densities owing to the positive metallicity - density gradient present in the vzw models @xcite .
photo - ionised @xmath14 should exist around the mean density ( assuming the metallicity is sufficiently high ) , but it should not exceed even 1 m , well below the detectability threshold of cos .
photo - ionised @xmath13 exists at higher overdensities but is still very weak . a detection of @xmath13 or particularly @xmath14 ,
while likely rare and exceedingly difficult , would provide the most unambiguous uv tracer of hot halo gas at @xmath238 k. the phase space traced by @xmath2 at @xmath90 is shown for the constant , slow , and no - wind models in figure [ fig : zmodelionphase ] . while the cw model appears qualitatively similar to vzw with most absorbers being diffuse and photo - ionised , there are significantly fewer absorbers at @xmath239 .
stronger outflows create fewer strong absorbers , because metals preferentially reach lower overdensities where they can not rapidly cool .
even though the igm is generally hotter in this model relative to vzw , absorbers still remain predominantly photo - ionised . despite the metal content of the cw whim being @xmath240 times higher than in the vzw model , the limited phase space traced by the @xmath2
does not probe this enrichment , which can only be directly detected in species with x - ray transitions ( e.g. @xmath241 and @xmath242 ) .
slow winds produce @xmath2 statistics nearly indistinguishable from the vzw model as reflected in the phase space plot .
this same simulation produces a similar @xmath29 galactic stellar mass function to the vzw model below @xmath83 @xcite , thus the match with the @xmath2 statistics is consistent with sub-@xmath83 galaxies producing the @xmath2 absorbers ( od09 ) .
the no - wind simulation shows that winds are required to enrich the diffuse igm ; no absorbers exist in this phase .
overall , the conclusion that the @xmath2 detected by cos ( at least under the s / n assumptions here ) will be predominantly photo - ionised is robust to variations in the wind model and the cooling rates ( as we will show in [ sec : zcool ] ) .
even @xmath5 , at columns just below what is currently accessible , becomes dominated by photo - ionised absorbers .
@xmath14 at cos sensitivities probes hot gas , but not at low whim densities , and anyway these lines should be rare . a key conclusion from this work
is that the bulk of the diffuse collisionally ionised whim gas , i.e. the so - called missing baryons , is only traceable through high ionisation x - ray transitions .
we will explore such x - ray absorption in a forthcoming paper .
the central theme in our exploration of the physical conditions of metal absorbers is the complex link between the enrichment patterns of metals and their observational tracers .
simulations show that even the most commonly observed low-@xmath87 metal absorption line , @xmath2 , traces only a small fraction of the metals and baryons outside of galaxies ( od09 ; * ? ? ?
* ; * ? ? ?
furthermore , even when one uses equation [ equ : omega ] to sum the observed absorbers to find the cosmic density of a particular ion , @xmath243 , this does not necessarily yield the true physical mass density of that ion .
we compare the mass - weighted cosmic ion density summed via the contribution of every sph particle outside of galaxies , @xmath244 ( the grey shading in figure [ fig : zionphase ] ) to @xmath243 ( equation [ equ : omega ] , the volume - weighted observational probe represented by the coloured circles for individual absorbers ) .
ideally the two methods should yield the same answer ; i.e. @xmath243 should recover the true @xmath245 .
we list the comparisons in table [ table : omegacomp ] for a number of different ions and wind models along with the percentage of the @xmath245 recovered ( @xmath246 ) in our simulated observational sample corresponding to s / n=30 cos observations over a pathlength of @xmath247 and any identified line below @xmath248 .
this pathlength appears to adequately sample the densities from which the species we discuss arise .
lccccc ion & redshift & model & @xmath249 & @xmath250 & @xmath251 + + @xmath2 & @xmath252 & vzw & 36.7 & 57.6 & 64 + `` & '' & cw & 20.0 & 30.7 & 65 + `` & '' & sw & 41.6 & 65.8 & 63 + `` & '' & nw & 0.11 & 0.30 & 37 + `` & '' & vzw - turb & 45.9 & 57.6 & 80 + @xmath3 & @xmath252 & vzw & 11.5 & 34.0 & 34 + `` & '' & cw & 5.0 & 12.2 & 41 + `` & '' & sw & 8.5 & 24.2 & 35 + `` & '' & vzw - turb & 24.8 & 34.0 & 73 + @xmath4 & @xmath252 & vzw & 2.3 & 10.3 & 22 + `` & '' & cw & 0.9 & 2.6 & 35 + `` & '' & sw & 1.3 & 5.2 & 25 + `` & '' & vzw - turb & 6.9 & 10.3 & 67 + @xmath5 & @xmath253 & vzw & 4.5 & 10.7 & 42 + `` & '' & cw & 4.0 & 8.8 & 45 + `` & '' & sw & 7.6 & 15.5 & 49 + `` & '' & vzw - turb & 3.4 & 10.7 & 32 + + [ table : omegacomp ] between 22 and 80% of @xmath245 is recovered , depending on the species and the wind model .
lower ionisation species often have lower @xmath246 , which at least double for @xmath28 and @xmath254 when turbulence is added to the vzw model .
this is because these species ( i ) arise from higher densities that are more stochastically sampled in our sight lines , and ( ii ) depend heavily on the derived column densities for the strongest absorbers , from where most of the cosmic density of an ion arises .
@xmath4 is the most stochastic absorber we follow , and arises from the highest densities ( @xmath255 ) .
conversely , @xmath197 and @xmath200 are higher ionisation species that trace lower densities with less stochasticity .
this explains the lower dispersion of recovery fractions between the models .
recovery of @xmath200 ( 32 - 49% between @xmath158 ) suffers because of sensitivity issues ; the majority of @xmath5 is not well - sampled at diffuse overdensities because it is too weak to be detected .
@xmath14 is an even more extreme example with vzw having @xmath256 ; only a portion of the collisionally ionised @xmath14 halo component is recovered .
@xmath2 traces a statistically well - sampled phase space with the most sensitivity to provide the most complete census of @xmath245 , but still manages to recover at most @xmath116 of the total .
another practical issue results from saturated absorbers that have uncertain column densities .
autovp typically estimates lower column densities than the true value , and @xmath243 becomes under - estimated . adding turbulent broadening raises
the column density where lines become saturated and , therefore , @xmath257 rises and also becomes more accurate as we discuss in [ sec : turb ] .
this effect can be quite dramatic for species with high stochasticity such as @xmath3 and @xmath4 at @xmath233 , where applying turbulent broadening more than doubles @xmath246 . in summary
, limited sensitivity and stochasticity can lead to inaccuracies in the determination of the cosmic ion abundance .
sensitivity issues always lead to an underestimate , while stochasticity leads to more uncertainty in an ion s cosmic density .
additionally , saturated absorbers can hide significant amounts of true absorption , leading to under - estimates of cosmic ion densities , especially for low ionisation species .
hence while cosmic ion densities are a useful characterisation of the overall metal evolution in the igm , detailed comparisons between models and data are more robustly accomplished using absorber statistics .
a fundamental difference between the @xmath20 and metal - line forests is the latter s high sensitivity to outflow models . in this section
we explore such sensitivities of the model predictions to variations in some important physical and chemical assumptions .
we will mainly focus on @xmath2 , but will also consider other species .
od09 already explored a range of model variations for @xmath2 , finding that this ion consistently traced the diffuse , @xmath6 k igm .
given that other groups find different results for @xmath2 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , we concentrate our study on varying four aspects of the modelling : the ionisation background , metal - line cooling , turbulent broadening , and metal inhomogeneity .
we present the ewds and the cdds for the variations that we explore in figures [ fig : ew_ovi ] and [ fig : cdd_ovi ] , respectively , and provide cosmic @xmath2 ion densities at @xmath233 in table [ table : omega_ovi ] .
our conclusion that photo - ionisation is responsible for the majority of the absorption by high - ionisation species is sensitive to the shape and intensity of the ionisation background at far - uv and soft x - ray energies .
no direct observational constraints exist between the observed slope of extreme uv ( euv ) spectra of quasars @xcite and the x - ray background @xcite .
the ionisation potential required to photo - ionise @xmath2 at 8.4 rydbergs lies in this unconstrained gap ( j. m. shull , private communication ) .
lccc model & @xmath249 & @xmath250 & @xmath251 + + vzw & 36.7 & 57.6 & 64 + vzw-@xmath258 & 36.2 & 53.4 & 68 + vzw-@xmath259 & 42.5 & 75.3 & 56 + vzw-@xmath260 & 41.5 & 110.5 & 38 + vzw-16 & 37.3 & 56.6 & 66 + vzw - pi & 26.2 & 39.6 & 66 + cw-16 & 23.9 & 35.6 & 67 + cw - pi & 7.7 & 11.5 & 67 + vzw - turb & 45.9 & 57.6 & 80 + vzw - pi - turb & 32.5 & 39.6 & 82 + cw - pi - turb & 11.5 & 11.5 & 100 + vzw-0sig & 34.4 & 57.6 & 60 + vzw - z01 & 51.5 & 71.4 & 72 + + [ table : omega_ovi ] we use the cuba package @xcite to generate three alternative ionisation backgrounds varying the slope of the euv portion of the quasar spectrum . figure [ fig : ionbkgd ] displays these three other backgrounds , which are referred to as hm2005 backgrounds ( after haardt & madau 2005 ) with @xmath261 , 1.57 , and 2.0 , where the spectrum above 1 ryd has an intensity @xmath262 . the hm2005 background is included as an input into the latest package of cloudy and uses the @xmath263 power law based on @xcite , but the range of slopes explored here is reasonable given the observed range and uncertainty in quasar spectral shapes ( f. haardt , private communication ) .
the fiducial hm2001 spectrum uses @xmath264 and is most similar to the hm2005 @xmath265 spectrum .
the hardest background , @xmath261 , has 16 times more intensity at the @xmath2 ionisation potential relative to the softest background , @xmath265 , while leaving the background at the lyman limit and , therefore , the @xmath20 forest nearly unchanged .
figure [ fig : phaseionbkgd ] shows the phase space distribution of @xmath2 absorbers under these two extreme assumptions , and demonstrates that the peak density for @xmath197 shifts from @xmath266 to @xmath267 when moving from the softest to the hardest backgrounds ( cf .
grey @xmath156-axis histograms ) .
however , typical @xmath2 absorbers ( @xmath268 ) do not reflect such a dramatic shift .
recovery fractions of @xmath197 also decline from 56% to 32% ( table [ table : omega_ovi ] ) , mainly because @xmath2 absorbers from haloes have underestimated column densities due to saturation .
the top panel of figure [ fig : cdd_ovi ] displays the @xmath2 cdd for the four backgrounds . as the euv intensity
is turned up , photo - ionised @xmath2 moves to higher overdensities creating more strong absorbers and fewer weak absorbers ; a power law fit to the cdd would give a shallower slope .
the @xmath261 background model appears consistent with the ewd in the top panel of figure [ fig : ew_ovi ] without the need for turbulent broadening , but the linewidths are still underestimated for strong absorbers compared to the observed @xmath86-parameters ( not shown ) . when @xmath261 , strong photo - ionised @xmath2 absorbers populate haloes while the @xmath265 background predicts barely any photo - ionised halo absorbers .
if photo - ionised @xmath2 is clearly identified within haloes , it suggests that a hard meta - galactic ionisation background is penetrating halo gas .
collisionally ionised @xmath2 is a poor tracer of whim gas in all cases , because even the weakest ionising background efficiently photo - ionises oxygen in its collisionally ionised temperature range , peaking at @xmath269 k , to higher ionisation states .
this is why @xmath2 fails to trace the copious amounts of metals pushed by cw winds into the low - density @xmath270 k whim ; this oxygen ionises to @xmath241 and beyond . in summary ,
reasonable variations in the hardness of the euv background are unlikely to alter our fundamental conclusion that most @xmath2 absorbers detectable with cos arise in photo - ionised gas .
even if the ionisation background photo - ionises most @xmath2 at @xmath215 k to higher states outside haloes , there still exists a potential for some @xmath2 to trace whim temperatures .
@xcite , @xcite , and @xcite all predict a significant component of collisionally ionised @xmath2 at overdensities as low as @xmath145 .
one possible explanation is the efficiency of metal - line cooling in the @xmath271 k temperature regime . to explore this ,
we now compare two implementations of metal - line cooling ( i ) the assumption of collisionally ionised equilibrium ( cie ) , and ( ii ) cooling in the presence of the @xcite photo - ionising background ( pi ) .
our main simulations employ the first implementation , namely that @xcite cie metal cooling rates are added to hydrogen and helium rates calculated in the presence of the @xcite background .
however , in the presence of a photo - ionising background , the number of bound electrons is reduced , leading to less efficient cooling .
@xcite in their 5.3 demonstrate that photo - ionised cooling times using @xcite rates in owls ( overwhelmingly large simulations ) , rates that include the effects of the uv background on the metal cooling , are significantly longer for @xmath24 and @xmath98 gas at @xmath272 .
this leads to a higher equilibrium temperature where cooling balances photo - heating .
photo - ionised rates are clearly the more physically correct cooling rates , therefore we compare the results of using our normal metal cooling implementation ( adding cie rates ) to those using photo - ionised metal cooling rates based on the @xcite solar abundance cooling tables .
to do so , we evolve four @xmath80 simulation volumes at the same resolution as our @xmath65 simulation volumes : vzw and cw winds with cie cooling ( vzw-16 , cw-16 ) and pi ( vzw - pi , cw - pi ) cooling .
we present the ewds and cdds for all but the cw-16 model along with the 48 @xmath82 vzw simulation in the second panels of figures [ fig : ew_ovi ] and [ fig : cdd_ovi ] , respectively .
the vzw-16 simulation is statistically identical to the larger vzw simulation for all @xmath2 observations .
this comes as no surprise because od09 demonstrated that @xmath2 is mostly associated with low - mass , @xmath273 , galaxies , which are well sampled in both volumes .
this justifies using the smaller volume to explore these physical variations . comparing the vzw-16 and vzw - pi simulation
, there exists 70% as much @xmath2 absorption in the latter , specifically @xmath197 , @xmath274 versus @xmath275 , with the reduction evenly distributed in column density ( figure [ fig : cdd_ovi ] , middle panel ) .
examination of the @xmath2 phase space for vzw - pi in figure [ fig : ophase_cool ] ( upper left panel ) reveals that most of the @xmath2 absorption still arises from the diffuse @xmath2 phase . however , there are subtle differences compared to the standard vzw model ( bottom right panel ) .
the locus of cool metals at @xmath276 ( i.e. @xmath277 ) is no longer flat at @xmath63 k , below which cie metal - line cooling no longer contributes , but declines from @xmath278 k to below @xmath117 k for increasing overdensity as seen also in owls ( cf .
* their figure 1 ) .
@xmath2 absorbers , therefore , trace somewhat hotter temperatures , but are still within the diffuse phase .
somewhat more @xmath2 exists in the whim phase owing to this less efficient cooling , but still only a handful of identified absorbers actually trace this phase .
examining the two different cooling models in the cw simulations reveals much larger differences ( figure [ fig : ophase_cool ] , right panels ) ; @xmath197 declines by a factor of almost three using pi cooling , from @xmath279 to @xmath280 .
the main difference is that cw winds push metals to lower overdensities where photo - ionisation plays a larger role in retarding cooling .
less oxygen cools below @xmath281 k where photo - ionised @xmath2 ionisation fractions are the largest . as an aside
, we note that galaxy growth remains relatively unaffected by this change .
the amount of stars formed at @xmath1 is reduced by only 8% using pi cooling relative to cie cooling .
this mostly arises because the reduced metal - line cooling at lower densities hinders the recycling of wind material back into galaxies @xcite .
a small reduction in overall star formation was also noted by @xcite when making the same comparison in their grid - based simulation ( cf . the red and black curves in their figure 1 ) . in summary , including photo - ionised cooling rates does not alter our fundamental conclusions about the phases @xmath2 and other absorbers trace in our favoured vzw model @xmath2 still traces mostly diffuse , photo - ionised gas .
we also demonstrate that using a different wind model ( cw ) , similar to that used by @xcite , produces a much larger change in @xmath2 absorber properties when using photo - ionised metal cooling rates , because this model pushes many more metals into very diffuse gas where the photo - ionised cooling rates have a larger impact .
we note that the vzw model does a much better job than the cw model in fitting the observed @xmath1 galactic stellar mass function below @xmath282 @xcite , from which most @xmath2 absorption arises ( od09 ) .
hence , we favour the vzw model , although our tests do illustrate that the interpretation of observed @xmath2 still could depend on the details of the input physics .
in od09 we introduced a heuristic prescription to add sub - grid turbulent broadening to the @xmath86-parameters of @xmath2 , after the conclusion of the simulation , using specexbin .
this added turbulent broadening is a function of the physical density of the gas .
the prescription adds an additional turbulent @xmath86-parameter in quadrature where @xmath283 ^ 2 + 15674 { { \rm log}}[n_{\rm h } ] + 43610}~{\;{\rm km}\,{\rm s}^{-1}},\ ] ] over the range @xmath284 to @xmath285 and @xmath286 over the range @xmath287 to @xmath288 .
furthermore , @xmath159 retains its maximum value of @xmath289 at higher densities .
see od09 4.4 for an in - depth discussion of the physical motivation for this model and its relation to the turbulence observed on kpc scales in @xmath3 paired absorbers along lensed quasar sight lines at @xmath290 @xcite .
od09 argue that the turbulence is related to outflows and decays on timescales of the order of a hubble time .
we already showed results for the vzw - turb model in figures [ fig : ew ] through [ fig : ewcompsys ] .
here we combine it with the vzw and cw models with photo - ionised cooling rate ( vzw - turb - pi , cw - turb - pi ) shown in the third panels of figures [ fig : ew_ovi ] and [ fig : cdd_ovi ] .
photo - ionised metal - line cooling rates reduce vzw @xmath2 line frequencies and @xmath197 by the same factor , one - third , either including turbulence or without it .
the vzw - turb - pi model agrees well with the observed @xmath2 cdd and ewd , and better than the vzw - turb model , which over - produces weak @xmath2 absorbers .
the vzw - turb - pi model is our favoured model for @xmath2 as it has the most physically motivated combination of outflows , ionisation background , physical metal - line cooling , and turbulent broadening to explain both @xmath2 and @xmath20 forest observations ( d10 ) , as well as other observations of low-@xmath87 galaxies @xcite .
we show the @xmath3 , @xmath4 , and @xmath5 cdds for this model along with the vzw - pi model in figure [ fig : cdd_c4_si4_ne8 ] . as before
, our favoured model continues to overproduce the incidence of the lower - ionisation species .
we do not believe this discrepancy can be mitigated by cosmic variance , because even our @xmath80 box is sufficient to sample the gas and galaxies responsible for the absorption of these ions ( cf . vzw and vzw-16 show nearly identical results ) .
however , the idea of adding turbulent broadening for lower ionisation species that trace higher densities may not be as valid as in the case of @xmath2 .
for instance , using lensed quasars probing the @xmath291 igm , @xcite observed sub - kpc sub - structure in lower ionisation species including @xmath3 and @xmath4 that did not exist in @xmath2 .
at low redshifts , all of the @xmath4 and most of the @xmath3 in our models have densities where our turbulence model predicts @xmath292 , but the observations of @xmath3 suggest much narrower linewidths ( e.g. * ? ? ?
hence @xmath2 may be tracing more diffusely distributed gas that is more affected by large - scale turbulence injected by outflows , while @xmath3 and @xmath4 could be tracing individual high - density cloudlets .
regardless , models both with and without turbulence overpredict the @xmath3 and @xmath4 absorption . figure [ fig : zionphase_bturb_pi ] shows the phase - space diagram for metals and absorbers in the vzw - turb - pi model .
@xmath4 and stronger @xmath3 absorbers trace much higher densities than @xmath2 .
this results in many more strong @xmath3 and @xmath4 components than in the vzw - pi model , because autovp finds stronger @xmath85s for otherwise non - broadened saturated profiles and weaker absorbers are subsumed into the stronger ones . the histograms in figure [ fig : zionphase_bturb_pi ] show that these absorbers mostly arise in cool halo gas .
a difficulty with our simulations is that we may not resolve cooling instabilities in the halo gas ( e.g. * ? ? ?
* ) that could result in much of the carbon being in dense cold cloudlets , which would favour lower ionisation states than @xmath3 .
this is in principle easily possible as only 3 - 5% of the carbon outside of galaxies resides in @xmath3 between @xmath157 ( vs. 13 - 15% at @xmath293 ) , and hence a small change in the ionisation rate at @xmath294 ryd could dramatically change the @xmath3 statistics
. it will be interesting to compare our predictions of @xmath33 and @xmath295 to observations when a sufficient sample becomes available , to see if we significantly underpredict the amount of carbon in these states corresponding to our overprediction of @xmath3 .
another possibility is that the @xcite type ii carbon yields we use are too high .
for example , the @xcite carbon yields are @xmath296 dex lower , which could explain some of the differences with respect to the observations .
the agb stellar yields we use can not explain the over - estimate , because carbon production from agb stars is sub - dominant to type ii sne in our simulations @xcite .
yet another potential explanation is that a substantial fraction of carbon and silicon are depleted onto dust , while the more volatile oxygen is not .
@xcite find that the vzw model can explain the observational evidence for intergalactic dust @xcite if the igm dust - to - gas ratio is about half that in the milky way ism , which about 25% of intergalactic metals
( by mass ) in dust .
@xmath5 traces similar overdensities as @xmath2 and hence should be affected by the same turbulence , and hence the vzw - turb - pi model should be the most appropriate . but
as figure [ fig : cdd_c4_si4_ne8 ] shows , the predicted column density distributions do not vary significantly with pi cooling or turbulence . in detail , between @xmath158 , @xmath200
slightly increases with pi cooling , but declines by a factor of 20% with the addition of turbulence to @xmath297 , because already weak lines are broadened enough so that autovp fails to identify them .
most of the @xmath5 absorption still traces photo - ionised gas , but the reduced cooling raises the bulk of @xmath5 absorbers from @xmath298 k to @xmath299 k and leads to 22% compared to 7% of the @xmath5 absorption arising from @xmath300 k gas ( figure [ fig : zionphase_bturb_pi ] ) .
this percentage could grow even larger if our simulations adequately resolved interfaces between cool and hot gas within haloes , which they do not as we discussed in [ sec : absphase ] .
we show the cw - turb - pi model to demonstrate that turbulent broadening affects other wind models in much the same way as the vzw model , by making more wide @xmath2 absorbers and slightly fewer weak absorbers ( cf .
cw - pi model ) .
this more appropriate cooling rates applied to the cw model , combined with turbulent broadening to match observed line widths indicates this model far under - produces @xmath2 , especially weaker systems , when compared to observed ewds and cdds .
the degree of metal inhomogeneity in the igm remains highly uncertain , but dynamically very important .
the distribution of metals affects gas cooling , which is critical for galaxy formation because the temperature of the gas determines the efficiency of gas accretion onto galaxies @xcite .
more metal mixing leads to more star formation as more efficient metal - line cooling affects more baryons @xcite . in our simulations , we do not mix metals , and each wind particle retains its own metals as it propagates out of galaxies . the homogeneity of metals impacts the statistics of absorbers , which we explore here .
we consider two ad hoc models for distributing metals throughout the igm , by replacing particle metallicities in our simulation with prescriptive metallicities in our vzw model . in the first ( vzw - z01 ) ,
we assume a uniform metallicity of @xmath301 in the igm ( vzw - z01 ) , which is the global @xmath1 igm metallicity in our vzw model .
note that this is similar to @xmath24 painted onto earlier simulations to reproduce older observed @xmath2 absorber frequencies @xcite .
the second model ( vzw-0sig ) takes the average [ o / h ] of the vzw model in each @xmath211 dex pixel of phase space ( upper left panel of figure [ fig : ophase_dist ] at @xmath302 ) and paints this metallicity onto all sph particles within that phase space cell , with zero dispersion .
both prescriptions preserve the total metal mass , but neither model is dynamically self - consistent , because metal cooling would alter the physical state of the gas and feedback would have to enrich the baryons in the first place .
however , our purpose here is to explore extreme cases of metal homogeneity to see if absorption line statistics can distinguish the level of metal inhomogeneity .
figures [ fig : ew_ovi ] and [ fig : cdd_ovi ] , bottom panels , show the ewds and the cdds of these models , compared to our original vzw model .
the vzw - z01 model , as found in od09 , produces too many weak absorbers , too few strong absorbers ( in terms of column density ) , and a large alignment fraction with @xmath25 ( figure [ fig : align_ovi ] ) .
collisionally ionised @xmath2 ( @xmath215 k ) accounts for a quarter of @xmath197 and two - thirds of lines above @xmath303 .
simply painting metals onto a simulation uniformly overestimates @xmath2 and other species in cie , because the most commonly observed uv resonance lines ( @xmath3 , @xmath2 , and soon @xmath5 ) are also some of the most important coolants .
painting the average metallicity by phase space pixels explores how the dispersion in metallicity among sph particles affects the line statistics .
this model shows negligible differences versus the original vzw case , showing that the fact that metals are not mixed in our simulations has no impact on @xmath2 absorber statistics .
we reiterate that this test does not dynamically account for differences in the cooling rates that would result from such metal mixing , and hence it is not self - consistent ; this test isolates the impact of spatial inhomogeneities in the metals .
figure [ fig : ophase_dist ] provides a broader view of metal inhomogeneity in our original vzw simulation .
the upper left shows the oxygen metallicity in phase space , this is similar to figure [ fig : zphasespaces ] except here we only consider oxygen and the general trends are as before .
the upper right panel shows the fraction of baryons enriched with oxygen , @xmath304 , using a threshold of [ o / h]=@xmath305 .
@xmath304 is not very sensitive to this threshold and is used to discard sph particles enriched to very low levels by delayed feedback .
practically all the icm gas and most gas , both hot and cold , inside of haloes is enriched .
interestingly , the @xmath304 distribution for the nw simulation ( not shown ) looks nearly identical to the vzw model above @xmath119 k , indicating that enrichment mechanisms other than outflows ( e.g. dynamical stripping , delayed feedback ) are getting hot baryons , i.e. hot halo and whim gas , above this threshold metallicity . in the igm outside of haloes ,
most particles remain pristine , as the typical fractions of enriched particles are @xmath306 . in photo - ionised gas within voids
, the fraction of enriched particles can drop below @xmath307 .
the lower left panel shows the average metallicity of only the enriched particles , [ o / h]@xmath308 .
this demonstrates that much of the enriched portion of the diffuse igm has [ o / h]=@xmath309 .
since the outflows carry out metals that are representative of the ejecting galaxy , this indicates that the dominant galaxies enriching the igm tend to have sub - solar metallicities ; these galaxies are generally well below @xmath221 ( as found in * ? ? ?
however , there is a selection effect that highly enriched igm particles undergo stronger metal - line cooling , dropping them out of the warmer phases onto the photo - ionised locus .
hence enriched particles in the underdense regions actually have higher metallicities than the whim gas in moderately overdense regions , since these are the only particles that can cool at such low densities .
nevertheless , the @xmath304 shows that such particles are quite rare .
the lower right panel shows the log - normal dispersion , @xmath62([o / h]@xmath308 ) of the metals around the median log[[o / h]@xmath308 ] for the enriched particles only .
both @xcite and @xcite found that @xmath3 and @xmath2 enrichment at @xmath311 could be described by a log - normal distribution , but applied to all baryons .
in general , regions with dispersions below 0.5@xmath62 are enriched via outflows consisting of ism gas having similar metallicities .
the higher dispersions correspond to regions enriched via other mechanisms such as dynamical stripping and delayed feedback from intracluster stars .
the association of @xmath25 with @xmath2 offers a potentially powerful way to explore metal inhomogeneity .
we show alignment statistics of @xmath2 components with nearby @xmath25 components in figure [ fig : align_ovi ] .
we plot the fraction of @xmath2 absorbers with an associated @xmath25 absorber at or below the velocity difference indicated on the @xmath156-axis , with the criteria that both absorbers have @xmath312 m and @xmath233 . overlaid are stis data from @xcite using the same selection criterion .
the data are unable to discriminate between vzw and vzw-0sig , but disfavour the vzw - z01 model , which tends to produce more alignment .
od09 showed that @xmath2 absorbers trace similar diffuse densities and temperatures as @xmath25 , but that the @xmath2-traced enriched gas is inhomogeneous and clumpier than the diffuse baryons traced by the @xmath20 forest .
the difference in metal inhomogeneity between the vzw and vzw-0sig models does not show up in the ewd and cdd distributions in figures [ fig : ew_ovi ] and [ fig : cdd_ovi ] , but the expected observational sample of the two most common species probed by cos is expected to provide the ability to statistically distinguish these two models .
the cw model shows lower alignment owing to faster winds pushing metals further from the @xmath20 in both physical and velocity space .
both turbulence and photo - ionisation - dependent cooling ( not shown ) do not alter the statistics from the vzw model .
alignment statistics versus @xmath25 fundamentally probe the level of photo - ionisation in metal absorption , because @xmath25 arises almost exclusively in photo - ionised gas . models that produce more collisionally ionised @xmath2 will therefore have significantly less alignment with @xmath25 .
current observations of @xmath2-@xmath25 alignment from @xcite shown in figure [ fig : align_ovi ] suggest that the vzw model produces roughly the correct level of photo - ionised gas in relation to @xmath25 , and therefore strongly favours our conjecture that much of the @xmath2 seen with cos is photo - ionised .
our favoured model includes momentum - conserved winds and uses photo - ionisation - dependent cooling rates .
the vzw - turb - pi model matches the @xmath2 statistics including the @xmath233 ewd , cdd , @xmath197 , and @xmath86-parameters when turbulent broadening is added as argued and theoretically motivated in od09 .
we further advocate vzw winds for their ability to provide an adequate fit to the @xmath1 galactic stellar mass function below @xmath282 @xcite , from which most @xmath2 absorption arises ( od09 ) , as well as other properties of the low - redshift galaxy population @xcite ; these galaxy properties are not expected to be greatly impacted by adding igm turbulence or pi cooling .
the hm2001 ionisation background multiplied by 1.5 appears capable of both reproducing the observed @xmath20 forest ( d10 ) and @xmath2 that is principally photo - ionised .
the vzw wind model also has successes fitting higher redshift data , including the @xmath19 mass - metallicity relationship @xcite , the @xmath313 galaxy luminosity function @xcite , and observations of igm metal enrichment at redshifts @xmath314 @xcite .
the most obvious failing of the vzw model is its inability to quench star formation in @xmath315 galaxies and create the observed red sequence , although this likely has to do with our simulations lacking a mass - dependent quenching mechanism such as agn feedback @xcite .
however , massive galaxies are not responsible for the vast majority of @xmath2 enrichment ( od09 ) , because they enrich a comparatively small volume of the universe , and so this quenching of massive galaxies is not expected to significantly impact our igm enrichment levels .
nevertheless , this model also has significant problems that are shared by the other wind models we consider here . @xmath3 and
@xmath4 observations pose a problem in that our favoured model produces too many absorbers compared to @xcite
. it might be possible to alter the ionisation background at 3 - 4 rydbergs to mitigate this discrepancy while still satisfying the other observational constraints .
updated ionisation backgrounds generated with the new version of cuba @xcite are worth exploring in the future .
alternatively , our assumed yields could be too high , or carbon and silicon could be preferentially depleted onto dust @xcite . however
, the discrepancy may result from limitations present in our simulations , namely that we fail to resolve cooling instabilities in halo gas where such lower - ionisation absorbers are expected to arise .
quantifying @xmath33 and @xmath295 absorption may help distinguish between these possibilities .
our favoured model also under - predicts the frequency of @xmath5 lines with column densities similar to those observed so far @xcite , which again may be a result of insufficient resolution within the halo gas , in the sense that we fail to resolve metal - bearing interfaces between cold clumps embedded in hot , tenuous halo gas .
relatedly , od09 ( 5.4 ) found a lower frequency of complex , collisionally ionised @xmath2 absorbers , which may also arise in shock interfaces within hot halo gas , as implied by some observations .
in contrast , absorbers arising in more diffuse gas should be more robustly modelled in our simulations .
hence , we are confident that , when probed to sufficiently low column densities , most high ionisation lines of @xmath2 and @xmath5 will mostly trace diffuse photo - ionised gas .
since there have been a number of recent publications exploring @xmath2 and other metal - line statistics in cosmological simulations to @xmath1 , we now compare and contrast our results with these works .
our discussion of alternative feedback methods begins with @xcite , who generate statistics for @xmath2 and @xmath3 from @xmath316 in a 50 @xmath82 @xmath317-cell uniform mesh simulation .
their winds are implemented by distributing gas and energy onto the 27 nearest cells @xcite , which instantaneously distributes metals over a 73 comoving @xmath222 cube .
clearly this model does not self - consistently track the dynamics of enriching this cube , but it nevertheless provides a heuristic way to distribute metals into the igm .
their favoured wind efficiency is @xmath318 , where @xmath319 represents the fractional rest mass energy of all stars formed put into thermal feedback .
their total wind energy is then @xmath320 erg/@xmath321 , which @xcite argue is roughly comparable to the total sne energy in a chabrier initial mass function ( imf ) . @xcite similarly use fixed - mesh eulerian simulations and parametrise star formation and feedback the same way . they can reproduce much of the observed @xmath322 @xmath2 cdd with a wind energy efficiency of @xmath323 .
we calculate their total @xmath324 erg/@xmath321 , which is nearly twice the sne energy available in a chabrier imf if all stars @xmath325 explode as @xmath326 erg sne , but about equivalent if all stars @xmath327 explode as sne .
@xcite advocate the need of distributing feedback over 27 cells ( as in @xcite ) to fit the observed star formation rate density history , whereas limiting the feedback to one cell makes it much less efficient at curtailing star formation .
this means that in their case , metals are instantaneously distributed over a 99 to 294 @xmath222 cube depending on their simulations . our simulations , in contrast ,
attempt to directly mimic observations of galaxies that indicate wind speeds of hundreds of km / s emanating from galaxies ( e.g. * ? ? ?
* ) , and scaling with the circular velocity ( e.g. * ? ? ?
we are inevitably limited by resolution and numerics in that we can not fully resolve how winds escape from galaxies , so we are forced to employ parametrisations for these winds . the adopted parametrisations ( including hydrodynamic decoupling )
are intended to represent observed winds at the scales at which they are observed .
it is interesting that our winds typically travel @xmath142 physical kpc @xcite , which is similar to the scales over which @xcite and @xcite assume outflows distribute gas .
the difference is that some of our winds can travel much farther or not as far , depending on the self - consistently tracked interaction between the gas and the ambient medium ; we do not directly parametrise how far winds and metals travel .
moreover , in these fixed mesh simulations the distance metals travel is explicitly governed by the spatial resolution , while in our case the wind speeds are based on the galaxy mass , which is a well - converged quantity .
for instance , the lower - resolution mesh simulations of @xcite produce a significant @xmath2 signal without any mechanism for distributing gas to nearby cells , in contrast to our no - wind case that produces essentially no igm metal absorbers .
the direct sensitivity of fixed mesh simulations to ad hoc numerical parameters makes it difficult to interpret their conclusions about the physics driving igm enrichment . adaptive mesh refinement ( amr ) simulations could potentially provide a less resolution - sensitive implementation of winds . finally , we point out that our feedback model has quite modest energy requirements relative to the above mesh simulations .
both @xcite and @xcite strain the limits of available supernova energy .
our cw model and the feedback prescription in @xcite ( @xmath328 , @xmath56 ) require substantial fractions of sn energy to be converted directly to kinetic winds @xmath329 assuming the ( conservative ) @xmath330 lower mass limit for sne using the 6 @xmath321 lower mass limit for sne , and the cw model gives an efficiency of @xmath331 using this limit ; however we scale all models to the @xmath330 mass limit in our comparison . ] .
comparatively , the vzw model has @xmath332 at @xmath147 and @xmath333 at @xmath1 , which is meant as an illustrative comparison since this wind prescription simulates energy generated during the lifetimes of o and b stars .
r48n384vzw value published in table 1 of @xcite is a miscalculation , overestimating the efficiency by 50% . ] because vzw feedback energy scales with @xmath62 , the lower mass galaxies that are primarily responsible for the enrichment of the diffuse igm have even less of an energy requirement .
for instance , a @xmath334 galaxy has @xmath335 at @xmath147 and 0.21 at @xmath1 .
note that @xcite in their 5.4 misinterpret the vzw model as requiring more energy than in the sn energy budget ; this is true only for central galaxies in haloes more massive than @xmath336 at @xmath147 , or for between 2 - 4% of all wind particles launched at any redshift .
these galaxies are not responsible for enriching the igm , and in any case should not have significant ongoing star formation today @xcite .
absent in our simulations is any attempt to simulate the mixing of metals . for the purposes of cooling and generating mock spectra
, we use particle metallicities , which is in contrast to @xcite , who smooth metallicities over sph particle smoothing lengths when calculating metal - line cooling .
their metal smoothing was introduced in @xcite , who calculate that smoothed metallicities are @xmath337th the individual particle metallicities when all surrounding particles in their kernel are pristine . given that our metal - enriched igm wind particles
are often surrounded by pristine gas , the cooling times of our metals could be as much as @xmath338 times shorter using our individual particle metallicities if metal - line cooling dominates total cooling .
we suggest this could be the source of the greatest differences with the results of @xcite , who find the majority of @xmath2 above @xmath119 k. our cw - pi model is most similar to their `` reference '' model with notable other differences including a slightly higher wind speed ( @xmath339 vs. @xmath340 ) and that we decouple our winds ; both of these should make it easier for winds to escape galaxies in our simulations .
nonetheless , both simulations have similar @xmath197 and similar overdensities bearing @xmath2 gas , but most of our @xmath2 gas resides around @xmath281 k , whereas theirs is at @xmath341 k with significantly more @xmath2 at temperatures @xmath342 k. fixed mesh simulations with distributed feedback also can lead to much smoother metallicities , resulting in longer cooling times and hotter metal - bearing gas .
we argue that the lack of metal smoothing is a fundamental difference causing a significant photo - ionised component of @xmath2 in our simulations versus others @xcite .
both @xcite and @xcite make a strong case that using cie metal - line cooling rates over - estimates cooling resulting in too much photo - ionised @xmath2 in od09 .
this is no doubt true , especially given the fact that in our cw simulations , photo - ionised metal - line cooling rates reduce @xmath197 by a factor of three .
but as we demonstrated in [ sec : zcool ] , for our favoured vzw case the difference is much less .
still , metal mixing can be an important determinant for the igm phase of @xmath2 absorbers .
treatments of metal diffusion have been implemented into other simulations @xcite , but it remains unclear to what extent metals mix in the igm .
interpreting observations of metal absorbers at higher redshift suggest that metals could remain in concentrated , small cloudlets with metallicities similar to galactic ism gas @xcite .
this scenario would favour our lack of mixing and the fact that winds carry ism gas directly into haloes and the igm . in summary ,
the necessity and specific implementation of metal mixing remains uncertain when considering the low-@xmath87 igm .
it appears that metal mixing is as determinative as implementations of metal - line cooling or feedback for interpreting igm metal - line observations .
we examine the absorption and physical conditions of metals residing in the @xmath16290% of baryons outside of galaxies between @xmath1 and @xmath343 .
we employ @xmath65 , @xmath344-particle cosmological simulations , the same simulations used by @xcite to explore the @xmath1 galactic stellar mass function and by @xcite to explore the @xmath20 forest .
our work builds on the previous results of @xcite , finding fundamentally the same conclusions for @xmath2 , and also makes predictions for @xmath4 , @xmath3 , @xmath5 , @xmath13 , and @xmath14 specifically tailored to the expected instrumental characteristics of the _ cosmic origins spectrograph _ ( cos ) .
we subdivide the gas phases between the true igm ( @xmath345 at @xmath1 ) and gas within dark matter haloes ( @xmath346 at @xmath1 ) , and between hot ( @xmath347 k , whim and hot halo ) and cool ( @xmath6 k , diffuse and condensed ) phases .
metal - line absorption is fundamentally different than its primordial big brother , the @xmath20 forest , owing to its much greater sensitivity to the processes of galaxy formation and evolution .
while galactic outflows suppress star formation and therefore metal production , they concurrently distribute metals into haloes and the igm .
our simulation without winds ( nw ) produces 2.1 times more metals by @xmath1 than the momentum - conserved wind ( vzw ) model , but the vzw simulation has 3.5 times more gaseous metals outside of galaxies , and _ 380 times more metals in the diffuse igm phase _ from which most igm metal absorbers arise .
the resulting enrichment patterns are quite sensitive to the details of the enriching outflows .
the combination of wind velocities and environment determines to how low a density the winds reach , e.g. , whether or not metals leave their dark matter haloes .
low velocities from dwarf galaxies easily reach the igm , while @xmath138 winds from @xmath348 galaxies often fail to escape their haloes because of the greater ambient density of the surrounding gas @xcite .
the wind velocity also determines the temperature to which the gas shock heats as the wind decelerates .
fast winds that shock gas to @xmath349 k are more likely to join the whim or hot halo gas .
the density and temperature where the metals are deposited in turn impacts their cooling rate , which can return enriched gas to photo - ionisation temperatures .
tracking all these processes dynamically within growing large - scale structure presents a complex and challenging problem that impacts our understanding of a wide range of phenomena related to the cosmic evolution of baryons .
our main results are as follows : + @xmath21 metals progressively are found in more overdense regions from @xmath38 . by @xmath1 in our favoured momentum - conserved wind ( vzw ) model ,
only 4% of metals are outside of haloes ( i.e. at densities below the threshold defined by equation [ eqn : deltath ] ) , compared with @xmath350 at @xmath19 . the constant wind ( cw ) model has many more metals outside of haloes , but they mostly reside in the difficult - to - detect whim phase .
the majority of metals today reside within galaxies , in contrast to the bulk of baryons that reside outside of haloes ( 3 ) .
@xmath21 for all the wind models , the mean metallicity of most of the baryonic phases is very slowly rising since @xmath351 .
the exception is the condensed phase , which shows a more rapid rise as winds progressively enrich halo gas closer to galaxies at later epochs . by @xmath1 , the igm metallicities in both the cool ( diffuse ) and the hot ( whim ) phases are @xmath8 in the vzw model , and @xmath17 in the 90% of baryons outside galaxies ( 3 ) .
@xmath21 the igm is enriched in an `` outside - in '' fashion where lower overdensities are enriched at higher redshift .
higher ionisation metal species generally trace older metals at lower overdensities owing to these species being primarily photo - ionised .
the @xmath352 relation for vzw outflows results in a continuously rising metallicity - overdensity gradient and a gradual age - overdensity anti - correlation , which contrasts with the constant wind simulations ( 3.3 ) .
@xmath21 constant ( cw ) , momentum - conserved ( vzw ) , and slow ( sw ) wind models all predict @xmath233 @xmath2 observational statistics ( equivalent width distributions , column density distribution , cosmic ion densities ) that agree within a factor of two with observations by stis and fuse .
by contrast , the no - wind model produces an @xmath2 incidence 100 times lower than the other feedback models and observations , which strongly emphasises the need for galactic superwinds to enrich the igm to the observed levels .
all wind models over - produce the @xmath3 and @xmath4 incidence relative to available observations ( 4 ) .
@xmath21 @xmath2 , @xmath3 , and @xmath4 absorbers predominantly ( by number ) arise in photo - ionised , cool gas .
photo - ionisation results in higher ionisation lines tracing lower gas densities : @xmath4 and @xmath3 arise primarily from halo gas at @xmath233 , while @xmath2 arises mainly in the diffuse igm ( 4,5.1 ) .
we predict a population of @xmath5 absorbers with @xmath353 tracing photo - ionised diffuse @xmath6 k gas at overdensities of @xmath145 .
stronger @xmath5 absorbers ( like stronger @xmath2 systems ) are more likely to be associated with @xmath354 k gas and may share more in common with the absorbers already detected ( 4,5.1 ) .
these results do not change when using metal - line cooling rates including the presence of a photo - ionising background ( 6.2,6.3 ) .
@xmath21 @xmath13 and @xmath14 absorbers are very rare , with frequencies of @xmath355 for cos s / n=30 spectra in all of our models .
these absorbers likely trace halo gas at @xmath356 k ( 4,5.1 ) .
these species and @xmath5 all have a significant photo - ionised component , which is consistent with the shape of the @xcite ionisation background ; however @xmath13 and @xmath14 are too weak in this state to be detected using cos ( 5.1.1 ) .
no uv tracer we examine consistently and predominantly traces whim gas , i.e. shock - heated gas outside of galaxy haloes .
@xmath21 using metal - line cooling rates accounting for photo - ionisation by the ionisation background @xcite reduces the incidence of @xmath2 absorbers at all column densities by one - third in the vzw model , which still provides an acceptable match with the observations .
the difference is more dramatic for cw winds since it pushes metals to lower overdensities ; @xmath2 statistics are reduced by two - thirds and arise from gas closer to @xmath357 k. @xcite and @xcite correctly argue that assuming collisional ionisation equilibrium for metal cooling rates ( with no photo - ionisation ) leads to over - efficient metal - line cooling ; however we find that the moderate vzw winds leave metals at higher overdensities where metal - line cooling is still efficient even in the presence of photo - ionisation and results in most of the @xmath2 absorbers tracing @xmath358 k gas ( 6.2 ) .
@xmath21 we reiterate the result of od09 that heuristic turbulent broadening is required to reproduce the observed @xmath2 linewidths , but we suggest that this prescription may not be applicable to @xmath3 and @xmath4 primarily arising from haloes .
@xmath2 arises from longer columns with potentially many cloudlets with random / turbulent velocities at sub - particle scales , while lower ionisation species are more likely to trace individual higher density cloudlets ( 6.3 ) .
@xmath21 absorber alignment statistics of @xmath25 with @xmath2 provide an independent constraint on the level of metal inhomogeneity in the igm , which we quantify in 6.4 .
we disfavour a uniform metallicity of @xmath359 from comparisons with existing data , and future cos observations can potentially quantify the fraction of enriched gas and the dispersion of metals in the igm ( 6.5 ) .
@xmath21 the metal - line statistics to be probed by cos rarely vary by a factor of more than two between the different feedback models , despite enrichment patterns varying by much greater factors ( e.g. , 6.4 times more metals in the @xmath1 whim of cw vs. vzw ) . the relatively restricted density - temperature phase space traced by @xmath2 , @xmath3 , @xmath4 , and @xmath5
are enriched similarly in all wind models relative to the total distribution of metals outside galaxies ( 3.2,5.1.2 ) .
@xmath21 the observed cosmic ion density @xmath155 summed from absorbers typically underestimates the total @xmath155 of an ion .
the stochastic effects of high overdensities being probed rarely in a volume - averaged measurement plus a large fraction of an ion s cosmic density being contained in saturated lines make accurate determinations of @xmath28 and @xmath254 more difficult .
we estimate recovery fractions of 56 - 69% for @xmath197 and 30 - 39% for @xmath200 in s / n=30 cos observations .
these fractions may be sensitive to the quality of the spectra and the fitting method used to estimate the column densities ( 5.2 ) .
@xmath21 exploring alternative ionisation backgrounds given the observed range of quasar emission slopes beyond the lyman limit , we find that the slope of the @xmath2 column density distribution becomes shallower using a harder background .
the ionisation background strength varies by a factor of 16 at the @xmath2 photo - ionisation edge ( 8.4 rydberg ) , but even for the weakest field @xmath2 remains primarily photo - ionised .
slopes of @xmath263 and harder predict more photo - ionised @xmath2 arising from halo gas ( 6.1 ) . + while metal lines observed by cos are more sensitive to the processes regulating galaxy evolution , they are also more uncertain in their modelling and interpretation compared to the @xmath20 forest .
our parameter exploration finds good fits to existing @xmath2 statistics , but appears to over - produce @xmath3 and possibly under - produce @xmath5 arising in hot haloes . like od09 ,
we favour a momentum - conserved wind model with turbulent broadening , although we note that metal - line cooling in the presence of the photo - ionising background reduces @xmath2 by a third and makes the average absorber hotter ( 30,000 vs. 15,000 k ) ; however our @xmath2 results are still consistent with the current data .
we further advocate this model for its ability to reproduce the galactic stellar mass function below @xmath360 @xcite , which are the same galaxies that are primarily responsible for igm enrichment ( od09 ) .
we double down on our prediction of a primarily photo - ionised origin for most uv absorbers ( od09 ) , but realise the risky nature of our gamble .
our model variations show just how sensitive metal - line statistics are to the shape of the ionisation background at poorly constrained energies , the stiff behaviour of metal - line cooling , and a turbulence model that is surely over - simplified .
how metals mix in haloes and the igm is perhaps the most under - appreciated uncertainty for simulating such observations .
our scheme of launching ism gas with no mixing may be the largest difference with other works with smoother metallicities leading to hotter temperatures @xcite ; however , there is observational evidence and theoretical motivation that small , highly - enriched cloudlets remain un - mixed in the igm @xcite .
the other numerical uncertainty is the accuracy of the basic gadget-2 method of implementing galactic winds by one - at - a - time particle ejection , also used in many other sph codes .
we have considered several variations of the scaling of outflow rate and ejection speed with galaxy mass , but we have not altered this basic scheme .
the work presented here represents a broad brush approach to simulating metal line absorbers along random sight lines sampling the volume - weighted universe . with cos ,
new statistics can be generated including correlation functions and line of sight variance statistics , probing the stochasticity of various ion species and perhaps the phase of gas that they trace .
relating the distribution of metal absorbers to the surrounding large - scale structure , particularly the low - mass galaxies responsible for enriching the igm , will provide independent constraints on the nature of outflows .
studying the galaxy - absorber connection uniquely links the physics regulating the growth of galaxies to the dominant reservoir of primordial and enriched baryons in the igm , which we will explore in the next paper in this series .
we expect confronting the galaxy - absorber connection in particular with these and future simulations will provide unique tests that can both better distinguish wind models from each other , and push the limits of our scheme .
our claim that photo - ionisation can explain the majority of the high ion species explored here will face a significant test as we focus on the circumgalactic environment within several hundred kpc of galaxies . given that interface physics and chemistry is not resolved in our simulation while it appears fundamental for absorption in galactic high - velocity clouds , suggests we are underestimating this phase of metal - absorbing gas .
conversely , this phase may represent a small minority of the universal quantity of diffuse metal - enriched absorbing gas , which is the view we advocate here .
we thank kristian finlator , joop schaye , mike shull , thorsten tepper - garcia , chris thom , jason tumlinson , and rob wiersma for useful discussions and contributions .
we especially acknowledge kathy cooksey for detailed discussions and her effort in re - formulating her data specifically for our paper , francesco haardt for providing and explaining his cuba code , charles danforth for providing updated forms of his data , and mark fardal for his diligence in debugging and testing code .
we also are grateful for the hospitality at the university of cape town and the south african astronomical observatory where much of this work was completed .
partial support for this work came from nasa atp grant nnx10aj95 g , hst grants hst - go-11598 and hst - go-12248 , nasa adp grant nnx08aj44 g , and nsf grant ast-133514 .
the simulations used here were run on university of arizona s sgi cluster , ice .
aguirre , a. , hernquist , l. , schaye , j. , weinberg , d. h. , katz , n. , & gardner , j. 2001a , apj , 560 , 599 asplund m. , grevesse n. , & sauval a. j. , 2005 , in barnes t. g. iii , bash f. n. , eds , asp conf .
vol . 336 .
cosmic abundances as records of stellar evolution and nucleosynthesis .
, san francisco , p. 25
bouch , n. , lehnert , m. d. , aguirre , a. , proux , c. , & bergeron , j. 2007 , mnras , 378 , 525 burles , s. & tytler , d. 1996 , apj , 460 , 584 cen , r. & ostriker , j. p. , 1999 ,
apj , 514 , 1 cen , r. , tripp , t. m. , ostriker , j. p. , & jenkins , e. b. 2001 , apjl , 559 , l5 cen , r. , nagamine , k. , & ostriker , j. p. 2005 , apj , 635 , 86 cen , r. & ostriker , j. p. 2006 ,
apj , 650 , 560 cen , r. & fang , t. 2006 , apj , 650 , 573 cen , r. , & chisari , n. e. 2011 , apj , 731 , 11 chen , x. , weinberg , d. h. , katz , n. , dav , r. 2003 , apj , 594 , 42 chieffi , a. & limongi , m. 2004 , apj , 608 , 405 choi , j .- h . , & nagamine , k. 2011 , mnras , 410 , 2579 cooksey , k. l. , prochaska , j. x. , chen , h .- w . ,
mulchaey , j. s. , & weiner , b. j. 2008 , apj , 676 , 262 cooksey , k. l. , thom , c. , prochaska , j. x. , & chen , h .- w .
2010 , apj , 708 , 868 cooksey , k. l , prochaska , j. x. , thom , c. , & chen , h .- w .
2011 , apj , 729 , 87 cowie l. l. , songaila a. , kim t .- s .
, & hu e. m. 1995 , aj , 109 , 1522 danforth , c. w. , shull , m. j. 2005 , apj , 624 , 555 danforth , c. w. , shull , m. j. , rosenberg , j. l. , & stocke , j. t 2006 , apj 640 , 716 danforth , c. w. & shull , m. j. 2008 , apj , 679 , 194 dav , r. , hernquist , l. , weinberg , d. h. , katz , n. 1997 , apj , 477 , 21 dav , r. , hellsten , u. , hernquist , l. , katz , n. , & weinberg , d. h. 1998 , apj , 509 , 661 dav , r. , hernquist , l. , katz , n. , & weinberg , d. h. 1999 , apj , 511 , 521 dav , r. et al .
2001 , apj , 552 , 473 dav , r. , finlator , k. , & oppenheimer , b. d. 2006 , mnras , 370 , 273 dav , r. & oppenheimer , b. d. 2006 , mnras , 374 , 427 dav , r. , oppenheimer , b. d. , & sivanandam , s. 2008 , mnras , 391 , 110 dav , r. , oppenheimer , b. d. , katz , n. , kollmeier , j. a. , & weinberg , d. h. 2010 , mnras , 408 , 2051 dav , r. , oppenheimer , b. d. , & finlator , k. 2011 , mnras , accepted dav , r. , finlator , k. , & oppenheimer , b. d. 2011 , mnras , accepted dekel , a. & birnboim , y. 2006 , mnras , 368 , 2 dodorico , v. , calura , f. , cristiani , s. , & viel , m. 2010 , mnras , 401 , 2715 fang , t. & bryan , g. l. 2001 , apj , 561 , l31 fang , t. , buote , d. a. , humphrey , p. j. , canizares , c. r. , zappacosta , l. , maiolino , r. , tagliaferri , g. , & gastaldello , f. 2010 , apj , 714 , 1715 ferland , g. j. , korista , k. t. , verner , d. a. , ferguson , j. w. , kingdon , j. b. , & verner , e. m. 1998 , pasp , 110 , 761 finlator , k. & dav , r. 2008 , mnras , 385 , 2181 finlator , k. , oppenheimer , b. d. , & dav , r. 2010 , mnras , accepted fukazawa , y. , makishima , k. , tamura , t. , ezawa , h. , xu , h. , ikebe , y. , kikuchi , k. , & ohashi , t. 1998 , pasj , 50 , 187 fukigita , m. & peebles , p. j. e. 2004 , apj , 616 , 643 gabor , j. m. , dav , r. , finlator , k. , & oppenheimer , b. d. 2010 , mnras , 407 , 749 gabor , j. m. , dav , r. , oppenheimer , b. d. & finlator , k. 2011 , mnras , submitted gnat , o. & sternberg , a. 2007 , apjs , 168 , 213 greif , t. h. , glover , s. c. o. , bromm , v. , & klessen , r. s. 2009 , mnras , 392 , 1381 haardt , f. & madau , p. 2001 , in clusters of galaxies and the high redshift universe observed in x - rays , recent results of xmm - newton and chandra " , xxxvith rencontres de moriond , eds .
neumann & j.t.t .
haardt , f. , & madau , p. 2011 , arxiv:1105.2039 heckman , t. m. , norman , c. a. , strickland , d. k. , & sembach , k. r. 2002 , apj , 577 , 691 hellsten , u. , hernquist , l. , katz , n. , & weinberg , d. h. 1998 , apj , 499 , 172 hopkins , a. m. & beacom , j. f. 2006 , apj , 651 , 142 howk , j. c. , ribaudo , j. s. , lehner , n. , prochaska , j. x. , & chen , h .- w .
2009 , mnras , 396 , 1875 hui , l. & gnedin , n. y. 1997 , mnras , 292 , 27 jarosik , n. et al .
2010 , apjs , submitted , apjs , 192 , 14 katz , n. , weinberg , d. h. , & hernquist , l. 1996 , apjs , 105 , 19 katz , n. , kere , d. , dav , r. , & weinberg , d. h. 2003 , the igm / galaxy connection .
the distribution of baryons at z=0 , 281 , 185 kere , d. , katz , n. , weinberg , d. h. , & dav , r. 2005 , mnras , 363 , 2 kere , d. , katz , n. , fardal , m. , dav , r. , & weinberg , d. h. 2009 , mnras , 395 , 160 kitayama , t. & suto , y. 1996 , apj , 469 , 480 kobayashi , c. , springel , v. , & white , s. d. m. 2007 , mnras , 376 , 1465 kollmeier , j. a. , weinberg , d. h. , dav , r. , & katz , n. 2003 , apj , 594 , 75 lopez , s. , ellison , s. , dodorico , s. , & kim , t .- s .
2007 , a&a , 469 , 61 madau p. , ferrara a. , & rees m. j. 2001 , apj , 555 , 92 maller , a. h. , & bullock , j. s. 2004 , mnras , 355 , 694 marigo , p. 2001
, a&a , 370 , 194 martin , c. l. 2005 , apj , 621 , 227 mnard , b. , scranton , r. , fukugita , m. , & richards , g. 2010 , mnras , 405 , 1025 mulchaey , j. s. , & chen , h .- w .
2009 , apjl , 698 , l46 murray , n. , quatert , e. , & thompson , t. a. 2005 , apj , 618 , 569 narayanan , a. , wakker , b. p. , & savage , b. d. 2009 , apj , 703 , 74 narayanan , a. , et al .
2010 , apj , 730 , 15 nath , b. b. & trentham , n. 1997 , mnras , 291 , 505 nicastro , f. , et al .
2005 , nature , 433 , 495 oppenheimer , b. d. & dav , r. a. 2006 , mnras , 373 , 1265 oppenheimer , b. d. , dav , r. , & finlator , k. 2007 eas , 24 , 157 oppenheimer , b. d. & dav , r. a. 2008 , mnras , 387 , 587 oppenheimer , b. d. & dav , r. a. 2009 , mnras , 395 , 1875 ( od09 ) oppenheimer , b. d. , dav , r. , & finlator , k. 2009 , mnras , 396 , 729 oppenheimer , b. d. , dav , r. , kere , d. , fardal , m. , katz , n. , kollmeier , j. , weinberg , d. h. 2010 , mnras , 406 , 2325 peterson , j. r. , kahn , s. m. , paerels , f. b. s. , kaastra , j. s. , tamura , t. , bleeker , j. a. m. , ferrigno , c. , & jernigan , j. g. 2003 , apj , 590 , 207 pettini , m. , madau , p. , bolte , m. , prochaska , j.x . ,
ellison , s.l .
, & fan , x. 2003 , apj , 594 , 695 prochaska , j. x. , chen , h .- w . , howk , j. c. , weiner , b. j. , & mulchaey , j. 2004 , apj , 617 , 718 rauch , m. , sargent , w. l. w. , & barlow , t. a. 2001 , apj , 554 , 823 rauch , m. 1998 , ara&a , 36 , 267 ryan - weber , e. v. , pettini , m. , madau , p. , & zych , b. j. 2009 , mnras , 395 , 1476 savage , b. d. , lehner , n. , wakker , b. p. , semback , k. r. , & tripp , t. m. 2005 , apj , 626 , 776 scannapieco , e. , ferrara , a. , & madau , p. 2002 , apj , 574 , 590 schaye , j. , theuns , t. , rauch , m. , efstathiou , g. , & sargent , w. l. w. 2000 , mnras , 318 , 817 schaye , j. , aguirre , a. , kim , t .- s . , theuns , t. , rauch , m. , & sargent , w.l.w
2003 , apj , 596 , 768 schaye , j. , carswell , r. f. , & kim , t .- s .
2007 , mnras , 379 , 1169 schaye , j. , et al .
2010 , mnras , 402 , 1536 scott , j. e. , kriss , g. a. , brotherton , m. , green , r. f. , hutchings , j. , shull , j. m. , & zheng , w. 2004 , apj , 615 , 135 shen , s. , wadsley , j. , & stinson , g. 2010 , mnras , 407 , 1581 simcoe , r. a. , sargent , w. l. w. , & rauch , m. 2004 , apj , 606 , 92 simcoe , r.a . ,
sargent , w.l.w . , rauch , m. , & becker , g. 2006 , apj , 637 , 648 smith , b. d. , hallman , e. j. , shull , j. m. , & oshea , b. w. 2011 , apj , 731 , 6 sommer - larsen , j. & fynbo , j. p. u. 2008 , mnras , 385 , 3 songaila , a. , cowie , l.l .
1996 , aj , 112 , 335 songaila , a. 2001 , apj , 561 , l153 springel , v. & hernquist , l. 2003a , mnras , 339 , 289 springel , v. & hernquist , l. 2003b , mnras , 339 , 312 springel , v. 2005 , mnras , 364 , 1105 steidel , c. c. , erb , d. k. , shapley , a. e. , pettini , m. , reddy , n. , bogosavljevi , m. , rudie , g. c. , & rakic , o. 2010 , apj , 717 , 289 strickland , d. k. , & heckman , t. m. 2009 , apj , 697 , 2030 sutherland , r. s. & dopita , m. a. 1993 , apjs , 88 , 253 tegmark , m. , silk , j. , evrard , a. 1993 , apj , 417 , 54 telfer , r. c. , kriss , g. a. , zheng , w. , davidsen , a. f. , & tytler , d. 2002 , apj , 579 , 500 tepper - garcia , t. , richter , p. , schaye , j. , booth , c. m. , dalla vecchia , c. , theuns , t. , & wiersma , r. p. c. 2011 , mnras , accepted tescari , e. , viel , m. , dodorico , v. , cristiani , s. , calura , f. , borgani , s. , & tornatore , l. 2011 , mnras , 411 , 826 thom , c. & chen , h .- w .
2008 , apjs , 179 , 37 tornatore , l. , borgani , s. , viel , m. , & springel , v. 2010 , mnras , 402 , 1911 tripp , t. m. , savage , b. d. , & jenkins , e. b. 2000 , apj , 534 , l1 tripp , t. m. , semback , k. r. , bowen , d. v. , savage , b. d. , jenkins , e. b. , lehner , n. , & richter , p. 2008 , apjs , 177 , 39 tytler , d. 1987 , apj , 321 , 49 verner , d. a. , & yakovlev , d. g. 1995 , a&as , 109 , 125 verner , d. a. , ferland , g. j. , korista , k. t. , & yakovlev , d. g. 1996 , apj , 465 , 487 wiersma , r. p. c. , schaye , j. , theuns , t. , dalla vecchia , c. , & tornatore , l. 2009 , 399 , 574 wiersma , r. p. c. , schaye , j. , & smith , b. d. 2009 , mnras , 393 , 99 wiersma , r. p. c. , schaye , j. , dalla vecchia , c. , booth , c. m. , theuns , t. , & aguirre , a. 2010 , mnras , 409 , 132 woosley , s. e. & weaver , t. a. 1995 , apjs , 101 , 181 worsley , m. a. , et al .
2005 , mnras , 357 , 1281 zu , y. , weinberg , d. h. , dav , r. , fardal , m. , katz , n. , kere , d. , & oppenheimer , b. d. 2011 , mnras , 412 , 1059 | we investigate the metallicity evolution and metal content of the intergalactic medium ( igm ) and galactic halo gas from @xmath0 using 110-million particle cosmological hydrodynamic simulations .
we focus on the detectability and physical properties of uv resonance metal - line absorbers observable with _
hubble s _ cosmic origins spectrograph ( cos ) .
we confirm that galactic superwind outflows are required to enrich the igm to observed levels down to @xmath1 using three wind prescriptions contrasted to a no - wind simulation .
our favoured momentum - conserved wind prescription deposits metals closer to galaxies owing to its moderate energy input , while the more energetic constant wind model enriches the warm - hot igm ( whim ) 6.4 times more . despite these significant differences ,
all wind models produce metal - line statistics within a factor of two of existing observations .
this is because @xmath2 , @xmath3 , @xmath4 , and @xmath5 absorbers primarily arise from @xmath6 k , photo - ionised gas that is enriched to similar levels in the three feedback schemes .
@xmath2 absorbers trace the diffuse phase with @xmath7 , which is enriched to @xmath8 at @xmath1 , although the absorbers themselves usually exceed @xmath9 and arise from inhomogeneously distributed , un - mixed winds .
turbulent broadening is required to match the observed equivalent width and column density statistics for @xmath2 . @xmath3 and @xmath4 absorbers trace primarily @xmath10 k gas inside haloes ( @xmath11 ) , although there appear to be too many @xmath3 absorbers relative to observations .
we predict cos will observe a population of @xmath5 photo - ionised absorbers tracing @xmath6 k , @xmath12 gas with equivalent widths of 10 - 20 m . @xmath13 and @xmath14 are rarely detected in cos s / n=30 simulated sight lines ( @xmath15 ) , although simulated @xmath14 detections trace halo gas at @xmath16 k. in general , the igm is enriched in an outside - in manner , where wind - blown metals released at higher redshift reach lower overdensities , resulting in higher ionisation species tracing lower - density , older metals . at @xmath1 ,
the 90% of baryons outside of galaxies are enriched to @xmath17 , but the 65% of unbound baryons in the igm have @xmath18 and contain only 4% of all metals , a large decline from 20% at @xmath19 , because metals from early winds often re - accrete onto galaxies while later winds are less likely to escape their haloes .
we emphasise that our results are sensitive to how metal mixing is treated in the simulations , and argue that the lack of mixing in our scheme may be the largest difference with other similar publications .
[ firstpage ] galaxies : formation ; intergalactic medium ; cosmology : theory ; methods : numerical |
scanning tunneling microscopy@xcite ( stm ) has proven to be one of the most useful tools to measure and characterize low energy electron spectroscopy , at least on open surfaces where the microscope tip has a direct access to the physical region under study . in a series of recent works , kondo impurities @xcite on the surface of noble metals
have been studied using the stm.@xcite in the kondo problem there is a characteristic temperature @xmath0 that separates the low temperature from the high temperature regimes . in the low temperature regime ( @xmath1 )
, the spectral density of a kondo impurity develops a resonance at the fermi energy , known as the kondo resonance , and consequently locally changes the low energy spectroscopic properties of the system .
when the microscope tip is placed on top of the impurity , the characteristic conductance vs voltage @xmath2 shows a typical fano - like@xcite structure that is identified as the fingerprint of the kondo effect . in this way
, the kondo behavior of several transition metal impurities on different noble metal surfaces has been identified . in particular ,
the temperature dependence of the kondo effect has been measured and , by studying co impurities on different cu surfaces , the scaling of the kondo temperature with the host electron density has been shown.@xcite stm has also been used to create nanostructures on noble metal surfaces by manipulation of single atoms .
this allows to positionate impurities at special points of the surface or to engineer the environment of the impurity . in recent experiments
the kondo effect for a @xmath3 impurity inside a quantum corral has been observed.@xcite the fano like structure of the conductance observed when the stm tip is on top of a kondo impurity , can be explained with a simple phenomenological model which includes an impurity resonant state at the fermi energy .
this resonance , that represents the kondo resonance , has a width @xmath4 . in this model ,
the differential conductance though the tip @xmath5 is given by : @xmath6 here @xmath7 is a background conductance , @xmath8 is a constant , @xmath9 and @xmath10 is the fano parameter that depends on the electronic structure and on the tunneling matrix elements . for a more quantitative estimation , a many body calculation including the electron - electron interactions leading to the kondo resonance
is needed . moreover ,
if the impurity is placed on a surface that is structured at a nanoscopic scale , the local density of states may strongly depend on energy , an effect that changes the shape of the kondo resonance @xcite and consequently of the voltage dependence of the stm conductance .
a nanostructure with a characteristic length @xmath11 has a characteristic energy scale @xmath12 where @xmath13 is the fermi velocity . in systems with @xmath14 we expect strong effects .
quantum corrals on the @xmath15 @xmath16 surface can be engineered with the desired energy scale @xmath17 , however it is not clear how much the surface states and bulk states contribute to the kondo effect . despite of some notorious effects like the observation , in elliptical corrals , of the so - called mirages,@xcite
some experiments indicate that the surface states play almost no role in the development of the kondo effect.@xcite if so , the kondo screening would not be affected by the corral that confines only surface states .
a detailed analysis of the stm conductance can be used to elucidate this point , however to do that the surface nanostructure and the kondo correlations have to be treated on the same footing . in what follows we present results for the tip conductance on top of a kondo impurity .
using the numerical renormalization group , we evaluate the finite temperature conductance for an impurity on open and clean surfaces and on structured ones .
in this section we present the model and briefly discuss how the stm conductance is calculated using wilson s renormalization group .
we then present the results for different situations .
our starting point is an extended anderson model for magnetic impurities described with the following hamiltonian:@xcite @xmath18 where the operator @xmath19 creates an electron with spin @xmath20 at the impurity orbital with energy @xmath21 and coulomb repulsion @xmath22 , @xmath23 creates an electron in an extended state with quantum numbers @xmath24 and @xmath25 and energy @xmath26 .
the last term is a falicov - kimbal like term which takes into account a coulomb repulsion between the electrons at the impurity orbital and the ones at extended states .
the quantum number @xmath24 includes the band index and crystal momentum . in the case of impurities on the cu ( 111 ) surface
there is a surface band that is relevant for determining the stm conductance.@xcite the stm conductance at low temperatures is given by@xcite @xmath27 where @xmath28 is the voltage drop from the tip to the sample , and @xmath29 is the tip density of states that is assumed to be constant around the fermi energy . in the linear response regime ,
the quantity @xmath30 is given by @xmath31 here @xmath32 is the retarded green function of the operator @xmath33 , @xmath34 and @xmath35 are matrix elements for tunneling of an electron from the tip to the conduction band states and impurity orbital respectively and @xmath36 is the field operator that creates an electron on the surface at coordinate @xmath37 . in what follows
we consider that the tip is on top of the impurity and take @xmath38to be the impurity coordinate .
the green function is calculated using the numerical renormalization group ( nrg ) . for a general description of the method
we refer to refs . and , here we give only some details that are specific to our calculation .
the starting point of the nrg is to reduce the hamiltonian ( [ ham ] ) to that of a linear chain @xmath39 the states described by the operators @xmath40 are wilson s orbitals centered at the impurity with the state @xmath41 associated with @xmath42 being a wannier like orbital centered at the impurity coordinate .
the parameters @xmath43 generate a logarithmic discretization of the conduction band , and @xmath44 describes the impurity potential scattering .
the hamiltonian ( [ rgham ] ) is diagonalized numerically using an iterative controlled scheme . at each iteration step
@xmath45 , a truncated hamiltonian with @xmath45 orbitals is used and the many body states within an energy scale @xmath46 are evaluated .
the thermodynamic and dynamical properties at each energy scale can then be calculated recursively .
as it is usually done to describe the kondo problem , hamiltonian ( [ rgham ] ) includes a single channel that is due to all the conduction bands . in the case of the @xmath47 surface
there is a surface band and bulk states that contribute to the density of states at the fermi energy .
wilson s orbitals are assumed to be a linear combination of surface and bulk states , in particular the wannier state @xmath41 contains a surface and a bulk component determined by the respective hybridization matrix elements @xmath48 . to calculate the stm conductance when the tip is on top of the impurity
, we take the field operator @xmath49 to be @xmath50 and calculate the retarded green function of the operator @xmath51 the matrix element @xmath52 depends on how much the wannier orbital @xmath41 contains of the surface or bulk states . to be more specific on the nature of the wannier orbital @xmath41 and on the matrix element @xmath52 , a detailed model of the electronic structure of the metal host and of the electron transfer from the tip to the surface
is required . the retarded green function of eq .
( [ green ] ) is calculated as in ref . and @xmath53_{\omega = \varepsilon _ { f}+ev } \nonumber \\ & = & \sum_{\lambda , \lambda ^{\prime } } |m_{\lambda , \lambda ^{\prime } } |^{2}% \frac{(e^{-e_{\lambda } /k_{b}t}+e^{-e_{\lambda ^{\prime } } /k_{b}t})}{z(t)}%
\times \nonumber \\ & & \delta ( \varepsilon _ { f}+ev-(e_{\lambda ^{\prime } } -e_{\lambda } ) ) \label{ro}\end{aligned}\ ] ] where @xmath54 is the grand partition function @xmath55 and @xmath56 . at each energy scale
@xmath46 the discrete spectra is smoothed by replacing the delta functions by a smooth distribution . in what follows we use a logarithmic gaussian that has been proven to give the best results . here
we present the results for a kondo impurity on an open surface . in an energy scale of several @xmath57the bare density of conduction states
is constant around the fermi energy .
then the calculation can be done with the usual band discretization first proposed by wilson .
the results obtained for the conductance are shown in figs .
[ gv ] , [ gvu ] and [ gvt ] .
the stm conductance depends on the following parameters : @xmath4 that gives the low energy scale where the local spectroscopic properties are dominated by the impurity , the temperature @xmath58 and the fano parameter @xmath59 that combines the tunneling matrix elements and some aspects of the band structure .
if the kondo resonance is represented by a simple resonant state , the fano parameter is given by @xmath60 where @xmath61 is an effective band propagator.@xcite for a system with electron - hole symmetry the real part of @xmath62 at the fermi energy is zero and the parameter @xmath10 is proportional to the matrix element @xmath35 that describes the direct tunneling into the localized state . in this case ,
@xmath10 accounts for the interference between the two tunneling channels .
in general , real systems have a non zero @xmath63 and there is no need to assume a direct tunneling to the localized state to have an asymmetric ( @xmath64 ) fano resonance . in the nrg approach ,
the use of a symmetric conduction band simplifies the calculation and most of the numerical work on the kondo effect has been done with this symmetry .
we have introduced the potential scattering term @xmath44 to break the electron - hole symmetry of the band .
the fano resonance then depends on the conduction band parameters and in agreement with ref .
, with a non zero @xmath44 we obtain asymmetric line shapes even for @xmath65 . , @xmath66 , @xmath67 and ( a ) @xmath68 and different values of @xmath69 as indicated in the box ; ( b ) @xmath65 and @xmath70 .
in both ( a ) and ( b ) a fano lineshape is obtained .
, width=283 ] in fig .
[ gv ] the conductance vs applied voltage is shown for @xmath71 and @xmath72 , all parameters are given in units of half the bandwidth @xmath73 . in fig .
[ gv](a ) the results for @xmath74 and a set of the ratios @xmath75 are shown , this ratio determines the structure of the conductance . in fig .
[ gv](b ) , by independently varying @xmath44 for @xmath65 we show that in a full nrg many body calculation , the low energy stm conductance is still given by a single parameter @xmath10 as given by equation ( [ q ] ) . as @xmath76 is increased the resonance narrows indicating a lowering of the kondo temperature .
this is due to a decrease in the local density of states at the fermi energy .
the increase of @xmath76 also induces an increase of the fano parameter . for simplicity , in what follows we take symmetric band ( @xmath77 ) where the fano parameter is given by @xmath78 , where @xmath79 is the density of states of the band .
our calculation , that up to now mainly reproduces known results , shows that the nrg gives reliable results when used to evaluate the stm conductance . in what follows we present some results that include the behavior of the conductance when the energy of the localized orbital approaches the fermi energy , the effect of the falicov - kimbal term @xmath80 and the temperature dependence of the conductance . , and @xmath81 .
as @xmath82 approaches the fermi level , the fano resonance becomes wider .
other parameters are @xmath83 , @xmath84 , @xmath67 , and @xmath85 . , width=283 ] we start by discussing the conductance dependence on the localized orbital energy @xmath21 . as @xmath21
is increased and approaches the fermi energy , first the kondo temperature increases and then the impurity enters an intermediate valence regime . in the conductance shown in fig .
[ gve ] these effects are clearly observed .
a notorious effect for intermediate values of @xmath21 is the occurrence of an asymmetry for finite voltages : away from the fermi level the values of @xmath86 are smaller for @xmath87 than for @xmath88 .
this effect is due to the localized orbital resonance that is close to the fermi energy .
as @xmath21 increases even more ( not shown ) the resonance crosses the fermi level , the impurity occupation is reduced and the conductance shows a broad structure with a width that is given by the resonance width @xmath89 rather than by the kondo temperature . as we now show , the effect of a non zero electron - electron interaction @xmath80 is equivalent to a shift in the impurity level .
the parameter @xmath80 is irrelevant , that means that in the kondo regime the zero temperature fixed point does not depend on this parameter and the low temperature thermodynamic behavior is universal as for @xmath72 .
however this interaction may change some details in the low energy spectral structure for energies of the order or larger than @xmath90 .
we have analyzed the effect of @xmath80 on the stm conductance and the results are summarized in fig .
as @xmath80 increases the fano line shape changes and two different effects can be observed : _ i _ )
the width of the conductance minimum increases indicating an increase in the kondo temperature and _ ii _ ) the voltage dependence of the conductance becomes more asymmetric .
these two points are consistent with an effective shift the localized level toward the fermi energy .
( thin solid line ) , @xmath91 ( dotted line ) , @xmath92 ( dashed line ) , @xmath93 ( solid thick line ) and @xmath94 .
( a ) @xmath95 ( b ) @xmath96 .
other parameters as in fig .
2.,width=283 ] the asymmetry in @xmath86 obtained when the resonant level approaches the fermi level has not been discussed in detail in the literature , however experimental results for @xmath3 on @xmath47 and for @xmath97 on @xmath98 show this type of effect . in ref .
the raw experimental has a marked asymmetriy in the conductance that was attributed to a broad resonance lying just above the fermi energy .
our results may explain the asymmetry as due to the same localized orbital that generates the kondo resonance .
now we present finite temperature results and analyze the temperature dependence of the conductance . to do this
we calculate the finite temperature spectral densities following refs . and and use the finite temperature extension of eq .
( [ gv0 ] ) @xmath99 where @xmath100 is the fermi function . in fig .
[ gvt](a ) the conductance for a small @xmath10 and @xmath101 is shown for different temperatures .
as the temperature increases the width of the low voltage structure increases . in agreement with the experimental results of ref .
, our results show a width that increases according to the low temperature fermi liquid theory . in the inset of fig .
[ gvt ] ( a ) the numerical width is shown for different temperatures , the thick continuous line is a fit with the fermi liquid expression @xmath102 . in fig .
[ gvt](b ) the temperature dependence of a system with @xmath103 is shown .
there is remarkable resemblance between some of these curves and the experimental results of @xmath97 on @xmath98 . and different temperatures : @xmath104 ( solid line ) , @xmath105 ( dotted line ) , and @xmath106 ( dashed line ) .
inset : fermi liquid theory fit to the width @xmath107 of the fano resonance as a function of temperature .
( b ) linear conductance spectra in an asymmetric situation @xmath108 , @xmath83 and @xmath109 for different temperatures as indicated ( the spectra at different temperatures have been shifted vertically for clarity).,width=283 ] we now present results for the stm conductance when the kondo impurity is on a structured surface .
the main effect of a nanostructure on the surface is to produce an energy dependence of the local density of states on a scale @xmath110 that may be comparable with @xmath4 . if this happens the kondo effect may be very much affected as discussed in previous works .
possible realizations of this situation may be obtained by placing impurities in quantum corrals , on small islands or in any structure with a characteristic size of the order of the hundred .
the problem then becomes much more complicated mainly because of two reasons : on the one hand , the nanostructure of the host and the kondo correlations have to be treated on the same footing . on the other hand , a detailed model for the metallic surface and the effect of the nanostructure on the bulk and surface states is needed .
in general , in surfaces like the @xmath47 where there is a surface band , we expect structures like the quantum corrals or small islands to confine surface states keeping the bulk states nearly unaffected .
if so , these structures would change the kondo effect according to the participation of the surface states on the magnetic screening of the impurity .
this point is still controversial : while the observation of quantum mirages seems to indicate that surface modes actively participate in the kondo screening , other experimental results suggest that the kondo effect is due to bulk states .
the stm conductance could be used to clarify this point .
due to the lack of a detailed model suitable to be treated with the renormalization group approach , we resort to a simple and phenomenological model to study the effect of structured local densities of states .
the model reduces the problem to a single effective conduction band . as
before , we assume that wilson s orbitals are built as linear combination of the surface or bulk states .
the surface nanostructure affects mostly the surface component of these states . in order to describe a local density of states with a structure in the desired energy range ,
we resort to the method described in refs . and .
the calculation includes a potential barrier that confines the electrons .
the local density of states so obtained in the absence of impurity has the form of resonant states .
this type of resonances have been observed using the stm in atomic islands or terraces on the @xmath47 surfaces.@xcite if surface states participate actively on the kondo screening , the modulation of the relevant local density of states would be large indicating that the active electrons are confined by the nanostructure ( a large potential barrier ) .
conversely , if the kondo effect is mostly due to the bulk states , the local density of states modulation would be very week . in this case
we use a small effective barrier .
the model is meant to make only qualitative estimations , nevertheless as we show below some important conclusions can be drawn from our analysis . and @xmath111 ( a ) at - resonance ( b ) off - resonance .
other parameter are @xmath94 , @xmath67 , @xmath112 , and @xmath113 . for comparison the lineshape for an open surface
is shown ( thin line).,width=283 ] as stated above , a nanostructure introduces a new length scale and the corresponding energy scale @xmath114 since the number of microscopic parameters of the model is large , we just distinguish to cases : large structures with @xmath115 and small nanostructures with @xmath116 .
another important aspect of the kondo effect in nanostructures concerns the position of the fermi level relative to the structure of the local density of states . in this respect
we consider the two extreme cases : the fermi energy lying at a resonance ( at - resonance case ) or between two resonances ( off - resonance case ) .
the quantum corrals on @xmath117 with characteristic sizes of the order of @xmath118 correspond to the small nanostructure category .
the elliptical corrals of ref .
have the fermi energy at a resonance .
figure [ gvb ] illustrates the case of a relatively large nanostructure ( @xmath119 ) for the at - resonance and off - resonance cases .
the calculation was done with a small @xmath10 and an intermediate barrier .
the low temperature stm conductance shows a fano like structure modulated by a superstructure of peaks separated approximately by the characteristic energy @xmath17 . for the at - resonance case the central valley slightly broadens while for the off - resonance case narrow structures at low voltages are obtained .
as the barrier height decreases the amplitude of the oscillations in the host local density of states decreases and the conductance continuously approaches the open surface result .
figure [ gvb1 ] shows the results for a small system ( @xmath120 ) . in this case
the superstructure is not visible since @xmath17 is out of scale . for the at - resonance ( fig .
[ gvb1](b ) ) case and large barriers , the kondo temperature aincreases and the fano structure is broadened .
this is due to an increase of the local density of states at the fermi energy @xmath121 .
the kondo temperature has an exponential dependence with the density of states and consequently the broadening is very important for intermediate or large barriers . for the off - resonance case ( fig .
[ gvb1](b ) ) , the density of states decreases and the structure narrows , for large barrier a narrow fano structure with a large @xmath10 mounted on a background with a pronounced minimum at @xmath121 is obtained . , @xmath122 and different widths @xmath123 of the resonant states ( a ) at - resonance without barrier ( thin solid line ) and @xmath124 , @xmath125 , @xmath126 ( thick solid line ) ( b ) off - resonance without barrier ( thin solid line ) and @xmath124 , @xmath125 ( thick solid line).,width=283 ]
we have analyzed the stm conductance @xmath86 for kondo impurities on clean and structured surfaces . starting with an anderson model we used the nrg to calculate the physical properties and voltage dependence of the conductance .
for clean surfaces we showed that the numerical approach gives reliable results and reproduces the fano type stm response . in particular , we have studied in some detail how @xmath86 depends on different microscopic parameters and on temperature . as the impurity level is shifted away from the symmetric case @xmath127 , either by changing its energy or by means of a falicov - kimbal interaction , the fano line is distorted . for a set of physical parameters , both the shape and the temperature dependence of the conductance
are in good agreement with experimental observations .
the temperature dependence is also in good agreement with the low temperature fermi liquid theory . to discuss the stm conductance of impurities in structured surfaces we used an effective one channel model . in surfaces like the @xmath47 one , where there is a surface band , this effective channel represents a linear combination of surface and bulk sates .
since a surface structure perturbs much more the surface modes than the bulk modes , the sensitivity of the kondo effect to the structure gives an idea of the surface mode participation in the kondo screening .
we have shown that small structures like the quantum corrals , that are quite efficient in confining surface electrons , should produce important changes in the kondo temperature and on the stm response if the surface modes were active in the kondo physics .
the experiments show that the fano line has essentially the same width when the impurity is on a clean surface or inside a corral , only small changes in the parameter @xmath10 are observed.@xcite this fact is a clear indication that the surface modes can sense the kondo effect but are not important in determining it .
this conclusion is in agreement with estimation of the microscopic parameters for surface and bulk states.@xcite | we study the scanning tunneling microscopy response for magnetic atoms on open and structured surfaces using wilson s renormalization group .
we observe fano resonances associated with kondo resonances and interference effects . for a magnetic atom in a quantum corral coupled to the confined surface states , and experimentally relevant parameters , we observe a large confinement induced effect not present in the experiments .
these results suggest that the kondo screening is dominated by the bulk electrons rather than the surface ones . |
s. bertone@xmath0 , c. le poncin lafitte@xmath1 , v. lainey@xmath2 and m .- c .
angonin@xmath1@xmath1 syrte - obs .
de paris - cnrs / umr8630 , upmc , france + @xmath3 inaf - astronomical observatory of turin , university of turin , italy + @xmath2 imcce - obs . de paris - cnrs / umr8028 , upmc , france e - mail : [email protected] deep space data processing during the last decade has revealed the presence of anomalies in the form of unexpected accelerations in the trajectory of probes @xcite .
the hypothesis made trying to solve this puzzle can be summarized in two main approaches : whether these anomalies are the manifestation of some new physics @xcite , or something is mismodeled in the data processing @xcite .
+ we investigate moyer s book @xcite , which describes the relativistic framework used by space agencies for data processing .
we know that the ephemeris of a space mission is built from subsequent measures involving the light time of a signal traveling between the earth and the probe and the solution of the inverse problem . since the ephemeris is used for both operational ( space probe navigation ) and scientific goals ( measurements for testing fundamental physics ) , a well defined model is then mandatory for both the interpretation of physical data and the orbit reconstruction . in this article
, we suggest an improvement of the light time modeling focusing on the treatment of the so - called `` transponder delay '' .
+ this paper is structured as follows .
+ in section 2 , we give a brief overview of light time computation as described by the moyer s book ; we show that the transponder s delay ( i.e. the time delay between the reception and retransmission of the light signal on board the satellite ) is not accurately taken into account in this model
. in section 3 we present an alternative , more precise , modeling . finally , in section 4 , we compare both modelings to highlight their differences and give some conclusions in section 5 .
+ throughout this work we will suppose that space - time is covered by some global barycentric coordinates system @xmath4 , with @xmath5 , @xmath6 being the speed of light in vacuum , @xmath7 a time coordinate and @xmath8 .
greek indices run from 0 to 3 , and latin indices from 1 to 3 . here
@xmath9/@xmath10 represents the position / velocity of body @xmath11 at time @xmath7 , where @xmath11 can take the value @xmath12 ( ground station ) or @xmath13 ( spacecraft ) .
primed values are related to the _ moyer s modeling _ , while we will generally use non - primed values for our proposed modeling .
deep space navigation is based on the exchange of light signals between a probe and at least one observing ground station .
the calculation of a coordinate light time , as resumed from @xcite , is quite simple : a clock starts counting as an uplink signal is emitted from ground at @xmath14 .
the signal is received by the probe at @xmath15 and then , after a short delay , reemitted towards the earth where it is received by a ground station at @xmath16 .
the clock stops counting and gives the round - trip light time @xmath17 where @xmath18 , @xmath6 is the speed of light , @xmath19 is the shapiro delay @xcite , while @xmath20 and @xmath21 are the transponder delay and other corrections ( ex : atmospheric delay ... ) that we will not detail here , respectively .
the light time @xmath22 is then used to compute two physical quantities : * the _ ranging _ , related to the distance between the probe and the ground station can be computed using @xmath23 * the _ doppler _ , related to the velocity of the probe with respect to the earth , is obtained by differentiating two successive light time measurements , @xmath24 and @xmath25 , during a given count interval @xmath26 .
it has been showed that @xmath27 where @xmath28 is a transponder s ratio applied to the downlink signal when it is reemitted towards the earth and @xmath29 .
+ since the _ doppler _ signal results from the differentiation of the _ ranging _ signal , all constant or slowly changing terms like @xmath20 and @xmath21 obviously cancel out in this modeling .
nevertheless , the electronic delay of some microseconds @xmath20 due to on board processing of the incoming signal requires to consider a different position of the spacecraft at reemission time . in the following ,
we study its consequences on light time modeling for _ ranging _ and _ doppler _ calculations .
+ for this purpose , we introduce an improved light time model @xmath30 taking into account four events ( one more with respect to moyer s model ) : the emission from the ground station at @xmath31 , the reception by the probe at @xmath32 , the reemission at @xmath33 and the reception at ground at @xmath34 .
the additional event @xmath35 accounts for this small delay of @xmath20@xmath36@xmath37 @xmath38 ( at least for modern spacecraft ) so that we get @xmath39 similarly to section [ sec : moyermodel ] , we then use @xmath30 to compute _ ranging _ @xmath40 and _ doppler _ @xmath41 observables as to compare the two modelings presented in sections [ sec : moyermodel ] and [ sec : ourmodel ] , we shall define the difference between the computed light times = - = t_1 - t_1 , @xmath46 and @xmath47 .
let us then analyze the supplementary event @xmath35 . this term is implicitly related to @xmath20 by the first order development ^sc_3=^sc_2+t ^sc_2 + o ( t^2 ) , the implications of this mismodeling are given by @xmath48 where we used eq . , eq . and eq . into eq . and defining @xmath49 as the minkowskian direction between the ground station and the probe .
+ equation highlights the presence of an extra non - constant term , directly proportional to the transponder delay and neglected in moyer s model .
this term also depends on the position and velocity of both the probe and the ground station .
neglecting it would actually lead to a wrong determination of the epoch @xmath50 and to an error in both _ ranging _ and _
doppler_. in order to evaluate the magnitude of the additional term in eq .
, we computed @xmath51 ( giving the difference between the _ ranging _ calculated with the two models ) and @xmath52 ( related to the difference of the _ doppler _ calculated by the two models ) for the observation of a probe .
we used the real orbit of some probes ( rosetta , near , cassini , galileo ) during their earth flyby , which is a particularly favorable configuration .
we used the naif / spice toolkit @xcite to retrieve the ephemeris for probes and planets to be used in the computation .
computing eq . and
it s time derivative for the near probe during its earth flyby on the 23 january 1998 , we found a difference of the order of some @xmath53 for the probe distance @xmath54 calculated by the two models and a difference up to several @xmath55 at the instant of maximum approach for its velocity .
these results are shown in figures [ fig : range ] and[fig : dop ] .
+ in order to highlight the high variability of the transponder delay effect on doppler measurements , we computed @xmath56 for different probes in different configurations with respect to the observing station .
the results are exposed in figure [ fig : dop1 ] and show that this delay can not be simply calibrated at the level of light time calculation nor neglected in the _ doppler _ calculation .
it seems obvious from our results that the influence of the transponder delay can not be reduced to a simple calibration without taking some precautions .
it is indeed responsible for a tiny effect on the computation of light time and has an impact on both _ ranging _ and _ doppler _ determination .
we represent it by a more complete modeling , considering four epochs instead of three . in order to test the amplitude and variability of this effect on real data ,
we compute its influence on some real probe - ground station configurations during recent earth flybys ( near , rosetta , cassini and galileo ) .
+ the observables calculated using moyer s model and our improved model show differences of the order of several @xmath53 and of @xmath57 for the _ ranging _ and the _ doppler _ , respectively .
such an error is acceptable for most operational goals at present .
anyway , we shall highlight that this error is directly proportional to the transponder delay and that for past missions , whose data are still largely used for scientific purposes , transponders were more than @xmath58 times slower that today . in the future too , increasing ephemeris precision @xcite should be followed by the development of faster transponders or by the use of a more precise model .
_ acknowledgements .
_ the authors are grateful to the anonymous referees for their detailed review , which allowed to improve the paper .
s. bertone and c. le poncin - lafitte are grateful to the financial support of cnrs / gram .
j. d. anderson , j. k. campbell , j. e. ekelund , j. ellis , and j. f. jordan . _ anomalous orbital - energy changes observed during spacecraft flybys of earth .
_ physical review letters , 100(9):091102 , march 2008 .
j. d. anderson , p. a. laing , e. l. lau , a. s. liu , m. m. nieto , and s. g. turyshev .
_ indication , from pioneer 10/11 , galileo , and ulysses data , of an apparent anomalous , weak , long - range acceleration .
_ physical review letters , 81:28582861 , october 1998 .
l. iess , m. di benedetto , n. james , m. mercolino , l. simone , and p. tortora .
_ astra : interdisciplinary study on enhancement of the end - to - end accuracy for spacecraft tracking techniques .
_ acta astronautica , volume 94 , issue 2 , p. 699 - 707 | during the last decade , the precision in the tracking of spacecraft has constantly improved .
the discovery of few astrometric anomalies , such as the pioneer and earth flyby anomalies , stimulated further analysis of the operative modeling currently adopted in deep space navigation ( dsn ) .
our study shows that some traditional approximations lead to neglect tiny terms that could have consequences in the orbit determination of a probe in specific configurations such as during an earth flyby .
therefore , we suggest here a way to improve the light time calculation used for probe tracking . |
low dimensional quantum magnets@xcite which display collective singlet ground states are very topical , due to the exotic low temperature properties they display , as well as their relation to high temperature superconductivity@xcite .
quasi - two dimensional s=1/2 systems such as the shastry - sutherland system srcu@xmath6(bo@xmath2)@xmath6 exist@xcite , wherein orthogonal cu@xmath7 dimers are arranged on a square lattice .
this material displays a collective singlet ground state , relatively dispersionless triplet excitations and multiple triplet bound excited states .
quasi - one dimensional quantum magnets are more common , with s=1/2 chains based on organic molecules , such as ttf - cubdt@xcite and mem-(tcnq)@xmath6@xcite , based on cu@xmath7 ( 3d@xmath8 ) , such as cugeo@xmath2@xcite , and most recently based on ti@xmath7(3d@xmath9 ) , such as tiocl@xcite and tiobr@xcite .
these materials undergo spin - peierls phase transitions to a singlet ground state as the temperature is lowered .
related phenomena occurs in quasi - one dimensional quantum magnets with s=1 chains , such as nenp and csnicl@xmath2 @xcite , which enter a haldane singlet phase at low temperatures .
cugeo@xmath2 was the first inorganic spin - peierls system to be discovered .
the singlet ground state associated with cugeo@xmath2 below its spin - peierls phase transition of t@xmath10@xmath11 14.1 k has been well studied@xcite .
such a system is characterized by uniform chains of s=1/2 moments at high temperatures , which dimerise at low temperature to allow singlets to form .
this phase transition breaks translational symmetry , and a singlet - triplet gap is introduced into its magnetic excitation spectrum at its magnetic zone centre .
it possesses a much higher magnetic moment density than the pre - existing organic spin - peierls systems@xcite , and it can be grown in large single crystal form by several different growth techniques .
this has enabled detailed neutron scattering studies of the spin - peierls ground state and its excitations@xcite .
cugeo@xmath2 can also be grown in the presence of impurities , and studies of doped - cugeo@xmath2 have revealed the sensitivity of the spin - peierls ground state to different types of impurities@xcite . in particular they have revealed a remarkably rich temperature - impurity concentration phase diagram in which antiferromagnetic long range order coexists with either a dimerised or uniform structure at sufficiently low temperatures@xcite .
this occurs for both non - magnetic zn@xmath7@xciteand mg@xmath7@xcite substituting for cu@xmath7 , as well as for si@xmath12@xcite substituting for ge@xmath12 .
most of the work on impurities in cugeo@xmath2 has employed dopants which possess a similar or a smaller ionic radius than that of the host ion which they seek to replace . zn@xmath7 ,
mg@xmath7 , and cu@xmath7 have ionic radii of 0.74 @xmath13 , 0.66 @xmath13 , and 0.72 @xmath13 , respectively .
however , some work has also been done on low concentration substitution@xcite of cu@xmath7 with cd@xmath7 , whose ionic radius is much bigger , 0.97 @xmath13 , than that of cu@xmath7 .
the difference in ionic radii severely limits the solubility of cd in cugeo@xmath2 ; nonetheless small single crystals of cu@xmath0cd@xmath1geo@xmath2with x @xmath14 0.002 were grown and studied@xcite .
this previous study@xcite , on small single crystals grown from a flux , showed little change in t@xmath10 , and no coexisting antiferromagnetism at the low cd concentrations and base temperature that could be achieved .
however , interestingly , the critical properties of the spin peierls phase transition changed from three dimensional universality to mean field behaviour on doping with cd .
one interesting dimension of the spin - peierls problem is the relation between the singlet - triplet gap in the spin excitation spectrum , and the order parameter for dimerisation .
cross and fisher@xcite originally argued for the power law relation @xmath15(t ) @xmath11 ( @xmath16 ( t))@xmath17 with @xmath18=2/3 .
the discovery of the spin - peierls state in cugeo@xmath2 has allowed this relationship to be tested directly using inelastic and elastic neutron scattering to measure the temperature dependence of the gap energy , @xmath15(t ) , and the square of the order parameter for the spin - peierls dimerisation @xmath19 .
we report here inelastic neutron scattering measurements of the temperature dependence of the magnetic excitation spectrum at the magnetic zone centre , and of the x - ray diffraction measurements of the superlattice bragg peak intensity , in a new single crystal of lightly doped cu@xmath0cd@xmath1geo@xmath2(x
@xmath14 0.002 ) .
these results show that the simple power law relation between the gap and the dimerisation order parameter is not obeyed , and that a finite triplet gap exists at the spin - peierls phase transition itself .
a single crystal of cu@xmath0cd@xmath1geo@xmath2with x @xmath14 0.002 was grown by the self - flux method in a floating zone image furnace .
the crystal was grown at a rate of @xmath11 5 - 8 mm / hour with an oxygen pressure of 47 kpa .
earlier experience@xcite on flux - grown cu@xmath0cd@xmath1geo@xmath2indicated a low solubility of cd in the cugeo@xmath2host .
for that reason , naturally occurring cd was used in the crystal growth , even though the crystals were intended for neutron scattering studies , and cd has a high neutron absorption cross section .
initial neutron diffraction measurements on the sample showed strong bragg scattering , from a high quality crystal that was single throughout its volume .
its approximate dimensions were of 30 mm long by 5 mm in diameter , and mosaic spread was less than 0.4 degree .
these measurements confirmed that the crystal was orthorhombic with lattice parameters within error the same as the pure material ; _ _ a__= 4.81 , _ _
b__= 8.47 and _ _
c__= 2.94 at 4k .
x - ray diffraction measurements were performed on a small single crystal cut from the large crystal used in the neutron scattering measurements .
the crystal was mounted on the cold finger of a closed cycle refridgerator and aligned within a huber four circle goniometer .
the measurements with a rotating anode cu - k@xmath20 x - ray source and a pyrolitic graphite monochromator were performed at temperatures from 6.5 k to 14.5 k with a temperature stability of @xmath11 0.005 k. the primary purpose of these measurements was to precisely study the critical properties of the spin - peierls order parameter , as measured by the temperature dependence of the * q*=(@xmath21,5,-@xmath21 ) superlattice bragg peak intensity , and to determine the critical exponent , @xmath22 .
another small piece of crystal was cut off and used for magnetic characterization with squid magnetometry .
the characteristic falloff of the dc susceptibility signifying t@xmath10 near 14.1 k was observed .
figure 1 shows the comparison of the normalised susceptibility measurement for both cugeo@xmath2and cu@xmath0cd@xmath1geo@xmath2samples as a function of temperature .
it is clear from this data that the susceptibility of the doped sample is very similar to that of the pure material . at temperatures above t@xmath10 ,
the susceptibility of both samples show a broad maximum characteristic of short range , quasi - one - dimensional correlations . below 10
k the cu@xmath0cd@xmath1geo@xmath2susceptibility is @xmath11 20@xmath23 larger than that of the cugeo@xmath2sample , indicating that cd impurities are indeed present in the system .
they have the effect of freeing - up individual spins near the impurities , thereby increasing the susceptibility .
elastic and inelastic neutron scattering measurements were performed on the large single crystal of cu@xmath0cd@xmath1geo@xmath2at the canadian neutron beam centre , chalk river , using the n5 triple axis spectrometer .
the crystal was mounted in a @xmath24he cryostat with its ( 0,k , l ) plane coincident with the horizontal scattering plane , such that wavevectors near the * q*=(0,1,@xmath21 ) magnetic zone centre could be accessed .
the measurements were made with pyrolytic graphite as both monochromator and analyser crystals , a fixed final neutron energy of 14.7 mev , and with two pyrolytic graphite filters in the scattered beam to reduce higher order contamination .
soller slits determined the horizontal collimation and the resulting horizontal and vertical divergences of the beam were [ 38,36,36,212 ] and [ 58,73,146,636 ] respectively , in minutes of arc , using the convention [ source - monochromator , monochromator - sample , sample - analyser , analyser - detector ] .
elastic neutron scattering measurements were performed at the magnetic zone centre , * q*=(0,1,@xmath21 ) , to search for impurity - induced antiferromagnetic ordering at t=0.32 k. no evidence for magnetic ordering was found .
the lack of change in t@xmath10 in cu@xmath0cd@xmath1geo@xmath2as compared with cugeo@xmath2 , as well as the absence of magnetic order at t=0.32 k , can be used to set an upper limit for the cd concentration in the single crystal sample of cu@xmath0cd@xmath1geo@xmath2 . assuming that the cu@xmath0mg@xmath1geo@xmath2 phase diagram@xcite is applicable to cu@xmath0cd@xmath1geo@xmath2 , at least at low doping concentrations , an upper limit of x @xmath25 0.002 can be set .
we measured the temperature dependence of the @xmath26=(@xmath21,5,-@xmath21 ) superlattice bragg peak intensity , as shown in figure 2 for temperatures close to t@xmath10 .
this peak arises from the dimerisation pattern within the spin - peierls state in cugeo@xmath2 , and its amplitude is proportional to the square of the order parameter . as was done previously to examine the critical properties of doped cugeo@xmath2@xcite , this peak intensity as a function of temperature was fit to a modified power law as shown in equation 1 .
this modified power law includes a correction to scaling term@xcite , with the correction to scaling exponent @xmath27 set to its expected value of 0.5 and @xmath28 .
@xmath29 the solid line in figure 2 shows the fit of equation 1 to the data and clearly this expression describes the data very well for temperatures close to t@xmath10 .
the fit gives t@xmath10 = 14.15 @xmath30 k and a critical exponent @xmath31 .
this value is close to the mean - field value of @xmath32 , and is much larger than the values for @xmath33 from three dimensional universality @xcite that are known to characterize both pure cugeo@xmath2@xcite and lightly doped cugeo@xmath2 in which the dopants possess similar ionic radii to the host ions they replace@xcite .
this mean field result is similar to that found by lumsden et al.@xcite in which the critical behaviour of lightly doped single crystals of cu@xmath0cd@xmath1geo@xmath2grown by the flux method also showed mean field critical exponent @xmath22 values .
these results establish that some cd impurities are present in the crystal , and also provide a quantitative form for the spin - peierls order parameter as a function of temperature , which can then be compared to the temperature dependence of the triplet gap in the excitation spectrum obtained from inelastic neutron scattering .
constant-*q * inelastic neutron scattering scans were performed at the magnetic ordering wavevector * q@xmath34*=(0,1,@xmath21 ) in order to observe the temperature dependence of the triplet excitations at the magnetic zone centre .
the energies of the triplet excitations disperse with wavevector due to the three dimensional nature of the magnetic system . near the ordering wavevector
, this dispersion@xcite varies with the relative wave vector * * q**=*q*-**q**@xmath34 , as : @xmath35 where @xmath15 is the minimum triplet excitation energy or gap energy .
@xmath36 , @xmath37 and @xmath38 are reduced wave vectors expressed in r.l.u . where : + @xmath39 and @xmath40 @xmath41 and @xmath42 @xmath43 and @xmath44 + representative data at 4 k and 21 k , well below and well above t@xmath10 respectively , are shown in figure 3 , and the low temperature singlet - triplet gap of @xmath15 @xmath11 2 mev is identified in the 4 k data . at 21 k ,
the triplet excitation is completely absent and the finite energy peak in the inelastic scattering has been replaced with a weak continuum of scattering from quasi - elastic energies , out to the end of the scan , 4.8 mev .
we note that the triplet excitation at low temperatures exhibits an asymmetric tail to the high energy side .
figure 4 shows the temperature dependence of the quasi - elastic scattering at @xmath5=0.41 mev and at @xmath26=(0,1,@xmath21 ) .
it shows a critical " regime which extends from @xmath11 12 k to @xmath11 21 k within which quasi - elastic scattering is significantly - enhanced compared with either lower or higher temperatures .
we wish to isolate the triplet excitation at @xmath45=(0,1,@xmath21 ) , from the incoherent elastic scattering as well as from the background scattering , and thereby determine the gap energy , @xmath15 , as a function of temperature .
this requires a background subtraction for which we have two options .
we can use the low temperature scattering at t=4 k - suitably modified to exclude the resolution - limited triplet excitation , or we can use the scattering at 21 k. each of these has advantages . for
data the triplet excitation is sharp in energy , and so can be cleanly separated from the remaining scattering , comprised of incoherent elastic scattering from the sample , and energy - independent background scattering from fast neutrons .
however the use of the t=4 k data set as a background does not recognise that inelastic scattering persists above t@xmath10 , albeit in the form of a weak , quasi - elastic spin excitation spectrum for which the t=21 k data set is characteristic . in what follows ,
we employ both a suitably modified t=4 k data set ( lt background ) as well as the t=21 k data set ( ht background ) as the background data set to be subtracted from the signal so as to accurately estimate the scattering from the triplet excitation alone .
this will allow us to examine the sensitivity of the gap , @xmath46 , to the method of background scattering estimation .
figure 5 shows representative constant-*q * scans at ( 0,1,@xmath21 ) , for which a high or low temperature data sets have been subtracted from the scans at temperatures ranging from @xmath11 0.7 t@xmath10 ( 10 k ) to @xmath11 1.1 t@xmath10 ( 15 k ) .
this data was fit to two different forms for s(*q * , @xmath4 ) with the intention of determining the temperature dependence of the gap energy , @xmath15 , and the inverse lifetime , @xmath47 of the triplet excitations .
however a qualitative examination of the data in figure 5 shows a substantial , well defined inelastic peak to exist at @xmath11 1.6 mev or @xmath11 1.8 mev and t=13.83 k @xmath11 0.98 t@xmath10 , depending on whether the low temperature ( lt ) or high temperature ( ht ) data set is used as a background .
one would qualitatively conclude therefore that the gap remains finite at t@xmath10 in this sample of cu@xmath0cd@xmath1geo@xmath2 , a result that is borne out by a quantitative analysis of the excitation spectrum , discussed below .
the inelastic spectra , shown in figure 5 , were fit to two models of s(*q * , @xmath4 ) , each of which was convolved with the four - dimensional instrumental resolution function .
the finite resolution of the measurement combines with the dispersion of the triplet excitations to higher energies at wavevectors away from the magnetic zone centre , equation 2 , and results in the asymmetry of the triplet lineshape , with a high energy tail .
this is accounted for within our resolution convolution , where we employed the spin wave velocities ( see equation 2 ) determined previously@xcite .
the first lorentzian model was employed by regnault et al@xcite , in their analysis of the temperature dependence of the triplet excitation energy near the magnetic zone centre in pure cugeo@xmath2 .
the lorentzian ( lor ) profile is given by : @xmath48\ ] ] the second model was a damped harmonic oscillator ( dho ) given by : @xmath49\ ] ] where @xmath50 is the static susceptibility at * q*. for small damping , the relation between the gap energy in the dho and lor models is : @xmath51 the results of fitting the data to the lor model and the dho model are shown as the solid and dashed lines in figure 5 .
both models are reasonable descriptors of the data . however , the dho model is a better descriptor as its goodness - of - fit parameter , @xmath52 , is typically 10 to 40@xmath23 lower than for the lor model at all temperatures .
this is because the lor spectrum , used earlier @xcite , falls off too slowly with @xmath4 ; indeed its integral in frequency is divergent .
we conclude that the inelastic scattering is best described using the dho form for s(*q * , @xmath4 ) , equation 4 .
the values of the gap energy , @xmath15 , and inverse lifetime , @xmath47 , of the triplet excitations extracted from this analysis are plotted as a function of temperature in figure 6 .
the top panel shows the parameters resulting from an analysis of the data using the ht background , while the bottom panel shows the parameters relevant to the lt background . as can be seen from figure 6 , while the background data set used
influences the details of the fit parameters , it does not affect the overall trends and general features of the temperature dependence of the gap and inverse lifetime of the triplet excitations .
our analysis , employing two different forms of s(*q * , @xmath4 ) and two different background subtractions , results in four forms of the gap energy , @xmath15 , and inverse lifetime , @xmath47 , as a function of temperature , which can then be compared with theoretical expectations . these are plotted as a function of temperature in figure 6 , where the top panel shows the parameters arising from use of the ht background , and the bottom panel shows those arising from use of the lt background . as can be seen the gap energy , @xmath53 2 mev at 4 k , is independent of both the form of s(*q * , @xmath4 ) and the details of the background , provided the lifetime of the triplets is sufficiently long , as it is below t@xmath11 10 k. above @xmath11 10 k , differences between the fitted gap energies progressively increase as the energy width of the excitations , and hence the inverse lifetimes , become larger . however in all four gap vs temperature plots shown in figure 6 , the gap energy , @xmath15 , does not appear to go to zero at t@xmath54 14.15 k. rather the phase transition occurs where the gap energy , @xmath15 , and the energy width or inverse lifetime of the excitation , @xmath47 , cross .
figure 7 shows the gap energy , @xmath15 , plotted as a function of the spin - peierls order parameter as determined from the x - ray scattering determination of the temperature dependence of the superlattice bragg peak intensity shown in figure 2 .
this net intensity is proportional to the square of the order parameter , and consequently we have plotted the square root of the net intensity on the x - axis of figure 7 . for reference
, an x - axis label has been added to the top of figure 7 to denote the actual temperature .
the top panel of figure 7 shows the analysis using the dho form of s(*q * , @xmath4 ) , while the bottom panel shows that using the lorentzian form .
the systematic dependence of the gap on the form of s(*q * , @xmath4 ) and the details of the background can be seen in figures 6 and 7 . as t@xmath10 is approached , the dho form of s(*q * , @xmath4 ) produces a higher value of the gap energy as compared with the lorentzian form . for either form of s(*q * , @xmath4 ) , the use of the lt background results in a higher gap energy near @xmath55 , as compared to when the ht background is used .
as previously discussed , theoretical expectations exist for the relation between the gap energy and the spin - peierls order parameter ; @xmath46@xmath11@xmath56@xmath57 with @xmath18=@xmath58 .
this argument was originally made by cross and fisher@xcite in the context of the spin - peierls transition in ttf - cubdt@xcite .
we have therefore fit the data shown in figure 7 to @xmath46@xmath59@xmath60+@xmath61@xmath57 , with @xmath60 a free parameter and also with @xmath60 set equal to zero . this latter case , with the gap going to zero at t@xmath10 , is consistent with the original theoretical expectation@xcite .
the results of fitting our data to this expression are given in table 1 , for all four data sets ( dho and lor forms of s(*q * , @xmath4 ) , and both ht and lt backgrounds ) .
as can be seen , the fits give a finite value of @xmath60 , the gap energy at t@xmath10 , unless it is constrained to be zero .
only one of the four combinations of s(*q * , @xmath4 ) and background ( lor and ht background ) give behaviour which is roughly consistent with @xmath62@xmath11@xmath56@xmath57 with @xmath18=@xmath58 , and this case gives a finite gap at t@xmath10 of 0.34 @xmath63 0.02 mev .
the fits using the dho form of s(*q * , @xmath4 ) , which allows for the better quality description of the inelastic neutron scattering spectra , give either very low values of the exponent @xmath18 , or non - zero values of the gap at t@xmath10 ranging from @xmath11 0.4 to 0.75 @xmath64 the zero temperature value of the gap .
we therefore conclude that the predicted relation @xmath46@xmath11@xmath61@xmath57 with @xmath18=@xmath58 is not obeyed in cu@xmath0cd@xmath1geo@xmath2 , and that the gap is finite at t@xmath10 .
a question arises as to what role doping plays in this behaviour ?
systematic studies of the cu@xmath0mg@xmath1geo@xmath2 have shown a pseudogap " temperature regime to exist above t@xmath10 for the low dopant concentrations which allow a spin - peierls transition to occur@xcite .
this temperature regime is bordered from below by the appearance of long range spin - peierls order , and from above by signatures indicative of the presence of a gap , such as a suppression in the susceptibility .
this pseudogap regime broadens in temperature with increasing doping until the spin - peierls state is lost altogether beyond x@xmath11 0.03 in cu@xmath0mg@xmath1geo@xmath2 .
the low doping level present in the cu@xmath0cd@xmath1geo@xmath2sample studied here , and the observation of pseudogap - like behaviour in cu@xmath0mg@xmath1geo@xmath2 , suggest that the finite gap at t@xmath10 may be intrinsic to pure cugeo@xmath2 as well .
surprisingly , in a previous study@xcite the temperature dependence of @xmath46 for pure cugeo@xmath2 was found to be consistent with @xmath46@xmath11@xmath56@xmath57 with @xmath18=@xmath58 .
however these earlier measurements focused on the triplet excitations at a wavevector slightly displaced from the dimerization zone centre , and employed the less satisfactory lorentzian form for s(*q * , @xmath4 ) only .
it may be of interest to revisit this problem in pure cugeo@xmath2 .
it is notable that pseudo - gap behaviour has also been observed in the unconventional spin - peierls material tiocl@xcite .
this material exhibits both a low temperature dimerization into a singlet ground state below t@xmath65 and an intermediate temperature phase characterized by an incommensurate structural distortion . above this phase transition ,
a uniform phase exists which displays characteristics of a finite gap and the nmr signature for this pseudogap is maintained to @xmath11 1.3 t@xmath66@xcite .
@l*15@l & ht background&< background & + & @xmath18 & @xmath67&@xmath18 & @xmath67 + dho&0.43 @xmath63 0.06 & 0.85 @xmath63 0.04&1.0 @xmath63 0.4 & 1.54 @xmath63 0.04 + lorentzian&0.67 @xmath63 0.02&0.34 @xmath63 0.02&0.82 @xmath63 0.05 & 0.88 @xmath63 0.02 + dho&0.30
@xmath63 0.03 & 0 & 0.13 @xmath63 0.03 & 0 + lorentzian&0.81 @xmath63 0.04&0&0.48 @xmath63 0.01 & 0 +
inelastic neutron scattering and x - ray diffraction measurements were carried out on a lightly doped sample of cu@xmath0cd@xmath1geo@xmath2 .
these x - ray diffraction measurements of the superlattice bragg intensity below t@xmath10 confirmed the mean field behavior of the spin - peierls phase transition in this new large single crystal of cu@xmath0cd@xmath1geo@xmath2grown by floating zone image furnace techniques .
this result is consistent with earlier critical scattering measurements on small single crystals of cu@xmath0cd@xmath1geo@xmath2grown by flux techniques@xcite .
the order parameter measurements as a function of temperature were correlated with inelastic neutron scattering measurements of the singlet - triplet energy gap in this singlet ground state system , for the purpose of testing the relationship between the gap energy and the spin - peierls order parameter .
we investigated the sensitivity of the gap energy extracted from an analysis of the inelastic scattering , to the form used to model s(*q * , @xmath4 ) and to the form of the background scattering .
this analysis showed the inelastic scattering to be best described using a dho form for s(*q * , @xmath4 ) .
we find that the energy gap remains finite at t@xmath10 , as opposed to going to zero , as might have been anticipated on the basis of earlier theoretical expectations@xcite .
we hope that this result on lightly doped cu@xmath0cd@xmath1geo@xmath2can inform and motivate further work on spin - peierls and other singlet ground state systems , and shed light on the formation of the triplet gap at and above t@xmath10 .
we acknowledge the expert technical assistance provided by the staff at nrc - cnbc , chalk river .
this work was supported by nserc of canada .
10 see , for example : * dynamical properties of unconventional magnetic systems * 1998 , edited by a.t . skjeltorp and d. sherrington ,
nato asi series , series e , applied sciences * 349 * ( kluwer academic publishers , boston ) .
see , for example : orenstein j and millis a j 2000 , science , * 288 * , 468 .
shastry b s and sutherland b 1981 , physica b&c , * 108b * , 1069 .
miyahara s and ueda k 1999 , phys .
lett . , * 82 * , 3701 .
kageyama h , yoshimura k , stern r , mushnikov n v , onizuka k , kato m , kosuge k , slichter c p , goto t , and ueda y 1999 , phys .
* 82 * , 3168 .
gaulin b d , lee s h , haravifard s , castellan j p , berlinsky a j , dabkowska h a , qiu y and copley j r d 2004 phys .
lett . , * 93 * , 267202 .
bray j w , hart h r , interrante l v , jacobs i s , kasper j s , watkins g d , wee s h , and bonner j c 1975 phys .
* 35 * , 744 .
jacobs i s , bray j w , hart h r , interrante l v , kasper j s , watkins g d , prober d e and bonner j c 1976 phys .
b , * 14 * , 3036 . huizinga s , kommandeur j , sawatzky g a , thole b t , kopinga k , de jonge w j m and roos j 1979 , phys . rev .
b , * 19 * , 4723 .
lumsden m d and gaulin b d 1999 , phys .
b , * 59 * , 9372 . uchinokura k 2002 , j. phys .
: condens .
matter , * 14 * , r195
. hase m , terasaki i and uchinokura k 1993 , phys .
lett . , * 70 * , 3651 .
hirota k , cox d e , lorenzo j e , shirane g , tranquada j m , hase m , uchinokura k , kojima h , shibuya y and tanaka i 1994 , phys .
* 73 * , 736 .
pouget j p , regnault l p , ain m , hennion b , renard j p , veillet p , dhalenne g and revcolevschi a 1994 , phys .
lett . , * 72 * , 4037 .
kamimura o , terauchi m , tanaka m , fujita o and akimitsu j 1994 , j. phys .
jpn . , * 63 * , 2467 .
nishi m , fujita o and akimitsu j 1994 , phys . rev .
b , * 50 * , 6508 .
fujita o , akimitsu j , nishi m and kakurai k 1995 , phys .
lett . , * 74 * , 1677 .
regnault l p , ain m , hennion b , dhalenne g and revcolevschi a 1996 , phys .
b , * 53 * , 5579 .
lumsden m d , gaulin b d and dabkowska h a 1998 , phys .
b , * 57 * , 14097 .
birgeneau r j , kiryukhin v , and wang y j 1999 , phys .
b , * 60 * , 3643 . seidel a , marianetti c a , chou f c , ceder g and lee p a 2003 , phys . rev .
b , * 67 * , 020405 .
imai t and chou f c , cond - mat/0301425 .
clancy j p , gaulin b d , rule k c , castellan j p and chou f c 2007 , phys .
b , * 75 * , 100401(r ) .
sasaki t , mizumaki m , nagai t , asaka t , kato k , takata m , matsui y , and akimitsu j , cond - mat/0509358 .
ma s , broholm c , reich d h , sternlieb b j , and erwin r w 1992 , phys .
lett . , * 69 * , 3571 .
buyers w j l , morra r m , armstrong r l , hogan m j , gerlach p and hirakawa k 1986 , phys .
* 56 * , 371 .
lumsden m d , gaulin b d and dabkowska h a 1998 , phys .
b , * 58 * , 12252 and references therein .
aharony a and ahlers g 1980 , phys .
lett , * 44 * , 782 .
hase m , terasaki i , sasago y , uchinokura k and obara h 1993 , phys .
* 71 * , 4059 .
oseroff s b , cheong s - w , aktas b , hundley m f , fisk z and rupp l w 1995 , phys .
lett . , * 74 * , 1450 .
lussier j - g , coad s m , mcmorrow d f and paul d mck 1995 , j. phys . : condens .
matter * 7 * , l325 .
sasgo y , koide n , uchinokura k , martin m c , hase m , hirota k and shirane g 1996 phys .
b , * 54 * , r6835 . renard j p , le dang k , vellet p , dhalenne g , revcolevschi a and regnault l p 1995 europhys .
, * 30 * , 475 .
regnault l p , renard j p , dhalenne g and revcolevschi a 1995 europhys .
* 32 * , 579 .
poirier m , beaudry r , castonguay m , plumer m l , quirion g , razavi f s , revcolevschi a and dhalenne g 1995 phys .
b , * 52 * , r6971 .
masuda t , fujioka a , uchiyama y , tsukada i and uchinokura k 1998 , phys .
lett . , * 80 * , 4566 .
cross m c and fisher d s 1979 , phys .
b , * 19 * , 402 .
nishi m , nakao h , fujii y , masuda t , uchinokura k and shirane g 2000 , j. phys .
jpn . , * 69 * , 3186 .
collins m f 1989 , * magnetic critical scattering * , ( oxford university press , new york ) .
newman k g and riedel e k 1982 , phys .
b , * 25 * , 264 . | single crystals of the lightly - doped spin - peierls system cu@xmath0cd@xmath1geo@xmath2have been studied using bulk susceptibility , x - ray diffraction , and inelastic neutron scattering techniques .
we investigate the triplet gap in the magnetic excitation spectrum of this quasi - one dimensional quantum antiferromagnet , and its relation to the spin - peierls dimerisation order parameter .
we employ two different theoretical forms to model the inelastic neutron scattering cross section and @xmath3(*q * , @xmath4 ) , and show the sensitivity of the gap energy to the choice of @xmath3(*q * , @xmath5 ) .
we find that a finite gap exists at the spin - peierls phase transition . |
a _ circle packing _ in @xmath0 is defined to be a finite collection of pairwise interiorwise disjoint metric closed disks in the riemann sphere @xmath7 equipped with the constant curvature @xmath8 metric as usual .
there are no conditions on the radii of the disks .
we say that two closed disks are _ tangent _ if their boundary circles are tangent .
the _ contact graph _ of a circle packing @xmath9 is the graph @xmath10 whose vertex set is in bijection with the disks of @xmath9 , so that two vertices share an edge if and only if the corresponding disks are tangent .
recall that a graph is called _ planar _ if it can be homeomorphically embedded in the plane .
the contact graph @xmath10 of a circle packing in @xmath0 is always planar , because we may draw the vertices of @xmath10 at the metric centers of the disks they correspond to , and connect adjacent vertices with geodesic arcs .
it turns out that conversely , any planar graph @xmath10 can be realized by a circle packing @xmath9 in @xmath0 .
furthermore , if @xmath10 is the 1-skeleton of a triangulation of the 2-sphere @xmath11 , then this @xmath9 unique , up to action by mbius and anti - mbius transformations on @xmath0 .
these results together constitute what is probably the most important theorem in circle packing , known as the koebe @xcite
andreev @xcite thurston theorem . a major motivating factor for the study of circle packings
is that they are , in a certain precise sense , a discrete approximation to the riemann mapping , as originally conjectured by thurston and proved by rodin and sullivan in @xcite .
the close relationship between circle packing and e.g. classical complex analysis has been further confirmed many times , for example see @xcite . for further exposition and references on circle packing ,
see for example the articles and their bibliographies , or the book @xcite by stephenson .
in this article we present some tools for uniformizing circle packings to make the centers and radii of their disks algebraic .
precisely , a circle packing @xmath9 in @xmath0 is called _ algebraic _ if the following are all algebraic : * all tangency points between pairs of disks in @xmath9 , * all centers of disks in @xmath9 , and * all euclidean radii of disks in @xmath9 under stereographic projection .
there are two equivalent definitions of algebraicity of a point @xmath12 .
first , we may say that @xmath13 is algebraic if and only if either the image of @xmath14 under standard stereographic projection to @xmath1 is algebraic , or @xmath15 .
second , we may say that @xmath16 is algebraic if and only if @xmath17 .
the first main theorem of this article is : [ main1 ] [ every g admits an algebraic packing ] let @xmath10 be a finite , simple , planar graph .
then there exists an algebraic circle packing in @xmath0 having contact graph @xmath10 .
recall that a graph is called _ simple _ if it is undirected , does not have loops , and has no repeated edges .
a _ loop _ is an edge from a vertex to itself . to the best of our knowledge ,
theorem [ every g admits an algebraic packing ] has never appeared in print .
next , we consider circle packings on more general riemann surfaces .
a _ circle packing _ on a riemann surface @xmath2 is a collection of pairwise interiorwise disjoint metric closed disks on @xmath2 .
the following is well - known , see @xcite : [ dut ] let @xmath18 be a triangulation of a compact oriented topological surface without boundary
. then there exists a constant curvature closed riemann surface @xmath2 homeomorphic to @xmath18 , and a locally finite circle packing @xmath9 on @xmath2 , so that @xmath9 realizes @xmath18 .
furthermore @xmath2 is uniquely determined by @xmath18 up to action by conformal isomorphisms on @xmath2 , and @xmath9 is then uniquely determined by @xmath18 and @xmath2 up to action by conformal and anti - conformal automorphisms of @xmath2 .
a _ triangulation_[triangulations ] @xmath19 of a surface @xmath20 is a collection of triangular faces @xmath21 , having vertices @xmath22 and edges @xmath23 , with identifications along the edges so that the resulting object is homeomorphic to @xmath20 .
we require that two distinct faces of @xmath18 meet along a single edge , or at a single vertex , or not at all , and that there are no vertex nor edge identifications along the boundary of any one triangle .
then the 1-skeleton of @xmath18 is the graph @xmath24 .
we consider triangulations only up to their combinatorics .
we say that the circle packing @xmath9 _ realizes _
@xmath18 if the contact graph of @xmath9 is equal to the 1-skeleton of @xmath18 . call a riemann surface @xmath2 which comes from some triangulation @xmath18 via theorem [ dut ] _
circle packable_. in other words , we say that @xmath2 is _ circle packable _ if there exists a locally finite circle packing @xmath9 on @xmath2 realizing some triangulation of @xmath2 .
then theorem [ dut ] implies that there are many @xmath2 which fail to be circle packable : this is because for any @xmath25 the moduli space of genus @xmath26 riemann surfaces is uncountable , but there are only countably many triangulations , considered up to combinatorial equivalence , of a genus @xmath26 surface .
it is then natural to ask which compact riemann surfaces are circle packable . in @xcite , brooks proves that the circle packable compact riemann surfaces @xmath2 of given genus are dense in their moduli space .
the next two theorems we state here , originally due to @xcite*chapters 8 , 9 , give further information in the direction of an answer to this question : [ main2 ] [ circle packable tori have algebraic modulus ] suppose that @xmath27 is a flat circle packable torus .
then @xmath27 is similar or isometric to @xmath28 , with @xmath29 some algebraic number in the upper half - plane . in other words ,
any flat circle packable torus has algebraic modulus .
we also show that if @xmath9 is a circle packing in @xmath28 with @xmath29 in the upper half - plane , then the disks of @xmath9 have algebraic radii , and furthermore @xmath9 may be chosen so that all of its disks centers and tangency points lift to algebraic numbers in @xmath1 .
next : [ main3 ] [ circle packable compact surfaces are algebraic ] suppose that @xmath2 is a compact circle packable riemann surface without boundary of genus at least 2 , and of constant curvature @xmath3 .
then @xmath2 is isometric to the quotient of @xmath4 by a subgroup of @xmath30 .
here @xmath4 denotes the hyperbolic plane , and @xmath31 the group of orientation - preserving isometries of @xmath4 .
we also show that then we may take the ( hyperbolic ) centers of the circles to be algebraic , and also that each circle has a hyperbolic radius which is the logarithm of an algebraic number . beyond these results , and
the denseness theorem mentioned above , not much is known about which compact riemann surfaces are circle packable .
the analogous question in the open surface case has been answered completely by williams in @xcite , where he proves that every open complete constant curvature riemann surface @xmath2 admits a locally finite circle packing having contact graph triangulating @xmath2 .
more recently , kojima , mizushima , tan , and others have studied the closely related question of which projective structures on surfaces are packable .
an excellent recent survey of results and open problems in that area is @xcite .
another closely related question is the following : if @xmath32 is a complete orientable finite - volume non - compact hyperbolic 3-manifold , then each of its cusps is isometric to @xmath33 , where @xmath27 is some flat torus
. then one may ask which flat tori appear in this way .
it is known that only countably many do . in @xcite ,
nimershiem proves that every circle packable flat torus arises as the cusp of some @xmath32 as above , and gives an explicit construction .
we now move on to our methods of proof for theorems [ main1 ] , [ main2 ] , [ main3 ] .
we have three different methods .
each of the next three sections of this article addresses a different one of these three methods of proof .
each approach works to prove a subset of theorems [ main1 ] , [ main2 ] , [ main3 ] , but none seems to work to prove all three . in each case , we include a discussion of the obstructions to applying a given method to those of our three main theorems which it does not handle .
the first method , which will prove theorems [ main1 ] and [ main2 ] , is via a result from model theory : [ tarski ] suppose that @xmath20 is a first - order sentence in the theory of real - closed fields . if @xmath20 is true in one real - closed field , then it is true in every real - closed field .
a _ real - closed field _ is an ordered field @xmath21 , so that any positive element of @xmath21 has a square root in @xmath21 , and so that any odd - degree polynomial over @xmath21 has a root in @xmath21 .
the only two examples we care about here are the real numbers @xmath34 and the real algebraic numbers @xmath35 .
first - order sentence in the theory of real - closed fields _ is , roughly speaking , a finite logical statement , in which we may use the usual logical connectives and quantifiers ( @xmath36 , @xmath37 , @xmath38 , @xmath39 , @xmath40 , @xmath41 , @xmath42 ) , as well as the symbols from the theory of real - closed fields ( @xmath43 , @xmath44 , @xmath45 , @xmath46 , @xmath47 , @xmath48 ) , so that all variables in the statement are appropriately quantified .
our plan is to translate the statements of theorems [ main1 ] and [ main2 ] into appropriately constructed first - order sentences which we know to be true over @xmath34 by other means , and then to apply tarski s theorem [ tarski ] to obtain the truth of these sentences over @xmath35 .
theorem [ main1 ] follows almost immediately , but the situation for theorem [ main2 ] is more subtle .
this proof technique is covered in detail in section [ tarski s section ] .
swan provides a nice expository article on tarski s theorem [ tarski ] , including a proof , in @xcite , and cites the book @xcite of kreisel and krivine as his source .
the proof of tarski s theorem is by elimination of quantifiers .
our second technique works only to prove theorem [ main1 ] .
it proceeds via the following algebraicity result of thurston in the flavor of the mostow
prasad rigidity theorem , found in @xcite*proposition 6.7.4 : [ thurston algebraicity prop ] if @xmath49 is a discrete subgroup of @xmath50 such that @xmath51 has finite volume , then @xmath49 is conjugate to a group of matrices whose entries are algebraic .
we sketch a proof of proposition [ thurston algebraicity prop ] at the end of section [ thurston s section ] .
our proof of theorem [ main1 ] proceeds as follows .
we first complete our finite simple planar graph @xmath10 to the 1-skeleton of a triangulation @xmath18 , and circle pack the resulting triangulation in @xmath0 by the koebe
thurston theorem , denoting the resulting packing by @xmath9 .
then every connected component @xmath52 of @xmath53 is a curvilinear triangle .
for such a @xmath52 , let @xmath54 be the metric closed disk in @xmath0 containing @xmath52 , whose boundary circle passes through the corners of @xmath52 .
then the collection @xmath55 of all such @xmath54 is called the _ dual circle packing _ to @xmath9 .
let @xmath56 be the group generated by reflections in the boundary circles of the disks of @xmath9 and @xmath57 .
we argue that @xmath56 is a discrete , finite covolume subgroup of @xmath58 .
it follows that the group @xmath49 consisting of orientation - preserving elements of @xmath56 is also discrete and of finite covolume
. then proposition [ thurston algebraicity prop ] allows us to assume without loss of generality that @xmath59 .
it follows easily that the tangency points of pairs of disks in @xmath9 are algebraic .
we obtain the algebraic circle packing whose contact graph is our original graph @xmath10 as a sub - packing of @xmath9 , establishing theorem [ main1 ] .
we work out the details of this technique in section [ thurston s section ] .
the approaches via tarski s theorem [ tarski ] or proposition [ thurston algebraicity prop ] had not previously been used to prove algebraicity statements for circle packings .
the third and final technique we exposit is due to mccaughan , and was applied in his ph.d .
thesis @xcite*chapters 8 , 9 to prove theorems [ main2 ] and [ main3 ] .
although it was not noted by mccaughan , this approach also works to prove theorem [ main1 ] .
the main tool is the following lemma of real algebraic geometry : [ mccaughan ] let @xmath60 be a real algebraic variety defined over a field @xmath61 , in @xmath62 variables , and let @xmath63 be a point of @xmath60 which is isolated if we give @xmath60 the subspace topology from @xmath64 .
then the coordinates of @xmath63 are algberaic over @xmath65 .
then the idea is to express the existence of a suitably normalized circle packing in the complex or hyperbolic plane by a set of algebraic equations defining a variety .
one shows that this variety is non - empty and that the point of the variety corresponding to the circle packing is appropriately isolated , and then applies the lemma .
we work out the details of this approach , and also sketch a proof of lemma [ mccaughan ] , in section [ mccaughan s section ] .
* acknowledgements .
* we thank ralf spatzier for pointing us indirectly to thurston s proposition [ thurston algebraicity prop ] .
thanks to chris hall for referring us to tarski s theorem [ tarski ] .
thanks to sergiy merenkov for referring us to mccaughan s thesis @xcite .
thanks to andreas blass , dan hathaway , and scott schneider for helpful background references on logic and model theory .
[ tarski section ] we begin the section with some terminology from model theory , and then go on to prove theorems [ main1 ] and [ main2 ] using tarski s theorem [ tarski ] . at the end of the section we describe the main difficulty of applying this technique to prove theorem [ main3 ] .
we first give the precise definition of _ elementary statement _ , to go with the statement of tarski s theorem [ tarski ] .
first , an _ atomic predicate in the theory of real - closed fields _ is defined to be a relation of the form @xmath66 or @xmath67 , where @xmath26 is a polynomial with integer coefficients in the variables @xmath68 .
we fix once and for all a countable collection of free variables which we may use in our atomic predicates .
the collection of _ elementary predicates in the theory of real - closed fields _ is defined to be the smallest collection * containing all of the atomic predicates , * closed under negation @xmath38 , disjunction @xmath39 , and conjunction @xmath40 , * closed under existential quantification @xmath69 , and * closed under universal quantification @xmath70 .
in particular , elementary predicates are finitely long .
finally , an _ elementary statement in the theory of real - closed fields _ is an elementary predicate in that theory having no free variables , that is , all of whose variables have been appropriately quantified .
tarski s theorem [ tarski ] says that one can not distinguish between different real - closed fields using only elementary statements .
we consider a simpler example : an _ elementary statement in the theory of groups _ is defined in the analogous way , but the atomic predicates are just relations of the form @xmath71 .
we _ can _ distinguish between different groups in this restricted language : for example , the statement @xmath72 is true in @xmath73 , but is false in @xmath74 .
tarski s theorem says that this does not happen in the theory of real - closed fields . for a more thorough exposition on these definitions and on tarski s theorem ,
see @xcite*section 2 .
let @xmath10 be a finite simple planar graph .
the existence of a circle packing having contact graph @xmath10 may be expressed as follows .
enumerate the vertices of @xmath22 by @xmath75 . for
every vertex @xmath76 of @xmath22 , let @xmath77 be free variables , which will naturally represent the @xmath78-coordinate , @xmath79-coordinate , and euclidean radius , of the disk corresponding to @xmath76 .
* for @xmath80 , let @xmath81 be the atomic predicate @xmath82 .
these statement will ensure that the radii of our disks are positive . * enumerate the edges @xmath83 of @xmath10 . for every @xmath84 ,
let @xmath85 be the atomic predicate @xmath86 .
this statement will ensure that the disks corresponding to @xmath87 and @xmath88 are tangent . *
enumerate the pairs of vertices @xmath89 of @xmath22 which do _ not _ share an edge .
suppose there are @xmath90 such pairs .
for every such pair , let @xmath91 , with @xmath92 , be the atomic predicate @xmath93 .
this statement will ensure that the disks corresponding to @xmath87 and @xmath88 do not meet .
* then , let @xmath20 be the elementary statement : @xmath94 then the truth of @xmath20 if the existential symbol is taken to quantify over @xmath34 is equivalent to the existence of a circle packing in @xmath1 having contact graph @xmath10 .
furthermore @xmath20 is true when quantified over @xmath34 , by the koebe
thurston theorem , as we discussed before .
thus by tarski s theorem [ tarski ] , the statement @xmath20 is also true when quantified over @xmath35 .
theorem [ main1 ] follows immediately .
our approach is along the same lines as in the proof of theorem [ main1 ] , but there are added difficulties .
we begin by pointing out the complications , and outlining our plan .
let @xmath18 be a triangulation of a topological 2-torus .
note first that the existence of a circle packing on some flat torus @xmath27 realizing @xmath18 is equivalent to the existence of a doubly translation - periodic circle packing in the universal cover @xmath1 of @xmath27 , realizing the lift of the triangulation @xmath18 .
thus we wish to express , via an elementary statement in the theory of real - closed fields , the existence of a @xmath29 in the upper half - plane , and a @xmath95-periodic circle packing @xmath96 in @xmath1 , so that the image of @xmath97 in the quotient @xmath98 is a circle packing @xmath9 in the torus @xmath27 having contact graph equal to the 1-skeleton of @xmath18 .
however , our elementary statement can not have an infinite number of constituent atomic predicates , so we can not simply pick @xmath77 for every vertex of the infinite packing @xmath97 and quantify over them . instead
, our plan is to pick a sort of `` fundamental domain '' for @xmath97 , having only finitely many disks , so that the image of these disks under the action by @xmath95 is all of @xmath97 .
we now work out the proof in detail . beginning with the triangulation @xmath18 , let @xmath27 be a flat torus , and let @xmath9 be the circle packing in @xmath27 having contact graph equal to the 1-skeleton of @xmath18 , by the discrete uniformization theorem [ dut ] .
let @xmath97 be its lift to @xmath1 .
we now apply some normalizations to @xmath97 .
first , note that for example @xmath95 and @xmath99 act the same way on @xmath1 , so the choice of @xmath29 in the generators of the group which quotients @xmath97 to @xmath9 is not unique .
however , it is well - known ( see @xcite*section 7.2.3 , theorem 2 ) that every flat torus is equivalent ( by a similarity ) to exactly one @xmath100 for @xmath29 in the so - called _ fundamental region of the modular group _ , which we will denote by @xmath22 .
specifically , letting @xmath101 denote the upper half - plane , we have : @xmath102 normalize @xmath97 by euclidean similarities so that @xmath103 for some @xmath104 . by the essential uniqueness of @xmath2 in the discrete uniformization theorem [ dut ] , we have precisely one choice of @xmath105 for this normalization . next ,
pick once and for all a distinguished vertex @xmath106 of @xmath18 .
let @xmath107 denote the disk of @xmath9 corresponding to @xmath106 .
apply translations to @xmath97 so that some lift @xmath108 of @xmath109 is centered at the origin .
then in particular the lifts of @xmath107 in @xmath97 are exactly the images of @xmath108 under the action by @xmath95 .
furthermore , by the essential uniqueness of @xmath9 in the discrete uniformization theorem [ dut ] , the resulting packing @xmath97 is the unique one which is @xmath95-periodic , so that there is a lift of @xmath107 centered at the origin .
( actually , the last sentence is only true if @xmath28 has no non - trivial automorphisms .
if @xmath28 has non - trivial automorphisms then we will require one more normalizing condition to identify @xmath97 uniquely , but almost any reasonably chosen one will do , and it will be clear how to adapt the rest of our argument to include the extra normalization .
we therefore make no further mention of this point . )
we fix a fundamental domain in @xmath1 for the action by @xmath95 .
let @xmath110 be the parallelogram spanned by @xmath43 and @xmath29 , so that the following hold : * the only corner we include in @xmath110 is the bottom - left - hand corner , the origin .
* we include the open bottom and left sides , but neither the top nor the right side of the parallelogram @xmath110 . explicitly ,
if @xmath111 for @xmath112 , then we may express @xmath110 as : @xmath113 this expression will be helpful later .
let @xmath114 be the set of vertices of the contact graph of @xmath97 which correspond to disks whose centers lie in @xmath110 .
the notation in the following discussion can get messy , so to help we adopt the convention that if @xmath115 is a vertex of the contact graph of a packing , say @xmath9 , then we write @xmath116 to denote the disk to which @xmath115 corresponds . similarly , if @xmath20 is a subset of vertices of the contact graph of @xmath9 , then @xmath117 is the collection of disks of @xmath9 corresponding to vertices of @xmath20
. then for example the image of @xmath118 under the action by @xmath95 is all of @xmath97 , and furthermore no two disks of @xmath118 are identified under this action .
let @xmath119 be the set of vertices of the contact graph of @xmath97 which correspond to disks whose centers lie in @xmath120 , similarly @xmath121 .
note that then @xmath122 are pairwise disjoint .
let @xmath123 be the sub - packing of @xmath97 consisting only of those disks corresponding to vertices in @xmath124 .
this @xmath123 will act as our `` fundamental domain '' for the packing @xmath97 .
we now construct an elementary statement in the same style as in the proof of theorem [ main1 ] above .
enumerate the vertices @xmath125 of the contact graph @xmath126 of @xmath123 .
label the @xmath76 so that * the disks of @xmath123 having centers in @xmath110 are exactly @xmath127 ,
* those having centers in @xmath128 are @xmath129 , * those having centers in @xmath130 are @xmath131 , and * those having centers in @xmath132 are @xmath133 .
fix free variables @xmath134 , which will correspond to the centers and radii of the disks of @xmath123 in the natural way .
it will be helpful notationally to fix two additional free variables @xmath135 , which will stand for the real and imaginary parts of @xmath29 , and three additional free variables @xmath136 , which will eventually be set equal to @xmath77 for some @xmath137 , namely that @xmath137 for which @xmath138 corresponds to our distinguished disk @xmath139 .
* define @xmath140 for @xmath126 as we did for @xmath10 in the proof of theorem [ main1 ] . *
let @xmath76 be the vertex of @xmath126 corresponding to the disk @xmath139 .
let @xmath141 be the elementary predicate @xmath142 .
we do this for notational convenience , and to encode the normalization that @xmath108 is centered at the origin . *
recall that the disks @xmath143 and @xmath144 both lie in @xmath123 .
suppose that the vertices of @xmath126 corresponding to these two translates of @xmath108 are @xmath87 and @xmath88 respectively .
then let @xmath145 be the elementary predicate @xmath146 .
this statement further encodes the normalizations on our packing , and picks out @xmath147 for us .
( we write @xmath148 to avoid overloading the variable @xmath137 , which we use frequently as an indexing variable . ) * let @xmath149 be the elementary predicate expressing that @xmath147 lies in the fundamental region @xmath22 of the modular group , in terms of only @xmath150 and @xmath151 .
it is clear how to do this given the equation defining @xmath22 , although to write it out would be messy and unilluminating , so we leave it to the dedicated reader . *
enumerate the pairs of disks of @xmath123 which are identified under the action of @xmath95 .
suppose there are @xmath152 such pairs . for the @xmath137th such pair @xmath153 , with @xmath154
, we have that @xmath155 , for some pair of integers @xmath156 and @xmath157
. then let @xmath158 be the elementary predicate @xmath159 .
these statements encode the desired action of @xmath95 on our packing .
note that @xmath156 and @xmath157 are not variables in @xmath158 , but are actual integers . * for @xmath80 ,
let @xmath160 be the elementary predicate expressing that @xmath161 , in terms of only @xmath162 .
again , it is clear how to do this , given the equation defining @xmath110 .
similarly , * * for @xmath163 , let @xmath160 express that @xmath164 , * * for @xmath165 , let @xmath160 express that @xmath166 , and * * for @xmath167 , let @xmath160 express that @xmath168 .
* then , let @xmath20 be the elementary statement : @xmath169 clearly @xmath20 is true when quantified over @xmath34 , because the centers and radii of the disks of the packing @xmath123
satisfy @xmath20 .
then by tarski s theorem [ tarski ] the statement @xmath20 is true when quantified over @xmath35 as well .
we wish to show that there is a unique tuple of values for the @xmath170 making the constituent predicates of @xmath20 true , and it will then follow that in fact these must lie in @xmath171 , completing the proof .
suppose that @xmath172 is any solution to the statement @xmath20 .
that is , if we plug in these values for the variables into @xmath20 , then all of its constituent predicates evaluate to true .
let @xmath173 be the circle packing in @xmath1 having @xmath62 disks , so that the @xmath137th disk of @xmath173 has center @xmath174 and radius @xmath175 .
now , the disks of @xmath176 are exactly those whose centers and radii are @xmath177 for @xmath178 , so the disks of @xmath173 and @xmath176 are pairwise interiorwise disjoint . similarly , the disks of @xmath173 are pairwise interiorwise disjoint from those of @xmath179 and of @xmath180 .
it follows that @xmath181 is a circle packing in @xmath1 , which is invariant under the action of @xmath182 .
furthermore it is clear that @xmath183 has the same contact graph as our initial packing @xmath9 . finally , by construction of @xmath20 , we have that @xmath184 is normalized so that some lift of @xmath185 is centered at the origin , with @xmath186 .
it follows that @xmath187 , and theorem [ main2 ] is proved .
we conclude the section with a brief discussion of the main difficulty in applying this approach to prove theorem [ main3 ] : the natural approach is to try to follow the proof of theorem [ main2 ] .
then the @xmath77 can be taken to represent hyperbolic centers and radii in the poincar disk model of hyperbolic space .
it turns out that it is possible to express algebraically that the hyperbolic distance between @xmath174 and @xmath188 is @xmath189 , see our proof of theorem [ main3 ] in section [ mccaughan s section ] , so this is not an obstruction .
it is also possible to algebraically normalize the lift @xmath97 of our packing @xmath9 in @xmath2 , where @xmath2 is a compact complete constant curvature @xmath3 riemann surface .
suppose that @xmath190 for @xmath191 , so that @xmath192
. then the elements of @xmath49 are matrices over @xmath34 , and it is possible to express the action of such a matrix on a point @xmath193 algebraically in terms of its matrix entries and the coordinates of @xmath14 .
the difficulty in making the proof go through is in picking out generators for the quotient group @xmath49 so that @xmath194 . in our proof of theorem [ main2 ]
given immediately above , we were fortunate to be able to pick these generators to be @xmath43 and @xmath104 . in our setting , if we wish to write down an elementary sentence @xmath20 which is to have precisely one solution over @xmath34 , expressing the existence of @xmath97 and our selected generators for @xmath49 , then we need a normalization on our generating set which identifies it uniquely .
this normalization needs to be expressible via a finite list of elementary predicates in the variables we quantify over in @xmath20 .
it is not clear how to do this .
the proof in section [ mccaughan s section ] follows a similar strategy to the one we have described here
. however , the reason that the proof in section [ mccaughan s section ] goes through is that it expresses the existence of the desired normalized circle packing @xmath97 via polynomial equations defining an algebraic variety , and in that setting we may encode all of the infinitely many contacts between the disks of @xmath97 .
this is not possible in the present setting , as elementary sentences must be finite .
we begin the section with a proof of theorem [ main1 ] , using proposition [ thurston algebraicity prop ] .
this section relies on some basic , standard tools and facts from hyperbolic geometry . for a general reference on hyperbolic geometry , see for example @xcite . after the proof of theorem [ main1 ] , we sketch the proof of proposition [ thurston algebraicity prop ] , for the convenience of the reader . finally
, we briefly remark on the potential difficulty of applying this approach to theorems [ main2 ] and [ main3 ] .
let @xmath10 be a finite simple planar graph .
we show that there is a circle packing @xmath195 in @xmath1 having contact graph @xmath10 , so that all euclidean centers and radii of disks of @xmath195 , as well as all tangency points of pairs of disks of @xmath195 , are algebraic numbers .
first , let @xmath18 be a triangulation of @xmath11 , so that the 1-skeleton of @xmath18 has @xmath10 as a subgraph .
we will find a circle packing in @xmath1 realizing @xmath18 , which has the algebraicity properties described in the previous paragraph , and obtain @xmath195 as a sub - packing , completing the proof .
we recall a notational convention from the previous section , that if @xmath115 is a vertex of the contact graph of the circle packing @xmath9 , then @xmath116 denotes the disk of @xmath9 corresponding to the vertex @xmath115 .
let @xmath9 be any circle packing in @xmath0 realizing @xmath18 .
then all of the connected components of @xmath53 are curvilinear triangles , and furthermore these triangles are in natural bijection with the set of faces @xmath21 of @xmath18 .
we now construct the so - called _ dual circle packing _ to @xmath9 , denoted @xmath57 . for each such triangle @xmath196 ,
corresponding to the face @xmath197 , let @xmath198 denote the closed disk containing @xmath196 , whose boundary circle passes through the three corners of @xmath196 , that is , the three tangency points between pairs of the disks @xmath199 . then @xmath57 is defined to be the collection of disks @xmath200 .
it is well - known that the boundary circle @xmath201 is orthogonal to all three of @xmath202 , see figure [ orthogonality ] .
it follows that the disks of @xmath57 are pairwise interiorwise disjoint , thus that @xmath57 is a circle packing in @xmath0 .
furthermore the two packings @xmath9 and @xmath57 together completely cover @xmath0 .
( 0,-2.55)(5.0,2.55 ) ( 1.35,1.2)1.35 ( 3.84,1.19)1.16 ( 2.79,-1.14)1.41 ( 2.71,0.44)0.77 ( 2.7,1.19 ) ( 2.06,0.05 ) ( 3.38,0.13 ) ( 2.9814062,1.415)@xmath203 ( 2.4214063,0.875)@xmath204 ( 2.1414063,0.415)@xmath205 ( 2.0814064,-0.265)@xmath206 ( 3.4614062,-0.205)@xmath207 ( 3.2614062,0.495)@xmath208 ( 2.9214063,0.915)@xmath209 we now wish to define the _ reflection group _ associated to the packing @xmath9 .
we first establish some conventions and notation .
for the rest of this proof , we identify the riemann sphere @xmath0 with the boundary at infinity of hyperbolic 3-space @xmath210 .
recall that this identification is consistent with the identification of the group of mbius transformations @xmath211 with the group of orientation - preserving isometries @xmath212 of @xmath210 .
every circle @xmath213 in @xmath0 is now the boundary of a totally geodesic copy of @xmath4 in @xmath210 .
denote by @xmath214 the reflection of @xmath210 along this hyperbolic plane .
the induced boundary map of @xmath214 on @xmath0 is inversion along the circle @xmath213 in the sense of projective geometry .
note that @xmath215 , as it is orientation - reversing .
then the _ reflection group _ associated to @xmath9 , which we will denote by @xmath216 , is the group generated by all of the reflections @xmath214 , for @xmath217 .
the following lemma is well - known , but we sketch its proof for the convenience of the reader .
[ lattice ] the group @xmath56 is discrete and the associated orbifold @xmath218 has finite volume . given a round closed disk @xmath219 in @xmath0 , let @xmath220 be the hyperbolic half - space whose closure in @xmath221 is equal to @xmath219 .
let @xmath222 denote the closure , in @xmath210 , of the complement of @xmath223 .
then @xmath222 is a polyhedron in @xmath210 with finitely many faces and @xmath56 is the group of isometries of @xmath210 generated by reflections along the faces of @xmath222 .
the fact that the circles of @xmath9 meet those of @xmath57 orthogonally implies that the dihedral angles of @xmath222 are all equal to @xmath224 .
in particular , it follows from poincar s polyhedron theorem , see @xcite , that @xmath56 is a discrete group with fundamental domain @xmath222 . we still need to prove that @xmath222 has finite volume .
the cusps of @xmath222 are in bijection with the tangency points of pairs of disks of @xmath9 , equivalently the tangency points of pairs of disks of @xmath57 .
let @xmath152 be such a tangency point .
then there are two disks of @xmath9 , and two of @xmath57 , whose boundary circles pass through @xmath152 . call these disks @xmath225 . for a moment we consider hyperbolic space @xmath226 in the upper half - space model , normalized so that @xmath227 . in this model , under this normalization , each of the hyperbolic half - spaces @xmath228 becomes the intersection of @xmath229 with a vertical euclidean half - space .
furthermore because the disks of @xmath9 meet those of @xmath57 orthogonally , we in fact have that @xmath230 is an open rectangular column .
a straightforward computation shows that the hyperbolic volume contained in this column above any fixed euclidean height is finite .
thus @xmath222 has finite volume in a neighborhood of any of its cusps .
but it has finitely many cusps , and the complement in @xmath222 of some small open neighborhoods of its cusps is compact .
it follows that @xmath222 has finite volume , concluding the proof of lemma [ lattice ] .
next , let @xmath49 denote the group of orientation - preserving elements of @xmath56 .
then @xmath231 is an index 2 subgroup of @xmath56 .
it follows from lemma [ lattice ] that @xmath49 is discrete , and that @xmath232 has finite volume
. we can now apply proposition [ thurston algebraicity prop ] .
then there is a @xmath233 so that @xmath234 .
noting that the reflection group associated to the circle packing @xmath235 is exactly @xmath236 , we conclude that we may assume without loss of generality that @xmath237 .
it will also be helpful to assume , again without loss of generality , that @xmath238 lies outside of the disks of @xmath9 , so @xmath9 is a packing in @xmath1 .
the condition that a point @xmath14 is fixed by an element @xmath239 is algebraic in @xmath14 and the matrix entries of @xmath240 .
thus we get that any fixed point of @xmath241 is algebraic . on the other hand ,
if @xmath152 is a tangency point between some pair of disks @xmath242 , then @xmath152 is fixed by @xmath243 .
we conclude that all tangency points between disks of @xmath9 are algebraic .
it is then a simple computation to show that the euclidean centers and radii of the disks of @xmath9 are algebraic : this is because there is a unique triple of pairwise tangent disks , whose tangency points are @xmath244 , and the conditions defining this triple are algebraic in the euclidean @xmath78- and @xmath79-coordinates of their centers , their euclidean radii , and the coordinates of @xmath244 .
this concludes the proof of theorem [ main1 ] .
proposition [ thurston algebraicity prop ] is a particular case of a more general result of raghunathan @xcite .
we now sketch the proof of proposition [ thurston algebraicity prop ] in the case that @xmath49 is cocompact : consider @xmath49 as an abstract group and denote the embedding of @xmath49 inside @xmath211 by @xmath245 .
we consider @xmath245 as an element of @xmath246 .
the group @xmath50 is an affine algebraic group defined over @xmath247 . in particular , @xmath248 is also an affine algebraic variety defined over @xmath247 .
hence , the galois group @xmath249 acts on @xmath248 .
the variety @xmath248 has only finitely many irreducible components and the action of the galois group permutes them .
it follows that every irreducible component of @xmath248 is invariant by a finite index subgroup of @xmath249 .
this implies that each one of the irreducible components is defined over a finite extension of @xmath247 and hence defined over @xmath250 .
let @xmath60 be the irreducible component of @xmath248 containing @xmath245 . since @xmath60 is defined over @xmath250 and @xmath250
is algebraically closed , we get that @xmath60 contains an element @xmath251 which is defined over @xmath250 .
it follows from the mostow rigidity theorem that @xmath251 and @xmath245 are conjugated , and this concludes the proof of proposition [ thurston algebraicity prop ] in the case that @xmath49 is cocompact . in the general ( finite covolume ) case
, one has to replace the variety @xmath248 with the subvariety consisting of representations which map parabolic elements to elements whose trace squared is 4 .
the argument then is concluded using prasad s extension of the mostow rigidity theorem to the finite volume setting .
finally , we discuss the main difficulty in applying this approach to the settings of theorems [ main2 ] and [ main3 ] .
the natural approach to take is to construct a packing @xmath9 in a compact riemann surface @xmath2 , lift it to the universal cover ( either @xmath1 or @xmath4 ) of @xmath2 , obtain algebraicity in the universal cover , and then apply known rigidity results to obtain algebraicity in the quotient .
however , the reflection group of the lifted packing in the universal cover will not have finite covolume in @xmath210 , thus proposition [ thurston algebraicity prop ] does not apply .
furthermore the finite covolume hypothesis is essential in the proof of proposition [ thurston algebraicity prop ] .
we begin by applying lemma [ mccaughan ] to prove theorem [ main1 ] .
this lemma was previously applied by mccaughan in @xcite*chapters 8 , 9 to prove theorems [ main2 ] and [ main3 ] , and we give these proofs below .
he did not observe that the same lemma works to prove theorem [ main1 ] as well , but there is essentially nothing new in the argument we give .
after these three proofs , we sketch a proof of lemma [ mccaughan ] for the convenience of the reader .
let @xmath10 be a finite simple planar graph .
we wish to show the existence of a circle packing @xmath195 in @xmath1 , so that all euclidean centers and radii of the disks of @xmath195 are algebraic .
it will follow that the points of tangency between disks of @xmath195 are algebraic .
let @xmath18 be a triangulation of @xmath11 the 1-skeleton of which has @xmath10 as a subgraph .
let @xmath9 be a circle packing in @xmath1 realizing @xmath18 , by the koebe
thurston theorem .
we will show that @xmath9 can be taken to have the algebraicity properties described in the preceding paragraph , and we may obtain our desired @xmath195 as a sub - packing of @xmath9 , completing the proof of the theorem .
we now apply some normalizations to @xmath9 .
first , pick some face @xmath252 of @xmath18 .
normalize @xmath9 so that @xmath199 are disks of euclidean radius @xmath43 , having euclidean centers @xmath253 respectively , and so that the rest of the disks of @xmath9 lie in the bounded curvilinear triangular region formed between @xmath199 .
the resulting packing @xmath9 is the unique one having contact graph @xmath18 and satisfying these normalization conditions .
suppose that there are @xmath62 vertices of @xmath18 .
let @xmath60 be the algebraic variety in @xmath254 defined by equations of the form @xmath255 for every pair of vertices @xmath256 of @xmath18 sharing an edge , together with the extra equations @xmath257 .
let @xmath258 be the @xmath259-tuple consisting of the centers and radii of the disks of @xmath9
. then @xmath260 .
we wish to argue that @xmath63 is isolated in @xmath60 under the subspace topology from @xmath254 .
then theorem [ main1 ] will be proved , by lemma [ mccaughan ] . first note
that @xmath63 is well - separated from points of @xmath60 where some radius @xmath175 is negative or zero , because there are finitely many @xmath175 , and all of them are positive in @xmath63 .
suppose that @xmath261 is a sequence of points of @xmath60 , converging to @xmath63 , so that every radius coordinate of every @xmath261 is positive .
pass to a subsequence so that @xmath262 for all @xmath137 .
we claim that then eventually @xmath263 . to see why , let @xmath264 be the collection of disks having centers and radii given by the coordinates of @xmath261 in the natural way .
note first that if @xmath87 and @xmath88 are vertices of @xmath18 with no edge between them , then eventually @xmath265 and @xmath266 are disjoint , because their limiting disks are disjoint . on the other hand , if @xmath267 is an edge of @xmath18 , then the disks @xmath265 and @xmath266 are tangent for every @xmath137 , by the construction of @xmath60 . thus @xmath264 is eventually a circle packing realizing @xmath18 . furthermore if @xmath137 is big enough then the disks of @xmath264 are sufficiently close to their partners in @xmath9 that the disks of @xmath264 must lie in the bounded curvilinear triangular region formed between @xmath199 .
thus we get that eventually @xmath268 by the essential uniqueness of @xmath9 , concluding the proof .
let @xmath18 be a triangulation of the 2-torus , and let @xmath27 be a flat torus and @xmath9 a circle packing in @xmath27 realizing @xmath18 , by the discrete uniformization theorem [ dut ] .
we wish to show that @xmath27 is similar to the quotient of @xmath1 by @xmath95 with @xmath29 an algebraic number .
our proof will also show that then , supposing that @xmath269 , we have that the radii of the disks of @xmath9 are algebraic .
this proof uses ideas similar to those in the proof of theorem [ main2 ] of section [ tarski section ] .
suppose without loss of generality that @xmath270 , and let @xmath97 be a @xmath95-periodic packing in @xmath1 , so that @xmath271 . normalize @xmath97 so that , say , some distinguished disk @xmath107 of @xmath9 has a lift in @xmath97 centered at the origin . in this setting
it is not necessary that we apply enough normalizations to @xmath97 and @xmath29 that they are uniquely determined by our normalizing conditions .
define a fundamental parallelogram @xmath272 for the action by @xmath95 as in the proof of theorem [ main2 ] in section [ tarski section ] , see p. .
let @xmath114 be the set of vertices of the contact graph of @xmath97 corresponding to circles having euclidean center in @xmath110 .
enumerate the vertices @xmath75 of @xmath114 .
fix variables @xmath273 , and two free variables @xmath150 and @xmath151 , which will serve as the real and imaginary parts of @xmath29 .
let @xmath274 be the real algebraic variety in these variables defined by equations as follow : * every disk @xmath219 of @xmath97 may be written @xmath275 , for some @xmath276 and integers @xmath156 and @xmath157 .
( if @xmath219 has center in @xmath110 then @xmath277 . )
suppose that the distinct disks @xmath278 of @xmath97 meet .
write @xmath279 and @xmath280 .
then add @xmath281 ^ 2 + \left [ ( y_i + s_1 b ) - ( y_j + s_2 b ) \right]^2 = ( r_i + r_j)^2 $ ] to the defining equations of @xmath60 .
do this for every pair of distinct disks of @xmath97 which meet .
* to encode our normalization from above , let @xmath282 be the vertex of that disk of @xmath118 which is a lift of our distinguished disk @xmath107 . then add @xmath283 to the defining equations of @xmath60 .
let @xmath63 be the @xmath284-tuple given by the radii , and coordinates of the centers , of the disks of @xmath118 , and by the real and imaginary parts of @xmath29 , in the natural way .
then clearly @xmath285 .
furthermore , we argue , essentially as in the proof of theorem [ main1 ] given earlier in this section that @xmath63 is an isolated point of @xmath60 in the subspace topology from @xmath64 , concluding the proof of theorem [ main2 ] by lemma [ mccaughan ] .
let @xmath18 be a triangulation of a genus @xmath26 surface for @xmath286 , let @xmath2 be a complete compact constant curvature @xmath3 riemann surface , and let @xmath9 be a circle packing in @xmath2 realizing @xmath18 .
we wish to show that @xmath2 is isometric to @xmath287 for some @xmath288 . along the way we will also end up showing that the hyperbolic radius of any disk of @xmath9 is the logarithm of an algebraic number .
the proof proceeds essentially as did the proof of theorem [ main2 ] given in this section .
first , let @xmath97 be a circle packing in @xmath4 , so that the quotient of @xmath4 by some discrete group of isometries of @xmath4 is @xmath2 , and so that the image of @xmath97 under this quotient map is exactly @xmath9 .
pick once and for all a distinguished vertex @xmath106 of @xmath18 , and a neighbor @xmath289 of @xmath106 in @xmath18 , and normalize @xmath97 so that some lift @xmath290 of @xmath291 is centered at the origin , and so that the neighboring lift @xmath292 of @xmath293 is centered along the positive real axis .
this fixes @xmath97 uniquely .
let @xmath294 be the group of isometries of @xmath4 so that @xmath295 .
let @xmath296 be a set of generators for @xmath49 .
fix @xmath297 to be a fundamental domain for the action by @xmath49 , so that @xmath110 contains both the origin , and the hyperbolic center of @xmath292 .
we ask that no two points of @xmath110 are identified by the action by @xmath49 , and that @xmath110 tiles @xmath4 under the action of @xmath49 .
let @xmath114 be the set of vertices of the contact graph of @xmath97 corresponding to disks whose hyperbolic centers lie in @xmath110 .
enumerate the vertices @xmath75 of @xmath114 , and fix real variables @xmath298 . here
the @xmath174 will represent the hyperbolic centers of the disks , in the poincar disk model of hyperbolic space , and the @xmath299 where the @xmath175 represent their hyperbolic radii .
fix @xmath300 additional real variables which will represent the matrix entries of the @xmath301 matrices @xmath302 .
let @xmath303 be the real algebraic variety in all of these variables defined by equations as follows : * every disk @xmath219 of @xmath97 may be written @xmath304 for some @xmath305 and @xmath306 . in particular
the action of @xmath307 sends the hyperbolic centers of disks to hyperbolic centers .
suppose that the distinct disks @xmath278 of @xmath97 meet .
we wish to express , via a polynomial equation in our variables from above , that the distance between the hyperbolic centers @xmath308 of @xmath309 and @xmath310 is exactly @xmath189 . to do this , recall that if @xmath311 , then the hyperbolic distance between @xmath312 and @xmath313 may be expressed as @xmath314 , where : @xmath315 therefore the expression @xmath316 may be simplified to a polynomial .
furthermore @xmath317 , similarly for @xmath313 .
plugging these expressions for @xmath318 into our formula , we may again simplify to a polynomial expression in the @xmath319 , and the variables representing the matrix entries of the @xmath302 .
add the resulting polynomial to the defining equations of @xmath60 for every such pair of distinct disks of @xmath97 which meet .
* as before , encode our normalization on @xmath97 as follows : let @xmath76 be the vertex of @xmath114 corresponding to @xmath320 , similarly @xmath87 for @xmath292 .
add the polynomials @xmath321 to our defining equations for @xmath60 .
let @xmath60 be a variety in @xmath62 variables defined over a field @xmath61 , and let @xmath322 be a point of @xmath60 which is isolated in the subspace topology from @xmath64 on @xmath60 .
we wish to show that then the @xmath76 are algebraic over @xmath65 .
note that if @xmath323 is any field homomorphism @xmath324 which fixes @xmath65 , then @xmath325 .
suppose for contradiction that some @xmath76 fails to be algebraic over @xmath65 .
then we may write @xmath326 , where the @xmath327 are independent transcendentals over @xmath65 , and @xmath14 is algebraic over @xmath328 . in particular
, every @xmath76 is equal to some rational function in the variables @xmath329 .
let @xmath152 be the minimal polynomial for @xmath14 over @xmath328 .
then @xmath152 is irreducible over @xmath328 , thus @xmath330 at @xmath331 .
then by the implicit function theorem , in a neighborhood of @xmath331 we may write the @xmath14-coordinate of the vanishing set of @xmath152 as a smooth function of the @xmath327-coordinates . therefore , choose @xmath332 to be close to @xmath333 , so that @xmath334 for @xmath335 close to but unequal to @xmath14 .
we may ensure in our choice that the @xmath336 are independent transcendentals over @xmath65 .
then @xmath323 fixing @xmath65 and sending @xmath337 and @xmath338 is a field homomorphism from @xmath339 to @xmath34 , which sends @xmath63 to another point of @xmath60 , and which can be arranged by our choices earlier in this paragraph to send @xmath63 as close to itself as desired .
this contradicts the hypothesis that @xmath63 is isolated in @xmath60 . | the main purpose of this article is to demonstrate three techniques for proving algebraicity statements about circle packings .
we give proofs of three related theorems : ( 1 ) that every finite simple planar graph is the contact graph of a circle packing on @xmath0 , equivalently in @xmath1 , all of whose tangency points , centers , and radii are algebraic , ( 2 ) that every flat conformal torus which admits a circle packing whose contact graph triangulates the torus has algebraic modulus , and ( 3 ) that if @xmath2 is a compact riemann surface of genus at least 2 , having constant curvature @xmath3 , which admits a circle packing whose contact graph triangulates @xmath2 , then @xmath2 is isomorphic to the quotient of @xmath4 by a subgroup of @xmath5 .
the statement ( 1 ) is original , while ( 2 ) and ( 3 ) have been previously proved in @xcite*chapters 8 , 9 the ph.d .
thesis of mccaughan .
our first proof technique is to apply tarski s theorem , a result from model theory , which says that if an elementary statement in the theory of real - closed fields is true over one real - closed field , then it is true over any real closed field .
this technique works to prove ( 1 ) and ( 2 ) .
our second proof technique is via an algebraicity result of thurston on finite co - volume discrete subgroups of @xmath6 .
this technique works to prove ( 1 ) .
our first and second techniques had not previously been applied in this area .
our third and final technique is via a lemma in real algebraic geometry , and was previously used by mccaughan to prove ( 2 ) and ( 3 ) .
we show that in fact it may be used to prove ( 1 ) as well . |
since bright spiral disk galaxies contain most of the h i mass in the nearby universe ( @xcite ) , it has always seemed likely that these types of galaxies or at least their progenitors would be the origin of the high - redshift dly@xmath0 systems seen in background qso spectra .
the detection of @xmath4 dly@xmath0 systems ( @xcite ) has now made it possible to search for the galaxies which cause the absorption .
although redshifts are largely unavailable for the objects imaged , recent observations suggest that a wide range of morphological types are responsible for the absorption ( @xcite ) , and that bright spiral galaxies do _ not _ dominate the dly@xmath0 population . in particular , several low surface brightness ( lsb ) galaxies have been found close to qso sightlines ( @xcite ) , while in one case , only an lsb or dwarf galaxy can account for the complete non - detection of a suitable galaxy responsible for a @xmath5 dly@xmath0line ( @xcite ) .
further evidence that dly@xmath0 systems may not originate in bright spiral galaxies , even at higher redshift , comes from measurements of the metallicity of the absorbing gas .
the abundances of alpha elements in dly@xmath0 systems do not appear to be enhanced relative to iron group elements but similar to the solar value ( molaro , centurion & vladillo 1998 ; centurion et al .
1998 ) , while the overall metallicity seems to remain low even at @xmath6 ( pettini et al . 1998 ) .
these results suggest that dly@xmath0 lines may actually arise in systems which are less common than the main star - forming disks we see around us today . in this _
letter _ we use our model of disk formation and evolution to explore differences in metallicity , dust content , and neutral - gas mass density between high surface brightness ( hsb ) and lsb galaxies .
we find that we are able to better reproduce the observational data with models of lsb galaxies , and predict how future abundance measurements may be able to further differentiate between the two .
several authors have shown how differences in surface brightness between galaxies can be readily understood if lsb galaxies are hosted in dark halos with values of the spin parameter , @xmath7 , of the dark halo ( @xmath8 , where @xmath9 is the angular momentum , @xmath10 is the total energy and @xmath11 is the total mass of the halo ) larger than those of hsb galaxies ( @xcite ) .
jimenez et al . ( 1998 ) used a detailed chemo - spectro - photometric disk model to show ( abandoning the assumption of a constant @xmath12 ratio ) that not only surface brightness , but also color , color gradients and metallicity of lsb disks can be explained if the spin parameter is larger for lsb than for hsb galaxies . in order to investigate the properties of both hsb and lsb disks , and to determine whether their characteristics are akin to those identified in dly@xmath0systems , we have computed models of galactic disks with @xmath13 ( hsbs ) and @xmath14 ( lsbs ; see @xcite ) .
the model is described in detail in jimenez et al .
( 1997 ) and we only summarize it here .
we assume that the specific angular momentum of baryonic matter is the same as the dark matter and that gas settles into a given dark halo until centrifugally supported .
we also assume that the dark matter profile is that found in numerical simulations ( @xcite ) , although using the isothermal sphere profile does not significantly change the computed initial surface density for the settling disk .
once the initial surface density of the disks has been computed , we used the schmidt law ( @xcite ) to compute the star formation rate . in particular , the sfr law adopted is @xmath15 , with @xmath16 and @xmath17 , and where @xmath18 is the gas surface density and @xmath19 is the total mass surface density .
the hsb disks have a initial burst of star formation which then declines with time , while the lsb galaxies have constant , but less active , star - formation . the primordial gas infall rate in the disk
is assumed to be the same function of total surface density as used in models of the milky way ( see @xcite ) .
the infall rate is higher in the center than in the outermost regions of the disk , with the infall law expressed as @xmath20 , where @xmath21 is the time - scale for the formation of the disk at a radius @xmath22
. the values of @xmath21 are chosen to fit the present time radial distribution of the gas surface density in the disk . in analogy with what is required for the disk of the milky way
, we assumed @xmath21 is increasing towards larger radii ( see table 2 of @xcite ) .
@xmath23 is the abundance of element @xmath24 in the infalling gas and the chemical composition is assumed to be primordial .
the parameter @xmath25 is obtained by requiring the surface density now to be @xmath26 , so that @xmath27 . the evolution of several chemical species ( h , d , he , c , n , o , ne , mg , si , s , ca , fe and zn ) , as well as the global metallicity @xmath28 , is followed by taking into account detailed nucleosynthesis prescriptions ( jimenez et al .
the imf is taken from scalo ( 1986 ) .
for simplicity , we have adopted an einstein - desitter universe ( @xmath29 , @xmath30 , @xmath31 km s@xmath32 mpc@xmath32 ) .
non - zero values of @xmath33 actually strengthen the results discussed below .
the initial goal of our modelling was to explore any differences in gas metallicity between hsb and lsb disks , to determine whether the measurements of [ zn / h ] now available for dly@xmath0 systems could discriminate between the two types of galaxies . in figure 1
we plot the values of [ zn / h ] found by pettini et al .
( 1997 and refs . therein ; 1998 ) against the redshift of the absorption systems .
our models enable us to plot the metallicity of gas disks as a function of redshift for both a given galactic radius , @xmath22 , and for a formation redshift .
hence , in figure 1_a _ , we plot the evolution in the metallicity of an lsb disk , forming at @xmath34 , at @xmath35 , 1.0 , 2.5 , and @xmath36 kpc .
we take @xmath36 kpc to be the outer limit at which an lsb would still give rise to a dly@xmath0 line ; our models predict @xmath37(h i)@xmath38 @xmath39 at this radius for an lsb baryonic mass of @xmath40 .
this agrees well with observed h i surface densities toward nearby lsb galaxies at similar radii ( de blok , mcgaugh , & van der hulst 1996 ) . in figure
1_b _ , we show the same variation in metallicity for @xmath41 6 kpc but with the galaxy forming at @xmath423.2 , 2.2 and 1.4 .
figure 1_a _ clearly shows that the predicted metallicities of lsb galaxies fit the observed data very well , particularly at outer radii , where the absorption cross - section will be largest and hence where the majority of qso sightlines will intercept .
figure 1_b _ demonstrates , however , that such lsb disks must form at @xmath43 , and evolve slowly for such a model to be consistent with the data .
in contrast , figure 1_c _ shows the variation of metallicity with redshift for an hsb galaxy , again forming at @xmath44 , for @xmath450 , 4.0 , 6.0 and @xmath46 kpc .
the figure shows that hsb galaxies have enriched their isms much faster than lsb galaxies , leading to gas metallicities higher than observed .
however , this does not rule out hsb disks as potential absorbers .
figure 1_d _ shows that if hsb galaxy halos form over the range of redshifts measured for dly@xmath0 systems , as is indeed the case in cdm cosmogonies ( e.g. , @xcite ) , then it is possible to reproduce the [ zn / h ] measurements . in this case , most of the dly@xmath0 systems would be proto - galaxies in the very early stages of their evolution it takes only 0.3 gyr for hsb disks to reach [ zn / h ] @xmath47 @xmath48 , close to the average measured value .
we note that in our models , however , these proto - galaxies have not yet formed rotating disks , hence we would not expect the velocity profiles of metals absorption lines in dly@xmath0 systems to be indicative of rotating disks .
the fact that the profiles of metal lines _ do _ appear to support rotating thick - disk models ( @xcite ) tends to favor lsb disks .
we note , however , that it remains possible that the initial merging of galaxy halos in a cdm - like scenario could still give rise to the observed line profiles ( haehnelt , steinmetz , & rauch 1998 ) .
one possible way to distinguish between slowly evolving lsb disks and a continuously forming population of hsb galaxies is to consider the evolution of the dust content of each galaxy type .
figure 2 shows the dust formation rate for hsb and lsb disks at different radii ( each radius normalized to @xmath49 m@xmath50 ) .
the amount of dust produced by each generation of stars was accounted for by using the model of draine & lee ( 1984 ) . as expected , hsb disks form about twice as much dust as their lsb counterparts due to their higher metallicity . more importantly , however , hsb galaxies produce most of their dust during the first gyr , reaching a maximum at an age of 0.3 gyr . on the other hand ,
lsb galaxies produce virtually no dust at radii larger than @xmath51 kpc .
therefore , a population of continually forming proto - disks should quickly begin to obscure background qsos , removing hsb disks from dly@xmath0 samples , as suggested by other researchers ( @xcite ) .
lsb galaxies , of course , remain largely unobscured , and would show up readily against background qsos . if , in reality , it is a mixture of lsb and hsb galaxies which actually give rise to dly@xmath0 systems , such an effect would at least explain why lsb galaxies have so readily been found responsible for low redshift dly@xmath0 absorbers .
our model predicts one more difference between the observable properties of dly@xmath0 systems depending on whether hsb or lsb disks are responsible .
dly@xmath0systems have been used to infer the evolution in the cosmological mass density of neutral gas in the universe , @xmath52 , from @xmath53 . in figure 3
we reproduce the values of @xmath52 as a function of redshift ( @xcite ) .
we also show how h i gas is consumed by stars in our disk models , both for hsb ( dashed line ) and lsb ( continuous line ) galaxies .
the curves are normalized such that that @xmath52 for lsb galaxies is zero at @xmath54 .
the true value of @xmath52 for lsb galaxies clearly lies somewhere between 0 ( an extreme limit which would assume that lsb disks have used up all their gas at odds with the observations ) and that observed for galaxies today , a value dominated by the contribution from spirals and irregular galaxies ( rao & briggs 1993 ) . as can be seen in figure 3
, however , this range is extremely small , and it matters little whether we set @xmath52 for lsb disks to zero or the value observed for all galaxy types .
surprisingly , the hsb galaxies do not reproduce the data particularly well the amount of gas needed at high - redshift to reproduce the neutral gas density measured in the local universe is much higher than observed .
on the other hand , lsb disks fit the data much better , since they have less efficient star formation and thus transform less h i into stars .
it would seem , therefore , that dly@xmath0 systems are very _ inefficient _ star producers .
the figure is also important in demonstrating that there need not be many lsb galaxies in the nearby universe to account for the number of dly@xmath0 systems found at high redshift . although plausible arguments have been made that the size and number density of lsb galaxies in the local universe are sufficient to explain the frequency of dly@xmath0 systems at high redshift ( @xcite ) , our models show that even if _ no _ lsb disks remained today , their gradual slow consumption of h i over time best fits the observed values of @xmath52 , and that hsb disks simply convert h i into stars too efficiently to account for those measurements .
since the star formation histories in lsb and hsb galaxies are so different , our models make it possible to predict differences in the @xmath55nucleid to iron - peak element ratios for the two types of galaxies . in particular , we would expect that at a fixed epoch hsb disks have [ @xmath0/fe ] ratios around zero or less and lower than [ @xmath0/fe ] ratios in lsb disks .
this because at the same epoch lsb disks have lower fe abundances than @xmath0-element abundances relative to hsb disks .
this difference is mostly due to the fact that @xmath0-elements are produced on very short timescales by type ii sne whereas fe is produced on long timescales ( from several tenths of million years to several gyr ) by type ia sne .
therefore , the different star formation history affects iron more than @xmath0-elements .
this is contrary to what happens if one looks at the [ @xmath0/fe ] vs. [ fe / h ] relation at a fixed [ fe / h ] instead of at a fixed epoch . in this case , galaxies with lower star formation rate shows lower [ @xmath0/fe ] ratios ( @xcite ) than systems with higher star formation rates .
this effect is clearly seen in figure 4 where we plot the redshift evolution of [ @xmath0/zn ] for lsb ( continuous line ) and hsb ( dashed line ) galaxies formed at @xmath56 and 4.0 , which encompasses a reasonable range in redshift for disk formation .
as expected , the lsb disks have super - solar [ @xmath0/zn ] ratios during most of their evolution .
the hsb galaxies exhibit super - solar ratios only during their first gyr , but have mainly sub - solar ratios for the rest of their lives .
it can be seen that any measured [ @xmath0/zn ] ratio at a given redshift is unique in determining whether the absorption originates in an lsb or hsb disk although we only plot the curves after the first 0.4 gyr of the disks formation , before which the [ @xmath0/zn ] ratio move vertically to higher values .
unfortunately , only two [ @xmath0/zn ] ratios have been measured in dly@xmath0 systems ( for @xmath57 systems ) , both of which are from observations of s ii lines ( see , e.g. , @xcite and refs . therein ) , plotted as diamonds in figure 4 .
both measurements have large errors , and it is clear that many , more accurate observations are needed in the future to determine the nature of the dly@xmath0 population in the universe . 99 de blok , w. j. g. , mcgaugh , s. s. , van der hulst , j. m. 1996 , mnras , 283 , 18 p. , le brun v. , bergeron j. , deharveng j .- m . 1998 , , 333 , 841 centurion m. , bonifacio p. , molaro p. , vladilo g. 1998 , j. j. , spergel d. n. , summers f. j. 1997 , , 482 , 659 b. t. , lee h. m. 1984 , , 285 , 89 s. m. , efstathiou g. 1980 , , 193 , 189 s. m. , pei y. c. 1993 , , 402 , 479 haehnelt , m. g. , steinmetz , m. , & rauch , m. , 1998 , apj , 495 , 647 c. , bothun g. 1997 , , 35 , 267 jimenez r. , heavens a. , hawkins m. , padoan p. 1997
, , 292 , l5 r. , padoan p. , matteucci f. , heavens a. f. 1998 , , 299 , 123 a. 1982 , , 200 , 585 , robert c. j. 1998 , , 498 , 541 k. m. , wolfe a. m. , altan h. , barcons x. , chen h .- w .
, fernandez - soto a. , meyer d. m. , ortiz - gil a. , savaglio s. , webb j. k. , yahata n. 1997 .
, 114 , 1337 v. , bergeron j. , boisse p. , deharveng j. m. 1997 , , 321 , 733 f. , 1991.in : _ stellar evolution and galactic chemical evolution , sp conf .
ser . 20 : frontiers of stellar evolution _ , pp 539 .
h. j. , mao s. , white s. d. m. 1997 , , 295 , 319 p. , centurion m. , vladilo g. 1998 , , 293 , l37 j. f. , frenk c. s. , white s. d. m. 1997 , , 490 , 493 j. a. , heavens a. f. 1985 , , 217 , 805 pettini m. , ellison s. , steidel c. , bowen d. 1998 , j. x. , wolfe a. m. 1997 , , 487 , 73 s. , briggs f. 1993 , , 419 , 515 rao s. , turnshek d. 1998 , k. c. , blades j. c. 1995 , , 445 , l95 j. m. 1986 , 11 , 1 c. c. , pettini m. , dickinson m. , persson s. e. 1994 , , 108 , 2046 l. j. , mcmahon r. g. , irwin m. j. 1996 , , 283 , l79
d. a. , 1997.in : _ structure & evolution of the igm from qso absorption line systems _ , 263 , eds petitjean p. , charlot s. , france : edit .
a. m. , prochaska j. x. 1998 , , 494 , l15 * figure 1 : * redshift evolution of [ zn / h ] for hsb ( bottom panels ) and lsb ( upper panels ) galaxies , as measured at different radii in a disk .
_ a ) _ the evolution in metallicity of lsb galaxies formed at @xmath58 well fits the values measured by pettini et al .
( 1997 ; 1998 ) .
also shown as a dashed box at @xmath59 is the range in metallicity of our own galaxy ( @xcite ) ; _ b ) _ lsb disks that formed late , however , do not fit the data . _
c ) _ conversely , hsb disks that formed at @xmath58 become metal - rich too quickly to explain the observed values ; _ d ) _ only if there is a continuously forming population of hsb disks between @xmath60 can they account for the measured metallicities .
* figure 2 : * the dust formation rates for lsb and hsb galaxies , for several galactic radii . the dust production in lsb disks at @xmath61 kpc is negligible , while hsb disks produce a significant amount of dust at all radii in the early stages of their evolution .
such dust production may well obscure background qsos and remove evolved hsb galaxies from dly@xmath0 samples . * figure 3 : * the evolution of neutral h i with redshift ( @xcite ) derived from dly@xmath0 systems ( squares ) . also plotted are the predictions from our model for hsb ( dashed line ) and lsb ( continuous line ) galaxies , both normalized to the present day neutral h i abundances ( @xcite ) .
both curves have been computed using an average of the radii larger than 6 kpc ( hsb disks ) and 2.5 kpc ( lsb disks ) .
we find that hsb galaxies require the existence of much more neutral gas at high redshift than is observed , because of their high star formation rates .
lsb galaxies , however , provide an excellent fit to the data because their gas consumption is lower .
this supports the idea that lsb galaxies may be responsible for most of the observed dly@xmath0 lines even at high redshift . *
figure 4 : * predictions from our models for [ @xmath0/zn ] ratios in dly@xmath0systems as a function of redshift .
the diamonds correspond to measurements of [ s / zn ] from the literature .
lsb galaxies have super - solar values of [ @xmath0/zn ] since their low star formation rate means fewer type ia snae produce smaller yields of zn , even though the initial burst of star formation produced the same amount of @xmath0 elements from type ii snae . | we use a model of galaxy disk formation to explore the metallicities , dust content , and neutral - gas mass density of damped lyman-@xmath0 ( dly@xmath0 ) absorbers .
we find that the [ zn / h ] abundance measurements of dly@xmath0 systems now available can be reproduced either by a population of low surface brightness ( lsb ) galaxies forming at redshifts @xmath1 , whose chemical contents evolve slowly with time and whose star formation rates are described by continuous bursts , or by high surface brightness ( hsb ) galaxies which form continuously over an interval of @xmath2 ( and no higher ) . although , in reality , a mixture of galaxy types may be responsible for low@xmath3 dly@xmath0systems , our models predict that hsb galaxies form more dust , more rapidly , than lsb galaxies , and that hsb galaxies may therefore obscure background qsos and not give rise to dly@xmath0 lines , as suggested by other researchers .
significantly , we find that the rate at which hsb disks consume neutral gas is too fast to explain the observed evolution in the neutral gas mass density with redshift , and that the consumption of hydrogen by lsb galaxies better fits the data .
this further suggests that lsb disks may dominate the dly@xmath0population at high - redshift . |
cographs have been defined independently by several authors since the 1970 s and are usually defined as @xmath4-free graphs , as proved in @xcite .
the original definition of cograph is based on a recursive construction described as follows : any single vertex graph is a cograph ; if @xmath5 is a cograph , so is its complement graph @xmath6 ; if @xmath5 and @xmath7 are cographs , so is their disjoint union . the _ disjoint union _ @xmath8 of graphs @xmath9 and @xmath10 is a graph operation such that @xmath11 and @xmath12 .
cographs can be defined alternatively in the following way .
the _ join _
@xmath13 of @xmath9 and @xmath10 is an operation such that @xmath11 and @xmath14 .
note that @xmath15 .
one can obtain a structural decomposition of a cograph by means of the operations _ join _ and _ disjoint union _ , as follows .
in @xcite a special tree , called _ cotree _ , is used to represent a cograph @xmath5 , in which leaf nodes are associated with the vertices of @xmath5 , and each internal node is labeled 0 ( _ type-0 node _ ) or 1 ( _ type-1 node _ ) indicating operations of join or disjoint union on their children , respectively .
furthermore , every cotree must be such that the nodes in a root - leaf path have alternate labels , which ensures that each cograph is associated with only one cotree ( up to permutation of siblings ) . on the other hand , each cotree refers to a single cograph .
as cographs can be identified by cotrees , given a rooted tree @xmath16 we use the following notation .
the root of @xmath16 is denoted by @xmath17 . given a node @xmath18 ( other than the root ) , we denote by @xmath19 the only path from @xmath18 to the root .
the nodes of @xmath19 ( except @xmath18 ) are named _
ancestors _ of @xmath18 in @xmath16 . the immediate ancestor of @xmath18 in @xmath19
is called the _
parent _ of @xmath18 and denoted by @xmath20 .
we set @xmath21 .
if @xmath18 is an internal node , its _ children _ are the elements of the set @xmath22 . if @xmath23 , its _ siblinghood _ is the set @xmath24 .
define also @xmath25 .
we denote by @xmath26 the subtree of @xmath16 rooted at @xmath18 containing all the nodes @xmath27 for which @xmath18 is an ancestor of @xmath27 . when there is no ambiguity we omit the index @xmath16 from the above definitions . if two graphs @xmath9 and @xmath10 are isomorphic we write @xmath28 .
in @xcite the authors prove that given two leaves @xmath18 and @xmath27 of a cotree , the closest ancestor to both ( that is common to @xmath19 and @xmath29 ) is type-1 if and only if @xmath18 and @xmath27 are adjacent in the associated cograph .
hence a cograph is connected if and only if the root of the cotree is type-1 . in the context of enumerative combinatorics , the work @xcite gives an asymptotic approximation for the number of cographs with a given number of vertices .
another approach consists of the generation of _ all _ graphs with a given property ( e.g. , `` cographs with @xmath0 vertices '' ) .
observe that this kind of approach provides , as a by - product , the exact number of graphs satisfying the given property .
formally , a _ generation _ ( or _ enumeration _ ) algorithm @xmath30 generates the sequence @xmath31 of all graphs satisfying a required property @xmath32 , where @xmath33 for @xmath34 .
there are some alternative definitions of efficiency for generation algorithms @xcite , the weakest one being _ polynomial total time _ , i.e. , the total running time is polynomial in the combined size of @xmath35 .
the strongest definition is _ polynomial delay _ , which means that the time between the generation of two consecutive elements is polynomial only in the size of the next output element
. some important examples of enumeration algorithms with polynomial delay are the generation of all minimum spanning trees @xcite , all maximal independent sets @xcite , and all cycles of a graph @xcite . in @xcite ,
the author describe sufficient conditions that guarantee linear delay generation of certain structures of a graph @xmath5 .
more precisely , the author proves the following theorem : if @xmath36 is a family of graphs of bounded tree - width then for every monadic second - order formula @xmath37 , where @xmath38 are the free set variables in @xmath39 , there is an algorithm that takes as input an @xmath0-vertex graph @xmath5 in @xmath36 and generates with linear delay all the @xmath40-tuples of sets that satisfy @xmath39 in @xmath5 , after an @xmath41 preprocessing step .
the formula @xmath39 can be interpreted as a query on @xmath5 with parameters @xmath38 .
for instance , the result of the query @xmath42 , where @xmath43 , @xmath44 , and @xmath45 , is the collection of all pairs @xmath46 such that @xmath47 and @xmath48 are disjoint independent sets of @xmath5 .
( in other words , the result of the query can be viewed as the collection of all induced bipartite subgraphs of @xmath5 . )
motivated by the above discussion , in this work we develop an algorithm that generates all ( mutually nonisomorphic ) unlabeled cographs with @xmath0 vertices .
say that two cotrees are _ isomorphic _ if they are associated with the same cograph . since a cotree is a compact representation of the associated cograph size representation of a cograph , because adjacency relations are implicit from the structure of cotrees .
] , our algorithm in fact generates all mutually non - isomorphic cotrees with @xmath0 leaves , for a given positive integer @xmath0 .
the delay of our algorithm ( time spent between two consecutive outputs ) is @xmath1 .
the time needed to generate the first output is also @xmath1 , which gives an overall @xmath2 time complexity , where @xmath3 is the number of unlabeled cographs with @xmath0 vertices .
the algorithm avoids the generation of duplicates ( isomorphic outputs ) and produces , as a by - product , a linear ordering of unlabeled cographs with @xmath0 vertices . to the best of the authors knowledge ,
this is the first explicit description of a linear - delay cograph generation algorithm .
since each output element is associated with a cograph with exactly @xmath0 vertices , the @xmath1 delay attained by our algorithm is best possible in practice .
the remainder of this work is organized as follows . in sections 2 and 3 ,
we define and study special orderings of nodes and trees using concepts of integer partitions . in section 4
, we describe our cograph generation algorithm and discuss its complexity and correctness .
section 5 contains our conclusions .
as discussed above , any path in a cotree alternates the types of its nodes ; thus they are only determined by the type of the root .
let @xmath49 be the set of rooted trees where each internal node has at least two children .
then each element of @xmath50 refers to exactly two distinct cographs : one with a type-1 root and another with a type-0 root , unless the tree consists only of the root ( and , in this case , represents the trivial graph ) . on the other hand , each cograph
is associated with a single tree of @xmath50 , up to permutation of nodes in the same siblinghood . in the next paragraphs
we introduce a standard way to configure such a tree . for @xmath51 and each node @xmath18 of @xmath16
we define @xmath52 as the number of leaves of @xmath26 . if @xmath18 is a leaf , @xmath53 .
a tree is said _ labeled _ if each node @xmath18 in @xmath16 is labeled @xmath52 .
figure [ fig : ordered - tree ] shows an example of labeled tree .
[ fact : sum - of - children ] if @xmath18 is an internal node then @xmath54 .
let @xmath55 , and consider the partition @xmath56 .
notice that each cograph with @xmath0 vertices can be represented by a tree in @xmath57 .
the fact below is straightforward and will be useful to calculate the complexities of the algorithms .
[ fact : number - of - internal - nodes ] for each @xmath0 the number of nodes of a tree in @xmath57 is at most @xmath58 .
facts [ fact : sum - of - children ] and [ fact : number - of - internal - nodes ] guarantee that labeling a tree can be done in @xmath1 . the notion of _ integer partition _ was introduced by euler . in this work
we assume that a partition must have at least two components , as described below .
[ def : particao ] let @xmath59 be a positive integer . the _ integer partition _ of @xmath0 is a non - decreasing sequence of positive integers @xmath60 such that @xmath61 and @xmath62 .
the set of all partitions of @xmath0 is denoted by @xmath63 .
we will use a _ lexicographic ordering _ of partitions .
[ def : ordenacaoparticao ] given distinct elements @xmath64 , where @xmath65 and @xmath66 , let @xmath67 , where @xmath68 . if @xmath69 then @xmath70 , otherwise @xmath71 . as definition [ def : ordenacaoparticao ]
is based on comparison of integers , the law of trichotomy applies to partitions : [ fact : trichotomy - partitions ] if @xmath72 then exactly one of the following holds : @xmath70 , @xmath73 , or @xmath71 .
the elements of @xmath74 listed in increasing order are : @xmath75 [ fact : min - and - max - partitions ] the minimum element of @xmath63 is @xmath76 , and the maximum is @xmath77 .
it is useful for our purposes to relate partitions of integers with distribution of siblinghoods in a tree .
let @xmath51 , and assume that for every internal node @xmath18 of @xmath16 its children @xmath78 are such that @xmath79 .
we say that @xmath80 is the _ partition induced by _
@xmath18 _ in _ @xmath16 .
[ def : nodes - inequality ] let @xmath81 , and let @xmath82 two nodes of @xmath16 .
define an order relation on the nodes of @xmath16 recursively as follows : * case 1 : if @xmath83 then @xmath84 ; * case 2 : if @xmath85 then @xmath86 ; * case 3 : if @xmath87 , @xmath88 * * let @xmath89 , and let @xmath90 be the partition induced by @xmath18 in @xmath16 .
analogously , let @xmath91 , and let @xmath92 be the partition induced by @xmath27 in @xmath16 . consider the following subcases : * * case 3.1 : if @xmath93 then @xmath84 ; * * case 3.2 : if @xmath94 @xmath95and , consequently , @xmath96 @xmath97 * * * case 3.2.a : if @xmath98 for all @xmath99 then @xmath86 ; * * * case 3.2.b : otherwise , let @xmath100 . if @xmath101 + then @xmath84 else @xmath102 .
if @xmath86 we will say that @xmath18 and @xmath27 are _ equivalent _ , which allows us to write the equivalence class @xmath103=\{u \in v(t)\mid u\equi v\}$ ] . also , we use the following notation : ( a ) @xmath104 if and only if @xmath86 or @xmath84 ; ( b ) @xmath105 if and only if @xmath106 . it is worth noting that definition [ def : nodes - inequality ] can be easily extended to nodes @xmath18 and @xmath27 in different trees @xmath107 and @xmath108 ( simply consider an auxiliary tree with a new root @xmath109 with subtrees @xmath110 and @xmath111 ) .
from fact [ fact : trichotomy - partitions ] , we have : _ ( node trichotomy ) [ fact : trichotomy - nodes ] if @xmath82 are nodes of a tree @xmath51 then exactly one of the following holds : @xmath84 , @xmath86 , or @xmath105 . _ throughout the text , if we use the symbol of equality between nodes we are referring to the same node . for two trees
@xmath112 , write @xmath113 if they can be converted into identical trees by sibling permutations . definition [ def : nodes - inequality ] gives the intuitive idea that two nodes @xmath18 and @xmath27 are equivalent when @xmath114 .
this is guaranteed by the following lemma : [ lem : equality - isomorphism ] if @xmath82 are nodes of @xmath115 then @xmath114 if and only if @xmath86 .
the proof is trivial when @xmath18 is a leaf .
suppose @xmath18 is an internal node of @xmath16 and let @xmath90 be the partition induced by @xmath18 in @xmath16 , where @xmath89 and @xmath116 for @xmath117 .
analogously , let @xmath92 be the partition induced by @xmath27 in @xmath16 , where @xmath91 .
suppose @xmath114 .
then @xmath118 and @xmath119 , for some integer @xmath120 .
we use induction to prove that @xmath86 holds for every @xmath120 .
the base case @xmath121 easily follows , since @xmath122 .
now assume that @xmath86 holds for every @xmath123 .
since @xmath114 , we can arrange siblings so that @xmath124 and @xmath125 for @xmath126 .
thus , by the induction hypothesis , @xmath98 holds , which implies @xmath86 by case 3.2.a in definition [ def : nodes - inequality ] .
conversely , suppose that @xmath86 .
then @xmath127 for some @xmath40 .
we use induction on @xmath40 again to prove that @xmath114 .
the result is valid for the base case @xmath128 .
suppose now that it is valid for every @xmath123 .
using node trichotomy , we have that @xmath98 for @xmath129 , with @xmath125 .
thus , by the induction hypothesis , @xmath124 , i.e. , @xmath114 .
[ def : ordered - tree ] let @xmath51 , and let @xmath18 be a node of @xmath16 .
the siblinghood @xmath130 is said to be _ ordered _ if @xmath131 .
if all the siblinghoods of a tree are ordered , we say that the tree is ordered . the tree in figure [ fig : ordered - tree ]
is ordered .
ordered trees establish a standard way to represent cotrees , as shown in the following proposition : [ propo : cographs - and - trees ] each cograph is associated with a single ordered tree in @xmath50 .
let @xmath5 be a cograph and let @xmath16 be its associated cotree .
in @xcite the authors prove that @xmath16 is unique up to isomorphism .
let @xmath132 and @xmath133 be ordered trees , both isomorphic to @xmath16 . as the nodes of @xmath132 and @xmath133 have an order of arrangement , they are the same tree , up to permutation of _ equivalent _ nodes of the same siblinghood . but lemma [ lem : equality - isomorphism ] ensures that this permutation generates identical subtrees with exactly the same arrangement of nodes
. therefore @xmath132 and @xmath133 are identical .
now we introduce a total ordering on trees of @xmath57 .
[ def : ordering - trees ] let @xmath134 .
define @xmath135 if @xmath136 , and @xmath137 if @xmath138 .
[ fact : minimum - tree ] the minimum element of @xmath57 is the tree @xmath139 whose root induces the partition @xmath140 ( see fact [ fact : min - and - max - partitions ] ) .
this tree is associated with the graphs @xmath141 and @xmath142 .
proposition [ propo : cographs - and - trees ] ensures that one can generate the unlabeled cographs with @xmath0 vertices by generating the elements of @xmath57 , whose first element is the minimum tree @xmath139 described in fact [ fact : minimum - tree ] .
based on the total order established in definition [ def : ordering - trees ] , a way to generate all the elements of @xmath57 is to generate the tree @xmath108 which is _ immediately next _ to @xmath139 , and then the tree @xmath143 which is immediately next to @xmath108 , and so on , until the generation of the maximum tree in @xmath57 . the notion of immediately next element in a set equipped with a total order is formalized in the definition below .
let @xmath144 be a non - empty , finite , and totally ordered set .
given @xmath145 , we say that @xmath146 is the element _ immediately next _ to @xmath147 in @xmath144 if @xmath148 , @xmath149 , and for any @xmath150 such that @xmath151 it holds that @xmath152 .
if there no element immediately next to @xmath147 in @xmath144 then @xmath147 is the maximum element of @xmath144 .
we now apply the above definition to the set @xmath63 . given a partition @xmath153 ,
algorithm [ alg : immediately - next - partition ] determines the partition immediately next to @xmath90 , if it exists , or decides that @xmath90 is the maximum partition in @xmath63 ( see fact [ fact : min - and - max - partitions ] ) .
@xmath154 @xmath92 immediately next to @xmath90 in @xmath63 , if it exists , or @xmath155 @xmath156 ; @xmath157 @xmath158 ; @xmath159 ; @xmath160 ; @xmath161 ; * if * @xmath162 * then * @xmath163 ; * else * @xmath164 ; @xmath165 special case * if * @xmath166 * then * ; * else * @xmath167 ; [ thm : next - partition - correctness ] if @xmath168 , algorithm [ alg : immediately - next - partition ] returns the partition immediately next to @xmath147 . in case
the algorithm returns , @xmath147 is the maximum element of @xmath63 .
let @xmath169 .
cases @xmath170 and @xmath165 are trivial . by definition of partition , necessarily @xmath171 .
suppose @xmath172 and @xmath173 .
consider the following cases : _ case 1 _ : @xmath174 . in this case , the algorithm returns the partition @xmath175 , where @xmath176 for @xmath177 and @xmath178 .
clearly , @xmath149 and @xmath179 .
we prove that @xmath146 is the partition immediately next to @xmath147 .
let @xmath180 such that @xmath151 .
our aim is to prove that @xmath152 . by definition
there is a minimum index @xmath181 such that @xmath182 . if @xmath183 then @xmath184 , and it follows that @xmath185 .
if @xmath186 then @xmath187 for @xmath188 and @xmath189
. thus , @xmath190 .
also , note that @xmath191 has @xmath192 terms , otherwise we would have @xmath193 .
this implies @xmath194 , which is a contradiction .
therefore , @xmath195 and this implies @xmath196 , that is , @xmath197 . finally , if @xmath198 then @xmath194 , which is a contradiction .
hence , @xmath152 .
_ case 2 _ : @xmath199 .
in this case , lines 6 and 7 generate the partition @xmath200 , where @xmath176 for @xmath188 , @xmath201 , and @xmath202 .
notice that @xmath203 , and then @xmath204 . let @xmath205 and @xmath109 be the quotient and the rest , respectively , of the integer division of @xmath206 by @xmath207 ( line 8) .
we have two sub - cases to consider : _ case 2a _ : @xmath162 . in this case
, the algorithm returns the partition @xmath208 note that @xmath209 has @xmath210 terms .
note also that : * \(a ) @xmath211 , if @xmath177 ; * \(b ) @xmath212 , if @xmath213 ; * \(c ) @xmath214 .
thus : @xmath215 therefore @xmath216 and @xmath217 .
we prove that @xmath209 is immediately next to @xmath147 in @xmath63 .
let @xmath218 such that @xmath151 , and let @xmath181 be the minimum index such that @xmath182 .
if @xmath183 then @xmath219 , i.e. , @xmath220 . if @xmath186 then @xmath189 , i.e. , @xmath221 .
thus , @xmath222 . finally , if @xmath198 then @xmath194 , which is a contradiction . thus @xmath222 .
_ case 2b _ : @xmath223 .
in this case the algorithm returns the partition @xmath224 . since @xmath225 , we have @xmath149 . again , we prove that @xmath146 is immediately next to @xmath147 in @xmath63 .
let @xmath218 such that @xmath151 , and let @xmath181 be the minimum index such that @xmath182 .
if @xmath183 then @xmath226 , i.e. , @xmath185 . if @xmath186 then @xmath189 , i.e. , @xmath227 .
thus , @xmath152 . finally , if @xmath198 then @xmath228 , a contradiction . thus @xmath152 . to conclude the proof , if @xmath172 and @xmath229 then the algorithm returns .
[ coro : next - partition - complexity ] let @xmath168 .
then @xmath230 runs in @xmath1 time .
the only case that does not have a constant - time complexity appears in line 9 , whose worst case occurs when the input is @xmath231 . in such a situation , @xmath232 and the number of operations performed is @xmath233 . in order to develop a procedure for determining the tree immediately next to the current tree being generated , we introduce the concept of _ pivot node _ , which indicates the place of the current tree where changes will be made .
let @xmath81 be an ordered tree .
a node @xmath18 of @xmath16 is said to be _ exhausted _ if it is a leaf or the partition induced by @xmath18 in @xmath16 is the maximum element of @xmath234 .
[ fact : exhausted - node ] given an ordered and labeled tree @xmath235 , verifying whether a node @xmath18 of @xmath16 is exhausted can be done in constant time ( see fact [ fact : min - and - max - partitions ] ) . for an ordered tree @xmath236
we define the _ inverted post - order traversal _ of @xmath16 by the following recursive procedure : if @xmath237 , traverse ( in the given sequence ) the subtrees @xmath238 in inverted post - order , and then visit @xmath17 .
the inverted post - order traversal can be done in @xmath1 time , by fact [ fact : number - of - internal - nodes ] .
[ def : tree - pivot ] let @xmath51 be an ordered tree such that the inverted post - order traversal of @xmath16 visits its nodes in the sequence @xmath239 , where @xmath240 .
the _ pivot _ of @xmath16 , if it exists , is the node @xmath241 with minimum index such that @xmath241 is not exhausted .
the square node in figure [ fig : ordered - tree ] is the pivot of the tree .
given an ordered and labeled tree @xmath235 , the procedure for determining the pivot of @xmath16 can be simply done using the definition . in other words , simply perform an inverted post - order traversal and check whether the current visited node is exhausted or not .
we call this procedure @xmath242 . if there is no pivot in the tree , the procedure returns .
[ fact : pivot - complexity ] the search for the pivot node in a tree @xmath115 can be done in @xmath1 time .
another step towards our cograph generation algorithm is to describe a procedure that replaces a subtree @xmath26 of a given ordered @xmath115 by another subtree @xmath132 rooted at @xmath18 with the same number of leaves as @xmath26 .
the procedure receives as input a node @xmath18 of @xmath16 and a partition @xmath243 , and replaces @xmath26 by a subtree @xmath132 such that @xmath244 , @xmath245 , and , for @xmath246 , @xmath116 and @xmath247 consists of @xmath248 leaves ( if @xmath249 ) .
the procedure returns the new tree @xmath250 obtained from @xmath16 in this way .
node @xmath18 of an ordered tree @xmath115 ; partition @xmath243 new tree @xmath250 obtained from @xmath16 where @xmath26 is replaced by a subtree @xmath132 such that @xmath90 is the partition induced by @xmath18 in @xmath132 replace the children of @xmath18 by @xmath78 such that @xmath90 becomes the partition induced by @xmath18 * if * @xmath249 * then * insert @xmath248 leaves as children of @xmath241 @xmath16 by fact [ fact : sum - of - children ] , we have : [ fact : rebuild - complexity ] let @xmath243 .
then procedure @xmath251 runs in @xmath252 time .
[ lem : maximum - tree - no - pivot ] a tree @xmath115 is the maximum element of @xmath57 if and only if @xmath16 contains no pivot node .
suppose @xmath16 contains a pivot node @xmath18 such that the partition induced by @xmath18 in @xmath16 is @xmath253 .
thus there is @xmath254 such that @xmath149 .
let @xmath250 be the tree returned by @xmath255 .
clearly , @xmath256 and therefore @xmath16 is not the maximum element of @xmath57 .
the other direction is trivial .
now we are able to describe the procedure that generates the tree immediately next to a given tree @xmath16 , which is crucial to our cograph generation algorithm .
we use the following notation : let @xmath257 be the ordered siblinghood of a node @xmath18 .
assume @xmath258 for some @xmath259 and define the sets @xmath260 and @xmath261 .
note that @xmath262 is a partition of @xmath263 . ordered and labeled tree @xmath235 @xmath264 such that @xmath132 is the tree immediately next to @xmath16 @xmath265 @xmath266 , where @xmath90 is the partition induced by @xmath18 @xmath267 @xmath268 * if * @xmath269 * then * copy subtree @xmath270 in @xmath271 * else * @xmath272 , where @xmath273 @xmath274 @xmath16 there is no immediately next tree algorithm [ alg : immediately - next - tree ] finds the tree pivot @xmath18 and the partition that is immediately next to the one induced by @xmath18 .
next , it `` restarts '' each node @xmath27 visited during @xmath242 to the lowest possible configuration for @xmath27 .
this idea is based on the fact that comparison between trees is done from left to right in each siblinghood , while the searching for the pivot occurs from right to left .
figure [ fig : next - tree ] depicts the tree immediately next to the tree depicted in figure [ fig : ordered - tree ] . before checking the correctness of algorithm [ alg :
immediately - next - tree ] , we prove the following lemma .
[ lem : next - tree - keeps - ordering ] let @xmath115 be an ordered tree that is not the maximum element of @xmath57 .
then the tree returned by @xmath275 is ordered .
let @xmath132 be the tree returned by @xmath275 .
note that @xmath264 because its number of leaves is the same as in @xmath16 .
let @xmath18 denote the pivot of @xmath16 ( whose existence is guaranteed by lemma [ lem : maximum - tree - no - pivot ] ) .
denote its corresponding node in @xmath132 by @xmath276 ( line 5 ) . in general
, for each @xmath277 in @xmath16 , denote by @xmath278 its corresponding node in @xmath132 , if it exists ( lines 9 and 10 ) . for each @xmath278 ancestor of @xmath276 , we prove that the siblinghoods @xmath279 and @xmath280 are ordered .
since other siblinghoods did not change , their orderings follow directly from the ordering of @xmath16 .
moreover , by the same reason , the set @xmath281 is already ordered for each @xmath278 .
then it remains to analyze the orderings of @xmath282 and @xmath283 .
in fact , for every @xmath284 the following cases are valid : 1 . if @xmath285 and @xmath286 then the operation in line 9 and lemma [ lem : equality - isomorphism ] guarantee @xmath287 .
2 . if @xmath288 and @xmath289 then the partition induced by @xmath290 is minimum in @xmath291 ( line 10 ) ; therefore , @xmath292 .
thus in both cases the ordering of the nodes is maintained , i.e. @xmath293 is ordered .
it remains to analyze the ordering of @xmath280 .
write @xmath294 , where @xmath295 .
similarly , write @xmath296 , where @xmath297 . as the algorithm changes only the nodes of @xmath298 , the ordering of @xmath16 guarantees that @xmath299 is ordered .
* claim : * @xmath276 is correctly positioned in @xmath280 .
to prove the claim we need to show that @xmath300 . by theorem
[ thm : next - partition - correctness ] and case @xmath301 of definition [ def : nodes - inequality ] we have @xmath302 .
but by the ordering of @xmath16 , @xmath303 is valid , and thus @xmath304 . on the other hand , if @xmath305 then by line 9 we have @xmath306 , otherwise @xmath307 and from line 10 and definition [ def : nodes - inequality ] we have @xmath308 .
hence , the claim follows .
following the same analysis used for @xmath309 , we conclude that @xmath310 is also ordered , because both are handled by the operation in lines 9 and 10 .
this fact and the above claim guarantee that @xmath280 is ordered .
hence , the lemma follows .
[ thm : next - tree - correctness ] let @xmath115 .
if @xmath311 then @xmath16 is the maximum element in @xmath57 ; otherwise , @xmath275 returns the tree immediately next to @xmath16 in @xmath57 .
if the procedure returns , the result is guaranteed by lemma [ lem : maximum - tree - no - pivot ] .
otherwise , let @xmath132 be the output of the algorithm . since @xmath132 is ordered , we use lemma [ lem : next - tree - keeps - ordering ] to compare @xmath16 and @xmath132 .
let @xmath18 be the pivot of @xmath16 , and @xmath276 the node corresponding to @xmath18 in @xmath132 .
as in the previous lemma , for each @xmath277 in @xmath16 we denote by @xmath278 its corresponding node in @xmath132 .
* claim * : @xmath312 . to prove the claim , first note that theorem [ thm : next - partition - correctness ] and definition [ def : nodes - inequality ] guarantee @xmath313 .
if @xmath314 then it is clear that @xmath312 . otherwise we write @xmath315 and @xmath316 , where @xmath317 and @xmath318 . since the algorithm only makes changes to @xmath319 , we have @xmath320 for @xmath321 .
hence , by case 3.2.b in definition [ def : nodes - inequality ] , @xmath322 .
let @xmath323 .
write @xmath324 and @xmath325 , where @xmath326 and @xmath327 .
similarly , we have @xmath328 for @xmath329 .
then , since @xmath330 , by case 3.2.b in definition [ def : nodes - inequality ] it follows that @xmath331 . in general ,
the same argument can be successively applied to each ancestor @xmath277 of @xmath18 to conclude that @xmath330 .
if @xmath332 then @xmath312 , and the claim follows . now prove that @xmath132 is immediately next to @xmath16 in @xmath57 .
initially , the algorithm takes the pivot @xmath18 of @xmath16 and builds @xmath276 , whose induced partition by line 5 is immediately next to the partition induced by @xmath18 .
let @xmath277 be such that @xmath333 or @xmath277 is an ancestor of @xmath18 .
then , for each @xmath334 , the algorithm builds @xmath335 as follows : if @xmath336 then @xmath290 is equivalent to @xmath278 ; if @xmath337 then @xmath338 is exhausted from the definition of pivot .
therefore , @xmath338 is built so that it is the minimum node , as detailed below and similarly to the proof of the previous lemma : 1 . if @xmath339 and @xmath286 then @xmath287 ( line 9 ) .
2 . if @xmath288 and @xmath289 then @xmath290 is built from the minimum partition @xmath291 ; therefore , @xmath292 ( line 10 ) . since @xmath340 and @xmath341
, it follows that @xmath331 . by the cases above , for each @xmath342 we have that @xmath343
is exhausted in @xmath16 .
moreover , @xmath344 is built so that it is the minimum node with @xmath345 leaves .
hence , @xmath346 is immediately next to @xmath347 .
as this argument applies to every ancestor @xmath277 of @xmath18 and to @xmath18 itself , it follows that @xmath132 is immediately next to @xmath16 in @xmath57 .
[ coro : next - tree - complexity ] let @xmath115 .
then algorithm @xmath275 runs in @xmath1 time .
the worst case occurs when @xmath16 has a pivot other than @xmath17 .
corollary [ coro : next - partition - complexity ] together with facts [ fact : pivot - complexity ] and [ fact : rebuild - complexity ] ensure that all the operations outside the loop in lines 713 are done in @xmath1 time . the * for * loop in lines 811 performs the operations in lines 1011 , for each @xmath348 . assume that @xmath349 .
the worst case complexity of lines 811 occurs when @xmath350 and @xmath351 , and thus is given by @xmath352 ( recall fact [ fact : sum - of - children ] ) .
let @xmath353 be the path from the pivot to the root of the tree , where @xmath354 and @xmath355 .
the * repeat * command in lines 713 executes the internal * for * loop in lines 811 for each @xmath277 of @xmath19 .
then , using the idea developed in the previous paragraph , we conclude that : ( a ) for @xmath356 , the internal * for * loop in lines 811 is executed for every @xmath357 in @xmath358 total time , since @xmath359 ; ( b ) the runtime for @xmath360 , @xmath361 , is : @xmath362 where @xmath363 is the complexity of the @xmath364 previous iterations .
the process terminates when the last ancestor ( @xmath365 ) is reached .
therefore , the overall time complexity of @xmath275 is @xmath366 .
as discussed in the beginning of the previous section , the generation of the elements in @xmath57 can be done as follows : starting from the minimum tree @xmath139 in @xmath57 ( fact [ fact : minimum - tree ] ) , apply algorithm [ alg : immediately - next - tree ] to generate the tree immediately next to @xmath139 in @xmath57 , and successively repeat the application of the algorithm until reaching the maximum element of @xmath57 ( characterized by the absence of pivot ) . since the entire generation procedure starts with an ordered tree ( recall that @xmath139 is trivially ordered ) , lemma [ lem : next - tree - keeps - ordering ] ensures that subsequent trees are all ordered as well , i.e.
, there is no need of extra work to order trees along the generation .
in addition , by corollary [ coro : next - tree - complexity ] , each new generated tree is determined in @xmath1 time .
based on such arguments , we present below a formal description of our cograph generation algorithm .
let @xmath367 denote the family of all cographs , and consider the partition @xmath368 , where @xmath369 .
below we establish a total order on the members of @xmath370 .
first , note that any cotree @xmath16 with @xmath0 leaves can be viewed as a member ( ordered tree ) of @xmath57 , and therefore we can apply the total ordering in definition [ def : ordering - trees ] to cotrees .
[ def : cograph - ordering ] let @xmath371 , and let @xmath372 be their respective ordered cotrees .
say that @xmath373 if @xmath374 @xmath375 , or @xmath376 @xmath137 , @xmath377 is type-1 , and @xmath378 is type-0 . in addition , say that @xmath379 if @xmath137 and @xmath377 and @xmath378 are of the same type . from the above definition ,
it is easy to see that @xmath373 implies @xmath380 . in adition , @xmath379 if and only if @xmath28 . for @xmath115 , let @xmath381 ( resp . , @xmath382 ) be the cotree associated with @xmath16 by setting @xmath17 as a type-0 ( resp
. , type-1 ) node .
also , let @xmath383 and @xmath384 be the cographs associated with cotrees @xmath385 and @xmath386 , respectively .
integer @xmath387 all cographs with @xmath0 vertices @xmath388 @xmath389 output @xmath390
@xmath391 @xmath392 [ propo : corretude_geradorcografos ] let @xmath393 be an integer
. then : @xmath394 the sequence @xmath395 generated by algorithm [ alg : cograph - generation ] contains all the cotrees with @xmath0 leaves ; @xmath396 the associate sequence @xmath397 contains all the cographs with @xmath0 vertices , where no two graphs in @xmath398 are isomorphic ; @xmath399 the delay of algorithm [ alg : cograph - generation ] is @xmath1 ; @xmath400 the time spent by algorithm [ alg : cograph - generation ] to output the first element is @xmath1 .
\(a ) by theorem [ thm : next - tree - correctness ] , the tree @xmath401 is immediately next to @xmath402 in @xmath57 . since the algorithm starts with the minimum tree and stops with the maximum tree in @xmath57 , the sequence @xmath403 determined by the algorithm contains all the elements of @xmath57 .
therefore , the sequence @xmath395 generated by the algorithm contains all the cotrees with @xmath0 leaves .
\(b ) by item ( a ) , the associated sequence @xmath404 of cographs contains all the members of @xmath370 .
now , note that the @xmath402 s are in increasing order according to definition [ def : ordering - trees ] , and by lemma [ lem : equality - isomorphism ] are pairwise nonisomorphic .
hence , @xmath398 is in _ strictly _ increasing order according to definition [ def : cograph - ordering ] .
this implies that no two cographs in @xmath398 are isomorphic .
\(c ) by fact [ fact : number - of - internal - nodes ] , each @xmath402 contains @xmath1 nodes ; hence , obtaining @xmath385 and @xmath386 from @xmath402 can be done in @xmath1 time .
in addition , by corollary [ coro : next - tree - complexity ] , @xmath405 runs in @xmath1 time .
therefore , algorithm [ alg : cograph - generation ] has delay @xmath1 .
\(d ) it is easy to see that @xmath139 can be determined in @xmath1 time .
this implies that the time spent to output @xmath406 is @xmath1 .
[ coro ] algorithm [ alg : cograph - generation ] determines a linear order on @xmath370 .
figure [ fig : four - vertex - cographs ] depicts in increasing order all cographs with @xmath407 vertices generated by @xmath408 .
based on the fact that a cotree is a compact , @xmath1 size representation of an @xmath0-vertex cograph , in this work we described an algorithm for generating all unlabeled cographs with @xmath0 vertices , via the generation of cotrees .
the delay of our algorithm is @xmath1 and the time needed to generate the first output is also @xmath1 , yielding an overall @xmath2 time complexity , where @xmath409 . the algorithm avoids the generation of duplicates ( isomorphic outputs ) and produces , as a by - product , a total ordering of @xmath370 . to the best of the authors knowledge ,
this is the first practical description of a linear - delay cograph generation algorithm .
the algorithm was implemented in language @xmath410 and executed on an amd fx-6100 six - core processor at 3.30ghz with 8 gb of ram , running _
windows 7 _ operating system .
table [ tab : number - of - cographs ] shows the values @xmath411 calculated by the algorithm , for @xmath412 .
we remark that the results in table [ tab : number - of - cographs ] agree with the results presented in @xcite , for @xmath413 . in a future work
, we will describe an application of our generation algorithm in the search of counter - examples for spectral graph theory conjectures on cographs .
d. eppstein . representing all minimum spanning trees with applications to counting and generation .
tech . rep .
95 - 50 ( 1995 ) , univ . of california ,
irvine , ca 92717 , dept . of information and computer science .
available at http://www.ics.uci.edu/@xmath414eppstein/pubs/epp-tr-95-50.pdf .
b. kimelfeld and y. sagiv .
efficiently enumerating results of keyword search . proc . of the 10th international symposium dblp 2005 , trondheim , norway .
_ lecture notes in computer science _ 3774 ( 2005 ) 5873 . | cographs have always been a research target in areas such as coloring , graph decomposition , and spectral theory . in this work
, we present an algorithm to generate all unlabeled cographs with @xmath0 vertices , based on the generation of cotrees .
the delay of our algorithm ( time spent between two consecutive outputs ) is @xmath1 .
the time needed to generate the first output is also @xmath1 , which gives an overall @xmath2 time complexity , where @xmath3 is the number of unlabeled cographs with @xmath0 vertices .
the algorithm avoids the generation of duplicates ( isomorphic outputs ) and produces , as a by - product , a linear ordering of unlabeled cographs wih @xmath0 vertices .
cographs , enumarative combinatorics , enumerative algorithms |
vector meson leptoproduction at large energies @xcite is one of the important processes which is sensitive to the gpds . at small @xmath0-bjorken ( @xmath0 ) the process factorizes @xcite into a hard meson leptoproduction off gluons and gpd ( fig.1 ) .
the generally used standard collinear approximation ( sca ) leads to some essential problems .
the cross section for longitudinally polarized virtual photons which predominates for the large photon virtualities exceeds the data by a large factor if calculated in the sca @xcite .
only the amplitude with a longitudinally polarized virtual photon and vector meson ( ll ) amplitude can be calculated in the sca .
the transition amplitudes for transversally polarized photons @xmath1 ( tt ) and @xmath2 ( tl ) exhibit in the sca the infrared singularities @xcite which breakdown factorization .
here we use the modified perturbative approach ( mpa ) @xcite which includes the quark transverse degrees of freedom accompanied by sudakov suppressions . in the mpa
one can solve all mentioned problems .
transverse quark degrees of freedom and sudakov factors suppress the contribution from the end - point region in the ll amplitude .
the transverse quark momentum regularizes the end - point singularities in tt and tl amplitudes .
this leads to reasonable agreement of our results with h1 and zeus data @xcite for electroproduced @xmath3 and @xmath4 mesons at small @xmath0 @xcite . in this report
, we discuss the spin effects in the @xmath3 meson leptoproduction . within the mpa
we calculate the three @xmath5 , @xmath6 and @xmath7 transition amplitudes , and afterwards cross sections and sdme .
the leading - twist wave function @xmath8 describes the longitudinally polarized vector mesons . to study the transversally polarized light mesons
, we use the higher order @xmath9- dependent wave function @xmath10 proposed in @xcite @xmath11 @xmath12 @xmath13 \phi'_v(k,\tau){\nonumber}.\ ] ] here @xmath14 is a polarization of the vector meson with momentum @xmath15 , @xmath16 is the fraction of the meson momentum carried by the quark , @xmath17 is the scale in the higher order wave function which should be about the meson mass @xmath18 .
we use here @xmath19 the gluon contribution to the leptoproduction amplitudes for @xmath20 and positive proton helicity reads as a convolution of the hard subprocess amplitude @xmath21 and a large distance unpolarized @xmath22 and polarized @xmath23 gluon gpds ( fig.1 ) : @xmath24\ , h^g(\xb,\xi , t ) \right . { \nonumber}\\ & + & \hspace{2.5mm}\left .
\left[\,{\cal h}^v_{\mu'+,\mu + } \ , - { \cal h}^v_{\mu'-,\mu -}\,\right]\ , \widetilde{h}^g(\xb,\xi ,
t)\ , \right\}\,.\end{aligned}\ ] ] here @xmath25 ( @xmath26 ) denotes the helicity of the photon ( meson ) , @xmath27 is the momentum fraction of the transversally polarized gluons and the skewness @xmath28 is related to bjorken-@xmath0 by @xmath29 .
the flavor factors are @xmath30 and @xmath31 .
the polarized @xmath23 gpd at small @xmath27 is much smaller than the unpolarized gpd @xmath32 and will be important in the @xmath33 asymmetry only .
the subprocess amplitude @xmath21 is represented as the contraction of the hard part @xmath34 , which is calculated perturbatively , and the non - perturbative meson wave function @xmath35 @xmath36 the wave function is chosen in gaussian form @xmath37}\,.\ ] ] here @xmath38 , @xmath39 is the decay coupling constant and the @xmath40 parameter determines the value of average transverse momentum of the quark in the vector meson .
generally , values of @xmath41 are different for longitudinal and transverse polarization of the meson .
the subprocess amplitude is calculated within the mpa @xcite where we keep the @xmath42 terms in the denominators of the amplitudes and the numerator of the tt amplitude .
the gluonic corrections are treated in the form of the sudakov factors which additionally suppress the end - point integration regions .
the model leads to the following hierarchy of the amplitudes : @xmath43 note , that the @xmath44 and @xmath45 transitions are small and neglected in our analysis .
the gpd is a complicated function which depends on three variables .
for the small momentum transfer @xmath46 we use the double distribution form @xcite @xmath47\ , \frac{\beta}{\xi}\,f(\beta,\alpha=\frac{\xb-\beta}{\xi},t)\ ] ] with the simple factorizing ansatz for the double distributions @xmath48 @xmath49}{(1-|\beta|)^{3}}.\ ] ] in this model , we calculate gpd @xcite through the gluon distribution @xmath50 and take the cteq5 m results @xcite for it .
for the three @xmath28 values the gluon gpd is shown in fig .
the @xmath51- dependence of the amplitudes is important in analyses of experimental data . for simplicity
we parameterize it in the exponential form @xmath52 for @xmath53 transitions .
experimentally , only the slope of the @xmath54 cross section is measured whereas information about individual slopes @xmath55 is absent .
we suppose that @xmath56 .
estimations for the @xmath3 meson production amplitudes are obtained using @xmath57 , @xmath58 ; @xmath59 ; @xmath60 . [ cols="^,^ " , ] @xmath61 ^ 2 $ ] the @xmath62 dependence of sdme on the @xmath3 production at @xmath63 and @xmath64 .
data are taken from h1 and zeus . in fig.4
, we present our results for the three sdme at desy energy range . description of experimental data is reasonable .
our results for other sdme can be found in @xcite the @xmath33 asymmetry for a longitudinally polarized beam and target is sensitive to the polarized gpd .
really , the leading term in the @xmath33 asymmetry integrated over the azimuthal angle is determined through the interference between the unpolarized gpd @xmath32 and the polarized @xmath23 distributions .
we expect a small value for the @xmath33 asymmetry at high energies because it is of the order of the ratio @xmath65 which is small at low @xmath0 .
gluon contribution to the @xmath33 asymmetry of vector meson production at @xmath66 ( solid line ) and @xmath67 ( dashed line ) ; @xmath68 .
the shaded bands reflect the uncertainties in the @xmath33 due to the error in @xmath69 [ yourname_fig4 ] at compass energies @xmath70 the gluon contribution to the @xmath33 asymmetry of vector meson production is small ( fig.5 ) . at hermes energies @xmath71 the major contribution to
the amplitudes comes from the region @xmath72 where @xmath73 is not small , which leads to a large value of the @xmath33 .
the asymmetry is shown in fig.5 together with the uncertainties in our predictions due to the error in the polarized gluon distribution .
note that our predictions should be valid for the @xmath4 production . in the case of @xmath3 production
the polarized quark gpd should be essential in the @xmath33 asymmetry in the low energy range .
light vector meson electroproduction at small @xmath0 was analyzed here within the gpd approach .
the amplitudes were calculated using the mpa with the wave function dependent on the transverse quark momentum .
the transverse quark momentum regularizes the end - point singularities in the amplitudes with transversally polarized photons .
this gives a possibility to study spin effects in vector meson production . within the gpd approach we find a fine description of the @xmath62 dependence of the cross section , @xmath74 ratio and sdme for the @xmath3 meson production at hera energies .
more results for @xmath3 and @xmath4 production can be found in @xcite .
some comparison with the two - gluon exchange model @xcite can be found in @xcite too .
we would like to point out that study of sdme gives important information on different @xmath75 hard amplitudes . by analyses of @xmath51 dependencies of sdmes
the diffraction peak slopes of the @xmath6 and @xmath7 amplitudes can be defined .
unfortunately , the data on spin observables have large experimental errors .
the new experimental results with reduced errors in sdme are extremely important .
information about polarized gluon distribution can be obtained from the @xmath33 asymmetry on @xmath4 meson production at hermes and compass energies which is sensitive to @xmath76 .
goloskokov , p. kroll : euro .
j. c*42 * ( 2005 ) 281 .
x. ji : phys .
d*55 * ( 1997 ) 7114 ; + a.v .
radyushkin : phys .
b*380 * ( 1996 ) 417 ; + j.c .
collins : _ et al .
_ , phys .
rev . d*56 * ( 1997 ) 2982 .
l. mankiewicz , g. piller and t. weigl : eur .
j. c*5 * ( 1998 ) 119 . l. mankiewicz and g. piller : phys .
rev . d*61 * ( 2000 ) 074013 ; + i.v .
anikin and o.v .
teryaev : phys .
lett . b*554 * ( 2003 ) 51
. j. botts and g. sterman : nucl .
phys . b*325 * ( 1989 ) 62 .
c. adloff et al .
[ h1 collaboration ] : eur .
j. c*13 * ( 2000 ) 371 ; + s. aid et al .
[ h1 collaboration ] : nucl .
b * 468 * ( 1996 ) 3 .
j. breitweg et al .
[ zeus collaboration ] : eur .
j. c*6 * ( 1999 ) 603 ; + s. chekanov et al .
[ zeus collaboration ] : nucl.phys . b*718 * ( 2005 ) 3 .
goloskokov , p. kroll : proc . of 16th international spin physics symposium , trieste , 2004 , world scientific , 2005 , 421 .
j. bolz , j.g .
krner and p. kroll : z. phys .
a*350*(1994 ) 145 .
i. v. musatov and a. v. radyushkin : phys .
d*61 * ( 2000 ) 074027
. j. pumplin , d. r. stump , j. huston , h. l. lai , p. nadolsky and w. k. tung : jhep * 0207 * ( 2002 ) 012 .
s. j. brodsky et al . :
d * 50 * ( 1994 ) 3134 . | we study light vector meson photoproduction at small @xmath0 on the basis of the generalized parton distribution ( gpd ) .
our results on the cross section and spin density matrix elements ( sdme ) are in fair agreement with desy experiments . |
in this paper , we investigate numerically the behavior of granular material at the surface of an asteroid during close approach to the earth .
we focus on the specific case of asteroid ( 99942 ) apophis , which will come as close as @xmath3 earth radii on april 13th , 2029 .
we provide predictions about possible reshaping and spin - alteration of and surface effects on apophis during this passage , as a function of plausible properties of the constituent granular material .
studies of possible future space missions to apophis are underway , including one by the french space agency cnes calling for international partners ( e.g. , michel et al .
2012 ) , with the aim of observing this asteroid during the 2029 close encounter and characterizing whether reshaping , spin - alteration , and/or surface motion occur .
the numerical investigations presented here allow for estimation of the surface properties that could lead to any observed motion ( or absence of motion ) during the actual encounter .
apophis made a passage to the earth at @xmath4 earth radii in early 2013 . at that time
, the herschel space telescope was used to refine the determination of the asteroid s albedo and size ( mller et al .
according to these observations , the albedo is estimated to be about 0.23 and the longest dimension about @xmath5 m , which is somewhat larger than previous estimates ( @xmath6 m , according to delbo et al .
concurrent radar observations improved the astrometry of the asteroid , ruling out the possibility of a collision with the earth in 2036 to better than 1 part in @xmath7
. however , wlodarczyk et al . (
2013 ) presented a possible path of risk for 2068 .
this finding has put off any crisis by @xmath8 years and makes exploring apophis in 2029 to be more for scientific interest . to date , nothing is known about the asteroid s surface mechanical properties , and this is why its close passage in 2029 offers a great opportunity to visit it with a spacecraft , determine its surface properties , and , for the first time , observe potential modifications of the surface due to tidal effects . and as apophis approaches , it is likely that international interest in a possible mission will increase , since such close approaches of a large object are relatively rare .
the case for tidally induced resurfacing was made by binzel et al .
( 2010 ; also see demeo et al .
2013 ) and discussed by nesvorn et al .
( 2010 ) to explain the spectral properties of near - earth asteroids ( neas ) belonging to the q taxonomic type , which appear to have fresh ( unweathered ) surface colors .
dynamical studies of these objects found that those bodies had a greater tendency to come close to the earth , within the earth - moon distance , than bodies of other classes in the past @xmath9 kyr .
the authors speculated that tidal effects during these passages could be at the origin of surface material disturbance leading to the renewed exposure of unweathered material .
we leave a more general and detailed investigation of this issue for future work , but if this result is true for those asteroids , it may also be true for apophis , which will approach earth on friday , april 13 , 2029 no closer than about 29,500 km from the surface ( i.e. , @xmath10 earth radii , or @xmath3 earth radii from the center of the planet ; giorgini et al .
it is predicted to go over the mid - atlantic , appearing to the naked eye as a moderately bright point of light moving rapidly across the sky .
our aim is to determine whether , depending on assumed mechanical properties , it could experience surface particulate motions , reshaping , or spin - state alteration due to tidal forces caused by earth s gravity field .
the classical roche limit for a cohesionless fluid body of bulk density @xmath11 @xmath12 to not be disrupted by tidal forces is @xmath13 earth radii , so we do not expect any violent events to occur during the rocky asteroid s 2029 encounter at @xmath3 earth radii . the presence of granular material ( or regolith ) and boulders at the surface of small bodies has been demonstrated by space missions that visited or flew by asteroids in the last few decades ( e.g. , veverka et al .
2000 ; fujiwara et al .
it appears that all encountered asteroids to date , from the largest one , the main belt asteroid ( 4 ) vesta by the dawn mission , to the smallest one , the nea ( 25143 ) itokawa , sampled by the hayabusa mission , are covered with some sort of regolith .
in fact , thermal infrared observations support the idea that most asteroids are covered with regolith , given their preferentially low thermal inertia ( delbo et al .
there even seems to be a trend as a function of the asteroid s size based on thermal inertia measurements : larger objects are expected to have a surface covered by a layer of fine regolith , while smaller ones are expected to have a surface covered by a layer of coarse regolith ( clark et al .
this trend is consistent with observations by the near - shoemaker spacecraft of the larger ( @xmath14 km mean diameter ) eros , whose surface is covered by a deep layer of very fine grains , and by the hayabusa spacecraft of the much smaller ( @xmath15 m mean diameter ) itokawa , whose surface is covered by a thin layer of coarse grains .
however , interpretation of thermal inertia measurements must be made with caution , as we do not yet have enough comparisons with actual asteroid surfaces to verify that the suggested trend is systematically correct .
thus , we are left with a large parameter space to investigate possible surface motion during an earth close approach of an asteroid with unknown surface mechanical properties .
our approach is to consider a range of simple and well - controlled cases that certainly do not cover all possibilities regarding apophis surface mechanical properties , but rather aim at demonstrating whether , even in a simple and possibly favorable case for surface motion , some resurfacing event can be expected to occur during the passage .
for instance , instead of considering a flat granular surface , we consider a sandpile consisting of a size distribution of spherical grains ( section [ s : initialcondition ] ) and vary the grain properties in order to include more or less favorable cases for motion ( from a fluid - like case to a case involving rough particles ) .
slight disturbances may manifest as very - small - scale avalanches in which grain connections readjust slightly , for example .
the forces acting on the sandpile are obtained by measuring all `` external '' perturbations during the encounter , including body spin magnitude and orientation changes , for cases in which the global shape remains nearly fixed , and again assuming simple and favorable configurations of the asteroid .
indirectly , the encounter may also lead to internal reconfigurations of the asteroid , which in turn produce seismic vibrations that could propagate to the surface and affect the regolith material .
these secondary modifications are not modeled here , although it may be possible in future work to account for this by shaking the surface in a prescribed manner . in any case , for this particular encounter , we demonstrate ( section [ s : reshpeff ] ) that any global reconfiguration will likely be small to negligible in magnitude . in the following , we first present , in section [ s : method ] , the numerical method used to perform our investigation , including the initial conditions of the sandpile adopted to investigate surface motion , the representation of the encounter , and the mechanical environment .
results are described in section [ s : results ] , including potential reshaping of the asteroid , tidal disturbances for apophis encounter in 2029 , which is a function of the sandpile properties , spin orientation changes , and the dependency of the location of the sandpile on the asteroid to the outcome of the encounter .
we also show the responses of the sandpiles for artificially close approaches ( @xmath16 and @xmath17 earth radii ) to demonstrate that our method does predict significant alteration of the sandpiles when this is certainly expected to happen .
the investigation is discussed in section [ s : discuss ] and conclusions are presented in section [ s : concl ] .
we use ` pkdgrav ` , a parallel @xmath2-body gravity tree code ( stadel 2001 ) adapted for particle collisions ( richardson et al . 2000 ; 2009 ; 2011 ) .
originally collisions in ` pkdgrav ` were treated as idealized single - point - of - contact impacts between rigid spheres .
a soft - sphere option was added recently ( schwartz et al .
2012 ) ; with this new functionality , particle contacts last many timesteps , with reaction forces dependent on the degree of overlap ( a proxy for surface deformation ) and contact history this is appropriate for dense and/or near - static granular systems with multiple persistent contacts per particle .
the code uses a 2nd - order leapfrog integrator , with accelerations due to gravity and contact forces recomputed each step .
various types of user - definable confining walls are available that can be combined to provide complex boundary conditions for the simulations .
the code also includes an optional variable gravity field based on a user - specified set of rules .
the spring / dash - pot model used in ` pkdgrav ` s soft - sphere implementation is described fully in schwartz et al .
briefly , a ( spherical ) particle overlapping with a neighbor or confining wall feels a reaction force in the normal and tangential directions determined by spring constants ( @xmath18 , @xmath19 ) , with optional damping and effects that impose static , rolling , and/or twisting friction .
the damping parameters ( @xmath20 , @xmath21 ) are related to the conventional normal and tangential coefficients of restitution used in hard - sphere implementations , @xmath22 and @xmath23 .
the static , rolling , and twisting friction components are parameterized by dimensionless coefficients @xmath24 , @xmath25 , and @xmath26 , respectively .
plausible values for these parameters are obtained through comparison with laboratory experiments ( also see section [ s : initialcondition ] ) .
careful consideration of the soft - sphere parameters is needed to ensure internal consistency , particularly with the choice of @xmath18 , @xmath19 , and timestep a separate code is provided to assist the user with configuring these parameters correctly .
the numerical approach has been validated through comparison with laboratory experiments ; e.g. , schwartz et al . ( 2012 ) demonstrated that ` pkdgrav ` correctly reproduces experiments of granular flow through cylindrical hoppers , specifically the flow rate as a function of aperture size , and found the material properties of the grains also affect the flow rate .
also simulated successfully with the soft - sphere code in ` pkdgrav ` were laboratory impact experiments into sintered glass beads ( schwartz et al .
2013 ) , and regolith , in support of asteroid sampling mechanism design ( schwartz et al .
2014 ) .
we use a two - stage approach to model the effect of a tidal encounter on asteroid surface particles ( regolith ) .
first , a rigid ( non - deformable ) object approximating the size , shape , and rotation state of the asteroid is sent past the earth on a fixed trajectory ( in the present study , the trajectory is that expected of ( 99942 ) apophis see section [ s : representme ] ; note that the actual shape of apophis is poorly known beyond an estimate of axis ratios , so we assume an idealized ellipsoid for this study ) . all forces acting at a target point designated on the object surface are recorded ( section [ s : representme ] ) .
then , a second simulation is performed in the local frame of the target point , allowing the recorded external forces to affect the motion of particles arranged on the surface ( in the present study we consider equilibrated sandpiles ) .
this two - stage approach is necessary due to the large difference in scale between the asteroid as a whole and the tiny regolith particles whose reactive motion we are attempting to observe .
we approximate the regolith , which in reality likely consists of a mixture of powders and gravel ( clark et al .
2002 ) , by a size distribution of solid spheres .
we mimic the properties of different materials by adjusting the soft - sphere parameters ( section [ s : initialcondition ] ) .
the soft - sphere approach permits simulation of the behavior of granular materials in the near - static regime , appropriate for the present case in which the regolith particles generally remain stationary for hours and suffer a rapid disturbance only during the moments of closest approach to the planet .
in particular , the model permits a detailed look at the responses of local individual particles even when the tidal effects are too weak to cause any macroscopic surface or shape changes ; this gives insight into the limit of tidal resurfacing effects . in order to assess the effect of tidal encounters on a small surface feature
, we carried out numerical simulations in a local frame consisting of a flat horizontal surface ( i.e. , the local plane tangential to the asteroid surface at the target point ) with a `` sandpile '' resting on top .
the sandpile consists of @xmath2 = 1683 simulated spherical particles with radii drawn from a power - law distribution of mean @xmath27 cm and @xmath28 cm width truncated ( minimum and maximum particle radii @xmath29 and @xmath30 cm , respectively ) . between the sandpile and the floor is a rigid monolayer of particles drawn from the same distribution and laid down in a close - packed configuration in the shape of a flat disk .
this rigid particle layer provides a rough surface for the sandpile to reduce spreading ( fig .
[ f : sandpile ] ) .
three different sandpiles were constructed using the material parameters described below .
particles were dropped though a funnel from a low height onto the rough surface to build up the sandpile .
these sandpiles were allowed to settle until the system kinetic energy dropped to near zero .
this approach eliminates any bias that might arise from simply arranging the spheres by hand in a geometrical way ; the result should better represent a natural sandpile .
figure [ f : sandpile ] . for this study , we compared three different sets of soft - sphere parameters for the sandpile particles ( table [ t : ssdemparams ] ) .
our goal was to define a set of parameters that spans a plausible range of material properties given that the actual mechanical properties of asteroid surface material are poorly constrained . in the specific case of apophis , very little is known beyond the spectral type , sq ( binzel et al .
there are no measurements of thermal properties that might give an indication of the presence or absence of regolith on apophis .
consequently , we chose three sets of material parameters that span a broad range of material properties .
the first set , denoted `` smooth '' in the table , consists of idealized frictionless spheres with a small amount of dissipation ( 5% , chosen to match the glass beads case ) .
this is about as close to the fluid case that a sandpile can achieve while still exhibiting shear strength arising from the discrete nature of the particles ( and the confining pressure of surface gravity ) .
it is assumed this set will respond most readily to tidal effects due to the absence of friction between the particles and the non - uniform size distribution .
the second set , `` glass beads , '' is modeled after actual glass beads being used in a set of laboratory experiments to calibrate numerical simulations of granular avalanches ( richardson et al .
2012 ) . in this case
@xmath23 was measured directly , which informed our choice for @xmath21 ( schwartz et al .
2012 ) , and @xmath24 and @xmath25 were inferred from matching the simulations to the experiments .
the glass beads provide an intermediate case between the near - fluid smooth spheres and the third parameter set , denoted `` gravel . ''
table [ t : ssdemparams ] .
the gravel parameters were arrived at by carrying out simple avalanche experiments using roughly equal - size rocks collected from a streambed . in these experiments , the rocks ( without sharp edges ) were released near the top of a wooden board held at a 45@xmath31 incline .
the dispersal pattern on the stone floor was measured , including the distance from the end of the board to approximately half of the material , the furthest distance traveled by a rock along the direction of the board , and the maximum angle of the dispersal pattern relative to the end of the board ( fig .
[ f : rockslide]a c ) .
a series of numerical simulations was then performed to reproduce the typical behavior by varying the soft - sphere parameters ( fig .
[ f : rockslide]d f ) .
the values used in table [ t : ssdemparams ] for `` gravel '' were found in this way .
the @xmath24 and @xmath25 values are quite large , reflecting the fact that the actual particles being modeled were not spheres .
a correspondingly smaller timestep is needed to adequately sample the resulting forces on the particles in the simulations .
the value of @xmath22 was measured by calculating the average first rebound height of sample gravel pieces that were released from a certain height ; @xmath23 was not measured but since the rocks were rough it was decided to simply set @xmath32 .
this exercise was not meant to be an exhaustive or precise study ; rather , we sought simply to find representative soft - sphere parameters that can account plausibly for the irregularities in the particle shapes . in any case , we will find that such rough particles are difficult to displace using tidal forces in the parameter range explored here , so they provide a suitable upper limit to the effects .
similarly , we do not consider cohesion in this numerical study , which would further resist particle displacement due to tidal forces .
figure [ f : rockslide ] . a new adaptation of ` pkdgrav ` developed in this work allows for the simulation of sandpiles located on an asteroid surface , based on the motion equations of granular material in the noninertial frame fixed to the chosen spot .
detailed mechanics involved in the local motion of the sandpile during a tidal encounter were considered and represented in the code , including the contact forces ( reaction , damping , and friction between particles and/or walls ) and the inertial forces derived from an analysis of the transport motion of a flyby simulation . here
we provide a thorough derivation of the relevant external force expressions .
figure [ f : frames ] .
figure [ f : frames ] illustrates four frames used in the derivation , the space inertial frame ( spc ) , mass center translating frame ( mct ) , body fixed frame ( bdy ) and local frame ( loc ) .
( [ e : posspc2loc ] ) gives the connection between the spatial position and the local position of an arbitrary particle in the sandpile , for which @xmath33 is the vector from spc s origin to the particle , @xmath34 is the vector from spc to mct / bdy , @xmath35 is the vector from mct / bdy to loc , and @xmath36 is the vector from loc to the particle : @xmath37 equations ( [ e : velspc2loc])([e : accspc2loc ] ) are derived by calculating the first- and second - order time derivatives of eq .
( [ e : posspc2loc ] ) , which denote the connections of velocity and acceleration between the spatial and local representations , respectively .
@xmath38 indicates the angular velocity vector of the asteroid in spc or mct .
the operator @xmath39 denotes the time derivative with respect to the inertial frame spc , while @xmath40 denotes the time derivative with respect to the body - fixed frame bdy or loc : @xmath41 @xmath42+\frac{\tilde \mathrm{d}}{\mathrm{d } t } { { \mbox{\boldmath{$\omega$ } } } } \times ( \mathbf{l}+\mathbf{r})+2 { { \mbox{\boldmath{$\omega$ } } } } \times \frac{\tilde \mathrm{d}}{\mathrm{d } t } \mathbf{r}+\frac{\tilde \mathrm{d}^2 } { \mathrm{d } t^2 } \mathbf{r}.\ ] ] the dynamical equation for an arbitrary particle in the inertial frame spc can be written as eq .
( [ e : spcdyneqn ] ) , which shows that the forces a particle in the sandpile feels can be grouped into three categories : @xmath43 denotes the local gravity from the asteroid ; @xmath44 denotes the attraction from the planet ; and @xmath45 represents all the contact forces coming from other particles and walls : @xmath46 we get the corresponding dynamical equation in the local frame loc ( eq .
( [ e : locdynequ ] ) ) by substituting eq .
( [ e : accspc2loc ] ) into eq .
( [ e : spcdyneqn ] ) , @xmath47-\frac{\tilde \mathrm{d}}{\mathrm{d } t } { { \mbox{\boldmath{$\omega$ } } } } \times ( \mathbf{l}+\mathbf{r})-2 { { \mbox{\boldmath{$\omega$ } } } } \times \frac{\tilde \mathrm{d}}{\mathrm{d } t } \mathbf{r}.\ ] ] the terms on the right - hand side of eq .
( [ e : locdynequ ] ) represent the mechanical environment of a particle in the local sandpile during a tidal encounter , and can be interpreted as follows .
@xmath45 is the reaction force from the walls and surrounding particles , which is directly exported by ` pkdgrav ` .
@xmath43 is approximated by a constant value at the loc s origin that acts as the uniform gravity felt throughout the whole sandpile , since the latter s size is negligible compared with the asteroid dimensions .
the difference between the planetary attraction and the translational transport inertia force , @xmath48 , is the tidal force that plays the primary role of moving the surface materials .
the term @xmath49 $ ] is the centrifugal force due to the rotation of the asteroid , @xmath50 is the librational transport inertia force ( ltif ) that results when the rotational state changes abruptly during the encounter , and @xmath51 is the coriolis effect that only acts when particle movement occurs .
figure [ f : rubblepile ] .
in the numerical scheme , we derived the component dynamical equations by projecting eq .
( [ e : locdynequ ] ) to loc and treating each term separately . the local gravity @xmath43 and loc origin @xmath35 are constant vectors , which were initialized in the simulations as parameters .
the complications arising from the asteroid motion ( tidal force , angular velocity , and acceleration ) were solved within a simulation in advance , using a rigid rubble pile as the asteroid model and exporting the physical quantities required in each step . to specify apophis s encounter with earth in 2029 , we computed the planet trajectory by tracing back from the perigee conditions , @xmath3 @xmath52 distance and @xmath53 km / s encounter speed ( giorgini et al .
2008 ) , where @xmath52 indicates the average earth radius ( @xmath54 km ) . as fig .
[ f : rubblepile ] illustrates , a tri - axial ellipsoid model with axis ratio 1.4:1.0:0.8 was employed to approximate the overall shape of apophis ( scheeres et al .
2005 ) , corresponding to an equivalent radius of @xmath55 m ( mller et al .
a rubble pile bounded by the tri - axial model was constructed for the flyby simulation .
the rubble pile model consisted of @xmath2 = 2855 equal - size spherical particles in a close - packed configuration .
the total mass was set to @xmath56 kg ( bulk density @xmath11 @xmath12 ) with an initial rotation around the maximum principle axis of inertia of period @xmath57 h ( tholen et al . 2013 ) .
three markers ( non - coplanar with the origin ) were chosen from the surface particles to track the variation of the rubble pile s attitude in spc .
a complete list of its motion states during the encounter was exported to a data file , which was accessed by the subsequent simulations with local sandpiles .
richardson et al . ( 1998 ) showed the complicated behavior of a rotating rubble pile due to a tidal encounter .
generally , a variation in rotational state is usually induced because of the coupling effect between librational motion and orbital motion . especially for elongated bodies like apophis ,
the terrestrial torques during its flyby may force a strong alteration of the rotational state in a short period ( scheeres et al .
the modifications in structure of the rubble pile are also notable .
the tidal encounter outcomes show a strong dependence on the progenitor elongation , the perigee distance , and the time spent within the roche sphere .
the outcomes also depend on the spin orientation of the progenitor to the extent that retrograde rotators always suffer less catastrophic consequences than prograde ones ( richardson et al .
1998 ) . however , the parameter space considered in that study ( richardson et al .
1998 ) did not include the apophis scenario in 2029 , since the focus was on the conditions for full - scale distortion or disruption .
for this study , we first redid the flyby simulations ( section [ s : representme ] ) using a true ( non - rigid ) rubble pile model with the apophis encounter parameters of @xmath3 @xmath52 and @xmath53 km / s .
we quantified the tidal reshaping effect by measuring the maximum net change of a normalized shape factor defined by eq .
( [ e : shpfact ] ) ( hu et al . 2004 ) , in which @xmath58 , @xmath59 , @xmath60 are the principal inertia moments , with @xmath61 : @xmath62 the shape factor @xmath63 $ ] roughly describes the mass distribution of the body ( 0=oblate , 1=prolate ) .
the initial rubble - pile model ( equilibrated , prior to the encounter ) had @xmath64 , and its relative change , defined by @xmath65 , was measured for several simulations parameterized by bulk density and constituent material type ( table [ t : fullscalechange ] ) .
the corresponding roche limits are provided in the table , with the values estimated for the case of a circular orbit .
the reshaping effects show consistent results for all three types of materials , varying between a noticeable change in shape ( magnitude @xmath66 ) and a negligible change in shape ( magnitude @xmath67 ) , with the sharpest transition occurring for a critical bulk density @xmath68 @xmath12 with corresponding roche limit of @xmath3 @xmath52 .
this agrees with the analysis of apophis disruption limit using a continuum theory for a body made up of solid constituents ( holsapple et al .
note we find that even a bulk density as low as @xmath69 @xmath12 did not dislodge any particles enough for them to end up in a new geometrical arrangement , due to the very short duration of the tidal encounter .
figure [ f : newpile ] .
table [ t : fullscalechange ] . since idealized equal - size spheres can arrange into crystal - like structures that may artificially enhance the shear strength of the body ( tanga et al .
2009 ; walsh et al .
2012 ) , we carried out a second suite of simulations with rubble piles made up of a bimodal distribution of spheres ( fig .
[ f : newpile ] ) .
the model consisted of @xmath70 big particles of radius @xmath71 m and @xmath72 small particles of radius @xmath73 m , bounded by the same tri - axial ellipsoid as shown in fig . [
f : rubblepile ] , and arranged randomly .
as before , the bulk density and material type were varied , with notable differences in outcome compared to the equal - size - sphere cases ( table [ t : fullscalechange ] ) .
particles using the smooth " parameter set are not able to hold the overall shape of apophis ( the 1.4:1.0:0.8 tri - axial ellipsoid ) ; instead the rubble pile collapses and approaches a near - spherical shape . for the other two parameter sets , however , the overall shape of the rubble pile is maintained .
figure [ f : comp ] shows the relative change of shape factor @xmath74 as a function of bulk density for both types of simulations .
as shown , the results from the unimodal and bimodal rubble piles present similar trends : before reaching the roche limit density , the reshaping remains small ( the magnitudes are @xmath75@xmath76 with a certain amount of stochasticity ) ; after that , the reshaping effect sharply increases to around @xmath77 .
the bimodal rubble piles show less resistance to the tidal disturbance , since they have less - well - organized crystalline structures compared to the unimodal cases ; they therefore exhibit larger shape modifications ( walsh et al .
2012 ) . figure [ f : comp ] .
regardless , catastrophic events were not detected for either type of rubble pile until the bulk density was as small as @xmath69 @xmath12 , for all material parameter sets .
therefore , the reshaping effects on apophis in 2029 should be negligibly small for bulk densities in the likely range ( @xmath78@xmath79 @xmath12 ) .
even so , minor internal reconfigurations resulting from the tidal encounter may produce seismic waves that could propagate and affect the configuration of surface regolith .
since we are only looking at localized areas on the surface , isolated from the rest of the body , vibrations emanating from other regions , in or on the asteroid , are not evaluated in the current work , although again we expect these to be small or non - existent for the specific case of the apophis 2029 encounter .
only the external " forces acting directly on the surface particles in the considered localized region are taken into account .
we will look into the effects of seismic activity , which may stem from other regions of the asteroid , on an actual high - resolution rubble pile in future work . in this study ,
we focus on external forces , outside of the rubble pile itself ; the rigid rubble pile model is employed as a reasonable simplification for the purpose of measuring the external forces on a surface sandpile .
the right - hand side of eq .
( [ e : locdynequ ] ) provides a description of the constituents of the mechanical environment experienced by a sandpile particle , including local gravity , tidal force , centrifugal force , ltif , and the coriolis effect .
note the coriolis force does not play a role before particle motion begins , so it can be ignored when examining the causes of avalanches / collapses .
the centrifugal force and ltif show weak dependence on local position @xmath36 , so these terms can be simplified by substituting @xmath80 , since the dimension of the sandpile is much smaller than that of the asteroid .
this leads to an approximation of the resultant environmental force ( eq . (
[ e : envirforce ] ) ) , which provides a uniform expression of the field force felt throughout the sandpile : @xmath81 the variation of @xmath82 shows a common pattern for different encounter trajectories and different locations on the asteroid : it stays nearly invariable for hours as the planet is approaching / departing and shows a rapid single perturbation around the rendezvous .
this feature enables us to evaluate the mechanical environment by measuring the short - term change of @xmath82 , and moreover , to make a connection between these external stimuli with the responses of sandpiles in the simulations .
spherical coordinates @xmath83 are used to represent the resultant force ( fig .
[ f : accs ] ) , in which @xmath84 is the magnitude and @xmath85 and @xmath86 denote the effective gravity slope angle and deflection angle , respectively .
figure [ f : accs ] .
the effects of the environment force must be considered in the context of the sandpile modeling .
a rough approximation of the catastrophic slope angle ( eq . ( [ e : critcang ] ) ) can be derived from the conventional theory of the angle of repose for conical piles ( brown et al .
1966 ) , namely that avalanches will occur when the effective gravity slope angle @xmath85 exceeds @xmath87 : @xmath88 here @xmath89 denotes the resting angle of the ( conical ) sandpile , assumed to be less than the repose angle @xmath90 , and @xmath91 denotes the critical slope angle , which , if exceeded by the effective gravity slope angle , results in structural failure of the sandpile , i.e. , an avalanche
. essentially , the avalanches of the sandpile depend primarily on the instantaneous change of the slope angle @xmath85 of the resultant force and should be only weakly related to the magnitude @xmath84 and deflection angle @xmath86 .
the global distribution of changes of @xmath85 were examined for the duration of the encounter ( see below ) .
it would be difficult to make an exhaustive search over all asteroid encounter orientations due to the coupling effects between the asteroid s rotation and the orbital motion .
instead , we chose @xmath92 representative trajectories along the symmetry axes of the apophis model ( at perigee ) for study , which serves as a framework for understanding the influence of encounter orientation and the strength of tidal disturbance at different locations .
( the technique used to match the orientation of the asteroid at perigee is to first run a simulation in reverse starting at perigee with the required orientation , thereby providing the correct asteroid state for a starting position far away . )
figure [ f : traject ] shows the @xmath92 trajectories along three mutually perpendicular planes , including both the prograde and retrograde cases .
this choice is based on the symmetry of the dynamical system , and all @xmath92 trajectories have a speed of @xmath53 km / s at a perigee distance of @xmath3 @xmath52 .
figure [ f : traject ] .
figure [ f : traject ] presents these representative trajectories in order , each corresponding to a special possible orientation of the encounter .
the effective gravity slope angle changes throughout the surface of the tri - axial ellipsoid model were recorded for each trajectory , with particular attention to the location and time of the maximum change .
we found that these maximum changes concentrate at several minutes around perigee .
figure [ f : maxcsa ] shows the maximum values of slope angle change along these trajectories , indicating that the largest perturbation on slope angle is less than @xmath93 .
we verified that the achievable range of the effective gravity slope angle during the encounter ( section [ s : representme ] ) is within the safe limit predicted by eq .
( [ e : critcang ] ) for all three sandpiles ( section [ s : initialcondition ] ) ; that is , a slope change below @xmath93 is not enough to trigger any massive avalanches throughout the sandpiles .
figure [ f : maxcsa ] .
figure [ f : maxcsa ] also shows a general dependence of the tidal perturbation on the orientation of the encounter trajectory , namely that the three step levels seen in the figure correspond to three directions of the relative planet motion at perigee .
trajectories @xmath94@xmath95 correspond to the direction of the body long axis , which leads to relatively strong tidal effects ; trajectories @xmath96@xmath97 correspond to the direction of the intermediate axis , which leads to moderate effects ; while trajectories @xmath98@xmath92 correspond to the direction of the short axis , which leads to relatively weak tidal effects .
figure [ f : patterns ] illustrates the distribution of effective gravity slope angle change at perigee for the three trajectory sets , with a denoting trajectories @xmath94@xmath95 , b denoting trajectories @xmath96@xmath97 , and c denoting trajectories @xmath98@xmath92 .
we confirm that the four trajectories in each set present visually the same distribution around perigee , therefore we chose the patterns due to trajectories 1 , 5 , and 9 for demonstration of sets a , b , and c , and generalize these three patterns as representative .
several points can be inferred from fig .
[ f : patterns ] .
first , the direction to the planet at perigee largely determines the global distribution of tidal perturbation , that is , the strongest effects tend to occur along the long axis and the weakest effects along the short axis .
second , at perigee , the largest slope change occurs near areas surrounding the pole for the most favorable orientation ( set a ) , while the largest slope change at the pole occurs several minutes before or after perigee ( see fig . [
f : pxpypz ] ) .
these maximum slope changes are about equal in magnitude ( the change is only slightly smaller at the pole compared to the area immediately surrounding it , but not enough to make a difference to the avalanches ) , so for simplicity we just use the poles themselves as our testing points .
third , the duration of strong tidal effects depends on the eccentricity of the encounter trajectory .
we checked the trajectories shown in fig .
[ f : traject ] and found the duration of strong perturbation , defined for illustration as the period when the force magnitude stays above @xmath99 of the peak value , is tens of minutes long ; correspondingly , the responses of the surface material are also transitory and weak .
figure [ f : patterns ] .
the three poles p@xmath100 , p@xmath101 and p@xmath102 ( see figure [ f : traject ] ) were chosen as the test locations for the sample sandpiles since the effective gravity slope angle is near zero at these locations , which enables the sandpiles to hold their initial shapes . figure [ f : pxpypz ] illustrates the time variation of slope angle at the poles during the encounter .
the results derived from all @xmath92 trajectories in fig . [ f : traject ] are shown and to the same scale .
the most perturbed pole is p@xmath101 on the medium axis , which gains the largest slope angle change for most cases , except in trajectories \{3,4,9,10}. it is notable that trajectories in the same plane always share similar variation at the three poles , such as \{1,2,5,6 } , \{3,4,9,10 } , and \{7,8,11,12}. the curves at p@xmath100 and p@xmath101 show similar doublet shapes since those poles are both located in the rotational plane .
the variation at p@xmath102 is more complicated due to the fact that the spin pole is not perpendicular to the orbit plane .
figure [ f : pxpypz ] .
figure [ f : pxpypz ] can be used to estimate the changes in the mechanical environment at these pole locations . since the slope angle change is the primary mechanism to drive avalanches on the sandpiles ,
the results derived from different trajectories serve as a framework to determine the magnitude of tidal perturbation at the candidate locations and locations in between , which turns out to be quite small for the apophis encounter ( less than about @xmath103 in slope for these cases ) .
as analyzed above , catastrophic avalanches solely due to external forces acting directly on the surface particles may never occur on apophis during the 2029 encounter since the tidal perturbation will be very weak , however it might still have the potential to trigger some local tiny landslides of the surface materials . generally , the regolith experiences more activity than the constituents deep in the asteroid due to the dynamical environment on the surface , including the microgravity , maximum tidal and centrifugal acceleration , and smaller damping forces from the surroundings due to the smaller confining pressure , therefore we conclude that resurfacing due to external perturbations should occur before wholesale reshaping during the encounter ( discussed in section [ s : reshpeff ] ) . in the following sections we detail our numerical examination of the local sandpiles for the predicted apophis encounter scenario to estimate the limit and magnitude of the material responses in 2029 .
one problem that must be confronted in the numerical simulations is that the soft - sphere sandpiles exhibit slow outward spreading due to the accumulation of velocity noise , and this slow spreading may eventually lead to a collapse of the whole sandpile .
unfortunately , our integration time required is long ( hours ) for a granular system , thus the numerical noise has to be well limited in our method .
two techniques are used in this study .
first , as described in section [ s : initialcondition ] , a rough pallet made up of closely packed spheres in a disk fixed to the ground is used to reduce spreading of the bottom particles .
second , we introduce a critical speed in the code , below which all motions are considered to be noise and are forced to zero . we found a speed threshold @xmath104 m / s by launching hours - long simulations of a static sandpile for different material properties and finding the corresponding minimum value of the critical speed that limits the numerical spreading .
simulations of sandpiles without a flyby were carried out first , serving as a reference for subsequent flyby simulations .
sandpiles in this situation were confirmed to stay equilibrated for a long period , which suggests the avalanches in our numerical experiments ( if any ) would be attributed entirely to the effects of the tidal encounter .
we performed local simulations with the sandpiles generated in section [ s : initialcondition ] to consider the response of different materials to the apophis encounter .
the local frames at the three poles p@xmath100 , p@xmath101 and p@xmath102 ( fig .
[ f : traject ] ) were used to position the sample sandpiles , and the flyby simulation data derived from the @xmath92 fiducial trajectories were employed as the source of external perturbations for the local simulations .
there are @xmath105 possible combinations in total for different materials , locations , and trajectory orientations , covering a wide range of these undetermined parameters . in all the simulations , the start distance of apophis from the center of the planet was set to @xmath106 @xmath52 to ensure the sandpiles were equilibrated fully before it approached the perigee . for our study , we concentrated on the connections between the sandpile s responses and significant variables , e.g. , spin orientation and sandpile locations , by which we will get a better sense of the surface processes due to a tidal encounter .
the sandpiles located at pole p@xmath101 for trajectory 2 ( fig .
[ f : traject ] ) were examined in detail , which is the case suffering the strongest tidal perturbation ( fig .
[ f : pxpypz ] ) .
figure [ f : avalanches ] illustrates the disturbances on sandpiles of three different materials for this configuration .
each panel includes a snapshot of the sandpile after the tidal encounter with the displaced particles highlighted and a diagram showing the time variations of the total kinetic energy with planetary distance .
intensive events can be identified by noting the peaks of the kinetic energy curves , and the particles involved in these events are marked with different colors to show the correspondence between the disturbed site and occurrence time .
figure [ f : avalanches ] .
the scale of the disturbances proves to be tiny even for the strongest " case in fig .
[ f : avalanches ] . as illustrated , the displaced particles are few in number and mostly distributed on the surface of the sandpile .
we found that the maximum displacement of a given particle is less than @xmath107 times its radius ; that is , these disturbances only resulted from the collapse of some local weak structures and the small chain reaction among the surrounding particles .
moreover , the gravel " sandpile proves to be unaffected by the tidal encounter , because the static friction is quite large ( table [ t : ssdemparams ] ) and all particles are locked in a stable configuration , in which case the total kinetic energy of the sandpile remains very small ( @xmath108 j ) .
the glass beads " sandpile suffered small but concentrated disturbances that involved very few particles .
accordingly , the displacements of these particles are relatively large .
smooth " sandpile presents near - fuild properties with many surface particles experiencing small - amplitude sloshing .
the motion eventually damps out and the displacements of the involved particles are tiny ( smaller than that of glass beads " ) .
based on the detailed analysis of this representative scenario , we performed simulations at other poles and at other orientations , constructing a database to reveal any connections between disturbances and these parameters .
however , the results show little association between these two : for the gravel " sandpile , no disturbances were detected for any location and any orientation due to the large @xmath109 ; for the glass beads " sandpile , some small - scale avalanches can always be triggered , while the occurrence ( site and time ) of these avalanches seems to be widely distributed and independent of location and orientation ; and for the smooth " sandpile , surface particles can feel the tidal perturbation and show slight sloshing , but no visible avalanches eventually resulted , because the initial low slope angle in this case imposes a relatively stable structure that is always able to recover from the external perturbations .
to illustrate the effect of stronger perturbations on our model sandpiles , we carried out a few simulations of closer approaches , specifically at @xmath16 and @xmath17 @xmath52 , for the same encounter speed of @xmath53 km / s .
these scenarios , though unlikely to occur in 2029 based on current understanding of apophis orbit , are presented here to illustrate the magnitudes of tidal resurfacing effects over a wider range of perturbation strengths .
flyby simulations of rubble piles were first carried out to quantify the reshaping effects , using the monodisperse and bidisperse particle rubble - pile models ( see section [ s : reshpeff ] ) .
the bulk densities for both models were set to @xmath110 g / cc .
table [ t : reshp24 ] presents the relative net changes of shape factor @xmath65 for these rubble piles of different materials at different perigee distances .
table [ t : reshp24 ] .
table [ t : reshp24 ] shows results consistent with section [ s : reshpeff ] , namely that the bimodal rubble pile shows greater fluidity and larger shape changes during the encounter . to be more specific , the magnitude of reshaping effects at perigee distance @xmath16 @xmath52 ( still larger than the roche limit of @xmath111 @xmath52 for a fluid body ) remains small ( @xmath112 ) , and that at perigee distance @xmath17 @xmath52 ( smaller than the roche limit ) achieves a significant level of @xmath113@xmath77 .
it is notable that the rubble piles did not experience any irreversible distortion even for an approach distance as close as @xmath17 @xmath52 , because the duration of strong tidal effects is relatively short ( see section [ s : reshpeff ] ) .
anyway , in this section we still adopt the assumption of rigidity to measure the quantities required by the local simulations , which is acceptable since these scenarios are fictitious and only designed to exhibit some massive resurfacing effects . for the same reason , in this section we do not present systematic local simulations for different sandpile locations and orientations as done for the real encounter scenario ; instead , we simply place the sample sandpiles ( see fig . [
f : sandpile ] ) at the pole of the body long axis p@xmath100 , and choose trajectory @xmath94 ( see fig .
[ f : traject ] ) as the encounter trajectory to give rise to the maximum tidal effects .
figures [ f : refrsh4f ] and [ f : refrsh2f ] illustrate the responses of the three sandpiles for perigee distance @xmath16 @xmath52 and @xmath17 @xmath52 , respectively .
each panel includes snapshots of the sandpile before and after the tidal encounter , and the overall shape of the sandpile is traced with a white border for emphasis .
diagrams showing the time variation of the total kinetic energy with planetary distance are also included , in which the intensive avalanches can be identified by noting the peaks of the kinetic energy curves ( note the different vertical scales ) .
figure [ f : refrsh4f ] .
[ f : refrsh2f ] . as illustrated in fig .
[ f : refrsh4f ] , the shape changes of the three sandpiles are still small ( but visible ) for the encounter at @xmath16 @xmath52 perigee , and involve many more particles than the encounter at @xmath3 @xmath52 ( fig . [ f : avalanches ] ) . and
accordingly , the magnitudes of the total kinetic energy increase consistently for the three sandpiles of different materials , except for the gravel case at @xmath16 @xmath52 , where it appears the frictional `` lock '' established when the pile was first created is disturbed enough to cause a stronger distortion than might otherwise be expected , causing a sharp spike in the kinetic energy plot . essentially , the `` gravel '' sandpile experienced a more significant collapse than the `` glass bead '' sandpile .
as stated in section [ s : initialcondition ] , the three sandpiles were constructed in a manner that allowed for some inherent randomness , thus their energy state and strength may differ to some degree , and appearently the `` gravel '' sandpile was closer to its failure limit than the other two .
this adds an extra element of stochasticity to the results , an aspect to be explored in future work .
figure [ f : refrsh2f ] shows the results from the encounter at @xmath17 @xmath52 perigee , for which the encounter trajectory partly entered the roche sphere ( @xmath111 @xmath52 ) . in this case , the shapes of the three sandpiles are highly distorted during the encounter , with the involved particles slumped towards the direction where the planet is receding .
the corresponding changes in total kinetic energy become extremely large when the massive avalanches occur , especially for the `` smooth '' particles .
the results of this section suggest that a @xmath16 @xmath52 encounter may alter the regolith on apophis surface slightly , and a @xmath17 @xmath52 encounter may produce a strong resurfacing effect ( of course , we would expect considerable global distortion as well , if the asteroid is a rubble pile ) .
the shape changes of the sandpiles depend on the orientation of the encounter trajectory ; i.e. , particles can be dragged away along the direction in which the planet recedes .
the argument about whether a terrestrial encounter can reset the regolith of neas has been discussed for a while ( e.g. , binzel et al .
2010 , nesvorn et al .
important evidence is provided by measurements of the spectral properties , which suggest that asteroids with orbits conducive to tidal encounters with planets show the freshest surfaces , while quantitative evaluation of this mechanism is still rare and rough .
the two - stage approach presented in this paper enables the most detailed simulations to date of regolith migration due to tidal effects , which we have applied to make a numerical prediction for the surface effects during the 2029 approach of ( 99942 ) apophis .
we confirm that the shape modification of apophis due to this encounter will likely be negligibly small based on the results of systematic simulations parameterized by bulk density , internal structure , and material properties .
the analysis of the global mechanical environment over the surface of apophis during the 2029 encounter reveals that the tidal perturbation is even too small to result in any large - scale avalanches of the regolith materials , based on a plausible range of material parameters and provided that external forces acting directly on the surface dominate any surface effects due to seismic activity emanating from other regions of the body .
future work will explore these second - order surface effects and their relevance to this encounter , and to cases of tidal resurfacing of small bodies in general .
it is notable that our approach is capable of capturing slight changes in the regolith ( modeled as sandpiles ) ; thus , through numerical simulation , we find that this weak tidal encounter does trigger some local disturbances for appropriate material properties .
these possible disturbances , though local and tiny , are essentially related to the nature of the sandpile , which is actually a bunch of rocks and powders in a jammed state .
the sandpile is formed by a competition between the constituents to fall under gravity and eventually reach an equilibrium with mutual support .
it can hold a stable shape under a constant environment force at low sandpile densities ( compared with close packing ) , which is primarily due to interior collective structures , called bridges , that are formed at the same time as the sandpile and are distributed non - uniformly throughout the sandpile ( mehta 2007 ) .
the soft - sphere method provides a detailed approximation to a real sandpile in terms of the granular shape and contact mechanics , which can well reproduce the bridges that dominate the structure of the sandpile .
accordingly , we can describe the nuanced responses due to the changes in @xmath83see section [ s : gcme ] and fig .
[ f : accs ] .
changing the slope angle @xmath85 is the most efficient way to break the equilibrium of substructures in the sandpile , while changing the magnitude @xmath84 and deflection angle @xmath86 may also play a role in the causes of avalanches .
changes in @xmath84 can readjust the interparticle overlaps and cause collapse of some bridges , which is a principal mechanism of compaction by filling the voids during small structural modifications ( mehta et al .
this dependence was recently demonstrated in experiments of measuring the angle of repose under reduced gravity ( kleinhans et al .
similarly , the weak links in bridges of the sandpile may also be disturbed and broken during the sweeping of @xmath86 in the plane .
in addition , mehta et al .
( 1991 ) pointed out that even slight vibration caused by the environment force may result in collapse of long - lived bridges , which is also responsible for the sandpile landslides during the encounter .
although the magnitudes of the avalanches are different for the three materials we tested , they are nonetheless all very small .
we note the occurrence of these local avalanches shows little dependence on the orientation / location of the sandpile , because all the perturbations are small in magnitude and the sandpile behavior , these small - scale avalanches , actually depends on the presence of weak bridges inside the sandpile , which in turn could be sensitive to the changes of the environmental force s magnitude , deflection angle , and slope angle in a somewhat random way . in any case
, this numerical study shows that tidal resurfacing may not be particularly effective at moderate encounter distances .
we predict that overall resurfacing of apophis , regolith will not occur if the only source of disturbance is external perturbations .
mini - landslides on the surface may still be observed by a visiting spacecraft with sufficiently sensitive monitoring equipment .
this provides a great motivation for @xmath0 @xmath1 exploration of apophis in 2029 ( michel et al .
this paper provides a numerically derived prediction for the surface effects on ( 99942 ) apophis during its 2029 approach , which is likely to be one of the most significant asteroid encounter events in the near future .
a two - stage scheme was developed based on the soft - sphere code implementation in ` pkdgrav ` to mimic both a rubble pile s ( rigid and flexible ) responses to a planetary flyby and a sandpile s responses to all forms of perturbations induced by the encounter .
the flyby simulations with the rubble pile indicate that reshaping effects due to the tidal force on apophis in 2029 will be negligibly small for bulk densities in the expected range ( @xmath78@xmath79 @xmath12 ) .
the resultant environmental force felt by the sandpile on the asteroid surface was approximated with a uniform analytical expression , which led to an estimate of the changes in the global mechanical environment .
three typical patterns of perturbation were presented based on the asteroid body and spin orientation at perigee , showing a general dependence of the magnitude of tidal perturbation on the orientation of the trajectory .
twelve fiducial trajectories were used to calculate the magnitude of the tidal perturbation at three poles of the tri - axial ellipsoid model , indicating that the strongest tidal perturbation appears where the local slope is originally steep and that the duration of the strong perturbation is short compared with the whole process .
the tidal perturbation on surface materials is confirmed to be quite weak for the 2029 encounter , therefore large - scale avalanches may never occur .
however , we showed this weak perturbation does trigger some local tiny landslides on the sample sandpiles , though the involved particles are few in number and are distributed on the surface of the sandpile .
these small - scale avalanches result from the breaking of weak substructures by slight external perturbations , therefore the occurrence of these local avalanches is widely distributed and presents little dependence on the encounter parameters .
the simulations of closer approaches show that an encounter at @xmath17 @xmath52 is capable of triggering some massive avalanches of the sandpiles , i.e. , to alter the regolith on apophis surface significantly ( although the entire body would also undergo significant shape change in this case ) .
further research will be performed to generalize our work over a wide range of possible asteroid - planet encounter conditions . and as stated above , we will also investigate whether even slight internal perturbations in the asteroid during tidal encounters may contribute to noticeable surface motions .
this material is based on work supported by the us national aeronautics and space administration under grant nos .
nnx08am39 g , nnx10aq01 g , and nnx12ag29 g issued through the office of space science and by the national science foundation under grant no .
some simulations in this work were performed on the yorp " cluster administered by the center for theory and computation in the department of astronomy at the university of maryland .
patrick michel acknowledges support from the french space agency cnes .
ray tracing for figs.1 , 2 , 4 , 5 and 12 was performed using pov - ray ( http://www.povray.org/ ) .
1999 catastrophic disruptions revisited icarus 142 520 2009 spectral properties and composition of potentially hazardous asteroid ( 99942 ) apophis icarus 200 480485 2010 earth encounters as the origin of fresh surfaces on near - earth asteroids nature 463 331334 1966 principles of powder mechanics oxford : pergamon 2007 albedo and size determination of potentially hazardous asteroids : ( 99942 ) apophis icarus 188 266269 2013 mars encounters cause fresh surfaces on some near - earth asteroids icarus 227 112122 2006 the rubble - pile asteroid itokawa as observed by hayabusa science 312 13301334 2008 predicting the earth encounters of ( 99942 ) apophis icarus 193 119 2006 tidal disruptions : a continuum theory for solid bodies icarus 183 331348 2004 numerical determination of stability regions for orbital motion in uniformly rotating second degree and order gravity fields planetary and space science 52 685692 2010 fragment properties at the catastrophic disruption threshold : the effect of the parent body s internal structure icarus 207 5465 2011 static and dynamic angles of repose in loose granular materials under reduced gravity journal of geophysical research 116 e11004 2010 do planetary encounters reset surfaces of near earth asteroids ?
icarus 209 510519 2007 granular physics cambridge university press 1991 vibrated powders : a microscopic approach physics review letter 67 394397 2004 cooperativity in sandpiles : statistics of bridge geometries journal of statistical mechanics : theory and experiment p10014 2001 collisions and gravitational reaccumulation : forming asteroid families and satellites science 294 16961700 2012 probing the interior of asteroid apophis : a unique opportunity in 2029 special session 7 of the xxviii iau general assembly , 2931 august , 2012 , beijing ( china ) , abstract book 1998 tidal distortion and disruption of earth - crossing asteroids icarus 134 4776 2000 direct large - scale @xmath2-body simulations of planetesimal dynamics icarus 143 4559 2009 numerical simulations of asteroids modeled as gravitational aggregates planetary and space science 57 183192 2011 numerical simulations of granular dynamics : i. hard - sphere discrete element method and tests icarus 212 427437 2012 numerical simulations of landslides calibrated against laboratory experiments for application to asteroid surface processes american astronomical society , dps meeting 44 105.06 2005 abrupt alteration of asteroid 2004 mn4 s spin state during its 2029 earth flyby icarus 178 281283 2012 an implementation of the soft - sphere discrete element method in a high - performance parallel gravity tree - code granular matter 14 363380 2013 numerically simulating impact disruptions of cohesive glass bead agglomerates using the soft - sphere discrete element method icarus 226 6776 2014 low - speed impact simulations into regolith in support of asteroid sampling mechanism design i. : comparison with 1-g experiments planet .
space sci .
2001 cosmological @xmath2-body simulations and their analysis university of washington 141 2009 rubble - pile reshaping reproduces overall asteroid shapes the astrophysical journal 706 197202 2013 improved astrometry of ( 99942 ) apophis acta astronautica 90 5671 2000 near at eros : imaging and spectral results science 289 20882097 2012 spin - up of rubble - pile asteroids : disruption , satellite formation , and equilibrium shapes icarus 220 514529 2013 the potentially dangerous asteroid ( 99942 ) apophis monthly notices of the royal astronomical society 434(4 ) 30553060 .soft - sphere parameter sets used to represent three different material properties of sandpiles in the simulations . here
@xmath24 is the coefficient of static friction ( @xmath114 , where @xmath115 is the material friction angle ) , @xmath25 is the coefficient of rolling friction , @xmath22 is the normal coefficient of restitution ( 1 = elastic , 0 = plastic ) , @xmath23 is the tangential coefficient of restitution ( 1 = no sliding friction ) , @xmath18 is the normal spring constant ( @xmath116 ) , and @xmath19 is the tangential spring constant ( also @xmath116 ) . [
cols="^,^,^,^",options="header " , ]
* figure [ f : sandpile ] : * snapshots of sandpiles constructed using three different materials , which are ( a ) gravel , " ( b ) glass beads , " and ( c ) smooth , " with corresponding soft - sphere parameters listed in table [ t : ssdemparams ] .
the same constituent spheres are used in each sandpile both for the free particles ( green ) and the rigid particles ( white ) .
the values of average slope and pile height after equilibrium is achieved are indicated in the snapshots .
snapshots of the avalanche experiment and corresponding simulations . in the experiments ,
@xmath117 similar - size gravel pieces were piled up on the slope each time and released all at once by removing the supporting board .
snapshots from both the experiment and simulation include frames from the beginning ( a , d ) , middle ( b , e ) , and end ( c , f ) of the avalanche event , respectively .
sketch of the frames used for deriving the motion equations of the local sandpile , indicated with different colors .
spc ( black ) is an inertial frame with the origin and axes set by ` pkdgrav ` .
mct ( blue ) is a noninertial translating frame with the origin at the mass center of the asteroid and @xmath118 , @xmath119 , @xmath120-axes parallel to spc .
bdy ( green ) is a noninertial frame that is fixed to the asteroid and initially coincides with mct .
loc ( red ) is a frame fixed to the asteroid surface , taking the origin to be where the sandpile is located and choosing the @xmath120-axis to be normal to the surface at that point .
diagram of asteroid ( 99942 ) apophis. the tri - axial ellipsoid ( light gray ) denotes the overall shape model , while the arranged particles ( khaki ) denote the rubble - pile model used for numerical simulations .
the highlighted particles ( red ) are markers used for determining the attitude of the asteroid in spc . snapshot of the rubble - pile model for apophis with bimodal particles in irregular packing .
the tri - axial ellipsoid ( light gray ) denotes the overall shape model , and the arranged particles ( khaki ) show the rubble - pile model used for numerical simulations . the relative change of shape factor @xmath74 as a function of bulk density during the encounter .
the solid line ( with triangular points ) shows the results for the original rubble pile with equal - sized particles .
the dashed line ( with square points ) shows the results for the bimodal rubble pile with irregular packing .
sketch of the environmental force that the sandpile feels in loc .
the plane ( gray ) indicates the local tangential plane .
the red arrow indicates the environmental force given by eq .
( [ e : envirforce ] ) .
label @xmath84 indicates its magnitude , and @xmath85 and @xmath86 indicate the slope angle ( elevation ) and deflection angle ( azimuth ) of the effective gravity , respectively .
sketch of the fiducial trajectories in different orientations .
these trajectories ( blue numbered lines ) are placed in the perpendicular planes ( yellow ) of the apophis model ( gray ellipsoid ) , matching the three axial directions at perigee .
the blue arrows show the prograde and retrograde directions .
the red arrow denotes the angular velocity .
p@xmath100 , p@xmath101 , p@xmath102 indicate the poles along the three axes @xmath118 , @xmath119 , @xmath120 , respectively .
maximum values of slope angle change for the fiducial trajectories .
the stars denote the maximum values , and the dashed lines denote the levels of tidal perturbation ( the average of the 4 values in each case ) , which divide the @xmath92 trajectories into @xmath79 basic catagories according to the spin orientation at perigee .
see fig .
[ f : traject ] for the trajectory orientations .
global distribution of slope angle change at perigee for the three trajectory sets a , b and c , each including 4 trajectories of the same spin orientation at perigee ( see text for an explanation of each set ) .
a uniform colormap ranging up to @xmath121 is used for the three plots .
time variation of the effective gravity slope angle at the poles p@xmath100 ( dashed lines ) , p@xmath101 ( solid lines ) , and p@xmath102 ( dotted lines ) during the encounter .
each subgraph corresponds to one fiducial trajectory marked with its number on the upper left ( refer to fig .
[ f : traject ] ) .
snapshots of sandpiles after the tide - induced avalanches in 2029 , with corresponding time variations of their total kinetic energy .
the three panels include the results of sandpiles constructed using the three materials ( a ) gravel , " ( b ) glass beads , " and ( c )
smooth , " respectively .
particles that moved more than @xmath69 of a particle radius during the avalanches are highlighted .
the peaks in the total kinetic energy curves indicate occurrence of avalanches ; the peaks are labeled , and corresponding particles that moved appreciably are marked in different colors ( red , blue , orange ) in the snapshots .
the dotted line in each graph shows the evolution of the planet - asteroid distance during the encounter .
snapshots of the sample sandpiles before and after the avalanches induced by the encounter at @xmath16 @xmath52 perigee , with corresponding time variations of their total kinetic energy .
the white lines in the snapshots indicate the original and final shapes of the sandpiles .
the peaks in the total kinetic energy curves indicate occurrence of avalanches , and the dotted line shows the variation of the planet - asteroid distance during the encounter . | asteroid ( 99942 ) apophis close approach in 2029 will be one of the most significant small - body encounter events in the near future and offers a good opportunity for @xmath0 @xmath1 exploration to determine the asteroid s surface properties and measure any tidal effects that might alter its regolith configuration .
resurfacing mechanics has become a new focus for asteroid researchers due to its important implications for interpreting surface observations , including space weathering effects .
this paper provides a prediction for the tidal effects during the 2029 encounter , with an emphasis on whether surface refreshing due to regolith movement will occur .
the potential shape modification of the object due to the tidal encounter is first confirmed to be negligibly small with systematic simulations , thus only the external perturbations are taken into account for this work ( despite this , seismic shaking induced by shifting blocks might still play a weak role and we will look into this mechanism in future work ) .
a two - stage approach is developed to model the responses of asteroid surface particles ( the regolith ) based on the soft - sphere implementation of the parallel @xmath2-body gravity tree code ` pkdgrav ` . a full - body model of apophis
is sent past the earth on the predicted trajectory to generate the data of all forces acting at a target point on the surface .
a sandpile constructed in the local frame is then used to approximate the regolith materials ; all the forces the sandpile feels during the encounter are imposed as external perturbations to mimic the regolith s behavior in the full scenario .
the local mechanical environment on the asteroid surface is represented in detail , leading to an estimation of the change in global surface environment due to the encounter .
typical patterns of perturbation are presented that depend on the asteroid orientation and sense of rotation at perigee .
we find that catastrophic avalanches of regolith materials may not occur during the 2029 encounter due to the small level of tidal perturbation , although slight landslides might still be triggered in positions where a sandpile s structure is weak .
simulations are performed at different locations on apophis surface and with different body- and spin - axis orientations ; the results show that the small - scale avalanches are widely distributed and manifest independently of the asteroid orientation and the sandpile location .
we also include simulation results of much closer encounters of the apophis with earth than what is predicted to occur in 2029 , showing that much more drastic resurfacing takes place in these cases . * keywords * : asteroids , dynamics ; asteroids , surfaces ; near - earth objects ; regoliths ; tides , solid body |
optical quantum technologies such as quantum computation , quantum sensing , and quantum cryptography make use of the intrinsic quantum - mechanical behavior of non - classical light , such as quadrature - squeezed , precise photon - number , or entangled photon - pair states to surpass the performance afforded by classical light .
proposals have been made that use frequency entanglement between photon pairs to enhance timing measurements @xcite , improve clock synchronization @xcite , and cancel dispersion in interferometers @xcite .
frequency - time entangled photon pairs have also been used in fundamental tests of quantum mechanics itself , e.g. through violation of bell inequalities @xcite .
advancement both in understanding of the character of quantum mechanics and facilitating development of new quantum technologies are underpinned by methods to prepare a variety of photonic quantum states , ranging from highly - entangled photon pairs to completely uncorrelated twin beams @xcite . for these sources to be useful the complete quantum state of the output light
must be controlled , including the spatial , spectral , and polarization degrees of freedom .
the most widespread sources of non - classical light have been based on nonlinear optical interactions , especially parametric downconversion ( pdc ) and four - wave mixing ( fwm ) , both in the stimulated and spontaneous regimes .
historically , spontaneous pdc ( spdc ) has been the work horse for quantum experiments due to the relative ease of experimental setup , inexpensive and widely available equipment , and a high yield of usable photon pairs .
more recently , waveguide spdc has been developed to take advantage of increased spectral flexibility , dramatically increased brightness and the ability to mode match into optical fibers @xcite .
waveguided sources can be significantly brighter than bulk sources ( per mw of pump power ) due in part to the strong transverse spatial confinement of the pump ( which increases the interaction strength ) and longer guide lengths available for conversion ( which increases the interaction length ) .
however , spdc in both bulk crystals and waveguides is limited by the natural dispersion of the nonlinear optical materials to spectral domains in which birefringent or quasi - phase matching can be accomplished @xcite .
further , bulk crystal sources are usually multi - mode in character @xcite and demand careful alignment and pump configuration @xcite .
recent progress in photon - pair generation by spontaneous fwm ( sfwm ) in optical fibers has gained much attention @xcite .
fiber sources are expected to be better suited for integration with optical - fiber and waveguide networks that are being developed for quantum information protocols @xcite , because of the similar spatial modes of the sources and networks .
there have been three distinct approaches , aside from the scheme presented here , for the production of photon pairs using sfwm in optical fibers .
these can be categorized by fiber structure and approach to phase matching : photonic - crystal fiber ( pcf ) @xcite , dispersion - shifted fiber ( dsf ) @xcite , and standard single - mode fiber ( smf ) @xcite .
our approach uses birefringent single - mode fiber ( bsmf ) .
the majority of the three previous methods all share a common aspect in that they utilize waveguide dispersion in a symmetric guide to achieve phase matching .
the critical parameter for this is the fiber zero - dispersion wavelength ( zdw ) @xcite . in principle
, the zdw can be controlled by manipulation of the fiber transverse - spatial structure in pcfs @xcite and addition of impurities in dsfs @xcite .
the latter can only tune the fiber zdw to wavelengths longer than the innate fiber zdw .
sfwm in smfs harnesses the natural zdw of silica fibers near 1310 nm using modulation instability ( mi ) to achieve phase matching . to date , the dsf and smf approaches have produced photons in the near - infrared spectral domain , where high - efficiency single - photon detectors are not readily available .
furthermore , these techniques have been plagued by raman scattering that adds noise to the signal of interest .
suppression of raman noise can be accomplished by cooling the optical fiber @xcite , although this adds a layer of experimental difficulty .
thus , using the zdw to achieve phase matching restricts the ability to tune the pump laser and wavelengths at which photons can be produced .
in contrast , the scheme presented here is flexible in the pump wavelength allowing photon pairs to be generated at arbitrary wavelengths . instead of waveguide dispersion
, it makes use of birefringence in an asymmetric guide to achieve phase matching , as previously demonstrated in the large gain regime @xcite . besides the potential to tune fiber dispersion , pcfs have high transverse spatial mode confinement ( @xmath0 mode diameter ) compared to standard optical fibers designed for visible light ( @xmath1 mode diameter ) , making the effective interaction stronger by roughly a factor of four .
however , the exotic fiber structure that gives rise to the controlled dispersion in pcf leads to transverse spatial modes that are not well matched with integrated photonic circuits @xcite in which the produced photons are to be used .
this limits the coupling efficiency between pcf and smf , currently 65% , that can be achieved @xcite .
transverse - spatial mode mismatch does not significantly affect dsf , smf , and bsmf techniques .
furthermore , unavoidable fluctuations in pcf manufacturing process may have a large effect on their dispersion due to the small core size and strong dispersion dependence on the spatial structure of the fiber , leading to a random distribution of two - photon spectral states from the same fiber draw @xcite .
indeed , the ability to create pcf with desired dispersion is still not a well - developed technology , and thus experimentalists must search the catalog of currently available fibers to find the fiber with dispersion that comes closest to their needs .
then the pump central wavelength and bandwidth must be carefully tuned to achieve the joint spectral state desired @xcite .
dispersion of standard fibers is approximately equal to that of their constituent material in bulk when the wavelengths involved are smaller than the core diameter , and thus do not suffer as much from small fluctuations .
nonclassical light sources based on fwm in standard birefringent fiber represents a useful alternative to pcf , dsf , smf and bulk pdc sources .
photon pairs generated in this way are better mode matched to integrated optical networks compared to both bulk and pcf sources .
birefringent fiber has nearly 30 years more development than pcf , which makes them significantly more cost effective , and more likely to be uniform from draw to draw . by utilizing fiber birefringence for phase matching
, photons can be created far from the raman scatter , but still in the visible and near infrared where currently available single - photon - counting detectors can be used .
the added control over the sfwm spectral state allowed by fiber birefringence can be tuned through stress induced birefringence , e.g. by transverse compression . in this
paper theoretical and experimental results showing how sfwm in standard birefringent fibers can be used to produce a range of two - photon spectral states are presented . the high heralding efficiency , low raman background , high brightness , ability to tune the joint spectral state , and readily available components make this source useful in current applications .
we begin by laying out the theoretical description of sfwm in birefringent fibers .
a discussion of raman scattering and its role in sfwm follows .
finally , the results of experiments to characterize sfwm in off - the - shelf birefringent fibers are presented .
in the process of spontaneous four - wave mixing two photons from potentially two different pump pulses ( @xmath2 and @xmath3 ) are converted into a pair of daughter photons , known as the signal ( @xmath4 ) and idler ( @xmath5 ) , moderated by a third - order nonlinear optical material .
this process occurs when energy is conserved and is most probable when the wave - vector mismatch is zero , i.e. @xmath6 @xmath7 where @xmath8 and @xmath9 ( @xmath10 ) are the angular frequency and wave vector evaluated at frequency @xmath8 of mode @xmath11 , @xmath12 is the nonlinear parameter of the fiber , @xmath13 is related to the polarization of the pumps relative to the signal and idler fields , and @xmath14 is the peak power of pulse pump 1 ( 2 ) .
frequencies that satisfy both energy conservation , eq .
( [ eq : e - cons ] ) , and zero wave - vector mismatch ( i.e. @xmath15 ) , eq .
( [ eq : phasematching ] ) , are said to be phase matched .
there are several configurations in which phase matching can be achieved @xcite .
this paper focuses on the case in which a single pump pulse polarized along one of the principal axes of a birefringent fiber , either the fast ( f ) or slow ( s ) axis , generates signal and idler photons polarized along the opposite fiber axis , s or f , respectively , as previously examined in pcf @xcite and in standard birefringent fibers in the strong pump regime @xcite . in this cross - polarized birefringent phase - matching case
, the wave - vector mismatch can be written to explicitly bring out the contribution of the fiber birefringence , @xmath16 , as @xmath17 here the dispersion of the two principal axes is modeled with the same material dispersion , given by the refractive index @xmath18 , and a constant offset , @xmath16 , added to the slow axis , along which the pump is assumed to be polarized .
the factor of @xmath19 arises from the self- and cross - phase modulation between the various modes ( @xmath20 ) , and @xmath21 is the pump peak power @xcite .
note that this approximation can be justified when far from the zero - dispersion wavelength , where the material dispersion dominates cross - polarized phase matching @xcite , and when the birefringence @xmath16 changes little over the wavelength range of interest .
the phase - matching contours , that is signal and idler frequency pairs as a function of the pump central frequency that satisfy energy conservation and phase matching ( @xmath15 ) , are plotted in fig .
[ fig : bi - phasematch ] for birefringent phase matching in silica fiber , eq .
( [ eq : bi - phasematching ] ) . to make this graph , the sellmeier equations for silica @xcite are used , with @xmath22 , the pump polarized along the slow axis , and assuming self- and cross - phase modulation terms are negligible .
note that phase matching is not possible when the pump is polarized along the fast axis in the normal dispersion regime considered here @xcite . .
( b ) magnified view of inset box in ( a ) .
the upper red line ( @xmath23 ) shows the idler ( @xmath24 ) theory ( experiment ) and the lower blue line ( @xmath25 ) shows the signal ( @xmath26 ) theory ( experiment ) for the fibercore hb800 g fiber with @xmath27 0.2 m and @xmath22 .
experimental pump power was 15mw , and theory plots neglect the @xmath28 contribution in eq .
( [ eq : bi - phasematching ] ) . experimental uncertainties are smaller than the data symbols.,scaledwidth=85.0% ] the two - photon state of the signal and idler modes of the electromagnetic field emitted from the fiber , ignoring the vacuum component , can be written as @xcite @xmath29 where @xmath30 is the two - photon spectral amplitude of the signal and idler , and @xmath31 is a two - photon state with frequencies @xmath32 and @xmath33 .
the transverse - spatial and polarization modes , assumed independent of frequency and factorable between each mode ( @xmath4 and @xmath5 ) , are omitted from eq .
( [ eq : state ] ) for clarity .
the two - photon spectral amplitude can be written in terms of the pump spectral amplitude @xmath34 and the phase - matching function @xmath35 @xcite , as @xmath36 the phase - matching function , @xmath37 , depends on fiber length , @xmath38 , birefringence , @xmath16 , and dispersion as can be seen from the definition of the wave - vector mismatch , eqs .
( [ eq : phasematching ] ) and ( [ eq : bi - phasematching ] ) .
the two - photon spectral state , @xmath30 , is completely determined by the pump - pulse spectral amplitude and fiber properties through the phase - matching function . for cross - polarized phase matching in silica fiber the birefringence , @xmath16 , and
pump central frequency , @xmath39 , determine the central frequencies of the signal and idler , positioning the joint - spectral state of the photon pair .
the overall shape of the joint - spectral state is governed primarily ( i.e. to first order in detuning from the ideal phase - matched frequencies ) by the pump bandwidth , @xmath40 , the fiber length , @xmath38 , and the angle , @xmath41 , the phase - matching function makes with the @xmath32 axis .
the relation between pump amplitude , phase - matching function and joint spectrum is illustrated in fig .
[ fig : joint - spec ] . here
the pump intensity , @xmath42 ( assuming a gaussian pump ) , phase - matching probability , @xmath43 , and the joint - spectral probability , @xmath44 are plotted for three different pump bandwidths and three different fiber lengths .
these figures are calculated using eqs .
( [ eq : bi - phasematching ] ) and ( [ eq : jsa ] ) .
note that the width of the pump spectral amplitude is governed by the pump bandwidth , @xmath40 , the width of the phase - matching function is proportional to the inverse fiber length , @xmath45 , and the angle that the phase - matching function makes with the @xmath32 axis is @xmath46 .
this angle is related to the group velocities of the signal , idler and pump modes , and thus to the higher - order dispersion properties of the fiber . here
@xmath47 represents the group - velocity mismatch between the pump and signal ( idler ) , given by @xmath48 , \label{gvmismatch}\ ] ] where @xmath49 is the first - order derivative of mode @xmath11 s wave vector with respect to frequency evaluated at the central frequency of mode @xmath11 . 0.25 , 1 , and 1.5 nm fwhm )
, phase - matching function for different fiber lengths ( left column of plots , @xmath50 21.3 , 5.7 and 3.8 cm ) and theoretical joint spectra ( lower - right 3@xmath513 array of plots ) as a function of signal and idler wavelengths for birefringent phase matching .
plots are modeled using a pump central wavelength of 704 nm , the bulk silica dispersion , @xmath22 , as measured for the fibercore hb800 g fiber , and neglecting the @xmath28 contribution in eq .
( [ eq : bi - phasematching ] ) .
the pump bandwidths and fiber lengths were chosen to display the variety of spectral states that can be realized using the birefringent phase - matching technique . ] of particular interest for photonic quantum computation is the ability to herald pure - state single photons @xcite .
for this the joint - spectral amplitude must be factorable , i.e. @xmath52 @xcite .
approximating the phase - matching function by a gaussian , @xmath53 , ( @xmath54 ) , allows approximate determination of the regime in which a factorable joint - spectral state can be achieved for a given fiber .
the pump bandwidth should be appropriately chosen to eliminate correlations , balancing pump and phase - matching bandwidths . in this gaussian approximation
a factorable joint - spectral amplitude can be attained for a pump bandwidth ( @xmath55 spectral amplitude ) given by @xmath56 note that this only works when the group - velocity mismatch terms ( @xmath57 ) have opposite signs , that is , one mode must be faster and the other slower than the pump @xcite , so that the phase - matching angle , @xmath41 , lies in the range @xmath58 $ ] . in silica bsmf , this angle is always less than @xmath59 in the visible and near infrared making it possible to create heralded single photons in pure quantum states over a broad spectral range . in actual experiments ,
the phase - matching function is not a gaussian , but a sinc function , with significant amplitude far from the central peak .
this causes the joint - spectral state to contain more than a single set of signal and idler schmidt modes @xcite , implying a heralded photon from a realistic source to be produced in a mixed rather than pure state . to obtain a pure state in an experiment the tails of the sinc function
can be spectrally filtered out leading to the theoretical purity limit of 1 .
this purity increase comes at the expense of overall production rate and heralding efficiency . using our model of birefringent - phase - matched sfwm for @xmath60 cm and @xmath22 , with a gaussian pump centered at 704 nm , 0.5 nm bandwidth ( spectral intensity ) full - width at half maximum ( fwhm )
, the purity can be increased from approximately 0.86 to 0.99 using 90% transmittance hard - edge filters centered on the signal and idler wavelengths with 0.7 nm and 1.4 nm spectral bandwidths fwhm , respectively .
the production rate is reduced by 18% and heralding efficiencies are reduced by about 10% in each beam . in addition to the complication arising from the wings in the phase - matching function , there is a another subtlety that arises in realistic experiments the pump spectral shape may not necessarily be well approximated by a gaussian in a given experiment .
for example , in ref .
@xcite , the pump spectrum took on a nearly rect - function profile due to the nature of the spectral filtering used to prepare the pump . such a spectral amplitude ( with constant phase )
leads to significantly reduced purity for heralded photons compared to a gaussian profile , by roughly 16% , due to the asymmetry introduced by the pump hard " edges .
it is therefore crucial to take care when modeling and performing actual experiments to create the desired two - photon state .
another nonlinear process , raman scattering , can occur simultaneously with sfwm and produce photons at the signal and idler wavelengths .
the process of raman scattering can be viewed as a pseudo - three - wave mixing process , in which one of the waves is a vibrational excitation of the material , an optical phonon .
there are two ways in which this can occur ( up or down conversion ) . in the quantum picture
the two raman scattering processes can be described as the splitting of a pump photon with frequency @xmath61 , into a stokes photon of frequency @xmath62 , and a phonon of frequency @xmath63 , or the annihilation of a pump photon and a phonon of frequency @xmath63 , creating an anti - stokes photon with frequency @xmath64 . at room temperature and with small detuning ( @xmath65 )
the raman process can be stimulated by thermal phonons .
however , at room temperature and for larger phonon frequencies , the phonon population is small , and spontaneous raman scattering dominates . in this regime ,
the idler photon from sfwm is typically contaminated with light at the stokes frequency , @xmath66 . the raman gain profile for silica
allows significant stimulated stokes photon production approximately 100 nm detuned from the pump wavelength when pumping in the visible to near infrared @xcite .
one approach to eliminate stimulated raman contamination from the idler wavelengths is to cool the fiber @xcite .
however , this is difficult in practice and adds significant experimental complications .
utilizing birefringent phase matching enables production of the idler sufficiently far from the pump to avoid contamination by raman scatter .
the number of raman photons produced in the spontaneous regime scales linearly with the pump power @xmath67 , and fiber length @xmath38 , i.e. @xmath68 . the number of idler photons scales quadratically as a function of the pump power and linearly with fiber length , i.e. @xmath69 @xcite .
this is illustrated in fig .
[ fig : ramanscatter ] , where the experimentally obtained signal , idler , and raman powers as a function of the pump power are shown for a 1 m fiber length .
the signal - to - noise ratio , defined by the ratio of the number of idler photons and raman scattered photons , is proportional to the pump power , @xmath70 . for higher pump powers
the probability that detection of an idler photon originates from sfwm over raman scattering is increased . however , for the purpose of heralding a single photon , coincidence detection events that do not originate from sfwm are the most detrimental .
the most significant contribution to such coincidences arises from the detection of a signal photon and stokes photon simultaneously .
the rate of these unwanted coincidences depends solely on the probability to detect a stokes photon per sfwm event , which scales linearly with pump power .
therefore , the raman contamination of coincidence events actually increases with increased power .
thus , a compromise must be struck between source brightness and background noise suppression .
it is therefore crucial to reduce raman contamination by other means , such as generating photons spectrally far from the pump , which cross - polarized birefringent phase matching can achieve . ) and idler ( red @xmath23 ) normalized power as a function of the pump power for 1 m fiber ( fibercore hb800 g ) and corresponding fit lines . in the spontaneous regime shown here ,
the raman scatter scales linearly with the pump power , while the sfwm scales quadratically.,scaledwidth=50.0% ]
since birefringence plays a central role in the sfwm experiment , this was measured for several commercially - available fibers over a spectral region from 690 nm to 850 nm .
this was done by launching into the fiber broadband ( 10 nm to 20 nm ) femtosecond laser pulses from a ti : sapphire laser ( central wavelength centered at 10 nm increments from 690 nm to 850 nm ) plane - polarized at @xmath71 with respect to the fiber fast axis .
the output light from the fiber passed through a polarizing beam splitter prior to entering a spectrometer , where spectral interference between the two polarization modes was observed , as shown in fig .
[ fig : birefringence ] . for
a fiber of length @xmath38 and pump with central wavelength @xmath72 , the birefringence can be calculated from the spectral fringe spacing , @xmath73 , using @xmath74 .
the birefringence measurements showed that @xmath16 is constant within experimental errors over the wavelength range examined .
there was a small , but not insignificant spread in the measured birefringence of the different fibers , from @xmath75 to @xmath76 .
because the detuning of the signal and idler photons from the pump central wavelength is determined by the birefringence , the fiber with the largest measured birefringence ( fibercore hb800 g ) was used in the sfwm experiments .
this ensures that photon pairs are produced as far from the pump as possible to avoid unwanted raman contamination .
the birefringence measurements are summarized in table [ table ] .
note that the quoted birefringence is only a lower bound given by the manufacturer .
c c c c manufacturer & fiber model & quoted & measured + & & @xmath16 [ 10@xmath77 & @xmath16 [ 10@xmath77 + [ 0.5 ex ] fibercore & hb750 & 3.0 & 3.9 @xmath78 0.2 + fibercore & hb780pm & 3.5 & 4.3 @xmath78 0.2 + fibercore & hb800 g & 4.2 & 4.3 @xmath78 0.2 + 3 m & fs - ls-4616 & 4.0 & 4.0 @xmath78 0.2 + [ table ] in order to demonstrate the capability of cross - polarized phase matching in standard birefringent single - mode optical fibers the experimental setup depicted in fig .
[ fig : exp_setup ] ( a ) was implemented .
femtosecond pulses from a ti : sapphire laser ( 80 mhz repetition rate , 704 nm central wavelength , and 10 nm bandwidth fwhm ) were spectrally filtered using a folded 4@xmath79 prism pulse shaper with a hard - edge filter ( adjustable slit ) , resulting in a nearly square pump spectrum of 2.5 nm bandwidth fwhm . to control the polarization of the light launched into the fiber
, the shaped pulses were passed through a polarizing beam splitter ( pbs ) and half - wave plate before being focused into the birefringent single - mode fiber ( bsmf ) ( 10 cm length ) with an aspheric lens .
the light emitted from the fiber consisting of the pump , raman , signal and idler polarized along the slow , fast and slow , and fast axes respectively , was collimated with a second aspheric lens . to split the signal and idler from the pump an achromatic half - waveplate ( ahwp ) oriented the polarizations of the pump and some of the raman vertically and the signal and idler horizontally , which were then split at a pbs .
the signal and idler were split at a dichroic mirror ( dm ) . to further suppress the pump and any raman scatter near the pump wavelength
the signal photon was passed through a broadband interference filter ( if - 36 nm bandwidth fwhm centered at 607 nm wavelength ) while the idler was sent through an angle - tuned long - pass filter ( lpf - 830 nm transition - edge wavelength ) .
the signal and idler photons were then coupled into single - mode fibers ( smfs ) designed for 633 nm and 830 nm , respectively .
prism pulse shaper and passes through a polarizing beam splitter ( pbs ) and half - wave plate ( hwp ) to control the polarization launched into the birefringent single - mode fiber ( bsmf ) .
an achromatic half - wave plate ( ahwp ) and pbs separate the pump from the sfwm , which are subsequently split at a dichroic mirror ( dm ) and further filtered using a band - pass filter ( bpf ) ( long - pass filter ( lpf ) ) in the signal ( idler ) beam .
the signal ( green , top ) and idler ( red , bottom ) are directly coupled into single - mode fibers ( smf-633 and smf-830 ) , which can be directed to ( b ) a spectrometer via 50:50 fiber coupler to measure the marginal spectra of the photons , ( c ) single - photon counting modules ( spcms ) directly and via 50:50 fiber coupler to measure coincidence rates and conditional @xmath80 , and ( d ) monochromators set to @xmath26 and @xmath24 with outputs coupled to spcms via multi - mode fibers .
count rates and coincidences are registered using a field - programmable gate array ( fpga ) connected to a personal computer ( not shown).,scaledwidth=85.0% ] observation of the cross - polarized birefringent phase - matched sfwm was initially performed by mixing the signal and idler photons at a 50:50 fiber beam splitter and connecting one output port to a fiber - coupled spectrometer as depicted in fig .
[ fig : exp_setup ] ( b ) .
figure [ fig : marginals ] ( a ) shows a typical spectrum obtained from such a measurement for @xmath81 nm , @xmath82 nm , and 15mw average pump power .
note that the signal near 611 nm is essentially free of raman background with a signal - to - noise ratio of @xmath83 , limited by detector noise .
the idler , at 830 nm , is slightly contaminated with raman background ( signal - to - noise ratio of approximately 20 ) .
raman background adds little to false coincidence counts , but affects the signal heralding efficiency , decreasing it from @xmath84 27.9% @xmath78 0.2% to 26.5% @xmath78 0.2% , including the detector efficiency .
these relatively high efficiencies are a direct result of the standard single - mode fiber used , which allows good spatial mode matching between source and fiber networks .
considerable gains in heralding efficiency should be possible by performing all spectral filtering and sorting in fiber @xcite , which is difficult with pcf . by tuning the pump central wavelength the phase - matching contours of fig .
[ fig : bi - phasematch ] were mapped out using this experimental configuration . in some situations
multimode sfwm was exhibited
more than one sfwm peak was observed in the signal and idler spectrum as shown in fig .
[ fig : marginals ] ( b ) .
this behavior is due to the fact that two different spatial modes were excited by the sfwm .
this was attributed to poor coupling of the pump and short fiber lengths that allow non - propagating modes to be transmitted .
this effect may prove useful for space - time coupling and generation of different spatial - spectral correlations via fwm in fibers , but has not been further explored .
( b ) multi - mode sfwm signal and idler peaks with residual pump and raman scatter through an edge filter for 5 nm bandwidth fwhm pump centered at 704 nm with 4 cm fiber length.,scaledwidth=85.0% ] the production rate of photon pairs , conditional second - order correlation function @xmath80 , and relative heralding efficiency of the sfwm photons were measured using the experimental setup depicted in fig .
[ fig : exp_setup ] ( c ) .
the signal smf was coupled directly to a single - photon counting module ( spcm - a ) , while the idler smf was sent to a 50:50 fiber beam splitter whose outputs were coupled to a pair of spcms ( spcm - b and spcm - c ) .
the individual photodetection- and coincidence - count rates between the detectors were recorded using field - programmable gate array ( fpga ) electronics connected to a personal computer . for an average input pump power of 15mw (
704 nm central wavelength and 3 nm bandwidth fwhm ) and fiber length of 10 cm , individual rates of @xmath85 163000/s and @xmath86 87000/s and coincidence count rate of @xmath87 23000/s were measured .
this leads to raw heralding efficiencies @xmath88 and @xmath89 , where uncertainties arise from poisson count statistics .
this is much brighter ( @xmath90pairs / s / mw ) than bulk sources ( @xmath91pairs / s / mw ) with the same average pump power @xcite , and comparable to pcf sources ( @xmath92pairs / s / mw ) @xcite .
the conditional second - order correlation function @xmath80 is a useful measure of the non - classical nature of a light source @xcite .
a value of @xmath93 indicates that when the signal detector fires ( heralding the existence of an idler photon ) the likelihood there is more than one photon in the idler mode is negligible . for classical light @xmath94 , so that the degree of non - classicality of a light source is given to some extent by violation of this inequality .
the various detection events obtained in 300s using the experimental setup shown in fig .
[ fig : exp_setup ] ( c ) , are given in table [ table2 ] .
c c @xmath95 = 24652543 & @xmath96 + @xmath97 = 12353888 & @xmath98 + @xmath99 = 11022920 & @xmath100 + & @xmath101 + [ table2 ] the conditional second - order coherence function is given by @xcite @xmath102 where @xmath103 are the singles , two - fold and three - fold coincidence counts in a given time for detectors @xmath104 .
( [ eq : g2 ] ) and the data in table [ table2 ] , the second - order coherence function is calculated to be @xmath105 , where errors assume poisson count statistics .
this result shows significant suppression below the classical limit of 1 , indicating that the light source indeed emits photon pairs with little contamination from higher photon numbers and raman scatter .
the two - photon joint spectrum was measured using the experimental setup shown in fig .
[ fig : exp_setup ] ( d ) .
the smfs containing the signal and idler beams were directed into two separate monochromators with 600 lines / mm gratings .
the output of the monochromators were directly coupled into multimode fibers ( mmfs ) with 100 @xmath106 m diameter , and sent to spcms @xcite .
the position of the mmfs determined the central wavelength of the coupled light .
the coincidence count rate between the two mmf outputs was recorded as a function of the fiber positions at the output of the monochromators , and thus wavelength settings ( @xmath26 and @xmath24 ) . the experimental joint spectrum for an input pump power of 14mw ( 704 nm central wavelength and 5.4 nm bandwidth fwhm ) and fiber length @xmath60 cm ( fibercore hb800 g ) is shown in fig .
[ fig : jointspec ] ( a ) . the theoretical prediction for this case is shown in fig . [
fig : jointspec ] ( b ) , calculated using our simplified model with birefringence , @xmath22 .
the measured and theoretical joint spectra are well - matched qualitatively .
a more quantitative measure of this is the fidelity or overlap , which for two density matrices @xmath107 and @xmath108 is given by @xmath109 .
assuming flat phase , this can be written in terms of the experimental and theoretical joint spectral probabilities , @xmath110 and @xmath111 , as @xcite @xmath112 using eq .
( [ classical_fidelity ] ) the overlap between the measured joint spectrum and theoretical joint spectrum is @xmath113 , calculated directly with no fit parameters .
this implies that our model predicts the joint spectrum of the photon pair considerably well .
the measured joint spectrum clearly shows frequency correlations between the signal and idler photons , indicating the joint spectral amplitude is not factorable . to quantify the degree of factorability ( or de - correlation ) , the schmidt number can be calculated by taking a singular - value decomposition of the joint spectrum assuming a constant phase @xcite .
this yields a lower bound on the purity of a heralded photon , which for this two - photon spectral state is 0.22 .
note that for a pump bandwidth of approximately 0.4 nm fwhm , as calculated from eq .
( [ eq : pumpbandwidth ] ) , using the fiber length and dispersion from this experiment , an approximately factorable two - photon state could be produced with this configuration .
this would enable heralded pure - state photons with no need for tight spectral filtering from the fiber .
we have experimentally demonstrated photon - pair generation in standard birefringent single - mode fiber utilizing cross - polarized birefringent phase matching . this is a useful alternative to previous methods and offers unique advantages by producing photons in a region where waveguide dispersion can be neglected for phase matching .
a model using only the material dispersion and a constant birefringence works well in the spectral region of experiments presented here .
the predicted phase - matched wavelengths and joint spectral amplitude of the two - photon state produced by sfwm are well matched to the experimental measurements .
the model predicts a broad range of two - photon spectral states ranging from correlated to completely factorable .
all of this can be accomplished at wavelengths well within the spectral sensitivity of photon - counting detectors and the spectral range of a ti : sapphire pump laser .
indeed , a key advantage of this method is that the joint spectral shape of the produced two - photon state is relatively independent of the pump central wavelength and depends mainly on the fiber length and pump bandwidth , two relatively easy experimental parameters to control .
this technique could be easily used to create heralded single - photons with narrow bandwidths and at telecommunications wavelengths by using narrower pump bandwidth and longer wavelength pump , respectively . the experimentally observed high heralding efficiencies demonstrate the potential use in integrated optical networks .
this source is anticipated to be useful for a broad audience of experimentalists in quantum optics and information .
this work was supported by the epsrc through the qip irc , the ec under qap , the ist directorate and the royal society .
fibercore provided fiber samples for part of this research .
we thank h. j. mcguinness , m. g. raymer , c. j. mckinstrie , and e. brainis for useful discussions . | we study both experimentally and theoretically the generation of photon pairs by spontaneous four - wave mixing ( sfwm ) in standard birefringent optical fibers . the ability to produce a range of two - photon spectral states , from highly correlated ( entangled ) to completely factorable , by means of cross - polarized birefringent phase matching ,
is explored . a simple model is developed to predict the spectral state of the photon pair which shows how this can be adjusted by choosing the appropriate pump bandwidth , fiber length and birefringence .
spontaneous raman scattering is modeled to determine the tradeoff between sfwm and background raman noise , and the predicted results are shown to agree with experimental data . -.3
cm 2.25pc -.3 cm 10 url # 1`#1`urlprefix[2][]#2 s. e. harris , `` chirp and compress : toward single - cycle biphotons , '' phys . rev .
lett . * 98 * , 063602 ( 2007 ) .
k. a. odonnell and a. b. uren , `` observation of ultrabroadband , beamlike parametric downconversion , '' opt . lett . *
32 * , 817819 ( 2007 ) .
m. b. nasr , s. carrasco , b. e. a. saleh , a. v. sergienko , m. c. teich , j. p. torres , l. torner , d. s. hum , and m. m. fejer , `` ultrabroadband biphotons generated via chirped quasi - phase - matched optical parametric down - conversion , '' phys .
rev .
lett . * 100 * , 183601 ( 2008 ) .
v. giovannetti , s. lloyd , and l. maccone , `` quantum - enhanced positioning and clock synchonization , '' nature * 412 * , 417419 ( 2001 ) .
v. giovannetti , l. maccone , j. h. shapiro , and f. n. c. wong , `` generating entangled two - photon states with coincident frequencies , '' phys . rev .
lett . * 88 * , 183602 ( 2002 ) .
o. kuzucu , m. fiorentino , m. a. albota , f. n. c. wong , and f. x. krtner , `` two - photon coincident - frequency entanglement via extended phase matching , '' phys .
rev .
lett . * 94 * , 083601 ( 2005 ) .
j. d. franson , `` nonlocal cancellation of dispersion , '' phys .
rev .
a * 45 * , 31263132 ( 1992 ) .
a. m. steinberg , p. g. kwiat , and r. y. chiao , `` dispersion cancellation in a measurement of the single - photon propagation velocity in glass , '' phys .
rev .
lett . * 68 * , 24212424 ( 1992 ) .
r. erdmann , d. branning , w. grice and i. a. walmsley , `` restoring dispersion cancellation for entangled photons produced by ultrashort pulses , '' phys .
rev .
a , * 62 * , 053810 ( 2000 )
. m. b. nasr , b. e. a. saleh , a. v. sergienko , and m. c. teich , `` demonstration of dispersion - canceled quantum - optical coherence tomography , '' phys .
rev .
lett . * 91 * , 083601 ( 2003 ) .
j. d. franson , `` bell inequality for position and time , '' phys .
rev .
lett .
* 62 * , 22052208 ( 1989 ) . y. h. shih , a. v. sergienko , and m. h. rubin , `` einstein - podolsky - rosen state for space - time variables in a two - photon interference experiment , '' phys . rev .
a * 47 * , 12881293 ( 1992 ) .
p. g. kwiat , a. m. steinberg , and r. y. chiao , `` high - visibility interference in a bell - inequality experiment for energy and time , '' phys .
rev .
a * 47 * , r2472r2475 ( 1993 ) .
w. p. grice , r. erdmann , i. a. walmsley , and d. branning , `` spectral distinguishability in ultrafast parametric down - conversion , '' phys .
rev . a * 57 * , r2289r2292 ( 1998 ) . c. k. law , i. a. walmsley , and j. h. eberly , ``
continuous frequency entanglement : effective finite hilbert space and entropy control , '' phys .
rev .
lett . * 84 * , 53045307 ( 2000 ) .
w. p. grice , a. b. uren , and i. a. walmsley , `` eliminating frequency and space - time correlations in multiphoton states , '' phys .
rev .
a * 64 * , 063815 ( 2001 ) .
a. valencia , a. cer , x. shi , g. molina - terriza , and j. p. torres , `` shaping the waveform of entangled photons , '' phys .
rev .
lett . * 99 * , 243601 ( 2007 ) .
k. garay - palmett , h. j. mcguinness , o. cohen , j. s. lundeen , r. rangel - rojo , a. b. uren , m. g. raymer , c. j. mckinstrie , s. radic , and i. a. walmsley , `` photon pair - state preparation with tailored spectral properties by spontaneous four - wave mixing in photonic - crystal fiber , '' opt . express * 15 * , 14,87014,886 ( 2007 ) .
k. banaszek , a. b. uren and i. a. walmsley , `` generation of correlated photons in controlled spatial modes by downconversion in nonlinear waveguides '' , opt . lett . , * 26 * , 13671369 ( 2001 ) a. b. uren , ch .
silberhorn , k. banaszek and i. a. walmsley , `` conditional preparation of single photons for scalable quantum - optical networking , '' phys .
rev .
lett . , * 93 * , 093601 ( 2004 ) .
s. tanzilli , h. d. riedmatten , h. tittel , h. zbinden , p. baldi , m. d. micheli , d. ostrowsky , and n. gisin , `` highly efficient photon - pair source using periodically poled lithium niobate waveguide , '' electron . lett . * 73 * , 26 ( 2001 ) .
j. e. sharping , m. fiorentino , and p. kumar , `` observation of twin - beam - type quantum correlation in optical fiber , '' opt . lett . * 26 * , 367369 ( 2001 ) .
j. fan , a. migdall , and l. j. wang , `` efficient generation of correlated photon pairs in a microstructure fiber , '' opt
. lett .
* 24 * , 33683370 ( 2005 )
. j. fan and a. migdall , `` generation of cross - polarized photon pairs in a microstructure fiber with frequency - conjugate laser pump pulses , '' opt .
express * 13 * , 5777 ( 2005 ) .
j. g. rarity , j. fulconis , j. duligall , w. j. wadsworth , and p. s. j. russell , `` photonic crystal fiber source of correlated photon pairs , '' opt .
express * 13 * , 534544 ( 2005 ) .
j. fulconis , o. alibart , w. j. wadsworth , p. s. j. russell , and j. g. rarity , `` high brightness single mode source of correlated photon pairs using a photonic crystal fiber , '' opt .
express * 13 * , 75727582 ( 2005 ) .
x. li , p. l. voss , j. e. sharping , and p. kumar , `` optical - fiber source of polarization - entangled photons in the 1550 nm telecom band , '' phys .
. lett . * 94 * , 053601 ( 2005 ) .
j. chen , x. li , and p. kumar , `` two - photon - state generation via four - wave mixing in optical fibers , '' phys .
rev . a * 72 * , 033801 ( 2005 ) .
a. t. nguyen , k. p. huy , e. brainis , p. mergo , j. wojcik , t. nasilowski , j. v. erps , h. thienpont , and s. massar , `` enhanced cross phase modulation instability in birefringent photonic crystal fibers in the anomalous dispersion regime , '' opt .
express * 14 * , 82908297 ( 2006 ) .
j. fan , m. d. eisaman , and a. migdall , `` bright phase - stable broadband fiber - based source of polarization - entangled photon pairs , '' phys .
rev .
a * 76 * , 043836 ( 2007 ) .
j. fulconis , o. alibart , j. l. obrien , w. j. wadsworth , and j. g. rarity , `` nonclassical interference and entanglement generation using a photonic crystal fiber pair photon source , '' phys .
rev .
lett . *
99 * , 120501 ( 2007 ) .
x. li , l. yang , l. cui , z. y. ou , and d. yu , `` fiber - based source of photon pairs at telecom band with high temporal coherence and brightness for quantum information processing , '' opt . lett .
* 33 * , 593595 ( 2008 ) .
e. a. goldschmidt , m. d. eisaman , j. fan , s. v. polyakov , and a. migdall , `` spectrally bright and broad fiber - based heralded single - photon source , '' phys .
rev .
a * 78 * , 013844 ( 2008 ) .
o. cohen , j. s. lundeen , b. j. smith , g. puentes , p. j. mosley , and i. a. walmsley , `` tailored photon - pair generation in optical fibers , '' phys .
rev .
lett . *
102 * , 123603 ( 2009 ) .
a. r. mcmillan , j. fulconis , m. halder , c. xiong , j. g. rarity , and w. j. wadsworth , `` narrowband high - fidelity all - fibre source of heralded single photons at 1570 nm , '' opt .
express * 17 * , 61566165 ( 2009 ) .
m. halder , j. fulconis , b. cemlyn , a. clark , c. xiong , w. j. wadsworth , and j. g. rarity , `` nonclassical 2-photon interference with separate intrinsically narrow band fibre sources , '' opt . express * 17 * , 46704676 ( 2009 ) .
a. s. clark , j. fulconis , j. g. rarity , w. j. wadsworth , and j. l. obrien , `` all - optical - fiber polarization - based quantum logic gate , '' phys .
rev .
a * 79 * , 030303(r ) ( 2009 ) .
a. politi , m. j. cryan , j. g. rarity , s. yu , and j. l. obrien , `` silica - on - silicon waveguide quantum circuits , '' science * 320 * , 646649 ( 2008 ) . j. c. f. matthews , a. politi , a. stefanov , and j. l. obrien , `` manipulation of multiphoton entanglement in waveguide quantum circuits , '' nature photonics * 3 * , 346350 ( 2009 ) .
g. d. marshall , a. politi , j. c. f. matthews , p. dekker , m. ams , m. j. withford , and j. l. obrien , `` laser written waveguide photonic quantum circuits , '' opt .
express * 17 * , 12,54612,554 ( 2009 ) .
b. j. smith , d. m. kundys , n. thomas - peter , p. g. r. smith , and i. a. walmsley , `` phase - controlled integrated photonic quantum circuits , '' opt .
express * 17 * , 13,63913,645 ( 2009 ) .
a. politi , j. c. f. matthews , and j. l. obrien , `` shor s quantum factoring algorithm on a photonic chip , '' science * 325 * , 1221 ( 2009 ) .
l. zhang , a.b .
uren , r. erdmann , k.a .
odonnell , c. silberhorn , k. banaszek , and i.a .
walmsley , `` generation of highly entangled photon pairs for continuous variable bell inequality violation , '' j. mod . opt . * 54 * , 707 ( 2007 ) .
a. b. uren , c. silberhorn , k. banaszek , i. a. walmsley , r. erdmann , w. p. grice , and m. g. raymer , `` generation of pure - state single - photon wavepackets by conditional preparation based on spontaneous parametric downconversion , '' laser physics * 15 * , 146161 ( 2005 ) .
p. j. mosley , j. s. lundeen , b. j. smith , p. wasylczyk , a. b. uren , c. silberhorn , and i. a. walmsley , `` heralded generation of ultrafast single photons in pure quantum states , '' phys .
rev .
lett . * 100 * , 133601 ( 2008 ) .
j. e. sharping , j. chen , x. li , and p. kumar , `` quantum - correlated twin photons from microstructure fiber , '' opt .
express * 12 * , 30863094 ( 2004 ) .
j. e. sharping , m. fiorentino , a. coker , p. kumar , and r. s. windeler , `` four - wave mixing in microstructure fiber , '' opt
. lett .
* 26 * , 10481050 ( 2001 ) .
m. fiorentino , p. l. voss , j. e. sharping , and p. kumar , `` all - fiber photon - pair source for quantum communications , '' ieee photon .
tech .
lett .
* 14 * , 983985 ( 2002 ) .
h. takesue and k. inoue , `` 1.5-m band quantum - correlated photon pair generation in dispersion - shifted fiber : suppression of noise photons by cooling fiber , '' opt .
express * 13 * , 78327839 ( 2005 ) . c. liang , k. f. lee , t. levin , j. chen , and p. kumar , `` ultra stable all - fiber telecom - band entangled photon - pair source for turnkey quantum communication applications , '' opt .
express * 14 * , 69366941 ( 2006 ) .
s. d. dyer , m. j. stevens , b. baek , and s. w. nam , `` high - efficiency , ultra low - noise all - fiber photon - pair source , '' opt .
express * 16 * , 99669977 ( 2008 ) .
s. d. dyer , b. baek , and s. w. nam , `` high - brightness , low - noise , all - fiber photon pair source , '' opt .
express * 17 * , 10,29010,297 ( 2009 ) .
j. b. altepeter , j. chen , and p. kumar , `` entangled state engineering in single - mode fibers , '' in _ conference on lasers and electro - optics / quantum electronics and laser science conference and photonic applications systems technologies _ , osa technical digest ( cd ) , ( optical society of america , 2008 ) , paper qma4 .
m. a. hall , j. b. altepeter , and p. kumar , `` generation of o - band polarization entanglement in smf-28 , '' in _ conference on lasers and electro - optics / international quantum electronics conference _ , osa technical digest ( cd ) ( optical society of america , 2009 ) , paper itue5 .
m. a. hall , j. b. altepeter , and p. kumar , `` drop - in compatible entanglement for optical - fiber networks , '' opt .
express * 17 * , 14,55814,566 ( 2009 ) .
r. h. stolen and j. e. bjorkholm , `` parametric amplification and frequency conversion in optical fibers , '' ieee j. quantum electron . *
18 * , 10621072 ( 1982 ) .
g. p. agrawal , _ nonlinear fiber optics _ , 4th ed .
( academic press , burlington , ma , 2006 ) . p. s. russell , `` photonic - crystal fibers , '' j. lightwave technology * 24 * , 47294749 ( 2006 ) .
k. f. lee , j. chen , c. liang , x. li , p. l. voss , and p. kumar , `` generation of high - purity telecom - band entangled photon pairs in dispersion - shifted fiber , '' opt . lett . * 31 * , 19051907 ( 2006 ) .
r. h. stolen , m. a. bosch , and c. lin , `` phase matching in birefringent fibers , '' opt .
lett .
* 6 * , 213215 ( 1981 ) .
s. g. murdoch , r. leonhardt , and j. d. harvey , `` polarization modulation instability in weakly birefringent fibers , '' opt . lett . * 20 * , 866868 ( 1995 ) .
s.kasap and p.capper , eds . , _ springer handbook of electronic and photonic materials _
( springer , new york , 2006 ) . c. sller , b. brecht , p. j. mosley , l. zang , a. podlipensky , n. y. joly , p. s. russell , and c. silberhorn , `` bridging visible and telecom wavelengths with a single - mode broadband photon pair source , '' quant - ph:0908.2932v1 .
p. j. mosley , j. s. lundeen , b. j. smith , and i. a. walmsley , `` conditional preparation of single photons using parametric downconversion : a recipe for purity , '' new j. phys .
* 10 * , 093011 ( 2008 ) .
q. lin , f. yaman , and g. p. agrawal , `` photon - pair generation by four - wave mixing in optical fibers , '' opt .
lett . * 31 * , 12861288 ( 2006 ) .
q. lin , f. yaman , and g. p. agrawal , `` photon - pair generation in optical fibers through four - wave mixing : role of raman scattering and pump polarization , '' phys .
rev .
a * 75 * , 023803 ( 2007 )
. j. j. thorn , m. s. neel , v. w. donato , g. s. bergreen , r. e. davies , and m. beck , `` observing the quantum behavior of light in an undergraduate laboratory , '' am . j. phys .
* 72 * , 12101219 ( 2004 ) .
a. b. uren , c. silberhorn , j. l. ball , k. banaszek , and i. a. walmsley , `` characterization of the nonclassical nature of conditionally prepared single photons , '' phys .
rev .
a * 72 * , 021802(r ) ( 2005 ) .
p. grangier , g. roger , and a. aspect , `` experimental evidence for a photon anticorrelation effect on a beam splitter : a new light on single - photon interferences , '' europhys . lett . * 1 * , 173179 ( 1986 ) .
y .- h .
kim and w. p. grice , `` measurement of the spectral properties of the two - photon state generated via type ii spontaneous parametric downconversion , '' opt
. lett . * 30 * , 908910 ( 2005 ) .
m. a. nielson and i. l. chuang , _ quantum computation and quantum information _ , ( cambridge university press , cambridge , 2000 ) . |
bell s theorem shows that the correlations exhibited by quantum mechanical systems go beyond what is achievable by any local realistic theory ( i. e. any local hidden variable model ) @xcite .
this result has deep implications and still spurs several subfields of research @xcite . on the one hand , from a foundational perspective
, it has been questioned what physical principle in nature could then be responsible for giving rise to precisely the particular set of correlations allowed by quantum mechanics . since it has been shown that this set is strictly smaller than what could be achieved by just imposing the no - signaling principle ( i. e. the impossibility of instantaneous propagation of information ) @xcite , several works have considered more restrictive physically - motivated axioms @xcite .
although these approaches rule out several subsets of no - signaling correlations , no definite answer has yet been found as there still exist supra - quantum correlations compatible with these principles @xcite . on the other hand , from a more practical point of view and in the context of quantum information theory , it has been realized that quantum nonlocality can be regarded as a resource for device - independent quantum information processing ( diqip ) .
the tasks that can be carried out in this way include key distribution for cryptography @xcite , randomness generation @xcite and dimensionality certification ( see e. g. @xcite ) . in order to understand which physical principles constrain the set of quantum correlations and to elucidate what are the ultimate limitations behind diqip protocols , a fundamental question arises :
can an efficient mathematical description of this set be found ?
it turns out that this is a highly nontrivial question .
so far , the only systematic and general form to bound this set is given by the npa hierarchy @xcite .
this provides a hierarchy of semi - definite programs that approximate the quantum set from the outside .
although this constitutes an extremely powerful tool , it has been mainly exploited numerically .
thus , many properties of the quantum set remain unknown , since verifying general statements can not be usually approached by numerical means .
it would be therefore desirable to have simple analytical conditions constraining the set of quantum correlations .
these could be moreover used to exclude some general subsets of no - signaling correlations from the quantum set based on their analytical properties or to bound the maximal efficiency diqip tasks with a given structure can attain . the aim of this article is to fill this gap : i provide simple general analytical conditions any quantum behaviour should satisfy which rely on standard matrix analysis .
although these conditions emerge from the first step of the npa hierarchy and are , therefore , only necessary , i will consider examples showing their strength and non - triviality .
furthermore , i will provide several different applications of this result : a proof of the separation between the quantum set and extremal nonlocal no - signaling correlations ( a question recently raised in @xcite ) , a systematic way to obtain quantum bounds on arbitrary bell inequalities ( generalizing the recent result @xcite ) and the possibility to do robust self - testing of bipartite maximally entangled states by building bell inequalities that are maximally violated by them .
we will consider the standard bipartite @xmath0 bell scenario @xcite in which two parties @xmath1 and @xmath2 that could have interacted in the past remain now uncommunicated .
party @xmath1 ( @xmath2 ) can freely choose questions from a finite alphabet @xmath3 ( @xmath4 ) .
given any of these inputs @xmath5 and @xmath6 , each party can obtain an outcome @xmath7 and @xmath8 , where the output alphabets are also finite .
the central object here , referred to as behaviour , is the joint conditional probability distribution of obtaining the outputs @xmath9 given the choice of inputs @xmath10 , @xmath11 ( for which we will use the shorthand @xmath12 ) .
this list of @xmath13 numbers must fulfill @xmath14 @xmath15 and @xmath16 @xmath17 .
moreover , since communication among the parties is not possible during the choice of input and recording of the output , the marginal of each party must be independent of the other s action , @xmath18 @xmath19 and @xmath20 @xmath21 .
the set of all behaviours satisfying these no - signalling conditions will be denoted by @xmath22 .
particular elements of this set are the @xmath23 different local deterministic behaviours ( ldbs ) @xmath24 in which for every party a unique output occurs with probability 1 for every choice of input .
the convex hull of these behaviours gives rise to the local set @xmath25 @xcite . on the other hand ,
the quantum set @xmath26 is given by all behaviours that can be obtained by performing measurements on bipartite quantum states of unrestricted dimension @xmath27 , i. e. @xmath28 for some projectors @xmath29 such that @xmath30 and @xmath31 equal the identity in each party s hilbert space @xmath17 .
the crucial observation mentioned in the introduction is that @xmath32 .
although @xmath33 is a convex set , it is in general very hard to decide from the definition whether a given behaviour is in @xmath33 or not . on the contrary , @xmath34 and @xmath35
are both convex polytopes with vertices given by the ldbs in the first case , to which we have to add some non - local vertices in the second case . following the standard notation , we will refer to these extremal non - local behaviours as popescu - rohrlich ( pr ) boxes . in order to present our results
, we will use some further notation .
we will arrange every @xmath36 to form the @xmath37 real matrix @xmath38 where in the standard notation of quantum mechanics @xmath39 and @xmath40 denotes the computational basis of @xmath41 and similarly for the other alphabet elements .
thus , @xmath42 can be partitioned as a block matrix @xmath43 with blocks @xmath44 normalization imposes that the entries in each block add up to one while no - signaling that the sum of the elements in the same row ( column ) for blocks in the same row ( column ) is equal .
we will furthermore consider different matrix norms .
following the schatten p - norm notation , @xmath45 will be the trace norm ( i. e. the sum of all singular values ) while @xmath46 the spectral norm ( i. e. the maximal singular value ) .
we are now in the position to state our main result . in every @xmath0 scenario , if @xmath47 then @xmath48 .
we will actually prove the following stronger result : for every @xmath47 and @xmath49 it must hold that @xmath50 the relevance of this result will be discussed later on . for the moment , let us simply point out that theorem 1 follows from it by noticing that @xmath51 where the maximization is over all isometries @xcite . in order to prove that inequality ( [ ineq1 ] ) is true , we will show that the inequality holds @xmath52 , i. e. the first step of the npa hierarchy ( notice that @xmath53 ) .
it is worth mentioning that theorem 1 can be proved without invoking this @xcite .
however , we have chosen to present this proof because it furthermore clarifies the relative strength of the condition .
behaviours in @xmath54 must fulfill @xcite that a @xmath55 real positive semidefinite matrix @xmath56 exists with @xmath57 and @xmath58 , where @xmath59 ( @xmath60 ) is a vector of @xmath61 ( @xmath62 ) with entries given by @xmath63 ( @xmath64 ) with @xmath65 ( @xmath66 ) in lexicographical order .
and @xmath67 have to be fixed to zero .
see @xcite for details .
] thus , defining @xmath68 , for any given @xmath69 the maximum value of @xmath70 attainable in @xmath33 can not be larger than @xmath71 where @xmath72 ( @xmath73 ) and @xmath74 ( @xmath75 ) a @xmath76 ( @xmath77 ) matrix
whose only nonzero entry is the @xmath78 ( @xmath79 ) with value 1 .
this is the primal form of a semi - definite program ( sdp ) @xcite with cost function @xmath80 and we will denote its solution by @xmath81 .
the dual form of this sdp corresponds to @xmath82 where @xmath83 is a @xmath84 real vector yielding the value @xmath85 . by duality , for any feasible @xmath83 ( i. e. satisfying the constraint above )
, it must hold that @xmath86 .
thus , to finish the proof it suffices to construct a feasible @xmath83 yielding the value @xmath87 .
this is the case for @xmath88 where @xmath89 is the @xmath90-dimensional vector with all entries equal to one . to see that it is feasible amounts to checking that @xmath91 this is indeed true because , since the upper left corner in strictly positive definite , by schur s complement condition @xcite this is equivalent to @xmath92 which is obviously true given that the maximal eigenvalue of @xmath93 is precisely @xmath94 .
one of the appealing properties of theorem 1 and inequality ( [ ineq1 ] ) is that they have a very compact form .
however , the reasoning used in this proof can be applied to obtain other stronger but more complicated conditions .
let us denote by @xmath95 the matrix constructed using the same prescription as @xmath42 in eq .
( [ pmatrix ] ) but with entries given by @xmath96 we will now prove that for any choice of matrix @xmath49 , @xmath97 should hold @xmath98 and , hence in every @xmath0 scenario , if @xmath47 then @xmath99 this proof goes along similar lines as the previous one , so we will only sketch the details .
the set @xmath54 is actually equivalent to the positive semidefiniteness of @xmath100 @xcite , that is , @xmath101where now @xmath102 and @xmath103 .
thus , the value of @xmath104 in @xmath33 can not be larger than the primal sdp @xmath105 with dual @xmath106 where the vectors @xmath107 have entries @xmath108 and @xmath109 .
the same choice of @xmath83 as in the previous proof does the job .
let us finish this section analyzing the strength of theorems 1 and 2 by considering a few examples .
as mentioned before , our conditions are deduced from the definition of the first step of the npa hierarchy , @xmath54 .
obviously then , the best we can expect from them is to separate this set from its complement in @xmath35 .
therefore , since @xmath110 , it is clear that there exist supra - quantum behaviours satisfying the conditions of theorems 1 and 2 ( i. e. they are necessary for a behaviour to be in @xmath33 but not sufficient ) .
we have performed numerical explorations in the ( 2222 ) scenario that show that the gap between @xmath54 and the behaviours satisfying our conditions is small .
theorem 2 only provides a slightly better approximation of @xmath54 than theorem 1 . a more detailed example
can be found in figure 1 . to my knowledge , the only previous instance of analytical means to constrain @xmath33 is given by the results of @xcite ( which emerge from @xmath54 as well ) . however , the application of these techniques is not completely straightforward as they rely on some choice of data processing .
still , the inequality emerging from this approach reproduces @xmath54 in the ( 2222 ) scenario .
notwithstanding , figure 1 shows that our conditions give a much tighter restriction already in the ( 3322 ) example considered in @xcite .
this is also apparent in the ( 2233 ) case , where the results presented in @xcite fail to completely reproduce @xmath54 for the isotropic behaviours obtained by mixtures of a fully random box and the pr box @xmath111 ( see eq .
( [ pr2d ] ) below ) . on the contrary ,
theorems 1 and 2 are tight in this case .
thus , it is interesting to point out that while inequalities ( [ ineq1 ] ) and ( [ ineq2 ] ) depend on a proper choice of @xmath69 for each case , these examples indicate that theorems 1 and 2 , which are straightforward to apply in general , still provide remarkably strong conditions . to show further the usefulness of these results , in the remaining sections we provide several applications of them . , where @xmath112 and @xmath113 are two pr boxes and @xmath114 is the fully mixed behaviour .
horizontal lines achieve the same value for the @xmath115 bell inequality ( see @xcite for details ) .
crosses : the boundary of @xmath34 and @xmath33 ( equal in this slice , @xmath116 ) . dashed line : the boundary of @xmath54 ( @xmath117 ) . solid line : the conditions of theorems 1 and 2 ( they are equivalent in this case , @xmath118 ) .
dashed - dotted line : the inequality of @xcite with the sign binning data processing ( @xmath119 ) .
a more clever data processing for this case only manages to lower this condition to @xmath120 @xcite . ]
the previous examples already give a good idea of the strength of the conditions derived here . to analyze in full generality their non - triviality ,
notice that , due to the convexity of the trace norm , @xmath121 must attain its maximum value in @xmath35 at the vertices of the polytope .
hence , the ideal situation would be that the ldbs achieve the maximal possible value ( @xmath122 ) and that all pr boxes violate this bound .
actually , it has been only recently shown in @xcite ( see also @xcite ) that all pr boxes ( including the multipartite case ) are not in @xmath33 .
it was left as an open question there whether there exists a separation between them and @xmath33 .
the interest of this question relies on the fact that pr boxes are the most advantageous resources in diqip applications such as cryptography @xcite .
thus , the closer @xmath33 can be to these behaviours the more efficient these applications can be . in the following
we show that all ldbs attain the @xmath122 bound and that all pr boxes in @xmath123 and @xmath124 scenarios violate the bound .
thus , besides showing that the bound is not improvable and in general not trivial , we further provide a simple proof in these cases of the result of @xcite .
moreover , we show that there actually exists a quantitative separation between these pr boxes and @xmath33 , answering in these scenarios the question raised therein . in every @xmath0 scenario , if @xmath125 then @xmath48 , with equality for ldbs .
every ldb is of the form @xmath126 where the lists @xmath127 ( @xmath128 ) have @xmath129 ( @xmath130 ) entries equal to 1 and 0 otherwise .
hence , @xmath131 notice that the fact that @xmath48 holds @xmath132 follows then by the convexity of the trace norm without the need of invoking theorem 1 . evidently , that the inequality is fulfilled in @xmath34 was already obvious from theorem 1 .
the important observation here is that all ldbs attain the bound , hence showing that it can not be improved .
we analyze now the values @xmath121 might take for pr boxes .
let us consider first the @xmath123 scenario .
taking @xmath133 to be the circulant @xmath134 matrix @xmath135 the pr boxes in this case are given by @xmath136 up to relabelings of the inputs and the outputs @xcite .
since these transformations amount to certain permutations of the rows or columns of @xmath42 that leave the trace norm invariant , it suffices to compute it for the matrix given in eq .
( [ pr2d ] ) . using the pinching inequality @xcite
, we obtain that @xmath137the conditions under which equality is attained in the pinching inequality are given in theorem 8.7 of @xcite and it is easily checked that they are not met in this case .
hence , we obtain that @xmath138 , which amounts to the non - quantumness of these pr boxes by theorem 1
. moreover , by a more refined use of the pinching inequality , we obtain the following stronger result , which shows the existence of a finite gap between these pr boxes and @xmath33 . in every @xmath123 scenario , it holds @xmath98 that @xmath139 @xmath140 .
we need to show that @xmath141 @xmath139 . by permutation matrices , that leave the trace norm invariant , we can map @xmath142 to @xmath143 , which has now blocks given by @xmath144 we therefore have that @xmath145 where @xmath146 hence
, using now the pinching inequality we obtain the desired result @xmath147 .
it might be interesting to note that @xmath148 . to see this ,
notice that for the frobenius norm @xmath149 and that @xmath150 for any @xmath151 matrix @xmath152 .
numerics suggest that the above estimates can be improved to @xmath153 , that would change the bound in theorem 4 to @xmath154 .
let us move now to the @xmath124 scenario .
the corresponding pr boxes have all been determined in @xcite ( see table ii therein ) .
we denote an arbitrary one of them by @xmath155 .
one sees that ( up to relabelings ) these matrices always have the following structure : they have a @xmath156 block in the diagonal given by @xmath157 followed by @xmath158 @xmath159 blocks in the diagonal , which are either @xmath160 , @xmath161 or @xmath162 .
since the latter blocks all have unit trace norm , it follows again by the pinching inequality that @xmath163 .
this shows again by virtue of theorem 1 that there is a quantitative separation between the pr boxes in these scenarios and @xmath33 : in every @xmath124 scenario , it holds @xmath98 that @xmath164 .
it might be interesting to mention as well that in the @xmath124 scenario for the fully nondeterministic boxes it holds that @xmath165 .
for these behaviours there are @xmath166 non - vanishing entries with value @xmath167 @xcite .
hence , similarly as before , @xmath168 , obtaining the above estimate .
one can see that the left - hand - side of inequality ( [ ineq1 ] ) defines an arbitrary bell expression , i. e. any linear combination of the elements @xmath11 . since @xmath34 , @xmath33 and @xmath35 are compact convex sets , there always exist such expressions separating them , i. e. @xmath169 depending on whether @xmath170 with @xmath171 .
the most characteristic one is the chsh inequality ( see below ) in the @xmath172 scenario for which @xmath173 @xcite , @xmath174 @xcite and @xmath175 @xcite . while to determine the optimal value of @xmath176 and @xmath177 it suffices to check over the corresponding vertices , to determine the optimal value of @xmath178 , known as tsirelson bounds , is a less straightforward task . and @xmath177 can be computationally hard as the number of vertices increases exponentially with the number of inputs .
] however , this is very relevant to identify optimal diqip performances in the context of quantum games @xcite .
thus , a remarkable feature of inequality ( [ ineq1 ] ) is that it provides a systematic way to construct quantum upper bounds to arbitrary bell inequalities .
actually , our result resembles that of @xcite , but the latter only holds for the particular class of bell inequalities based on correlators .
later on , we will discuss this relation further . to give a hint of the usefulness of inequality ( [ ineq1 ] )
, we will show now that it allows to obtain tsirelson s bound for the chsh inequality .
this can be expressed by a matrix @xmath179 with blocks @xmath180 it turns out that @xmath181 and , hence , we obtain the trivial bound 4 .
nevertheless , given that behaviours in @xmath35 must fulfill several different constraints , equivalent bell inequalities can be expressed up to rescaling and addition of an offset .
thus , if we take ( @xmath182 is a matrix with all entries equal to 1 ) @xmath183 it holds that @xmath184 @xmath185 . since @xmath186
, we have then that @xmath98 , @xmath187 and , hence , @xmath188 .
since the chsh tsirelson bound is achievable by a quantum behaviour arising from certain measurements on a maximally entangled two - qubit state and @xmath189 is orthogonal , this also shows that for this behaviour @xmath190 , achieving the bound of theorem 1 .
it is a natural question to ask which other behaviours in @xmath191 can attain it .
we computed @xmath121 for the quantum behaviours yielding the largest known value for several two - outcome bell inequalities given in @xcite but , in general , the bound of theorem 1 is not attained . interestingly , when this occurs , the behaviour arises from a maximally entangled state of qubits .
this seems to extend for scenarios with more outcomes . in particular , in ( 2233 ) the quantum behaviour maximally violating the cglmp inequality @xcite was given in @xcite .
however , for it we find that @xmath192 , while the maximal value @xmath190 is attained for the behaviour that yields the maximal cglmp value when restricted to a maximally entangled two - qutrit state @xcite .
this leads to consider whether for every bipartite maximally entangled state of dimension @xmath193 , latexmath:[\[\label{maxent } exists measurements such that the corresponding behaviour attains the bound of theorem 1 . in the following we show that this is indeed the case @xmath139
. we will use the construction of @xcite that provides measurements on a @xmath195 scenario ( i. e. @xmath196 and @xmath197 ) such that acting on @xmath198 lead to the behaviour @xmath199\right)^{-1},\ ] ] where @xmath200 , @xmath201 , @xmath202 and @xmath203 .
this construction is enough to show that for a maximally entangled state of any dimension , there exist a behaviour for which the bound of theorem 1 can be saturated .
however , it should be stressed that other behaviours arising from maximally entangled states can have this property as well . for any
maximally entangled state @xmath198 , the corresponding @xmath195 behaviour given in eq .
( [ maxentbeh ] ) attains the bound of theorem 1 : @xmath204 @xmath139 . in order to obtain @xmath205
we will compute its @xmath206 singular values , which we will denote by @xmath207 for reasons that will become clear later .
to that aim we will first show that @xmath208 is normal ( i. e. @xmath209 ) implying that the singular values correspond to the absolute values of the eigenvalues . by fixing the inputs @xmath210 , the matrix of the behaviour ( [ maxentbeh ] )
can be partitioned as @xmath211 where @xmath212\right)^{-1},\nonumber\\ p(ab|12)&=\left(2d^3\sin^2\left[\pi(a - b-1/4)/d\right]\right)^{-1},\nonumber\\ p(ab|12)&=\left(2d^3\sin^2\left[\pi(a - b+3/4)/d\right]\right)^{-1}.\end{aligned}\ ] ] thus , to ease the notation we will rewrite eq .
( [ pd ] ) as @xmath213 notice that the @xmath134 matrices @xmath214 are circulant , as for every fixed @xmath210 it holds that @xmath215 where it should be understood here that @xmath216 .
we will use the following properties of @xmath134 circulant matrices @xcite : they all have the same eigenvectors @xmath217 corresponding to eigenvalues @xmath218 where @xmath219 are the @xmath193-th roots of unity .
this particularly implies that all circulant matrices are normal and commute with each other .
hence , it is easy to check that @xmath208 is normal if @xmath220 . to see that this is indeed the case , consider that eq .
( [ eign ] ) tells us that the eigenvalues of our matrices are given by @xmath221 these summation formulas are computed in the appendix , obtaining @xmath222 notice then that @xmath223 corresponding to the same eigenvector , which in addition to the fact that these matrices are normal , implies indeed that @xmath220 . hence , @xmath208 is normal and , therefore , its singular values are the absolute value of its eigenvalues .
we compute now then the latter .
since all circulant matrices are diagonalized by the same unitary @xmath224 ( i. e. @xmath225 with @xmath226 diagonal for every circulant matrix @xmath152 ) , it holds then that the matrix @xmath227 has the same eigenvalues as @xmath208 .
using now the schur complement condition that tells us that @xmath228 we have that @xmath229 and , hence , @xmath230 thus , using eqs .
( [ eigs ] ) , we obtain that @xmath231 which means that @xmath232 hence , we finally obtain that @xmath233 this result is interesting because it shows that the bound of theorem 1 is attainable in @xmath191 and that behaviours arising from maximally entangled states are extremal in this sense .
notwithstanding , it has some further application .
if a real square matrix @xmath42 has singular value decomposition given by @xmath234 then @xmath235 with @xmath236 orthogonal ( as so are @xmath224 and @xmath237 ) .
thus , by choosing @xmath238 , we can always construct a bell expression such that for any given @xmath12 it holds that @xmath239 .
remarkably , if we happen to have a quantum behaviour such that @xmath240 ( i. e. it saturates the bound of theorem 1 ) , then the aforementioned prescription immediately yields a bell expression which is maximized in @xmath33 by @xmath42 .
this is because @xmath241 and , hence , by inequality ( [ ineq1 ] ) , there can not exist any other quantum behaviour @xmath242 such that @xmath243 .
thus , this allows to construct bell inequalities which are maximally violated in @xmath33 by different behaviours of interest .
theorem 6 shows that this is possible for maximally entangled states of any dimension @xcite .
for example , following this prescription for the behaviour ( [ maxentbeh ] ) with @xmath244 leads to a bell expression @xmath245 which is then maximized in @xmath33 by @xmath246 and is given by the following coefficient matrix .
hence , @xmath224 and @xmath237 are not orthogonal ( they are not square ) and nor is @xmath69 . still , it holds that @xmath247 as @xmath241 . ] @xmath248 notice that this bell inequality separates @xmath34 from @xmath33 as it is straightforward to find that the maximal value of @xmath245 under @xmath34 is @xmath249 .
a different example of a bell inequality maximally violated by @xmath246 can be found in @xcite .
thus , theorem 6 also shows that that for a maximally entangled state of any dimension there always exists a bell inequality that is maximally violated in @xmath33 by it and how to construct it @xcite . in this sense , one can then devise a diqip protocol for which maximally entangled states are optimal within @xmath33 .
this might also be of relevance in the context of self - testing @xcite if it turned out that the behaviour ( [ maxentbeh ] ) is the only one maximizing these bell expressions in @xmath33 .
self - testing arises when a certain behaviour is the unique to attain a particular bell value .
this allows to check the performance of a quantum set - up without trusting any of the devices , particularly when it can be made robust @xcite . using the techniques of @xcite with the bell inequality @xmath245 and its generalizations for other dimensions , it could be possible to check whether robust self - testing of maximally entangled states is possible in this way .
notice , moreover , that this observation above that allows to construct a bell expression such that @xmath239 can be used in other contexts .
for instance , the results of theorems 4 and 5 imply that for every pr box considered there we can write down constructively a bell expression , and hence a potential diqip protocol , whose performance has a quantitative gap with any quantum behaviour .
if we restrict ourselves ourselves to two - outcome scenarios ( @xmath250 ) and taking @xmath251 , all behaviours in @xmath35 can be alternatively characterized by the correlators @xmath252 and the marginal expectations @xmath253 as mentioned before , it has been shown in @xcite that for the particular class of correlator bell inequalities , @xmath254 must hold @xmath98 and every real @xmath255 matrix @xmath69
. we will denote by @xmath256 the @xmath255 matrix with entries @xmath257 .
it can be shown that the result above can also be proved using similar techniques as in theorems 1 and 2 @xcite .
for that , one just needs to consider that behaviours in @xmath54 must fulfill @xcite that a @xmath258 real positive semidefinite matrix @xmath259 exists with @xmath260 and @xmath261 and proceed as in the proofs of theorems 1 and 2 to upper bound @xmath262 .
this does not only provide an alternative proof of inequality ( [ eqepping ] ) but it also shows that this bound can not give stronger constraints than @xmath54 .
moreover , as in theorems 1 and 2 this leads to the following condition latexmath:[\[\label{ineqc1 } that this condition is strictly weaker than theorem 1 ( i. e. every behaviour detected as non - quantum by the above inequality is also non - quantum by theorem 1 ) as we show in the following .
for every two - outcome behaviour @xmath36 , if @xmath264 , then @xmath265 .
using again the mapping from @xmath42 to @xmath266 as in the proof of theorem 4 , we obtain that @xmath267\nonumber\\ & = \frac{\sqrt{m_am_b}+||c||_1}{2},\end{aligned}\ ] ] where the inequality comes from corollary 3 in @xcite .
the above result agrees with intuition since with @xmath256 we are not fully characterizing @xmath12 .
this suggests to consider on the analogy of theorem 2 a condition including the marginals @xmath268 and @xmath269 .
indeed , using similar arguments based on the first step of the npa hierarchy one can show that , for every matrix @xmath270 , it must hold for every quantum behaviour that @xmath271 this particularly implies the following .
for every two - outcome behaviour @xmath47 it holds that @xmath272 where @xmath273 has entries @xmath274 .
we have to show that inequality ( [ ineqc2 ] ) is true for quantum behaviours .
this follows from the fact that @xmath54 is equivalent @xcite to the positive semi - definiteness of @xmath275 where @xmath276 ( @xmath277 ) is an @xmath129(@xmath130)-dimensional vector with entries @xmath268 ( @xmath269 ) . by schur
s complement condition this leads to @xmath278 with @xmath279 and @xmath280 .
proceeding as in the proof of theorems 1 and 2 we can upper bound @xmath281 to obtain the desired result .
we have shown that the first step of the npa hierarchy allows to obtain simple analytical conditions constraining the set of quantum behaviours in general bipartite bell scenarios , whose strength and non - triviality have been illustrated .
since not all problems in quantum nonlocality and diqip can be addressed numerically , we expect these conditions to be of utility , filling the hitherto lack of such general tools .
in fact , we have applied these conditions here to obtain a variety of results . in sec .
[ nontsec ] we have shown that our bounds are tight and , in general , non - trivial , and we have used them to prove that there exists a finite gap ( whose size we have estimated ) between the quantum set and pr boxes in several general scenarios answering a question raised in @xcite . in sec .
[ tsisec ] we have provided a systematic construction of quantum bounds for arbitrary bell inequalities and we have shown that for a maximally entangled state of any dimension one can obtain a behaviour that attains the bound of theorem 1 .
interestingly , this can be translated into a bell inequality whose tsirelson bound is reached by such a state .
this could be applied for robust self - testing of maximally entangled states using the techniques developed in @xcite .
finally , in sec . [ corrsec ] we studied the particular case of correlation scenarios and established some links with the results of @xcite .
several other ideas will be further investigated in the future . given a bell inequality ,
it would be interesting to find a procedure to find the best form of @xmath69 in ( [ ineq1 ] ) to obtain its quantum upper bound and when it can be optimal .
it is also worth further research to characterize which behaviours in @xmath191 attain the bound in theorem 1 : do they only arise from maximally entangled states ?
it would be also desirable to extend this approach to the multipartite setting .
last , it is worth studying whether stronger analytical conditions as those derived here can be obtained by considering further steps of the npa hierarchy .
i thank a. acin and his collaborators for sharing with me a draft of their work on bell inequalities maximally violated by maximally entangled states @xcite .
this research was funded by the spanish mineco through grants mtm 2010 - 21186-c02 - 02 , mtm2011 - 26912 and mtm2014 - 54692 and the cam regional research consortium quitemad+cm s2013/ice-2801 .
here we prove the summation formulas used in the proof of theorem 6 : @xmath282 which follow from @xmath283 to verify eq .
( [ eqapp ] ) , first notice that the geometric sum yields @xmath284 and , therefore , @xmath285 thus , we find that @xmath286 notice that the inner sum is equal to zero , unless @xmath287 for any integer @xmath288 for which the sum is equal to @xmath193 .
given the values the indices take , this can only happen for @xmath289 or @xmath290 , hence obtaining @xmath291 that leads to the desired result .
masanes , s. pironio , and a. acin , nature comm . * 2 * , 238 ( 2011 ) ; e. hnggi and r. renner , arxiv:1009.1833 ( 2010 ) ; s. pironio , ll .
masanes , a. leverrier , and a. acin , phys .
x * 3 * , 031007 ( 2013 ) ; u. vazirani and t. vidick , phys .
lett . * 113 * , 140501 ( 2014 ) .
s. pironio _
et al . _ ,
nature * 464 * , 1021 ( 2010 ) ; r. colbeck and r. renner , nature phys .
* 8 * , 450 ( 2012 ) ; r. gallego , ll .
masanes , g. de la torre , c. dhara , l. aolita , and a. acin , nature comm . * 4 * , 2654 ( 2013 ) . after the publication of the first version of this preprint and before devising the proof of theorem 6 , a. acin informed me that his group is finishing a work on bell inequalities maximally violated by maximally entangled states using different techniques ( private communication ) .
m. mckague , t. h. yang , and v. scarani , j. phys .
* 45 * 455304 ( 2012 ) ; c. a. miller and y. shi , arxiv:1207.1819 ( 2012 ) ; t. h. yang and m. navascus , phys . rev .
a * 87 * , 050102(r ) ( 2013 ) ; b. w. reichardt , f. unger , and u. vazirani , nature * 496 * , 456 ( 2013 ) . | the characterization of the set of quantum correlations in bell scenarios is a problem of paramount importance for both the foundations of quantum mechanics and quantum information processing in the device - independent scenario .
however , a clear - cut ( physical or mathematical ) characterization of this set remains elusive and many of its properties are still unknown .
we provide here simple and general analytical conditions that are necessary for an arbitrary bipartite behaviour to be quantum .
although the conditions are not sufficient , we illustrate the strength and non - triviality of these conditions with a few examples . moreover , we provide several applications of this result : we prove a quantitative separation of the quantum set from extremal nonlocal no - signaling behaviours in several general scenarios , we provide a relation to obtain tsirelson bounds for arbitrary bell inequalities and a construction of bell expressions whose maximal quantum value is attained by a maximally entangled state of any given dimension . |
the profile of hii region around an isolated source of uv photons is an old topic in astrophysics .
a classical result was given by strmgren ( 1939 ) , who showed that the profile of the spherical hii region of a point source embedded in a uniformly distributed hydrogen gas with number density @xmath0 at radial coordinate @xmath1 can be approximately described by a step function as @xmath2 = \left \ { \begin{array}{ll } 1 , & { \rm if \ } r < r_s \\ 0 , & { \rm if \ } r > r_s \\ \end{array } \right .\ ] ] where the fraction of ionized hydrogen @xmath3 , @xmath4 being the number density of ionized hydrogen . that is , hydrogen gas is sharply divided into two regions : within a sphere with strmgren radius @xmath5 around the source , hydrogen is fully ionized , while outside the sphere hydrogen atoms remain neutral .
@xmath5 is determined by the balance between ionization and recombination @xmath6 where @xmath7 is the emission of ionizing photons of the source , and @xmath8 is the recombination coefficient of hii ( osterbrock & ferland 2005 ) .
the sharp boundary at @xmath9 is the ionization front ( i - front ) separating the hii and hi regions .
the problem of the strmgren sphere has once again attracted many studies recently , because the formation of hii regions around high redshift quasars , galaxies and first stars is crucial to understanding the evolution of the reionization ( cen & haiman 2000 ; madau & rees 2000 ; ciardi et al .
2001 ; ricotti et al . 2002 ; wyithe & loeb , 2004 ; kitayama , et al .
2004 ; whalen et al . 2004
; yu , 2005 ; yu & lu 2005 ; alvarez et al . 2006 , iliev et al .
2006 ) . unlike the static solution eq.([eq1 ] ) , new studies focus on the dynamical behavior of the ionized sphere , such as the time - dependence of the ionization profile @xmath10 , the propagation of the i - front . besides the hii region and the i - front
, a high kinetic temperature region also exists around the uv photon source due to the photon - heating of gas .
similar to the i - front , there is also a t - front separating heated and un - heated gas .
the temperature profile @xmath11 and the t - front are important for probing reionization .
for instance , the region with high kinetic temperature @xmath12 and low @xmath13 would be the 21 cm emission region associated with sources at the reionization epoch ( tozzi et al .
2000 ; wyithe et al .
2005 ; cen , 2006 ; chen & miralda - escude 2006 ) .
although the possible existence of a 21 cm emission shell around high redshift quasars and first stars has been addressed qualitatively or semi - analytically in these references , a serious calculation seems to be still lacking .
many numerical solvers for the radiative transfer equation have been proposed ( ciardi et al . 2001 , gnedin & abel 2001 , sokasian et al .
2001 , nakamoto et al .
2001 ; razoumov et al .
2002 , cen 2002 , maselli et al .
2003 , shapiro et al . , 2004 ;
rijkhorst et al .
2005 ; mellema et al .
2006 ; susa 2006 , whalen & norman 2006 ) .
these solvers provide numerical results of the i - front .
however , the results are still diverse due to the usage of different approximations .
some of the results show that the time scale of the i - front evolution is sensitively dependent on the intensity of source ( white et al .
2003 ) , while some yield intensity - independent evolution ( e.g. mellema et al .
this is because the retardation of photon propagation is ignored in the later , while the recombination is ignored in the former .
therefore , it is worth to re - calculate this problem without above - mentioned assumptions .
it has been pointed out that to study the dynamical features of the i- and t - fronts , one would need to apply high - resolution shock - capturing schemes similar to those developed in fluid dynamics ( razoumov & scott 1999 ) .
the finite difference weno scheme is an algorithm satisfying this requirement .
moreover , although the fraction of the remained neutral hydrogen within the ionized sphere is extremely small , it is not zero .
the small fraction is important to estimate the ly-@xmath14 photon leaking at high redshifts .
therefore , a proper algorithm for the dynamical properties of the ionized sphere should be able to , on the one hand , effectively capture sharp profile of ionization and temperature around the i- and t - fronts , and , on the other hand , give a precise value of the remained neutral hydrogen between the discontinuities .
this can also be satisfied by the weno algorithm .
the weno algorithm has proved to have high order of accuracy and good convergence in capturing discontinuities and complicated structures in fluid as well as to be significantly superior over piecewise smooth solutions containing discontinuities ( shu 2003 ) .
we have shown that the weno algorithm is indeed effective for solving radiative transfer problem with discontinuities with high accuracy .
for instance , it can follow the propagation of a sharp i - front and the step function cut - off of retardation ( qiu et al .
we now develop this method to solve both ionization and temperature profiles . it is not a trivial generalization . because the rate equations of heating - cooling and ionization - recombination are stiff , the time integration of the weno scheme can not be directly implemented .
a proper strategy to save computational cost on time integration will be developed .
a long - term motivation of this work , as we mentioned in ( qiu et al .
2006 ) , is to develop a weno solver of hydrodynamic / radiative transfer problems , similar to the development of a hybrid algorithm of hydrodynamic / n - body simulation based on weno scheme ( feng et al .
2004 ) .
the paper is organized as follows .
section 2 describes the problem and equations needed to be solved .
section 3 presents the weno numerical scheme .
section 4 gives the solutions of the temperature and ionization profiles and the evolution of energy spectrum of photons .
a discussion and conclusion are given in section 5 .
details of the atomic processes are listed in the appendix .
to demonstrate the algorithm , we consider the ionization of a uniformly distributed hydrogen gas in space @xmath15 with number density @xmath0 by a point uv photon source located at the center @xmath16 . adding helium component in the gas is straightforward and will not change the algorithm for radiative transfer .
if the time scale @xmath17 of the growth of the ionized sphere is less than @xmath18 , @xmath19 being the hubble parameter , the expansion of the universe can be ignored .
the radiative transfer equation of the specific intensity @xmath20 is then ( see the appendix ) @xmath21 where @xmath22 is the frequency of photon .
the source term , @xmath23 , is given by @xmath24 where @xmath25 is the energy distribution of photons emitted by the central source per unit time within the frequency range from @xmath22 to @xmath26 .
we assume the energy spectrum of uv photons to be of a power law @xmath27 , and @xmath28 is the ionization energy of the ground state of hydrogen @xmath29 ev .
integration of @xmath30 over @xmath22 gives the total intensity ( energy per unit time ) of ionizing photons emitted by the source , @xmath31 .
the absorption coefficient of eq.([eq3 ] ) is @xmath32 , where the cross section @xmath33 and @xmath34 @xmath35 .
the evolution of the number density of neutral hydrogen hi , @xmath36 , is governed by the ionization equation , @xmath37 where @xmath38 is the fraction of neutral hydrogen , and @xmath39 is the number density of electrons . obviously , @xmath40 . in eq.([eq5 ] )
, @xmath41 is the recombination coefficient and @xmath42 is the collision ionization rate .
the photoionization rate @xmath43 is given by @xmath44 the kinetic temperature of baryon gas is determined by the equation @xmath45 where @xmath46 erg @xmath47 is the boltzmann constant and the temperature @xmath12 is in unit of k. the details of the heating @xmath48 and cooling @xmath49 are given in the appendix .
in the numerical implementation , it is convenient to introduce the dimensionless variables of time , space and frequency defined by @xmath50 , @xmath51 and @xmath52 . @xmath53 is the optical depth of ionizing photon in neutral hydrogen gas with density @xmath0 .
therefore , @xmath54 and @xmath55 are respectively , the time and distance in units of mean free flight time and mean free path of ionizing photon @xmath56 in the non - ionized background hydrogen gas @xmath0 . for the @xmath57cdm model , @xmath58 @xmath59 , where @xmath60 is redshift , @xmath61 myrs and @xmath62 mpc .
correspondingly , the intensity is rescaled by @xmath63 .
thus , eqs.([eq3 ] ) , ( [ eq5 ] ) , ( [ eq6 ] ) and ( [ eq7 ] ) become @xmath64 and @xmath65 @xmath66 @xmath67 where we have assumed @xmath68 and @xmath69 the point source condition eq.([eq4 ] ) requires @xmath70 .
if we take @xmath71 , the total intensity of the source is given by @xmath72 in our numerical calculation , we solve the system of equations ( [ eq8 ] ) , ( [ eq9 ] ) and ( [ eq11 ] ) for the specific intensity @xmath73 , the fraction of the neutral hydrogen @xmath74 and the temperature @xmath12 as functions of the radius @xmath55 , frequency @xmath75 and time @xmath54 .
we will drop the prime in the variables @xmath73 , @xmath55 , @xmath75 and @xmath54 in this section below , when there is no ambiguity , and keep prime in the variables in showing the numerical results . to solve the radiative transfer equation
, we adopt the fifth - order finite difference weno scheme with anti - diffusive flux corrections .
the fifth - order finite difference weno scheme was designed in ( jiang @xmath76 shu 1996 ) and the anti - diffusive flux corrections to the high order finite difference scheme was designed in ( xu @xmath76 shu 2005 ) .
the objective of the anti - diffusive flux corrections is to sharpen the contact discontinuities in the numerical solution of the weno scheme as well as to maintain high order accuracy .
a fourth order quadrature formula is used in the computation of integration in equations ( [ eq10 ] ) and ( [ eq1112 ] ) .
the third order tvd runge - kutta time discretization is used in time integration for the system of equations ( [ eq8 ] ) , ( [ eq9 ] ) and ( [ eq11 ] ) .
we now describe our numerical algorithm in more detail .
the computational domain is @xmath77 \times [ 1 , \nu_{max}]$ ] , where @xmath78 and @xmath79 are chosen such that @xmath80 for @xmath81 or @xmath82 . in our computation ,
@xmath83 and @xmath84 .
the computational domain is discretized into a uniform mesh in the @xmath85-direction and into a smooth non - uniform mesh in the @xmath22-direction .
the uniform mesh in the @xmath85-direction is @xmath86 and the non - uniform mesh in the @xmath22-direction is taken to be @xmath87 which is allowed because only integration , e.g. in equations ( [ eq10 ] ) and ( [ eq1112 ] ) , is involved in the computation with respect to the @xmath22-variable . in our model , large @xmath22 contributes little in equations ( [ eq10 ] ) and ( [ eq1112 ] ) , therefore the non - uniform mesh is designed in a way such that the mesh becomes coarser for larger @xmath22 . to approximate the spatial derivative in equation ( [ eq8 ] ) , the weno scheme with anti - diffusive flux corrections is used
. specifically , to calculate @xmath88 , the variable @xmath22 is fixed and the approximation is performed along the @xmath85-line @xmath89 where the numerical flux @xmath90 is obtained with the procedure given below .
we can use the upwind fluxes without flux splitting in the fifth order weno approximation because the wind direction is fixed ( positive ) . to obtain the sharp resolution of the contact discontinuities , the anti - diffusive
flux corrections are used in our code .
first , we denote @xmath91 where @xmath0 and @xmath92 are fixed .
the numerical flux from the regular weno procedure is obtained by @xmath93 where @xmath94 are the three third order fluxes on three different stencils given by @xmath95 and the nonlinear weights @xmath96 are given by @xmath97 with the linear weights @xmath98 given by @xmath99 and the smoothness indicators @xmath100 given by @xmath101 @xmath102 is a parameter to avoid the denominator to become 0 and is taken as @xmath103 times the maximum magnitude of the initial condition @xmath104 in the computation of this paper .
the reconstruction of the finite difference weno flux on the downwind side @xmath105 is obtained in a mirror symmetric fashion with respect to @xmath106 as that for @xmath107 .
the anti - diffusive flux corrections are based on the fluxes obtained from the regular weno procedure .
it is given by @xmath108 where @xmath109 is the cfl number and the minmod function is defined as @xmath110 @xmath111 in eq.([eq41 ] ) is the discontinuity indicator between 0 and 1 , defined as @xmath112 where @xmath113 with @xmath114 being a small positive number taken as @xmath115 in our computation . @xmath116 and @xmath117 are the maximum and minimum values of @xmath118 for all grid points . with the definition
above , we will have @xmath119 .
@xmath120 in the smooth regions and @xmath121 is close to 1 near strong discontinuities .
the purpose of the anti - diffusive flux corrections is to improve the resolution of contact discontinuities without sacrificing accuracy and stability of the original weno scheme . the integration in equations ( [ eq10 ] ) and ( [ eq1112 ] ) is approximated by a fourth order quadrature formula @xmath122 where @xmath123 , and the weights @xmath124
are given by @xmath125 again we refer to ( qiu et al .
2006 ) for details of implementation . when considering the time integration for the system of equations ( [ eq8 ] ) , ( [ eq9 ] ) and ( [ eq11 ] ) , we start with the third order tvd runge - kutta time discretization in ( shu @xmath76 osher 1988 ) . for the system of odes
@xmath126 this time discretization reads @xmath127 the runge - kutta method needs to be modified considering the modification on the anti - diffusive flux @xmath128 by @xmath129 where the spatial derivative @xmath130 term in operator @xmath131 is defined by ( [ shuadd60 ] ) with the anti - diffusive flux @xmath132 given by eq.([eq41 ] ) , and @xmath130 in the operator @xmath133 is defined by the modified anti - diffusive flux @xmath134 as @xmath135 and @xmath130 in the operator @xmath136 is defined by the modified anti - diffusive flux @xmath137 ,
@xmath138 here @xmath139 , @xmath140 .
the difficulty of a direct implementation of the scheme lies in the stiffness of equations ( [ eq9 ] ) and ( [ eq11 ] ) . especially for strong sources ,
when j(r=0 ) is large , one needs very small time step @xmath141 to guarantee the stability of the numerical scheme , therefore huge computational cost for long time integration . by observing that * though ( [ eq9 ] ) and ( [ eq11 ] ) are stiff odes , ( [ eq8 ] ) involving the spatial derivative is not a stiff equation . * the weno procedure with the anti - diffusive flux corrections in approximating the spatial derivative is the major cost at each time step . * implicit numerical method in time evolution has milder time step restriction than the explicit one .
we have settled down with the following strategies to save the computational cost * we use different time scales to solve j in ( [ eq8 ] ) and to solve @xmath142 and t in ( [ eq9 ] ) and ( [ eq11 ] ) . the time step for j , say @xmath141 , is larger , while the time step for @xmath142 and t , say @xmath143 with @xmath144 being the number of small time steps in a large time step , is much smaller . by doing this
, we take advantage of the non - stiffness of ( [ eq8 ] ) and eliminate the cost of the weno procedure with flux corrections for each small time step . * in each small time step @xmath145 , we use a semi - implicit numerical method for strong sources to proceed in time for @xmath142 and t , which greatly releases the severe time step restriction for the sake of the stability of the scheme .
the number of small time steps @xmath145 in one large time step @xmath141 is thus greatly reduced , hence the saving of the computational cost .
more precisely , we modify the third order tvd runge - kutta time discretization in the following procedure 1 .
we have the initial condition of j , @xmath142 and @xmath12 at t=0 .
let us denote @xmath146 @xmath147 2 . for a large time step @xmath141 , we evolve @xmath148 , @xmath149 and @xmath150 by three inner stages to @xmath151 , @xmath152 and @xmath153 . *
@xmath151 : + @xmath154 , @xmath155 , @xmath151 are updated according to ( [ rk1_a ] ) , ( [ rk2_a ] ) , ( [ rk3_a ] ) . * @xmath152 , @xmath153 : + let us denote @xmath156 and @xmath157 instead of ( [ rk1 ] ) , @xmath158 and @xmath159 are evolved by @xmath160 where @xmath161 is the evolution of the solution by the euler - forward method for @xmath144 times , with the euler - forward method at each time step defined as @xmath162 instead of ( [ rk2 ] ) , the second inner stage @xmath163=@xmath164 is updated by @xmath165 with @xmath166 being the evolution of the solution by the euler - forward method for @xmath144 times based on @xmath154 as @xmath167 finally , instead of ( [ rk3 ] ) , @xmath152 and @xmath153 are computed as @xmath168 with @xmath169 obtained in a similar manner as in the previous step but based on @xmath155 @xmath170 3 .
when the source is strong , ( [ eq9 ] ) and ( [ eq11 ] ) suffer from severe time step restriction , we use a semi - implicit scheme instead of ( [ euler - for ] ) in our code to release the severe time step restriction and to save computational cost . the procedure of the implementation is the same as described above except that ( [ euler - for ] ) is replaced by @xmath171 ( [ fhi_im ] ) is computed by solving a quadratic equation with the root located between 0 and 1 , since @xmath172 is a quadratic function of @xmath142 , and ( [ t_im ] ) is computed by the newton iteration method .
( [ euler - for2 ] ) and ( [ euler - for3 ] ) are modified in a similar way as ( [ euler - for ] ) .
the boundary conditions are implemented as follows * inflow boundary condition at r=0 : + @xmath173 . *
the boundary condition at @xmath174 : + @xmath175 in this subsection , we perform a grid refinement convergence study for the numerical scheme we proposed above , to assess the accuracy of the computational result for the typical mesh sizes that we will use in next section .
we test our numerical schemes for the case with the strongest source intensity , i.e. @xmath176 erg @xmath177 .
this is the toughest situation for our numerical simulation , as the scheme suffers severe time step restriction due to the stiffness of the ode .
we use both the multi - time - scale strategy and semi - implicit time discretization described in the previous subsections , and compare the results obtained with @xmath178 points in @xmath85 and @xmath179 points in @xmath22 , which is the typical mesh size used in next section , and with a _ coarser _ mesh consisting of @xmath180 points in @xmath85 and @xmath181 points in @xmath22 , in figure [ fig1 ] .
the results with these two mesh sizes match very well , indicating that our numerical results are already numerically convergent at the coarser mesh .
the numerical results reported in next section with the finer mesh should therefore be reliable .
( left ) and @xmath182 ( right ) with the source of @xmath183 erg s@xmath184 at time @xmath185 myrs ( @xmath186 ) .
the power - law frequency spectrum has the index @xmath187 , and the reionization redshift is taken to be @xmath188 .
the solid line is the numerical result with the typical mesh @xmath189 , @xmath190 , and the circles indicate the numerical result with a coarser mesh @xmath191 , @xmath192 . ,
title="fig:",width=188 ] ( left ) and @xmath182 ( right ) with the source of @xmath183 erg s@xmath184 at time @xmath185 myrs ( @xmath186 ) .
the power - law frequency spectrum has the index @xmath187 , and the reionization redshift is taken to be @xmath188 .
the solid line is the numerical result with the typical mesh @xmath189 , @xmath190 , and the circles indicate the numerical result with a coarser mesh @xmath191 , @xmath192 . ,
title="fig:",width=188 ] we also test the accuracy of our multi - time - scale strategy by comparing the numerical results of evolving ( [ eq8 ] ) , ( [ eq9 ] ) and ( [ eq11 ] ) together with single , extremely small time step , and that of updating ( [ eq8 ] ) , ( [ eq9 ] ) and ( [ eq11 ] ) by the multi - time - scale strategy with ( @xmath145 , @xmath141 ) .
these two approaches produce numerical solutions in good agreement , see figure [ fig2 ] .
( left ) and @xmath182 ( right ) with the source of @xmath193 erg s@xmath184 at time @xmath194 myrs ( @xmath195 ) .
the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the numerical mesh is @xmath191 and @xmath190 .
the solid line is the numerical solution with single , extremely small time step , and the circles indicate the numerical solution with the multi - time - scale strategy . ,
title="fig:",width=188 ] ( left ) and @xmath182 ( right ) with the source of @xmath193 erg s@xmath184 at time @xmath194 myrs ( @xmath195 ) .
the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the numerical mesh is @xmath191 and @xmath190 .
the solid line is the numerical solution with single , extremely small time step , and the circles indicate the numerical solution with the multi - time - scale strategy . , title="fig:",width=188 ] additional grid refinement study , not reported here to save space , has been performed to assure the numerical convergence of the results reported in next section
the results in this section are obtained with the numerical algorithm described in the previous section .
the code is stable for all the cases reported in this section . as indicated in the previous section 3.7 , and we have performed grid refinement study for some representative examples to ensure that the results reported are numerically converged solutions . in our calculation , we use the explicit time discretization to solve ( [ eq9 ] ) and ( [ eq11 ] ) when @xmath196 erg @xmath177 , and semi - implicit scheme when @xmath197 erg @xmath177 , @xmath198 erg @xmath177 and @xmath183 erg @xmath177 in each of the small time step @xmath145 in the multi - time - scale strategy .
we calculate the profiles of @xmath200 and @xmath201 around a point source .
the result is presented in figure [ fig3 ] , which shows the time - dependence of the profiles @xmath201 and @xmath200 for sources with intensity , respectively from top to bottom panels , @xmath202 , @xmath203 , @xmath204 and @xmath205 erg s@xmath184 with @xmath187 and @xmath188 .
the emission rate of the number of photons is roughly @xmath206 , @xmath207 , @xmath208 and @xmath209 s@xmath184 .
( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom : for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs . the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points .
, title="fig:",width=188 ] ( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom : for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs . the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points .
, title="fig:",width=188 ] ( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom : for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs . the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points .
, title="fig:",width=188 ] ( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom : for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs . the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points . ,
title="fig:",width=188 ] ( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom : for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs . the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points .
, title="fig:",width=188 ] ( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom :
for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs .
the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points .
, title="fig:",width=188 ] ( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom : for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs . the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points . , title="fig:",width=188 ] ( left ) and @xmath200 ( right ) , in which @xmath85 and @xmath55 are the physical and dimenssionless distance , respectively . from top to bottom : for sources of @xmath210 , @xmath211 , @xmath198 and @xmath183 erg s@xmath184 at time @xmath212 ( dash dot dot ) , 0.36 ( long dash ) , 0.53 ( dash dot ) , 0.71 ( dash ) , and 0.89 ( solid line ) myrs .
the power - law frequency spectrum has the index @xmath187 , and the redshift is taken to be @xmath188 .
the mesh has @xmath213=4000 , @xmath214=200 grid points .
, title="fig:",width=188 ] figure [ fig3 ] shows that the ionized sphere generally has a sharp i - front .
the time - dependent profile @xmath201 may still be approximately described as a strmgren sphere with time - dependent size @xmath215 as @xmath216 the code can well reveal the propagation of the i - front @xmath217 . unlike the solution eq.([eq1 ] ) , @xmath199 is not zero in @xmath218 .
although the neutral hydrogen @xmath199 remained in @xmath218 is very small , it can not be ignored , because the heating of gas is given by the hard photons absorption of neutral hydrogen hi [ eq.(a8 ) ] . the high temperature within the ionized sphere is actually maintained by the ionization of the small fraction therein by all the ionizing photons [ eq.([eq22a ] ) ] .
figure [ fig3 ] indicates that @xmath199 within @xmath218 is sensitive to the intensity of the source .
this is important for investigating leakage of the ly-@xmath14 photon .
figure [ fig3 ] shows that the temperature profiles @xmath11 within the i - front @xmath218 are simple .
the temperature keeps around a constant @xmath219 k. an interesting feature is that for strong sources , the temperature at the center ( @xmath220 ) is not the highest and is even lower than the constant temperature .
this is because the ionizing ( soft ) photon flux at the center is very strong , the number of @xmath221 is extremely low ( see figure [ fig3 ] ) , and then , the heating rate is lower .
a common feature of @xmath11 is that the size of heated region is generally larger than that of ionized region at all time .
that is , hydrogen gas is already significantly heated before being ionized .
the gas is firstly heated up , and then ionized .
the region between the t - front and the i - front is a pre - heating layer , in which the temperature can be as high as @xmath222 k , or even larger . for a source of @xmath223 erg s@xmath184 at @xmath224 , the physical size of the ionized region at time 0.36
mys is about 0.06 mpc , while the physical size of the region with temperature larger than @xmath225 k is about 0.08 mpc . when @xmath185 mys , the two physical sizes are , respectively , about 0.1 mpc , and 0.16 mpc , or comoving size 1 mpc , and 1.6 mpc .
therefore , for strong sources , the time scales and comoving length scales of the pre - heating layer are in the range of cosmological interest .
an important feature of the ionization profile is that @xmath226 can be approximately written as @xmath227 $ ] , where @xmath228 is time - independent , and @xmath215 is only a function of @xmath17 .
that is , all the time - dependence of @xmath199 can be described by the i - front @xmath215 . in the region
@xmath218 , the ionization profile @xmath229 is time - independent .
it can be given by a static or stationary solution of the radiative transfer equation .
this property has been applied in some codes for radiative transfer , in which the ionization field is given by the static solution of the radiative transfer equation for a given matter distribution .
for the temperature profile , one might also approximately define a t - front function by a step function as @xmath230 $ ] .
however , unlike the i - front , the temperature profile can not be rewritten as @xmath231 $ ] , where @xmath232 is time - independent .
that is , the @xmath85-dependence of the function @xmath200 can not be separated with @xmath17 .
especially , the function @xmath11 in the range between the i- and t - fronts actually always depends strongly on @xmath17 .
figure [ fig3 ] shows that , for a very strong source @xmath233 erg s@xmath184 , we have @xmath234 , i.e. the ionizing front moves with a speed close to the speed of light . for a weak source @xmath235 erg s@xmath184
, the profile @xmath201 does not show a significant expansion of the i - front if time @xmath17 is larger than a few of free flight times of ionizing photons .
this result is also evident in figure [ fig4 ] on @xmath236 versus @xmath17 .
@xmath236 is defined by @xmath237 .
it denotes the size , within which , i.e. @xmath238 , 90% hydrogen are ionized .
therefore , it can be used for the i - front . for a strong source , such as @xmath239 ergs s@xmath184 , we have approximately @xmath240 , which implies that the ionizing region grows with an ionizing front propagating with almost the speed of light . for weak sources
, @xmath236 are also following @xmath17 at very small @xmath17 , but become @xmath241 when @xmath17 is large .
the weaker the sources , the earlier the stage of @xmath241 takes place .
this point is more clearly shown in the right panel of figure [ fig4 ] .
we see that the speed @xmath242 is close to @xmath243 when @xmath17 is small , and then the speed decreases with @xmath17 by a power law @xmath244 with @xmath245 .
one can define a time @xmath246 , larger than which , the speed @xmath242 starts to decrease with @xmath17 by the power law . at time @xmath247 , the ionizing sphere is still expanding , but very slowly .
it will finally approach a solution , of which the ionization equilibrium is approximately established , and the ionized sphere becomes time - independent .
the time @xmath246 depends on the source intensity .
the stronger the source , the larger the time @xmath246 . , which is the solution of @xmath237 .
the light solid line is @xmath240 .
right panel : @xmath248 vs. @xmath17 .
other parameters and the mesh in the numerical simulation are the same as those in figure [ fig3 ] . the sources are taken to be @xmath249 ( dash dot dot ) , @xmath211 ( dash dot ) , @xmath198 ( dash ) , and @xmath183 ( solid line ) erg s@xmath184 .
time @xmath54 is dimensionless .
, title="fig:",width=226 ] , which is the solution of @xmath237 .
the light solid line is @xmath240 .
right panel : @xmath248 vs. @xmath17 .
other parameters and the mesh in the numerical simulation are the same as those in figure [ fig3 ] . the sources are taken to be @xmath249 ( dash dot dot ) , @xmath211 ( dash dot ) , @xmath198 ( dash ) , and @xmath183 ( solid line ) erg s@xmath184 .
time @xmath54 is dimensionless .
, title="fig:",width=226 ] obviously , the speed of the propagation of t - front can not be larger than the speed of light , and therefore , the t - front will approximately coincide with the i - front when @xmath250 . only in this period ,
the ionized sphere is the same as the heated sphere . when @xmath247 , the t - front starts to exceed the i - front , and the pre - heating layer is formed .
therefore , the formation of pre - heated layer happens later for a stronger source .
figure [ fig5 ] gives a comparison of @xmath199 and @xmath11 at time @xmath185 myrs for sources with different intensity .
we can see that the pre - heating layer has been well established at time @xmath185 myrs for all sources with @xmath251 erg s@xmath184 , while the t - front of the source with @xmath252 erg s@xmath184 is still about the same as the i - front .
one can expect that for a source of @xmath253 erg s@xmath184 , a preheated layer will be formed at time @xmath254 myrs and on comoving distance @xmath255 mpc from the source .
( left panel ) and @xmath232 ( right panel ) at time @xmath185 myrs for sources with intensity @xmath249 ( dash dot dot ) , @xmath211 ( dash dot ) , @xmath198 ( dash ) , and @xmath183 ( solid line ) erg s@xmath184 . @xmath85 and @xmath55 are the physical and dimensionless distance , respectively .
the power - law frequency spectrum has index @xmath187 .
the redshift is taken to be @xmath188 .
the mesh in the numerical simulation is the same as that in figure [ fig3 ] .
, title="fig:",width=226 ] ( left panel ) and @xmath232 ( right panel ) at time @xmath185 myrs for sources with intensity @xmath249 ( dash dot dot ) , @xmath211 ( dash dot ) , @xmath198 ( dash ) , and @xmath183 ( solid line ) erg s@xmath184 . @xmath85 and @xmath55 are the physical and dimensionless distance , respectively .
the power - law frequency spectrum has index @xmath187 .
the redshift is taken to be @xmath188 .
the mesh in the numerical simulation is the same as that in figure [ fig3 ] .
, title="fig:",width=226 ] the formation of the pre - heated layer with high @xmath12 and high @xmath74 is due to the lack of soft photons beyond the i - front , while hard photons are still abundant in that region . in other words , the energy spectrum of the photons
is significantly hardened around the i - front .
we now directly demonstrate the evolution of the photon frequency spectrum , as our code can effectively reveal the evolution of radiations in the @xmath22-space .
the left panel of figure [ fig6 ] gives the frequency spectra of 1 . ) source @xmath256 erg s@xmath184 at time @xmath185 myrs and physical distance @xmath257 , 0.11 , and 0.14 mpc , and 2 . )
source @xmath258 erg s@xmath184 at time @xmath185 myrs and physical distance @xmath259 and 0.24 mpc .
we can see a significant @xmath85-dependence of the frequency spectrum . at a small @xmath85 the spectra are still almost the same as the original power - law spectrum @xmath260 with @xmath187 , while at large @xmath85 , they significantly deviate from the original power - law .
all photons of @xmath261 , are exhausted within @xmath262 mpc . at high frequency @xmath263 ,
the spectra are still of power - law with @xmath187 , while at @xmath264 they are substantially dropped .
it shows a peak at @xmath265 , and looks like a spectrum with self - absorption .
however , unlike a self - absorption spectrum , the position of the peak is not fixed in the frequency - space , it moves to higher frequency with time .
namely , the photon energy spectrum is harder at later time .
vs. @xmath266 at time @xmath185 myrs . 1 . ) left panel : for the source @xmath267 erg s@xmath184 , at physical distance @xmath257 ( dash dot ) , 0.11 ( dash ) and 0.14 ( solid line ) mpc ; and 2 . ) right panel : for the source @xmath258 erg s@xmath184 at physical distance @xmath259 ( dash ) and 0.24 ( solid line ) mpc .
redshift is @xmath188 .
the mesh in the numerical simulation is the same as that in figure [ fig3 ] .
, title="fig:",width=226 ] vs. @xmath266 at time @xmath185 myrs . 1 . )
left panel : for the source @xmath267 erg s@xmath184 , at physical distance @xmath257 ( dash dot ) , 0.11 ( dash ) and 0.14 ( solid line ) mpc ; and 2 . ) right panel : for the source @xmath258 erg s@xmath184 at physical distance @xmath259 ( dash ) and 0.24 ( solid line ) mpc .
redshift is @xmath188 .
the mesh in the numerical simulation is the same as that in figure [ fig3 ] .
, title="fig:",width=226 ] the spectral hardening can be measured by the index of power law defined as @xmath268 figure [ fig7 ] plots @xmath14 vs. @xmath22 for the frequency spectra of figure [ fig6 ] . obviously , in the band @xmath269 , @xmath14 becomes smaller at larger @xmath85 .
both figures [ fig6 ] and [ fig7 ] show that the frequency spectra of photons depend strongly on @xmath17 and @xmath85 when @xmath17 and @xmath85 are close to or inside the pre - heating layer . vs. @xmath266 for the source @xmath258 erg s@xmath184 at physical distance @xmath259 ( dash ) and 0.24 ( solid line ) mpc .
redshift is @xmath188 .
the mesh in the numerical simulation is the same as that in figure [ fig3 ] .
, width=226 ]
we have described weno scheme which is able to solve the phase - space distribution function of photons in an isolated ionized patch around individual source in the early stage of reionization .
this algorithm can produce robust results for the propagation of the i- and t - fronts .
it can also give stable results for the time - dependent distribution of the small fraction of neutral hydrogen within the i - front .
we have developed the method to deal with the stiffness of rate equations in time integration .
consequently , the computational speed is acceptable .
this algorithm can be applied to problems with a wide range of intensity of sources , from @xmath270 to 10@xmath271 erg s@xmath184 with a power law spectrum at redshift @xmath188 . since this algorithm treat the frequency space as well as physical space , it can also be used for sources with other frequency spectrum .
we have not considered helium and secondary effects in the atomic processes yet , however the algorithm has no practical difficulty to include these factors .
we show that a common feature of the uv photon sources in the reionization epoch is to form a preheating region . in the first stage the i- and t - fronts in baryon gas are coincident , and propagate with the speed about the same as the speed of light .
when the frequency spectrum of the uv photons is hardened , the evolution enters into the second stage .
the propagation speeds of both the ionizing and heating fronts are less than that of light , but the t - front is always moving faster than the ionizing front . in the spherical shell between the i- and t - fronts , the kinetic temperature of gas can be as large as @xmath272 , while atoms are almost neural . obviously , the shell would be of interest in the search for the 21 cm emission from the reionization epoch .
the details of the preheating region are sensitively dependent on the parameters of the problems . for instance
, the radius and thickness of the preheated shell do not show a scaling relation on their dependence of the source intensity .
therefore , an algorithm , which can properly handle models with various parameters is necessary . using these results
, one can make comments on some numerical solvers used for the radiative transfer equations .
several solvers are based on the approximation of omitting the time - derivative term of the radiative transfer equation ( nakamoto et al . 2001 ; cen 2002 ; razoumov et al .
2002 ; rijkhorst et al .
2005 ; susa 2006 ) .
they calculate the static ionization field for a given uniform or non - uniform density fields of cosmic baryon matter .
these codes essentially are generalizing the static solution eq.([eq1 ] ) to the case of inhomogeneous background distribution of baryon matter , and multiple sources .
such an approximation is useful to calculate the ionization field for each given inhomogeneous mass field of baryon matter , but it will no longer be a good approximation when the retardation effect due to photon propagation is important . for the problem of ionization profiles around a point source ,
the retardation of photon is not always negligible .
for instance , one finds the following parameters have been used in the model : the ionization and heating at comoving distance @xmath273 h@xmath184 mpc from a point source of age @xmath274 yrs at redshift @xmath275 ( e.g. cen , 2006 ) .
these parameters already violated the retardation constraint @xmath276 .
the retardation effect can be seen from the time - dependence of the i - front , @xmath277 .
if the retardation effect is negligible , an analytical solution of @xmath277 is given by e.g. whalen & norman ( 2006 ) @xmath278^{1/3},\ ] ] where @xmath279 is the static radius of the strmgren sphere given by eq.(2 ) , and @xmath280 is recombination time scale . according to eq.([shuadd61 ] ) , the time scale of the @xmath281 evolution is @xmath282 , which is independent of the source intensity @xmath30 .
that is , @xmath283 only if @xmath284 , regardless @xmath7
. however , figure [ fig4 ] shows that time scale of the @xmath236 evolution is @xmath7-dependent .
this result also holds if we replace @xmath236 by , e.g. @xmath285 , and is true for a monochromatic source too . considering the retardation effect , an analytical solution of @xmath277
is approximately given by the following algebraic equation ( white et al .
2003 ) @xmath286^{1/3}.\ ] ] according to eq.([shuadd62 ] ) , when @xmath287 , the speed of the i - front , @xmath288 , and when @xmath289 , @xmath290 .
it is qualitatively consistent with @xmath242 shown in figure [ fig4 ] .
first , the time scale of the @xmath236 evolution is approximately @xmath291 , the larger the @xmath7 , the longer the @xmath292 .
second , @xmath293 decreases with @xmath17 by a power law @xmath294 when @xmath17 is large .
however , the power law index @xmath295 .
this is expected , because eq.([shuadd62 ] ) does not include the effect of recombination , which leads to a slower decrease than @xmath242 .
a common assumption used in eqs.([shuadd61 ] ) and ( [ shuadd62 ] ) is that the distribution @xmath199 is described by a step function like eq.([eq1 ] ) . obviously , it ignores the neutral hydrogen hi probably remained within @xmath281 .
the tiny fraction of hi may also be hardly calculated well by monte - carlo codes ( ciardi et al .
2001 ; maselli et al .
2003 ) , which yield large numerical errors due to possion shot noise .
the weno algorithm has been successfully applied to kinetic equations of the distribution function in the phase space with one or two spatial dimensions and two or three phase space dimensions with acceptable computational speed ( carrillo et al . 2006 ) .
we believe that it is not difficult to implement the weno algorithm for radiative transfer problems with similar dimensions in the phase space . in our calculation , the evolution the cosmic baryon gas is not tracked by the hydrodynamic equations , but is simply assumed to have a uniform distribution with density @xmath0 .
this treatment would be reasonable if the typical time scale of the relevant hydrodynamic effects are less than that of the i- and t - fronts .
for instance , dynamical effects associated with sonic propagation would be negligible in solving the propagation of the i- and t - fronts .
of course , one can expect that richer results will be yielded if the weno scheme for radiative transfer problems can be incorporated with the euler hydrodynamics .
since the weno scheme for the cosmological hydrodynamical simulation has already been well established , it would be possible to develop a unified radiation / n - body / hydrodynamics code for cosmological problems . * acknowledgments .
* this work is supported in part by the us nsf under the grants ast-0506734 and ast-0507340 .
llf acknowledges support from the national science foundation of china under the grant 10573036 .
for an ionized sphere associated with a point photon source , the radiation transfer ( rt ) equation is ( bernstein , 1988 , qiu et al .
2006 ) @xmath296 where @xmath297 is the specific intensity , @xmath298 the cosmic factor , @xmath299 , @xmath22 the frequency of photon and @xmath300 a unit vector in the direction of photon propagation . in eq.([eqa1 ] ) , we take @xmath301 .
@xmath302 and @xmath23 are , respectively , the absorption and sources of photons . the absorption coefficient of eq.([eqa1 ] )
is given by @xmath303 where the cross section @xmath304 .
\5 . cooling .
since only the recombination cooling is important , we have @xmath314 ^ 2 \\
\nonumber & + & 1.42\times 10^{-27 } t^{1/2}[1-f_{\rm hi}]^2 \\
\nonumber & + & 2.45\times 10^{-21}t^{1/2 } e^{-157809.1/t}(1+t_5^{1/2})^{-1 } ( 1-f_{\rm hi})f_{\rm hi } \\\nonumber & + & 7.5\times 10^{-19 } e^{-118348/t}(1+t_5^{1/2})^{-1 } ( 1-f_{\rm hi})f_{\rm hi}\end{aligned}\ ] ] where @xmath308 .
the terms on the r.h.s . of eq.([eq16 ] ) are , respectively , the recombination cooling , collisional ionization cooling , collisional excitation cooling and bremsstrahlung .
both @xmath48 and @xmath49 are in the unit of ergs @xmath315 s@xmath184 . | we develop a numerical solver for radiative transfer problems based on the weighted essentially nonoscillatory ( weno ) scheme modified with anti - diffusive flux corrections , in order to solve the temperature and ionization profiles around a point source of photons in the reionization epoch .
algorithms for such simulation must be able to handle the following two features : 1 .
the sharp profiles of ionization and temperature at the ionizing front ( i - front ) and the heating front ( t - front ) , and 2 .
the fraction of neutral hydrogen within the ionized sphere is extremely small due to the stiffness of the rate equations of atom processes .
the weno scheme can properly handle these two features , as it has been shown to have high order of accuracy and good convergence in capturing discontinuities and complicated structures in fluid as well as to be significantly superior over piecewise smooth solutions containing discontinuities . with this algorithm
, we show the time - dependence of the preheated shell around a uv photon source . in the first stage
the i - front and t - front are coincident , and propagate with almost the speed of light . in later stage , when the frequency spectrum of uv photons is hardened , the speeds of propagation of the ionizing and heating fronts are both significantly less than the speed of light , and the heating front is always beyond the ionizing front . in the spherical shell between the i- and t - fronts , the igm is heated , while atoms keep almost neutral .
the time scale of the preheated shell evolution is dependent on the intensity of the photon source .
we also find that the details of the pre - heated shell and the distribution of neutral hydrogen remained in the ionized sphere are actually sensitive to the parameters used .
the weno algorithm can provide stable and robust solutions to study these details .
, , , cosmology : theory , gravitation , hydrodynamics , methods : numerical , shock waves 95.30.jx , 07.05.tp , 98.80.-k |
studies of the supernova remnants ( snrs ) in the large magellanic cloud ( lmc ) are essential for elucidating the details of snr evolution , nucleosynthesis , the nature and environments of supernova progenitors , and so on . as probes of the interstellar medium ( ism ) ,
snrs also yield information on the energy balance and chemical composition of their environments , which serves as input to larger questions of galactic chemical evolution and the star formation history of the cloud .
the relative closeness of the lmc plus its well - determined distance ( we use 50 kpc throughout ) means that accurate physical quantities can be derived for individual remnants in the galaxy .
also , due to the generally low interstellar absorption toward the lmc , it is possible to detect x - rays in the important 0.52.5 kev energy band , which includes emission lines from some of the most abundant metals in the universe , such as k - shell transitions of highly - ionized atoms of oxygen , neon , magnesium , silicon and sulfur and l - shell lines of ionized iron .
the solid - state imaging spectrometer ( sis ) onboard the _ advanced satellite for cosmology and astrophysics _ ( ) , with its superior energy resolution compared to previous broadband x - ray spectrometers , gives us the first real opportunity for studying these issues quantitatively in the x - ray regime .
mathewson et al .
( 1983 , 1984 , 1985 ) identified 32 snrs in the lmc on the basis of optical , radio , and x - ray observations .
the remnants range in size from 2 pc to about 100 pc , and therefore span the range of evolutionary phases , from young ejecta - dominated remnants , through middle - aged adiabatic remnants , to old remnants in the radiative phase .
at least 12 of the lmc remnants are bright enough x - ray emitters ( @xmath5 erg s@xmath6 @xmath7 in the 0.154.5 kev band ) , that we can perform detailed spectroscopic study with .
first results have already been reported from on three small lmc remnants , n103b , [email protected] , and [email protected] ( hughes et al .
1995 ) , that show strong emission lines originating from ejecta characteristic of thermonuclear supernovae , i.e. , type ia sne .
for the research described herein , we observed seven middle - aged remnants , n23 , n49 , n63a , dem71 , n132d , [email protected] , and n49b , with and performed detailed spectral analysis of their sis data .
this is all the lmc snrs larger than 5 pc ( radius ) and brighter than @xmath8 ergs s@xmath6 @xmath7 over the soft kev x - ray band , which includes the three brightest thermal x - ray emitting remnants in the lmc ( n132d , n63a , and n49 ) , plus four less luminous ones that cover a range of evolutionary states .
previous x - ray spectroscopic studies of n49 , n63a , n132d , and n49b were reported by clark et al.(1982 ) based on solid state spectrometer ( sss ) data from the .
the sss data clearly showed weak emission lines of mg , si , and s indicative of thermal emission , although the analysis , which used somewhat unphysical two - temperature collisional ionization equilibrium plasma models , left some doubt about the reliability of the derived elemental abundance values .
more recently hwang et al . (
1993 ) analyzed all the available data on n132d using a phenomenological nonequilibrium ionization ( nei ) model and derived abundances of the heavy elements that were considerably lower than the solar values .
both analyses suggest that the bulk of the x - ray emission from these remnants comes from swept - up ism rather than supernova ejecta and , as we show below , our data confirm this finding .
this observational fact gives us confidence that the snr abundances we measure indeed correspond to those of the ism in the lmc .
the elemental abundances of the magellanic clouds are of great importance to extragalactic astronomy . as the nearest major galaxies to our own
, they are well - studied and serve as the stepping stones for studies of more distant systems . in the current chemical composition of the clouds we see at least some of the material returned to the ism by stars of all types through winds and supernovae and , using models of stellar evolution and nucleosynthesis , we can constrain the history of star formation in these galaxies . it
is well known that the clouds have undergone quite different star formation histories from the galaxy , based on their differing chemical properties ( see , for example , tsujimoto et al .
1995 ) . in the past 20 years with the advent of high quality optical and uv spectra ,
the study of the chemical abundances of the lmc has made great progress .
the earlier results of this era ( summarized in dufour 1984 ) showed that the metallicity of lmc regions was significantly less than solar ; the observed oxygen abundance , for example , was only about the solar value .
more recent studies ( luck & lambert 1992 , russell & dopita 1992 , hereafter rd92 ) have , in general , tended to confirm these findings using optical spectra and detailed modeling of regions , snrs , cepheids , and supergiant stars in the lmc .
however , even though considerable effort has been expended in this area of research over the past decade , there remains considerable room for improvement .
in particular , certain critical elements , such as mg and si , are not easily studied in the optical and uv bands due to a paucity of appropriate atomic transitions .
the work we present here represents the first systematic attempt to measure the gas - phase lmc abundances in a new , independent manner using x - ray spectroscopy of snrs .
our spectral analysis also allows us , for the first time , to derive accurate values for the temperature , ionization timescale , and intrinsic x - ray emissivity of the hot gas in lmc snrs .
when combined with their physical radii , which are accurately known because of the well - established distance to the lmc , we have four independent observational constraints on the dynamical evolution .
only three of these quantities are necessary to fully specify the evolutionary state according to the sedov ( 1959 ) similarity solution , so in fact the observational data result in an overconstrained system .
this allows us to go beyond merely assuming the sedov solution and instead allows to actually establish whether or not the sedov model is a correct representation for the evolution of the snrs in our sample . in the following
we present our analysis of the selected lmc snr sample . in
2 we briefly present the observational details .
in contrast to the phenomenological models used in previous x - ray studies of lmc snrs , we have developed realistic self - consistent models that incorporate the dynamical evolution of remnants .
section 3 provides a description of the spectral model we developed and compares our calculations to similar published ones .
results of the sedov model fits , as well as considerations of the applicability of the model , are given in 4 .
discussions of the derived abundances are in 5 . in 6 we examine the high energy x - ray emission ( @xmath43 kev ) of the remnant sample .
we comment on individual remnants in 7 and summarize in the final section .
has four identical x - ray telescopes ( xrt ) and four focal plane detectors .
two of the focal plane instruments are solid - state imaging spectrometers ( sis0 and sis1 ; ccd cameras ) and the other two are gas imaging spectrometers ( gis2 and gis3 ; gas scintillation proportional counters ) . a summary of the observatory and its complement of instruments has been given by tanaka , inoue , & holt ( 1994 ) .
the log of observations for our remnant sample is given in table 1 .
all but one of the remnants were observed during the performance verification phase or the first guest observer cycle .
the seven remnants in our sample were all found to have soft spectra ( mean electron temperatures @xmath9 kev ) , so we used only the sis data for the present spectral analysis .
the sis was operated in 1-ccd faint mode for all observations .
the data were corrected for dark - frame error , the echo effect , and charge - transfer inefficiency and hot and flickering pixels were removed using standard procedures .
data were rejected during orbital times of high background ( south atlantic anomaly passages ) or when the geomagnetic cutoff rigidity was less than 6 gevc@xmath6 . in order to avoid contamination due to light leaks through the optical blocking filters , we excluded all data taken when the satellite viewing direction was 10@xmath10 or less from the bright rim of the earth .
x - ray spectra were extracted from circular regions of radius @xmath11 centered on each snr with the exception of n49 and n49b . due to their close relative proximity ( n49 and n49b
are separated on the sky by merely @xmath12 ) , the spectra of these two remnants were obtained from @xmath13 radius regions . in all cases background spectra
were constructed from source - free regions in the sis field of view of the same observation .
we show the background - subtracted sis0 spectra of our remnant sample in figure [ fig1 ] .
the x - ray spectra in figure [ fig1 ] are all clearly thermal in nature , showing characteristic k - shell line emission from highly ionized atoms of oxygen , neon , magnesium , silicon , and sulfur as well as a broad blend of iron l - shell emission near 1 kev .
the spectra of the different snrs are remarkably similar to each other , while they differ sharply from the young ejecta - dominated remnants which we studied in our earlier work on the lmc snrs ( hughes et al .
in fact , whereas the spectra of the ejecta - dominated snrs appear qualitatively more consistent with an abundance distribution corresponding to type ia sn ejecta , the seven remnants in the current study are more consistent with the roughly solar abundances of the lmc . as our first approach to a quantitative analysis of these spectra we used a phenomenological model , free from any specific dynamical model of snr evolution , for including nei effects in the spectral fitting ( hayashi et al .
1995 ) . in this approach
the x - ray - emitting plasma is assumed to have been shock heated some time ago @xmath14 to a constant temperature @xmath15 .
the ionization timescale , which is the parameter characterizing deviations from ionization equilibrium , is given by @xmath16 , where @xmath17 is the mean electron density .
in simple terms these parameters are determined from the data in the following manner : @xmath15 is largely constrained by the shape of the continuum emission while @xmath16 is constrained by the emission line ratios ( principally the ratio of helium - like to hydrogen - like k@xmath18 lines ) .
in addition to the temperature and ionization timescale , the elemental abundances , an emission measure , and the line - of - sight absorption column density were required to completely specify the model .
the code is based on the nei model from hughes & helfand ( 1985 ) ; we use the implementation described in hughes & singh ( 1994 ) .
no remnant could be adequately described by a single - component nei plasma model , i.e. , with one @xmath15 and @xmath16 for all species . in particular
, the degree of ionization appeared to vary from element to element .
similar discrepancies of this kind were found in fits of the phenomenological nei plasma model to the data of the young oxygen - rich snr [email protected] in the small magellanic cloud ( hayashi et al .
we therefore modified our fits to the lmc snrs to allow for variable ionization timescales for the different elemental species displaying prominent line emission in the observed band , while at the same time retaining a single value of @xmath15 .
this model provided improved fits , the results of which allowed us to estimate the quantity of x - ray emitting plasma in the various snrs . assuming a constant density shell filling of the entire volume , using the fitted elemental abundances ( which were all substantially less than solar , hayashi et al .
1995 ) , and taking the remnant radii from mathewson et al .
( 1983 ) ( see also table 2 ) , we obtained masses ranging from several tens to several hundred solar masses .
clearly these are evolved remnants that have swept up a large quantity of ism .
nevertheless , the apparent dependence of @xmath16 on elemental species as implied by this phenomenological nei model , makes us question its physical plausibility .
taken at face value , one interpretation of the ionization timescale variation might be that the various elemental species are in spatially disjoint zones that have undergone differing thermodynamic histories . since the amount and approximate composition of the x - ray emitting plasma are consistent with being dominated by ism material ( which should be chemically homogeneous ) , rather than sn ejecta ( which may not be ) , we are confident that this explanation can be rejected .
an alternate and indeed more physically plausible explanation is that the x - ray emitting plasma is chemically homogeneous , but contains a broad range of temperatures and ionization timescales .
this situation would arise as a natural consequence of incorporating realistic models for the dynamical evolution of snrs , since our spectra are integrated over the entire x - ray emission from each remnant .
apparent variation of the average value of @xmath16 with composition , for example , would then arise as a result of the strong dependence of ionization , recombination , and excitation rates , as well as other atomic physics quantities , on elemental species . in effect , each species would tend to pick out that value of @xmath16 which produces a maximum of intensity in the relevant emission lines .
some support for this scenario comes from the fact that , at least for n132d , a better fit to the data was obtained using a two component phenomenological nei model that contained emission from gas at both @xmath190.5 kev and @xmath192 kev ( hayashi 1997 ) ( see [ n132d ] below ) .
we consider the sedov ( 1959 ) similarity solution for an adiabatic blast wave expanding into a uniform , homogeneous ism as the description for the dynamical evolution of our snrs .
as is well known this phase of evolution begins when the mass of swept - up ism grows to be much more than the mass of sn ejecta ( so that the ejecta are dynamically insignificant ) ( gull 1973 ) and it lasts until radiative cooling becomes important , which occurs roughly when the shock temperature drops below @xmath19@xmath20 k ( cox 1972 ) .
based on our estimates of the swept - up mass from the phenomenological model fits , the seven lmc snrs in the sample are all old enough that they satisfy the first constraint and thus are likely to be in the adiabatic phase of evolution . in [ results ] below
we show that the remnants are not so old that radiative cooling has affected their dynamical evolution , and so they also satisfy the second condition . in order to calculate the emergent nei x - ray spectrum ,
the time history of the electron density and temperature throughout the model remnant is required .
for this calculation we divided the remnant into a number of spherically symmetric shells spaced linearly in the nondimensional variable @xmath21 , where @xmath22 and @xmath23 are the velocity and radial position of a specific shell , respectively , and @xmath14 is the remnant s age .
this choice of integration variable ensures a fine spacing near the shock front where the emission measure of the remnant is the largest and the time variation of the relevant quantities is the most rapid . for simplicity , the time integration was constrained so that on successive time steps a single new spatial shell was shock heated and incorporated into the remnant .
consequently the more finely the remnant was divided spatially , the shorter was the time step . based on some trial studies with the spectrum of n132d
, it was found that 30 shells provided an optimal compromise between execution speed and calculational accuracy and so this value was adopted for the analysis .
the temperature and density of each interior shell at each time step was determined from the analytical expression for the interior variation of these quantities ( sedov 1959 ) .
it was assumed that the composition was spatially uniform and that hydrogen and helium were fully ionized .
the mean molecular weight per particle was also assumed constant across the remnant , which is a good approximation for the low metallicity environment of the lmc , where the gas is dominated by hydrogen and helium . with the time dependence of the electron density and temperature
now fully specified , the ionization rate equations were integrated using the matrix diagonalization method ( e.g. , masai 1984 ; hughes & helfand 1985 ) to determine the final ( observed ) ionization state as a function of position .
then the radial run of temperature , coupled with the ionization state , was used to calculate the total emergent x - ray spectrum summed over the entire model snr .
we used the ionization and recombination rate coefficients and the x - ray emission code of masai ( 1984 ) .
the unshocked ism was assumed to be in ionization equilibrium at a temperature of @xmath24 k , but the final results are very insensistive to the precise value of this quantity .
again experiments with the spectrum of n132d found no significant difference in the best - fit parameters when the temperature defining the initial equilibrium ionization fractions was varied from @xmath25 k to @xmath20 k. this is because ionization of the first few ion states proceeds very rapidly ; most of the time is spent in the helium- and hydrogen - like ionic stages for the medium @xmath26 atomic species that are important in our analysis ( i.e. , the abundant elements from oxygen to sulfur , see below ) .
emission from thirteen elements ( h , he , c , n , o , ne , mg , si , s , ar , ca , fe , and ni ) is included in our model .
the abundances of helium and carbon can not be determined from our data since there are no significant emission lines from these species in the x - ray band .
furthermore , we found that the abundances of nitrogen , argon , calcium , and nickel were not well constrained by the data .
consequently we fixed the abundances of these elements to the lmc values determined by rd92 : he 10.94 , c 8.04 , n 7.14 , ar 6.29 , ca 5.89 , and ni 6.04 , where we quote values on a scale of @xmath27 .
the baseline abundances we assume for the other species are o 8.82 , ne 7.92 , mg 7.42 , si 7.52 , s 7.2 , and fe 7.6 , which are the cosmic values from allen ( 1973 ) .
the precise values of the carbon , nitrogen , argon , calcium , and nickel abundances have virtually no effect on our measurements of the other elemental abundances , unless the listed species are extremely overabundant relative to solar , which is unlikely , again given the low metallicity of the lmc .
the helium abundance does have some impact on the absolute values of our derived abundances , since a significant fraction of the continuum emission ( roughly 30% at 1 kev ) is provided by helium . roughly speaking a 10% increase ( decrease ) in the helium abundance
translates into a 5 - 10% increase ( decrease ) in the derived abundances of the other elements . on the other hand ,
the relative abundance values we derive are affected to a much lesser extent by changes in the helium abundance .
the parameters of the model are the shock temperature ( @xmath28 ) , the ionization timescale ( @xmath29 , where @xmath30 is the preshock ism number density of hydrogen atoms and @xmath31 is the ionization age of the remnant ) , the abundances of the elements listed above , and a normalization factor ( @xmath32 , where @xmath33 and @xmath34 are the angular radius and distance of the remnant , respectively ) .
the quantities @xmath28 and @xmath29 are constrained by the data in the same manner as described above for the phenomenological nei model and the normalization factor is essentially determined from the x - ray flux .
when the radius of each remnant is included , we have four independent observational constraints on the three independent parameters of the sedov model , namely age , ambient ism density , and initial explosion energy .
this overconstrained system allows us to carry out an internal check on whether or not the sedov solution is a correct description of the snrs in our sample . in effect
what we do is to explore the consistency between the age derived from the dynamics of the sedov model ( which depends on @xmath28 and @xmath33 ) and that derived from the fitted ionization timescale ( converted to an age using @xmath30 derived from the normalization factor ) .
one complication is that none of the remnants in our sample displays the ideal , spherical symmetry expected of the sedov model , so an additional multiplicative factor is needed to account for the fraction of volume within the radius of the remnant that is filled with x - ray emitting gas when the fitted normalization factor is converted to a mean density .
the last parameter in the spectral fits is the line - of - sight absorbing column density @xmath35 .
we use the cross sections of morrison & mccammon ( 1983 ) .
the precise relationship between the amount of energy imparted to ions and electrons at supernova shock fronts is still one of the major unsolved issues in the physics of strong collisionless shocks ( shklovsky 1968 ) .
whether the electrons attain the same mean thermal velocity behind the shock front as the ions , in which case they will have a considerably lower temperature , or very rapidly equilibrate with the ions , as proposed by mckee ( 1974 ) , is still not entirely clear .
therefore we have constructed two classes of models that encompass the plausible extremes of these possible situations .
the first is a full equilibration model , where the electron and ion temperatures are assumed to be equal .
the other model assumes that the electrons and ions attain the same mean thermal velocity at the shock and then share energy solely through coulomb collisions ( see spitzer 1978 ) and that , consequently , the electron temperature lags the ion temperature ( see , for example , itoh 1978 ; cox & anderson 1982 ) .
however , as it turns out for our sample of lmc remnants , with the possible exception of the ionization timescale , the fitted parameters do not depend strongly on which model is assumed and , in particular , the derived abundances are almost the same in the two models . as part of our model verification we computed a number of simple diagnostic quantities and compared our results with those from the similar sedov - phase nei models of hamilton , sarazin , & chevalier ( 1983 ) .
we found reasonable agreement ( within a factor @xmath36 ) between the two models for the continuum flux at 2 kev , the he - like k@xmath18 line flux , the fe - l line flux , and the x - ray luminosity above 0.1 kev .
our model gives a slightly higher ionization state than the hamilton et al .
model , which is most likely due to the difference in the rate coefficients used by the models .
in figure [ fig1 ] we show the sis spectra and the best - fit nei spectral models for all the remnants in our sample .
sedov models assuming full electron - ion temperature equilibration are plotted ; the other set of models assuming coulomb temperature equilibration would be indistinguishable from these . for display purposes only we averaged the sis0 and sis1 data and corresponding best - fit spectral models . for the actual analysis , in all cases but one , both the sis0 and sis1 spectra were fitted jointly .
the one exception is n63a , where there was a pronounced gain discrepancy for the sis1 data . rather than introducing an _ ad hoc
_ gain correction we decided to use only the sis0 data for this snr . because of its high flux and reasonably long exposure ( see table 1 ) , n63a has the best statistics of any remnant in the sample and so the loss of the sis1 data does not seriously impede our ability to determine its spectral parameters . in all cases
the sis data were rebinned so that each spectral channel contained at least 20 counts after background subtraction .
this allows us to use @xmath37 as the figure - of - merit function for assessing goodness of fit and determining best fit quantities and confidence levels . in table 2
, we give the best - fit values and 90% confidence level statistical uncertainties for the spectral parameters and elemental abundances for both the full equilibration model and the model assuming energy exchange between electron and ions through coulomb collisions . from a statistical point of view , the reduced @xmath37 values given in table 2 are formally unacceptable .
however , systematic errors in the experimental data due to calibration uncertainties , for example , have not been included , nor have we included uncertainties due to the atomic physics or the incompleteness of the x - ray line lists used in the spectral models .
this last point is particularly important since the pattern of residuals for all the snrs show significant excesses at photon energies of @xmath190.9 kev and @xmath191.2 kev which are due to missing fe l - shell lines in the models . as a representation of the data
, however , the sedov model does a considerably better job than the phenomenological nei model discussed above ( [ phenomenological ] ) , which was specifically introduced in order to merely reproduce the observed spectra with little astrophysical motivation .
although the phenomenological model requires a larger number of free parameters , the sedov model provides fits with comparable or even better quality . given our present state of knowledge of the instrumental calibration and the atomic physics of x - ray spectral line emission , we believe that the sedov model describes the spectra about as well as can be expected .
our fitted column densities from the spectra agree quite well with estimates of the total column toward each remnant based on the gas in the galaxy ( heiles & cleary 1979 ) and in the lmc ( mcgee & milton 1966 ) .
in particular , the total column of gas is greatest toward n49 , n63a , and n49b ( values of @xmath19@xmath38 atoms @xmath7 ) and is least toward n23 , dem71 , n132d , and [email protected] ( values of @xmath19@xmath39 atoms @xmath7 ) , which correlates quite well with the x - ray results . the quantity of absorbing gas in the galaxy along the line - of - sight to the lmc sets the minimum column density we should expect to see .
this is @xmath40 atoms @xmath7 , which appears to be consistent with the columns derived for dem71 , n23 , and perhaps [email protected] .
the reader will notice that both the fitted shock temperature and ionization timescale are consistently higher for the coulomb equilibration timescales model compared to the full equilibration one , as a direct consequence of the lag of the electron temperature with respect to the ion temperature in the former case .
however , the difference in temperature is small , @xmath4130% , and comparable to the statistical uncertainty in the parameters .
the ionization timescale differs by a much larger amount , up to factors of @xmath193 , the consequences of which we discuss below when we compare the ionization and dynamical ages .
the other fitted parameters , i.e. , column density , normalization and the elemental abundances , also agree to within the statistical errors . we note that the quality of the fits as indicated by the reduced @xmath37 values is nearly the same under the two different assumptions we have made about electron - ion equilibration timescales , which is unfortunate since it means that our study can not use this to discriminate between the two cases . from the fitted normalization of the x - ray spectra ( @xmath42 ) and shock temperature ( @xmath43 )
, we estimate the ism number density of hydrogen ( @xmath30 ) , the sedov dynamical age ( @xmath44 ) , the initial explosion energy ( @xmath45 ) , and the swept - up mass ( @xmath46 ) , using the following equations : @xmath47 @xmath48 @xmath49 @xmath50 where @xmath51 is the solid angle subtended by the remnant shell with respect to the remnant center , and is a factor that is intended to take account of possible incompleteness of the spherical structure .
spherical symmetry ( @xmath52 ) was assumed in order to compute the values of the derived quantities given in table 2 .
the physical radii in table 2 are based on the angular x - ray size from mathewson et al .
( 1983 ) and our assumed distance of 50 kpc to the lmc .
in calculating the swept - up mass we assumed that the mean ionic weight was @xmath53 .
figure [ fig2 ] plots the ambient ism hydrogen number density and derived explosion energy versus radius for the several remnants .
the ambient density is uncorrelated with remnant size for our sample .
there are some potential biases which work against finding snrs in the upper right corner of this figure ( radiative cooling ) or the lower left ( rapid expansion ) , but given the lack of a statistically significant correlation , these issues are moot .
the range of derived @xmath54 values vary over an order of magnitude from a high of @xmath55 erg for n132d to a low of @xmath56 erg for n23 . recall that @xmath54 depends on @xmath57 and that we assumed the maximum value for @xmath51 ( @xmath58 , i.e. , spherical symmetry )
, so the explosion energies we quote are in some sense lower bounds . in particular , n23 , which has the lowest derived @xmath54 value , also is the remnant that has the most asymmetric x - ray image .
uncertainty in the distance to individual snrs due to the thickness of the lmc , for example , is probably not much more than 10% , which translates to a @xmath1925% error in @xmath54 . as for the ambient density ,
we find no statistically significant evidence for a dependence of the derived explosion energy on remnant size .
this is in contrast to optical studies of lmc snrs ( dopita 1979 , danziger & leibowitz 1985 ) that have claimed a rather strong dependence of @xmath54 on remnant size , even for some of the same remnants as in our sample .
these results were never particularly convincing ( at least to us ) since they were based on extrapolating the pressure behind the shock , as estimated from the density - sensitive [ ] lines , to estimate the energy density in the interior of the entire remnant .
blair , kirshner , & chevalier ( 1981 ) have pointed out that the apparent relation between @xmath54 and radius could be explained if the [ ] emitting regions are dominated by magnetic pressure , rather than thermal pressure as assumed otherwise .
our explosion energy values are based on the well constrained observables of shock temperature , radius , and emission measure , and rely on the applicability of the sedov evolutionary model .
whether or not this is justified is the next issue we consider .
one of the assumptions of sedov - phase remnant models is that they are adiabatic , or energy conserving , which implies that radiative cooling is insignificant .
cox ( 1972 ) identified a number of evolutionary phases in the life of an snr as it makes the transition from an adiabatic remnant to a fully radiative one . in his picture , the onset of radiative cooling occurs when the temperature distribution begins to deviate from the adiabatic model ( temperature sag phase ) at an age given by @xmath59 where the cooling coefficient @xmath60 is of order @xmath61 . using @xmath54 and @xmath62 from table 2 , @xmath63 varies from @xmath64 yr for n63a to @xmath65 yr for [email protected] .
for all the remnants in our sample , @xmath63 is at least 50% larger than the dynamical age quoted in table 2 showing that , at least from this point of view , the sedov model is justified . as we mentioned in
[ sedovmodel ] , we can estimate the ionization age ( @xmath31 ) of our remnant sample from the fitted @xmath29 values , by employing the @xmath30 value derived from the emission measure ( eqn .
because of the difference in fitted ionization timescales between the full and coulomb temperature equilibration models , there is a significant difference in the inferred ionization ages as well . in figure [ fig3 ]
we plot the correlation between ionization age ( for both models ) and dynamical age . for all remnants
, the full equilibration model predicts an ionization age that is considerably lower than the dynamical age with differences ranging from @xmath191.3@xmath66 ( for [email protected] ) to @xmath199@xmath66 ( for n49b ) . on the other hand , under the assumption of coulomb equilibration timescales there is excellent agreement ( a difference of @xmath411.2@xmath66 ) between the ionization and dynamical ages for n23 , n49 , dem71 , and [email protected] although the ionization ages of n132d , n63a , and n49b all continue to remain low .
this discrepancy for the latter three remnants can not be resolved by invoking clumping of the x - ray emitting gas , since that would increase our estimate for @xmath30 and consequently lower the inferred ionization ages even further .
similarly , reducing @xmath51 from its spherically symmetric value of @xmath67 would also result in reduction of the ionization ages as well .
we believe that the most likely reason for the discrepancy is that the sedov model has overestimated the dynamical ages of these remnants .
in this case , it is not coincidental that the three remnants with significant differences between their ionization and dynamical ages are also the remnants with unusually high inferred values of initial sn explosion energy : @xmath4@xmath68 erg .
a plausible explanation for both discrepancies is that these remnants exploded within preexisiting low - density cavities in the ism .
the low interior density allows the sn blast wave to propagate rapidly to the cavity wall , where it then encounters denser gas , begins to slow down , and emits copious amounts of x - rays .
this scenario naturally results in a reduction of both the dynamical age and the initial explosion energy from that inferred assuming the sedov model .
note that hughes ( 1987 ) proposed precisely this same model to account for the difference between the sedov dynamical age and the kinematic age determined from the expansion of high velocity oxygen - rich filaments in n132d ( more on this in [ n132d ] below ) .
it is encouraging that our independent estimate of n132d s age based on its ionization timescale confirms this earlier result and it gives us confidence in the identification of n63a and n49b as two more examples of this phenomenon in the lmc .
this turns out to be extremely important , since it allows us to identify these three remnants as the core - collapse supernovae of massive stars whose strong stellar winds modified the surrounding ism .
the four other snrs , n23 , n49 , dem71 , and [email protected] , are fully consistent with the sedov model under the assumption that the electrons and ions do not attain the same temperature at the shock front , but rather exchange energy slowly through coulomb collisions .
this hypothesis appears to be supported by other observations of snrs ; perhaps the best evidence to date comes from the remnant of sn1006 in the galaxy .
based on comparing shock models to the observed far - uv emission lines of the remnant , laming et al.(1996 ) find that the electrons at the shock front in sn1006 can attain a temperature of no more than @xmath1920% of the ion temperature and that even lower equilibration values are prefered by their data .
this is the strongest observational evidence in the literature against full electron - ion equilibration at sn shock fronts .
we close this section by noting that the consistency of these four snrs with the sedov model appears to extend even to the value we find for their average initial explosion energy @xmath69 erg , where the uncertainty comes from the rms object - to - object scatter .
figure [ fig4 ] displays the complete set of abundance results for the seven lmc remnants , where we have chosen to average the fitted abundances ( and errors ) that come from our two different assumptions about the timescale for electron - ion temperature equilibration .
overall , we find abundances that are less than cosmic by factors of 2 to 5 , as expected for interstellar gas in the lmc . at first glance
the different remnants yield rather similar abundances for the same elemental species , but a closer look reveals some notable exceptions that we discuss further below . in order to check for systematic effects that might have biased the fitted abundances we carried out the following correlation analysis .
first we determined the mean metallicity for each remnant by calculating the error - weighted average of the fitted abundance values ( presented in the bottom half of table 2 ) , including both sets of values from the full and coulomb equilibration models .
we searched for correlations between these mean metallicities and the various fitted or derived parameters using the spearman rank - order test .
the correlation coefficents ( @xmath70 ) for mean metallicity versus @xmath35 , @xmath43 , @xmath71 , and @xmath72 are @xmath73 , 0.04 , 0.61 , and 0.21 , respectively , although none of the correlations were statistically significant ( among these the highest significance was @xmath1993% for metallicity versus @xmath71 ) .
this is very encouraging since it shows that there are no obvious systematic biases in the derived abundances that are directly linked to the spectral fits .
likewise there are no statistically significant correlations between mean metallicity and any of the derived parameters @xmath54 , @xmath62 , and @xmath46 .
there are , however , two statistically significant correlations : mean metallicity versus age ( @xmath74 , significance @xmath1999.6% ) and and mean metallicity versus radius ( @xmath75 , significance @xmath1999.97% ) . plots of these latter two correlations , plus the correlation of metallicity versus swept - up mass , are shown in figure [ fig5 ] .
the error bars displayed are estimates based on the weighted rms scatter from the individual abundance measurements of the six species considered .
the correlation in the left panel of figure [ fig5 ] shows that the smaller remnants are somewhat enhanced in metallicity with respect to the larger ones .
in fact , this correlation also holds when the elemental species are considered individually .
specifically , the oxygen , neon , silicon , and iron abundances show correlations with radius that are significant at the 99.97% , 94% , 96% and 95% confidence levels , respectively , while the magnesium and sulfur abundances are essentially uncorrelated with remnant size .
note that these correlations can be seen in figure [ fig4 ] since the data points corresponding to the individual remnants are plotted for each species from left to right in order of remnant size .
regions in the lmc are known to show a significant correlation between distance from the galaxy center and their oxygen abundance based on measurements of [ ] and [ ] from warm photoioinized gas .
the abundance gradient has a value of @xmath76 dex / kpc ( kobulnicky 1997 ) .
we determined the deprojected radial distances of our remnants from the center of the lmc assuming a flat disk geometry , an inclination angle of 27@xmath77 between the normal to the disk and the line - of - sight , position angle of 165@xmath77 , and center at 5@xmath7824@xmath79 , @xmath069@xmath7748@xmath80 .
the remnants in our sample cover galactocentric radial distances from 0.3 kpc to 3.3 kpc , which would suggest a 30% difference in abundance based on the regions .
in fact , the mean metallicity is uncorrelated with distance from the center of the lmc and our data are best described by a model with no gradient in chemical abundance with galactocentric distance .
the interpretation we favor is that the metallicity - radius correlation indicates the presence of newly produced sn ejecta in at least some of the remnants of the sample .
this is not particularly surprising , since we know that small , young galactic remnants like tycho and cas a show very strong and prominent x - ray emission from supernova ejecta , while larger ones , like the cygnus loop and vela , do nt .
furthermore , earlier work by us ( hughes et al .
1995 ) on the x - ray spectra of the three lmc snrs n103b , [email protected] , and [email protected] ( with radii between 3 and 4 pc ) revealed them to be dominanted by ejecta .
it is perhaps a little surprising that the correlation is so good with radius or age but not with swept - up mass .
apropos to this point is the observation that the only remnant out of these seven previously known to contain newly minted sn ejecta is n132d , which has the largest swept - up mass in our sample !
clearly the fate of the metals ejected by a sn is complex , which makes prediction of their emission properties as a function of time rather difficult .
such a study is beyond the scope of our work here and we direct the interested reader to tenorio - tagle ( 1996 ) , where a discussion of some of the relevant issues for type ii supernovae is given .
figure [ fig5 ] puts the level of enrichment ( i.e. , the difference between the average metallicity of the smaller snrs compared to the larger ones ) at a value of @xmath190.1 .
interestingly , this difference is comparable to the amount of metals that models predict should be produced and ejected by certain mass ranges of progenitor stars . for example if we take the nucleosynthetic yields for a progenitor with a main - sequence mass of 13 @xmath81 from thielemann , nomoto , & hashimoto ( 1996 ) and dilute it with 200 @xmath81 of swept - up material ( the median value in our sample ) , then the fractional abundances of o , ne , and mg would increase by @xmath190.1 . in the same way the abundances of si , s , and fe would increase by an amount equal to @xmath190.4 .
the nucleosynthetic yields of the lighter elements , in particular , are a strong function of main sequence mass and , thus , diluting the ejecta from a 25 @xmath81 progenitor with 200 @xmath81 of ism would result in the o abundance increasing by @xmath191 .
the preceding points are intended to indicate the order of magnitude agreement between the amount of enrichment we actually see in the lmc snr sample and the amount that might be expected in the very simplest scenario imaginable for the mixing of ejecta and ism in snrs .
in addition to this statistical evidence for ejecta enrichment , there is direct evidence as well . in particular , we refer the reader to figure [ fig4 ] where the anomalously high iron abundance for dem71 can be clearly seen .
although some other individual abundances are slightly discrepant with respect to their neighbors ( e.g. , the magnesium abundance for n49b ) , only the case of the iron abundance of dem71 is statistically significant .
this will be discussed in more detail below ( [ dem71 ] ) .
another question concerning our abundance measurements that must be addressed is the possible depletion of metals onto dust grains in the interstellar medium .
interstellar gas is believed to be significantly depleted in magnesium , silicon , calcium , and iron ( with abundances roughly an order of magnitude less than solar ) , but only slightly depleted in oxygen and sulfur .
depletions depend on the mean gas density of the environment , in general , growing larger with increasing gas density ( mathis 1990 ) .
an important constraint is that neon and argon , as inert , noble gases , are not expected to form grains and thus should not be depleted in the gas - phase .
our abundance results appear to be in conflict with this picture of the dusty ism .
in fact , the abundance we find for neon is essentially the same as that of magnesium , silicon , and sulfur , while oxygen and iron are no more than about a factor of two less .
vancura et al . (
1994 ) study the effects of interstellar grains on the predicted x - ray and ir emission of supernova shock waves and find that the x - ray spectra of snrs are substantially influenced by the gradual destruction of dust grains , primarily through collisions with protons and he nuclei , in the post - shock flow .
their models are based on a planar shock geometry and do not include the deceleration of the blast wave , so their results are not directly applicable to our measurements . nevertheless , with this as a warning we attempt a comparison .
the fraction of initial mass remaining in grains is a strong function of the swept - up column density behind the shock . as an estimate of the swept - up column for our remnants we use the quantity @xmath82 , which simply assumes that the entire mass of shocked gas in the snr is contained in a uniform density shell of thickness @xmath83 and takes the average value of the swept - up column at the inner and outer edges of the shell .
the smallest value of this quantity among the remnants in our sample is @xmath84 @xmath7 ( for [email protected] ) while the largest is @xmath85 @xmath7 ( for n63a ) .
according to figure 3 in vancura et al .
( 1994 ) the fraction of initial mass that should have remained in grains is 60% for our minimum swept - up column and 20% for the maximum one . for the species magnesium , silicon , and iron , which are supposed to be nearly entirely depleted in the gas - phase ,
then the mean abundances should vary by about a factor of 2.4 over this range in swept - up column .
we just do nt see such large variations in the abundances of these elements as a function of swept - up column density .
furthermore , there is not even a significant correlation between swept - up column and abundance for magnesium and iron , although the correlation for the silcon abundance versus swept - up column density is significant at the 97.4% confidence level .
weighing all the evidence presented in this and the preceding paragraph leads us to conclude that we have not detected evidence for the destruction of swept - up interstellar dust grains on the expected timescales in these lmc snrs .
this suggests that either the gas - phase of the ism is not significantly depleted in metals or the timescale for the destruction of dust grains behind sn shock waves is shorter than previously thought .
we determine the gas - phase abundances of the lmc by averaging , for each elemental species , the fitted abundances from the seven remnants . as before we include both sets of values corresponding to the full and coulomb equilibration models .
the uncertainties quoted come from the weighted rms scatter among the 14 measurements . for sulfur ,
the values from [email protected] and n49b are not included , since these snrs provided only upper limits .
table 3 lists the average abundances and uncertainties expressed as @xmath86 = 12 + \log ( { \rm m / h})$ ] and figure [ fig6 ] plots the abundances , relative to the cosmic values we assumed , versus elemental species .
the third column of table 3 lists the lmc chemical abundances of oxygen , neon , and sulfur determined from uv / optical spectra of regions as summarized in a review by dufour ( 1984 ) .
the agreement between the x - ray snrs and regions for the different elements is impressive .
more recently , rd92 presented the results of a large study to establish a consistent set of global abundances for the large and small magellanic clouds covering the elements from @xmath87 to @xmath88 .
this work was based on spectral analyses of f supergiants for the abundances of the heavier elements ( e.g. , mg , si , and elements with @xmath89 ) ( russell & bessell 1989 ) and the modelling of spectra from regions and snrs for the lower @xmath26 elements ( e.g. , he , o , ne , and s ) ( russell & dopita 1990 ) . our values and the ones from rd92 are also in general agreement with the possible exception of magnesium and silicon , where the rd92 values appear anomalously high compared to the other species , although it should be noted that the abundance of silicon was quoted in rd92 as being highly uncertain .
it is now clear that these species are not anomalously abundant in the ism of the lmc compared to , say , neon and sulfur .
thus the results from x - ray snrs have proved to be especially valuable in filling this important gap in our knowledge of the abundances of the intermediate mass elements and in providing a set of accurate abundance values determined in a consistent manner for elemental species over a broad range of atomic number .
now a few statements about the reliability of our abundance estimates .
the abundances of neon , magnesium , silicon , and sulfur that we derive from the x - ray spectra of snrs are least subject to systematic uncertainty since they come from fits to k - shell emission lines of well understood helium - like and hydrogen - like ionic species .
our iron abundance is based on the l - shell blend at @xmath191 kev , so there is some systematic uncertainty due to inadequate or poorly known atomic physics and the incompleteness of the line lists used in the spectral models .
our oxygen abundance is also somewhat less reliable because the oxygen k - shell lines are near the low energy cutoff of the sis detectors where the efficiency is low and the calibration is less certain .
furthermore , of all the species considered , the oxygen abundance is most sensitive to the precise variation of temperature and ionization through the interior of our sedov models .
for example , over the outer 10% radius of the sedov models ( from which nearly all of the x - ray emission comes ) , the ionization fraction of o viii ( hydrogen - like ) varies by more than an order of magnitude ( decreasing from the edge in toward the center ) . by contract ,
si xiv ( also hydrogen - like ) is the dominant ionic species and varies by less than a factor of 2 over the same radial range .
thus , our models predict that the oxygen emission should be coming from a narrow shell near the edge of the snr , while the silicon emission should be distributed over a larger radial extent .
future higher spatial resolution x - ray observations with will be able to measure these variations and thereby provide more accurate abundance results that are less dependent on the model details .
however , the comparison between our results and previous ones shown in figure [ fig6 ] suggests that the systematic error in the oxygen and iron abundances derived from our work is not much more than @xmath19 0.10.2 , with our values being slightly lower than the optical / uv ones .
x - ray imaging at energies above @xmath194 kev has been one of the most important recent developments in the search for pulsar - powered synchrotron nebulae ( psn ) in supernova remnants .
studies with in the past 2 years have confirmed the existence of , or directly detected , psn in several galactic snrs including kes 73 ( blanton & helfand 1996 ) , [email protected] ( vasisht et al . 1996 ) , w44 ( harrus , hughes , & helfand 1996 ) , cta-1 ( slane et al .
1997 ) , and msh 11 - 6_2 _ ( harrus , hughes , & slane 1998 ) .
the reason for this is simple : above 4 kev thermal emission from shock - heated gas is negligible , while hard power - law emission from a psn remains relatively strong .
the discovery of a psn in a supernova remnant is pivotal , since it conclusively establishes the remnant as the product of a massive - star core - collapse sne .
the only lmc snrs that can , without question , be put in this class are the snrs [email protected] and n157b both of which contain rapidly rotating pulsars with periods of 50.3 ms ( seward , harnden , & helfand 1984 ) and 16.1 ms ( marshall et al .
1998 ) , respectively , and whose x - ray emissions are dominated by the nonthermal power - law from the psn . because of the known distance to the lmc , these psn have well - determined soft ( 0.154.5 kev ) x - ray luminosities of @xmath90 ( [email protected] ) and @xmath91 ( n157b ) .
a factor of @xmath922500 separates the luminosity of [email protected] from that of the faintest known galactic psn which surrounds psr b1853 + 01 in w44 .
it is quite possible that psn with luminosities in this range are present in some of the lmc snrs in our sample even though all are clearly dominated by thermal emission .
however the arcmin imaging capability of , which is adequate for spatially resolving thermal and nonthermal emission in galactic snrs , is insufficient for doing so for lmc snrs , where a 2 pc diameter psn would subtend merely 8@xmath93 .
thus any information on psn from will have to come from the integrated spectra of individual lmc remnants . at this time we do not claim unambiguous detection of nonthermal x - ray emission for any of the 7 snrs in this study , because the sis emission observed above @xmath193 kev is comparable to the fluctuations in background . in order to set quantitative limits , we fitted the high energy portion of the sis spectra to a power - law model ( @xmath94 ) with a fixed photon index of @xmath95 .
for the four fainter snrs ( n23 , dem71 , [email protected] , and n49b ) the band fitted was 310 kev , while for the brighter ones ( n49 , n63a , and n132d ) we used the 410 kev band .
acceptable fits were obtained in all cases .
the 3-@xmath66 upper limits to the flux of this component were then used to calculate unabsorbed luminosities in the soft x - ray band as shown in table 4 .
these luminosity values are in the range expected for psn ( seward & wang 1988 ) . the high limit for n63a is rather interesting since this is one of the remnants , along with n132d and n49b , that is likely to have arisen from a core - collapse supernova and therefore might be expected to contain a psn .
note that the high upper limit value for n132d largely reflects the short exposure on the remnant .
the limits we set for the others are consistent with them _ not _ containing rapidly spinning young pulsars , in agreement with our age estimates assuming sedov evolution .
current theories predict that type ia sne should not produce pulsars , so their remnants should not contain psn .
it is possible that these snrs may show nonthermal x - ray emission from electrons accelerated to high energies at the shock front by the first - order fermi process as observed in sn1006 ( koyama et al .
this emission tends to be steeper spectrum ( @xmath96 to @xmath97 ) and fainter ( sn1006 s soft @xmath98 is @xmath19@xmath99 erg s@xmath6 ) than that of psn and therefore much more difficult to detect in lmc snrs .
the upper limit to the high energy luminosity for dem71 , which is the only securely identified remnant of a type ia sn in the sample , is about an order of magnitude higher than sn1006 s luminosity .
n23 is not a particularly well studied remnant .
chu & kennicutt ( 1988 , hereafter ck88 ) tentatively classify it as the remnant of a massive star sne ( pop i ) based on the high density of ob stars nearby and its proximity to a modest molecular cloud ( cohen et al .
recently banas et al .
( 1997 ) searched for co emission on smaller spatial scales ( @xmath1920@xmath93 ) than the cohen survey , but found nothing significant that could be associated with the remnant .
we note that n23 has one of the lowest column densities in the sample ( a factor of 34 less than n49 and n49b ) , which may make direct association with a molecular cloud unlikely . as the smallest remnant in our sample
, n23 is an important object that supports the statistical evidence for sn ejecta enrichment discussed above .
images of n49 in the optical , x - ray , and radio bands ( vancura et al . 1992 ; dickel et al .
1995 ) are in general agreement as far as the broad picture of its morphology is concerned .
the remnant is brightest in the southeast and trails off in brightness toward the northwest . on smaller scales , however , individual features differ in relative brightness across the electromagnetic spectrum .
the asymmetry and high x - ray and optical flux are due to the enhanced ambient density toward the southeast , where a moderate - sized molecular cloud sits ( banas et al .
the large x - ray absorbing column density we derive , further indicates a high density environment .
this is also true of the mean optical reddening of n49 , @xmath100 ( vancura et al .
1992 ) , which is quite a bit more than the reddening due to the galaxy , @xmath101 ( fitzpatrick 1985 ) .
we note that the x - ray absorbing column density and interstellar reddening toward n49 are in excellent agreement with the ratio of these quantities , @xmath102 atoms @xmath7 mag@xmath6 , given by ryter , cesarsky , & audouze ( 1975 ) . by comparing shock models to their optical and uv spectra of n49 , vancura et al .
( 1992 ) estimate the abundances of several species . for oxygen and sulfur ,
their abundances are some 50% higher than ours , while their iron abundance is in numerical agreement with ours .
the other species in common ( neon , magnesium , and silicon ) agree to within a factor of two , but the optical results are based on lines from only a single ion and are therefore potentially less reliable . since vancura et al .
( 1992 ) quote no abundance uncertainties it is difficult to assess the significance of these differences .
vancura et al . (
1992 ) also develop a self - consistent scenario for the evolution of n49 .
optical emission comes from slow shocks ( velocity range of 40270 ) driven into dense clouds ( 20940 @xmath103 ) distributed in thin sheets that are located at places where the sn blast wave is interacting with the nearby molecular cloud .
x - ray emission comes from a fast shock ( 730 ) propagating through the intercloud medium ( preshock density of 0.90 @xmath103 ) .
they derive a sedov age of roughly 5400 yr by assuming an explosion energy of @xmath104 ergs . in this picture
the prevailing ism was largely unaffected by the progenitor star , in direct contrast to the view espoused by shull et al .
( 1985 ) , where the sn blast wave is interacting with the dense shell of the progenitor star s strmgen sphere .
our results show that the x - ray data on n49 are consistent with sedov evolution in two ways .
first , the initial sn explosion energy we derive from our fitted shock temperature is consistent with the canonical value of @xmath104 ergs .
second the two ages we determine for the remnant , i.e. , the ionization age and the sedov dynamical age , are entirely consistent with each other .
thus the x - ray , optical , and uv data all tend to support the vancura et al .
( 1992 ) picture for the evolution of n49
. there is at least one other interesting aspect of n49 , namely that it is positionally coincident with the 5 march 1979 @xmath105-ray burst event , one of only three known soft @xmath105-ray repeaters ( sgr ) .
rothschild et al .
( 1994 ) claim that the burst position of sgr0525@xmath066 is coincident with an unresolved source at the northern edge of n49 , based on soft x - ray images from the high resolution imager ( hri ) .
our results do not preclude the possibility that this source has a hard power - law spectrum .
in fact , the upper limit we derive for the power - law emission from n49 ( table 4 ) corresponds to an hri count rate of 0.022 s@xmath6 , which is greater than the actual count rate , @xmath106 s@xmath6 , of the unresolved source in the hri image .
in addition , our upper limit to the hard power - law emission implies a luminosity @xmath107@xmath108 ergs s@xmath6 ( 210 kev band ) , which is comparable to the persistent x - ray luminosity of ax1805.7@xmath02025 ( @xmath109 ergs s@xmath6 ; murakami et al .
this source is the counterpart to one of the other sgrs ( sgr1806@xmath020 ) and is believed to be an isolated neutron star powering a radio nebula in the snr [email protected] ( kulkarni et al .
getting back to n49 , dickel et al .
( 1995 ) claim to have found no evidence for a radio or optical counterpart to sgr0525@xmath066 .
they do note that a compact radio feature lies near the quoted position of the unresolved x - ray source , but they dismiss it as being unrelated largely due to a @xmath110 difference in the radio and x - ray positions .
however this conclusion may be premature , since a few arcseconds is roughly the accuracy for absolute position determination by .
it is probably fair to say that the issue of the counterpart to sgr0525@xmath066 has not been completely resolved and awaits further study with .
n63a is the second brightest x - ray emitting snr in the lmc and its derived shock temperature , ambient density , and abundances ( table 2 ) turn out to be very similar to those of the other bright lmc snrs , n49 and n132d .
however , unlike both n49 and n132d , n63a s morphology varies quite a bit across the electromagnetic spectrum . at optical wavelengths
the remnant consists of three bright lobes of emission embedded in the extended diffuse region n63 .
two of the three lobes consist of shock - heated gas while the third is a photoionized region .
the shock - heated gas appears to be of normal composition for the lmc and is expanding at moderate velocities , @xmath19250 km s@xmath6 ( shull 1983 ) . in the radio band n63a
shows a thick asymmetric shell - like structure @xmath1965@xmath93 in diameter with enhancements at the positions of the bright optical features ( dickel et al .
the x - ray image ( mathewson et al . 1983 ) also displays a thick shell of roughly the same size as the radio one but without any obvious enhancements corresponding to features at the other wavelengths .
curiously , the bright optical region is only about the size of the x - ray or radio shells and is displaced westward by about 15@xmath93 from their respective centers . based on its location in an region , as well as its positional coincidence with the ob association ngc 2030 , it is not surprising that this remnant is considered to have had a pop i progenitor ( ck88 ) .
this could also be consistent with our suggestion that n63a exploded within a wind - blown cavity in the ism .
we derive the highest ambient ism density for this remnant , nearly 4 @xmath103 , yet other signs of such a high density environment are not obvious .
for example , unlike both n49 and n132d , there is no evidence for molecular emission near n63a from either the columbia survey of the lmc ( cohen et al .
1988 ) or from higher angular resolution sest data ( israel et al .
it is clear that a lot more work will be needed before we obtain a comprehensive picture of this enigmatic snr and its environment .
there is a class of snrs that show only hydrogen lines in their optical spectra with virtually no emission from collisionally excited forbidden lines , e.g. , [ ] and [ ] , as usually seen from snrs .
tycho s remnant and sn1006 , believed to be from type ia sne , are examples of such so - called balmer - dominated snrs in the galaxy .
there are also four such remnants in the lmc ( tuohy et al . 1982 ) .
two of them , [email protected] , and [email protected] , show strong x - ray emission lines of the metals si , s , ar , ca , and fe ( hughes et al . 1995 ) , which are known to be characteristic of type ia supernova ejecta .
the young inferred ages of these snrs ( @xmath41 1500 yr ) are consistent with the ejecta - dominated nature of their x - ray emission .
dem71 is another lmc balmer - dominated snr that is older and more evolved than the two just discussed .
based on its age , size , and the estimated amount of gas swept - up by the blast wave ( see table 2 ) , dem71 is believed to be in a transition state in which the x - ray emission is changing from being dominated by sn ejecta to being dominated by swept - up interstellar gas . ( the fourth lmc example of this class , [email protected] , is a much weaker x - ray source and is not in the current sample , so it will not be discussed further . )
the balmer - dominated optical spectra can be interpreted rather well in the context of a model in which a high - velocity interstellar shock overtakes at least partially neutral interstellar gas ( chevalier & raymond 1978 ) .
when examined in detail the h@xmath18 line from these remnants shows a distinctive two - component line profile , consisting of narrow and broad components .
the width of the broad component provides a nearly direct measurement of the shock velocity . for dem71 , smith et al .
( 1991 ) determined a probable range for the shock velocity of 300800 km s@xmath6 , corresponding to shock temperatures of 0.110.76 kev .
this range is in very good agreeement with our x - ray - derived value of @xmath111 kev .
it is now reasonably secure that balmer - dominated snrs are the remnants of type ia sne . in light of this point
it is very interesting that dem71 is the only remnant in the current sample that shows direct evidence for enhanced abundances .
as clearly shown in figure [ fig4 ] the iron abundance of dem71 is about a factor of 2 higher than all the others in the sample .
this corresponds to an amount of `` extra '' iron present in dem71 of roughly @xmath112 , which is about 10% of the total amount of iron believed to be ejected by a ia supernovae ( nomoto , thielemann , & yokoi 1984 ) .
this discovery provides further support for a causal connection between balmer - dominated snrs and type ia sne and it highlights the important of such objects for understanding the chemical enrichment of the ism .
follow - up observations with and searches for coronal [ ] @xmath1135303 emission are planned in order to determine the spatial distribution of the excess iron in dem71 and to search for other elemental species that might be enhanced .
n132d is the one of the brightest lmc remnants and is probably the best studied of them all .
its popularity with optical observers is due in part to the fact that it contains high - velocity oxygen - rich ejecta that identifies the remnant as having come from the core - collapse sn of a massive star .
this ejecta lies near the geometric center of the remnant and is surrounded by a larger limb - brightened shell of material with normal lmc abundances ( see references in morse , winkler , & kirshner 1995 ) .
these two components also appear , more or less , in the radio ( dickel & milne 1995 ) and the x - ray ( hughes 1987 ) images of n132d , although individual features when examined in detail do vary with wavelength .
it is not known if the x - ray emitting features near the center of the remnant are oxygen - rich .
one clue , suggesting that they may not be enhanced , was first presented by blair , raymond , & long ( 1994 ) who noted that the x - ray peaks near the center were generally associated with low - velocity , normal composition , optical filaments , rather than with high - velocity oxygen - rich ones .
observations of n132d ( morse et al . 1996 ) provide a clear confirmation of this effect .
the high - velocity oxygen - rich filaments provide an independent estimate of the age of n132d .
the most detailed analysis of the velocity field of n132d has been carried out by morse et al .
( 1995 ) who find an expansion velocity of @xmath191650 for filaments with a mean radius of 5.3 pc , that results in a kinematic age of @xmath193200 yr assuming undecelerated motion .
this value is considerably less than the sedov dynamical age we quote in table 2 .
this discrepancy between the kinematic age and the sedov dynamical age was noted before ( hughes 1987 ) and it led to the suggestion that n132d exploded within a low - density cavity evacuated by the stellar wind and ionizing radiation of the high - mass progenitor . in this scenario ,
the outer shell of emission is the main blast wave that has reached the edge of the cavity and is now interacting with the high density material there .
the high density of the ambient medium can be explained by the presence of a giant molecular cloud associated with n132d ( banas et al .
our work with the spectra provides yet another independent estimate of the age , based on the ionization timescale , which is also considerably less than the sedov dynamical age and supports the cavity scenario .
our derived n132d abundances for the elements from oxygen to iron are higher by , at most , a factor of three than those derived by hwang et al .
( 1993 ) using data from the .
there are several reasons for this discrepancy .
first our sis data are of considerably higher quality ( in all respects : higher sensitivity , lower background , better spectral resolution , more accurate calibration , and coverage of a broader x - ray band ) than the sss data used by hwang et al .
our sedov models provide a more realistic description of the physical conditions in the snr than the single - temperature single - timescale phenomenological nei models used previously .
the fitted column density in our work is about a factor of 2 more than that of hwang et al . , which tends to increase our fitted abundances relative to theirs .
note that both our fitted column density and hwang et al.s are consistent with the total column of neutral hydrogen toward n132d , @xmath114 @xmath7 .
we believe our x - ray determination of the absorbing column density is more accurate because it is based on a more realistic spectral model and because the data are better calibrated in the important band below 1 kev .
additionally , we can point to the excellent correlation we get between the column and the x - ray column for the seven snrs in our sample to support the general accuracy of our results .
another determination of n132d s abundances was recently presented by favata et al .
( 1997 ) using x - ray spectral data from bepposax .
these results are in strong disagreement with the results we present here and thus the discrepancy requires some explanation .
first we note that the column density derived by favata et al .
, @xmath115 @xmath7 ( but with large error ) , is considerably larger than the total column of neutral hydrogen toward n132d quoted above . because of their higher column density
their abundances will be spuriously high .
their two - component nei model is a somewhat better model than the single - component model used by hwang et al .
( 1993 ) , but it is not as realistic as our sedov - phase model .
however , the difference in spectral models does not fully explain the difference in derived abundances .
we also applied a two - component nei model to the sis data of n132d and found best - fit temperatures of 0.50.6 kev ( component a ) and 2.23.5 kev ( component b ) with a ratio of emission measures of 0.050.06 ( b / a ) , values that are similar to favata et al .
however , this two - component nei model still resulted in low values for the abundances using the sis data : @xmath116 for oxygen , @xmath117 for neon , and @xmath118 for iron .
an oxygen abundance as high as 1.2 ( as derived from the bepposax data ) would produce a huge k@xmath18 oxygen line in the sis that we just do nt see .
we believe that the main difference in our results arises from the poorer spectral resolution of the bepposax data compared to the data from the sis .
the abundances are determined from the emission line strengths , so clearly better spectral resolution is important .
however , the derived abundances are sensitive to the ionization timescales which can only be constrained accurately if lines from helium- and hydrogen - like ionic species can be resolved .
we resolve these lines in the sis data , while the bepposax data does not .
we also have an astrophysical objection to the large abundances quoted by favata et al .
if their abundances were correct , it would require that we are seeing all of the enriched ejecta from a massive progenitor .
in particular , according to the favata et al .
results , the amount of `` extra '' mass in oxygen above the amount in the swept - up ism is @xmath194 @xmath81 which is comparable to the entire mass of oxygen ejected by a 25 @xmath81 progenitor ( thielemann , nomoto , & hashimoto 1996 ) .
since we see oxygen - rich filaments in the optical band , at least some of the ejecta have cooled below @xmath19@xmath25 k and therefore can not be contributing to the x - ray emission
. clearly definitive answers to questions about the amount and distribution of enriched x - ray emitting ejecta in n132d must await observations .
this remnant is located in the western edge of the lmc and , like n23 above , is also rather poorly studied .
it has been tentatively classified as the remnant of a type ia sne ( pop ii ) by ck88 due to the lack of nearby ob stars , regions , and co emission .
this interpretation is consistent with the low preshock density we derive for [email protected] ( @xmath119 @xmath103 , the lowest value in the sample ) , although a lack of evidence for an overabundance of iron , as seen for dem71 , is a ( weak ) countervailing factor .
the line - of - sight column density is low , similar to the values found for n23 and dem71 , and probably arises mostly from the interstellar medium of the galaxy .
optically , n49b consists of two sets of bright knots separated on the sky by approximately @xmath120 and embedded in a patchy network of diffuse nebulosity .
its x - ray appearance is entirely different , showing a somewhat limb - brightened , nearly - circular region of emission that encompasses both sets of optical knots .
on the basis of its location within the region dem 181 , ck88 identify n49b with a pop i progenitor , a picture that is consistent with our finding that n49b is the result of an explosion within a preexisting cavity in the ism .
the relatively high column density toward this remnant suggests that its association with molecular emission ( cohen et al .
1988 ) may be more than an accidental projection effect .
we have carried out a systematic analysis of the sis x - ray spectra of seven luminous snrs in the lmc .
our work has resulted in the following conclusions .
( 1 ) : : the spectral data are described well by a nei model that includes the time evolution of density and temperature given by the sedov similarity solution for the dynamical evolution of snrs . this model was a better description of the data than any single or multiple component phenomenological nei model that we tried .
( 2 ) : : the x - ray data provide 4 independent constraints on the 3 independent parameters of the sedov model .
the observational constraints are : radius , shock temperature , emission measure , and ionization timescale .
the parameters of the sedov model are age , initial explosion energy , and ambient ism density .
there is an additional ambiguity in the models that arises from the unknown relationship between the energy imparted to electrons and ions at the shock front .
we model two extreme situations : full equilibration ( in which the electron and ion temperatures are assumed to be equal ) and coulomb equilibration ( in which the electrons gain energy from the ions though coulomb collisions only ) .
( 3 ) : : four of the remnants in the sample , n23 , n49 , dem71 , and [email protected] , are fully consistent with sedov evolution under the assumption of coulomb equilibration . for this group of remnants
the mean explosion energy is @xmath3 ergs in agreement with the canonical value .
the key to discriminating between the full and coulomb equilibration models comes from comparing the sedov dynamical age with the age determined from the ionization timescale .
( 4 ) : : when the derived parameters of the other three snrs , n63a , n132d , and n49b , are examined in detail , they appear to be internally inconsistent .
in particular the age determined by the ionization timescale is considerably less than the sedov dynamical age , regardless of whether one assumes the full or coulomb equilibration models .
furthermore the initial sn explosion energies are large @xmath4@xmath68 ergs .
both of these discrepancies can be explained by invoking a scenario in which the remnants exploded within preexisting cavities or bubbles in the ism and that the x - ray emission we see now comes from the blast wave interacting with the dense material at the cavity wall .
this was previously suggested for n132d to explain the discrepancy between this remnant s sedov age and the kinematic age determined from the expansion of high - velocity oxygen - rich filaments .
our independent estimate of n132d s age from its ionization timescale confirms this earlier result and it gives confidence that n63a and n49b are actually two more examples of this phenomenon .
this result is not particularly surprising since it is precisely the manner in which massive stars are predicted to modify their environment .
( 5 ) : : we find statistical evidence for enrichment by supernova ejecta in the sense that smaller remnants show a somewhat higher mean metallicity than the larger ones . on the other hand , the mean metallicity does not correlate with swept - up mass . in the case of the balmer - dominated snr dem71 , which is likely to be the remnant of a type ia supernova , the derived abundance of iron is about a factor of two larger than the other remnants in the sample .
this corresponds to an excess of iron over the amount in the ism of @xmath190.06 @xmath81 , which is about 10% of the total amount of iron ejected by a type ia sn .
( 6 ) : : all things being considered , however , the middle - aged , evolved snrs in our sample are in general dominated by swept - up ism and so can be used to estimate the mean lmc gas - phase abundances .
we find that the common elements from oxygen to iron have abundances 0.20.4 times solar , consistent with previous results based on optical and uv data , but without the anomalous overabundance of mg and si seen by others .
this work demonstrates the validity of using x - ray spectra of snrs to measure the current chemical composition of interstellar gas and it has the potential to provide a significant new constraint on the chemical evolution and star formation history of the cloud .
( 7 ) : : we set limits on the hard ( @xmath43 kev ) x - ray flux of the remnants that correspond to luminosities of from @xmath121 to @xmath122 in the 0.24.0 kev band .
these limits indicate that the remnants in the sample do not contain very luminous pulsar - powered synchrotron nebulae , as is consistent with our picture of them as evolved middle - aged snrs .
however , the quoted luminosity range is consistent with the possiblity that they contain weaker synchrotron nebulae with pulsars losing energy at a slower rate than is the case in the crab nebula , for example .
the discovery of such pulsar - powered synchrotron nebulae would definitively identify the progenitor as a massive star that underwent a core collapse sn .
we gratefully acknowledge the numerous scientists , engineers , and other members of the team that designed , developed , and built .
we thank t. murakami for allowing us early access to the data on n49 and d. helfand for useful discussions and comments on the manuscript .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
this research was partially supported by nasa grants nag5 - 2684 and nag5 - 0638 and nsf grant int-9308299 to jph .
allen , c. w. 1973 , astrophysical quantities , 3rd edition ( london : the athlone press ) , 31 banas , k. r. , hughes , j. p. , bronfman , l. , & nyman , l .- .
1997 , apj , 480 , 607 blair , w. p. , kirshner , r. p. , & chevalier , r. a. 1981 , apj , 247 , 879 blair , w. p. , raymond , j. c. , & long , k. s. 1994 , apj , 423 , 334 blanton , e. l. , & helfand , d. j. 1996 , apj , 470 , 961 chevalier , r. a. , & raymond r. c. 1978 , , 225 , l27 chu , y .- h .
, & kennicutt , r. c. 1988 , aj , 96 , 1874 ( ck88 ) clark , d. h. , tuohy , i. r. , long , k. s. , szymkowiak , a. e. , dopita , m. a. , mathewson , d. s. , & culhane , j. l. 1982 , apj , 255 , 440 cohen , r. s. , dame , t. m. , garay , g. , montani , j. , rubio , m. , & thaddeus , p. 1988 , apj , 331 , l95 cox , d. p. 1972 ,
apj , 178 , 159 cox , d. p. , & anderson , p. r. 1982 , apj , 253 , 268 .
i. j. , & leibowitz , e. m. 1985 , mnras , 216 , 365 davies , r. d. , elliott , k. h. , & meaburn , j. 1976 , mem .
, 81 , 89 .
dickel , j. r. , chu , y .- h . , gelino , c. , beyer r. , burton , m. g. , milne , d. k. , spyromilio , j. , green , d. a. , wilkinson , c. , & junkes , n. 1995 , apj , 448 , 623 dickel , j. r. , & milne , d. k. 1995 , aj , 109 , 200 dickel , j. r. , milne , d. k. , junkes , n. , & klein , u. 1993 , a&a , 275 , 265 dopita , m. a. 1979 , apjs , 40 , 455 dufour , r. j. 1984 , in iau symp . 108 , structure and evolution of the magellanic clouds , eds . s. van den bergh & k. s. de boer ( dordrecht : reidel ) , 353 favata , f. , vink , j. , parmar , a. n. , kaastra , j. s. , mineo , t. 1997 , aa , 324 , l45 fitzpatrick , e. l. 1985 , apj , 299 , 219 gull , s. f. 1973 , mnras , 161 , 47 hamilton , a. j. s. , sarazin , c. l. , & chevalier , r. a. 1983 , apjs , 51 , 115 harrus , i. , hughes , j. p. , helfand , d. j. 1996 , apj , 464 , l161 harrus , i. , hughes , j. p. , slane , p. o. 1998 , apj , 499 , in press ( astro - ph/9801204 ) hayashi , i. 1997 , ph.d .
thesis , kyoto university .
hayashi , i. , koyama , k. , ozaki , m. , miyata , e. , tsunemi , h. , hughes , j. p. , & petre , r. 1994 , pasj , 46 , l121 hayashi , i. , koyama , k. , hughes , j. p. , & murakami , t. 1995 , in uv and x - ray spectroscopy of astrophysical and laboratory plasmas , ed .
k. yamashita & t. watanabe ( tokyo : universal academy press ) , 319 heiles , c. , & cleary , m. n. 1979 , aust .
j. phys .
astrophys .
suppl . , 47 , 1 henize , k. g. 1956 , , 2 , 315 hughes , j. p. , & helfand , d. j. 1985 , apj , 291 , 544 hughes , j. p. 1987 ,
apj , 314 , 103 hughes , j. p. , hayashi , i. , helfand , d. , hwang , u. , itoh , m. , kirshner , r. , koyama , k. , markert , t. , tsunemi , h. , & woo , j. 1995 , apj , 444 , l81 hughes , j. p. , & singh , k. p. 1994 , apj , 422 , 126 hwang , u. , hughes , j. p. , canizares , c. r. , & markert , t. h. 1993 , apj , 414 , 219 israel , f. p. , et al .
1993 , a&a , 276 , 25 itoh , h. 1978 , pasj , 30 , 489 kobulnicky , c. 1997 , in `` abundance profiles : diagnostic tools for galaxy history '' the proceedings from a workshop at u. laval , october 1997 , in press ( astro - ph/9711077 ) koyama , k. , et al .
1995 , nature , 378 , 255 kulkarni , s. r. , frail , d. a. , kassim , n. e. , murakami , t. , & vasisht , g. 1994 , nature , 368 , 129 laming , j. m. , raymond , j. c. , mclaughlin , b. m. , & blair , w. p. 1996
, apj , 472 , 267 luck , r. e. , & lambert d. l. 1992 , apjs , 79 , 303 marshall , f. , et al .
1998 , iau circular no . 6810 masai , k. 1984 , ap&ss , 98 , 367 mathewson , d. s. , ford , v. l. , dopita , m. a. , tuohy , i. r. , long , k. s. , & helfand d. j. 1983 , apjs , 51 , 345 mathewson , d. s. , ford , v. l. , dopita , m. a. , tuohy , i. r. , mills , b. y. , & turtle a. j. 1984 , apjs , 55 , 189 mathewson , d. s. , ford , v. l. , tuohy , i. r. , mills , b. y. , turtle a. j. , & helfand , d. j. 1985 , apjs , 58 , 197 mathis , j. s. 1990 , ara&a , 28 , 37 mcgee , r. x. , & milton , j. a. 1966 , aust .
j. phys .
, 19 , 343 mckee , c. f. 1974 , apj , 188 , 335 morrison , r. , & mccammon , d. 1983 , apj , 270 , 119 morse , j. a. , blair , w. p. , dopita , m. a. , hughes , j. p. , kirshner , r. p. , long , k. s. , raymond , j. c. , sutherland , r. s. , & winkler , p. f. 1996 , aj , 112 , 509 morse , j. a. , winkler , p. f. , & kirshner , r. p. 1995 , aj , 109 , 2104 murakami , t. , tanaka , y. , kulkarni , s. r. , ogasaka , y. , sonobe , t. , ogawara , y. , aoki , t. , & yoshida , a. 1994 , nature , 368 , 127 nomoto , k. , thielemann , f .- k .
, & yokoi , m. 1984 , apj , 286 , 644 rothschild , r. e. , kulkarni , s. r. , & lingenfelter , r. e. 1994 , nature , 368 , 432 russell , s. c. , & bessell , m. s. 1989 , apjs , 70 , 865 russell , s. c. , & dopita , m. a. 1990 , apjs , 74 , 93 russell , s. c. , & dopita , m. a. 1992 , apj , 384 , 508 ( rd92 ) ryter , c. , cesarsky , c. j. , & audouze , j. 1975 , apj , 198 , 103 sedov , l. i. 1959 , similarity and dimensional methods in mechanics ( new york : academic press ) seward , f. d. , & wang , z .- r .
1988 , apj , 332 , 199 seward , f. d. , harnden , f. r. , & helfand , d. j. 1984 , apj , 287 , l19 shklovsky , i. s. 1968 , supernovae ( new york : wiley - interscience ) shull , p. 1983 , apj , 275 , 592 shull , p. , dyson , j. e. , kahn , f. d. , & west , k. a. 1985 , mnras , 212 , 799 slane , p. o. , seward , f. d. , bandiera , r. , torii , k. , & tsunemi , h. 1997 , apj , 485 , 221 smith , r. c. , kirshner , r. p. , blair , w. p. , & winkler , p. f. 1991 , apj , 375 , 652 spitzer , l. , jr .
1978 , physical processes in the interstellar medium ( new york : wiley - interscience ) tanaka , y. , inoue , h. , & holt , s. s. 1994 , pasj , 46 , l37 tenorio - tagle , g. 1996 , aj , 111 , 1641 thielemann , f .- k . , nomoto , k. , & hashimoto , m. 1996 , apj , 460 , 408 tsujimoto , t. , nomoto , k. , yoshii , y. , hashimoto , m. , yanagida , s. , & thielemann , f .- k .
1995 , mnras , 277 , 945 tuohy , i. r. , dopita , m. a. , mathewson , d. s. , long , k. s. , & helfand , d. j. 1982 , apj , 261 , 473 vancura , o. , blair , w. p. , long , k. s. , and raymond , j. c. 1992 , apj , 394 , 158 vancura , o. , raymond , j. c. , dwek , e. , blair , w. p. , long , k. s. , and foster , s. 1994 , apj , 431 , 188 vasisht , g. , aoki , t. , dotani , t. , kulkarni , s. r. , & nagase , n. 1996 , apj , 456 , l59 l l l l l & other @xmath123 & date of & exposure & source count rate @xmath124 + snr & name & observation & time ( ksec ) & sis0/sis1 ( count s@xmath6 ) + [email protected] & & 95 november 25 & 38.7 & 0.14/0.11 + [email protected] & dem71 & 95 october 8 & 28.7 & 0.50/0.39 + [email protected] & n23 & 93 november 9 & 32.9 & 0.35/0.29 + [email protected] & n49b & 94 october 10 & 37.4 & 0.35/0.25 + [email protected] & n49 & 94 october 10 & 37.4 & 0.59/0.55 + [email protected] & n132d & 93
september 23 & 5.2 & 4.07/3.39 + [email protected] & n63a & 93 november 21 & 18.0 & 3.16/2.62 + + + | we present the results of x - ray spectroscopy of a flux - limited sample of seven middle - aged supernova remnants ( snrs ) in the large magellanic cloud ( lmc ) : n23 , n49 , n63a , dem71 , n132d , [email protected] , and n49b .
we constructed self - consistent nonequilibrium ionization snr models assuming a sedov solution for the dynamical evolution , and then applied the resulting spectral models to the data obtained by the solid - state imaging spectrometer onboard the _ advanced satellite for cosmology and astrophysics_. all the remnants were reasonably well described by the model , allowing us to derive accurate values for their physical parameters , i.e. , ages , densities , initial explosion energies , and metal abundances .
the derived explosion energies vary from @xmath1 erg to @xmath2 erg . a restricted subset of the sample exists for which the ionization and sedov dynamical ages agree quite well under the assumption that the electron and ion temperatures are not fully equilibrated at the shock front ; for these four snrs the mean value of the initial explosion energy is @xmath3 erg .
we show that it is likely that the other three remnants exploded within preexisting cavities in the interstellar medium .
the limits on high energy x - ray emission ( @xmath43 kev ) that we present indicate that these snrs do not contain very luminous pulsar - powered synchrotron nebulae , in general agreement with our picture of them as evolved , middle - aged remnants .
we find statistical evidence for enrichment by supernova ejecta in the sense that smaller remnants show a somewhat higher mean metallicity than the larger ones . in the case of dem71 ,
the putative remnant of a type ia supernova , the derived abundance of iron is about a factor of two larger than the other remnants in the sample .
these things being said , however , the derived abundances are in general dominated by swept - up interstellar material and so we use the snr sample to estimate the mean lmc gas - phase abundances .
we find that the astrophysically common elements from oxygen to iron are less abundant than the solar values by factors of 24 .
overall these results are consistent with previous ones based on optical and uv data , but do not show the anomalous overabundance of magnesium and silicon seen by others .
10 pt |
crystalline materials often have highly singular strain fields at crack tips , dislocations , and grain boundaries that require the accuracy of atomistic modeling only in small regions surrounding these defects .
however , these localized defects interact through long - ranged elastic fields with a much larger region where the strain gradients are sufficiently small to allow accurate and efficient approximation by coarse - grained continuum finite element models .
this has motivated the development of numerical methods that couple atomistic regions with continuum regions to compute problems with length scales that are sufficiently large for accurate and reliable scientific and engineering application @xcite . the quasicontinuum energy ( qce ) @xcite couples an atomistic model to a finite element continuum model based on the cauchy - born strain energy density .
the cauchy - born strain energy density reproduces the atomistic energy density for a uniformly deformed lattice .
the qce atomistic - to - continuum coupling also reproduces the atomistic energy density for a uniformly deformed lattice simply by modifying the volume of the triangles or tetrahedra within the cut - off radius of the boundary of the atomistic region by using the voronoi cell @xcite .
the qce energy remains the most popular quasicontinuum energy since the modification of the volume of the triangles near the interface to reproduce the atomistic energy density can be explicitly achieved and has been implemented in the code @xcite for multi - dimensional problems with many - body potentials .
a quasicontinuum energy is said to be _ patch test consistent _ if it reproduces the zero net forces at each atom for a uniformly deformed lattice .
the fully atomistic and cauchy - born energies are patch test consistent by symmetry , but the symmetry is broken in the qce atomistic - to - continuum interface .
the nonzero forces in the qce atomistic - to - continuum interface of a uniformly deformed lattice are called _ ghost forces _ @xcite . the _ ghost force correction method _ ( gfc ) was developed to improve the accuracy of qce @xcite , and it has been shown to converge to the patch test consistent force - based quasicontinuum method ( qcf ) @xcite .
gfc has been implemented in the code @xcite and further developed to utilize the efficiency of atomistic - to - continuum modeling and continuum mesh _ a posteriori _
adaptivity @xcite .
however , it has been shown that the force - based quasicontinuum method ( qcf ) gives a non - conservative force field @xcite , so the stability of a deformation that satisfies the qcf equations can not be simply determined by checking whether it is a local minimum of a quasicontinuum energy @xcite .
the linearization of the qcf method has also been shown to be indefinite and to have unusual stability properties @xcite which presents a further challenge to the notion of stability and to the development of efficient and reliable iterative methods @xcite .
the potential for more reliable and efficient iterative solution methods and more direct extensions to finite temperature dynamical methods @xcite make energy - based quasicontinuum methods more desirable than force - based quasicontinuum methods .
several quasicontinuum energies have been proposed which satisfy the patch test for a limited class of problems , but a quasicontinuum energy that satisfies the patch test for many - body potential interactions in several space dimensions with general coupling interfaces and coarsening has yet to be developed . the quasi - nonlocal energy ( qnl ) @xcite gives an explicit and implemented algorithm for close range interactions ( up to next - nearest neighbor interactions for a chain and up to fourth - nearest neighbor interactions for a fcc lattice ) by restoring the symmetry of the interactions .
ghost forces remain for nonplanar interfaces and coarsening @xcite .
the geometric consistency approach gives geometric and algebraic conditions to allow finite range interactions , but an implemented algorithm has only been given for low index planar interfaces .
an efficient and implemented algorithm has yet to be given for a general nonplanar interface surrounding a defect .
ghost forces still remain for nonplanar interfaces and coarsening @xcite .
further , the generalization of the geometric consistency approach to three - dimensional problems seems to depend on special decompositions of three - dimensional space into congruent tetrahedra ( for example , the hexagonal lattice is such a special decomposition of two - dimensional space ) .
the qce energy , on the other hand , can be implemented for any decomposition of three - dimensional space into tetrahedra ( see ( 21 ) in @xcite ) , and so can the blended quasicontinuum energy ( bqce ) which we propose below . in fact , a code for the bqce energy can be obtained by simply modifying any qce code and suppressing the coarsening in the blending interface .
we have done that for our two - dimensional qce code @xcite .
the bqce code will then be able to utilize any adaptive atomistic - to - continuum and continuum coarsening features of the qce code .
more recently , generalizations have been given for the explicit extension of the qnl energy to allow finite range interactions @xcite .
the consistent a / c coupling @xcite gives an implemented patch test consistent quasicontinuum energy for two - dimensional problems with general pair potential interactions , atomistic - to - continuum coupling interface , and coarsening .
no generalizations of these or other patch test consistent methods to three - dimensional crystals , multi - lattices , or multi - body potentials are yet known .
since there remain important problems for which no patch test consistent methods are known , we feel it is important to develop general strategies for reducing the error of coupled energies which can be applied to a broad class of methods and problems .
we propose that by blending the atomistic and corresponding cauchy - born continuum energies of qce in an interfacial region with thickness of a small number @xmath0 of blended atoms , a blended quasicontinuum energy ( bqce ) can be developed with the potential to significantly improve the accuracy of the qce energy near lattice instabilities such as dislocation formation and motion .
this blended quasicontinuum energy can be implemented for any problem for which a qce energy has been developed , and we think that most existing codes for qce could be easily modified to implement bqce .
the blended quasicontinuum energy is not patch test consistent , but our error estimates in section [ subsec : modeling ] show that the error due to the ghost force can be reduced significantly by optimizing the blending function over a blending region of small size @xmath7 therefore , we think that the use of bqce is a good strategy for reducing the error of qce when applied to problems for which no patch test consistent methods are known . moreover , patch test consistent quasicontinuum methods tend to be more complicated and more difficult to implement than qce .
thus , even for problems for where a patch test consistent quasicontinuum energy is known , we think that the bqce energy remains an attractive quasicontinuum energy .
the bqce method is similar to many other methods in which atomistic and continuum energies are smoothly blended over an interfacial region , with some features specific to the quasicontinuum setting .
we call such a region a _ blending region _ , and we call such a method a _ blended _ method .
( other authors call the blending region a handshake region " , an interface " , a bridging region " , or an overlap " . )
we call the weights which blend the atomistic and continuum contributions to the energy _ blending functions_. for example , the atc coupling @xcite , the bridging domain method @xcite , and the arlequin method for coupling particle and continuum models @xcite are all energy - based blended methods which are similar to bqce . by contrast ,
in other schemes , the atomistic and continuum models are coupled abruptly across a sharp interface consisting of only a few atoms .
such schemes include the energy - based quasicontinuum ( qce ) method @xcite , the quasinonlocal quasicontinuum ( qnl ) method @xcite , the generalized qnl method @xcite , and the consistent a / c coupling @xcite . in this paper , we present an error analysis of the blended quasicontinuum energy ( bqce ) .
we determine precisely how the error of the bqce method depends on the size of the blending region and how it depends on the blending functions .
we analyze the accuracy of bqce applied to the problem of a one - dimensional chain of atoms with periodic boundary conditions and a next nearest neighbor pair interaction model .
our choice of a one - dimensional analysis allows us to explicitly investigate the accuracy of the bqce energy for deformations up to lattice instability @xcite , which is crucial for the computation of critical strains in lattice statics as well as for the computation of the dynamics of defect nucleation and movement .
we focus on modeling error and do not consider coarse - graining in our analysis . instead
, we assume that the continuum energy is a cauchy - born energy discretized by piecewise linear finite elements with nodes at every atom .
we refer the reader to @xcite for an analysis of coarse - graining for a problem similar to our one - dimensional chain .
in addition , we constrain the displacements of atoms in the blending region to exactly match the continuum displacement field . in some other methods , the displacements of atoms in the blending region are coupled weakly to the continuum displacement field using lagrange multipliers or a penalty method .
this is the case in the arlequin method @xcite , in the bridging domain method @xcite , and in some implementations of the atc coupling @xcite .
computational experiments comparing various methods of weak coupling are given in @xcite .
a numerical analysis of a mixed finite element method for a weakly coupled problem is given in @xcite .
blending can also improve the accuracy of quasicontinuum energies which are more accurate than qce , but may not satisfy the patch test for general interactions , nonplanar interfaces , or coarsening .
we thus also define and analyze a blended generalization of qnl which we call the _ blended qnl _ ( bqnl ) energy .
the bqnl method is an energy - based , blended coupling which passes the patch test for the one - dimensional problem analyzed in this paper , but does not generally satisfy the patch test as described above .
our analysis extends the techniques developed in @xcite for the patch test consistent qnl to obtain precise estimates of the error for bqce , which does not satisfy the patch test .
most importantly , our estimates can be used to optimize the blending function and blending interface size of bqce to obtain accurate solutions up to lattice instabilities .
our error estimate in theorem [ thm : a priori existence for bqce ] shows that the @xmath1 strain error for the bqce method can be reduced by a factor of @xmath4 where @xmath0 is the number of atoms in the blending region .
this result is suggested by the results of @xcite on the decay of the coupling error for qce .
it has been shown that the qce method suffers from a o@xmath5 error in the critical strain at which the lattice loses stability ( which models fracture or the formation of a defect ) @xcite .
we prove in theorem [ thm : aprstab ] that the error in the critical strain of bqce can be reduced by a factor of @xmath6 where @xmath0 is the number of atoms in the blended interface region . in remark
[ rem : ghost force ] , we use our modeling error estimates to derive blending functions for the bqce method which minimize the error due to the ghost force .
we give our _ a priori _ error estimate for bqnl in theorem [ thm : a priori existence for bqnl ] . in theorem
[ thm : a priori existence for bqnl ] , we show that the @xmath1 strain error for the bqnl method can be reduced by a factor of @xmath8 . we prove in theorem [ thm : aprstab ] that the bqnl method does not suffer from a critical strain error .
we begin by presenting a model of a one - dimensional lattice of atoms .
our reference configuration is the integer lattice @xmath9 with lattice spacing @xmath10 .
the admissible deformations are n - periodic , mean - zero displacements of uniformly strained states of @xmath9 .
precisely , for @xmath11 , we let @xmath12 be the set of n - periodic , mean - zero displacements . for @xmath13
we let @xmath14 be the uniformly deformed state with _
macroscopic strain _
@xmath15 , and we let @xmath16 be the set of admissible deformations with fixed macroscopic strain @xmath15 .
we will fix @xmath17 throughout the remainder of the paper , so that the reference length of a period is one , independently of @xmath18 . for notational convenience
, we have let the deformations @xmath19 be functions of @xmath20 instead of @xmath9 even though we still think of @xmath9 as the reference configuration .
we also remark that the displacements " @xmath21 which appear in the definition of @xmath22 are not displacements measured relative to the reference lattice @xmath9 , but are instead displacements relative to the uniformly deformed state @xmath23 . in the language of mechanics
, one would say that our choice of the deformation space @xmath22 means that we have imposed periodic boundary conditions with macroscopic strain @xmath15 . "
periodic boundary conditions are mathematically similar to dirichlet boundary conditions in that constraints are applied to the space of admissible deformations . in our case ,
the constraint is given by the macroscopic strain @xmath15 which models the macroscopic scale stretching or contraction of the chain .
periodic boundary conditions are useful because they eliminate surface effects near the boundaries of the chain . for a deformation @xmath24
we define a _ stored energy per period _ which we call @xmath25 .
our stored energy @xmath25 sums the energies of all the interactions between first nearest neighbors and second nearest neighbors in the lattice .
we compute the energies of interaction using a potential @xmath26\rightarrow{\mathbb r}$ ] which we assume satisfies 1 .
@xmath27 , { \mathbb r});$ ] 2 .
there exists @xmath28 so that @xmath29 for all @xmath30 , and @xmath31 for all @xmath32 3 .
@xmath33 and its derivatives decay at infinity .
examples of commonly used interaction potentials which satisfy the above conditions include the lennard - jones potential and the morse potential .
specifically , we define @xmath34 in , we call the terms of the form @xmath35 _ first nearest neighbor interactions _ , and we call terms of the form @xmath36 _ second nearest neighbor interactions_. the reader will observe that we have scaled the energy by @xmath18 and the interatomic distances by @xmath37 .
this is done for two related reasons .
first , quasicontinuum methods are only needed when when @xmath38 is large ( equivalently , when @xmath18 is small ) . by introducing the scaling above
, we are able to distinguish those parts of the error which are most significant in the practically relevant limit of large @xmath38
. the reader should bear this in mind when interpreting our error estimates in sections [ subsec : modeling ] , [ subsec : stability ] , and [ subsec : a priori existence ] .
second , we desire that in the limit as @xmath18 tends to zero ( equivalently , as @xmath38 tends to infinity ) , the atomistic energy converges to the fully continuum cauchy - born energy .
this is explained in detail in section [ subsec : cauchy born approx ] below . roughly speaking
, this ensures that the atomistic energy is compatible with the continuum energy to which it is coupled in the bqce and bqnl schemes .
the compatibility of the two energies is reflected in the fact that the modeling error of the cauchy - born energy is of order @xmath39 ( see remark [ rem : clarification of modeling ] ) .
now suppose we want to model the response of our lattice to a dead load .
we let @xmath40 denote the algebraic dual of @xmath41 , and we equip the space @xmath41 with the _ @xmath42 inner product _ given by @xmath43 generally speaking , a dead load @xmath44 should be thought of as an element of @xmath40 . however , since @xmath41 is a hilbert space with the inner product , any @xmath45 is represented by some @xmath46 .
that is , for any @xmath45 there exists @xmath46 so that for all @xmath47 , @xmath48 . thus , we will think of dead loads as elements of @xmath41 . the total energy for an atomic chain subject to a dead load @xmath46
is then given by @xmath49 and the equilibrium deformation of the atoms solves the minimization problem : @xmath50 let @xmath51 denote the first variation of the energy @xmath52 , and let @xmath53 denote the second variation .
we will say that a solution @xmath19 of the minimization problem is _ strongly stable _ if @xmath54>0 $ ] for all @xmath55 , and we will call a deformation @xmath56 an _ equilibrium _ if it solves @xmath57 = < f , y > \mbox { for all } u \in \mathcal{u}.\ ] ] we call an energy @xmath58 _ local _ if it can be written in the form @xmath59 where @xmath60 \rightarrow { \mathbb r}$ ] is called a _
strain energy density_. otherwise , we say that an energy is _
nonlocal_. observe that energy is nonlocal .
the development and use of a strain energy density @xmath33 allows coarse - graining in the continuum region by utilizing the full range of algorithms , codes , and analysis developed for the continuum finite element method .
( an alternative quasicontinuum approach uses quadrature to approximate the nonlocal coarse - grained energy without the construction of a strain energy density @xcite .
peridynamics offer another promising fully nonlocal approach to coarse - graining @xcite . )
thus , it is useful to devise local energies which approximate .
the key to developing such energies is to observe that as long as @xmath61 varies slowly between neighboring lattice points , we have @xmath62 if we replace the second nearest neighbor terms @xmath63 by @xmath64 in , we then obtain the _ cauchy - born energy _
@xmath65 which is a local energy commonly used to approximate .
an alternative derivation of the cauchy - born energy is as follows .
suppose that @xmath66 \rightarrow { \mathbb r}$ ] is a @xmath67 deformation of @xmath68 $ ] , and define @xmath69 .
it can be shown that @xmath70 } \psi^{cb}\left ( \frac{dy}{dx}(x ) \right ) dx = : \psi^{cb}(y),\ ] ] where @xmath71 is called the _ cauchy - born strain energy density_. we call the energy @xmath72 the _ fully continuum cauchy - born energy_. we refer the reader to @xcite for a detailed discussion of the convergence of @xmath73 to @xmath74 as @xmath18 goes to zero .
now suppose that @xmath75 is in the lagrange @xmath76 finite element space on @xmath68 $ ] with nodes at the reference position of each atom ; so the nodes are @xmath77 for @xmath78 . in that case
, it is easy to show that @xmath79 thus , we see that the cauchy - born energy @xmath80 is a finite element approximation of the fully continuum cauchy - born energy @xmath72 .
consequently , we will refer to @xmath80 as a continuum " energy .
we say that the energy @xmath80 is not coarse - grained since @xmath80 has the same number of degrees of freedom as the atomistic energy @xmath81 : both depend on the deformed position of each atom .
suppose we want to approximate a solution to the minimization problem which has defects in localized regions but which is smooth throughout most of the lattice .
efficiency requires that most of the lattice be modeled using the cauchy - born energy , while accuracy requires that the defects be treated using the nonlocal energy .
thus , we desire a coupling of the two models .
we distinguish two approaches to deriving coupled energies .
we call these the _ strain - energy based _ approach and _ bond based _ approach . in the strain - energy based approach ,
one first defines an atomistic _ energy per atom_. for our one - dimensional chain , an appropriate energy per atom is @xmath82 we call @xmath83 the _ atomistic energy at atom @xmath84_. observe that @xmath83 is half the total energy of all bonds involving atom @xmath84 , so @xmath85 .
the energy per atom should be interpreted as the energy per length @xmath18 at atom @xmath84 ; thus , @xmath86 is analogous to a continuum strain - energy density . to define a coupled energy we now blend the atomistic energy per atom with the continuum energy
schematically , we choose some blending functions @xmath87 \rightarrow [ 0,1]$ ] and @xmath88 $ ] . then for a deformation @xmath89 \rightarrow { \mathbb r}$ ] with corresponding atomistic deformation @xmath90 , we define @xmath91 } \alpha(x ) \psi^{cb } \left ( \frac{dy}{dx}(x ) \right ) dx + { \epsilon}{\sum_{\xi = 1}^n}\beta_\xi \phi^a_\xi(y^{\epsilon}).\ ] ] in equation , one should imagine that the function @xmath92 is zero in a small region containing any defects , and one throughout the bulk of the lattice .
one should imagine that @xmath93 is one near defects , and zero throughout the bulk of the lattice .
for an abruptly coupled energy , one should imagine that @xmath92 is the characteristic function of the continuum region , and @xmath93 is the characteristic function of the atomistic region . by contrast , for a blended energy , one should imagine that @xmath92 and @xmath93 are more smooth .
the energy - based quasicontinuum ( qce ) method @xcite , the bridging domain method @xcite , and the arlequin method @xcite all take essentially the form .
moreover , although the atc method @xcite was derived in variational form , the equilibrium condition is the euler - lagrange equation of an energy similar to .
we will now explain the qce energy in detail .
let @xmath68 = \mathcal{a } \cup \mathcal{c}$ ] be a partition of the domain @xmath68 $ ] into an atomistic region @xmath94 and a continuum region @xmath95 .
suppose that @xmath75 is in the @xmath76 lagrange finite element space with nodes at every lattice site , and let @xmath96 be the corresponding deformation of the atomistic lattice @xmath9 .
in this special case , the qce energy reduces to @xmath97 we call @xmath98 the _ continuum energy at _ @xmath84 .
our blended energy - based quasicontinuum energy ( bqce ) is based on a blend of the atomistic energy at @xmath84 and the continuum energy at @xmath84 .
let @xmath99 $ ] be a blending function .
for @xmath19 we define @xmath100 we observe that the qce energy is simply the bqce energy whose blending function @xmath101 is the characteristic function of the continuum region
. we will show in theorem [ thm : modeling ] that bqce does not pass the patch test where we recall that a quasicontinuum energy passes the _ patch test _ if there are no forces under uniform strain ; for our one - dimensional problem an energy @xmath102 passes that patch test if @xmath103 for all @xmath104 . we will call a method which passes the patch test _ patch test consistent_. we will now discuss the qnl and bqnl energies in detail .
the atomistic and continuum energies of the bond between the nearest neighbors at reference positions @xmath77 and @xmath105 are taken to be @xmath106 .
this is the same as the energy of that bond in the fully atomistic model .
the atomistic energy of the bond between the second nearest neighbors at reference positions @xmath107 and @xmath105 is taken to be @xmath108 , and the continuum energy of that bond is taken to be @xmath109 . as discussed in section [ subsec : cauchy born approx ] , this choice of continuum bond energy is related to the cauchy - born rule .
the coupled energy is then given by @xmath110 where @xmath68 = \mathcal{a } \cup \mathcal{c}$ ] is a partition of @xmath68 $ ] .
our blended quasinonlocal qc ( bqnl ) energy is based on a smooth blending of the atomistic and continuum bond energies used to define the qnl energy .
let @xmath111 $ ] be a blending function .
we define @xmath112 we remark that the qnl energy is the bqnl energy whose blending function is the characteristic function of the atomistic region @xmath94
. we will now give a simple proof that bqnl passes the patch test .
we discuss the modeling error of bqnl in more detail in section [ subsec : modeling ] .
the patch test consistency of bqnl is a consequence of the following result .
[ lem : bqc is affine closure of qnl ] the set of bqnl energies is the affine hull of the set of qnl energies .
in particular , any bqnl energy may be expressed as an affine combination of qnl energies with different atomistic and continuum regions .
the reader may verify that the set of bqnl energies is an affine space .
thus , to prove the lemma it suffices to express every bqnl energy as an affine linear combination of qnl energies with different atomistic and continuum regions .
let @xmath113 be the bqnl energy with blending function @xmath93 , and let @xmath114 be the qnl energy with atomistic region @xmath115 and continuum region @xmath116 for @xmath117 .
we compute @xmath118 and so we derive @xmath119 then we have @xmath120 this expresses @xmath121 as an affine combination of qnl energies , since we recall that the cauchy - born energy , @xmath80 , is the qnl energy whose continuum region is the entire domain . in light lemma [ lem
: bqc is affine closure of qnl ] , one expects that bqnl will inherit many of the properties of the qnl energy . in particular , since any bqnl energy is an affine combination of qnl energies , and since the qnl energy is patch test consistent @xcite , the bqnl energy passes the patch test .
moreover , we show in remark [ rem : accuracy of critical strain ] that the bqnl energy predicts the critical strain of the atomistic energy as accurately as the qnl method .
[ rem : construction of consistent blended energies ] ( construction of patch test consistent blended methods ) .
lemma [ lem : bqc is affine closure of qnl ] suggests a general method for constructing patch test consistent , blended methods from patch test consistent methods with a sharp interface : one can define a patch test consistent , blended method by taking a convex combination of patch test consistent energies with different atomistic and continuum regions .
of course , it may be that the method so constructed is not practical .
nevertheless , we feel that this observation could be useful in deriving patch test consistent , blended couplings . for our analysis
, it will be convenient to define a single energy which incorporates both bqce and bqnl as special cases .
we will use the _ blended quasicontinuum ( bqc ) energy _
@xmath122 where @xmath123 $ ] are _ blending functions_. we observe that the bqce energy with blending function @xmath101 is the same as the bqc energy with blending functions @xmath124 the bqnl energy with blending function @xmath125 is the bqc energy with blending functions @xmath126 here we collect all the notation and auxiliary theorems which we will use below .
the reader may feel free to skim this section at first and return only as necessary . for differences , we write @xmath127 for means
, we write @xmath128 for @xmath19 , we will think of @xmath129 , @xmath130 , @xmath131 , and @xmath132 as n - periodic functions defined on the reference configuration .
we will also use certain bounds on the potential function @xmath33 and its derivatives : @xmath133 we will denote the convex hull of the set @xmath134 by @xmath135 we define the following norms @xmath136.\end{aligned}\ ] ] correspondingly , we let @xmath137 denote the space @xmath41 equipped with the norm @xmath138 , and we let @xmath139 denote the space @xmath41 equipped with the norm @xmath140
. additionally , let @xmath141 denote the topological dual of the banach space @xmath75 , and let @xmath142 where @xmath143 $ ] with @xmath144 .
we let @xmath145 denote the atomistic region , @xmath146 denote the continuum region , and @xmath147 denote the interface .
we let @xmath148 be the @xmath138 norm taken over the set @xmath149 for @xmath150 .
we denote the closed ball of radius @xmath151 at @xmath152 in @xmath153 by @xmath154 for @xmath153 one of the spaces @xmath137 or @xmath139 .
we will write @xmath155 for the first variation of a differentiable energy functional @xmath156 .
the first variation is a map from @xmath22 into @xmath40 ; we will let @xmath157 $ ] denote the first variation of @xmath156 at the deformation @xmath19 evaluated on the test function @xmath158 .
we will use the letter @xmath21 to denote a test function belonging to @xmath41 throughout the remainder of the paper .
the letter @xmath159 will be used to denote a deformation belonging to @xmath22 .
we warn the reader that in expressions such as @xmath157 $ ] , @xmath21 denotes an arbitrary test function , not the displacement corresponding to the deformation @xmath159 .
similarly , we will let @xmath160 denote the second variation of @xmath156 .
the second variation can be interpreted either as a map from @xmath22 into the space of bilinear forms on @xmath41 , or as a map from @xmath22 into @xmath161 .
we will let @xmath162 $ ] denote the second variation of @xmath156 at @xmath19 evaluated on the test functions @xmath163 .
we will need the following version of the inverse function theorem which appears as lemma 1 in @xcite .
[ thm : ift ] _ ( inverse function theorem ) _ let @xmath153 and @xmath75 be banach spaces , let @xmath134 be an open subset of @xmath153 , and let @xmath164 be a @xmath165 function .
let @xmath166 and suppose : 1 .
@xmath167 , and @xmath168 2 .
@xmath169 3 .
@xmath170 for all @xmath171 with @xmath172 ; 4 .
@xmath173 .
then there exists @xmath174 so that @xmath175 and @xmath176 . in section 5 , we will use theorem [ thm : ift ] to prove _ a priori _ existence with error estimates for the bqce and bqnl energies
. we will apply theorem [ thm : ift ] using bounds on @xmath125 derived from the modeling estimates in section [ subsec : modeling ] and bounds on @xmath177 derived from the stability estimates given in section 4 .
in theorem [ thm : modeling ] , we estimate the @xmath178 norms of the modeling errors of the bqce and bqnl energies .
[ thm : modeling ] let @xmath19 with @xmath179
let @xmath180 be the bqce energy with blending function @xmath101 .
recall from equation that @xmath180 is the bqc energy with blending functions @xmath181 , and @xmath182 .
we have @xmath183 where @xmath184 , @xmath185 for @xmath186 2 .
let @xmath121 be the bqnl energy with blending function @xmath93 .
we have @xmath187 where @xmath184 , @xmath188 for @xmath186 we begin by writing down the first variations of @xmath189 and @xmath81 . for @xmath19 and @xmath190
we have @xmath191 & = { \epsilon}{\sum_{\xi = 1}^n}{\phi}'(y'_\xi)u'_\xi + 2{\alpha_{\xi}}{\phi}'(2y'_\xi)u'_\xi + \beta_\xi { \phi}'(y'_\xi + y'_{\xi + 1})(u'_\xi + u'_{\xi + 1})\\ & = { \epsilon}{\sum_{\xi = 1}^n}\ { { \phi}'(y'_\xi ) + 2{\alpha_{\xi}}{\phi}'(2y'_\xi ) + \beta_{\xi-1}{\phi}'(y'_{\xi-1}+y'_{\xi } ) + \beta_\xi{\phi}'(y'_\xi+y'_{\xi+1 } ) \}u'_\xi . \end{split}\ ] ] the first variation of @xmath81 is a special case of the above .
we have @xmath192 = { \epsilon}{\sum_{\xi = 1}^n}\ { { \phi}'(y'_\xi ) + { \phi}'(y'_{\xi-1 } + y'_{\xi})+{\phi}'(y'_\xi + y'_{\xi + 1 } ) \}u'_\xi.\ ] ] using these formulas we compute the _ modeling error _
@xmath193 : & = { \delta}{\phi}(y)[u]-{\delta}{\phi_{\alpha , \beta}}(y)[u]\\ & = { \epsilon}{\sum_{\xi = 1}^n}\left [ \{(1-\beta_{\xi-1}){\phi}'(y'_{\xi-1}+y'_{\xi})-{\alpha_{\xi}}{\phi}'(2y'_{\xi})\ } + \{(1-\beta_{\xi}){\phi}'(y'_{\xi } + y'_{\xi+1})-{\alpha_{\xi}}{\phi}'(2y'_{\xi})\ } \right ] u'_\xi.\notag \ ] ] now we expand the terms @xmath194 in at @xmath195 using taylor s theorem .
we obtain @xmath196 where @xmath197 , and @xmath198 .
when we substitute and into , we obtain @xmath199 & = { \epsilon}{\sum_{\xi = 1}^n}2 \ { 1 - { \alpha_{\xi}}-{\bar{\beta}}_\xi \ } { \phi}'(2y'_\xi ) u'_\xi \\ & \quad + { \epsilon}\ { ( 1-\beta_\xi){\phi}''(2y'_\xi)y''_\xi -(1-\beta_{\xi-1}){\phi}''(2y'_\xi)y''_{\xi-1 } \ } u'_\xi \\ & \quad + \frac{{\epsilon}^2}{2 } \ { ( 1-\beta_\xi){\phi}'''({\mathcal{o}}_{2,\xi})(y''_\xi)^2 + ( 1-\beta_{\xi-1}){\phi}'''({\mathcal{o}}_{1,\xi})(y''_{\xi-1})^2 \ } u'_\xi . \end{split}\ ] ] expanding the term of order @xmath18 on the right hand side of , we compute @xmath200 = { \epsilon}{\sum_{\xi = 1}^n}2 & \ { 1 - { \alpha_{\xi}}-{\bar{\beta}}_\xi \ } { \phi}'(2y'_\xi ) u'_\xi \\ & + \{-{\epsilon}\delta\beta_\xi{\phi}''(2y'_{\xi})y''_\xi + { \epsilon}^2 ( 1-\beta_{\xi-1 } ) { \phi}''(2y'_{\xi})y'''_{\xi } \ } u'_\xi \\ & + \frac{{\epsilon}^2}{2}\{(1-\beta_\xi){\phi}'''({\mathcal{o}}_{2,\xi})(y''_\xi)^2+(1-\beta_{\xi-1}){\phi}'''({\mathcal{o}}_{1,\xi})(y''_{\xi-1})^2\}u'_\xi . \end{split}\ ] ] we recall from equation that the bqnl energy @xmath121 with blending function @xmath93 is the bqc energy @xmath201 with @xmath202 . in that case
, we compute that for the bqnl energy @xmath121 , the term of order zero in equation vanishes , and so @xmath203 - { \delta}{\phi^{qnl}_{\beta}}(y)[u ] = { \epsilon}{\sum_{\xi = 1}^n}\ { - & { \epsilon}\delta\beta_\xi{\phi}''(2y'_{\xi})y''_\xi + { \epsilon}^2 ( 1-\beta_{\xi-1 } ) { \phi}''(2y'_{\xi})y'''_{\xi } \ } u'_\xi \\ + & \frac{{\epsilon}^2}{2}\{(1-\beta_\xi){\phi}'''({\mathcal{o}}_{2,\xi})(y''_\xi)^2+(1-\beta_{\xi-1}){\phi}'''({\mathcal{o}}_{1,\xi})(y''_{\xi-1})^2\}u'_\xi .
\end{split}\ ] ] therefore , by hlder s inequality , @xmath204 - { \delta}{\phi^{qnl}_{\beta}}(y)[u]| \leq \left \ { { \epsilon}{\bar{c}_2}{\|\delta \beta_\xi y''_\xi\|}_{{{\ell^p_{{\epsilon } } } } } + { \epsilon}^2 { \bar{c}_2}{\|(1-\beta_{\xi-1})y'''_\xi\|}_{{{\ell^p_{{\epsilon } } } } } + { \epsilon}^2 { \bar{c}_3}{\|(1-\beta_{\xi})(y''_\xi)^2\|}_{{{\ell^p_{{\epsilon } } } } } \right \ } { \|u'\|}_{{{\ell^q_{{\epsilon}}}}}\ ] ] where @xmath205 .
thus , @xmath206 - { \delta}{\phi^{qnl}_{\beta}}(y)[u]\|}_{{{\mathcal{u}^{{-1},{p } } } } } } \leq { \epsilon}{\bar{c}_2}{\|\delta \beta_\xi y''_\xi\|}_{{{\ell^p_{{\epsilon } } } } } \\ + { \epsilon}^2 \{{\bar{c}_2}{\|(1-\beta_{\xi-1})y'''_\xi\|}_{{{\ell^p_{{\epsilon } } } } } + { \bar{c}_3}{\|(1-\beta_{\xi})(y''_\xi)^2\|}_{{{\ell^p_{{\epsilon } } } } } \}.\ ] ] this proves the first claim made in the statement of the theorem . on the other hand , for the bqce energy @xmath180 there is a ghost force arising from the term of order zero on the right hand side of equation .
using formulas , we compute @xmath207 substituting this expression in equation , we derive @xmath203 - { \delta}{\phi^{qce}_{\gamma}}(y)[u ] = { \epsilon}{\sum_{\xi = 1}^n}&\delta^2\alpha_\xi { \phi}'(2y'_\xi ) u'_\xi \\ & + \ { - { \epsilon}\delta\beta_\xi{\phi}''(2y'_{\xi})y''_\xi + { \epsilon}^2 ( 1-\beta_{\xi-1 } ) { \phi}''(2y'_{\xi})y'''_{\xi } \ } u'_\xi \\ & + \frac{{\epsilon}^2}{2}\{(1-\beta_\xi){\phi}'''({\mathcal{o}}_{2,\xi})(y''_\xi)^2+(1-\beta_{\xi-1}){\phi}'''({\mathcal{o}}_{1,\xi})(y''_{\xi-1})^2\}u'_\xi . \end{split}\ ] ] thus , @xmath208 - { \delta}{\phi^{qce}_{\gamma}}(y)[u]| & \leq \bigg\ { \bar{c}_1 { \|\delta^2\alpha_\xi\|}_{{{\ell^p_{{\epsilon } } } } } + { \epsilon}{\bar{c}_2}{\|\delta \beta_\xi y''_\xi\|}_{{{\ell^p_{{\epsilon } } } } } \\ & + { \epsilon}^2 { \bar{c}_2}{\|(1-\beta_{\xi-1})y'''_\xi\|}_{{{\ell^p_{{\epsilon } } } } } + { \epsilon}^2 { \bar{c}_3}{\|(1-\beta_{\xi})(y''_\xi)^2\|}_{{{\ell^p_{{\epsilon } } } } } \bigg\ } { \|u'\|}_{{{\ell^q_{{\epsilon}}}}}. \end{split}\ ] ] this proves the first claim made in the statement of the theorem .
estimate implies that bqnl is patch test consistent . for
observe that if @xmath23 is the uniform deformation , then @xmath209 for all @xmath210 .
thus , by , @xmath211 . on the other hand , observe that the term @xmath212 which appears on the right hand side of estimate does not vanish under uniform strain .
this reflects the fact that bqce is not patch test consistent .
[ rem : clarification of modeling ] we will now give a detailed interpretation of the each term in estimates and .
first , we consider the term @xmath213 observe that the function @xmath214 is supported in the interface and continuum , and that it is identically one throughout the continuum .
this suggests that the term arises from the error of the cauchy - born model .
recall that the cauchy - born energy is the bqnl energy with blending function @xmath215 .
therefore , by the modeling estimate for bqnl given in theorem [ thm : modeling ] we have @xmath216 consequently , we will call the term @xmath217 the _ cauchy - born error_. next , we consider the term @xmath218 .
we observe that @xmath219 is supported in the interface .
thus , the term @xmath218 arises from the error caused by coupling the atomistic and continuum models in the interface .
we will call this term the _ coupling error_. finally , we consider the term @xmath220 which appears in the modeling estimate for bqce but not in the estimate for bqnl .
let @xmath23 be the uniform deformation @xmath221 .
we call @xmath222 the _ ghost force _ associated with the energy @xmath180 , and we observe that under uniform strain formula reduces to @xmath223 = { \epsilon}{\sum_{\xi = 1}^n}-\delta^2\alpha_\xi { \phi}'(2f ) u'_\xi.\ ] ] thus , we see that @xmath224 is the @xmath178 norm of the ghost force . we will call @xmath220 the _ ghost force error_. [ rem : ghost force ] we will now analyze the dependence of the ghost force error on the size of the interface and the shape of the blending function . our analysis will lead to an estimate for the rate at which the ghost force error decreases with the number of atoms in the interface , and to an optimal family of blending functions for the bqce method . consider the bqce energy @xmath180 with blending function @xmath101 as depicted in figure [ fig : blendingfigure ] .
graph of the blending function @xmath101 with interface region @xmath225 .
, width=415 ] each transition between the atomistic and continuum models results in some ghost force
. we will consider the ghost force which arises from a single transition from the atomistic model to the continuum model .
the total ghost force is , of course , the sum of the ghost forces due to each transition .
let @xmath226 be part of the interface , which we will denote simply as @xmath227 is the following .
assume that atoms @xmath228 and @xmath229 are in the atomistic region , and that atoms @xmath230 and @xmath231 are in the continuum .
that is , suppose that @xmath232 for @xmath233 , and that @xmath234 for @xmath235 .
we will estimate the ghost force due to the transition which occurs over region @xmath227 .
first , we construct a blending function which implements such a transition .
let @xmath236 \rightarrow [ 0,1]$ ] be a twice continuously differentiable function such that @xmath237 , @xmath238 , and @xmath239 .
extend @xmath240 to a function defined on @xmath241 by taking @xmath242 for @xmath243 and @xmath244 for @xmath245 .
now let @xmath246 , and define @xmath247 $ ] by @xmath248 .
we call @xmath240 the _ shape _ of the blending function @xmath249 . we will show @xmath250)}\ ] ] for all @xmath251 $ ] .
suppose @xmath252 , and compute @xmath253 by taylor s theorem , @xmath254 where @xmath255 , \\ 0 & \mbox{otherwise}. \end{cases}\ ] ] we note that holds for @xmath256 and @xmath257 only since we have assumed that @xmath258 .
now observe that @xmath259 is a non - negative function whose mass is one , and that @xmath260 is convex for @xmath261 .
therefore , jensen s inequality implies @xmath262 this yields the estimate @xmath263)}.\end{aligned}\ ] ] for all @xmath252 . by a similar argument , one can show @xmath264)}.\ ] ] recall from remark [ rem : clarification of modeling ] that the ghost force error due to the transition over @xmath227 is @xmath265 by minkowski s inequality and the estimates above , we have @xmath266 ) } \label{eqn : ghost force estimate},\ ] ] for all @xmath267 $ ] .
this gives the dependence of the ghost force error on @xmath18 , the blending function , and @xmath268 .
we remark that inequality is sharp . in fact
, if @xmath240 is sufficiently smooth , then @xmath269 converges to @xmath270)}$ ] as @xmath268 tends to infinity .
we pause here to explain why we assumed @xmath239 .
it can be shown that no estimate of order @xmath271 holds unless @xmath239 .
we leave the proof of this fact to the reader , and will instead give an illustrative example .
suppose that @xmath240 were the linear polynomial with @xmath237 and @xmath272 , so @xmath240 is not differentiable at @xmath273 or @xmath274 .
then @xmath275 in section [ subsec : a priori existence ] , we use an estimate of the @xmath276 norm of the modeling error in order to obtain convergence results . thus , we are particularly interested in the @xmath276 norm of the ghost force error . by estimate ,
if @xmath240 satisfies @xmath239 then the @xmath276 norm of the ghost force error decreases with @xmath268 as @xmath277 . on the other hand ,
if @xmath240 is the linear blending function then we see that the ghost force error decreases as @xmath278 .
we conclude that if a large blending region is desired , it is best to choose a blending function which satisfies @xmath279 so that the faster rate of decrease is obtained . in particular , we suspect that the linear polynomial @xmath240 is a poor choice for the shape of the bqce blending function .
we will now use estimate to derive an optimal shape for the blending function of the bqce energy . as discussed above
, we will use the @xmath276 modeling estimate to prove our error results in section [ subsec : a priori existence ] .
thus , we would like to find a family of blending functions which minimizes the @xmath276 ghost force .
estimate suggests that we should choose the shape @xmath240 which minimizes @xmath280)}$ ] subject to @xmath237 , @xmath272 , and @xmath239 .
the euler - lagrange equation for this minimization problem is @xmath281 thus , the optimal shape @xmath240 is the cubic polynomial which satisfies the constraints @xmath237 , @xmath272 , and @xmath239 .
[ rem : coupling error ] now we examine the dependence of the coupling error on the number of atoms in the interface and the shape of the blending function .
first , we consider the coupling error of the bqnl energy . following the notation established in remark [ rem : ghost force ] ,
let @xmath282 be contained in the interface .
let @xmath283 \rightarrow [ 0,1]$ ] be a continuously differentiable function with @xmath284 and @xmath285 .
extend @xmath286 to a function defined on @xmath241 by taking @xmath287 for @xmath243 and @xmath288 for @xmath245 .
now let @xmath289 $ ] be defined by @xmath290 . using an argument similar to the one given in remark [ rem : ghost force ]
, one can show @xmath291 for all @xmath292 $ ] . the result above yields the estimate @xmath293 this gives the dependence of the coupling error of the bqnl energy on @xmath18 , the blending function , and the size of the interface .
we now address the coupling error of the bqce method .
let the blending function , @xmath249 , of the bqce energy be defined as in remark [ rem : ghost force ] .
we recall from equation that the bqce energy with blending function @xmath249 is the bqc energy , @xmath294 , with @xmath295 thus , using minkowski s inequality and the result for the bqnl energy , we make the estimate @xmath296 this gives the dependence of the coupling error of the bqce method on @xmath18 , the blending function , and the size of the interface .
[ higher ] suppose we let the measure of the interface be a fixed fraction @xmath297 of the total measure of the domain as @xmath18 tends to zero
. then the number of atoms @xmath268 in the interface would be approximately @xmath298 , and estimates and would reduce to @xmath299 we observe that the forces arising on each atom due to the bqc energy can be decomposed into a sum of central forces which satisfy newton s third law . in particular , the forces arising from both the bqce and bqnl energies satisfy newton s third law .
first we derive expressions for the second variations of the atomistic and bqc energies .
we have @xmath300 & = { \epsilon}{\sum_{\xi = 1}^n}{\phi}''(y'_{\xi } ) |u'_\xi|^2 + 4 \alpha_\xi { \phi}''(2y'_\xi)|u'_\xi|^2 + \beta_\xi { \phi}''(y'_{\xi } + y'_{\xi+1})|u'_\xi + u'_{\xi+1}|^2 \\ & = { \epsilon}{\sum_{\xi = 1}^n}{\phi}''(y'_{\xi})|u'_\xi|^2 + 4\alpha_\xi { \phi}''(2y'_\xi)|u'_\xi|^2 + \beta_\xi { \phi}''(y'_{\xi } + y'_{\xi + 1 } ) [ 2 |u'_\xi|^2 + 2 |u'_{\xi + 1}|^2 - { \epsilon}^2 |u''_\xi|^2 ] \notag\\ & = { \epsilon}{\sum_{\xi = 1}^n}{\bar{a}}_\xi |u'_\xi|^2 + { \epsilon}^2 \bar{b}_\xi |u''_\xi|^2,\notag \ ] ] where @xmath301 \mbox { and } \\ \bar{b}_\xi & : = - \beta_\xi { \phi}''(y'_{\xi } + y'_{\xi+1 } ) .
\end{split}\ ] ] the second variation of the atomistic energy is , of course , a special case of the above .
we have @xmath302 & = { \epsilon}{\sum_{\xi = 1}^n}{\phi}''(y'_{\xi } ) + 4 \left \{{\frac{1}{2}}{\phi}''(y'_{\xi } + y'_{\xi+1 } ) + { \frac{1}{2}}{\phi}''(y'_{\xi } + y_{\xi -1 } ) ] \right \ } |u'_\xi|^2 + ( -{\epsilon}^2 { \phi}''(y'_{\xi } + y'_{\xi+1 } ) |u''_\xi|^2 ) \\ & = { \epsilon}{\sum_{\xi = 1}^n}a_\xi |u'_\xi|^2 + { \epsilon}^2 b_\xi |u''_\xi|^2,\end{aligned}\ ] ] where @xmath303 \mbox { and } \\ b_\xi & : = - { \phi}''(y'_{\xi } + y'_{\xi+1 } ) .
\end{split}\ ] ] we begin our analysis with a lemma bounding @xmath304 . in remark
[ rem : clarification of modeling ] , we interpreted each term which appears in the estimate below , and in remarks [ rem : ghost force ] and [ rem : coupling error ] we explained how each term depends on @xmath18 , the blending function , and the number of atoms in the interface .
we concluded that @xmath305 where @xmath268 is the number of atoms in the interface .
the reader should keep these scalings in mind throughout section [ subsec : stability ] .
[ lem : difference of hessian cfcts ] .... .... 1 .
let @xmath180 be the bqce energy with blending function @xmath101 .
we have @xmath306 where @xmath181 , @xmath307 and @xmath308 , @xmath309 for @xmath186 2 .
let @xmath121 be the bqnl energy with blending function @xmath93 .
we have @xmath310 where @xmath308 , @xmath311 for @xmath186 the proof of the lemma is extremely similar to the proof of our modeling estimate above . for the general bqc energy @xmath189 we compute @xmath312 using formulas and .
then we expand all terms above at @xmath195 using taylor s theorem .
so , @xmath313 where @xmath197 , and @xmath198 .
now we expand the second term in curly braces on the right hand side of .
we obtain @xmath314 we now consider the case of the bqce energy . as a consequence of equation , we have @xmath315 thus , by substituting into , we see that for the bqce energy @xmath316 this proves the first claim made in the statement of the lemma . on the other hand , we recall from the proof of theorem [ thm : modeling ] that for the bqnl energy , @xmath317 therefore , for the bqnl energy , the first term in curly braces in equation vanishes , and we have @xmath318 this proves the second claim made in the statement of the lemma .
now we derive estimates relating the @xmath319 coercivity constants of the hessians of @xmath189 and @xmath81 .
define @xmath320 , \qquad c_\beta(y ) = \inf_{{\|u'\|}_{l^2_{\epsilon } } = 1 } { \delta}^2 { \phi^{qnl}_{\beta}}(y ) [ u , u ] , \\
c_\gamma(y ) = \inf_{{\|u'\|}_{l^2_{\epsilon } } = 1 } { \delta}^2 { \phi^{qce}_{\gamma}}(y ) [ u , u].\end{gathered}\ ] ] for our _ a priori _ error estimate , we would like to show that if @xmath321 is a strongly stable minimizer of @xmath322 then @xmath323 and @xmath324 .
however , such a general result was not proved for the qnl method in @xcite . instead , a weaker stability result that is restricted to `` elastic states '' without defects @xcite was proved .
we now extend this _ a priori _ stability result to the bqc method .
[ rem : concave second nbrs ] we will place an additional condition on the set of admissible deformations in order to prove our _ a priori _ and _ a posteriori _ stability estimates in theorems [ thm : aprstab ] and [ thm : a posteriori stability ] .
we will assume that @xmath325 under this assumption , the constants @xmath326 and @xmath327 in the expressions for the atomistic and bqc hessians are nonnegative .
assumption is justified since @xmath328 only under extreme compression , and in that case the second nearest neighbor pair interaction model itself can be expected to be invalid .
the authors of @xcite , @xcite , and @xcite all consider energies similar to , and they all make assumptions similar to ( see section 2.3 of @xcite for further discussion of this point ) .
[ thm : aprstab ] let @xmath329 .
assume that @xmath330 . 1 .
let @xmath180 be the bqce energy with blending function @xmath101 .
we have @xmath331 where @xmath181 , @xmath182 , and @xmath308 , @xmath309 for @xmath186 2 .
let @xmath121 be the bqnl energy with blending function @xmath93 .
we have @xmath332 where @xmath308 , @xmath311 for @xmath186 let @xmath333 with @xmath334 .
recall that by formula we have @xmath335 = { \epsilon}{\sum_{\xi = 1}^n}{\bar{a}}_\xi |u'_\xi|^2 + { \epsilon}^2 \bar{b}_\xi |u''_\xi|^2\ ] ] for any bqc energy @xmath189 .
since we assume @xmath336 , all the coefficients @xmath327 are nonnegative , so @xmath335 \geq { \epsilon}{\sum_{\xi = 1}^n}{\bar{a}}_\xi |u'_\xi|^2.\ ] ] now we assume that the energy @xmath189 is a bqce energy .
then using the first part of lemma [ lem : difference of hessian cfcts ] we have @xmath337 & \geq { \epsilon}{\sum_{\xi = 1}^n}a_\xi |u'_\xi|^2 - 2 { \bar{c}_2}{\|\delta^2\alpha_\xi\|}_{\ell^\infty } \notag\\ & \quad \quad -{2{\epsilon}{\bar{c}_3}{\|\delta\beta_\xi y''_\xi\|}_{\ell^\infty }
-2 { \epsilon}^2 \ { { \bar{c}_3}{\|(1-\beta_{\xi-1})y'''_\xi\|}_{\ell^\infty } + { \bar{c}_4}{\|(1-\beta_{\xi})(y''_\xi)^2\|}_{\ell^\infty } \}}\notag \\ & \geq \underline{a } -2 { \bar{c}_2}{\|\delta^2\alpha_\xi\|}_{\ell^\infty } \notag\\ & \quad \quad- { 2{\epsilon}{\bar{c}_3}{\|\delta\beta_\xi y''_\xi\|}_{\ell^\infty } -2 { \epsilon}^2 \ { { \bar{c}_3}{\|(1-\beta_{\xi-1})y'''_\xi\|}_{\ell^\infty } + { \bar{c}_4}{\|(1-\beta_{\xi})(y''_\xi)^2\|}_{\ell^\infty } \}}.\notag\end{aligned}\ ] ] this proves the first claim made in the statement of the theorem . for the bqnl method
, we make a similar estimate using the second part of lemma [ lem : difference of hessian cfcts ] .
we have @xmath338 & \geq { \epsilon}{\sum_{\xi = 1}^n}a_\xi |u'_\xi|^2 - { 2{\epsilon}{\bar{c}_3}{\|\delta\beta_\xi y''_\xi\|}_{\ell^\infty } -2 { \epsilon}^2 \ { { \bar{c}_3}{\|(1-\beta_{\xi-1})y'''_\xi\|}_{\ell^\infty } + { \bar{c}_4}{\|(1-\beta_{\xi})(y''_\xi)^2\|}_{\ell^\infty } \}}\notag\\ & \geq \underline{a } - { 2{\epsilon}{\bar{c}_3}{\|\delta\beta_\xi y''_\xi\|}_{\ell^\infty }
-2 { \epsilon}^2 \ { { \bar{c}_3}{\|(1-\beta_{\xi-1})y'''_\xi\|}_{\ell^\infty } + { \bar{c}_4}{\|(1-\beta_{\xi})(y''_\xi)^2\|}_{\ell^\infty } \}}.\end{aligned}\ ] ] this proves the second claim made in the statement of the theorem .
[ rem : accuracy of critical strain ] let @xmath23 be the uniform deformation @xmath221 .
let @xmath339 be the strain at which the lattice loses stability in the atomistic model , so @xmath340 we call @xmath339 the _ critical strain _ of the atomistic model , and we define the critical strains of the cauchy - born , bqce , and bqnl energies similarly .
the significance of the critical strain is discussed in @xcite . under uniform strain ,
the @xmath276 coercivity constant @xmath341 of the cauchy - born energy is @xmath342 furthermore , by formula , for the atomistic model under uniform strain @xmath343 .
thus , by theorem [ thm : aprstab ] , we see @xmath344 where @xmath345 is the constant which arises in estimate , and @xmath268 is the number of atoms in the interface . in the language of @xcite , estimates imply that there is no error in the critical strain of the bqnl energy , and that the error in the critical strain of the bqce energy decreases as @xmath346 . for an _ a posteriori _ existence result , one would like to show that if @xmath347 is a strongly stable minimizer of @xmath189 , then @xmath348 .
we will not present an _ a posteriori _ existence result , but we give an _ a posteriori _ stability estimate anyway . using this estimate , one can prove _ a posteriori _ existence theorems similar to theorems [ thm : a priori existence for bqnl ] and [ thm : a priori existence for bqce ] .
[ thm : a posteriori stability ] _ ( a posteriori stability ) _ assume that @xmath349 . 1 .
let @xmath180 be the bqce energy with blending function @xmath101 .
we have @xmath350 where @xmath181 , @xmath182 , and @xmath308 , @xmath309 for @xmath186 2 .
let @xmath121 be the bqnl energy with blending function @xmath93 .
we have @xmath351 where @xmath308 , @xmath311 for @xmath186 the proof of the theorem is extremely similar to the proof of theorem [ thm : aprstab ] .
we use lemma [ lem : difference of hessian cfcts ] and that @xmath352 whenever @xmath353 .
we are now ready to prove our _ a priori _ error estimates .
our results generalize theorem 8 and its proof given in @xcite . before stating the theorem , it is convenient to establish some notation .
let @xmath159 be a minimizer of the energy @xmath354 .
we will use theorem [ thm : aprstab ] in the proof of theorem [ thm : a priori existence for bqnl ] .
thus , we must assume that @xmath159 is an elastic state .
that is , we assume that @xmath355 for @xmath356 . see the discussion preceding theorem [ thm : aprstab ] for more explanation .
now define the energy @xmath357 by @xmath358 .
we realize that the statement of theorem [ thm : a priori existence for bqce ] may seem slightly complicated , but the basic idea of the theorem is quite simple .
in essence , theorem [ thm : a priori existence for bqce ] states that if @xmath159 is a stable minimizer of the atomistic energy which is sufficiently smooth in the continuum region , and if the ghost force error is sufficiently small , then there exists a stable minimizer @xmath359 of the bqce energy which is close to @xmath159 .
the conditions and make the hypotheses that @xmath159 must be sufficiently smooth in the continuum region and that the ghost force must be small precise .
[ thm : a priori existence for bqce ] let @xmath360 , @xmath159 , and @xmath361 be as above .
assume that @xmath362 , and that @xmath330 .
there exist constants @xmath363 and @xmath364 so that if @xmath365 and @xmath366 then there exists a locally unique minimizer @xmath367 of @xmath360 which satisfies @xmath368.\ ] ] let @xmath369 by @xmath370
. we will apply theorem [ thm : ift ] to @xmath371 with @xmath372 . in order to apply theorem [ thm : ift ]
, we need to find a bound @xmath125 on the residual @xmath373 , a bound @xmath177 on @xmath374 , and a lipschitz constant @xmath375 for @xmath376 on the ball @xmath377 .
we will find @xmath125 using the modeling estimates given in theorem [ thm : modeling ] , and we will find @xmath177 using the stability estimates given in theorem [ thm : aprstab ] together with the condition .
once we have these constants , we must verify the condition @xmath378 .
as we will see , this condition holds if @xmath379 and @xmath380 are taken sufficiently small .
once these facts are established , theorem [ thm : ift ] implies that there exists @xmath381 such that @xmath382 and @xmath383 . finally , we show that @xmath367 is a strongly stable minimizer of @xmath360 which proves theorem [ thm : a priori existence for bqnl ] . *
modeling error : @xmath384 .
* first , we find @xmath125 so that @xmath385 . by theorem [ thm : modeling ] , we have @xmath386 so we take in theorem [ thm : ift ] @xmath387
stability : @xmath388 . * second , we find @xmath177 so that @xmath389 by theorem [ thm : aprstab ] , we have @xmath390 then if we take @xmath391 , we can combine and to find @xmath392 in that case , @xmath393
so we can use @xmath394 * 3 .
lipschitz bound of @xmath376 on @xmath395 .
* now we bound the lipschitz constant of @xmath376 on @xmath396 .
the lipschitz bound follows easily if we can assume that @xmath397 for all @xmath398 with @xmath399 .
we will choose @xmath380 so that implies . to that end , observe that for @xmath399 we have @xmath400 where @xmath401 . then we need only choose @xmath402 to ensure that holds . to see the lipschitz bound , let @xmath403 . then expand @xmath404 - { \delta}^2{\phi^{qce}_{\gamma}}(y+w_2)[u , v]|\ ] ] using formula .
one sees easily that @xmath404 - { \delta}^2{\phi^{qce}_{\gamma}}(y+w_2)[u , v]| \leq l ' { \|w'_1 - w'_2\|}_{\ell^\infty } { \|u'\|}_{{{\ell^2_{\epsilon } } } } { \|v'\|}_{{{\ell^2_{\epsilon}}}}\ ] ] where @xmath405 . in that case
, we have @xmath406 where @xmath407
the fixed point condition : @xmath173 .
* we show that if @xmath380 is sufficiently small then @xmath408 . we have @xmath409 observe that could be restated as @xmath410 with @xmath125 as defined in part ( 1 ) above .
thus , inequality holds whenever @xmath411 .
* 5 . the error estimate . *
take @xmath412 , and @xmath413 .
then by theorem [ thm : ift ] and the conclusions of parts ( 1 - 4 ) above , there exists @xmath414 with @xmath415 and @xmath416 it remains only to show that the equilibrium @xmath367 is a locally unique minimizer of @xmath360 . we will show that @xmath417 is positive definite .
we have @xmath418 the second inequality follows from , the lipschitz bound , and the error estimate @xmath416 the third inequality follows from the condition @xmath173 .
we will now restate theorem [ thm : a priori existence for bqce ] using the estimates given in remarks [ rem : ghost force ] and [ rem : coupling error ] . for simplicity ,
assume that all transitions between the atomistic and continuum models occur over regions which contain @xmath268 atoms .
let @xmath92 and @xmath93 be the blending functions so that @xmath419 . ( see for the precise definitions of @xmath92 and @xmath93 given @xmath101 . )
using the estimates given in remarks [ rem : ghost force ] and [ rem : coupling error ] , let @xmath420 be a constant so that @xmath421 and let @xmath422 be a constant so that @xmath423 in essence , corollary [ cor : error estimate for bqce ] states that if @xmath159 is a stable minimizer of the atomistic energy which is sufficiently smooth in the continuum region and if the ghost force error is small , then there exists a minimizer @xmath367 of the bqce energy @xmath360 which is close to @xmath159 .
[ cor : error estimate for bqce ] let @xmath360 , @xmath159 , and @xmath361 be as in theorem [ thm : a priori existence for bqce ] , and let @xmath420 , @xmath422 , and @xmath268 be as above .
there exist constants @xmath379 and @xmath380 so that if @xmath424 and @xmath425 then there exists a locally unique minimizer @xmath367 of @xmath360 which satisfies @xmath426.\end{aligned}\ ] ] we give a similar _ a priori _ error estimate for the bqnl method in theorem [ thm : a priori existence for bqnl ] . following the notation adopted in theorem [ thm : a priori existence for bqce ] , let @xmath427 be the energy given by @xmath428 .
let @xmath159 be a minimizer of the energy @xmath354 such that @xmath355 for @xmath356 .
[ thm : a priori existence for bqnl ] let @xmath429 , @xmath159 , and @xmath361 be as above .
assume that @xmath362 , and that @xmath330 .
there exist constants @xmath363 and @xmath430 so that if @xmath431 then there exists a locally unique minimizer @xmath432 of @xmath427 which satisfies @xmath433.\ ] ] the proof of theorem [ thm : a priori existence for bqnl ] is similar to the the proof of theorem [ thm : a priori existence for bqce ] ; we simply use the modeling and stability estimates for bqnl in place of those for bqce .
we have proposed a smoothly blended version of the quasicontinuum energy ( qce ) which we call the blended quasicontinuum energy ( bqce ) .
bqce blends the atomistic and corresponding cauchy - born continuum energies used in the qce method over an interfacial region whose thickness is a small number @xmath0 of blended atoms .
we analyze the accuracy of bqce applied to the problem of a one - dimensional chain with next - nearest neighbor interactions . for this test problem
, we show how to choose the optimal blending function for weighting the atomistic and cauchy - born continuum energies .
if bqce is implemented using the optimal blending function , the critical strain error of bqce can be reduced by a factor of @xmath6 .
thus , we believe that bqce could be used to accurately compute the deformation of crystalline solids up to lattice instabilities . we are continuing the development of the bqce energy by modifying the code developed to study the accuracy of quasicontinuum methods for two benchmark problems the stability of a lomer dislocation pair under shear and the stability of a lattice to plastic slip under tensile loading @xcite .
we note that the potential to significantly improve the accuracy of existing quasicontinuum codes by easily implemented modifications is a very desirable feature of the bqce approach .
we think our theoretical analysis of the accuracy near instabilities for one - dimensional model problems can successfully explain most of the computational results for these multi - dimensional benchmark problems .
we expect that some form of the stability and error estimates in this paper can be generalized to atomistic models with finite range interactions by extending the techniques given in @xcite .
we are investigating the extension of our one - dimensional analysis to the multi - dimensional setting , but we expect that any multi - dimensional analysis would likely be restricted to perturbations from the ground state which are far from lattice instability
. we will thus need to rely on our one - dimensional analysis to attempt to understand the computational results from our multi - dimensional benchmark studies @xcite .
we acknowledge the useful suggestions of xingjie li and dr .
christoph ortner .
m. luskin and c. ortner .
linear stationary iterative methods for the force - based quasicontinuum approximation . in b.
engquist , o. runborg , and r. tsai , editors , _ numerical analysis and multiscale computations _ , volume 82 of _ lect .
notes comput .
_ springer verlag , to appear .
arxiv:1104.1774 .
b. van koten , x. h. li , m. luskin , and c. ortner .
a computational and theoretical investigation of the accuracy of quasicontinuum methods . in i.
graham , t. hou , o. lakkis , and r. scheichl , editors , _ numerical analysis of multiscale problems_. springer , to appear . | the development of patch test consistent quasicontinuum energies for multi - dimensional crystalline solids modeled by many - body potentials remains a challenge . the original quasicontinuum energy ( qce ) @xcite has been implemented for many - body potentials in two and three space dimensions , but it is not patch test consistent .
we propose that by blending the atomistic and corresponding cauchy - born continuum models of qce in an interfacial region with thickness of a small number @xmath0 of blended atoms , a general quasicontinuum energy ( bqce ) can be developed with the potential to significantly improve the accuracy of qce near lattice instabilities such as dislocation formation and motion . in this paper ,
we give an error analysis of the blended quasicontinuum energy ( bqce ) for a periodic one - dimensional chain of atoms with next - nearest neighbor interactions .
our analysis includes the optimization of the blending function for an improved convergence rate .
we show that the @xmath1 strain error for the non - blended qce energy ( qce ) , which has low order @xmath2 where @xmath3 is the atomistic length scale @xcite , can be reduced by a factor of @xmath4 for an optimized blending function where @xmath0 is the number of atoms in the blending region .
the qce energy has been further shown to suffer from a o@xmath5 error in the critical strain at which the lattice loses stability @xcite .
we prove that the error in the critical strain of bqce can be reduced by a factor of @xmath6 for an optimized blending function , thus demonstrating that the bqce energy for an optimized blending function has the potential to give an accurate approximation of the deformation near lattice instabilities such as crack growth . |
in this paper , we consider transverse knots in @xmath0 , i.e. knots that are transverse to the contact planes of the standard contact structure @xmath1 .
a simple `` classical '' invariant is given by the self - linking number @xmath2 of a transverse knot .
however , if @xmath3 , @xmath4 are two transverse knots that are smoothly isotopic and share the same self - linking number , @xmath3 and @xmath4 do not have to be transversally isotopic : this phenomenon was first discovered in @xcite and @xcite , and more examples were recently obtained in @xcite .
the goal of this paper is to study contact manifolds that arise as cyclic covers branched over transverse knots and links in @xmath0 .
( such cyclic covers carry natural contact structures lifting @xmath5 . )
the main question we would like to address is suppose that transverse knots @xmath6 are smoothly isotopic , and @xmath7 .
fix @xmath8 .
are @xmath9-fold cyclic covers branched over @xmath3 and @xmath4 contactomorphic ?
[ quest ] finding two such non - contactomorphic covers would imply that the induced contact structure on the branched cover is an efficient invariant of transverse knots . on the other hand , a positive answer to
the above question for any pair of knots means that the cyclic branched covers are insensitive to the subtler structure of transverse knots . while we found no examples of non - contactomorphic branched covers , we are able to answer question [ quest ] positively in many special cases .
in particular , we show that branched cyclic covers of any degree are contactomorphic for all examples of birman
menasco @xcite , and that branched double covers for many examples of @xcite are also contactomorphic .
we prove the @xmath9-fold cyclic branched covers of transverse links @xmath3 and @xmath4 are contactomorphic for all @xmath9 if : * @xmath10 and @xmath11 are a positive and a negative transverse push - offs of a legendrian link @xmath12 and its legendrian mirror @xmath13 , or * @xmath3 and @xmath4 are given by transverse 3-braids related by a negative flype .
moreover , the branched double covers are contactomorphic for arbitrary transverse braids related by a negative flype .
[ introthm ] in fact , we are able to prove a little more ( see section [ distinguish ] ) .
we also note that all examples of birman menasco satisfy the second condition of theorem [ introthm ] .
( a _ negative flype _ is a braid move introduced in @xcite ; see also figure [ neg - flype ] in section [ distinguish ] . )
let @xmath14 denote the natural contact structure on the branched @xmath9-fold cyclic cover @xmath15 of a transverse link @xmath12 as explained in subsection [ subsec - induced - str ] .
we describe the contact manifolds @xmath15 in two ways .
first , we give an open book decomposition supporting @xmath14 .
if @xmath12 is represented as a transverse @xmath16-braid , an open book decomposition of @xmath17 can be obtained as a lift of the open book for @xmath18 whose binding is the braid axis , and the page is a disk meeting @xmath12 transversely at @xmath16 points .
the monodromy for the resulting open book can be read off the braid word .
more precisely , a positive crossing in the braid contributes @xmath19 positive dehn twists to the monodromy , while a negative crossing contributes @xmath19 negative dehn twists .
second , we give contact surgery diagrams @xcite for these contact manifolds .
we find that a positive ( resp .
negative ) crossing in the braid corresponds to a legendrian surgery ( resp .
@xmath20 contact surgery ) on @xmath19 standard legendrian unknots .
interestingly , it turns out that the linking between these @xmath19 unknots depends on the sign on the crossing : while for a negative crossing the surgery is performed on _ unlinked _ unknots , for a positive crossing the unknots are linked .
( this phenomenon arises in the smooth setting as well , where the construction can be thought of as a version of the montesinos trick for higher order covers . )
we refer the reader to lemma [ dehn twist lemma ] and theorem [ surgthm ] for precise statements .
this description yields a few properties of @xmath9-fold cyclic branched covers ; we can determine whether they are tight or overtwisted in certain cases and describe the homotopy invariants of the contact structures .
[ tight - ow ] the contact manifold @xmath21 is stein fillable if the transverse link @xmath12 is represented by a quasipositive braid ; it is overtwisted if @xmath12 is obtained as a transverse stabilization of another transverse link .
in fact , in section [ properties ] we prove overtwistedness for a much wider class of contact structures .
[ homotp ] fix @xmath8 .
let @xmath22 be the @xmath23 structure induced by @xmath24
. then @xmath25 .
the three - dimensional invariant @xmath26 is completely determined by the topological link type of @xmath12 and its self - linking number @xmath27 . the present paper continues the work started by the third author in @xcite , where question [ quest ] was studied for the case of branched double covers .
( the paper @xcite was written before the advent of heegaard floer transverse invariants @xcite , and the only explicit examples of non - transversely simple knots available then were the 3-braids of @xcite . ) the techniques from @xcite are useful for the higher order covers as well ; in particular , theorems [ homotp ] and [ tight - ow ] are direct extensions of results of @xcite .
in this section , we fix notation and collect the necessary facts about transverse knots , open books and contact surgeries , referring the reader to @xcite for more details .
we assume that all 3-manifolds are closed , connected and oriented , and all contact structures are co - oriented
. it will be helpful to represent transverse links by closed braids .
for this , consider the symmetric version of the standard contact structure @xmath28 with @xmath29 .
then , any closed braid about @xmath30-axis can be made transverse to the contact planes ; moreover , any transverse link is transversely isotopic to a closed braid in @xmath28 @xcite .
equivalently , we can consider transverse braids in the contact structure @xmath1 , for example assuming that our braids are satellites of a fixed standard legendrian unknot with @xmath31 . to define the self - linking number @xmath27 , trivialize the plane field @xmath32 , and
let the link @xmath33 be the push - off of @xmath12 in the direction of the first coordinate vector for @xmath32 .
then , @xmath27 is the linking number between @xmath12 and @xmath33 .
given a closed braid representation of @xmath12 , we have @xmath34 where @xmath35 ( @xmath36 ) is the number of positive ( negative ) crossings , and @xmath37 is the braid index .
the _ stabilization _ of a transverse link represented as a braid is equivalent to the negative braid stabilization , i.e. adding an extra strand and a negative kink to the braid .
if @xmath38 is the result of stabilization of @xmath12 , then @xmath39 note that the positive braid stabilization does not change the transverse type of the link .
abusing notation , we will often identify a transverse link with its braid word , writing it in terms of the standard generators @xmath40 and their inverses .
another useful way to think about transverse knots is as push - offs of legendrian knots .
indeed , a given legendrian knot yields two transverse knots ( a positive and negative push - off ) , whose self - linking number is @xmath41 .
this description is used in @xcite .
an open book decomposition of a 3-manifold @xmath42 is a pair @xmath43 of a surface @xmath44 with non - empty boundary @xmath45 and a diffeomorphism @xmath46 of @xmath44 with @xmath47 , such that @xmath48 is the mapping torus @xmath49 / \sim $ ] , where @xmath50 .
the surface @xmath44 is called a _ page _ and @xmath45 the _ binding _ of the open book . by the celebrated work of giroux @xcite , contact structures on @xmath42 up to an isotopy are in one - to - one correspondence with open book decompositions of @xmath42 up to stabilization .
stabilization of an open book consists of plumbing a right - handed hopf band , i.e attaching a 1-handle to a page and composing the monodromy with a right - handed dehn twist along an arbitrary curve intersecting the cocore of the handle at one point .
a right - handed dehn twist about a simple closed curve @xmath51 is the diffeomorphism @xmath52 that acts on the neighborhood @xmath53 of @xmath54 as @xmath55 , and fixes @xmath56 , see figure [ rh - twist ] .
( the term `` positive dehn twist '' is also common in the literature , but we avoid it since positive dehn twists will correspond to @xmath57 contact surgeries . ) a left - handed dehn twist about @xmath54 is the inverse of @xmath52 .
( 203 , 78 ) ( 0,0 ) about @xmath54 . , title="fig : " ] ( 35 , 3)@xmath54 ( -7 , 38)@xmath58 ( 95 , 27)@xmath59 we recall that the monodromy of an arbitrary open book can be written as a product of left - handed and right - handed dehn twists , and that a contact structure is stein fillable if and only if it admits an open book with the monodromy given by a product of right - handed dehn twists @xcite .
let @xmath60 be a null - homologous legendrian knot in a contact manifold @xmath61 .
performing a dehn surgery on @xmath60 , we cut out a tubular neighborhood of the knot @xmath60 ( i.e. a solid torus ) and glue it back in .
when the surgery coefficient is @xmath62 ( with respect to the contact framing on @xmath60 ) , this procedure is compatible with contact structures : the gluing can be done so that the contact structure on the solid torus matches the contact structure on its complement .
moreover , the resulting contact manifold is independent of choices , so the @xmath62 contact surgery is well - defined .
contact surgery is a very useful tool , as any contact manifold can be obtained from @xmath0 by a contact surgery on some legendrian link .
we also recall that @xmath57 contact surgery is in fact the same as legendrian surgery , while @xmath20 contact surgery is the operation inverse to it .
( unlike legendrian surgery , @xmath20 surgery does not preserve stein fillability or other similar properties of contact structures . )
homotopy invariants of a contact structure @xmath32 on @xmath63 encode information on the corresponding plane field .
first , we can consider the @xmath23 structure @xmath64 on @xmath63 induced by @xmath32 .
secondly , when @xmath65 is torsion , the three - dimensional invariant @xmath26 can be defined @xcite .
if @xmath66 is the boundary of an almost - complex 4-manifold @xmath67 , this invariant is given by @xmath68 the homotopy invariants of a contact structure can be read off its contact surgery presentation as follows @xcite .
let @xmath69 be the four - manifold obtained from @xmath70 by attaching the 2-handles as dictated by the @xmath62-surgery diagram .
consider an almost - complex structure @xmath71 defined on @xmath69 in the complement of @xmath72 balls lying in the interior of the ( + 1)-handles of @xmath69 .
as shown in @xcite , @xmath71 induces a @xmath23 structure @xmath73 which extends to all of @xmath69 , and the @xmath74 invariant of @xmath75 can be computed as @xmath76 this formula is very similar to the case where @xmath77 is almost - complex , except that there is a correction term of @xmath78 for each ( + 1)-surgery .
now , suppose that a @xmath79-handle is attached to the four - manifold @xmath69 in the process of legendrian surgery on a knot @xmath60 , and denote by @xmath80 $ ] the homology class that arises from the seifert surface of @xmath60 capped off inside the handle .
it is well - known @xcite that @xmath81 evaluates on @xmath80 $ ] as the rotation number of the legendrian knot @xmath60 .
furthermore , it is shown in @xcite that the same result is true for @xmath20-contact surgeries ( for the @xmath23 structure @xmath73 on @xmath69 described above ) .
there are two ways to describe a given contact structure , via an open book decomposition or a contact surgery diagram .
we will need to switch between the two descriptions . a contact surgery diagram consists of a legendrian link in @xmath18 with surgery coefficients .
we can find an open book decomposition of @xmath18 whose page contains this link .
thus components of the surgery link correspond to curves on the page ; we perform right - handed dehn twists on curves corresponding to legendrian surgeries , and left - handed dehn twists on those corresponding to @xmath20 contact surgeries .
the resulting open book is compatible with the contact structure given by the surgery diagram @xcite .
conversely , given an open book decomposition of a given contact manifold , we will need to obtain a contact surgery diagram .
to this end , we assume that the monodromy of the open book contains a sequence of dehn twists producing @xmath0 ( this can always be achieved by composing the given monodromy with a few dehn twists and their inverses ) .
we can then embed the page of the open book into @xmath18 ; if the curves on which the remaining dehn twists are to be performed become legendrian knots in @xmath18 , we can replace the dehn twists with @xmath62 contact surgeries to obtain the required surgery diagram .
( note that a `` compatible '' embedding will imply that the contact framing of the legendrian knots is the same as the page framing . )
we perform this procedure in detail in section [ surg for cov ] .
given a branched @xmath9-fold cyclic branched cover @xmath82 for a transverse knot @xmath60 , we describe the natural contact structure @xmath83 on @xmath82 as follows . in local coordinates
@xmath84 near the knot @xmath85 , we may write the covering map as @xmath86 , and assume that @xmath87 . we set @xmath83 to be the kernel of the pull - back form ; however , the pull - back form @xmath88 fails to be contact along the knot . to avoid this issue
, we can define a new contact form by interpolating between the form @xmath89 in a small tubular neighborhood of @xmath60 and the pull - back form on the branched cover away from @xmath60 .
its kernel is a contact structure which is independent of choices .
( this construction is explained in detail in @xcite for branched double covers and works for links and higher order covers with only notational changes . ) we can also describe the contact structure on @xmath82 via open books , by representing @xmath60 as a braid .
we then consider a branched cover of the standard open book for @xmath18 whose binding is the braid axis , and page a disk meeting the @xmath16-braid @xmath60 at @xmath16 points .
we adopt this approach in the next section , determining how the half - twist generators of the braid @xmath60 lift to the branched cover .
it is clear that the resulting contact structure is isotopic to the one described above .
let @xmath90 be a transverse link . identifying @xmath60 with a closed braid about the @xmath30-axis ,
let @xmath91 denote a braid representation of @xmath60 .
let @xmath92 be a disk .
then @xmath60 transversely intersects @xmath93 in @xmath16 points @xmath94 .
we may regard @xmath95 as a diffeomorphism of @xmath93 that exchanges @xmath96 in the neighborhood @xmath97 of @xmath98 and fixes @xmath99 ( figure [ sigma ] ) .
( 75,80 ) ( 0,0 ) and the map @xmath100 acting on the shaded region @xmath101.,title="fig : " ] ( 15 , 40)@xmath102 ( 40 , 58)@xmath103 ( 26 , 26)@xmath104 ( 55 , 43)@xmath105 ( 20 , 60)@xmath106 let @xmath107 be a monodromy map of @xmath93 .
the symmetric contact structure @xmath28 is supported by the open book decomposition @xmath108 of @xmath18 , whose binding is the @xmath30-axis ( braid axis ) and pages are disks @xmath93 .
fix @xmath8 , and let @xmath109 be the @xmath9-fold cyclic covering branched along @xmath60 .
the covering @xmath110 induces the open book decomposition @xmath111 of @xmath82 given by the lift of the open book @xmath108 .
the surface @xmath112 can be obtained by gluing @xmath9 copies of @xmath93 along slits as in figure [ dehn - twist ] .
( 218 , 325 ) ( 0,10 ) and simple closed curves @xmath113 s .
the region @xmath114 is shaded.,title="fig : " ] ( 145,99)@xmath115 ( 114,53)@xmath116 ( 117 , 35)@xmath117 ( 217 , 35)@xmath118 ( 52,304)@xmath119 ( 34,320)@xmath120 ( 49,320)@xmath121 ( 81,299)@xmath122 ( 87,290)@xmath123 ( -14,290)@xmath124 ( -8,297)@xmath125 ( 35,292)@xmath126 ( 26,275)@xmath127 ( 65,278)@xmath128 ( 31,254)@xmath129 ( 30,256)(-1,0)7 ( 25,214)@xmath130 sheet ( 182,304)@xmath131 ( 160,302)@xmath119 ( 164,320)@xmath121 ( 179,320)@xmath132 ( 211,299)@xmath123 ( 217,290)@xmath133 ( 116,290)@xmath125 ( 122,297)@xmath134 ( 167,289)@xmath126 ( 155,284)@xmath127 ( 195,278)@xmath128 ( 161,254)@xmath129 ( 160,256)(-1,0)7 ( 155,214)@xmath135 sheet ( 111,198)@xmath136 ( 74,165)@xmath137 ( 86,212)@xmath138 ( 112,212)@xmath139 ( 145,191)@xmath140 ( 151,182)@xmath141 ( 44,182)@xmath142 ( 49,189)@xmath143 ( 102,181)@xmath126 ( 89,176)@xmath127 ( 128,170)@xmath128 ( 94,146)@xmath129 ( 93,148)(-1,0)7 ( 78,108)@xmath144 sheet ( 16 , 316)@xmath145 ( -15 , 268)@xmath146 ( 145 , 316)@xmath147 ( 115,268)@xmath148 ( 14 , 97)@xmath149 ( -24 , 53)@xmath150 ( 25,0)@xmath151 sheet ( 155,0)@xmath152 sheet for example , denoting @xmath153 , we identify the edge @xmath154 of the @xmath155-th sheet with the edge @xmath154 of the @xmath156-th sheet . to compute the monodromy map @xmath157 we need the following lemma .
[ dehn twist lemma ] let @xmath158 for @xmath159 be a simple closed curve as in figure [ dehn - twist ] .
let @xmath160 be the right - handed dehn twist along @xmath113 .
then the lift @xmath161 of @xmath100 is @xmath162 ( where @xmath163 comes first and @xmath164 last )
. for simplicity , denote @xmath165 and @xmath166 .
we need to show that up to isotopy , @xmath167 ( 334,430 ) ( 0,0 ) and @xmath168 on @xmath169 .
, title="fig : " ] ( 14,418)@xmath170 ( 32,418)@xmath171 ( 78,418)@xmath171 ( 96,418)@xmath172 ( 201,418)@xmath173 ( 228,418)@xmath174 ( 266,418)@xmath174 ( 293,418)@xmath173 ( 14,306)@xmath175 ( 32,306)@xmath171 ( 78,306)@xmath171 ( 96,306)@xmath175 ( 201,306)@xmath176 ( 228,306)@xmath177 ( 266,306)@xmath177 ( 293,306)@xmath175 ( 337,288)@xmath163 ( 14,272)@xmath170 ( 32,272)@xmath171 ( 78,272)@xmath171 ( 96,272)@xmath172 ( 201,272)@xmath173 ( 228,272)@xmath174 ( 266,272)@xmath174 ( 293,272)@xmath173 ( 14,160)@xmath175 ( 32,160)@xmath171 ( 78,160)@xmath171 ( 96,160)@xmath175 ( 201,160)@xmath176 ( 228,160)@xmath177 ( 266,160)@xmath177 ( 293,160)@xmath175 ( 337,141)@xmath168 ( 14,124)@xmath170 ( 32,124)@xmath171 ( 78,124)@xmath171 ( 96,124)@xmath172 ( 201,124)@xmath173 ( 228,124)@xmath174 ( 266,124)@xmath174 ( 293,124)@xmath173 ( 14,12)@xmath175 ( 32,12)@xmath171 ( 78,12)@xmath171 ( 96,12)@xmath175 ( 201,12)@xmath176 ( 228,12)@xmath177 ( 266,12)@xmath177 ( 293,12)@xmath175 ( 28,386)@xmath126 ( 93,386)@xmath126 ( 221,386)@xmath126 ( 286,386)@xmath126 ( 28,239)@xmath126 ( 93,239)@xmath126 ( 221,239)@xmath126 ( 286,239)@xmath126 ( 28,91)@xmath126 ( 93,91)@xmath126 ( 221,91)@xmath126 ( 286,91)@xmath126 ( 28,336)@xmath127 ( 93,336)@xmath127 ( 221,336)@xmath127 ( 286,336)@xmath127 ( 25,189)@xmath127 ( 93,189)@xmath127 ( 221,189)@xmath127 ( 288,190)@xmath127 ( 25,42)@xmath127 ( 93,42)@xmath127 ( 223,42)@xmath127 ( 286,43)@xmath127 ( 11,373)@xmath178 ( 76,373)@xmath179 ( 270,373)@xmath180 cut @xmath112 into @xmath181 disks along oriented properly embedded arcs @xmath182 where @xmath183 and @xmath184 , dashed in figure [ dehn - twist ] .
we will check that after an isotopy , the map @xmath185 fixes each vertex and oriented edge of the graph @xmath186 .
our statement will then follow from the alexander method ( see e.g. ( * ? ? ?
* proposition 3.4 ) ) .
the alexander method is based on the observation that a diffeomorphism of @xmath93 fixing @xmath187 is isotopic to identity ; this observation is applied to each of the @xmath188 disks .
since the dehn twists are performed on curves @xmath189 which all lie in @xmath169 , we can assume that all the @xmath190-arcs except @xmath191 s are fixed by @xmath162 .
( 366,120 ) ( 0,10 ) . , title="fig : " ] ( 8,122)@xmath170 ( 26,122)@xmath171 ( 78,122)@xmath171 ( 96,122)@xmath172 ( 148,122)@xmath172 ( 166,122)@xmath192 ( 248,122)@xmath173 ( 275,122)@xmath174 ( 320,122)@xmath174 ( 346,122)@xmath170 ( 8,4)@xmath175 ( 26,4)@xmath193 ( 78,4)@xmath194 ( 96,4)@xmath175 ( 148,4)@xmath195 ( 166,4)@xmath194 ( 248,4)@xmath176 ( 275,4)@xmath177 ( 324,4)@xmath193 ( 346,4)@xmath176 ( 21,90)@xmath126 ( 90,90)@xmath126 ( 159,90)@xmath126 ( 269,90)@xmath126 ( 338,90)@xmath126 ( 19,36)@xmath127 ( 90,36)@xmath127 ( 159,37)@xmath127 ( 269,37)@xmath127 ( 338,37)@xmath127 ( 366,120 ) ( 0,10 ) obtained by finger moves applied to @xmath196.,title="fig : " ] ( 8,122)@xmath170 ( 26,122)@xmath171 ( 78,122)@xmath171 ( 96,122)@xmath172 ( 148,122)@xmath172 ( 166,122)@xmath192 ( 248,122)@xmath173 ( 275,122)@xmath174 ( 320,122)@xmath174 ( 346,122)@xmath170 ( 8,4)@xmath175 ( 26,4)@xmath193 ( 78,4)@xmath194 ( 96,4)@xmath175 ( 148,4)@xmath195 ( 166,4)@xmath194 ( 248,4)@xmath176 ( 275,4)@xmath177 ( 324,4)@xmath193 ( 346,4)@xmath176 ( 21,90)@xmath126 ( 90,90)@xmath126 ( 159,90)@xmath126 ( 269,90)@xmath126 ( 338,90)@xmath126 ( 19,36)@xmath127 ( 90,36)@xmath127 ( 159,37)@xmath127 ( 269,37)@xmath127 ( 338,37)@xmath127 therefore , we focus on @xmath169 shown in the top box of figure [ alpha - p ] to understand how the arcs @xmath191 change under the map @xmath185 .
( note that @xmath169 is the union of two shaded regions in figure [ dehn - twist ] , while in figure [ alpha - p ] one of these regions is shaded and the other is white ; the two regions are separated by the arcs @xmath191 ) .
the dehn twist @xmath163 changes the region @xmath169 , shown in the top box of figure [ alpha - p ] , to @xmath197 as in the second box
. then @xmath168 changes it to @xmath198 as in the bottom box .
applying all the dehn twists @xmath162 to @xmath169 , we obtain the region @xmath196 shown in figure [ pre - image2 ] .
next we isotope @xmath196 fixing the boundary of @xmath196 by a combination of two local finger moves near @xmath199 and @xmath200 , and obtain a region @xmath201 as in figure [ pre - image ] . to complete the proof ,
we observe that the region @xmath201 is precisely @xmath202 .
applying the above lemma repeatedly for different pairs of points @xmath96 , we can write down the monodromy of an arbitrary braid .
we denote the curve @xmath113 introduced in lemma [ dehn twist lemma ] by @xmath203 ( @xmath204 , @xmath205 when it is related to the twist of branch points @xmath96 and lies on the @xmath155-th and @xmath156-th sheets , and write @xmath206 for the right - handed dehn twist around @xmath203 . in particular , the @xmath113 curve in figure [ dehn - twist ] is renamed as @xmath207 , and the corresponding dehn twist is @xmath208 .
[ sigma1 ] let @xmath60 be the braid @xmath209 .
then the open book for the @xmath9-fold cover of @xmath60 given by lemma [ dehn twist lemma ] is the same as the open book for @xmath18 induced by the @xmath210-torus link fibration ; moreover , the images of the curves @xmath211 on the seifert surface of this torus link are as shown on figure [ torus - link ] .
each @xmath211 is an unknot in @xmath18 , with page framing @xmath212 .
we first observe that the branched @xmath9-fold cover of @xmath60 is @xmath0 .
this is easy to see : since @xmath60 is the transverse unknot with @xmath213 , it can be thought of as the binding of an open book decomposition of @xmath18 whose page is a disk .
the branched @xmath9-fold cover , then , is given by the same open book for any @xmath9 , yielding the standard contact structure on @xmath18 .
lemma [ dehn twist lemma ] produces a different open book for the @xmath9-fold cover of @xmath60 .
the page of this open book , together with the curves @xmath211 , can be embedded into @xmath18 as a seifert surface of the @xmath214-torus link shown on figure [ torus - link ] .
( 288 , 160 ) ( 0,0 ) of the open book with simple closed curves @xmath215 where @xmath216 and @xmath217.,title="fig : " ] ( 155 , 8)@xmath218 ( 127 , 40)@xmath219 ( 125 , 108)@xmath200 ( 127 , 144)@xmath199 ( 70 , 16)@xmath220 ( 123 , 10)@xmath221 ( 209 , 17)@xmath222 ( 60 , 50)@xmath223 ( 40 , 110)@xmath224 ( 90 , 110)@xmath225 ( 162 , 113)@xmath226 ( 210 , 140)@xmath152 sheet ( 31 , 144)@xmath130 sheet it is then clear that each @xmath227 is an unknot , with page framing = @xmath228 .
we claim that the torus knot fibration induces the monodromy of the open book given by lemma [ dehn twist lemma ] , i.e. the monodromy of this torus knot is the product of the dehn twists @xmath229 .
since the fiber surface of the torus knot can be obtained by plumbing together a sequence of right - handed hopf bands whose core curves are @xmath203 , it is clear that the monodromy of the torus knot is given by a composition of the right - handed dehn twists @xmath206 .
we need to determine the order in which the dehn twists are performed . and
@xmath230 lie on a fiber of the trefoil knot fibration ; @xmath231 and @xmath232 are their push - offs in the positive normal direction . ] to simplify the picture , we consider a model example where @xmath233 , @xmath234 .
let @xmath235 the right - handed trefoil knot and consider the fibration @xmath236 .
its monodromy is the product of the dehn twists around the curves @xmath237 and @xmath238 .
let @xmath239 , @xmath240 be pages of the corresponding open book .
assume that the curves @xmath54 and @xmath230 both lie on @xmath241 ; let @xmath231 and @xmath232 be their push - offs to the page @xmath242 for some small @xmath243 .
since @xmath244 is oriented as a mapping torus , this means that the curves are pushed off in the direction shown by arrow .
observe that @xmath231 and @xmath230 form a hopf link , while @xmath54 and @xmath232 are not linked .
suppose that the monodromy of the pictured trefoil is @xmath245 , and compose it with @xmath246 .
the result is of course the open book with trivial monodromy , which gives @xmath247 . on the other hand
, the composition of the two additional dehn twists corresponds to an integral surgery on @xmath18 performed on the link @xmath248 ( since @xmath249 follows @xmath250 , we need to place a copy of @xmath54 on the page following the page with @xmath230 ) .
the surgery coefficients are given by ( page framing)+1 , so we perform 0-surgery on both @xmath231 and @xmath230 ; but this surgery on the hopf link produces @xmath18 , not @xmath247 . by contrast , if we perform 0-surgeries on @xmath54 and @xmath232 which form a trivial link ( and correspond to composing the trefoil monodromy with @xmath251 ) , we obtain @xmath247 , so we conclude that the trefoil monodromy is @xmath252 .
similar argument for various pairs of curves
@xmath211 shows that the monodromy of the torus knot on figure [ torus - link ] is indeed @xmath253 .
the curves @xmath211 and @xmath254 , the push - off of @xmath255 , form a hopf link whenever @xmath256 or @xmath257 , and the trivial unlink otherwise .
( we return to this in remark [ rmk3.3 ] , see figure [ hopf - links ] for details . )
open books from the previous section will allow us to construct contact surgery diagrams for the branched covers . in proposition [ sigma1 ]
, we saw that the branched @xmath9-fold cover for the transverse braid @xmath258 is @xmath0 .
now consider a transverse @xmath16-braid @xmath259 , where @xmath37 is an arbitrary braid word . the branched @xmath9-fold cover for @xmath12 can be obtained from the branched cover for @xmath60 by performing additional dehn twists about curves @xmath260 in the open book decomposition of @xmath0 considered in lemma [ dehn twist lemma ] .
the goal of this subsection is to interpret these dehn twists as contact surgeries . forgetting the contact structure
, we can translate dehn twists into dehn surgeries along push - offs of the curves @xmath260 to successive pages of our open book .
we perform @xmath261-surgeries for left - handed and @xmath262-surgeries for right - handed dehn twists .
the order of push - offs is determined by the order of dehn twists , which in turn is dictated by the braid word @xmath37 and lemma [ dehn twist lemma ] . using honda s legendrian realization @xcite
, we can in principle find an isotopy that takes all @xmath260 to legendrian curves whose contact framing matches the page framing , so that @xmath261- and @xmath262-surgeries become contact @xmath62-surgeries .
this is almost what we need , but we want an explicit surgery diagram ; to this end , we give an explicit legendrian realization of our curves . indeed , following @xcite ( see ( * ? ? ?
* appendix ) for the same construction in the presence of a contact structure ) , we can embed the fiber surface of a torus link ( figure [ torus - link ] ) into @xmath18 as the page @xmath263 of an open book decomposition compatible with @xmath5 , and such that @xmath260 are all legendrian unknots with @xmath31 .
we simply draw this surface as in figure [ legreal ] ( assuming as usual that @xmath264 .
various legendrian push - offs of @xmath260 can then be thought of as lying on different pages of the same open book .
( 350 , 215 ) ( 0,0).,title="fig : " ] ( 55 , 15)@xmath265 ( 20 , 42)@xmath266 (-8 , 62)@xmath30 ( 117 , 78)@xmath223 to produce a contact surgery diagram of the @xmath9-fold branched cover for a transverse braid @xmath267 , we can now start with @xmath18 , write down the monodromy of the open book as dictated by the crossings of @xmath37 and lemma [ dehn twist lemma ] , and then perform legendrian surgeries on the successive legendrian push - offs of @xmath260 s , in the order corresponding to the order of dehn twists in the decomposition of the monodromy .
[ rmk3.3 ] in certain cases , it easy to see that the push - offs of different curves @xmath260 will be unlinked even if the curves themselves intersect on the surface @xmath263 .
indeed , consider the braid @xmath268 and the braid @xmath269 which differs from @xmath60 by an additional crossing .
the links @xmath60 and @xmath270 differ only in a small ball @xmath271 that contains this crossing ; the @xmath9-fold branched cover of @xmath271 is a genus @xmath19 handlebody , and the contact manifolds @xmath272 and @xmath273 differ only by a surgery on this handlebody .
( in fact , the surgery on the handlebody is equivalent to @xmath19 surgeries on the push - offs of @xmath215 where @xmath204 , corresponding to the given crossing ; the surgery curves are all contained in the handlebody . )
we also observe that @xmath271 can be thought of as neighborhood of the arc connecting two strands of the @xmath60 at the given crossing , and that the @xmath9-fold branched cover of @xmath271 is then a neighborhood of the branched @xmath9-fold cover of this arc . now , let @xmath274 and @xmath275 be two extra crossings added to @xmath60 , and @xmath276 and @xmath277 the corresponding arcs . untwisting the unknot @xmath60 , we can easily determine whether the lifts of @xmath276 and @xmath277 to the branched cover are linked ; if they are not , the corresponding surgery curves will not be linked either . if , however , @xmath276 and @xmath277 are linked , we have to examine the push - offs of the related curves @xmath215 to determine the surgery link .
we orient @xmath211 so that it goes from @xmath278 to @xmath279 on the @xmath155-th sheet of figure [ dehn - twist ] .
observe that @xmath280 ( see figure [ hopf - links ] . ) in all other cases , the curves @xmath215 and @xmath281 do not link each other .
( 340,150 ) ( 0,0 ) . in all cases
not shown , @xmath211 and @xmath282 do not link.,title="fig : " ] ( 110,130)@xmath283 ( 110,100)@xmath215 ( 110 , 60)@xmath215 ( 85 , 5)@xmath284 ( 170 , 100)@xmath285 ( 320 , 100)@xmath215 ( 180,30)@xmath215 ( 320,10)@xmath286 given a transverse @xmath16-braid @xmath12 ,
we can always write in the form @xmath287 , ( possibly after multiplying by the trivial word @xmath288 ) . we will use notation @xmath289 for the contact surgery diagram for the @xmath9-fold cover @xmath15 for a transverse @xmath16-braid @xmath12 , constructed by the above method .
thus @xmath289 is a collection of legendrian unknots equipped with contact surgery coefficients .
when @xmath9 is fixed , we often drop it from notation .
let @xmath239 , @xmath240 denote the pages .
examining the addition of an individual @xmath290 or @xmath291 to the braid word for @xmath12 , we obtain the following theorem .
[ surgthm ] fix @xmath8 .
suppose that @xmath287 is a transverse @xmath16-braid , and assume that @xmath292 for some @xmath293 .
pick @xmath294 .
denote the copy of @xmath295 in the page @xmath239 of the open book by @xmath296 .
\(1 ) suppose @xmath297 , @xmath298 .
define diagram @xmath299 as in figure [ u+ ] . then @xmath300 .
( 238,60 ) ( 0,0).,title="fig : " ] ( 30,63)@xmath301 ( 73,58)@xmath301 ( 120,52)@xmath301 ( 164,48)@xmath301 ( 206,43)@xmath301 ( -15,18)@xmath302 ( 33,9)@xmath303 ( 123,0)@xmath304 ( 169,-5)@xmath305
\(2 ) suppose @xmath306 , @xmath298 .
define diagram @xmath307 as in figure [ u- ]
. then @xmath308 . (
230,85 ) ( 0,0).,title="fig : " ] ( -5,0)@xmath309 ( 72,6)@xmath310 ( 168,33)@xmath311 ( 216,46)@xmath312 ( 24,38)@xmath313 ( 68,52)@xmath313 ( 112,66)@xmath313 ( 158,76)@xmath313 ( 203,91)@xmath313 ( here and below , we draw legendrian links as their front projections to the @xmath314 plane . ) the diagrams @xmath299 and @xmath307 may be linked to @xmath315 ; the way they link can be determined by drawing the corresponding legendrian push - offs of @xmath260 as explained in remark [ rmk3.3 ] .
this is a direct application of the algorithm developed above .
suppose that @xmath316 and @xmath317 ( i.e. , @xmath33 is a positive stabilization of @xmath12 representing the same transverse link )
. then @xmath318 .
note that every positively stabilized braid can be written in such form .
the braids @xmath12 and @xmath33 give rise to different initial open books for @xmath18 , the one corresponding to the braid @xmath319 and the other to @xmath320 , but the subsequent dehn twists corresponding to @xmath37 produce identical surgery diagrams .
[ cor - ot ] let @xmath316 , and @xmath321 ( i.e. , @xmath38 is a _ negative _ braid stabilization of @xmath12 , representing a transverse link stabilization ) .
define @xmath322 as in figure [ u - ot ] .
then the contact manifold represented by @xmath322 is an overtwisted 3-sphere , and @xmath323 .
we use the symbol `` @xmath324 '' to emphasize that @xmath322 is _ not _ linked to @xmath315 . [ u - ot ] ( 212,110 ) ( 0,10).,title="fig : " ] ( 23,0)@xmath325 ( 23,69)@xmath326 ( 77,16)@xmath327 ( 75,86)@xmath328 ( 178,46)@xmath329 ( 178,115)@xmath330 ( -12,43)@xmath313 ( -12,28)@xmath313 ( 77,44)@xmath313 ( 77,67)@xmath313 ( 128,82)@xmath313 ( 128,62)@xmath313 ( 178,77)@xmath313 ( 178,98)@xmath313 we write @xmath331 and apply part 2 of theorem [ surgthm ] twice . to show that the contact manifold represented by @xmath322 is an overtwisted 3-sphere
, we first use kirby calculus to see that the underlying smooth manifold is @xmath18 . using formula ( [ d3 ] ) , we compute @xmath332 .
( we have @xmath333 , @xmath334 , @xmath335 . )
since we know that @xmath5 is the unique tight contact structure on @xmath18 , and @xmath336 , it follows that the contact structure given by the diagram @xmath322 is overtwisted .
the branched cover of @xmath38 is then the connected sum of this overtwisted sphere and the branched cover of @xmath12 .
suppose that @xmath337 is an @xmath16-braid , and @xmath338 is an @xmath339-braid obtained from @xmath340 by an addition of a trivial @xmath339-th strand .
then @xmath341 , where @xmath342 is _ not _ linked to @xmath315 .
this follows from the identity @xmath343 , and the fact that the word @xmath37 does not contain @xmath344 .
we also observe that on the level of contact manifolds , we are taking a connected sum with @xmath345 , where the latter is equipped with its unique stein fillable contact structure .
it is now easy to obtain surgery diagrams of all @xmath9-fold branched covers of 2-braids .
a surgery diagram for the 5-fold cover of the transverse braid @xmath346 is shown in figure [ surgery - example ] . .
we perform legendrian surgery on each component . ] [ rem - different - choice ] even though every closed @xmath16-braid is isotopic to a braid containing a string @xmath347 , we may want to start with an open book corresponding to another version of transverse unknot , say @xmath348 ( this will be useful in subsection [ subsec - not ] ) . to obtain this other open book , we consider figure [ legreal ] and change the curves @xmath349 , so that they now go through the top two rows of the grid - like page .
this is shown on figure [ anotherlegreal ] ; the other curves @xmath350 for @xmath351 remain the same .
( 210 , 185 ) ( 0,0 ) produces an open book whose monodromy is @xmath352.,title="fig : " ] ( 37 , 15)@xmath265 ( 18 , 50)@xmath266 ( 3,75)@xmath30 ( 66 , 83)@xmath353 ( 168 , 133)@xmath354 for open books , this change corresponds to plumbing positive hopf bands together in a slightly different way to form the same page . analyzing the push - offs of the curves @xmath355 as in proposition [ sigma1 ]
, we see that the monodromy of the open book can now be expressed as @xmath356 , which by lemma [ dehn twist lemma ] corresponds to the braid @xmath352 as required .
another case worth mentioning is the initial unknot given by the braid @xmath357 . in this case
, we have the same open book as for the unknot @xmath358 , with the role of the curve @xmath359 played by @xmath360 . in principle , it is not necessary to single out the braid word that gives the unknot : we can as well start from the trivial braid and obtain @xmath361 as a result of surgery on @xmath362 . however , the presence of 1-handles seems to complicate matters .
in this section , we prove theorems [ tight - ow ] and [ homotp ] . the proofs are very similar to those of ( * ? ? ? * and 5 ) .
recall @xcite that a braid is called _ quasipositive _ if its braid word is a product of conjugates of the standard generators .
if @xmath12 is quasipositive , we can resolve a few positive crossings to convert the braid representing @xmath12 into a braid equivalent to a trivial one ( of the same braid index ) .
the @xmath9-fold cover branched over the trivial braid is a connected sum of several copies of @xmath363 , which is stein fillable ( @xmath364 here stands for the unique stein fillable contact structure on @xmath365 ) . putting the positive crossings back in , by lemma [ dehn twist lemma ] we see that the monodromy of the open book for @xmath17 is given by a composition of positive dehn twists .
it follows that the contact manifold is stein fillable .
the second part of the theorem follows from corollary [ cor - ot ] .
the fact that @xmath366 follows immediately : @xmath367 is the restriction to @xmath63 of the @xmath23 structure @xmath368 described in subsection [ cont - surgery ] ; @xmath81 evaluates as @xmath261 on each homology generator corresponding to either a @xmath57 or a @xmath20 surgery , because all surgeries are performed on standard legendrian unknots with rotation number @xmath261 . for
the second part of the theorem , suppose that two closed braids @xmath12 and @xmath33 are isotopic as smooth knots , and that @xmath369 . by the markov theorem for smooth knots @xcite
, @xmath33 can be obtained from @xmath12 by a sequence of braid isotopies and ( positive and negative ) braid stabilizations and destabilizations . braid isotopies and positive stabilizations preserve both @xmath2 and the @xmath74 invariant , since they do not change the transverse link type .
each negative stabilization ( resp . destabilization ) decreases ( resp .
increases ) the self - linking number by @xmath79 and the @xmath74 invariant by @xmath370 , since , as we saw in corollary [ cor - ot ] , transverse stabilization gives the connected sum with the overtwisted sphere in figure [ u - ot ] .
but if @xmath369 , every negative stabilization must be compensated by a negative destabilization .
it follows that @xmath371 .
fix @xmath8 .
let @xmath235 be a transverse link smoothly isotopic to the @xmath372 torus link , @xmath373 .
the branched cover @xmath374 is then the brieskorn sphere @xmath375 . if @xmath376 , then @xmath377 is stein fillable . otherwise @xmath377 is overtwisted . for different values of @xmath378 , these overtwisted structures have different @xmath74 invariants . by classification of transverse torus links , @xmath235 is given by a positive braid when it has the maximal self - linking number ;
otherwise it is isotopic to this link transversely stabilized @xmath379 times ( for some @xmath380 ) .
the number of stabilizations @xmath379 can be determined by the self - linking number .
we generalize the second part of theorem [ tight - ow ] , and show that the branched covers are overtwisted for a large family of transverse links .
[ neg - crossings ] suppose that the transverse link @xmath12 is represented by a closed braid such that its braid word @xmath381 contains a factor of @xmath382 but no @xmath383 s for some @xmath384 .
( this means that all the crossings in the braid diagram on the level between the @xmath385-th and the @xmath386-th strands are negative . )
then the branched @xmath9-fold cover @xmath15 is overtwisted for any @xmath8 .
we will use the right - veering monodromy criterion for tightness @xcite . more precisely
, we will show that the monodromy of the open book for the branched cover of @xmath12 given by lemma [ dehn twist lemma ] is not right - veering , which by ( * ? ? ?
* theorem 1.1 ) implies that the corresponding contact structure is overtwisted .
( this is in fact the criterion for overtwistedness that was first given in @xcite in terms of `` sobering arcs '' . )
the negative crossings between @xmath385-th and @xmath386-th strands yield the open book @xmath387 such that the dehn twists about curves @xmath295 are all left - handed for @xmath388 . , @xmath389 , and a left - handed hopf band .
] remove from @xmath12 all the ( negative ) crossings between the @xmath385-th and the @xmath386-th strands ( in other words , remove all the negative factors of @xmath383 from the braid word @xmath381 ) .
the link @xmath12 then splits into two links @xmath390 and @xmath271 such that the corresponding braid words @xmath391 and @xmath392 contain only generators @xmath393 with @xmath394 resp .
@xmath395 . let @xmath396 , @xmath389 be the open books for the branched covers @xmath397 and @xmath398 , and consider the plumbed sum of @xmath396 , @xmath389 , and a left - handed hopf band .
the resulting open book is shown on figure [ destab ] , and is in fact a negative stabilization of the connected sum @xmath399 .
it is easy to see that the monodromy of a negatively - stabilized open book is not right - veering .
the open book @xmath387 is obtained from this non - right - veering open book by a sequence of negative stabilizations and additional left - handed dehn - twists , and so can not be right - veering either ( because a composition of a right - handed dehn twist and a right - veering monodromy is right - veering ) .
if the branched @xmath9-fold covers of two transverse links @xmath3 and @xmath4 of the same topological type are both overtwisted and @xmath400 , then theorem [ homotp ] together with eliashberg s classification of overtwisted contact structures implies that @xmath401 is contactomorphic to @xmath402 .
we therefore have [ bmcont ] in table 1 in @xcite of transverse knots , all pairs ( except perhaps for the representatives of the knot @xmath403 ) give rise to contactomorphic @xmath9-fold branched covers for all @xmath8 . in view of the previous remark ,
showing that certain branched covers are overtwisted can be useful .
we thus illustrate two other ways to establish overtwistedness ( our examples below are all included in proposition [ neg - crossings ] , but the methods can be used for other links as well ) .
the first method applies in the rare cases where the classification of tight contact structures is known for the smooth manifold @xmath15 .
for example , this is the case for double covers of 2-bridge links : it is well known that these are lens spaces , and the tight contact structures on lens spaces were classified in @xcite .
consider the transverse 2-braid @xmath404 where @xmath405 ; its branched double cover is the lens space @xmath406 , with the contact structure @xmath32 given by the surgery diagram on figure [ k - neg ] ( where @xmath20 contact surgery is performed on each of @xmath407 successive push - offs of the legendrian unknot of @xmath31 ) . we compute the @xmath74 invariant of this contact structure . if @xmath69 is the 4-manifold corresponding to the surgery , then @xmath408 ( indeed , the intersection form for @xmath69 has zeroes on the diagonal and @xmath228 s for all other entries ; it is easy to see that the matrix has an eigenvalue @xmath409 of order @xmath155 and an eigenvalue @xmath410 of order 1 ) .
we also have @xmath411 , and @xmath412 .
therefore , from ( [ d3 ] ) we obtain @xmath413 . on the other hand , by @xcite , the lens space @xmath414 carries a unique tight contact structure @xmath364 ; this contact structure is the boundary of a linear plumbing ( also shown on figure [ k - neg ] ) .
the corresponding stein 4-manifold @xmath415 has @xmath416 , @xmath417 and @xmath418 , so @xmath419 .
it follows that the contact structure @xmath32 is not isotopic to @xmath364 , and therefore must be overtwisted .
( left ) and the unique tight contact structure on @xmath414 ( right ) . ] another way to prove overtwistedness is simply to find an overtwisted disk in the surgery diagram .
admitting that these pictures get unwieldy even for simple links , we exhibit such a disk for the overtwisted sphere @xmath420 described in figure [ u - ot ] ( i.e. the branched @xmath9-fold cover of @xmath421 ) .
indeed , the surface @xmath44 shown on figure [ otdisk ] induces the @xmath261-framing on each component of the surgery link @xmath420 , and the @xmath262-framing on the legendrian knot @xmath60 .
( we assume that all legendrian knots are oriented as the boundary of @xmath44 ) .
since @xmath20-contact surgery is @xmath261-framed dehn surgery , @xmath44 becomes a disk bounded by @xmath60 in the surgered manifold .
then the equality @xmath422 implies that this disk is overtwisted .
( 288 , 144 ) ( 0,0)otdisk ( 90 , 100)@xmath60 ( 213 , 46)@xmath44
we can now use the constructions from previous sections to examine the branched covers of certain transverse knots and prove theorem [ introthm ] ( see corollaries [ cor - a ] , [ cor - b ] , and theorem [ neg - flype - thm ] ) .
we already saw that for most examples of @xcite , the branched covers do not detect the difference between transverse knots .
we now consider the remaining pairs of non - isotopic transverse knots with the same classical invariants from @xcite , @xcite , @xcite , and try to distinguish them via the corresponding contact structures .
( top left ) and @xmath4 ( top right ) .
we assume that @xmath428 ; a @xmath20 contact surgery is performed on unknots marked with @xmath78 , legendrian surgery on all other components .
a diagram for the branched 3-fold cover of @xmath4 is shown at the bottom ; branched 3-fold cover of @xmath3 is obtained by contact surgery on a mirror of this link . to obtain @xmath9-fold covers , take @xmath19 copies of the surgery link for the double cover linked in a way similar to the 3-fold cover case . ]
we write the closed braids as @xmath429 observe that @xmath4 can be taken to @xmath430 by a transverse isotopy . using the method in theorem [ surgthm ] and the following corollaries
, we can draw surgery diagrams for the branched covers of @xmath3 and @xmath4 .
for example , double covers for the case where @xmath428 are shown on figure [ l1l2cover ] ( top ) ; we see that they are obtained by contact surgeries on two links which are legendrian mirrors of one another .
similarly , @xmath9-fold branched covers for @xmath3 and @xmath4 are also obtained by surgery on legendrian mirrors , since the corresponding diagrams are obtained by taking @xmath19 copies of the surgery link for the double cover linked as dictated by figure [ hopf - links ] . for example , the triple cover for @xmath4 is shown on figure [ l1l2cover ] ( bottom ) . for negative @xmath431 , @xmath432 or @xmath433 the pictures are similar ; besides , the case @xmath434 is already covered by corollary [ bmcont ] .
we have shown that the surgery link diagram @xmath435 for @xmath4 is the legendrian mirror of the link @xmath436 for @xmath3 .
now , observe that one link is taken to the other by the map @xmath437 .
this map reverses the sign of the standard contact form @xmath438 ( i.e. the orientation of contact planes on @xmath18 ) and extends to the map of branched covers that takes @xmath426 to @xmath439 .
alternatively , the previous proposition can be proved by using open books .
a careful examination of the monodromy shows that the open book for @xmath440 can be obtained from the open book for @xmath441 by reversing the orientation of the pages as well as the orientation of the @xmath442 direction in the mapping torus .
this operation preserves the orientation of the 3-manifold but reverses the orientation of contact planes .
let the braid @xmath4 be obtained by reading the braid word @xmath3 backwards , i.e. , if @xmath443 then @xmath444 .
then the contact structures @xmath426 and @xmath445 are conjugate to one another for any @xmath8 .
write @xmath446 then the braid word for @xmath4 is conjugate to @xmath447 in the surgery diagram for cover of @xmath3 , the part of the surgery link corresponding to @xmath448 will be above that for @xmath449 when @xmath450 ; for cover of @xmath4 , it will be below . in both cases ,
the surgery unknots corresponding to @xmath448 and @xmath449 with @xmath450 will be linked ( in exactly the same way ) iff @xmath451 ; using the braids - to - surgeries description from section [ surg for cov ] , cf .
figure [ hopf - links ] , we see that in fact the surgery links for the two branched covers are legendrian mirrors of one another .
it follows that the resulting contact structures @xmath426 and @xmath445 are conjugate to one another .
( 309,354 ) ( 0,0)reflection ( 40 , 355)@xmath130 sheet ( 230 , 355)@xmath152 sheet ( 40,228)@xmath455 sheet ( 220,228)@xmath456 sheet ( 40 , 0)@xmath152 sheet ( 230,0)@xmath130 sheet ( 40 , 323)@xmath457 ( 280 , 280)@xmath222 ( 40 , 70)@xmath458 ( 280 , 47)@xmath220 ( 155 , 175)@xmath459 we need to find an involution of the smooth manifold @xmath15 that induces the orientation reversal on contact planes . for a page @xmath460 of the open book described in lemma [ dehn twist lemma ]
, there is an orientation - reversing map @xmath461 that maps @xmath155-th sheet to the @xmath462-th sheet , acting as a reflection , and takes the curve @xmath215 to the curve @xmath463 ( see figure [ reflect ] ) .
if @xmath9 is odd , the @xmath464-th sheet is mapped to itself , and if @xmath16 is even , the curve @xmath465 is mapped to itself . moreover , @xmath466 , i.e. the involution @xmath459 takes right - handed dehn twists to the left - handed ones .
if @xmath467 is the lift of the half - twist @xmath393 as in figure [ sigma ] , we have @xmath468 write @xmath381 for the braid word for @xmath12 , and let @xmath469 be the braid word obtained by changing every half - twist generator @xmath393 to @xmath470 . the braids @xmath381 and @xmath469 are related by a braid isotopy ( rotating the braid ) , so if @xmath33 is the transverse link corresponding to the braid @xmath469 , then @xmath12 and @xmath33 are transversely isotopic .
however , we have @xmath471 if we extend the map @xmath459 to an orientation - preserving map @xmath472 \to p\times [ 0,1]$ ] , defined by @xmath473 , @xmath474 descends to open books , taking the open book @xmath475 to the open book @xmath476 .
the latter open book is compatible with the contact structure @xmath477 which is isotopic to @xmath478 .
it follows that @xmath75 and @xmath478 are isomorphic .
[ cor - a ] let @xmath12 be a legendrian link , @xmath13 its legendrian mirror , and consider the transverse push - offs @xmath479 and @xmath480 .
then the corresponding @xmath9-fold branched covers are contactomorphic for all @xmath9 .
the double branched covers of the birman
menasco 3-braids were studied in @xcite .
it was shown that these double covers are contactomorphic ; note , however , that there is a gap in the proof of ( * ? ? ?
* theorem 5 ) . in @xcite ,
transverse knots are given as push - offs of legendrian knots , and the latter are represented by grid diagrams of their ( smooth ) mirrors .
we recall how to obtain a positive transverse push - off of a legendrian knot given by such a grid diagram ( cf .
first , let the horizontal segments in the diagram go over the vertical segments ( this is opposite to the convention for grid diagrams and produces a front projection for the legendrian knot ) .
then keep every vertical segment oriented upwards ( i.e. has o above x ) , and replace every vertical segment oriented downwards by the complementary vertical segment .
the result is a braid that goes from the bottom to the top of the diagram and represents the positive push - off of the given legendrian knot . to obtain the braid for the negative transverse push - off , reverse the orientation of the legendrian knot ( by replacing o s by x s and vice versa in the grid diagram ) , and
repeat the above procedure .
we consider transverse push - offs @xmath483 and @xmath482 of the legendrian representatives of the pretzel knot @xmath484 , ( * ? ? ?
* figure 4 ) .
these are given by transverse closed braids @xmath485 a braid isotopy takes these braids to @xmath486 we can now draw surgery diagrams for the double branched covers of @xmath483 and @xmath482 ; they are shown on figure [ fig4ngozth ] .
recall remark [ rem - different - choice ] and figure [ anotherlegreal ] .
note that the two surgery links differ only in the circled region ; this corresponds to the fact that the braids for @xmath483 and @xmath482 differ only by exchanging two generators @xmath487 and @xmath100 ( together with a different choice of the open book ) .
we observe that the surgery links are in fact legendrian isotopic .
the isotopy can be performed via a sequence of legendrian reidemeister moves indicated on figure [ rmoves ] .
the transverse push - offs @xmath488 and @xmath489 of the legendrian representatives of the pretzel knot @xmath490 , ( * ? ? ?
* figure 5 ) can be treated in the same way . indeed , these are given by braids @xmath491 braid isotopic to @xmath492 as in the previous example , we make a different choice of the initial unknot , and then switch the two factors of @xmath100 and @xmath487 to relate the braids .
the surgery diagrams are very similar to the previous case ; the surgery links have more surgery components , but differ only in the circled region exactly as above , and can be related by a sequence of reidemeister moves .
it is conjectured in @xcite that all pretzel knots @xmath493 are not transversely simple , and if @xmath494 , @xmath495 are the legendrian representatives of @xmath493 similar to those considered above , then @xmath494 and @xmath495 are not transversely isotopic . our argument , however , clearly generalizes to show that the corresponding branched double covers are contactomorphic . moreover , our argument for the knots @xmath483 and @xmath482 will work for any two braids of the form @xmath496 where @xmath433 is any braid word on generators @xmath497 , and @xmath498 . indeed , such two closed braids are isotopic to @xmath499 and the corresponding surgery diagrams differ by the same local change as above , except that instead of the single legendrian unknot to be moved we have @xmath500 copies of its legendrian push - offs .
a similar sequence of reidemeister moves can be used to perform this local change .
we observe that two such braids are in fact related by a negative flype .
we thus have [ k1-k2 ] let @xmath501 , @xmath502 be two transverse braids related by the special kind of negative flype satisfying @xmath503 where @xmath433 is a word in @xmath497 , and @xmath72 is an integer .
then the branched double covers of @xmath501 and @xmath502 are contactomorphic .
the case @xmath498 is considered above .
when @xmath504 , the @xmath9-fold cyclic branched covers for @xmath501 and @xmath502 of any @xmath9 are overtwisted by proposition [ neg - crossings ] .
since they share the homotopy invariants , thus they are contactomorphic .
[ neg - flype - thm ] suppose @xmath501 and @xmath502 are related each other by a negative flype move sketched in figure [ neg - flype ] , i.e. , @xmath505 where @xmath432 and @xmath433 are any braid words in generators @xmath497 and @xmath506
. then the branched double covers @xmath507 and @xmath508 are contactomorphic .
consider positive stabilizations of @xmath501 and @xmath502 . since
a positive stabilization preserves transverse knot type , we use the same notations @xmath510 .
let @xmath511 ( resp .
@xmath512 ) be the braid words in @xmath513 obtained from @xmath432 ( resp .
@xmath433 ) by translation @xmath514 .
then we have @xmath515 similarly we have @xmath516 thus they satisfy the condition of proposition [ k1-k2 ] .
let @xmath6 ( resp .
@xmath517 ) be the legendrian @xmath518 ( resp .
@xmath519 ) knots studied in @xcite .
let @xmath520 be the legendrian @xmath521-cables of the @xmath521-torus knot found in @xcite @xcite .
the positive push - offs of every pair satisfy the condition of theorem [ neg - flype - thm ] .
therefore , double branched covers for each pair are contactomorphic .
similarly , the closed braids @xmath524 only differ in the dashed boxes sketched in figure [ neg - flype-2 ] and are related to each other by a negative flype .
( here we use the legendrian fronts for @xmath520 given in @xcite . )
we are unable to determine whether the branched double covers for some of the examples in @xcite are contactomorphic .
however , we can show that the heegaard floer contact invariants @xcite also fail to distinguish between these contact manifolds .
we use the following result of lawrence roberts @xcite , conjectured in @xcite . for a transverse link @xmath12 ,
let @xmath531 be its transverse invariant in reduced khovanov homology , @xcite . recall that by @xcite , there is a spectral sequence that relates the reduced khovanov homology of @xmath12 to the heegaard floer homology @xmath532 ( when one works with coefficients in @xmath533 .
then the element @xmath531 canonically corresponds to the heegaard floer contact invariant under this spectral sequence @xmath534 .
our result now follows from the fact that @xmath535 for all examples in question @xcite .
indeed , for these pretzel knots khovanov homology in the relevant bi - degree has rank 1 , and @xmath536 because all braids are quasipositive @xcite .
j. etnyre , _ lectures on open book decompositions and contact structures _ , floer homology , gauge theory , and low - dimensional topology , 103141 , clay math .
5 , amer . math . soc . , providence , ri , 2006 . e. giroux , _ gomtrie de contact : de la dimension trois vers les dimensions suprieures _ , in : _ proceedings of the international congress of mathematicians , vol .
ii ( beijing , 2002 ) _ , higher ed . press , beijing , 2002 , 405414 . | we study contact manifolds that arise as cyclic branched covers of transverse knots in the standard contact 3-sphere .
we discuss properties of these contact manifolds and describe them in terms of open books and contact surgeries . in many cases
we show that such branched covers are contactomorphic for smoothly isotopic transverse knots with the same self - linking number .
these pairs of knots include most of the non - transversely simple knots of birman menasco and ng ozsvth thurston . |
since the discovery of high - temperature superconductivity in cuprates , it has been a key interest to find the mechanism that causes the high @xmath5 and its unconventionality . after more than 20 years of intensive investigations , however , the origin of high-@xmath5 superconductivity in cuprates is still unsolved .
the discovery of relatively high @xmath5 in iron - based superconductors @xcite opens a different route for high - temperature superconductivity and thus provides a unique opportunity to re - clarify the physics of the unconventional superconductivity .
one of the features for the unconventional superconductivity is the appearance of the sign - switched gap function .
the presence of disconnected fermi surface ( fs ) sheets in iron - based superconductors is an ingenious condition for the existence of the sign - switched order parameters such as @xmath6-wave and _
d_-wave @xcite .
it is widely believed that the superconducting ( sc ) pairing is driven by the inter - fs - sheet scattering or equivalently by the antiferromagnetic ( afm ) fluctuations @xcite as the inter - fs - sheet scattering also leads to a strong afm correlation . the recent discovery of alkaline - intercalated iron - selenide superconductor a@xmath7fe@xmath1se@xmath2 with @xmath5 above 30k @xcite initiates an interesting branch for exploring superconductivity in iron - based superconductors . the chemically - stable structure a@xmath8fe@xmath9se@xmath2 exhibits a @xmath10 iron - vacancy order and block - checkerboard antiferromagnetism @xcite . by varying contents of iron or alkaline ,
the afm state becomes unstable and the sc state appears @xcite .
more recently , it is indicated that the antiferromagnetic phase and the superconducting phase may be separated in different space regions@xcite .
in particular , a scanning tunneling microscopy ( stm ) measurement @xcite shows that the sample would be spontaneously separated into the superconducting kfe@xmath2se@xmath2 and the insulating k@xmath11fe@xmath9se@xmath2 .
nevertheless , it still needs further clarification to see if the insulating region can be altered into a metal by chemical doping or by applying pressure .
first - principles calculations indicate that by applying pressure to samples with the @xmath10 iron - vacancy order , the system goes through two successive magnetic transitions , from the semiconducting block - checkerboard afm phase to the metallic stripe ( collinear ) afm phase and then to a metallic non - magnetic phase @xcite .
these results agree with transport measurements and indicate that a semiconductor - to - metal transition can be induced by the pressure@xcite .
the issue of whether the iron - vacancy - ordered a@xmath0fe@xmath1se@xmath2 supports superconductivity , however , remains unsettled as it is shown that as the system became more metallic under pressure , @xmath5 gets smaller , and whether it suggests the necessity of strong correlation for the emergence of superconductivity is an open question . on the other hand , nuclear magnetic resonance ( nmr ) experiments find singlet superconductivity and no pronounced spin fluctuations near @xmath5 @xcite .
the spin - lattice relaxation rate 1/__t__@xmath12_t _ shows the absence of the hebel - slichter coherence peak and a power - law behavior below @xmath5 , indicating an unconventional sc gap which is very likely gapless .
although salient spin fluctuations are not present in nmr , the possibility of spin - fluctuation - driven superconductivity should not be excluded since the information of spin fluctuations would be hidden in signals@xcite .
a recent raman measurement on the two - magnon scattering shows that the scattering rate grows as the temperature approaches @xmath5 and undergoes a sudden drop when the system enters the sc phase @xcite .
it thus supports that superconductivity arises from magnetic fluctuations .
theoretically , previous works @xcite on superconductivity in a@xmath0fe@xmath1se@xmath2 were based on the band structure of kfe@xmath2se@xmath2 @xcite .
however , the iron - vacancy order has been proved to induce a big change in the fs shape @xcite , though it is not confirmed so far by the angle - resolved photoemission spectroscopy ( arpes ) @xcite .
therefore , if superconductivity could happen in the iron - vacancy ordering phase , it is crucial to examine the pairing symmetry and pairing mechanism . in this work
, we investigate superconducting instability in iron - vacancy - ordered a@xmath0fe@xmath1se@xmath2 .
previously , based on the 12-orbital tight - binding model that fits the band structure of k@xmath0fe@xmath13se@xmath2 , we have studied the magnetic phase in the generalized hubbard model and succeeded in explaining the block - checkerboard afm instability from the stoner s theory @xcite .
here based on the same tight - binding model and starting from a non - magnetic metallic phase , we examine the superconductivity by the fluctuation - exchange ( flex ) approach @xcite . for singlet superconductivity , we examine the pairing coupling constants for _ s_-wave and _ d_-wave states from the effective scattering matrix .
we find that similar to other iron - pnictide superconductors @xcite , _ s_-wave and _ d_-wave are closely degenerate .
furthermore , the gap functions in both symmetries are highly anisotropic and have nodes . in addition , since the iron - vacancy order lowers the symmetry of a@xmath0fe@xmath1se@xmath2 to the group @xmath3 , the gap functions do not need to be reflection symmetric and this implies that there are degrees of freedom in defining origins of angles for the pairing momenta .
these extra degrees of freedom for the gap functions can further increase the gain of the condensation energy .
in addition , we find that the pairing symmetry is controllable by tuning the hund s rule coupling @xmath4 and the critical value at about @xmath14 .
the hund s rule coupling manipulates the phase transition or crossover in superconductivity , antiferromagnetism @xcite , or even the metallic transport @xcite , reflecting the substantial role played by the orbital - correlation in the multi - orbital electronic systems .
the vacancy ordered iron selenide a@xmath0fe@xmath1se@xmath2 is a system with one - fifth of fe being taken off and forming a characteristic @xmath15 pattern ( fig .
[ lattice](a ) ) .
the presence of the iron vacancy changes the space group from _ i4/mmm _ ( @xmath16 ) to _ i4/m _ ( @xmath3 ) . as a result
, the four - fold rotational symmetry is retained without the reflection symmetry with respect to the _ xz _ , _ yz _ and the diagonal planes .
the absence of reflection symmetry implies that there is no symmetry axis .
hence there is an extra degree of freedom in defining the zero value of angles for basis functions of @xmath3 .
for instance , the basis function @xmath17 becomes @xmath18 $ ] with @xmath19 being an free parameter . to implement the _
symmetry for a@xmath0fe@xmath1se@xmath2 , we have constructed a 12-orbital tight - binding model with four fe atoms ( labeled by _ a _ , _ b _ , _ c _ , and
_ _ d _ _ * * ) * * per cell and three @xmath20 orbitals ( @xmath21 , @xmath22 , and @xmath23 with x and y referring to fe - fe directions ) per fe to investigate the magnetic instability @xcite .
the resulting fs is shown in fig .
[ lattice](b ) with two hole pockets , @xmath24 the one centering on ( 0,0 ) and @xmath25 on ( @xmath26,@xmath26 ) , and four electron pockets , @xmath27 the one and its inversion around @xmath28(@xmath29,@xmath30 ) and @xmath31 around @xmath28(@xmath32,@xmath29 ) . to calculate the sc gap , we shall divide the momentum _
k _ space into a lattice with 200@xmath33 200 points and approximate states on the fs sheets by picking totally 216 _
k _ points with 44 points on fs-@xmath24 , 40 points on fs-@xmath25 , and 66 points both on fs-@xmath27 and @xmath34 .
these _ k _ points on fs sheets are characterized by angles , @xmath35 , @xmath36 , @xmath37 , and @xmath38 as illustrated in fig .
[ lattice](b ) .
all angles are relative to the horizontal @xmath39 axis except that @xmath40 is relative to the vertical @xmath41 axis due to that fs-@xmath31 is a 90@xmath42 rotation from fs-@xmath27 . on the average ,
the energy resolution in the * k * space with 200@xmath33 200 points within a 50mev window relative to the fermi energy is 0.6mev , while for the chosen 216 _ k _ points , the momentum resolution is about @xmath43 and we estimate the corresponding energy resolution is about 4mev .
vacancy ordering .
the solid circles denote fe atoms and the empty ones the vacancies .
every four nearest vacancies enclose the unit cells .
each unit cell contains four fe atoms , labeled by _ a _ , _ b _ , _ c _ , and _
d_. @xmath44 and @xmath45 are the primitive vectors .
( b ) the fermi surface results from the 12-orbital tight - binding model : two hole pockets ( @xmath46 and @xmath47 ) and four electron pockets ( @xmath48 and @xmath49 and their inverse partners ) . 216 _
k_-points on the fermi surfaces are selected and shown by circles ( zoom in to see ) .
points on fs-@xmath46 , @xmath47 , @xmath48 and @xmath49 are parameterized by the angles , @xmath50 , @xmath51 , @xmath52 , and @xmath53 , respectively.,width=352,height=247 ] we shall assume that the interaction between electrons on each fe atom is given by the generalized hubbard model , @xmath54 \right\ } , \label{h_i}\end{aligned}\]]where _
i _ is the index for four fe atoms and _ a _ is the index for three @xmath20 orbitals .
it was found @xcite that as the interaction turns on , the spin fluctuation is strong at @xmath55(@xmath26,@xmath26 ) and ( @xmath26,@xmath56 ) , in which the former is related to the checkerboard antiferromagnetism and the latter is related to the stripe antiferromagnetism .
these two magnetic states competes each other and the preference between them is controlled by the hund s rule coupling , @xmath4 .
the strong spin fluctuation is a possible pairing mechanism . to investigate the superconductivity arising from the exchange of spin and charge fluctuations , we follow the flex approach @xcite , in which the effective singlet pairing scattering matrix is given by @xmath57 _ { cd}^{ab}\text { } , \notag\end{aligned}\]]where
@xmath58 is a real frequency and _ a , b , c _ , and _
d _ as before are orbital indices . in eq . ( [ flex ] ) , the spin and charge vertices @xmath59 and @xmath60 are @xmath61 , @xmath62 , @xmath63 , @xmath64 , and @xmath65 , @xmath66 , @xmath67 , @xmath68 , where non - vanishing vertices are only between the same fe , and @xmath69 labels orbitals and @xmath70 . in the following
, we will take the relations @xmath71 and @xmath72 .
the random phase approximation ( rpa ) for spin and charge susceptibilities are given by @xmath73 ^{-1 } , \\
\chi _ { rpa}^{c}(\mathbf{q},\omega ) & = \chi _ { 0}(\mathbf{q},\omega ) \left [ 1+\gamma _ { c}\chi _ { 0}(\mathbf{q},\omega ) \right ] ^{-1},\end{aligned}\]]with the bare susceptibility being given as @xmath74 _ { ab}^{cd } & = -\frac{1}{n}{\sum_{\mathbf{k},\mu , \nu } } a_{c\mu } ( \mathbf{k})a_{a\mu } ^{\ast } ( \mathbf{k})a_{b\nu } ( \mathbf{k+q})a_{d\nu } ^{\ast } ( \mathbf{k+q } ) \notag \\ & \times \frac{n_{f}\left [ e_{\mu } ( \mathbf{k})\right ] -n_{f}\left [ e_{\nu } ( \mathbf{k+q})\right ] } { \omega + e_{\mu } ( \mathbf{k})-e_{\nu } ( \mathbf{k+q})+i\delta } , \end{aligned}\]]where @xmath75 is the orbital - band transformation matrix from @xmath76 * and * @xmath77 is the fermi - dirac distribution function . in this work
, we shall set the temperature to zero . here
@xmath78 is calculated by using the bare green s function not the dressed one .
the low energy physics of the scattering matrix is projected to scattering among the fs sheets .
we shall denote the scattering matrix with a cooper pair from @xmath79 on fs-@xmath80 scattered to @xmath81 on fs-@xmath82 by @xmath83 @xcite , @xmath84since we consider the even - parity pairing , we shall symmetrize the scattering matrix as @xmath85 $ ] . for a given gap function @xmath86 , a dimensionless coupling constant
is defined by @xcite @xmath87 here @xmath88and @xmath89are restricted within fs @xmath90 and @xmath91 , respectively .
@xmath92 is the fermi velocity , and @xmath93 ( = 216 ) is the number of @xmath94 on fermi surfaces we choose .
the coupling constant includes contributions from pocket - pocket scattering processes , and it is helpful to extract them and to denote by a matrix @xmath95 , satisfying @xmath96 , where @xmath90 , @xmath91 refer to fs sheets , @xmath90 , @xmath91=1 to @xmath97 , 2 to @xmath98 , 3 to @xmath99 , and 4 to @xmath100 . the stationary condition in eq .
( [ lambda ] ) ( @xmath101 /\delta g(\mathbf{k})=0 $ ] ) leads to the eigenvalue problem @xmath102 where the subscript @xmath103 stands for different solutions . for a given eigenvalue @xmath104
, its eigenfunction @xmath105 determines the symmetry of the gap , which could be _ s_-wave or _ d_-wave .
s_-wave state is characterized by @xmath106 and _
d_-wave by @xmath107 , where @xmath108 is the 90@xmath109 -rotational operation on @xmath94 . among the _ s_-wave solutions
, we pick up the largest one and set it to be @xmath110 and its eigenfunction to be @xmath111 .
similarly , for the _ d_-wave , they are denoted by @xmath112 and @xmath113 .
the eigenvalue equation eq .
( [ lambda ] ) is not identical to the bethe - salpeter equation in which magnitude of unity for the eigenvalue stands for the formation of superconductivity .
instead , eigenvalues in eq .
( [ lambda ] ) stand for the pairing strength .
therefore , the sc state is determined by largest one of @xmath110 and @xmath112 .
the scattering matrix shows some clues to the gap function . in order to gain more condensation energy ,
the gap should be larger in the region where the scattering matrix is maximal in magnitude . as the interaction is repulsive
, the gap becomes anisotropic and sign - changing .
[ scattering_matrix ] shows the `` weighted '' scattering matrix @xmath114 ^{-1}\widetilde{\gamma } _ { \mu \nu } ^{even}(\mathbf{k , k}^{\prime } ) $ ] for _ _ u__=1.0ev and @xmath4=0.2_u_. here because the inverse fermi velocity ( ifv ) , @xmath115 , is related to the density of states ( dos ) , it is used as a weighting factor at each point .
the scattering matrix contains intra and inter pocket scattering channels and different pockets are discriminated by the white dotted lines .
the angles , @xmath35 , @xmath36 , @xmath37 and @xmath40 parameterizing momenta @xmath94 ( @xmath116 ) , increase from 0@xmath42 to 360@xmath42 as the coordinate increases from the bottom ( left ) to the top ( right ) .
although we only show the result from one of the electron pocket @xmath27 ( @xmath34 ) , due to the nature of being even parity for @xmath117 , the result at the inversion point can be deduced . in fig .
[ scattering_matrix ] , the intra - electron - pocket ( @xmath27-@xmath27 or @xmath31-@xmath31 ) scattering is much stronger than other channels and gives peaks at two momenta , @xmath118 and @xmath119 as the antipodes ( on the major axis ) of the electron pockets .
the strong intensity of the scattering matrix at these two momenta is expected because large ifv happens about the antipodes of the elliptic fs ( @xmath115 is shown in fig .
[ s - wave ] ) .
one of the reasons why intra - electron - pocket scattering is strong is due to the much larger dos on the electron pockets .
this also explains why the inter - electron - hole - pocket scattering is stronger than the intra- and the inter - hole - pocket scattering .
furthermore , the intra - electron - pocket scattering comes from exchange of fluctuations at vectors @xmath120(0,0 ) and ( @xmath26,@xmath26 ) , and the latter is driven by the substantial checkerboard antiferromagnetic fluctuations .
the strength of dos , however , can not explain why the inter - electron - pocket ( @xmath27-@xmath31 ) scattering is much small because the stripe antiferromagnetic fluctuations at @xmath121(@xmath26,0 ) are not weak at all .
a detailed analysis shows that the reduction of the inter - electron - pocket scattering is due to the difference in the orbital - band matrix element . in fig .
[ orbital_weight ] , we show the orbital weight @xmath122 for every fs sheet . here
lines represent fe-_a / c _ and legends with circles and triangles represent fe-_b / d_. note that different curves exhibited in fig.3 are related by rotational symmetry@xcite .
for example , a 90@xmath123 rotation about the center of the unit cell located at the position of se changes fe - a to fe - b and @xmath21 to @xmath22 . as a result , as shown in figs .
[ orbital_weight](a ) or [ orbital_weight](b ) , weight curve of orbital-@xmath124 is equal to that of orbital-@xmath125 shifted by a phase @xmath29 .
similarly , since orbital-@xmath126 can be obtained by a 180@xmath123 rotation of orbital-@xmath125 , they have the same weight too .
this is the relation within the fermi surface of @xmath127 or @xmath128 . for relations between different fermi surfaces ,
since the same rotation changes fs-@xmath27 to fs-@xmath31 and relevant orbits also change under the rotation , the orbital components on different fs-@xmath129 would be highly different .
for instance , in figs .
[ orbital_weight](c ) and [ orbital_weight](d ) , the major component on fs-@xmath27 is orbital-@xmath130 , while it is @xmath131 on fs-@xmath31 and weights of other orbits are much smaller . because vertices in the fluctuation exchange formula in eq .
( [ flex ] ) only couple orbitals on the same fe , the inter - electron - pocket scattering will be heavily reduced .
we have examined the case of neglecting the orbital - band matrix elements in eq .
( [ gamma ] ) by setting @xmath132 and find that the inter - electron - pocket scattering becomes comparable to the intra - electron - pocket one .
therefore , the small inter - electron - pocket scattering is due to the small magnitude of the orbital - band matrix element . after a better understanding of the scattering matrix
, we now show the results of eigenvalues from the eigenvalue problem in eq .
( [ gap_eqn ] ) .
[ eigenvalues ] displays the largest eigenvalues of _ s_-wave ( @xmath133 ) and _ d_-wave ( @xmath134 ) as functions of _ u _ in three hund s rule couplings , @xmath4=0 , 0.2__u _ _ and 0.4__u__. @xmath133 is represented by solid blue lines while @xmath134 is represented by dashed red ones . as the interaction
_ u _ is increased , quantum fluctuations enhances the sc pairing and then @xmath135 s increase .
moreover , we observe that at different @xmath4 , the dimensionless coupling constants for different symmetry grow in different speeds . as shown in the insets where the differences of two eigenvalues @xmath136 are plotted , the _ d_-wave is always more stable against the _ s_-wave ( @xmath137 ) at @xmath4=0 , while the preference is reverse ( @xmath138 ) at @xmath4=0.4_u_. between them at @xmath4=0.2_u _ , the _ d_-wave is favored for _ _ [email protected] but becomes unfavorable above that .
the above results indicate that the hund s rule coupling is a parameter of controlling the sc symmetry as well as the afm order @xcite .
this is further supported by fig .
[ lambda_vs_jh ] , in which @xmath140 versus @xmath4 at different _ u _ is plotted . except for the small phase space with _
_ u__=1.0ev and @xmath141 , @xmath140 increases monotonously with @xmath4 , indicating that the hund s rule coupling favors the _ s_-wave .
however , @xmath140 is not a monotonic function of _
u_. there exists a critical @xmath4 , @xmath142 , separating two states ; below it , @xmath140 decreases as _ u _ increases , but above it , @xmath140 increases with _
u_. unless further symmetry is broken , _ s_-wave and _ d_-wave do not mix and the phase transition between them is first order . with the phase diagram of superconductivity , one still needs to know the gap function explicitly to understand its physics , especially in this new type of system with space group different from other iron - based superconductors .
we shall first examine the case of _ s_-wave solution in fig .
s - wave and then the _
d_-wave one in fig .
[ d - wave ] . following ref .
[ ] , we shall speak of _ s_-wave by @xmath143 with @xmath108 being the 90@xmath42-rotational operation on @xmath94 .
[ s - wave ] shows the _
s_-wave gap function @xmath144 along fs sheets @xmath24 , @xmath25 , and @xmath27 for _ _ u__=1.0ev and @xmath145=0.2_u_. @xmath144 on fs-@xmath31 not shown is the same as on fs-@xmath27 when @xmath146 .
blue squares are the eigenvalue - problem solution , while the solid red lines are the fit guided by eye .
one of the features is that such defined _ s_-wave gap have multi - nodes on fs-@xmath24 , @xmath27 and @xmath147 .
moreover , because there is no reflection symmetry , there is an extra degree of freedom in defining origins of angles . as a result ,
the fitting curves for _ s_-wave symmetry , chosen from lowest harmonic functions , are found to be @xmath148 + 0.0153 , \notag \\ g_{s}(\theta _ { \alpha 2 } ) & = & 0.028\cos \left [ 4\left ( \theta _ { \alpha 2}+5{{}^{\circ } } \right ) \right ] \notag \\ & + & 0.008\cos \left [ 8\left ( \theta _ { \alpha 2}-22{{}^{\circ } } \right ) \right ] -0.136 , \notag \\ g_{s}(\theta
_ { \beta 1 } ) & = & 0.13\left [ \cos ( k_{x}^{\prime } ) + \cos ( k_{y}^{\prime } ) \right ] \notag \\ & + & 0.28\cos ( k_{x}^{\prime \prime } ) \cos ( k_{y}^{\prime \prime } ) -0.03\text { , } \label{swave}\end{aligned}\]]where @xmath149 , @xmath150 and @xmath151 , @xmath152 .
here @xmath153 , @xmath154 , @xmath155 , and @xmath156 are the common _ s_-wave bases .
( because fs-@xmath27 is not located on symmetric point , a different type of bases is adopted . ) however , due to broken reflection symmetry , origins of angles get shifted by @xmath157 , @xmath158 , @xmath159 and @xmath160 .
these angles are the allowed degrees of freedom in the @xmath3 symmetry and their precise values are to minimize the total energy of the system .
similar to the @xmath161-wave proposed in iron - pnictide superconductor mazin2008 , where the sc order parameter on hole pockets and on electron pockets have opposite signs to make superconductivity stable , here we have the similar mechanism . in fig .
[ s - wave ] , we see that the mean value of the gap on fs-@xmath25 has an opposite sign to those on fs-@xmath24 and fs-@xmath129 s , which suggests these inter - pocket scatterings gain energy . due to their larger scattering strength ,
the @xmath162 s scattering processes gain most energy .
the @xmath163 s scattering processes are not favorable due to large cancellation by the oscillating behavior between positive and negative of gap on fs-@xmath24 . for the intra - pocket scattering processes , because the fermi surfaces are small and if the fluctuations are smooth over the wave vector * q * , the interaction would not able to change from repulsive to attractive .
however , the repulsive strength can be reduced by making the gap oscillatory and even with the sign being changed .
the reason why gap functions take the particular forms of eq .
( [ swave ] ) can be traced back to the behavior of the inverse fermi velocity ( ifv ) , which is shown as the dashed lines in fig .
[ s - wave ] .
clearly , the oscillatory behavior of gap functions results from oscillatory ifv . on fs-@xmath27 ,
the ifv is peaked at @xmath164225@xmath165 where the gap is a peak too .
the fact that the other point at @xmath16648@xmath42 with high ifv does not lead to a large gap is because the repulsion can be lowered in the intra- and inter - electron - pocket scattering processes .
due to that gap functions on fs-@xmath167 change slowly , it is clear that when the gap on fs-@xmath25 is in - phase with ifv and dose not change sign , one gains energy in @xmath162 s scatterings but loses energy in the intra - pocket scatterings .
the compromise between the intra- and inter - pocket scattering processes results in the gap function on fs-@xmath24 with the gap function being about @xmath168 out of phase relative to ifv . in the following , we list numerical values of the coupling matrix @xmath169 , which are consistent with the above explanations : @xmath170 \allowbreak , \]]where due to that @xmath171 is a symmetric matrix , the lower - triangular matrix elements are omitted for brevity , and @xmath172 .
similar to the _
s_-wave , the _ d_-wave that we shall speak of obeys @xmath174 with @xmath108 being the 90@xmath42-rotational operation on @xmath94 . in fig .
[ d - wave ] , we show the _ d_-wave gap function @xmath175 for _ _ u__=0.9ev and @xmath4=0 .
@xmath175 on fs-@xmath27 and on fs-@xmath31 * is * equal in magnitude but has opposite signs , _
@xmath176 when @xmath146 .
it is seen that the gap function of _ d_-wave always exhibits nodes on any fermi surface . the fitted functions in fig .
[ d - wave ] ( solid red ) are given by @xmath177 , \notag \\
{ \alpha 2 } ) & = & 0.18\cos \left [ 2\left ( \theta _ { \alpha 2}-39{{}^{\circ } } \right ) \right ] , \notag \\ g_{d}(\theta _ { \beta 1 } ) & = & 0.04\left [ \cos ( k_{x}^{\prime } ) -\cos ( k_{y}^{\prime } ) \right ] \notag \\ & -&0.3\sin ( 2k_{x}^{\prime \prime } ) \sin ( 2k_{y}^{\prime \prime } ) , \end{aligned}\]]where @xmath178 , @xmath179 , @xmath180 , and @xmath181 .
note that instead of @xmath153 ( which is correct for the _ s_-wave ) , _ d_-wave is represented by @xmath182 .
similar to the _ s_-wave case , although nodes appear on every fs sheet , they are shifted away from the diagonal directions .
the main cause for the occurrence of _ d_-wave differs from that for the _ s_-wave and is primarily due to reduction of repulsion in intra - pocket and inter - electron - pocket scatterings through its sign - changing character . as a result ,
both antipodes on fs-@xmath27 ( @xmath16448@xmath42 and 225@xmath42 ) have large gaps in fig .
[ d - wave ] , in comparison to a single hump of the _ s_-wave in fig .
[ s - wave ] .
the coupling constant matrix @xmath183 summarizes the above effects @xmath184 \allowbreak , \]]where @xmath185 .
note that not only the repulsion between intra - pocket electrons is reduced , due to the presence of nodes , the attraction between inter - pocket electrons is also reduced . although the _
d_-wave state might not gain much energy from inter - pocket scattering processes relative to the _ s _ -wave state , it saves more energy from the reduction of intra - pocket repulsion .
proximity of antiferromagnetism and superconductivity in a@xmath0fe@xmath9se@xmath186 implies that the sc state could be derived from spin fluctuations . under this assumption and
assuming that superconductivity exists in the @xmath187 iron - vacancy ordering state , we study the sc states from the effective pairing interaction in the flex approximation based on our previous 12-orbital tight - binding model @xcite . similar to that of iron - pnictide superconductors , for the spin - singlet superconductivity , _ s_-wave and _
d_-wave states are found to be close in energy . in particular , a quantum critical point is found at @xmath188 , below which the _ d_-wave prevails , while above which the _ s _ -wave wins over .
furthermore , unlike the iron - pnictide superconductors , the iron - vacancy order lowers the symmetry to the group @xmath3 so that the gap functions do not need to be reflection symmetric .
therefore , symmetric axes of the pairing momenta do not need to coincide with axes of the unit cell . as a result , the relative orientations of the pairing wave functions to the lattice , i.e.
, origins of angles for the pairing momenta , become new degrees of freedom for characterizing the superconducting gaps .
this implies that the complete order parameters for characterizing the superconductivity in a@xmath0fe@xmath9se@xmath2 consist of both the magnitude of the gap and the orientation of the pairing wave function relative to the underlying lattice . finally , while our work focuses on the superconductivity , the finding of quantum critical point controlled by @xmath4 in superconductivity is not an accident . in the magnetism
, we have shown that @xmath4 controls the magnetic phase transition from the stripe to the checkerboard antiferromagnetism @xcite .
in addition , haule and kotliar @xcite show that the crossover between coherence and incoherence is also determined by @xmath4 .
therefore , the hund s rule coupling plays the dominant role in determining localization , magnetism , and superconductivity of iron - based superconductors .
experimentally , it is observed that applying pressure drives the magnetic phase transition@xcite that accompanies the metal - to - semiconductor transition .
although in general , applying pressure affects itinerary electrons and could change @xmath189 and @xmath190 , the experimental results , when combined with our results on magnetism@xcite , indicate that increase of the pressure generally reduces @xmath191 in this system .
therefore , the phase tendency for iron - vacancy - ordered a@xmath0fe@xmath1se@xmath2 can be summarized in fig.[phase_diagram ] .
the tendency implies that by appropriately changing the applied pressure , transition between different pairing symmetry of superconductivity can be induced . while the phase tendency shown in fig .
[ phase_diagram ] needs further supports from experimental confirmation , our results provide a useful guide in clarifying the origin of unconventional superconducting gaps and searching for a possible quantum phase transition in iron - vacancy - ordered a@xmath0fe@xmath1se@xmath2 .
a. ricci , n. poccia , g. campi , b. joseph , g. arrighetti , l. barba , m. reynolds , m. burghammer , h. takeya , y. mizuguchi , y. takano , m. colapietro , n. l. saini , and a. bianconi , phys .
b * 84 * , 060511(r ) ( 2011 ) .
the relevant spin fluctuation in this system is the checkerboard afm fluctuation at * * q**=(@xmath26,@xmath26 ) , which is not revealed in the knight shift _ k _ , which is related to the spin susceptibility at * * q**=0 . for 1/__t__@xmath12_t _ on se or on the mono - valence element
a which locates at the center of the iron square , the form factor @xcite and iron deficiencies would suppress the effect of the checkerboard afm fluctuation . y. zhang , l. x. yang , m. xu , z. r. ye , f. chen , c. he , h. c. xu , j. jiang , b. p. xie , j. j. ying , x. f. wang , x. h. chen , j. p.
hu , m. matsunami , s. kimura , and d. l. feng , nat .
mater . * 10 * , 273 ( 2011 ) .
d. mou , s. liu , x. jia , j. he , y. peng , l. zhao , l. yu , g. liu , s. he , x. dong , j. zhang , h. wang , c. dong , m. fang , x. wang , q. peng , z. wang , s. zhang , f. yang , z. xu , c. chen , and x. j. zhou , phys .
lett . * 106 * , 107001 ( 2011 ) .
l. zhao , d. mou , s. liu , x. jia , j. he , y. peng , l. yu , x. liu , g. liu , s. he , x. dong , j. zhang , j. b. he , d. m. wang , g. f. chen , j. g. guo , x. l. chen , x. wang , q. peng , z. wang , s. zhang , f. yang , z. xu , c. chen , and x. j. zhou , phys .
b * 83 * , 140508(r ) ( 2011 ) . similar to the electron
phonon case where the electron - phonon coupling constant is related to a frequency integral of the phonon density of states divided by frequency , the imaginary part of @xmath192plays the role of the phonon density of states in the fluctuation - exchange coupling .
therefore , the integral of @xmath193 will produce @xmath194 by using the kramers - krnig relation @xcite . here
the real part in eq .
( [ gamma ] ) has taken account of the time reversal symmetry , @xmath195 . | based on an effective 12-orbital tight - binding model , we examine the superconducting states induced by the antiferromagnetic fluctuations for iron - vacancy - ordered a@xmath0fe@xmath1se@xmath2 .
it is shown that due to the broken reflection symmetry induced by the iron vacancies , new superconducting states with @xmath3 symmetry emerge .
in particular , we show that in the @xmath3 symmetry , symmetric axes of the pairing momenta do not need to coincide with axes of the unit cell . as a result ,
in addition to the magnitude of the pairing gap , the relative orientation of the pairing wave function to the lattice forms another degree of freedom for characterizing the superconducting gap and can further help in gaining the condensation energy .
nonetheless , similar to other iron - based superconductors , the singlet ground state is still dominated by _
s _
-wave or _
d_-wave , which are nearly degenerate with anisotropic gaps .
furthermore , _
s_-wave and _ d_-wave superconducting states are separated by a quantum critical point controlled by the hund s rule coupling @xmath4 . |
as is well known , the rare radiative decays of @xmath6 mesons is in particular sensitive to contributions from those new physics beyond the standard model(sm ) .
both inclusive and exclusive processes , such as the decays @xmath7 , @xmath0 and @xmath8 have been received some attention in the literature@xmath9}$ ] . in this paper
, we will present our results in technicolor theories . the one generation technicolor model ( ogtm)@xmath10}$]is the simplest and most frequently studied model which contained the parameters are less than sm .
same as other models , the ogtm has its defects such as the s parameter large and positive@xmath11}$ ] .
but we can relax the constraints on the ogtm form the @xmath12 parameter by introducing three additional parameters @xmath13@xmath14}$ ] .
the basic idea of the ogtm is : we introduce a new set of asymptotically free gauge interactions and the technicolor force act on technifermions .
the technicolor interaction at @xmath15 become strong and cause a spontaneous breaking of the global flavor symmetry @xmath16 .
the result is @xmath17 massless goldstone bosons .
three of the these objects replace the higgs field and induce a mass of @xmath18 and @xmath19 gauge bosons . and
at the new strong interaction other goldstone bosons acquire masses . as for the @xmath0 , only the charged color single and color octets have contributions .
the gauge couplings of the pgbs are determined by their quantum numbers . in table 1
we listed the relevant couplings@xmath20}$ ] needed in our calculation , where the @xmath21 is the corresponding element of @xmath22 matrix .
the goldstone boson decay constant @xmath23@xmath24}$ ] should be @xmath25 , which corresponds to the vacuum expectation of an elementary higgs field . at the lo in qcd
the effective hamiltonian is @xmath26 where , as usual , @xmath27 denotes the fermi coupling constant and @xmath28 indicates the cabibbo - kobayashi - maskawa matrix element.and the current - current , qcd penguin , electromagnetic and chromomagnetic dipole operators are of the form @xmath29 where @xmath30 and @xmath31 are color indices , @xmath32 labels su(3)c generators , e and @xmath33 refer to the electromagnetic and strong coupling constants , while @xmath34 and @xmath35 denote the qed and qcd field strength tensors , respectively . in fig.2 the shot - dash lines represent the charged pgbs @xmath2 and @xmath3 of ogtm .
we at first integrate out the top quark and the weak @xmath36 bosons at @xmath37 scale , generating an effective five - quark theory and run the effective field theory down to b - quark scale to give the leading log qcd corrections by using the renormalization group equation .
the wilson coefficients are process independent and the coefficients @xmath38 of 8 operators are calculated from the fig.2.the wilson coefficients are read@xmath39}$ ] @xmath40 with @xmath41 , @xmath42 and @xmath43.from the @xmath44 , we can see the situation of the color - octet charged pgbs is more complicate than that of the color - singlet charged pgbs , because of the involvement of the color interactions . where @xmath45 \\
b(y ) & = & \frac { -\frac{11}{36 } + \frac{53}{72}y -\frac{25}{72}y^2}{(1-y)^3}\nonumber \\ & + & \frac { -\frac{1}{4}y + \frac{2}{3}y^2 -\frac{1}{3}y^3 } { ( 1-y)^4}\log[y]\\ c(\delta)&=&\frac{\frac{1}{8 } -\frac{5}{8 } \delta-\frac{1}{4 } \delta^2}{(1-\delta)^3 } -\frac { \frac{3}{4}\delta^2}{(1-\delta)^4 } \log[\delta ] \\
d(y ) & = & \frac { -\frac{5}{24 } + \frac{19}{24}y -\frac{5}{6}y^2}{(1-y)^3}\nonumber \\ & + & \frac { \frac{1}{4}y^2 -\frac{1}{2}y^3}{(1-y)^4 } \log[y]\\ e(y ) & = & \frac { \frac{3}{2}-\frac{15}{8}y -\frac{15}{8}y^2 } { ( 1-y)^3 } + \frac{\frac{9}{4}y -\frac{9}{2}y^2}{(1-y)^4 } \log[y]\end{aligned}\ ] ] by caculate the graphs of the exchanged @xmath36 boson in the sm we gained the function @xmath46 and @xmath47;and by caculate the graphs of the exchanged color - singlet and color - octet charged pgbs in ogtm we gained the function @xmath6 , @xmath48 and @xmath49 . when @xmath50 , @xmath51 , the ogtm contribution @xmath6 , @xmath48 and @xmath49 have always a relative minus sign with the sm contribution @xmath46 and @xmath47 . as a result , the ogtm contribution always destructively interferes with the sm contribution . to calculate @xmath0 ,
one may follow a perturbative qcd approach which includes a proof of factorization , showing that soft gluon effects can be factorized into @xmath55 meson wave function ; and a systematic way of resumming large logarithms due to hard gluons with energies between 1gev and @xmath56 . in order to calculate the matrix element of eq(1 ) for the @xmath0
, we can work in the weak binding approximation and assume that both the @xmath57 and the @xmath58 quarks are at rest in the @xmath59 meson , and the @xmath57 quarks carries most of the meson energy , and its four velocity can be treated as equal to that of @xmath59 .
hence one may write @xmath57 quark momentum as @xmath60 where is the common four velocity of @xmath57 and @xmath55 .
we have @xmath61 the total amplitude is now separated into a cp - even and a cp - odd part @xmath66 we find that @xmath67 with @xmath68 , @xmath69 , and @xmath70 where @xmath71\nonumber \\ a_c & = & ( c_1+c_3-c_5)n_c+(c_2+c_4-c_6 ) \nonumber \\ a_s & = & { 1\over 4}\left[(c_3+c_4-c_5)n_c+(c_3+c_4-c_6)\right]\\ a_s & = & { 1\over 4}\left[(c_3+c_4-c_5)n_c+(c_3+c_4-c_6)\right].\end{aligned}\ ] ] the functions @xmath72 , @xmath73 and @xmath74 are defined by @xmath75 with @xmath76 in sm , with @xmath78 , and the other wilson coefficients are zero , we find @xmath79 which amounts to a branching ratio @xmath80 , for the given @xmath81 . in numerical calculations we use the corresponding input parameters @xmath82 , @xmath83 , @xmath84 , @xmath85 and @xmath86 , respectively .
the present experimental limit@xmath87}$ ] on the decay @xmath88 is @xmath89 which is far from the theoretical results .
so , we can not put the constraint to the masses of pgbs .
the constraints of the masses of @xmath2 and @xmath3 can be from the decay@xmath90}$ ] @xmath91 : @xmath92gev .
fig.3(4 ) denotes the @xmath95 about the mass of @xmath93 ( @xmath2 ) under different values of @xmath94 ( @xmath93 ) . from fig.3 and 4
, we find the the curves are much different from the the sm one .
it can be enhanced about 1 - 2 levels to the sm prediction in the reasonable region of the masses of pgbs .
this gives the strong new physics signals from the technicolor model .
the branching ratio of @xmath5 decrease along with the mass of @xmath93 and @xmath2 reduce .
this is from the decoupling theorem that for heavy enough nonstandard boson .
when @xmath96 and @xmath97 have large values , the contributions from ogtm is small.from the @xmath98 , we can see the functions @xmath6 , @xmath48 and @xmath49 go to zero , as @xmath99 , @xmath100.the branching ratio in the fig.(3 ) is changed much faster than that in the fig.(4).this is because the contribution to @xmath88 from the color octet @xmath93 is large when compared with the contribution from color singlet @xmath2 . as a conclusion ,
the size of contribution to the rare decay of @xmath88 from the pgbs strongly depends on the values of the masses of the charged pgbs .
this is quite different from the sm case . by the comparison of the theoretical prediction with the current data one can derived out the the contributions of the pgbs : @xmath2 and @xmath101 to @xmath88 and give the new physics signals of new physics . a n mitra .
lett . , 2000 , * b473 * : 297 - 304 junjie cao , zhenjun xiao , gongru lu .
rev . , 2001 ,
* d64 * : 014012 z.j.xiao , c.d.l , w.j.huo.phys .
, 2003 , * d67 * : 094021 l. reina , g. ricciardi , a. soni .
, 1997 , * d56 * : 5805 g. hiller , e.o .
lett . , 1997 , *
b409 * : 425 c .- h .
chang , g .-
lin , y .- p .
lett . , 1997 ,
* b415 * : 395 - 401 m.r .
ahmady , e. kou .
hep - ph/9708347 g. hiller , e.o .
iltan , mod.phys.lett .
, 1997.*a12 * : 2837 - 2846 s. choudlhury , j. ellis . hep - ph/9804300 t.m .
aliev , g. turan.phys .
, 1993 , * d48 * : 1176 t.m .
aliev , g. hiller , e.o .
nucl.phys.,1998 , * b515 * : 321 - 341 t.m .
aliev , e.o .
, 1998 , * d58 * : 095014 s. bertolini , j. matias .
, 1998 , * d57 * : 4197 - 4204 p. singer , d .- x .
, 1997 , * d56 * : 4274 s. dimopoulos .
nucl.phys.,1980 , * b168 * : 69 e. farhi , l. susskind.phys . rev . , 1979 , * d20 * : 3404 m.e .
peskint . takeuchi .
lett . , 1990 , * 65 * : 964 i. maksymyk , c.p .
, 1994 , * d50 * : 529 z.j.xiao , l.d.wan , j.m.yang , etal . phys .
, 1994 , * d49 * : 5949 e.eichten,i .
hinchliffe , k .
lane , etal .
, 1986 , * d34 * : 1547 c.d .
, 1996 , * d53 * : 2529 j.wicht , i. adachi , h. aihara , etal .
lett . , 2008 , * 100*:121801 | in this paper we calculate the contributions to the branching ratio of @xmath0 from the charged pseudo - goldstone bosons appeared in one generation technicolor model .
we find that the theoretical values of the branching ratio , @xmath1 , including the contributions of pgbs , @xmath2 and @xmath3 , are much different from the @xmath4 prediction .
the new physics effects can be enhance 2 - 3 levels to @xmath4 result .
it is shown that the decay @xmath5 can give the test the new physics signals from the technicolor model . |
despite important progress in the study of the structure and evolution of the galactic disk , many questions remain unanswered , such as the location of inner spiral arms or determination of the corotation distance .
open star clusters ( ocls ) play the most important role in the investigation of galactic disk properties , because their ages , distances , and colour excesses can be easily evaluated from photometric ccd observations . however , the number of investigated ocls with the heliocentric distances above 2 kpc is still small .
we developed and carry out a program of investigation of the perseus spiral arm based on the space and age distributions of open clusters .
we already obtained bvri ccd photometry and estimated the main physical parameters of six ocls in this region : king 13 , king 18 , king 19 , king 20 , ngc 136 , and ngc 7245 ( @xcite ) .
this study is concerned with another sample of six open clusters located toward the perseus spiral arm : berkeley 96 ( @xmath17 , @xmath18 ) , berkeley 97 ( @xmath19 , @xmath20 ) , king 12 ( @xmath21 , @xmath22 ) , ngc 7261 ( @xmath23 , @xmath24 ) , ngc 7296 ( @xmath25 , @xmath26 ) , and ngc 7788 ( @xmath27 , @xmath28 ) .
the webda database on open star clusters contains the following information on photometry of the clusters under study . *
berkeley 96 .
* @xcite measured photoelectric @xmath29 magnitudes of 10 stars and derived the cluster distance @xmath30 kpc and the colour excesses @xmath31 and @xmath32 mag . * berkeley 97 . *
@xcite evaluated the distance to the cluster @xmath33 kpc , its age @xmath34 myr , and colour excess @xmath35 mag , using @xmath36 magnitudes from 2mass point source catalog @xcite exclusively . * king 12 .
* @xcite obtained photoelectric @xmath29 magnitudes for 30 stars and found the cluster distance to be @xmath37 kpc and its colour excess @xmath4 to vary from 0.52 to 0.69 mag ( differential reddening ) .
@xcite measured photoelectric @xmath29 magnitudes of four stars in the cluster field .
* ngc 7261 . * based on @xmath29 photographic magnitudes of 147 stars , @xcite determined the cluster age @xmath38 myr , distance @xmath39 pc , and colour excesses @xmath40 and @xmath41 mag . using @xmath42 photoelectric photometry of five k - giants
, @xcite found the cluster age @xmath34 myr , distance @xmath43 pc , metal abundance @xmath44 = -0.7 $ ] , and colour excess @xmath45 mag . @xcite
obtained photoelectric @xmath29 photometry for 28 stars and photographic @xmath29 photometry for 53 stars ; @xcite measured @xmath29 photoelectric magnitudes for two stars ; @xcite measured photographic bv magnitudes for seven stars .
* ngc 7296 .
* based on @xmath46 and @xmath47 ccd photometry for about 140 stars , @xcite determined the cluster age @xmath48 myr , distance @xmath49 pc , and colour excesses @xmath50 and @xmath51 mag .
* ngc 7788 . * @xcite measured photoelectric @xmath29 magnitudes of five stars ; @xcite measured photographic @xmath29 magnitudes of 113 stars ; and @xcite found photographic @xmath52 magnitudes of 67 stars .
the clusters be 96 , king 12 , ngc 7261 , and ngc 7788 were studied by means of photographic and photoelectric photometry down to v - band limiting magnitudes of 14.516.5 mag . according to the estimates reported by the above authors and by @xcite for ngc 7788 , the distances to all these clusters exceed 2 kpc .
hence the published colour - magnitude diagram ( cmd ) isochrone fits involved only the upper parts of the main sequences of the clusters studied whose physical parameters could therefore be determined with large errors .
berkeley 97 was studied using only @xmath53 data , also down to a shallow limiting magnitude .
the only published color - magnitude diagram of ngc 7296 is based on ccd photometry and goes down to a v - band limiting magnitude of 16.8 mag .
ccd data is known to be often fraught with systematic errors and we therefore decided to repeat the observations of this cluster .
we also wanted to verify a very small value @xmath54 derived by @xcite .
all these factors justify our choice of the six open clusters listed above .
we observed six open clusters in the johnson / kron cousins broadband @xmath0 using the zeiss1000 telescope at the special astrophysical observatory of the russian academy of sciences equipped with a photometer based on an eev 4240 2k @xmath55 2k ccd .
the pixel size , readout noise , and gain were equal to 13.5 @xmath56 , @xmath57 , and @xmath58 per adu , respectively .
the scale and field of view were equal to @xmath59/pixel and about @xmath60 , and @xmath61/pixel and about @xmath62 during the 2003 and 2009 observations , respectively .
the scale and field of view size differences were due to changed photometer layout .
table 1 shows the log of observations .
its columns contain dates of observation , object names , air masses @xmath63 , and exposure times in each band .
figure 1 shows the images of all six clusters under study taken in @xmath64band with the exposure 300 s with the limiting magnitude of about 21 mag .
.log of observations [ cols="^,^,^,^,^ " , ] primary data processing , image extraction , and analysis were performed in the standard way using the eso - midas environment and the daophot / allstar program package @xcite . prior to the data reduction , we transformed the pixel coordinates for every star into celestial coordinates for the epoch j2000 .
in addition to the program clusters studied , the open cluster ngc 7790 was observed several times for each night as a photometric standard , and the ngc 6940 cluster was used once as a photometric standard ( see table 1 ) .
the standard magnitudes for stars in the fields of ngc 7790 and ngc 6940 were taken from stetson s database .
the ranges of magnitudes and colours for about 250 standard stars in the cluster field are @xmath65 and @xmath66 mag , respectively .
we used the following formulae @xcite to transform the magnitudes and colours from the instrumental to the standard system : @xmath67 @xmath68 @xmath69 @xmath70 @xmath71 where @xmath72 , @xmath73 , @xmath74 , and @xmath75 are the magnitudes in the instrumental system and @xmath76 , @xmath64 , @xmath77 , and @xmath78 are the magnitudes in the standard system .
the formula for the @xmath64 magnitude as a function of @xmath79 colour was used only for the stars without @xmath76 magnitudes .
the averages of the median coefficients were @xmath80 , @xmath81 , @xmath82 , @xmath83 , and @xmath84 for the observations done in 2003 and @xmath85 , @xmath86 , @xmath87 , @xmath88 , and @xmath89 for the observations done in 2009 ( the @xmath90 deviations are given as the errors ) .
we used mean values to recalculate the zero points @xmath91 , @xmath92 , @xmath93 , @xmath94 , and @xmath95 for the times of observations of the standard clusters .
the differences in the @xmath96 coefficients are presumably due to the changes in the photometer layout mentioned above .
as shown in table 1 , the standard clusters ngc 7790 and ngc 6940 , and all program clusters were observed at the same air mass @xmath97 .
this explains why we did not estimate the extinction coefficients . in the cases where
ngc 7790 was observed twice , before and after the program cluster , we calculated the zero points of the transformation expressions by a linear interpolation between zero point values calculated from the standard cluster .
weighted mean @xmath64 magnitudes and @xmath98 , @xmath79 , and @xmath99 colours were computed and published only for those stars in the fields of the six program clusters , for which the magnitudes and colours derived from different frames and reduced to the standard system differ by less than 0.1 mag . this value ( 0.1 mag )
seems to be the typical magnitude error for the faintest measured stars and , at the same time , close to 3@xmath100-variance of magnitudes measured on different frames .
the final values of the errors in the magnitudes and colours were calculated as the @xmath90 errors of the weighted mean on all frames ( taking into account the errors in the aperture corrections and the errors introduced by the transformation from the instrumental system to the standard one ) . however ,
if the @xmath90 deviation of the magnitude or colour calculated on all frames exceeded the @xmath90 error of the weighted mean , this @xmath90 deviation was taken as the final error .
the results of the photometric study of the clusters berkeley 96 , berkeley 97 , king 12 , ngc 7261 , ngc 7296 , and ngc 7788 are presented in tables 4 , 5 , 6 , 7 , 8 , and 9 , respectively .
tables 49 are available in the electronic form only and are accessible at the cds website and at the sternberg astronomical institute website . in this paper , we list a sample of few lines of table 4 only as an example .
tables 59 have the same layout .
the tables include the equatorial coordinates @xmath101 and @xmath102 in degrees at the epoch j2000 , @xmath64 magnitudes and @xmath98 , @xmath79 , and @xmath99 colours and their errors @xmath103 , @xmath104 , @xmath105 , and @xmath106 for individual stars .
we used 99.999 in all cells where the corresponding values are not available .
we derived @xmath64 magnitudes and @xmath98 , @xmath79 , and @xmath99 colours for the stars in the fields of berkeley 96 , berkeley 97 , king 12 , ngc 7261 , ngc 7296 , and ngc 7788 . in fig .
[ errors ] , the error in the magnitude , @xmath107 , and the errors in the colours , @xmath104 , @xmath105 , and @xmath106 , are plotted against the @xmath64 magnitude for each cluster ; the numbers on each plot indicate the number of stars . in most cases ,
the errors do not exceed 0.10 mag ; the magnitude limit is 20 - 21.5 mag , depending on the band .
the colour plots for ngc 7261 show narrow `` tails '' , because only one frame in each band was taken with the exposure 300 s , and no averaging was performed ; the same is seen for @xmath106 ( berkeley 96 ) . we compared the derived magnitudes and colours with published photometry and
show the magnitude and colour differences in fig .
[ liter ] .
our photometric data are in good agreement with all other photoelectric data - see plots for be 96 and king 12 , `` star '' symbols on the plots for ngc 7261 and ngc 7788 .
in contrast , the comparison of our ccd photometry with photographic values reported elsewhere exhibits rather wide spread , as shown by the `` dot '' symbols on the plots for ngc 7261 and ngc 7788 .
we also found systematic differences of approximately @xmath108 mag in @xmath64 and @xmath109 mag in @xmath79 between our values and those derived by @xcite for ngc 7296 .
differences in the @xmath98 colours show the systematic trend from @xmath110 to @xmath111 mag .
however , 32 common stars identified in the apass catalogue show small negative differences of about @xmath112 mag in @xmath64 and @xmath113 mag in @xmath98 , respectively . to estimate the cluster distance , age , and colour excess toward the cluster , we built the following colour magnitude diagrams ( cmd ) for each program cluster : @xmath114 , @xmath115 , and @xmath116 from our data , and @xmath117 , @xmath118 from the 2mass catalogue .
the faint stars with colour errors exceeding @xmath119 mag were excluded from our consideration .
we then fit each cmd to a solar - metallicity isochrone by @xcite using the method developed by @xcite . in this method we superimposed the isochrone on the colour
magnitude diagram in such a way that the corresponding plot for the radial density distribution exhibits a peak for the stars lying in the vicinity of the isochrone ( the distance in the colour index for such stars is less than @xmath119 mag ) and an almost flat distribution for all other stars ( supposed field stars ) .
the centers of clusters were found from density maps built using data from the 2mass psc .
we considered the position of the maximum of the density peak as the center of the cluster , and the distance from the cluster center , at which the star density in the cluster region becomes flat on the radial density distribution and equal to the density of field stars , as the cluster radius . in this way we evaluated the distance , colour excess , and age for each cluster on each cmd . to convert the colour excesses found from different colour magnitude diagrams to the colour excess @xmath4 and to calculate the distance modulus
, we used the following relations : @xmath120 ( @xcite ) and @xmath121 , @xmath122 , @xmath123 , @xmath124 , and @xmath125 ( @xcite ) .
figures [ izo1]-[izo6 ] present the @xmath126 , @xmath127 , @xmath128 ( or @xmath129 , see explanations below ) , and @xmath130 colour magnitude diagrams , on which the corresponding isochrones are plotted ( solid lines ) . for each cluster , table 2 gives the refined coordinates of the center , @xmath131 and @xmath132 , the cluster diameter @xmath133 in arcminutes , the age @xmath134 in myr , the colour excesses @xmath4 , and the true distance moduli @xmath135 determined from each cmd and recalculated using the relations above . on the whole ,
the parameters derived from different colour magnitude diagrams agree well , although for some clusters we can see small differences in @xmath4 values estimated separately from optical and infrared data .
we suppose that these differences can be due to greater contamination of the infrared cmds by field stars as compared to optical data , and to the uncertainties of color excess transformations used .
the final values of mean colour excesses and the true distance moduli , along with their @xmath90 deviations , are listed in table 3 .
all cmds yield approximately the same age estimates for each cluster , so we did not put any errors for the ages , which generally depend on the isochrones used .
we could not derive the cluster parameters from the @xmath118 diagram for ngc 7788 , because of the large dispersion of stars in this cmd .
let us compare our parameters with those from the literature .
table 3 contains the colour excesses @xmath4 , true distance moduli @xmath135 , distances @xmath74 in parsecs , and logarithms of the age @xmath136 from this paper and different published sources . for the clusters king 12 , ngc 7261 , and ngc 7788 , we include data from both @xcite and @xcite , because @xcite determined the distance moduli , and @xcite included these distances in his catalogue .
table 3 shows that only in the case of the king 12 cluster our parameters are in good agreement with those obtained in the previous study by @xcite . for the remaining five clusters investigated in this study
, the physical parameters have been considerably improved . for three clusters berkeley 97 , king 12 , and ngc 7788
we noticed gaps on their main sequences on all cmds built from our bv@xmath77@xmath78 ccd photometry . to verify these gaps , we extracted @xmath74 and @xmath75 data from the iphas survey ( @xcite ) and transformed these values from the sdss system into cousins @xmath137 , @xmath138 magnitudes using empirical colour transformations by @xcite .
we then built colour magnitude diagrams @xmath139 using both our photometric data and those derived from iphas .
figures [ izo2 ] , [ izo3 ] , and [ izo6 ] include these two cmds .
for each cluster , the gap on the main sequence is situated at the same values of magnitudes and colours on both diagrams @xmath139 and @xmath140 .
we used solar - metallicity evolutionary tracks and isochrones by @xcite to evaluate the mass intervals ( in solar - mass units ) that correspond to these ms gaps : [ 1.31.5 ] for berkeley 97 , [ 1.41.6 ] for king 12 , and [ 1.51.7 ] for ngc 7788 .
present - day luminosity and mass functions for these three clusters were built by star counts for all stars fainter than those at the ms turn - off . to evaluate the completeness of our photometric data
, we used the method described by @xcite .
artificial stars were added to our initial images in the @xmath76 and @xmath78 filters .
the fraction of artificial stars ( most of them are faint ) does not exceed 10% of the total population .
images with artificial stars are then processed using the same algorithm .
the portion of restored stars in each interval of magnitudes gives us the completeness value in the same magnitude intervals .
we used the least of the two values obtained by @xmath64 and @xmath78 images according to @xcite .
this method makes it possible to find the real pdlf with errors less than 3% in each magnitude interval where the completeness is more than 0.5 .
the members of each cluster were selected by photometric criteria using @xmath116 diagrams .
we built the pdlf for stars inside the cluster radius and for the field stars , and corrected both functions for the incompleteness factor and for the area difference .
the resulting pdlf is the difference between the cluster pdlf and the field stars pdlf . to convert the luminosity function into the mass function , we used stellar evolutionary tracks and solar - metallicity isochrones by @xcite .
figures [ izo7 ] , [ izo8 ] , and [ izo9 ] show the mfs for clusters berkeley 97 , king 12 , and ngc 7788 .
the values of the power law exponent ( logarithmic mass function slope )
@xmath101 was calculated by the @xmath141 solution for linear regression without taking into account escape points in the case of king 12 and ngc 7788 .
the gaps on each plot are plotted according to main sequence gaps on the colour - magnitude diagrams .
the escape point on fig .
[ izo8 ] and fig .
[ izo9 ] corresponds to the deficiency or the absence of stars in the gaps .
there is no such point on fig .
[ izo7 ] although a main sequence gap can be clearly seen in the case of berkeley 97 .
it can be explained by the insufficient subtraction of field stars . such gaps and breaks in mf plots
may be due to a discontinuity in the process of star formation inside the cluster .
lcccccc & berkeley 96 & berkeley 97 & king 12 & ngc 7261 & ngc 7296 & ngc 7788 + @xmath131 & @xmath142 & @xmath143 & @xmath144 & @xmath145 & @xmath146 & @xmath147 + @xmath132 & @xmath148 & @xmath149 & @xmath150 & @xmath151 & @xmath152 & @xmath153 + @xmath133 & @xmath154 & @xmath155 & @xmath156 & @xmath157 & @xmath158 & @xmath155 + @xmath159 & @xmath160 & @xmath161 & @xmath162 & @xmath163 & @xmath164 & @xmath163 + + @xmath165 & @xmath166 & @xmath167 & @xmath168 & @xmath169 & @xmath170 & @xmath171 + @xmath172 & @xmath173 & @xmath174 & @xmath175 & @xmath176 & @xmath177 & @xmath178 + @xmath179 & @xmath180 & @xmath181 & @xmath182 & @xmath183 & @xmath184 & @xmath171 + @xmath185 & @xmath186 & @xmath187 & @xmath188 & @xmath189 & @xmath190 & @xmath188 + @xmath191 & @xmath192 & @xmath167 & @xmath193 & @xmath194 & @xmath111 & @xmath195 + + @xmath165 & @xmath196 & @xmath197 & @xmath198 & @xmath199 & @xmath200 & @xmath201 + @xmath172 & @xmath201 & @xmath202 & @xmath203 & @xmath204 & @xmath205 & @xmath206 + @xmath179 & @xmath207 & @xmath208 & @xmath209 & @xmath210 & @xmath211 & @xmath212 + @xmath185 & @xmath213 & @xmath214 & @xmath215 & @xmath199 & @xmath216 & @xmath217 + @xmath191 & @xmath218 & @xmath219 & @xmath220 & @xmath221 & @xmath222 & @xmath195 + llllll & @xmath4 , mag & @xmath135 , mag & @xmath74 , pc & @xmath223 & source + berkeley 96 & @xmath6 & @xmath224 & @xmath5 & @xmath225 & this paper + & @xmath226 & @xmath227 & @xmath228 & & @xcite + berkeley 97 & @xmath8 & @xmath229 & @xmath7 & @xmath230 & this paper + & @xmath181 & @xmath231 & @xmath232 & @xmath233 & @xcite + king 12 & @xmath10 & @xmath234 & @xmath9 & @xmath235 & this paper + & @xmath236 & @xmath211 & @xmath237 & & @xcite + & @xmath238 & @xmath239 & & @xmath240 & @xcite + & @xmath238 & & @xmath241 & @xmath240 & @xcite + & @xmath238 & @xmath242 & @xmath243 & @xmath244 & @xcite + ngc 7261 & @xmath12 & @xmath245 & @xmath11 & @xmath246 & this paper + & @xmath247 & @xmath248 & @xmath249 & @xmath250 & @xcite + & @xmath169 & @xmath251 & @xmath252 & @xmath253 & @xcite + & @xmath254 & @xmath255 & & @xmath256 & @xcite + & @xmath254 & @xmath243 & @xmath257 & @xmath256 & @xcite + ngc 7296 & @xmath14 & @xmath258 & @xmath13 & @xmath259 & this paper + & @xmath260 & @xmath261 & @xmath262 & @xmath263 & @xcite + ngc 7788 & @xmath16 & @xmath264 & @xmath15 & @xmath246 & this paper + & @xmath265 & @xmath266 & & @xmath267 & @xcite + & @xmath265 & @xmath243 & @xmath268 & @xmath267 & @xcite + & @xmath247 & @xmath269 & @xmath243 & @xmath270 & @xcite + llllllllll @xmath131 , deg & @xmath132 , deg & @xmath64 & @xmath103 & @xmath271 & @xmath104 & @xmath272 & @xmath105 & @xmath273 & @xmath106 + 337.446900 & 55.392340 & 11.200 & 0.003 & 0.326 & 0.005 & 99.999 & 99.999 & 99.999 & 99.999 + 337.567690 & 55.446010 & 16.511 & 0.027 & 0.806 & 0.014 & 0.525 & 0.026 & 0.990 & 0.024 + 337.378080 & 55.451410 & 18.611 & 0.028 & 1.145 & 0.062 & 0.722 & 0.036 & 1.314 & 0.040 + 337.374210 & 55.451490 & 19.292 & 0.023 & 1.394 & 0.036 & 0.926 & 0.016 & 1.720 & 0.049
+ ... & ... & ... & ... & ... & ... & ... & ... & ... & ... +
we used our optical bv@xmath77@xmath78 ccd observations to obtain the @xmath64 magnitudes and @xmath98 , @xmath79 , and @xmath99 colours for a large number of stars in the fields of six open clusters : berkeley 96 , berkeley 97 , king 12 , ngc 7261 , ngc 7296 , and ngc 7788 ( see electronic tables 4 - 9 ) . for each cluster , we constructed optical and infrared ( based on 2mass data ) colour magnitude diagrams with the padova isochrones superimposed ( figs .
[ izo1]-[izo6 ] ) .
we used this technique to derive the ages , distances , and colour excesses of the clusters under study ( table 3 ) .
we found three clusters berkeley 97 , king 12 , and ngc 7788 to have gaps on their main sequences .
independent data extracted from the iphas ( @xcite ) catalogue confirmed these features on colour - magnitude diagrams for these three clusters .
we built the luminosity function and the mass function for these clusters and found the features on the mf plot .
we used the webda database operated at the institute for astronomy of the university of vienna and developed by e. paunzen and j .- c .
mermilliod ; 2mass data , which is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by nasa and nsf ; the apass database , located at the aavso website ; and the virtual observatory resource developed at sternberg astronomical institute ( moscow state university ) .
we thank l.n .
berdnikov , o.v .
vozyakova , and a.k .
dambis for valuable advice and the anonymous referee for useful comments .
becker w. , 1965 , mem .
soc . astron .
italiana , 36 , 277 dias w.s . , alessi b.s .
, moitinho a. and lepine j.r.d . , 2002 ,
a&a , 389 , 871 drew j.e .
, greimel r. , irwin m.j . , et al .
2005 , mnras , 362 , 753 dutra c.m . , santiago b.x . and bica e. , 2002 , a&a , 381 , 219 eggen o.j . , 1968 , royal obs .
137 fenkart r.p . ,
1968 , mem . soc . astron .
italiana , 39 , 85 frolov v.n .
, 1977 , izv .
pulkovo , 195 , 80 girardi l. , bertelli g. , bressan a. , et al . , 2002 , a&a , 391 , 195 glushkova e.v . ,
zabolotskikh m.v .
, koposov s.e .
et al . , 2010 , astronomy letters , 36 , 14 hardie r.h . , 1964 , photoelectric reductions in the book astronomical techniques , ed.w .
a. hiltner , university of chicago , p. 178 .
haug u. , 1970 , a&as , 1 , 35 he l. , whittet d.c.b .
, kilkenny d , and spencer jones j.h .
, 1995 , a&as , 101 , 335 hoag a.a . , johnson h.l .
, iriarte b. , et al .
, 1961 , publ . us .
obs . , 17 , 345 jennens p.a . ,
helfer h.l . , 1975 , mnras , 172 , 681 jordi k. , grebel e.k . and
ammon a. , 2006 , a&a , 460 , 339 kharchenko n.v . ,
piskunov a.e . ,
roser s. et al . , 2005 , a&a , 438 , 1163 koposov s.e .
, glushkova e.v . and zolotukhin i.yu . , 2008 , a&a , 486 , 771 kubinec w.r . , 1973 , publ .
warner swasey obs . , 1
, no . 3 loktin a.v . , gerasimenko t.p . and malysheva l.k . , 2001 , astron .
trans . , 20 , 607 mohan v. , pandey a.k . , 1984 ,
ap&ss , 105 , 315 netopil m. , paunzen e. , maitzen h.m .
, et al . , 2005 , astron .
nachr . , 326 , 734 del rio g. , 1984 , a&as , 56 , 289 sagar r. , richtler t. , 1991 , a&a , 250 , 324 skrutskie m.f .
, cutri r.m . , stiening r. , et al . , 2006 ,
aj , 131 , 1163 stetson p.b . , 1987 , pasp , 99 , 191 tadross a.l . , 2008 , mnras , 389 , 285 | we present @xmath0 ccd photometry in the fields of six galactic open clusters toward the perseus spiral arm .
these data , complemented with @xmath1 , @xmath2 , and @xmath3 magnitudes from 2mass , have been used to determine the ages , distances , and colour excesses @xmath4 for these clusters : 40 myr , @xmath5 pc , @xmath6 mag ( berkeley 96 ) ; 250 myr , @xmath7 pc , @xmath8 mag ( berkeley 97 ) ; 70 myr , @xmath9 pc , @xmath10 mag ( king 12 ) ; 160 myr , @xmath11 pc , @xmath12 mag ( ngc 7261 ) ; 280 myr , @xmath13 pc , @xmath14 mag ( ngc 7296 ) ; and 160 myr , @xmath15 pc , @xmath16 mag ( ngc 7788 ) .
we found gaps in the mass function of clusters be 97 , king 12 , and ngc 7788 in the mass intervals of [ 1.31.5 ] , [ 1.41.6 ] , and [ 1.51.7 ] solar masses , respectively .
[ firstpage ] galaxy : open clusters and associations : individual : berkeley 96 , berkeley 97 , king 12 , ngc 7261 , ngc 7296 , and ngc 7788 |
in causal dynamical triangulations ( cdt ) the quantum gravity path integral is regularized by approximating the spacetime geometry @xmath0 with a piecewise linear manifold @xmath1 @xcite .
the building blocks of four dimensional cdt are four - simplices , which form a simplicial manifold .
imposed global proper - time foliation allows a well - defined wick rotation .
spatial slices , enumerated by a discrete time coordinate @xmath2 , are built of equilateral tetrehedra with link length @xmath3 .
time - like links have length @xmath4 .
there are two kinds of simplices : @xmath5 and @xmath6 where the numbers denote the number of vertices in adjacent slices . the regularized partition function @xmath7 is written as a sum over causal triangulations @xmath1 , @xmath8 e^{i s^{eh}[g ] } \,\rightarrow \ , \sum_{{\mathcal{t } } } e^{- s[{\mathcal{t } } ] } , \ s[{\mathcal{t } } ] = - k_0\ n_0 + k_4\ n_4 + \delta\ ( n_{41 } - 6 n_0 ) , \ ] ] where @xmath9 , @xmath10 and @xmath11 are bare coupling constants , which are functions of @xmath12 and @xmath13 .
the discrete regge action @xmath14 $ ] is the einstein - hilbert action @xmath15 = \frac{1}{16 \pi g } \int { \mathrm{d}}t \int { \mathrm{d}}^3 x \sqrt{- g } ( r - 2 \lambda)$ ] evaluated on a simplicial manifold @xmath1 composed of @xmath16 vertices and @xmath17 simplices , among them @xmath18 of type @xmath5 , using monte carlo simulations , we can measure expectation values of observables within the cdt framework .
the simplest observable is the three - volume @xmath19 defined as the number of tetrahedra building slice @xmath2 . in the de sitter phase the average volume profile @xmath20 corresponds to a four - dimensional euclidean de sitter universe which emerges dynamically as a background geometry .
studies of the covariance matrix @xmath21 showed that the fluctuations of @xmath19 are well described by the discretized minisuperspace action@xcite , @xmath22 = \frac{1}{\gamma } \sum_t \left ( \frac{(n_{t+1 } - n_t)^2}{n_{t+1 } + n_t } + \mu n_t^{1/3 } - \lambda n_t \right ) .
\label{eq : sminid}\ ] ]
the cdt model is completely determined by a transfer matrix @xmath23 , @xmath24 } } = { \mathrm{tr}}{\mathcal{m}}^t , \quad \langle \tau_1 | { \mathcal{m}}| \tau_2 \rangle = \sum_{{\mathcal{t}}|_{\tau_1 , \tau_2 } } e^{-s[{\mathcal{t } } ] } .
\label{eq : z2}\ ] ] the matrix element @xmath25 denotes the transition amplitude between states corresponding to three - dimensional triangulations @xmath26 and @xmath27 in one time step .
it is given by the sum over all four - dimensional triangulations @xmath1 of a slab with boundaries @xmath26 and @xmath27 .
it depends both on the entropy factor and the action @xmath14 $ ] .
the probability of finding a configuration with spatial volumes @xmath28 , @xmath29 , \label{eq : ptncm}\ ] ] can be measured in monte carlo simulations . here
@xmath30 is a projection operator on the subspace spanned by subset @xmath31 of three - dimensional triangulations which are build of exactly @xmath32 three - simplices .
although it is misleading to think of the aggregated `` state '' @xmath33 as a normalized sum of vectors @xmath34 , the form of the effective action ( [ eq : sminid ] ) suggests that there exists an effective transfer matrix @xmath35 whose elements are labeled by the three - volumes . for simplicity ,
let us define the two - point function , @xmath36 by summing over all three - volumes except at the times @xmath2 and @xmath37 .
we will show that equation ( [ eq : ptwopoint ] ) provides a very good approximation to the measured data @xcite .
the effective transfer matrix elements can be measured in various ways , for example @xmath38 or @xmath39 .
the elements @xmath35 measured in different ways completely agree up to numerical noise , supporting equation ( [ eq : ptwopoint ] ) .
all measurements presented in this paper were performed using the second expression for coupling constants @xmath40 , @xmath41 and @xmath42 .
the minisuperspace effective action ( [ eq : sminid ] ) is directly related to the effective transfer matrix @xmath43 and suggests that @xmath44 } .
\label{eq : tmeff}\ ] ] empirical transfer matrix elements @xmath35 showed that for large @xmath19 , up to numerical noise , approximation ( [ eq : tmeff ] ) is very accurate and allows extraction of parameters @xmath45 , @xmath46 and @xmath47 . as a function of @xmath48 , and a linear fit @xmath49 ( red line ) .
right : the potential term .
measured @xmath50 and the fit of the potential term @xmath51 ( red line , stops at @xmath52 ) .
, title="fig:",scaledwidth=49.0% ] as a function of @xmath48 , and a linear fit @xmath49 ( red line ) .
right : the potential term .
measured @xmath50 and the fit of the potential term @xmath51 ( red line , stops at @xmath52 ) .
, title="fig:",scaledwidth=49.0% ] we can get a better estimation of the kinetic term by keeping the sum of the entries @xmath48 fixed .
the matrix elements show the gaussian dependence on @xmath32 , @xmath53 $ ] .
the denominator @xmath54 of the kinetic term can be measured .
linear behaviour @xmath49 is shown in fig .
[ fig : kinpot ] ( left ) .
the potential part of the effective lagrangian @xmath55 may be extracted from the diagonal elements of the transfer matrix , @xmath56 .
right part of fig .
[ fig : kinpot ] presents the measured effective lagrangian and the fit . in the bulk region ,
where @xmath19 is large enough , the theoretical expectation ( [ eq : tmeff ] ) fits very well .
the best fit gives @xmath57 , @xmath58 and @xmath59 .
the model of causal dynamical triangulations comes with a transfer matrix @xmath25 .
the measured distributions of three - volumes @xmath19 , e.g. @xmath60 , have an exact definition in terms of the full transfer matrix @xmath23 and the density matrix @xmath61 .
the data coming from monte carlo simulations seems to allow for a much simpler description in terms of an effective transfer matrix @xmath43 , labeled by abstract vectors @xmath33 referring only to the three - volume .
the effective transfer matrix @xmath43 allows to directly measure the effective action @xmath62 $ ] .
the new method makes measurements faster . over the whole range of @xmath19
the effective transfer matrix elements are given by ( [ eq : tmeff ] ) with high accuracy .
this result is fully consistent with the reduced minisuperspace action ( [ eq : sminid ] ) , although in cdt no degrees of freedom are frozen . | causal dynamical triangulations is a background independent approach to quantum gravity .
we show that there exists an effective transfer matrix labeled by the scale factor which properly describes the evolution of the quantum universe . in this framework
no degrees of freedom are frozen , but , the obtained effective action agrees with the minisuperspace model .
+ _ andrzej grlich _ + the niels bohr institute , copenhagen university + blegdamsvej 17 , dk-2100 copenhagen , denmark . +
email : [email protected] |
the area of circuit quantum electrodynamics ( circuit qed ) has experienced a series of expressive advances during the twelve years of its existence cir2,cir4,cir5,cir3,science . originally aimed at the coherent coupling of superconducting artificial two - level atoms ( _ qubits _ ) to the quantized electromagnetic ( em ) field confined in 1d resonator @xcite , this area currently embraces a plethora of experimental architectures with different kinds of multi - level artificial atoms and sophisticated assemblies of interconnected 3d and 1d resonators and waveguides transmon , majer , paik , apl , wilson .
diverse designs incorporating up to tens of josephson junctions give rise to artificial atoms with distinct properties regarding the dissipation mechanisms and energy levels structure @xcite .
still , all they have in common the capability of coherently coupling to the em field @xcite , and several atoms and resonators can be incorporated into single packages that store and process quantum information @xcite .
certain types of artificial atoms permit a real - time manipulation of the internal properties , such as the energy levels splitting or the atom field coupling strength @xcite .
combined with the ability of rapidly tuning _ in situ _ the resonator effective frequency by external magnetic flux @xcite this architecture enables one to access a novel regime of light matter interaction in which all the parameters of the system hamiltonian are regarded as time - varying functions controlled externally @xcite .
when these parameters are modulated by a single or a superposition of weak harmonic perturbations , under resonance conditions one can engineer new forms of light
matter interaction that were not reachable in the stationary case @xcite .
this attribute has already been employed for experimental implementation blais - exp , simmonds of quanta exchange between the qubit and the field in the dispersive regime ( known as red sideband transition or jaynes - cummings regime ) , as predicted theoretically in @xcite .
such coupling relies on the ` rotating ' or jaynes - cummings terms @xcite in the atom field interaction hamiltonian ( terms that conserve the total number of excitations ) .
more exotic forms of interaction can be engineered by resonantly amplifying the weaker effects of the ` counter - rotating terms ' ( crt ) in the hamiltonian @xcite , e.g. , generation of photon pairs from vacuum ( analogue of the dynamical casimir effect , dce rev1,rev2,rev3,rev4,camop ) , simultaneous creation of a single photon and a single atomic excitation ( blue sideband transition or anti - jaynes - cummings regime ) , generation from vacuum of highly entangled qubit field states with pairs of excitations , coherent annihilation of pairs of system excitations ( effect dubbed anti - dce in @xcite ) and other more sophisticated phenomena liberato9,etc1,etc2,etc3,relativistic , processing , entangles .
some isolated effects described above can be merged in a coherent manner via _ multi - tone modulations _ , giving rise to a general effective hamiltonian with adjustable coupling strengths @xcite .
besides , these effects could also be implemented using halved modulation frequencies by making use of the second - order resonances ( with respect to the modulation amplitudes ) , although for weak modulations the associated transition rates would diminish by at least one order of magnitude @xcite . in this paper
we describe how a new effect reliant only on the rotatingterms the _ two - photon exchange interaction _ could be implemented in nonstationary circuit qed by modulating any system parameter at frequencies of the order of twice the atom field average detuning , which can be much lower than the resonator natural frequency .
however , this phenomenon only takes place for an array of @xmath5 qubits ( mathematically equivalent to a multilevel artificial atom in the ladder configuration )
. closed analytical description of the system unitary dynamics is derived for @xmath6 ( section [ sec2 ] ) , but it is shown numerically that similar behavior persists for as low as @xmath7 qubits ( section [ sec3 ] ) .
the proposed scheme works only in the dispersive regime of light matter interaction , and because of effective _ kerr _ nonlinearities just one pair of system bare states can be coupled via single - tone modulations .
we also show analytically and numerically that the photon exchange rate can be enhanced by simultaneous modulation of several system parameters with properly adjusted phases .
the prospects of observing this phenomenon experimentally are discussed in section [ sec4 ] .
our system consists of a single mode of em field interacting with @xmath0 qubits , as described by the quantum dicke model @xcite or @xmath0-qubit quantum rabi model @xcite ( we set @xmath8)@xmath9 ~ , \label{h1}\]]where the index @xmath10 labels the identical noninteracting atoms .
we assume that the cavity frequency @xmath11 , the atomic transition frequency @xmath12 and the ( real ) atom cavity coupling strength@xmath13 are externally prescribed functions of time .
@xmath14 and @xmath15 are the annihilation and creation operators and @xmath16 is the photon number operator .
the qubit operators are @xmath17 , @xmath18 and @xmath19 , where @xmath20 and @xmath21 denote the ground and excited states of the @xmath10-th qubit , respectively . in the absence of crt @xmath22 the hamiltonian ( [ h1 ] )
is also known as _ tavis - cummings hamiltonian _ @xcite .
although in our approach we fully take into account the crt , they are not essential for the phenomenon described here .
we assume the simultaneous modulation of all the system parameters as@xmath23where @xmath24 is the bare value and @xmath25 is the respective modulation depth .
the sum runs over the modulation frequencies of the order @xmath26 and the parameters @xmath27 and @xmath28 are the relative weights and phase constants corresponding to the modulation of @xmath29 at frequency @xmath30 . to ease the notation we define the complex modulation depth@xmath31~,~x=\{\omega , \omega , g\}\]]that includes the weight and the phase of @xmath29-modulation at frequency @xmath32 . for @xmath3
we introduce the collective atomic operators via the holstein - primakoff transformation @xcite@xmath33@xmath34where the ladder operators @xmath35 and @xmath36 satisfy the bosonic commutation relation @xmath37=1 $ ] . to the first order in @xmath38 the hamiltonian ( [ h1 ] ) becomes@xmath39where we defined the collective coupling strength @xmath40 .
we write the solution of ( [ bone ] ) in the heisenberg picture as@xmath41 \nonumber\]]@xmath42 .
\label{an3}\]]here the time - dependent operators @xmath43 , @xmath44 are yet to be found and@xmath45where @xmath46 is the bare qubit field detuning .
next we propose the ansatz@xmath47@xmath48 in terms of the effective operators @xmath49 and @xmath50 , which to the first order in @xmath51 satisfy the bosonic commutation relations @xmath52=[\hat{b}(t),\hat{b}% ^{\dagger } ( t)]=1 $ ] .
we also defined small frequency shifts @xmath53 and @xmath54where @xmath55 .
the rapidly oscillating c - number functions read @xmath56 + c.c.&\end{aligned}\ ] ] @xmath57 + c.c.&\end{aligned}\ ] ] @xmath58 \right . & \\ & + \left ( \frac{\varepsilon _ { -}^{(j)\ast } } { \delta _ { -}}-\frac{\tilde{% \varepsilon}_{g}^{(j)\ast } } { \tilde{g}_{0}}\right ) \left [ \beta _ { -}\frac{% e^{-it(\beta + \eta ^{(j)})}-1}{\eta ^{(j)}+\beta } -\beta _ { + } \frac{% e^{it(\beta -\eta ^{(j)})}-1}{\eta ^{(j)}-\beta } \right ] & \\ & + \left .
\left ( \frac{\varepsilon _ { -}^{(j)}}{\delta _ { -}}-\frac{\tilde{% \varepsilon}_{g}^{(j)}}{\tilde{g}_{0}}\right ) \left [ \beta _ { -}\frac{% e^{-it(\beta -\eta ^{(j)})}-1}{\eta ^{(j)}-\beta } -\beta _ { + } \frac{% e^{it(\beta + \eta ^{(j)})}-1}{\eta ^{(j)}+\beta } \right ] \right\ } & , \end{aligned}\ ] ] where @xmath59 stands for the complex conjugate . substituting our ansatz into the heisenberg equations of motion@xmath60@xmath61and assuming @xmath62 one
can consistently eliminate the rapidly oscillating terms @xcite and obtain quite simple differential equations for the slowly - varying operators @xmath63 and @xmath64 .
after long algebraic manipulations we find that the dynamics can be described by the effective hamiltonian @xmath65 , whose structure is more evident in the resonant and dispersive regimes studied below .
in the resonant regime , @xmath66 , the original lowering operators read approximately@xmath67 \nonumber\]]@xmath68\]]and for @xmath69 the dynamics is governed by the effective hamiltonian@xmath70 + h.c.&,\end{aligned}\]]where @xmath71 stands for the hermitian conjugate .
this hamiltonian holds under the approximations@xmath72where @xmath73 ( @xmath74 is the maximum number of total system excitations . in our calculations we neglect frequency shifts smaller that @xmath75 and @xmath76 ( which appear due to the elimination of the rapidly oscillating terms ) ,
so ultimately the resonant modulation frequency @xmath77 must be found numerically in the vicinity of the analytical prediction .
we see that nothing peculiar occurs under the resonant modulation , @xmath78 , since many different terms contribute equally to the dynamics .
hence the lone two - photon exchange interaction can not emerge in the resonant regime . in the dispersive regime , @xmath79 , where @xmath80 , the physical operators read approximately @xcite@xmath81 & \\ & \hat{b}\simeq e^{-it(\omega _ { 0}-\tilde{\delta}_{-}-\tilde{\delta}_{+})}% \left [ \hat{b}e^{-it\nu } -i\hat{a}\frac{\tilde{g}_{0}\sin ( \delta _ { -}t/2)}{% \delta _ { -}/2}e^{-it(\delta _ { -}/2+\tilde{\delta}_{-}-\nu ) } \right ] & , \end{aligned}\]]so the effective operators @xmath82 and @xmath64 can be crudely interpreted as the slowly varying cavity and matter lowering operators .
for @xmath83 the dynamics is governed by the effective hamiltonian@xmath84 , \nonumber\end{aligned}\]]where @xmath85 is the detuning symbol and the transition rate associated to the two - photon exchange is@xmath86(we adopt the shorthand notation @xmath87 , @xmath88 ) .
the hamiltonian ( [ pec ] ) was obtained under the approximations@xmath89@xmath90@xmath91and we again neglected frequency shifts smaller that @xmath75 and @xmath76 .
the equation ( [ pec ] ) is the main result of this paper , as it shows that one can achieve the selective two - photon exchange by setting the modulation frequency to @xmath92 , where @xmath93 is a small correction . to determine the value of @xmath93 we write the wavefunction corresponding to the hamiltonian ( [ pec ] ) in the schrdinger picture@xmath94 , \nonumber\end{aligned}\]]where @xmath95 is the probability amplitude and @xmath96 and latexmath:[$% the operators @xmath64 and @xmath82 , respectively .
one obtains@xmath98 \right\ } e^{it\mathcal{d}x } & \\ & + \rho ^{(1)\ast } \sqrt{(m+2)(m+1)n(n-1)}c_{m+2,n-2 } & \\ & \times \exp \left\ { -it\delta _ { -}\left [ 8m-4n+10 + 3n\xi ^{2}(4n-6)\right ] \right\ } e^{-it\mathcal{d}x}&.\end{aligned}\]]in particular , for the atoms initially in the ground states ( @xmath99 ) we obtain@xmath100 \right\ } e^{-it\mathcal{d}x}\end{aligned}\]]@xmath101 \right\ } e^{it\mathcal{d}x } \\ & + & \rho ^{(1)\ast } \sqrt{12(n-2)(n-3)}c_{4,n-4 } \\ & & \times \exp \left\ { -it\delta _ { -}\left [ 34 - 4n+3n\xi ^{2}(4n-14)\right ] \right\ } e^{-it\mathcal{d}x}.\end{aligned}\ ] ] so if we choose@xmath102\]]only the states @xmath103 and @xmath104 become coupled with the transition rate @xmath105 , whereas for @xmath106 other transitions are suppressed due to the relation @xmath107 .
to verify our approximate expressions we solved numerically the schrdinger equation @xmath108 for the hamiltonian ( [ h1 ] ) . to simplify the notation we introduced the normalized _ dicke _ state with @xmath109 atomic excitations@xmath110 . in the dicke basis
the hamiltonian ( [ h1 ] ) reads@xmath111where @xmath112 and @xmath113 .
hence our system also describes the interaction between the field and a ladder - configuration atom with @xmath114 levels @xcite . and
@xmath115 , under the resonant modulation of @xmath116 for : a ) @xmath117 ; b ) @xmath118 .
c ) behavior of the occupation probabilities @xmath119 and @xmath120 for @xmath118 .
d ) behavior of @xmath121 and @xmath115 for @xmath122 , when our analytical results are not applicable .
according to formula ( ro ) the time of the first oscillation is @xmath123 , in reasonable agreement with the numerical data .
the modulation stimulates the transition @xmath124 , and the initial state is @xmath125.,scaledwidth=45.0% ] under the simultaneous modulation of @xmath116 and @xmath126 .
the transition rate is expected to double compared to the figure [ fig1]b , in good agreement with the numerical data .
a ) behavior of @xmath121 and @xmath115 .
b ) evolution of relevant cavity occupation probabilities .
c ) evolution of atomic occupation probabilities .
notice that @xmath127 and @xmath128 also exhibit small oscillations due to the spurious one - photon exchange.,scaledwidth=45.0% ] the system state is written as@xmath129where @xmath130 is the probability amplitude and @xmath131 stands for the cavity fock state ( in the following we assume @xmath132 , so @xmath131 is time - independent ) .
hence the probability of @xmath133 photons ( atomic excitations ) is @xmath134 ( @xmath135 ) , and the average number of photons [ atomic excitations ] reads @xmath136 [ @xmath137 .
the numerical solution is illustrated in the figures [ fig1 ] and fig2 for parameters @xmath138 , @xmath139 , @xmath140 , @xmath141 , @xmath142 and the initial state @xmath143 , where @xmath144 stands for the coherent state with @xmath145 . in figure [ fig1 ]
we consider only the modulation of @xmath116 ( setting @xmath146 ) and compare the dynamics for @xmath117 , @xmath147 and @xmath122 as function of the dimensionless time @xmath148 , where @xmath149 .
we adjusted numerically the value of the modulation frequency to @xmath150 for @xmath117 , @xmath151 for @xmath118 and @xmath152 for @xmath153 in order to maximize the two - photon coupling between the states @xmath154and @xmath155 .
the numerical results are in qualitative agreement with the analytical predictions , although the actual transition rate is roughly @xmath156 larger than the predicted one for @xmath157 , @xmath158 larger for @xmath118 and @xmath159 smaller for @xmath122 .
such discrepancy is not surprising , considering that the parameters used in the simulations lie outside the validity range of the analytical approach [ inequalities ( [ h1 ] ) ( [ h3 ] ) ] .
the apparent broad width of the curves in figures [ fig1 ] [ fig2 ] originates from the fast low - amplitude oscillations due to off - resonant one - photon exchange between the cavity and the atoms inherent to the dispersive regime . in figure [ fig2 ]
we consider the simultaneous modulation of @xmath116 and @xmath160 using the additional values @xmath161 , @xmath162 , @xmath163 and setting the modulation frequency @xmath164 to again optimize the coupling @xmath165 .
as predicted by equation ( [ ro ] ) the two - photon transition rate roughly doubles compared to the single - modulation case shown in figure fig1 , while other features remain unaltered .
it can be observed from figures [ fig2]b and [ fig2]c that the main transitions induced by the modulation occur between the states @xmath166and @xmath167 , although there are spurious transitions to other states owing to the off - resonant one - photon exchange [ illustrated by the small oscillations of probabilities @xmath127 and @xmath128 with the same slow rate ] . instead of the oscillatory behavior exemplified in figures [ fig1 ] and [ fig2 ] one can also induce a steady transition from the state @xmath168to @xmath155 by using the effective _ landau - zener _
transition @xcite in which the modulation frequency is varied linearly with time across the expected resonance .
this is demonstrated in figure [ fig3 ] for the case of two qubits and the modulation frequency @xmath169 , where @xmath170 and @xmath171 ( all other parameters are the same as in figure [ fig1 ] ) .
we plot the probabilities @xmath119 , @xmath172 , @xmath173 , @xmath120 and the mandel s factor of the cavity field state @xmath174/n_{a}$ ] .
we see that one can accomplish a steady transition between the desired states , though it takes longer than in figure [ fig1]d , and the cavity field state becomes super - poissonian for large times @xmath175 ) .
we showed analytically that effective two - photon exchange interaction between a single cavity mode and @xmath3 off - resonant qubits can be achieved by externally modulating the system parameters with frequencies @xmath176 , where @xmath177 is the bare atom
field detuning .
this effect relies only on the ` rotating ' terms in the tavis - cummings hamiltonian and does not occur in the resonant regime of light
matter interaction ( @xmath178 ) .
moreover , for single - tone modulations just one pair of system states is coupled because of kerr - like intrinsic nonlinearities , although several states could be coupled by using multi - tone modulations .
we verified numerically that the phenomenon persists event for @xmath179 ( when our approximate expressions are not valid ) , however the associated transition rate , equation ( [ ro ] ) , is quite small . for example , considering the modest circuit qed parameters @xcite @xmath180ghz , @xmath181 , @xmath182 and assuming @xmath183 , @xmath184 we get @xmath185khz .
this fact renders the described phenomenon very susceptible to the effects of dissipation @xcite and unwanted fluctuations in the modulation frequency , hindering its experimental verification . | we consider the interaction of a single - mode cavity field and @xmath0 identical qubits in the scenario when one or several system parameters undergo a weak external modulation .
it is shown that in the dispersive regime of light
matter coupling one can engineer the effective two - photon exchange interaction when the frequency of harmonic modulation is @xmath1 , where @xmath2 is the bare atom field detuning .
the effective hamiltonian and associated transition rate are derived analytically for @xmath3 , and a good agreement is found with numerical data even when @xmath4 . |
encoding information spin degrees of freedom has the potential to revolutionize information technology via the development of novel spintronic devices and possible future quantum information processors @xcite .
utilizing information encoded in spin degrees of freedom , however , requires efficient , controlled on - chip transfer of spin - based information .
in some types of spintronic devices , spin degrees of freedom are used in addition to motional degrees of freedom of electrons , and information encoded in the spin degrees of freedom can be transferred using conventional currents . in principle , however , information stored in spin states can propagate through a network of coupled spins without charge transport . as propagation of spin - based information
is governed by quantum - mechanics and the schrdinger equation , however , excitations in a spin network propagate , disperse and refocus in a wave - like manner , and controlling information transport is thus a quantum control problem . without any means to control the propagation of spin - based information in such networks information transport can be slow and inefficient .
control can be utilized to optimize transport in terms of maximizing transfer efficiency and speed @xcite . here , we consider how to control information propagation in a network of spins by optimizing spatially distributed potentials , which remain constant during the evolution , in contrast to dynamic control schemes , which require dynamic modulation or fast switching of the control potentials .
we restrict ourselves to spin-@xmath0 particles with two spin states labeled @xmath1 and @xmath2 as a simple prototype device for routing spin - based information .
the energies of the spin states of spin @xmath3 differ by small amounts @xmath4 .
nearby spins can interact , e.g. , by exchange coupling .
this leads to a model hamiltonian for a network of coupled spins of the form @xmath5,\ ] ] where @xmath6 is a parameter depending on the type of coupling , e.g. , @xmath7 for isotropic heisenberg coupling and @xmath8 for pure @xmath9 coupling .
@xmath10 , @xmath11 and @xmath12 are operators acting on the @xmath13 dimensional hilbert space of the @xmath14-spin network .
@xmath10 is a tensor product of @xmath15 identity operators with a single @xmath16 operator in the @xmath3th position , and similarly for @xmath11 and @xmath12 .
@xmath16 , @xmath17 and @xmath18 are the pauli spin operators @xmath19 for networks with uniform coupling all non - zero couplings have the same strength @xmath20 and we can set @xmath21 by choosing the frequencies in units of @xmath20 and time in units of @xmath22 .
it can be easily verified that a hamiltonian of the form ( [ eq : h ] ) commutes with the total excitation operator @xmath23 .
the hilbert space of the system therefore decomposes into excitation subspaces @xcite .
if we assume that only a single excitation ( or bit of information ) propagates through the network at any given time , then the hamiltonian can be reduced to the single excitation subspace hamiltonian @xmath24 where @xmath25 can be thought of as a matrix which is zero except for a @xmath26 in the @xmath27 position .
which @xmath28 are non - zero depends on the network topology . for a chain with nearest - neighbor coupling
we have @xmath29 unless @xmath30 , and similarly for a ring arrangement , except that for the latter we also have @xmath31 .
while a linear chain can be thought of as a type of quantum wire , a ring can be regarded as a basic routing element to distribute spin states , e.g. , to chains attached to various nodes of the ring ( see fig . [
fig : landscape ] ) .
dynamic control involves dynamically altering certain couplings @xmath28 or potentials @xmath4 .
this is a powerful tool and has been explored in previous work @xcite .
however , it typically requires the ability to rapidly modulate or switch fields .
furthermore , for networks with a high degree of symmetry such as rings with uniform coupling , controllability is generally limited by dynamic symmetries @xcite .
an alternative to this dynamic control is to shape the potential landscape to facilitate the flow of information from an initial state ( input node ) to a target state ( output node ) .
for example , for information encoded in nuclear spins or electron spins in quantum dots whose potential can be controlled by surface control electrodes , this could be achieved by varying the voltages to create a potential landscape as shown in fig .
[ fig : landscape ] .
fixing the network topology defined by the couplings @xmath28 and applying potentials @xmath4 that are constant in time , gives the constant hamiltonian @xmath32 and the information transfer is governed by the schrdinger equation @xmath33 where @xmath34 is a unitary propagation operator , initially equal to the identity operator @xmath35 .
the probability of transmission of information from the input node @xmath36 to the output node @xmath37 in time @xmath38 is then given by @xmath39 assuming the energies @xmath4 are controllable , we have a vector of control parameters @xmath40 and the objective is to find a @xmath41 such that @xmath42 at some time @xmath38 .
we can fix the time @xmath43 , require that @xmath44 where @xmath45 is an upper bound , or attempt to accomplish the transfer with maximum fidelity in minimum time .
these problems can in principle be solved in a straightforward manner using standard optimization tools although the optimization landscape is challenging , in particular when the goal is to find a control that achieves the highest possible fidelity in the shortest time possible as this effectively creates two objectives for the optimization .
we can reformulate the control problem by diagonalizing the hamiltonian , @xmath46 , where @xmath47 is a unitary matrix of the eigenvectors and @xmath48 a diagonal matrix of the eigenvalues of @xmath49 . then @xmath50 with @xmath51 and
our control objective is to ensure @xmath52 where @xmath53 is a global phase factor , which we introduce to cancel the the global phase of the output state . for a network with ring topology
, we can take the input state to be @xmath54 due to translation invariance , and the output state to be @xmath55 , @xmath56 .
. becomes @xmath57 thus , the optimization problem is equivalent to finding the a control vector @xmath41 and phase @xmath53 that minimize @xmath58 note that only the @xmath26st and @xmath59th component of the eigenvectors ( or @xmath26st and @xmath59th rows of @xmath47 ) and eigenvalues @xmath60 matter .
the upper bound on the fidelity of the transmission given no time constraints , referred to as information transfer fidelity @xcite , is given by @xmath61 where @xmath62 .
since it is a probability , @xmath63 .
here we use some eigenstructure assignment concept @xcite to show that , given @xmath64 and @xmath65 , there always exists an orthonormal reference frame @xmath66 such that the upper bound @xmath63 is achieved . the issue as to whether this reference frame can be achieved by an optimal biasing strategy
@xmath67 is open , but numerical exploration seems to indicate that the optimal biases in the sense of sec . [ s :
results ] reproduce this frame with high accuracy . instead of deriving the position of the reference frame @xmath68 relative to the input and output states , we reformulate the problem as the one of finding the position of the input and output states relative to a given ( orthonormal ) reference frame @xmath66 .
let @xmath69 , @xmath70 , @xmath71 , be the coordinates of the input and output states , resp .
, in the reference frame @xmath68 .
the problem is to maximize eq .
( [ e : itf ] ) subject to the constraints @xmath72 define the lagrange multipliers @xmath73 , @xmath74 , @xmath6 and the augmented functional @xmath75 the classical first order condition for optimality yields @xmath76 existence of a solution yields @xmath77 the _ crucial issue _ is to observe that @xmath73 should be independent of @xmath3 . at optimality ,
not all @xmath78 s could be of the same sign ( otherwise @xmath79 ) .
this implies that @xmath80 and @xmath81 . from there on , solving , and after some manipulation , it is found that @xmath82 as already said , the optimal biases @xmath4 s appear to reproduce this result .
the above eigenvector assignment is non - unique and dimension dependent .
when @xmath83 , the solution is already far from unique .
up to permutation , either @xmath84 is in @xmath85 and offset at a @xmath86 angle or @xmath87 and @xmath88 .
solving the optimization problem given by eq .
( [ eq : opt1 ] ) directly for a fixed target time @xmath89 is challenging even without constraints on the biases @xmath41 as the landscape is extremely complicated with many local extrema , resulting in trapping of local optimization approaches such as quasi - newton methods .
[ fig : ring9-errors ] shows the results of various runs for fixed times for a ring of @xmath90 spins , with the objective being to propagate the excitation from spin @xmath26 to spin @xmath91 , @xmath92 , @xmath93 and @xmath94 , respectively .
fixed time values @xmath89 from @xmath26 to @xmath95 with steps of @xmath96 were set and a quasi - newton optimizer with random initial values for the biases repeated @xmath97 times for each @xmath89 .
each point in this plot represents one run for the respective time @xmath89 and the achieved infidelity value @xmath98 .
there are many points for which the error is large ( i.e. , the transmission fidelity is low ) , indicating that the optimization for this run converged to a local extremum , making finding a good solution very expensive .
we also tested some global search algorithms that appeared successful for hard optimization problems such as a variant of differential evolution used in @xcite , but these proved to be much slower and produced generally worse results than the best solutions found by repeating a quasi - newton optimization algorithm for a relatively large number of initial states .
the results in fig .
[ fig : ring9-errors ] also indicate that there are only certain times for which we can expect to find a high information transfer fidelity . instead of fixing the time we can also add the time as additional parameter to optimize over and start for random initial biases and times .
the results of this are shown in fig .
[ fig : ring13_1 - 5 ] for the propagation from spin @xmath26 to @xmath94 in a ring of @xmath99 spins .
we report the solution with the highest fidelity overall and the fastest solution with a fidelity larger than @xmath100 . typically the highest fidelity solutions are found at longer times but good solutions for short times are also achieved .
however , many restarts of the optimization are required , and many runs fail with fidelities smaller than @xmath101 .
inspection of the good solutions found showed that many of these involved biases that were highly mirror symmetric w.r.t . the symmetry axis between the input and output spin on the ring .
to @xmath94 for a ring of @xmath99 spins over the spatial biases and times starting from random initial biases and times .
left - bottom shows the biases giving the best fidelity at time @xmath102 yielding the @xmath103 shown in the left - top image ( in blue vs. the natural evolution in green ) .
the two graphs in the middle show the fastest solution found with a fidelity greater than @xmath100 at time @xmath104 .
right - top shows the overall found solutions by the optimization plotting the time vs. the logarithm of the infidelity .
right - bottom plots a histogram of the logarithm of the infidelity . ] in principle the information transfer fidelity depends on all eigenvectors and eigenvalues of the hamiltonian
. however , the results in section [ sec : eigen ] show that the structure of the eigenvalues and eigenvectors must fulfill a specific condition to be able to maximize the transfer fidelity .
while there are many potential structures that fulfill the condition , we can choose a specific one to provide a guide for good initial values or a restricted domain for the population .
this significantly improved the solutions found by the optimal control algorithms .
to @xmath94 for a ring of @xmath99 spins over the spatial biases and times starting from random initial biases and times , where the biases are constrained to be symmetric across the axis going through the middle between initial and target state .
left - bottom shows the biases giving the best fidelity at time @xmath105 yielding the @xmath103 shown in the left - top image ( in blue vs. the natural evolution in green ) .
the two graphs in the middle show the fastest solution found with a fidelity greater than @xmath100 at time @xmath104 .
right - top shows the overall found solutions by the optimization plotting the time vs. the logarithm of the infidelity .
right - bottom plots a histogram of the logarithm of the infidelity . ]
the basic idea for imposing a specific eigenstructure is to quench the ring of @xmath14 spins into a chain from the initial spin to the target spin .
our previous work showed that this can be easily achieved by applying a very strong potential in the middle between the initial and target spin @xcite .
if we can control the potentials of all spins then we can generalize this to quench the ring just before the initial and after the target node , giving two options for a chain connecting the two nodes where either could provide a solution .
as it turns out , this gives rise to a general eigenstructure induced by applying mirror symmetric potentials across the axis going through the middle between initial and target state in the ring .
we consequently choose such symmetric potentials in combination with the approximate times where the maximum fidelity is achieved in the related chains as initial values for the optimal control algorithm .
this significantly improves the results and efficiently finds controls for maximum information transfer in minimum time for any initial and target spin .
this approach is also further justified by the results found using random initial values .
to @xmath94 in a ring of @xmath99 spins , optimizing both spatial biases and transfer times , starting with initial times corresponding to peaks in the chain transition and random biases .
left - bottom shows the biases giving the best fidelity at time @xmath106 yielding the @xmath103 in left - top graph ( in blue vs. the natural evolution in green ) .
the graphs in the middle show the fastest solution found with a fidelity greater than @xmath100 at time @xmath107 .
top - right plot shows the time - error distribution of the solutions found over @xmath97 runs and bottom right graphs shows a histogram of the logarithmic infidelity . ]
to @xmath108 in @xmath99-ring ( blue ) with a bias of @xmath109 on the spins from @xmath110 to @xmath99 vs. end - to - end natural transfer in @xmath59 chain ( red ) ] to @xmath94 for a ring of @xmath99 spins , optimizing both transfer times and spatial biases , starting with initial times derived from the peaks of the corresponding chain transition and biases that are random but subject to symmetry constraints .
left - bottom shows the biases giving the best fidelity at time @xmath111 yielding the @xmath103 shown in the left - top image ( in blue vs. the natural evolution in green ) .
middle graphs show the fastest solution found with a fidelity greater than @xmath100 at time @xmath112 .
right - top shows the overall found solutions by the optimization plotting the time vs. the logarithm of the infidelity .
right - bottom plot shows histogram of logarithmic infidelity . ]
the symmetry constraint can be easily applied by reducing the number of biases to be found to @xmath113 , symmetric across the symmetry axis between initial and target state .
[ fig : ring13_1 - 5_symmetry ] shows the results with this constraint similar to the results in fig .
[ fig : ring13_1 - 5 ] without the symmetry constraint .
similar shortest time solutions are found , while the best solution is at a different time .
this is not surprising considering that the solution found strongly depends on initial values .
more importantly , there are slightly fewer failed runs .
convergence of the optimization can be further improved by selecting constants , peaks or troughs as biases between initial and target spin on both sides of the ring randomly .
[ fig : ring_v_chain ] compares the information transfer probabilities from spin @xmath26 to @xmath59 in an @xmath14 ring with a constant bias on the spins @xmath59 to @xmath14 with the information transfer probabilities between the end nodes of a chain of length @xmath59 . from this example and theory
@xcite it is obvious that the maxima coincide and the stronger the bias the more similar the information transfer probabilities between the quenched ring and the chain .
hence , it seems likely that we can obtain better results if the initial times are taken from the largest peaks , say those over @xmath114 , of the chain transition , which can easily be approximated by evaluating the probabilities for the chain at a regular sampling .
[ fig : ring13_1 - 5_time ] shows the results for the @xmath26 to @xmath94 transition of the @xmath99 ring for random , unconstrained initial biases and the initial times taken from the peaks of the corresponding @xmath26 to @xmath94 transition in a @xmath94 chain .
this resulted in a faster time being found , but still quite a few failed runs .
combining this with the symmetry constraint gives the results shown in fig .
[ fig : ring13_1 - 5_symmetry_time ] , resulting in the same short time and fewer failed runs .
we can further add a constraint to limit the strength of the biases .
[ fig : ring13_1 - 5_cons ] shows the results . to achieve the same effect of quenching the ring to a chain
, all potentials are moved towards the maximum instead of pushing the two potentials at the two end spins to very high values .
overall , results for other rings and transitions are similar to the results presented for the @xmath26 to @xmath94 transition for a ring of size @xmath115 .
fig [ fig : shortest ] shows the shortest times achieved for transition fidelities greater than @xmath100 .
this indicates that the shortest transition times depend largely on the distance between the spins and not the size of the ring , consistent with quenching the ring into a chain .
although we have no proof that there is no shorter transition time , the results seem to indicate that the shortest time in the ring is closely related to the time of the natural evolution to the corresponding chain , in case of static bias controls . in special cases
we can further derive values for the expected minimum transfer time and the corresponding biases . if the distance between the initial and target spin is @xmath26 then quenching the ring as described reduces the network to a two - spin system with direct coupling and an effective hamiltonian of the form @xmath116 which undergoes rabi oscillations with the rabi frequency @xmath117 .
it can easily be shown that @xmath118 .
the maximum @xmath119 is assumed for @xmath120 , if and only if @xmath121 , or @xmath122 . indeed the numerical optimization results in fig .
[ fig : ring9-errors ] for transfer from @xmath26 to @xmath91 ( top left ) in a ring of size @xmath90 show that the first minimum is close to @xmath123 and occurs for @xmath124 .
to @xmath94 in a ring of @xmath99 spins , optimizing both spatial biases and transfer times , starting with initial times corresponding to peaks in the chain transition and biases that are random but subject to symmetry and amplitude constraints @xmath125 .
left - bottom shows biases giving the best fidelity at time @xmath111 yielding @xmath103 shown in the left - top image ( in blue vs. natural evolution in green ) .
the graphs in the middle show the fastest solution found with a fidelity @xmath126 at time @xmath112 .
top - right plot shows the time - error distribution of the solutions found over @xmath97 runs and bottom right graphs shows a histogram of the logarithmic infidelity . ] similarly , if the distance is @xmath91 the ring is reduced to a three - spin chain . in this case
we can easily show that , assuming zero - bias , @xmath127 , @xmath128 and thus @xmath129 , i.e. , we have perfect state transfer , for @xmath130 .
[ fig : ring9-errors ] ( top - right ) shows that the numerical results are indeed consistent with this solution with the first minimum for the @xmath131 transfer occurring for @xmath132 .
more generally , if the distance between the input and output nodes is @xmath133 and the biases satisfy @xmath134 then @xmath135 commutes with the permutation @xmath136 $ ] . if @xmath137 is the corresponding permutation matrix then @xmath138 and thus @xmath139 or @xmath140 . in particular , this means that the first and last row of @xmath47 are the same and eq .
( [ eq:3 ] ) becomes @xmath141 this expression vanishes if @xmath142 is a multiple of @xmath143 for @xmath144 . in the previous example , for a chain of length @xmath92 with no bias , @xmath145 and @xmath146 , hence we achieve perfect state transfer for @xmath147 setting @xmath148 . for rings of size @xmath149 and transitions from @xmath26 to @xmath150
we have demonstrated how static controls can be used to control the information flow in @xmath0-spin rings . compared to dynamic control , finding static controls is a considerably harder optimization problem due to the complexity of the optimization landscape .
careful selection of initial values and enforcement of the constraints derived from eigenstructure analysis substantially improve the performance of the algorithm . for rings ,
the constraints enforce symmetries that make the rings more similar to chains , and the timing of transmission peaks for corresponding chains give a good indication of the shortest possible times that can be achieved .
better understanding of the symmetry constraints , their relation to the reachability of target states and the design of global optimization algorithms that utilize the specific structure of the problem may yield even better solutions . | propagation of information encoded in spin degrees of freedom through networks of coupled spins enables important applications in spintronics and quantum information processing .
we study control of information propagation in networks of spin-@xmath0 particles with uniform nearest neighbor couplings forming a ring with a single excitation in the network as simple prototype of a router for spin - based information .
specifically optimizing spatially distributed potentials , which remain constant during information transfer , simplifies the implementation of the routing scheme .
however , the limited degrees of freedom makes finding a control that maximizes the transfer probability in a short time difficult .
we show that the structure of the eigenvalues and eigenvectors must fulfill a specific condition to be able to maximize the transfer fidelity , and demonstrate that a specific choice among the many potential structures that fulfill this condition significantly improves the solutions found by optimal control . |
the origin of the distribution of mass and angular momentum in disk galaxies is yet an unsolved astrophysical puzzle .
eggen , lynden - bell and sandage ( 1962 ) argued that spiral galaxies like the milky way formed by rapid infall of an initially uniform sphere of gas into a centrifugally supported disk .
it was soon realized that these disks would have characteristic exponential surface density distributions if the initial gas sphere would be in solid body rotation and if the gas would preserve its initial specific angular momentum distribution @xmath0 during infall ( mestel 1963 , crampin & hoyle 1964 , innanen 1966 , freeman 1970 ) . here
, @xmath0 is the cumulative mass of gas with angular momentum less or equal to @xmath1 . since these first pioneering studies our insight into galaxy formation has substantially changed and improved .
current cosmological models consider a dissipationless cold dark matter ( cdm ) component to dominate structure formation in the universe ( blumenthal et al .
1984 ) . initially small dark matter density perturbations in the early universe
decouple from the hubble flow , collapse into virialized dark matter halos and merge into larger and larger structures .
gas accumulates within the extended dark halos , dissipates its kinetic energy and settles into the equatorial plane as soon as centrifugal equilibrium is being reached , forming fast rotating disks that subsequently turn into stars ( white & rees 1978 ) .
fall & efstathiou ( 1980 ) argued that gas and dark matter should initially be well mixed . in this case , the specific angular momentum distribution of the gas should be equal to that of the dark halo .
if @xmath0 would be preserved during infall into the equatorial plane , the exponential disk scale length @xmath2 would be directly related to the specific angular momentum @xmath3 of the dark halo . here
@xmath3 is the dimensionless spin parameter ( e.g. peebles 1969 ) @xmath4 where j , e , and @xmath5 are the total angular momentum , energy and virial mass of the halo , respectively and g is newton s constant .
this shifted the focus to a more detailed investigation of the spin properties of dark halos and a determination of @xmath3 .
peebles ( 1969 ) suggested that halos would acquire non - negligible specific angular momentum by the gravitational interaction of their building blocks with neighboring structures .
subsequent cosmological n - body simulations ( barnes & efstathiou 1987 , efstathiou & barnes 1983 , zeldovich & novikov 1983 , cole & lacey 1996 ) confirmed this mechanism .
they showed that prior to collapse the angular momentum of a dark fluctuation grows roughly linearly with time , as predicted by linear tidal - torque theory ( white 1984 ) until it decouples from the hubble flow , collapses and virializes .
the @xmath3 distribution of virialized halos turned out to be well described by a log - normal ( steinmetz & bartelmann 1995 , cole & lacey 1996 , gardner 2001 , bullock et al .
2001 ) @xmath6 with median value of @xmath7 and dispersion @xmath8 .
a more partical spin parameter @xmath9 was proposed by bullock et al .
( 2001 ) : @xmath10 with @xmath11 the halo virial radius and @xmath12 its virial velocity .
the spin parameters @xmath3 and @xmath9 turn out to be very similar due to the fact that dark halos are well described by a universal density distribution ( navarro , frenk & white , 1997 , nfw ) and the @xmath9 distribution also follows a log - normal with best fit values @xmath13 and @xmath14 ( bullock et al . 2001 ) . given @xmath9 , and assuming that the disk gas will have the same specific angular momentum as the dark halo , the exponential disk scale length @xmath2 can be easily determined . adopting a flat rotation curve with velocity @xmath15 , the disk s specific angular momentum is ( mo et al .
1998 ) @xmath16 for typical nfw halos with concentrations @xmath17 , the peak velocity of the dark matter rotation curve which is in general of order the observed peak rotation velocity is @xmath18 . for a flat @xmath19cdm - cosmology with cosmological parameters @xmath20 , @xmath21 and a hubble constant of 70 km s@xmath22 mpc@xmath22
the virial parameters of dark halos are coupled by the relations ( navarro & steinmetz 2000 ) @xmath23 combining these questions leads to a relationship between the exponential disk scale length @xmath2 and the disk s rotational velocity @xmath15 : @xmath24 a more detailed investigation which takes into account adiabatic contraction of the dark halo ( jesseit et al .
2002 ) and the combined gravitational force of the disk and dark halo shows that equation 6 overestimates the disk scale length by a small amount of order 10% to 20% .
the dashed curve corresponds to @xmath25,width=384 ] figure 1 compares the observations ( courteau 1997 ) with the prediction of equation 6 , adopting a typical value of @xmath26 .
note that the slope reflects directly the spin parameter @xmath9 .
the theoretical predictions lead to a correlation that is steeper than observed .
the dashed line shows a best fit model which requires a value of @xmath27 .
if the rotation speeds of galactic disks are approximately the same as the maximum circular velocities of their dark halos , and if the gas would have the same specific angular momentum as the dark halo , cosmological models predict disk scale lengths that are about a factor of 1.4 larger than observed .
good agreement could be achieved if the gas retained only 70% of the available angular momentum during infall .
a similar conclusion was reached by navarro & steinmetz ( 2000 ) and mo , mao & white ( 1998 ) who investigated the structural properties of galactic disks within characteristic nfw - type dark halos in greater details .
numerical simulations have become one of the most powerful tools for exploring galaxy formation .
hydrodynamical simulations of disc formation were initiated by navarro @xmath28 benz ( 1991 ) , who included radiative cooling by hydrogen and helium , and attempted to account for star formation and feedback processes .
in contrast to the conclusion drawn in the previous chapter , the simulated galaxies however failed to reproduce their observed counterparts : the disks were found to be too small and were more centrally concentrated than actual galaxies ( navarro @xmath28 white 1993 ) .
in addition , star formation in the models was overly efficient , converting gas into stars too fast .
many of the shortcomings of the early modeling could be accounted for by the limited resolution of the simulations and the way in which feedback was treated . as star formation was very efficient in low mass halos at high redshifts ,
a large number of dense , compact stellar systems formed that lateron were collected in the innermost regions of larger galaxies that formed through the merging of smaller objects .
angular momentum loss by dynamical friction drove these star clusters to the center where they formed large , dense bulges instead of extended disks .
.2 in of model disks with scale length @xmath2 and rotational velocity @xmath29 is compared with observational data ( filled points ) .
figure adapted from navarro & steinmetz ( 2000).,width=384 ] the cosmological angular momentum problem was reinvestigated lateron in greater details by navarro @xmath28 steinmetz ( 2000 ) with high - resolution n - body / gasdynamical simulations that included star formation and feedback to examine the origin of the i - band tully - fisher relation for different comologies .
although the slope and the scatter could be well reproduced in the simulations , the models failed to match the zero point ( fig .
2 ) . again , the galaxies were too compact with respect to the observations even with realistic feedback formulations that were calibrated to reproduce the empirical correlations , connecting the local star formation rate with the gas surface density ( kennicutt 1998 ) .
a possible solution is suppression of early cooling of the gas by strong feedback from supernovae which prevents drastic angular momentum loss and produces better fits to the observations ( sommer - larsen et al .
2003 , abadi et al .
however , even in these simulations the disk systems typically contained denser and more massive bulges than observed late - type galaxies , indicating that the specific angular momentum problem is not completely solved .
in the fall @xmath28 efstathiou ( 1980 ) model , the scale size of a galaxy is determined by its angular momentum , which is acquired by tidal torques from neighboring objects in the expanding universe , prior to the collapse of the halo .
recent results from numerical n - body simulations have suggested that major mergers might be especially important to increase the mean angular momentum content of the halos and their spin parameters ( gardner 2001 ; vitvitska et al . 2002 ) .
this is due to the substantial amount of orbital angular momentum which a major merger adds to the system and which dominates the final net angular momentum of the remnant ( gardner 2001 ) .
minor mergers , in contrast , have little effects on the angular momentum budget of galaxies .
semi - analytical models of galaxy evolution ( kauffmann et al .
1993 ) can not investigate the angular momentum properties of a halo self - consistently .
they can however predict their merging history and by this their spin parameters , if the spin is coupled with the history of minor and major mergers .
this question has been investigated in details by vitvitska et al .
they followed the evolution of the cumulative mass and the spin parameter of the major progenitors for three high - resolution halos and found that the spin parameters clearly change with time . instead of a gradual increase
, @xmath3 is increasing with every major merger and subsequently decreasing as a result of minor merging .
they proposed that the net angular momentum of dark matter halos originates from statistically random merging with preferentially the last major merger dominating . in this picture , the evolution of the halo angular momentum is quite different from the standard tidal torques scenario , in which the angular momentum grows steadily at early times and its growth flattens lateron .
.2 in is shown in the upper panel for dark halos which experienced a major merger .
the lower panel shows the evolution of the total angular momentum @xmath30 , the total mass @xmath31 and the total energy @xmath32 with time .
the time axis @xmath33 is centered on the epoch of major merging.,width=384 ] this conclusion has recently been questioned by hetznecker & burkert ( in preparation ) .
figure 3 shows the evolution of the mean spin parameter @xmath9 , averaged over a large set of dark halos that experienced at least one major merger in their lifetime
. the time axis is centered at the epoch of the major merging .
note that indeed a substantial increase in @xmath9 is visible when the merger occurs .
however , afterwards , within the next 2 gyrs , @xmath9 decreases again , approaching the same low value that had been achieved prior to the merging event .
hetznecker & burkert argue that the temporary increase in @xmath9 is a result of the fact that dark halos are out of virial equilibrium during this epoch .
dark particles oscillate into and out of the virial radius and the most bound particle , with respect to which the angular momentum is typically defined , is not a good measure of the center of mass . neglecting unrelaxed dark halos ,
the authors find no difference between major and minor mergers and an average @xmath9-distribution that is on average shifted to somewhat lower values compared with previous studies that include unrelaxed halos .
most work has up to now focused on angular momentum properties of halos that had at least one dominant major merger during their evolution .
however , major merger events tend to destroy disks , producing spheroidal stellar systems , like bulges or early - type galaxies ( see e.g. review by burkert & naab 2003 ) .
not much work has been devoted to explore the angular momentum properties of halos that host pure disk galaxies , or bulgeless galaxies and that never experienced a major merger .
recently donghia @xmath28 burkert ( 2004 ) performed three n - body simulations in a @xmath19cdm cosmological universe and explored the angular momentum properties of halos that did not experience any major merger from redshift 3 until the present time and that are in principle good candidates to host bulgeless galaxies .
the authors traced each identified halo backward in time , following the mass of the most massive progenitor as a function of redshift during @xmath34 ( for details , see donghia @xmath28 burkert , these proceedings ) .
they found that disk - dominated late - type galaxies inhabiting halos that have not experienced a major merger have a distribution of @xmath9 that peaks around a value of 0.023 which is substantially smaller than expected from observed rotation curves of bulgeless disks . to demonstrate this they compared their results with the sample of van den bosch , burkert @xmath28 swaters ( 2001 ) who determined spin parameters for 14 late - type bulgeless disk galaxies .
.2 in of dark halos that experienced no major merger is shown by the filled histogram and compared to the spin distribution of observed bulgeless disk galaxies ( solid line ) .
the dotted curve shows the @xmath9 distribution if the spins of dark halos without major mergers would be multiplied by a factor of 3.15.,width=384 ] fig.4 shows the probability distribution of the spin parameter of halos that did not experience any major mergers since z=3 ( filled region ) and compares it with the normalized probability distribution of @xmath35 for the sample of galaxies measured by van den bosch , burkert @xmath28 swaters ( 2001 ) assuming a mass - to - light ratio of unity in the r band .
the galaxies show a distribution that follows a lognormal distribution with an average value for @xmath36 , and a dispersion of @xmath37 .
this is a factor of 3 larger than predicted by the numerical simulations .
again , galactic disks forming in these halos would be too small .
however , this time , the problem can not be solved by feedback processes which prevent the loss of specific angular momentum of clumpy infalling gas .
it is remarkable that multiplying the @xmath9 values of simulated halos by a factor 3.15 ( dotted histogram in fig .
4 ) reproduces the peak and dispersion of the observed @xmath35 distribution quite well .
halos without major mergers acquire their specific angular momentum through tidal torques in the early epochs of evolution ( barnes @xmath28 efsthatiou 1987 ) , when the density contrasts were small , in accordance with the prediction of the linear theory .
the net result are halos with typical spin parameters of @xmath38 whereas data for bulgeless galaxies indicate halos with values of @xmath39 , pointing out a new angular momentum problem , especially for bulgeless galaxies .
it s not clear how to overcome this problem .
it is known that the spin parameter distribution for the collapsed objects is insensitive to the shape of the initial power spectrum of density fluctuations , to the environment and the adopted cosmological model ( lemson @xmath28 kauffmann 1999 ) . in the fall @xmath28 efsthatiou ( 1980 )
model it is assumed that the gas and the dark matter in a protogalaxy have the same distribution of specific angular momentum . however the two components undergo different relaxation mechanisms : the dark matter experiences collisionless violent relaxation and the gas shocks and dissipates its kinetic energy and settles into the central regions .
van den bosch et al . ( 2002 ) used numerical simulations in a cold dark matter cosmology to compare the angular momentum distributions of dark matter and non - radiative gas .
they showed that gas and dark matter have identical angular momentum distributions after the protogalactic collapse phase if , and only if , cooling is ignored , in agreement with the standard assumptions .
in addition , they found that 5% to 50% of the mass has negative specific angular momentum .
realistic disks do not typically contain counterrotating material .
this suggests that during the cooling process the gas with negative specific angular momentum collides with material with positive specific angular momentum to build a bulge component .
thus , even without substructures and consequent dynamical friction a bulge would be likely to form in all galaxies due to this process , making the formation of bulgeless galaxies even more difficult .
.2 in ( circles with error bars ) and @xmath40 for edge - on numerical merger remnants resulting from collisions of disk galaxies .
the shaded area corresponds to the range occupied by bendo & barnes ( 2000 ) .
both distributions are not compatible with the observed one , even in this edge - on case .
, width=384 ] as outlined previously , numerical simulations indicate an angular momentum problem for disk galaxy formation that results from the fact that gas loses a substantial fraction of its angular momentum while settling into the equatorial plane .
cretton et al .
( 2001 ) however also found a related problem for fast rotating low - mass elliptical galaxies .
figure 5 shows the observed radial dependence of @xmath40 of their early - type galaxy sample with @xmath41 the observed line - of - sight rotational velocity of the stellar spheroid along the major axis and @xmath42 the local line - of - sight velocity dispersion .
the shaded region and the solid lines show the results of numerical simulations of unequal and equal - mass spiral galaxy mergers which reproduce many of the global properties of ellipticals remarkably well ( naab & burkert 2003 , burkert & naab 2003 ) .
note that none of these models leads to curves that rise as steeply as observed for many cases of low - mass ellipticals which show values of @xmath43 within 1 - 2 effective radii .
theoretical models instead predict values of @xmath44 .
like in the case of galactic disk formation , tidal torques and dynamical friction during the major merger event efficiently remove specific angular momentum from the baryonic stellar component , leading at the end to stellar spheroids with rotational velocities that do not exceed much the velocity dispersion .
.2 in of dark matter with angular momentum less than @xmath1 is fitted by equation ( 7 ) .
figure adapted from bullock et al .
( 2001).,width=384 ] early estimates assumed that the detailed specific angular momentum distribution of dark halos is well described by a hypothetical uniform sphere in solid body rotation .
this question was reinvestigated again by bullock et al .
( 2001 ) using a large statistical sample of halos that was drawn from a high - resolution @xmath19cdm simulation .
they found that the angular momentum distribution of dark halos has indeed a universal form which however strongly deviates from the previously expected uniformly rotating sphere .
the angular momentum distribution can be well fitted by ( fig .
6 ) @xmath45 where @xmath0 is the cumulative total mass of dark matter with angular momentum less than @xmath1 and @xmath46 is a free shape parameter .
the characteristic specific angular momentum @xmath47 is determined by @xmath48 with @xmath49 note , that this formula ignores any material with negative angular momentum ( chen et al .
2003 ) which makes the problem of low - angular momentum gas even worse , as discussed in chapter 5 .
following bullock et al .
( 2001 ) we explore the surface density distribution of galactic disks , adopting @xmath50 , a nfw halo with concentration @xmath51 and a disk baryon fraction of @xmath52 .
we keep @xmath53 as a free parameter within the measured range of @xmath54 and study its effect on the structure of the disk . assuming that the dark halo contracts adiabatically due to the gravitational force of the infalling gas , one can calculate the radius @xmath55 in the equatorial plane where gas of a given angular momentum @xmath56 reaches centrifugal equilibrium .
@xmath55 is given by the implicit equation @xmath57 where @xmath58 is the initial radius of the dark halo mass shell that after adiabatic contraction ends up at @xmath55 ( jesseit et al .
2002 ; blumenthal et al .
1984 ) : @xmath59 .2 in , @xmath60 and @xmath61 , respectively .
the upper left panel shows a log - log representation . the upper right panel shows the corresponding log - linear plot and
compares it with an exponential disk of scale length @xmath62 and @xmath63 , where @xmath64 is the dark halo virial radius .
the lower left panel shows the local disk scale length , normalized to the virial radius .
typical observed scale lengths are @xmath65 . the lower right panel shows the corresponding rotation curves adopting a dark halo mass of @xmath66 , which is characteristic for the milky way.,width=384 ] the upper left panel of figure 7 shows the normalized disk surface density profiles for @xmath67 , @xmath68 and @xmath69 .
the profiles agree with bullock et al .
( 2001 , see their fig .
20 ) . in this representation
it is difficult to estimate which profile , if any , would fit an exponential disk .
therefore the upper right panel shows again the same profiles , however now in a log - linear representation and focussing on the observed exponential disk regime of @xmath70 , where a disk scale length of @xmath71 has been assumed . an exponential disk with scale length of @xmath72 or @xmath73
is shown by the straight thin solid lines . even for large values of @xmath74
the profiles are not well fitted by an exponential .
all profiles instead rise above the exponential in the inner and outer regions .
the situation becomes even more clear in the lower left planel which shows the local exponential scale length defined as @xmath75 for all three values of @xmath53 . in all cases , @xmath76 is continuously and steeply increasing with radius , with no sign of a plateau at a characteristic scale length of order @xmath77 .
if one assumes that all material with small scale length @xmath78 forms a bulge component one can calculate the predicted bulge - to - disk ratio as function of @xmath53 .
we find that for all reasonable values of @xmath53 , galactic disks should harbour large bulges with bulge - to - disk mass ratios of more than 20% which is not consistent with the population of late - type disk galaxies .
viscous effects and secular evolution might change the surface density profiles of galactic disks and redistribute their specific angular momentum distribution .
it has indeed been shown e.g. by slyz et al .
( 2002 ) that exponential stellar disks would arise naturally from rather arbitrary initial conditions if the star formation timescale is equal to the viscous timescale .
it is however unlikely that viscous effects would increase the specific angular momentum of the gas in the innermost regions , by this reducing the bulge mass fraction .
although , on average , cosmological models of structure formation in a @xmath19cdm universe generate enough spin to explain the origin of extended galactic disks several problems still remain to be solved .
these include the loss of angular momentum during gas infall , the specific angular momentum distribution and exponential disk formation and the origin of bulge - less galaxies
. the last problem might be even more puzzling if many bulges formed by secular processes as summarized recently in kormendy & kennicutt ( 2004 ) .
feedback has been invoked as a mechanism to prevent the process of drastic angular momentum loss of infalling gas .
( van den bosch , burkert @xmath28 swaters 2002 ; maller @xmath28 dekel 2002 ; maller , dekel @xmath28 somerville 2002 ) .
however , donghia @xmath28 burkert ( 2004 ) have shown that the dark halos that experienced no major mergers have already too low an angular momentum to produce the observed disks and it is not clear which feedback process would increase the specific angular momentum of the gas beyond that of the dark component in order to explain this result .
selective outflow of especially low - angular momentum gas could provide another solution .
again it seems difficult to understand bulgeless galaxies in this context as in general a bulge component would be required to generate the kinetic energy to drive galactic winds .
another interesting scenario was proposed by katz et al .
( 2003 ) and birnboim and dekel ( 2003 ) who suggested that gas in high - redshift disk galaxies is accreted cold and without virial shocks directly from filaments .
more work along these lines is clearly required to understand the origin of galaxies and their angular momentum in greater details
. 1 abadi , m.g .
, navarro , j.f . , steinmetz , m. , eke , v.r .
2003 , apj , 591 , 499 barnes , j.e .
, @xmath28 efsthatiou , g. 1987 , apj , 319 , 575 bendo , g. & barnes , j. 2000 , mnras , 316 , 315 birnboim , y. , dekel , a. 2003 , mnras , 345 , 349 blumenthal , g.r . ,
faber , s.m . ,
primack , j.r . & rees , m.j .
1984 , nature , 311 , 527 bullock , j.s . ,
dekel , a. , kolatt , t.s . ,
kravtsov , a.v . ,
klypin , a.a . , porciani , c. , @xmath28 primack , j.r .
2001 , apj , 555 , 240 burkert , a. @xmath28 naab , t. 2003 , in galaxies and chaos , eds . g. contopoulos & n. voglis , springer , p. 327
chen , d. n. , jing , y.p .
, yoshikaw , k. 2003 , apj , 597 , 35 cole , s. & lacey , c. 1996 , mnras , 281 , 716 courteau , s. 1997 , aj , 114 , 2402 crampin , d.j , & hoyle , f. 1964 , apj , 140 , 99 cretton , n. , naab , t. , rix , h .- w . & burkert , a. 2001 , apj , 554 , 291 de jong , r.s . ,
lacey , c. 2000 , apj , 545 , 781 donghia , e. @xmath28 burkert , a. 2004 , apjl submitted , ( astro - ph/0402504 ) donghia , e. @xmath28 burkert , a. 2004 , this procceding .
efstathiou , g. & barnes , j. 1983 , in proc .
3rd moriond astrophys .
meeting , formation and evolution of galaxies and large structures in the universe , ed . j. audouze & j. tran thanh van ( dordrecht : reidel ) , p361 eggen , o.j .
, lynden - bell , d. & sandage , a.r . 1962 , apj , 136 , 748 fall , s.m . , efsthatiou , g. 1980 , mnras , 193 , 189 freeman , k.c .
1970 , apj , 160 , 811 gardner , j.p .
2001 , apj , 557 , 616 jesseit , r. , naab , t. & burkert , a. 2002 , apj , 571 , l89 innanen , k.a . 1966 , aj , 71 , 64 katz , n. , keres , d. , dave , r. , weinberg , d.h .
2003 , in `` the igm / galaxy connection : the distribution of baryons at z=0 '' , vol . 281 . ,
jessica l. rosenberg and mary e. putman kauffmann , g. , white , s.d.m .
& guiderdoni , b. 1993 , mnras , 264 , 201 kennicutt , r.c .
1998 , araa , 36 , 189 kormendy , j. & kennicutt , r.c .
2004 , araa , in press lemson , g. , @xmath28 kauffmann , g. 1999 , mnras , 302 , 111 maller , a.h . ,
@xmath28 dekel , a. 2002 , mnras , 335 , 487 maller , a.h . , @xmath28 dekel , a. , somerville , r. 2002 , mnras , 329 , 423 mestel , l. 1963 , mnras , 126 , 553 mo , h.j . , mao , s. & white , s.d.m .
1998 , mnras , 295 , 319 naab , t. , & burkert , a. 2003 , apj , 597 , 893 navarro , j. @xmath28 benz , w. 1991 , apj , 380 , 320 navarro , j. , frenk , c.s . &
white , s.d.m .
1997 , apj , 490 , 493 navarro , j.f . , steinmetz , m. 2000 , apj , 538 , 477 navarro , j. @xmath28 white , s.d.m .
1993 , mnras , 265 , 271 peebles , p.j.e .
1969 , apj , 155 , 393 slyz , a.d . ,
devriendt , j.e.g- , silk , j. & burkert , a. 2002 , mnras , 333 , 894 sommer - larsen , j. , gtz , m. , portinari , l. 2003 , apj , 596 , 47 steinmetz , r.s .
& bartelmann , m. 1995 , mnras , 272 , 570 van den bosch , f.c . , burkert , a. , @xmath28 swaters , r.a .
2001 , mnras , 326 , 1205 van den bosch , f. , abel , t. , croft , r.a.c . ,
hernquist , l. & white , s.d.m .
2002 , apj , 576 , 21 vitvitska , m. , klypin , a.a .
, kravtsov , a.v . ,
bullock , j.s .
, primack , j.r . ,
wechsler , r.h .
2002 , apj , 581 , 799 white , s.d.m .
1984 , mnras , 286 , 38 white , s.d.m . &
rees , m.j .
1978 , mnras , 183 , 341 zeldovich , y.b . &
novikov , i.d .
1983 , in relativistic astrophysics , ed .
g. steigman ( chicago : univ .
chicago press ) , p384 | the importance of angular momentum in regulating the sizes of galactic disks and by this their star formation history is highlighted .
tidal torques and accretion of satellites in principle provide enough angular momentum to form disks with sizes that are in agreement with observations .
however three major problems have been identified that challenge cold dark matter theory and affect models of galaxy evolution : ( 1 ) too much angular momentum is transferred from the gas to the dark halos during infall , leading to disks with scale lengths that are too small , ( 2 ) bulgeless disks require more specific angular momentum than is generated cosmologically even if gas would not lose angular momentum during infall , ( 3 ) gravitational torques and hierarchical merging produce a specific angular momentum distribution that does not match the distribution required to form exponential disks naturally ; some gas has exceptionally high angular momentum , leading to extended outer disks while another large gas fraction will contain very little specific angular momentum and is expected to fall into the galactic center , forming a massive and dominant bulge component .
any self - consistent theory of galaxy formation will require to provide solutions to these questions .
selective mass loss of low - angular - momentum gas in an early phase of galaxy evolution currently seems to be the most promising scenario .
such a process would have a strong affect on the early protogalactic evolution phase , the origin and evolution of galactic morphologies and link central properties of galaxies like the origin of central massive black holes with their global structure .
galaxies : disks , formation , kinematics and dynamics , structure - dark matter |
several classical special function inequalities , such as fejr s inequality @xcite or the askey - gasper inequality @xcite , assert the positivity of an object that can be defined by a linear recurrence with polynomial coefficients . even for the special case of linear recurrences @xmath4 with constant coefficients @xmath5 it is not always a simple matter to decide from the recurrence coefficients and the real initial values @xmath6 whether the solution is positive or not .
we call sequences @xmath0 that satisfy a recurrence of the form .
zeilberger @xcite gives them the more suggestive name _
@xmath7-finite sequences_. linear combinations
( with constant coefficients ) of recurrence sequences are recurrence sequences again , so positivity results are useful for comparing the magnitude of two sequences , too .
it is well known @xcite that the sequence @xmath0 can be written in terms of the roots @xmath8 of the _ characteristic polynomial _
@xmath9 of the recurrence as a generalized power sum @xmath10 where the @xmath11 are polynomials in @xmath1 with complex coefficients . given a recurrence of the form and initial values @xmath6 , the @xmath12 and the @xmath13 can be readily computed .
we refer to the @xmath12 that occur in with nonzero coefficient as _ characteristic roots of _ @xmath0 .
the characteristic roots of maximal modulus will be called _ dominating characteristic roots of _ @xmath0 .
[ ex : cfinite rec ] consider the recurrence @xmath14 its characteristic polynomial is @xmath15 and the solution is given by @xmath16 where the coefficients @xmath17 and @xmath18 depend on the real initial values @xmath19 .
we may ask ourselves whether @xmath20 approaches @xmath21 from one side only . if @xmath22 and @xmath23 do not both vanish , it is natural to expect ( and will be established in this paper ) that this does not hold , because the @xmath24 term seems to oscillate .
we pose the following conjecture .
[ cj : equ abs ] let @xmath0 be a recurrence sequence with no real positive dominating characteristic root .
then there are infinitely many @xmath1 with @xmath25 and infinitely many @xmath1 with @xmath26 .
the sequence @xmath0 might not oscillate if there is a real positive dominating characteristic root . see section [ se : pos real ] for more on this .
so far conjecture [ cj : equ abs ] has only been verified for one dominating characteristic root ( trivial ) and for one pair of conjugate complex roots @xcite .
we can not follow an argument from nagasaka and shiue @xcite , viz . that this special case should immediately imply the truth of the conjecture in general .
the main goal of this paper is to establish the following theorem by an extension of burke and webb s proof .
[ thm : main ] let @xmath0 be a recurrence sequence with at most four dominating characteristic roots , none of which is real positive
. then there are infinitely many @xmath1 with @xmath25 and infinitely many @xmath1 with @xmath26 .
the rest of the paper is organized as follows . in section [ se : pre ] we reduce theorem [ thm : main ] from multiple roots to simple roots and subsequently to a geometric statement about the distribution modulo one of integer multiples of a real vector @xmath27 , except for some special cases of theorem [ thm : main ] that are settled in section [ se : compl ] . in section [ se : irr ] we deduce the desired result from kronecker s approximation theorem , provided that one of @xmath28 is irrational .
the proof in the case where both are rational is the subject of sections [ se : rat ] and [ se : edrat ] . section [ se : pos real ] presents a metric result that deals with the case of a positive real root . in the conclusion we comment , among other things , on extending our approach
to conjecture [ cj : equ abs ] to an arbitrary number of dominating characteristic roots .
we write @xmath29 , @xmath30 , @xmath31 , @xmath32 , @xmath33 and @xmath34 for the sets of natural numbers ( including zero ) , integers , rational numbers , non - negative real numbers , real numbers and complex numbers , respectively .
the conjugate of a complex number @xmath35 is denoted by @xmath36 . whenever @xmath37 is a vector , we use the same letter with a subscript for its components , as in @xmath38 . for vectors
@xmath39 , @xmath40 of real numbers and a real number @xmath41 , we write @xmath42 we define the open rectangle parallel to the axes with side lengths @xmath43 centred at @xmath44 as @xmath45 for an open square parallel to the axes we write @xmath46 a _ lattice _ @xcite is a discrete subgroup @xmath47 .
its determinant is denoted by @xmath48 .
the lattice @xmath49 is defined in section [ se : rat ] .
let @xmath0 be as in theorem [ thm : main ] .
we order the characteristic roots @xmath50 of @xmath0 such that @xmath51 contain all real dominating characteristic roots , precisely one element of every pair of conjugate non - real dominating characteristic roots and no other roots .
note that this implies @xmath52 or @xmath53 .
moreover , let @xmath51 be ordered such that @xmath54 for some @xmath55 .
then we obtain @xcite @xmath56 where @xmath57 is the leading coefficient of @xmath11 .
this formula shows that theorem [ thm : main ] can be deduced from burke and webb s result ( @xmath58 ) and the following theorem ( @xmath59 ) .
observe that we can safely assume @xmath60 , since we can divide by the positive factor @xmath61 .
[ thm : main2 ] let @xmath62 , @xmath60 , @xmath63 .
let further @xmath64 be nonzero complex numbers and @xmath65 then there is @xmath66 such that @xmath67 for infinitely many @xmath1 and @xmath68 for infinitely many @xmath1 .
note that if @xmath69 was replaced by zero , it might happen that e.g. all negative values @xmath70 are so small in absolute value that the remainder term of @xmath3 , which comes from the characteristic roots of smaller modulus , takes over and makes the corresponding values @xmath3 positive .
this uniformity condition was missed by burke and webb @xcite .
they only argue that @xmath71 has infinitely many positive and infinitely many negative values , which is not sufficient , but their proof can be easily repaired .
now let @xmath72 , @xmath73 , @xmath22 , @xmath23 be as in theorem [ thm : main2 ] .
replacing @xmath74 by @xmath75 and vice versa if necessary , we may assume @xmath76 .
putting @xmath77 , we obtain by standard formulas @xmath78 where the coefficients are nonzero real numbers @xmath79 and the @xmath80 are given by @xmath81 we turn our attention to the signs of @xmath82 .
if we can prove that for every pair @xmath83 of @xmath84 s and @xmath85 s there are infinitely many @xmath1 such that the sign of @xmath82 equals @xmath86 for @xmath87 , we will have shown that @xmath88 oscillates , whatever the values of the @xmath57 ( and thus the @xmath89 ) are .
in other words , we are looking for @xmath1 such that @xmath900,\pi[\ ] ] or @xmath90\pi,2\pi[,\ ] ] respectively . to get the @xmath69 in theorem [ thm : main2 ]
, we have to shrink the intervals to @xmath91\epsilon,\pi-\epsilon [ \quad \text{and }
\quad ] \pi+\epsilon,2\pi-\epsilon[\ ] ] for some small @xmath92 , of course independent from @xmath1 .
now we rescale to the unit interval .
[ thm : xi0,xi1 ] let @xmath930,1\right[\backslash\{\tfrac{1}{2}\}$ ] such that @xmath94 and , if both @xmath95 and @xmath96 are rational , then the pair of their denominators ( written with the larger denominator first ) is none of @xmath97 , @xmath98 , @xmath99 .
then for all @xmath100 there is @xmath92 such that there are infinitely many @xmath1 with @xmath101 since the sine function is continuous , applying this theorem with @xmath102 and @xmath103 to make @xmath104 positive and @xmath105 for a negative sign proves theorem [ thm : main2 ] , unless one of the @xmath12 is a negative real number ( which implies @xmath106 ) or @xmath107 are rational numbers with denominators in @xmath108 .
section [ se : compl ] deals with these special cases of theorem [ thm : main2 ] . in the proof of theorem [ thm : xi0,xi1 ]
we distinguish the following three cases : 1 .
@xmath109 are linearly independent over @xmath31 .
[ case1 ] 2 .
@xmath110 are not both rational , but satisfy a linear relation @xmath111 with @xmath112 . [ case2 ] 3 .
@xmath95 and @xmath96 are both rational .
[ case3 ] section [ se : irr ] settles the first two cases .
the proof of theorem [ thm : xi0,xi1 ] in case 3 is the content of sections [ se : rat ] and [ se : edrat ] .
we remark that in order to prove conjecture [ cj : equ abs ] for one pair of conjugate complex dominating roots , it suffices to show that for every real number @xmath113 with @xmath114 and every real number @xmath115 there is @xmath92 such that for infinitely many @xmath1 @xmath116c-\tfrac{1}{4}+\epsilon , c+\tfrac{1}{4}-\epsilon\right [ \bmod 1.\ ] ] this is essentially what was done ( without @xmath117 , cf .
the introduction ) by burke and webb @xcite .
the closure of the set of integer multiples of a vector @xmath118 modulo one is described by a classical result from diophantine approximation .
[ thm : kronecker ] let @xmath95 , @xmath96 be real numbers . * if @xmath119 are linearly independent over the rationals , then the points @xmath120 , @xmath121 , lie dense in the unit square . *
if @xmath28 are not both rational , but satisfy a relation @xmath111 with @xmath112 and @xmath122 , then the points @xmath120 , @xmath121 , lie dense on the portions of the lines @xmath123 which lie within the unit square .
see e.g. niven ( * ? ? ?
* theorems 3.4 and 3.6 )
. for @xmath124 , @xmath125 and @xmath126 part ( i ) of theorem [ thm : kronecker ] settles case 1 of theorem [ thm : xi0,xi1 ] .
we proceed to case 2 .
let @xmath100 be arbitrary but fixed and @xmath127 be as in part ( ii ) of theorem [ thm : kronecker ] . since @xmath128 it suffices to find infinitely many @xmath120 in the set @xmath129 where @xmath92 is yet to be chosen .
first suppose that @xmath95 and @xmath96 are irrational .
then the parallel lines @xmath127 are neither horizontal nor vertical , since @xmath130 .
two adjacent lines @xmath127 , @xmath131 have horizontal distance @xmath132 and vertical distance @xmath133 . since @xmath94 , one of these quantities must be smaller than or equal to @xmath134 .
thus @xmath135 in fact this set is not only non - empty but contains a line segment . clearly , we can find @xmath92 such that the set @xmath136 still contains a line segment of length greater than zero .
filling this line segment densely with points @xmath120 requires infinitely many @xmath1 .
now let @xmath95 be rational and @xmath96 be irrational , and let @xmath137 be the denominator of @xmath95 .
this implies @xmath138 .
then the lines @xmath127 are vertical , and the horizontal distance between @xmath127 and @xmath131 is @xmath139 , since @xmath140 by the assumptions of theorem [ thm : xi0,xi1 ] . case 2 of theorem [ thm : xi0,xi1 ] is proved .
the main goal of this section and the next one is to prove the following theorem .
[ thm : genrat ] let @xmath141 , @xmath142 , @xmath143 , @xmath144 for @xmath87 and @xmath145 .
then there is @xmath146 ^ 2 $ ] such that for all @xmath121 @xmath147 provided that @xmath148 and there is no such @xmath149 if does not hold . to see that case 3 of theorem [ thm : xi0,xi1 ] follows from theorem [ thm : genrat ] , note that the purely periodic sequence @xmath150 assumes each of its finitely many values infinitely often .
the @xmath117 has disappeared because the set of all @xmath151 is finite and @xmath152 is open . if @xmath153 , we necessarily have @xmath154 , and we may take @xmath155 and @xmath156 arbitrary .
( see figure [ fi : spec cases ] for an example . ) if @xmath157 , it is easy to see that for all @xmath158 in question the set of integer multiples modulo one is one of the two sets @xmath159 obtained from @xmath160 and @xmath161 , respectively .
similarly , for @xmath162 it suffices to consider @xmath163 .
this is also true for @xmath164 , if we take @xmath165 instead of @xmath166 .
the number of @xmath158 s to check can be reduced further by taking advantage of some obvious symmetries . by the subsequent lemma
, the alternative with negative first entry can be discarded in each of the three cases .
figure [ fi : spec cases ] shows that in the remaining cases we may take @xmath167 , @xmath168 and @xmath169 , respectively .
[ le : symm ] define the maps @xmath170 and @xmath171 on @xmath172 by @xmath173 then for all real numbers @xmath95 , @xmath96 @xmath174 obvious . for @xmath175 , @xmath176 , @xmath177 and @xmath178 , respectively .
, title="fig : " ] for @xmath175 , @xmath176 , @xmath177 and @xmath178 , respectively .
, title="fig : " ] for @xmath175 , @xmath176 , @xmath177 and @xmath178 , respectively .
, title="fig : " ] for @xmath175 , @xmath176 , @xmath177 and @xmath178 , respectively . , title="fig : " ] we have shown this implication just for the sake of completeness .
the interesting part of theorem [ thm : genrat ] for our purpose is the converse implication .
its proof is the content of the remainder of this section and of the following section .
[ def : l ] let @xmath179 be a positive integer and @xmath180 be integers relatively prime to @xmath179
. then we define the _ lattice of multiples of @xmath181 modulo @xmath179 _ as @xmath182 alternatively @xcite , @xmath49 can be defined as the lattice generated by the vectors @xmath183 and @xmath184 .
the lattices @xmath49 will provide a convenient representation of the sets of integer multiples of rational numbers modulo one , which we encountered in theorem [ thm : genrat ] . for this purpose
we require a version of the well - known chinese remainder theorem for moduli that are not necessarily pairwise relatively prime .
[ thm : gcrt ] let @xmath185 be positive integers and @xmath186 be integers . then there is an integer @xmath187 with @xmath188 provided that @xmath189 see knuth ( * ? ?
* exercise 4.3.2.3 ) .
[ le : m to l ] let @xmath180 be integers and @xmath190 be positive integers with @xmath144 for @xmath87 and @xmath191 . then ( i ) & \ { n ( , ) 1 : n } = \ { ( , ) : l_g(a_1,a_2 ) , 0u_k < b_k } & + ( ii ) & l_g(a_1,a_2 ) = \ { ^2 : a_1u_2 a_2u_1 } & we have @xmath192 the latter equality and assertion ( ii ) follow from theorem [ thm : gcrt ] . . ]
[ ex : example l ] in example [ ex : cfinite rec ] the @xmath24 term yields @xmath193 and @xmath194 . the corresponding lattice @xmath195 is displayed in figure [ fi : example l ] .
let @xmath181 and @xmath196 be as in the assumptions of theorem [ thm : genrat ] , but such that @xmath197 is not in the set , and put @xmath191 . in the light of lemma [ le : m to l ] , it is an immediate consequence of the periodicity property @xmath198 that searching a point @xmath151 in a ` modded ' square @xmath199 amounts to looking for a point of the lattice @xmath49 in the rectangle @xmath200 with side lengths @xmath201 , @xmath202 .
we let @xmath57 absorb @xmath203 and write again @xmath204 for the arbitrary centre @xmath205 .
[ ex : point in rect ] if we want to show that the @xmath24 term in oscillates , we are lead to the problem of finding a point of @xmath206 in any rectangle @xmath207 , @xmath100 . if the numbers @xmath208 and @xmath209 are large , it is easy to find a point of @xmath49 in the rectangle , whereas @xmath210 is the most difficult case .
this is so because if we fix @xmath211 , @xmath212 and @xmath179 and enlarge @xmath208 and @xmath209 , the lattice @xmath49 remains invariant , while the rectangle becomes bigger . at first glance
, the problem seems to be easily reducible to the case of equal denominators @xmath210 .
in example [ ex : point in rect ] , if we could show that any square @xmath213 contains a point of @xmath206 , then it would follow at once that every rectangle @xmath207 contains a point of @xmath206 .
but we have already seen ( theorem [ thm : genrat ] ) that there are squares @xmath213 without points of @xmath206 .
in general , the catch is that even if @xmath214 satisfy the requirements of theorem [ thm : genrat ] and @xmath215 is not in the set , it may happen that @xmath216 violate the requirements of theorem [ thm : genrat ] or that @xmath217 is in .
therefore we choose a different approach for the case @xmath218 .
for relatively prime @xmath219 and @xmath220 the lattice @xmath49 equals @xmath221 .
all rectangles @xmath222 with @xmath100 have side lengths greater than one and therefore contain a point of @xmath221 . if @xmath223 , then @xmath211 and @xmath212 must be odd , hence @xmath224 since @xmath225 in this case , it is easy to see that this lattice contains a point of any rectangle @xmath222 .
from now on we assume @xmath226 .
the following proposition deals with the case @xmath227 . recall that @xmath228 , @xmath98 and @xmath99 need not be considered , because they are in the set .
[ pr:2 g 2 g ] let @xmath229 be as in theorem [ thm : genrat ] .
suppose @xmath230 , @xmath231 and @xmath232 .
then for all @xmath100 @xmath233 observe that by the periodicity property of @xmath49 it suffices to find a point of the lattice in the set @xmath234 let @xmath235 be the lower left corner of @xmath222 .
we assume w.l.o.g .
@xmath236 and define @xmath237p_2,p_2+\tfrac{1}{2}g[$ ] . then contains the set @xmath238 the interval @xmath239 contains at least two integers , since its length is @xmath240 .
since @xmath212 is invertible modulo @xmath179 , there are at least two points of @xmath49 in @xmath241 by part ( ii ) of lemma [ le : m to l ] , and at least one of them lies in .
now we consider values of @xmath219 that are at least @xmath242 , which completes the case @xmath218 of theorem [ thm : genrat ] .
[ pr:3 g ] let @xmath229 be as in theorem [ thm : genrat ] .
suppose @xmath243 and @xmath244 .
then for all @xmath100 @xmath233 it suffices to consider @xmath245 and @xmath246 .
proceeding analogously to the proof of proposition [ pr:2 g 2 g ] , we arrive at the set @xmath247 instead of .
the result follows from part ( ii ) of lemma [ le : m to l ] and @xmath248 .
in order to finish the proof of theorem [ thm : genrat ] , and thus the proof of theorem [ thm : xi0,xi1 ] , we will establish the following proposition .
[ pr : g ] let @xmath229 be as in theorem [ thm : genrat ] .
suppose @xmath249 .
then for all @xmath100 @xmath250 if @xmath49 contains one or two sufficiently short vectors , its points are dense enough so that the square @xmath251 is populated by at least one lattice point .
this is the basic idea of our proof of proposition [ pr : g ] . although there are algorithms @xcite tailored to @xmath49 for computing a reduced lattice basis , we do not know of any specialized a priori bounds for the norm of the basis elements . therefore , we appeal to the standard bound .
let @xmath252 be a subset of @xmath253 and @xmath254 be a lattice .
then the _ successive minima _ of @xmath252 w.r.t .
@xmath255 are defined for @xmath256 by @xmath257 in the following theorem , the term _ body _ denotes a set @xmath258 with non - empty interior such that @xmath252 is contained in the closure of its interior .
[ thm : mink2 ] if @xmath255 is an @xmath259-dimensional lattice in @xmath253 and @xmath258 is a bounded zero - symmetric convex body with volume @xmath260 , then @xmath261 see gruber and lekkerkerker s monograph ( * ? ? ?
* theorem 2.16.3 ) . from this theorem
we will deduce that @xmath49 must contain either two ` short ' linearly independent vectors or one ` very short ' nonzero vector .
if the first case occurs , we will apply the following result of bender @xcite .
[ le : bender ] let @xmath262 be a basis of a lattice @xmath263 , and let @xmath264 be the angle between @xmath265 and @xmath266 .
suppose further that @xmath267 is a bounded convex set such that the quotient of its area and its perimeter is greater than @xmath268 then @xmath269 contains a point of @xmath255 .
for the second case , where we find one vector of ` very small ' norm in @xmath49 , we could not find an applicable result in the literature that would ensure a lattice point in the square , so we provide one now .
[ le : one short vector ] let @xmath263 be a lattice and @xmath270 be a point of @xmath255 with @xmath271 and @xmath272 . let further @xmath273 be an open square with sides parallel to the axes and side length @xmath274 . if @xmath273 contains no point of @xmath255 , then @xmath275 there is a family @xmath276 of parallel equidistant lines with slope @xmath277 such that @xmath278 and the perpendicular distance between two adjacent lines of @xmath276 is @xmath279 ( * ? ? ?
* lemma iii.5 ) .
then the vertical distance between two adjacent lines is @xmath280 .
we claim @xmath281 if @xmath282 , then for each square @xmath283 there is a line in @xmath276 that goes through the left and the right edge of the square ( see figure [ fi : slope ] ) .
this settles the first case in the right hand side of .
if @xmath284 is larger than @xmath285 , there is a square that is not intersected by any line from @xmath276 .
we are left with the intermediate case @xmath286 . to achieve the minimum in
, we must certainly place @xmath287 such that there is no line from @xmath276 in the parallelogram @xmath288 of figure [ fi : slope ] .
but then there is always a line @xmath289 that intersects @xmath290 , say in the upper triangle of @xmath290 .
if no line intersects the lower triangle of @xmath290 , we can make the maximum in smaller by pushing @xmath287 downwards .
the smallest possible value of the maximum is achieved as soon as the intersections of @xmath287 with @xmath291 and the line from @xmath276 just below @xmath291 have equal length .
it is easy to see that these intersections both have horizontal length @xmath292 .
now that is established , let @xmath273 be an open square with sides parallel to the axes and side length @xmath293 our goal is to show @xmath294 .
if the first case in the right hand side of occurs , we are well off : since @xmath295 , the line segment in @xmath296 of horizontal length @xmath297 must contain a point of @xmath255 .
the third case in can not happen , since it would imply @xmath298 , contradicting .
as for the second case , @xmath299 implies @xmath300 hence @xmath294 .
( shaded ) and the parallelogram @xmath288 ( hatched ) , which lies between two lines of slope @xmath170 that go through the upper right and the lower left corner of @xmath287 , respectively . ]
we begin this proof , which is the core of the proof of theorem [ thm : main ] , by settling the cases where @xmath179 is at most @xmath301 .
the only numbers to consider are @xmath302 , since for smaller @xmath303 there are no @xmath180 that satisfy the requirements of theorem [ thm : genrat ] ( and hence of proposition [ pr : g ] ) .
first let @xmath304 . if we have proved the desired result for a pair @xmath184 , we need not consider the five pairs @xmath305 any more by lemma [ le : symm ] .
it is readily seen that under our restrictions on @xmath211 , @xmath212 all lattices @xmath306 are equal to @xmath307 modulo these symmetries .
similarly , for @xmath308 and @xmath309 it suffices to consider @xmath310 and @xmath311 , respectively . in all three cases it is easy to verify the desired result . from now on
we assume @xmath312 .
put @xmath313 and let @xmath314 be the unit circle .
it is not difficult to see ( * ? ? ?
* section 2 ) that the determinant of @xmath255 is @xmath315
. then theorem [ thm : mink2 ] shows @xmath316 first suppose @xmath317 .
the quotient of the area of @xmath251 and its perimeter is @xmath318 , hence we can apply lemma [ le : bender ] .
if , on the other hand , @xmath319 , then we have @xmath320 , which provides us with a nonzero point @xmath321 with @xmath322 .
w.l.o.g .
assume that @xmath323 satisfies @xmath271 and @xmath272 . according to lemma [ le : one short vector ]
, it suffices to show @xmath324 i.e. @xmath325 this inequality is satisfied for @xmath312 and @xmath326 which are all values of @xmath323 in question . observe that @xmath327 implies @xmath328 .
this completes the proof of theorems [ thm : genrat ] and [ thm : xi0,xi1 ] .
we remark that the successive minima approach from the preceding proof can be applied to the case of distinct denominators @xmath219 , @xmath220 , too .
however , the number of special cases that have to be checked separately is much larger than for equal denominators .
if @xmath96 from theorem [ thm : xi0,xi1 ] equals @xmath134 , which corresponds to @xmath329 and thus to a real negative dominating characteristic root @xmath73 in theorem [ thm : main ] , then the squares centred at @xmath330 or @xmath331 , @xmath156 , do not contain any point @xmath120 . but in this case we need not consider all squares : @xmath332 hence @xmath333 can be absorbed in @xmath334 , and we retain full generality if we assign a convenient value to @xmath333 . [ pr : eta real ] let @xmath3350,1\right[\backslash \{\tfrac{1}{2}\}$ ] .
then for all @xmath336 there are @xmath337 and @xmath338 such that for infinitely many @xmath121
@xmath339 let @xmath22 be an arbitrary real number with @xmath340 .
it suffices to show that there is @xmath121 with @xmath341c_1 - \tfrac{1}{4 } , c_1 + \tfrac{1}{4}\right [ \bmod 1.\ ] ] if @xmath95 is irrational , this follows immediately from theorem [ thm : kronecker ] .
now let @xmath95 be a rational number with denominator @xmath342 .
the desired result follows from @xmath343 and @xmath344 applying proposition [ pr : eta real ] with @xmath345 , where @xmath346 as usual , settles the case of theorem [ thm : main2 ] where @xmath73 is a negative real number .
clearly , the same argument applies if @xmath72 is negative real and @xmath73 is complex .
finally let us see what happens if the pair of denominators of @xmath347 in theorem [ thm : main2 ] is @xmath97 , @xmath98 or @xmath99 . according to theorem [ thm : genrat ]
, our argument with lattice points in squares fails for these values .
fortunately , it is straightforward to show directly that the purely periodic sequences @xmath88 arising from these values oscillate .
once again we can appeal to the symmetries noted in lemma [ le : symm ] .
indeed , swapping @xmath348 and @xmath349 does no harm , and the sign of @xmath350 can be absorbed in @xmath351 and @xmath352 .
it turns out that for each of the three pairs of denominators it suffices to consider one pair of numerators , namely @xmath353 for the denominators @xmath98 and @xmath99 , and @xmath354 for the denominators @xmath97 . in the following proposition ,
the cases ( i ) , ( ii ) and ( iii ) correspond to the pairs of denominators @xmath98 , @xmath99 and @xmath97 , respectively .
let @xmath351 , @xmath334 be nonzero real numbers and @xmath352 , @xmath333 be real numbers .
then the sequence defined by @xmath355 has a positive and a negative entry for each of the following values of @xmath356 .
\(i ) & _ 1= , _ 2= & + ( ii ) & _ 1= , _ 2= & + ( iii ) & _ 1= , _ 2= & it suffices to consider @xmath357 .
we set @xmath358 and @xmath359 for @xmath360 , suppose @xmath361 for all @xmath362 and derive a contradiction .
\(i ) from @xmath363 we deduce @xmath364 , so @xmath365 or @xmath366 . if @xmath366 , then @xmath367 implies @xmath368 , which contradicts @xmath369 .
if , on the other hand , @xmath365 , then @xmath368 follows from @xmath370 \(ii ) now @xmath371 shows @xmath372 , again contradicting @xmath369 .
\(iii ) since @xmath373 we have @xmath374
. then we obtain @xmath375 hence @xmath376 and @xmath377
. therefore @xmath366 or @xmath365 .
if @xmath366 , then the values @xmath378 and @xmath379 have opposite signs . it remains to consider the case @xmath365 .
for each of the four possible values of @xmath380 the inequalities @xmath381 form a linear system of inequalities in @xmath334 .
it is easy to check that none of these four systems are solvable .
the proof of theorem [ thm : main2 ] is complete , hence theorem [ thm : main ] is established .
if one of the dominating characteristic roots @xmath382 is real positive , conjecture [ cj : equ abs ] is not applicable . consider the sequence defined by @xmath383 where @xmath384 , @xmath385 are nonzero real numbers and @xmath386 are real numbers . here and throughout this section
we assume that the coefficient of the real positive root is positive ( and thus w.l.o.g
. equals one ) .
analogous considerations apply for a negative coefficient .
the behaviour of @xmath0 depends on how @xmath387 compares to @xmath388-w , w],\ ] ] where @xmath389 .
the sequence @xmath0 is positive for large @xmath1 if @xmath390 ( in particular , if @xmath391 ) , and it oscillates if @xmath392 .
if @xmath393 , the behaviour of @xmath0 depends on how well @xmath394 approximates @xmath85 and possibly on the @xmath395 term .
the preceding discussion gives a handy criterion only for @xmath391 , which was already obtained by burke and webb @xcite .
for @xmath396 we confine ourselves to showing how @xmath0 behaves for almost all values of the parameters @xmath350 , @xmath80 and @xmath89 . [
le : almost xi ] let @xmath397 and let @xmath398 be a sequence of positive real numbers such that @xmath399 converges .
then the set of inequalities @xmath400 has infinitely many solutions @xmath121 for almost no @xmath401 .
see cassels ( * ? ? ?
* lemma vii.2.1 ) . in order to apply the following theorem we require the dominating characteristic roots to be simple
this assumption makes the remainder term @xmath402 go to zero exponentially . parts ( i ) and ( iii ) hold for multiple roots as well , since they only require @xmath403 .
our proof of part ( ii ) , however , breaks down for @xmath58 in case of a multiple root , because then we can ensure only @xmath404 and this leads to a divergent series in lemma [ le : almost xi ] .
[ thm : almost no theta ] let @xmath385 be nonzero real numbers with @xmath405 , @xmath386 be real numbers and @xmath406 be a real sequence with @xmath407 for some @xmath408 .
* if @xmath391 , then for all @xmath409 the sequence @xmath0 defined by @xmath410 is positive for large @xmath1 . *
if @xmath411 , then for almost all @xmath409 the sequence @xmath0 is positive for large @xmath1 . *
if @xmath412 , then @xmath0 oscillates for almost all @xmath409 .
\(i ) is clear .
( iii ) follows from the @xmath259-dimensional version of theorem [ thm : kronecker ] ( see section [ se : conclusion ] ) , because @xmath413 are linearly independent over the rationals for almost all @xmath414 .
we proceed to prove ( ii ) .
suppose @xmath415 for all @xmath1 in an infinite set @xmath416 .
to make @xmath3 non - positive , @xmath104 has to be very close to @xmath85 if @xmath417 and very close to @xmath387 if @xmath418 . to be precise
, we must have @xmath419 for @xmath420 with @xmath421 by taylor expansion , we obtain @xmath422 removing finitely many elements from @xmath239 if necessary , we thus have @xmath423 where @xmath424 . since @xmath415 for @xmath425 , this implies @xmath426 hence for @xmath427 @xmath428 according to lemma [ le : almost xi ] this holds for almost no @xmath414 .
finer questions may be asked about the sets of measure zero alluded to in theorem [ thm : almost no theta ] . as for part ( ii ) of the theorem , we note that there are @xmath386 , @xmath402 and infinitely many @xmath414 such that @xmath0 oscillates for all nonzero @xmath385 with @xmath411 . to see this , define @xmath429
let @xmath430 be arbitrary and @xmath431 for some @xmath408 .
then @xmath432 for odd @xmath1 , and @xmath433 if @xmath1 is two times a common multiple of the denominators of @xmath384 .
the preceding example is a special case of the following proposition , which completely describes the behaviour of @xmath0 under the assumptions of part ( ii ) of theorem [ thm : almost no theta ] and the additional constraint @xmath434 for @xmath427 .
[ pr : theta rat ] let @xmath384 be real numbers such that @xmath435 is a rational number @xmath436 for @xmath427 , let @xmath386 be real numbers , let @xmath385 be nonzero real numbers with @xmath437 and define @xmath438 * if there is a @xmath439 such that @xmath440 is irrational , then @xmath0 is positive for large @xmath1 .
* suppose that @xmath441 is a rational number @xmath442 for @xmath427 .
if for all @xmath443 @xmath444 with @xmath445 then there are infinitely many @xmath1 with @xmath446 , where @xmath447 and the behaviour of @xmath0 depends in an obvious way on the sign of the @xmath395 term for these @xmath1 . if there are @xmath439 , @xmath448 such that does not hold , then @xmath0 is positive for large @xmath1 .
the purely periodic sequence @xmath88 satisfies @xmath361 for all @xmath362 .
if none of its finitely many values are zero , then @xmath0 is positive for large @xmath1 .
we have @xmath446 if and only if @xmath104 equals @xmath387 for the @xmath439 s with @xmath418 and @xmath85 for the @xmath439 s with @xmath417 , i.e. @xmath449 which is equivalent to @xmath450 clearly , this can not hold if one of the @xmath440 is irrational . under the assumption of part ( ii )
, we are lead to the system of congruences @xmath451 now the result follows from theorem [ thm : gcrt ] .
in order to extend our approach to conjecture [ cj : equ abs ] to @xmath259 dominating characteristic roots , we need to show that infinitely many @xmath452 lie in any given @xmath259-dimensional hypercube ( modulo one ) with side length @xmath453 .
theorem [ thm : kronecker ] generalizes in the following way ( * ? ? ?
* theorem iii.5.iv ) : the points @xmath120 lie dense in the set of all @xmath454 that satisfy @xmath455 for all integer vectors @xmath456 with @xmath457 . again
the case of rational @xmath458 with equal denominators @xmath459 will be the crux of the proof .
this case seems to become more and more difficult for fixed denominator @xmath179 as @xmath259 increases , since the set @xmath460 has @xmath179 elements for all @xmath259 , whereas the volume of the hypercube is @xmath461 .
theorem [ thm : mink2 ] is certainly a valuable tool .
hadwiger @xcite has extended bender s two - dimensional result ( lemma [ le : bender ] ) that we used in the proof of proposition [ pr : g ] to arbitrary dimension @xmath259 .
a significant extension of lemma [ le : one short vector ] is still needed .
anyway it is conceivable that the exceptional rational values @xmath462 for which the hypercube might contain no point of become unmanageable as @xmath259 increases ( cf .
section [ se : compl ] ) .
our results on a positive real characteristic root leave ample room for refinement . for instance
, one could try to extend part ( ii ) of theorem [ thm : almost no theta ] to the case of a multiple root or to continue the discussion begun in proposition [ pr : theta rat ] by relaxing the requirement that all @xmath463 be rational .
the skolem mahler lech theorem @xcite describes the structure of the zero set @xmath464 of a recurrence sequence .
it is the union of a finite set and finitely many arithmetic progressions .
there might be an analogue of this result for the set @xmath465 .
finally , we have excluded algorithmics so far .
we do not know whether the positivity of recurrence sequences is a decidable problem .
proving conjecture [ cj : equ abs ] and giving an effective criterion instead of the metric theorem [ thm : almost no theta ] would lead to a decision procedure . | we consider the following question : which real sequences @xmath0 that satisfy a linear recurrence with constant coefficients are positive for sufficiently large @xmath1 ? we show that the answer is negative for both @xmath0 and @xmath2 , if the dominating characteristic roots in the representation of @xmath3 as a generalized power sum comprise either two pairs of conjugate complex roots or one negative real root and one pair of conjugate complex roots .
the proof uses results from diophantine approximation and the geometry of numbers .
concerning the case of a real positive dominating root we show what the answer to the question of positivity is for almost all values of the other dominating roots , provided that all dominating roots are simple . _
keywords : _ recurrence sequence , inequality , lattice points . _
2000 mathematics subject classifications : _ primary : 11b37 ; secondary : 11p21 . |
the unification of the standard model ( sm ) gauge groups into a larger group , like in @xmath2 grand unified theories ( guts ) @xcite , is an attractive possibility of a new physics beyond the sm .
one of the important check points of guts is the gauge coupling unification which predicts that the gauge coupling constants of the sm become equal at the unification scale up to threshold corrections .
it is well known that , in the sm , there is no strong indication of the unification of the gauge coupling constants . in supersymmetric ( susy ) models ,
on the other hand , the situation changes because of the existence of the superparticles as well as up- and down - type higgses .
in particular , with the renormalization group equations ( rges ) of the minimal susy sm ( mssm ) , three gauge coupling constants more - or - less meet at the gut scale @xmath4 if the mass scale of the superparticles is @xmath5 @xcite . in simple gut models based on @xmath2 ,
quarks and leptons are embedded into full multiplets of @xmath2 . in particular , the right - handed down - type quarks and the left - handed lepton doublets are embedded into the anti - fundamental representations of @xmath2 , resulting in the unification of the down - type and charged lepton yukawa coupling constants . in particular , the unification of the yukawa coupling constants of @xmath0-quark and @xmath1-lepton is an interesting check point of ( susy ) guts .
indeed , the @xmath0-@xmath1 unification based on susy guts has been extensively studied in many literatures @xcite .
the renormalization group behaviors of the coupling constants are sensitive to the existence of new particles .
if full multiplets of @xmath2 are added at a single scale , the unification of the gauge coupling constants is unaffected ( at least at the one - loop level ) , although the values of the gauge coupling constants depend on the particle content .
contrary to the gauge coupling unification , the unification of the @xmath0 and @xmath1 yukawa coupling constants is expected to be significantly affected by new particles , because the renormalization group runnings of yukawa coupling constants are strongly dependent on the behaviors of the gauge coupling constants .
importantly , there are various candidates of such new particles , like new fermions ( as well as their superpartners ) to realize peccei - quinn symmetry @xcite , extra chiral superfields in gauge mediation models @xcite , and so on . in addition
, existence of extra matters at the mass scale of the superparticles is required if there exists a non - anomalous discrete @xmath6-symmetry @xcite .
thus , their effects on the renormalization group runnings of the @xmath0 and @xmath1 yukawa coupling constants are of great interest in particular from the point of view of the yukawa unification based on susy guts . in this paper , we study the @xmath0-@xmath1 unification in susy models with extra matters which have gauge quantum numbers under the sm gauge group .
we will see that the existence of the extra matters may significantly affect the renormalization group running of the @xmath0 and @xmath1 yukawa coupling constants , and hence modify the @xmath0-@xmath1 unification . as we will discuss
, the ratio of the @xmath0 to @xmath1 yukawa coupling constants may become very close to @xmath7 if the extra matters have yukawa couplings with mssm particles even though in a large fraction of the parameter space of the mssm , the yukawa coupling constant of @xmath0 becomes sizably smaller than that of @xmath1 at the gut scale .
we will also see that the ratio of the @xmath0 to @xmath1 yukawa coupling constants at the gut scale becomes smaller in models with extra matters if they do not have yukawa interactions .
.notations of chiral superfields used throughout this paper .
each column denotes , from left to right , the classification between mssm particles and extra matters , the notation of the @xmath2 multiplet , the notation of the multiplet of the sm gauge group @xmath8 , the representation under the unified gauge group @xmath2 , and the representation under @xmath9 . [ cols="^,^,^,^,^",options="header " , ] in order to calculate the renormalization group running of coupling constants from the weak scale to the gut scale @xmath10 , we consider several effective theories ; particle contents of all the effective theories used in our analysis are summarized in tab.[tab : effectiveth ] . in the present analysis , there are three important mass scales , i.e. , the gaugino mass scale @xmath11 , the sfermion mass scale @xmath12 , and the extra fermion mass scale @xmath13 , at which the effective theory changes from one to another .
as we have mentioned , the scalar masses are comparable to or larger than the gaugino masses , and hence @xmath14 . in addition , the masses of the scalars in the extra matter sector have two contributions , i.e. , susy invariant mass parameters and soft susy breaking masses ( denoted as @xmath15 and @xmath16 , respectively ) ; the scalar masses in the extra matter sector is @xmath17 .
thus , we assume that the scalars in the extra matter sector decouple from the effective theory at the renormalization scale @xmath18 .
the relevant effective theory for each renormalization scale depends on @xmath11 , @xmath12 , and @xmath13 , as summarized in tab.[tab : effective ] . for all the effective theories mentioned above , we use two - loop rges ; for susy models , we use the susyno package @xcite , while the rges for non - susy theories are calculated based on @xcite . in addition , at each energy threshold ,
one effective theory is matched to another , taking into account one - loop threshold corrections to lagrangian parameters . in the following ,
we summarize the important effects . at the sfermion mass scale @xmath12 , two higgs doublets in @xmath19
are matched to the sm - like higgs as @xmath20 where @xmath21 is the ratio of the vacuum expectation value of @xmath22 to that of @xmath23 .
the boundary condition for the higgs quartic coupling @xmath24 at @xmath12 is @xmath25 where @xmath26 is the threshold correction due to heavy scalar particles ( in particular , stops ) .
in addition , the mass of the pseudo - scalar higgs , which is a component of the heavy higgs multiplet @xmath27 , is determined at this scale as @xmath28_{q = m_s},\end{aligned}\ ] ] where @xmath29 is determined from the following radiative electroweak symmetry breaking condition : @xmath30 the yukawa coupling constants @xmath31 ( with @xmath32 , @xmath0 , and @xmath1 ) are matched to @xmath33 at @xmath34 , using the mixing angle @xmath35 . in our analysis ,
the threshold correction to the bottom yukawa coupling constant at @xmath36 is important .
the correction @xmath37 is defined by @xmath38 where @xmath39 and @xmath40 are the bottom quark yukawa coupling constants in the effective theory used just below and just above @xmath12 , respectively .
the most important contributions to @xmath37 come from the sbottom - gluino and stop - chargino diagrams @xcite ; at the leading order of the mass - insertion approximation , these contributions are given by @xmath41 \mu \tan\beta , \label{eq : deltab}\end{aligned}\ ] ] where @xmath42 and @xmath43 ( @xmath44 and @xmath45 ) are masses of lighter ( heavier ) stop and sbottom , respectively , and @xmath46 ( in our numerical analysis , we use the full one - loop expression of @xmath37 . ) the important point is that @xmath37 is approximately proportional to @xmath47 , resulting in the large correction to the bottom yukawa coupling constant in the models with heavy higgsinos or those with large @xmath21 .
we also include threshold corrections to the wino and bino masses at @xmath48 due to the higgs - higgsino loop diagram @xcite : @xmath49 where @xmath50 at @xmath51 and @xmath52 , we take into account one - loop threshold corrections to gauge coupling constants , gaugino masses , and scalar masses due to loop diagrams involving gauginos and extra matters .
then , at @xmath53 the sm - like higgs mass is evaluated as @xmath54 where @xmath55 is the vacuum expectation value of the sm - like higgs boson and @xmath56 is the threshold correction .
in this section , we show the results of our numerical study . in addition to the sm parameters
, the present model contains ten new parameters , @xmath57 , @xmath58 , @xmath59 , @xmath60 , @xmath61 , @xmath62 , @xmath63 , @xmath64 , @xmath65 , and @xmath66 , ignoring the yukawa and the trilinear couplings related to the extra matters . among them , @xmath63 and @xmath64 are determined at the sfermion mass scale @xmath12 to fix the vacuum expectation value of the sm - like higgs boson @xmath67 and @xmath57 .
we numerically solve rges from the weak scale to the gut scale .
our numerical calculation is based on the softsusy package @xcite , in which three - loop rges for the effective theory below the electroweak scale and two - loop rges for the mssm are implemented .
we have implemented into softsusy package two - loop rges for the other effective theories listed in tab .
[ tab : effective ] , i.e. , sm , @xmath68 , @xmath69sm , @xmath70 , and @xmath71 .
in addition , one - loop threshold corrections due to the diagrams with susy particles or extra matters in the loop are included at relevant thresholds . in our numerical calculation , @xmath72 is taken to be the geometric mean of the stop masses , while we take @xmath73 .
@xmath13 is set to the mass of the bottom - like extra fermion mass @xmath74 for models with @xmath75 , and is set to the geometric mean of top - like extra fermion masses for models with @xmath76 .
the gauge and yukawa coupling constants are determined based on @xcite .
in particular , we use the bottom quark mass of @xmath77 , the top quark mass of @xmath78 , and @xmath79 ( with @xmath80 ) .
let us now study the effects of extra matters on the @xmath0-@xmath1 unification in susy gut .
we first consider the case where the extra matters interact with the mssm particles only through gauge interactions .
because the boundary conditions for the yukawa coupling constants are fixed by using the fermion masses , @xmath81 and @xmath82 may differ in the present analysis . to quantify the difference
, we define @xmath83 we calculate @xmath84 as a function of model parameters , and study how it is affected by extra matters . if there is no source of the gut scale threshold corrections other than the splitting of the masses of gut scale particles , then @xmath85 .
thus , if @xmath84 is ( much ) larger than @xmath86 , it indicates a sizable threshold correction at the gut scale and/or a non - trivial flavor physics at the gut scale or below . if the yukawa interactions of the extra matters are negligible , the main effect of extra matters on the yukawa unification is through the enhancement of the gauge coupling constants at high energy scale . with extra matters ,
the coupling constants of @xmath87 become larger .
this can be understood by examining the rges of the gauge coupling constants ; in susy models , they are given by @xmath88 + \cdots,\end{aligned}\ ] ] where @xmath89 , and @xmath90 in the above equation denotes higher order effects .
one can see that , with non - vanishing @xmath91 or @xmath92 , the beta - function coefficients become larger , resulting in the enhancement of the gauge coupling constants at higher scale .
in particular , the enhancement of @xmath93 is the most important because of the largeness of the coupling constant @xmath93 itself .
the enhanced gauge coupling constants affect the renormalization group running of @xmath40 and @xmath94 , whose rges are given by @xmath95 where the mixings between different generations are neglected . with the low - scale values of the yukawa coupling constants being fixed to realize the observed fermion masses ,
the above equations indicate that the yukawa coupling constants at @xmath10 is more suppressed as the gauge coupling constants become larger . due to this effect
, @xmath40 is more suppressed than @xmath94 because @xmath93 only affects the running of @xmath40 . as a function of the mass scale of the extra matters for models with low @xmath21 ( left ) and high
@xmath96 ( right ) , taking @xmath97 ( red ) , @xmath98 ( green ) , and @xmath99 ( blue ) .
the horizontal axis denotes the value of @xmath65 for red and green lines and that of @xmath66 for blue lines .
we consider both signs of the susy invariant higgs mass : @xmath100 ( dotted ) and @xmath101 ( solid ) .
the boundary conditions used in the left figure are @xmath102 , @xmath103 , and @xmath104 .
@xmath96 is determined so that @xmath105 , which results in @xmath106 .
the boundary conditions for the right figure are @xmath107 , @xmath108 , and @xmath109 .
@xmath62 is determined so that @xmath105 , which results in @xmath110 . ] as a function of the mass scale of the extra matters for models with low @xmath21 ( left ) and high @xmath96 ( right ) , taking @xmath97 ( red ) , @xmath98 ( green ) , and @xmath99 ( blue ) .
the horizontal axis denotes the value of @xmath65 for red and green lines and that of @xmath66 for blue lines .
we consider both signs of the susy invariant higgs mass : @xmath100 ( dotted ) and @xmath101 ( solid ) .
the boundary conditions used in the left figure are @xmath102 , @xmath103 , and @xmath104 .
@xmath96 is determined so that @xmath105 , which results in @xmath106 .
the boundary conditions for the right figure are @xmath107 , @xmath108 , and @xmath109 .
@xmath62 is determined so that @xmath105 , which results in @xmath110 . ] in fig .
[ mexvsr ] , in order to investigate how these effects affect @xmath84 , we show @xmath84 as a function of the mass scale of extra matters .
the red , green , and blue lines correspond to the models with @xmath97 , @xmath98 , and @xmath99 , respectively .
( thus , the horizontal axis corresponds to @xmath65 for red and green lines and @xmath66 for blue lines . )
the dotted and solid lines are for models with @xmath100 and @xmath101 , respectively . for the left figure , we take msugra - like boundary conditions , @xmath111 , @xmath112 , and @xmath104 . here , @xmath21 is determined so that the sm - like higgs mass is given by the observed value @xmath105 ; then , it takes values in the range @xmath106 .
the right figure shows the results for the model with gaugino mediation boundary conditions @xcite , taking @xmath113 , @xmath114 , and @xmath104 . in this case , @xmath62 is tuned so that @xmath115 is equal to the observed higgs mass , which gives @xmath116 . as can be easily understood , the effects of extra matters on the runnings of @xmath40 and @xmath94 are more enhanced as the masses of the extra matters become smaller .
we can see that @xmath84 is suppressed by @xmath117 when the mass scale of the extra matters is at around the tev scale , while @xmath84 approaches to the mssm value when the mass scale @xmath13 becomes close to the gut scale . in the mssm , it is often the case that @xmath84 is significantly smaller than 1 in particular when @xmath21 is small .
as we have seen , the effects of extra matters make @xmath84 smaller if extra matters interact with mssm particles only through gauge interactions .
we also note here that the deviation between a solid line and the corresponding dotted line approximately shows ( twice ) the size of the threshold correction @xmath37 , since the sign of @xmath37 is determined by that of @xmath63 .
the size of @xmath37 is also affected by the change of @xmath13 because the mass spectrum of the mssm particles also depends on @xmath13 . with the model parameters for the solid lines in fig .
[ mexvsr ] ( right ) , for example , @xmath118 is enhanced as @xmath13 becomes smaller .
however , the suppression of @xmath84 due to the enhancement of @xmath93 at higher scale is more significant ; consequently , @xmath84 becomes smaller as @xmath13 decreases as shown in fig .
[ mexvsr ] .
so far , we have considered the case where the yukawa interactions of the extra matters are negligibly small .
however , the extra matters may couple to mssm chiral multiplets via yukawa couplings . with such new interaction , the renormalization group runnings of the coupling and mass parameters
may be changed , affecting the unification of yukawa coupling constants . in this subsection , we show that this is indeed the case . among several possibilities ,
we introduce the yukawa interaction for the extra matter in @xmath119 representation of @xmath2 .
we will see that , in such a model , the extra matters may help to make the @xmath0-@xmath1 unification successful . here
, we consider the case with @xmath120 , and study the effect of the yukawa interactions of extra matters .
we concentrate on the case where the extra matter scale is comparable to or higher than @xmath72 so that all the extra matters ( i.e. , fermions and scalars ) simultaneously decouple from the effective theory at a single scale @xmath13 .
for the study of such a case , it is instructive to use the fact that the yukawa interaction above the gut scale can be written in the following form : @xmath121 where @xmath122 s are coupling constants . here ,
@xmath123 and @xmath124 are chiral multiplets in the @xmath119 representation of @xmath2 , and are given by linear combinations of @xmath125 and @xmath126 .
notice that , in this basis , only @xmath123 couples to @xmath127 though the yukawa interaction . in addition , @xmath128 and @xmath129 are chiral multiplets containing up- and down - type higgses , respectively . in order to make our point clearer , we take @xmath130 in the following analysis . with such an assumption ,
the yukawa interaction below the gut scale is given by @xmath131 where @xmath132 , @xmath133 , and @xmath134 are chiral superfields embedded into @xmath123 .
( chiral multiplets embedded into @xmath135 are denoted as @xmath136 , @xmath137 , and @xmath138 . ) denoting the susy invariant mass terms for the extra matters as @xmath139 we obtain @xmath140 where @xmath141 with @xmath142 .
similar relations hold for @xmath143 and @xmath144 , with the mixing angles @xmath145 . at the mass scale of the extra matters ,
mssm yukawa coupling constants are given by @xmath146 in the present set up , the yukawa structure is like that of the mssm as eq .
is obtained from the yukawa interaction of the mssm by replacing @xmath147 and @xmath148 .
numerically , however , such a replacement may give significant effects on the yukawa unification .
this is because , as shown in eq .
, @xmath149 can be significantly larger than @xmath150 because @xmath151 .
such an enhancement of the coupling constant may have the following consequences : 1 .
@xmath152 is enhanced through the renormalization group effect while @xmath153 is not ( see eqs . and ) .
@xmath154 is suppressed through the renormalization group effect .
this can be understood from the rge of @xmath155 , which is given by @xmath156 where only the @xmath149-dependence of the beta - function at the one - loop level is shown in the above equation .
this may result in the enhancement of @xmath157 and @xmath118 . in order to study the effects of the yukawa couplings of the extra matters on the @xmath0-@xmath1 unification ,
we solve the rges numerically , taking into account the effects of extra yukawa couplings .
we assume that the threshold correction to the susy invariant masses of extra matters are negligible , and that they are unified at the gut scale ; we parameterize their boundary conditions as @xmath158 here , @xmath159 is approximately equal to @xmath160 ( with @xmath161 ) , although they slightly differ because of the renormalization group effects from the gut scale to the extra matter scale .
( in our numerical analysis , we have taken into account the effects of the renormalization group running of susy invariant mass parameters . )
in addition , for @xmath162 , the scalars in the extra matter sector have trilinear interactions .
we assume that , at the gut scale , the trilinear couplings are proportional to the corresponding yukawa coupling constants , and that the trilinear interactions above the extra matter scale are given by @xmath163 with the boundary conditions @xmath164 and @xmath165 .
furthermore , the susy breaking mass parameters above the extra matter scale can be written as @xmath166 notice that , with the present choice of parameters , there is no mixing term between @xmath167 and @xmath168 .
soft susy breaking parameters defined above and below the extra matter scale are matched at @xmath169 using eq .
( as well as the mixing angles for @xmath143 and @xmath144 ) .
we show examples of the running of yukawa coupling constants in fig.[fig : yukawarunning ] , which demonstrates the possibility to make the unification of the yukawa coupling constants successful due to the effects of the yukawa interactions of extra matters .
solid lines show the results for the present model while dotted lines denote the result for the mssm as a reference . here , we take @xmath170 , @xmath171 , @xmath104 , @xmath172 , and @xmath173 .
@xmath174 is used to adjust the sm - like higgs mass in each model to be the observed value , which gives @xmath175 for the model with @xmath176 and @xmath177 for the mssm .
the vertical dotted lines denote the matching scales in the model with extra matters : from left to right , they correspond to @xmath178 , and @xmath10 , respectively .
the large `` jumps '' of solid lines in fig .
[ fig : yukawarunning ] at @xmath36 are due to the threshold corrections , while those at @xmath52 are mainly due to the mixing effect represented in eq .
( [ ybprime ] ) and eq.([ytauprime ] ) since we plot @xmath179 and @xmath180 instead of @xmath181 and @xmath182 by solid lines in the range @xmath183 .
we can see from the figure that the enhancement of @xmath118 significantly modifies the prediction for yukawa unification .
together with the change in the running of @xmath40 as we discussed before , the mixing @xmath184 enlarges the prediction for @xmath185 , which is defined as @xmath186 it becomes almost @xmath7 in the present choice of parameters , while @xmath187 in the case of the mssm with the same choice of gut scale boundary conditions for susy breaking parameters . and @xmath188 ( left ) and those of @xmath39 , @xmath189 , @xmath190 and @xmath191 ( right ) in the model with @xmath120 ( solid lines ) and in the mssm ( dotted lines ) . here , we take @xmath170 , @xmath192 , @xmath104 , @xmath172 , and @xmath193 . here
, @xmath174 is tuned to adjust the sm - like higgs mass to be @xmath105 , which gives @xmath194 for the model with @xmath195 and @xmath196 for the mssm .
the vertical dotted lines denote the matching scales in the model with extra matters : @xmath53 , @xmath11 , @xmath12 , @xmath13 , and @xmath10 from left to right . for @xmath197 ( @xmath183 ) ,
the solid lines denote @xmath198 ( @xmath199 ) or @xmath200 ( @xmath201 ) with @xmath202 , or @xmath1 . ] and @xmath188 ( left ) and those of @xmath39 , @xmath189 , @xmath190 and @xmath191 ( right ) in the model with @xmath120 ( solid lines ) and in the mssm ( dotted lines ) . here , we take @xmath170 , @xmath192 , @xmath104 , @xmath172 , and @xmath193 . here , @xmath174 is tuned to adjust the sm - like higgs mass to be @xmath105 , which gives @xmath194 for the model with @xmath195 and @xmath196 for the mssm .
the vertical dotted lines denote the matching scales in the model with extra matters : @xmath53 , @xmath11 , @xmath12 , @xmath13 , and @xmath10 from left to right . for @xmath197 ( @xmath183 ) ,
the solid lines denote @xmath198 ( @xmath199 ) or @xmath200 ( @xmath201 ) with @xmath202 , or @xmath1 . ] in fig .
[ mixingvsr ] , we show the @xmath203 dependence of @xmath185 , taking @xmath170 , @xmath204 , and @xmath104 .
@xmath205 is tuned for each value of @xmath203 to adjust the sm - like higgs mass to be the observed value ; as a result , @xmath205 takes values in the range between @xmath206 and @xmath207 .
the red line shows the result for the case with @xmath208 and the green one shows that with @xmath209 .
we can understand from the figure that @xmath185 is enhanced with larger @xmath203 .
in the present choice of parameters with @xmath210 , @xmath211 is possible . on the other hand , for @xmath209 ,
the enhancement is not so important since in this case , the suppression of @xmath40 due to the enhancement of the gauge coupling constants is so large ( see fig.[mexvsr ] ) that it cancels the advantage of the mixing effect .
notice that the lines in fig .
[ mixingvsr ] terminate at some value of @xmath203 .
this is because , for larger value of @xmath203 , @xmath205 becomes larger in order to fix the sm - like higgs mass , which in turn may make the right - handed sbottom mass squared or the left - handed slepton mass squared being negative at @xmath34 , causing the tachyonic sfermion problem .
dependence of @xmath185 defined in eq .
( [ rbtaunew ] ) .
the parameters used are the same as fig .
[ fig : yukawarunning ] .
@xmath205 is again used to adjust the sm - like higgs mass to be @xmath105 at each value of @xmath203 .
the red and the green lines denote the model with @xmath208 and @xmath212 , respectively.,width=340 ]
in this letter , we have studied the @xmath0-@xmath1 unification in susy @xmath2 models with extra matters .
we have assumed that the extra matters are embedded into full @xmath2 multiplets .
we have seen that the extra matters may significantly affect the @xmath0-@xmath1 unification in particular when the mass scale of the extra matters is much lower than the gut scale .
we have first considered the case where the extra matters interact with the mssm particles only through gauge interaction . in such a case ,
the ratio of @xmath40 and @xmath94 at the gut scale , which we called @xmath84 , becomes suppressed as the mass scale of the extra matters becomes smaller .
this is because , with the extra matters , the @xmath3 gauge coupling constant is enhanced at higher scale , resulting in the suppression of the bottom yukawa coupling constant at the gut scale .
the suppression of @xmath84 has been found to be @xmath117 .
we have also studied the effects of the yukawa couplings of the extra matters with mssm particles . in the case
we have studied , the yukawa couplings above the mass scale of the extra matters are effectively enhanced , resulting in the change of the ratio of the ( effective ) @xmath0 and @xmath1 yukawa coupling constants . in particular , we have shown that a simple yukawa unification ( i.e. , @xmath211 ) can be realized via the effects of extra matters with yukawa interaction even though @xmath84 is significantly smaller than @xmath7 for the case without extra matters with the same gut scale boundary conditions .
_ acknowledgements _ : this work was supported by the grant - in - aid for scientific research c ( no.26400239 ) , and innovative areas ( no.16h06490 ) . the work of s.c . was also supported in part by the program for leading graduate schools , mext , japan .
h. georgi and s. l. glashow , phys .
* 32 * ( 1974 ) 438 .
h. georgi , h. r. quinn and s. weinberg , phys .
lett . * 33 * ( 1974 ) 451 .
a. j. buras , j. r. ellis , m. k. gaillard and d. v. nanopoulos , nucl .
b * 135 * ( 1978 ) 66 .
s. dimopoulos , s. raby and f. wilczek , phys .
d * 24 * ( 1981 ) 1681 .
m. b. einhorn and d. r. t. jones , nucl .
b * 196 * ( 1982 ) 475 .
w. j. marciano and g. senjanovic , phys .
d * 25 * ( 1982 ) 3092 .
u. amaldi , w. de boer and h. furstenau , phys .
b * 260 * ( 1991 ) 447 .
j. r. ellis , s. kelley and d. v. nanopoulos , phys .
b * 260 * ( 1991 ) 131 .
p. langacker and m. x. luo , phys .
d * 44 * ( 1991 ) 817 .
k. tobe and j. d. wells , nucl .
b * 663 * ( 2003 ) 123 [ hep - ph/0301015 ] .
h. baer , i. gogoladze , a. mustafayev , s. raza and q. shafi , jhep * 1203 * ( 2012 ) 047 [ arxiv:1201.4412 [ hep - ph ] ] .
h. baer , s. raza and q. shafi , phys .
b * 712 * ( 2012 ) 250 [ arxiv:1201.5668 [ hep - ph ] ] .
m. badziak , mod .
lett . a * 27 * ( 2012 ) 1230020 [ arxiv:1205.6232 [ hep - ph ] ] .
a. s. joshipura and k. m. patel , phys .
d * 86 * ( 2012 ) 035019 [ arxiv:1206.3910 [ hep - ph ] ] .
g. elor , l. j. hall , d. pinner and j. t. ruderman , jhep * 1210 * ( 2012 ) 111 [ arxiv:1206.5301 [ hep - ph ] ] .
h. baer , s. kraml and s. kulkarni , jhep * 1212 * ( 2012 ) 066 [ arxiv:1208.3039 [ hep - ph ] ] .
a. anandakrishnan , s. raby and a. wingerter , phys .
d * 87 * ( 2013 ) 055005 [ arxiv:1212.0542 [ hep - ph ] ] .
m. a. ajaib , i. gogoladze , q. shafi and c. s. un , jhep * 1405 * ( 2014 ) 079 [ arxiv:1402.4918 [ hep - ph ] ] .
d. j. miller and a. p. morais , jhep * 1412 * ( 2014 ) 132 [ arxiv:1408.3013 [ hep - ph ] ]
. i. gogoladze , a. mustafayev , q. shafi and c. s. un , phys .
d * 91 * ( 2015 ) 096005 [ arxiv:1501.07290 [ hep - ph ] ] .
s. chigusa and t. moroi , phys .
d * 94 * ( 2016 ) no.3 , 035016 [ arxiv:1604.02156 [ hep - ph ] ] .
r. d. peccei and h. r. quinn , phys .
* 38 * ( 1977 ) 1440 .
m. dine and a. e. nelson , phys .
d * 48 * ( 1993 ) 1277 [ hep - ph/9303230 ] .
m. dine , a. e. nelson and y. shirman , phys .
d * 51 * ( 1995 ) 1362 [ hep - ph/9408384 ] .
m. dine , a. e. nelson , y. nir and y. shirman , phys .
d * 53 * ( 1996 ) 2658 [ hep - ph/9507378 ] .
k. kurosawa , n. maru and t. yanagida , phys .
b * 512 * ( 2001 ) 203 [ hep - ph/0105136 ] .
m. asano , t. moroi , r. sato and t. t. yanagida , phys .
b * 705 * ( 2011 ) 337 [ arxiv:1108.2402 [ hep - ph ] ] .
m. ibe , t. moroi and t. t. yanagida , phys .
b * 644 * ( 2007 ) 355 [ hep - ph/0610277 ] .
m. ibe and t. t. yanagida , phys .
b * 709 * ( 2012 ) 374 [ arxiv:1112.2462 [ hep - ph ] ] .
n. arkani - hamed , a. gupta , d. e. kaplan , n. weiner and t. zorawski , arxiv:1212.6971 [ hep - ph ] .
r. m. fonseca , comput .
commun .
* 183 * ( 2012 ) 2298 [ arxiv:1106.5016 [ hep - ph ] ] .
m. e. machacek and m. t. vaughn , nucl .
b * 222 * ( 1983 ) 83 .
m. e. machacek and m. t. vaughn , nucl .
b * 236 * ( 1984 ) 221 .
m. e. machacek and m. t. vaughn , nucl
. phys .
b * 249 * ( 1985 ) 70 .
l. j. hall , r. rattazzi and u. sarid , phys .
d * 50 * ( 1994 ) 7048 [ hep - ph/9306309 , hep - ph/9306309 ] .
m. carena , m. olechowski , s. pokorski and c. e. m. wagner , nucl .
b * 426 * ( 1994 ) 269 [ hep - ph/9402253 ] .
t. blazek , s. raby and s. pokorski , phys .
d * 52 * ( 1995 ) 4151 [ hep - ph/9504364 ] .
g. f. giudice , m. a. luty , h. murayama and r. rattazzi , jhep * 9812 * ( 1998 ) 027 [ hep - ph/9810442 ] .
b. c. allanach , comput .
commun .
* 143 * ( 2002 ) 305 [ hep - ph/0104145 ] .
k. a. olive _ et al .
_ [ particle data group collaboration ] , chin . phys .
c * 38 * ( 2014 ) 090001 .
d. e. kaplan , g. d. kribs and m. schmaltz , phys .
d * 62 * ( 2000 ) 035010 [ hep - ph/9911293 ] .
z. chacko , m. a. luty , a. e. nelson and e. ponton , jhep * 0001 * ( 2000 ) 003 [ hep - ph/9911323 ] . | we consider @xmath0-@xmath1 unification in supersymmetric @xmath2 grand unified theories ( guts ) with extra matters .
the renormalization group runnings of @xmath0 and @xmath1 yukawa coupling constants may be significantly affected by the existence of extra matters .
if the extra matters interact with the standard model particles ( and their superpartners ) only through gauge interaction , the ratio of the @xmath0 to @xmath1 yukawa coupling constants at the gut scale becomes suppressed compared to the case without extra matters . this is mainly due to the change of the renormalization group running of the @xmath3 gauge coupling constant .
if the extra matters have yukawa couplings , on the contrary , the ( effective ) @xmath0 yukawa coupling at the gut scale can be enhanced due to the new yukawa interaction .
such an effect may improve the @xmath0-@xmath1 unification in supersymmetric guts .
ut-17 - 05 + february , 2017 + .75 in * * bottom - tau unification in supersymmetric @xmath2 models + with extra matters .75 in so chigusa and takeo moroi 0.25 in 0.25 in _ department of physics , university of tokyo , tokyo 113 - 0033 , japan _ .5 in |
the objective of this article is to study the growth rate of a version of legendrian contact homology and its implications to the dynamics of reeb flows on contact 3-manifolds .
it fits in a larger program which aims at understanding the relationship between sft - invariants of a contact structure and global dynamical invariants of reeb flows . in this work
we focus our attention on one dynamical invariant , the topological entropy .
the topological entropy @xmath1 is a non - negative number that one associates to a dynamical system and which measures the complexity of the system .
positivity of the topological entropy for a dynamical system implies that the system has some type of exponential instability .
we first recall some basic definitions from contact geometry .
a 1-form @xmath2 on a @xmath3-dimensional manifold @xmath4 called a _ contact form _ if @xmath5 is a volume form on @xmath4 .
the hyperplane @xmath6 is called the _
contact structure_. for us a _ contact manifold _ will be a pair @xmath7 such that @xmath8 is the kernel of some contact form @xmath2 on @xmath4 ( these are usually called co - oriented contact manifolds in the literature ) .
when @xmath2 satisfies @xmath9 , we will say that @xmath2 is a contact form on @xmath7 . on any contact manifold there
always exist infinitely many different contact forms . given a contact form @xmath2 , its _ reeb vector field _ is the unique vector field @xmath10 satisfying @xmath11 and @xmath12 . the reeb flow @xmath13 of @xmath2 is the flow generated by the vector field @xmath10 .
we will refer to the periodic orbits of @xmath13 as _ reeb orbits _
of @xmath2 .
the action @xmath14 of a reeb orbit is defined by @xmath15 .
a contact form is called _ hypertight _ if its reeb flow has no contractible periodic orbits .
an isotropic submanifold of @xmath7 is a submanifold of @xmath4 whose tangent bundle is contained in @xmath8 .
when this submanifold is of maximal possible dimension , it is called a _ legendrian submanifold _ of @xmath7 .
it turns out that the maximal possible dimension of an isotropic submanifold is @xmath16 .
given a contact form @xmath17 and a pair of legendrian submanifolds @xmath18 , a _
reeb chord _ from @xmath19 to @xmath20 is a trajectory @xmath21 of the reeb flow of @xmath17 that starts in @xmath19 and ends in @xmath20 .
we define the action @xmath22 of a reeb chord @xmath21 as @xmath23 .
a reeb chord @xmath21 is said to be _ transverse _ if the intersection @xmath24 is transverse at the endpoint of @xmath21 .
we study the topological entropy of reeb flows from the point of view of contact topology .
more precisely , we search for conditions on the topology of a contact manifold @xmath0 that force * all * reeb flows on @xmath0 to have positive topological entropy .
the condition we impose is on the behaviour of a contact topological invariant called strip legendrian contact homology .
we show that if a contact manifold @xmath0 admits a contact form @xmath25 and a pair of legendrian knots @xmath18 for which the strip legendrian contact homology is well defined and has _ exponential homotopical growth _ , then all reeb flows on @xmath0 have positive topological entropy .
the inspiration for studying the topological entropy of reeb flows from this perspective comes from the study of geodesic flows .
it is an important result in the theory of geodesic flows that if the based loop space of a manifold @xmath26 has a rich homology , then for every riemannian metric @xmath27 on @xmath26 the geodesic flow of @xmath27 on the unit tangent bundle @xmath28 of @xmath26 has positive topological entropy ; see paternain s book @xcite .
geodesic flows on the unit tangent bundle are examples of reeb flows .
the reason for this is that there exists a contact structure @xmath29 on @xmath30 such that for every riemannian metric @xmath27 on @xmath26 , the geodesic flow @xmath31 of @xmath27 coincides with the reeb flow of a contact form @xmath32 on @xmath33 . in @xcite macarini and schlenk showed that if the based loop space of@xmath26 has rich homology then every reeb flow on @xmath33 has positive topological entropy .
their result can be seen as an extension of the result previously known for geodesic flows in @xmath30 to the larger class of reeb flows on @xmath34 .
the strategy to estimate the topological entropy used in @xcite can be sketched as follows : exponential growth of lagrangian floer homology of the cotangent fiber @xmath35 + @xmath36 + exponential volume growth of the unit tangent fiber @xmath37 for all reeb flows in @xmath38 + @xmath36 + positivity of the topological entropy for all reeb flows in @xmath38 . to obtain the first implication macarini and schlenk use the fact that the contact manifold @xmath38 has the structure of a legendrian fibration , and apply the geometric idea of @xcite to show that the number of trajectories connecting a legendrian fiber to another legendrian fiber can be used to obtain a volume growth estimate .
inspired by @xcite we present in this article techniques that can be used to estimate the topological entropy for reeb flows on contact 3-manifolds which are not unit cotangent bundles . because most contact @xmath39-manifolds are not legendrian fibrations , it is usually not possible to apply the scheme presented above .
however a sufficiently small neighbourhood of a given legendrian knot in a contact @xmath39-manifold can always be given the structure of a legendrian fibration .
this local structure of a legendrian fibration allows us to use a localised version of the geometric idea of @xcite , and to prove that the exponential growth of the strip legendrian contact homology of a pair of legendrian knots on a contact 3-manifold @xmath0,implies exponential growth of the length of these legendrian knots for any reeb flow on @xmath40 .
applying yomdin s theorem ( see @xcite ) we obtain the following result , which is the main structural result of this paper .
[ theorem1 ] let @xmath41 be a contact 3-manifold , and @xmath25 be a hypertight contact form on @xmath7 adapted to the pair of disjoint legendrian knots @xmath18 .
assume that the strip legendrian contact homology @xmath42 has exponential homotopical growth rate with exponential weight @xmath43 .
then , the reeb flow of any @xmath44 contact form @xmath2 on @xmath7 has positive topological entropy .
moreover , if we denote by @xmath45 the positive function such that @xmath46 we have @xmath47 theorem [ theorem1 ] gives a criterion implying that all reeb flows on a contact 3-manifold have positive topological entropy .
to explain the significance of this result we quote a result due to katok @xcite and lima and sarig @xcite which shows that positivity of topological entropy of the 3-dimensional flow of a vector field without singularities has deep implications for the dynamics of such flows .
if @xmath48 is a smooth flow on a closed oriented 3-manifold generated by a non - vanishing vector field , then @xmath48 has positive topological entropy if , and only if , there exists a smale `` horseshoe '' as a subsystem of the flow . as a consequence
, the number of hyperbolic periodic orbits of @xmath48 grows exponentially with respect to the period . here
a `` horseshoe '' is a compact invariant set where the dynamics is semi - conjugate to that of the suspension of a finite shift by a finite - to - one map . in particular , the number of hyperbolic periodic orbits on a `` horseshoe '' of a 3-dimensional flow @xmath48 grows exponentially with respect to the period .
the recent work of lima and sarig @xcite proves a stronger result : there exists an invariant set @xmath49 of @xmath48 of full topological entropy where the dynamics of the restriction @xmath50 is semi - conjugate to the suspension of a countable shift by a finite - to - one map .
it is worthy noting that their set @xmath51 is not necessarily a horseshoe , since it may not be compact .
nevertheless , since the topological entropy on countable shifts is the supremum of the topological entropy of its finite subshifts , @xcite implies that @xmath48 has horseshoes with topological entropy as close to @xmath52 as we wish .
it follows from the result of katok , and lima and sarig , that if a contact 3-manifold @xmath0 satisfies the hypothesis of theorem [ theorem1 ] , then for every reeb flow on @xmath0 the number of hyperbolic periodic orbits grows exponentially with the period .
an interesting feature of this result ( and also of a result in @xcite ) is that it gives a method of proving the existence of many reeb orbits of a reeb flow which is not based on the use of a contact topological invariant defined using reeb orbits .
the appearance of the reeb orbits is obtained as a consequence of a multiplicity result for the existence of * reeb chords*. the other main result of the paper is the existence of a large family of contact 3-manifolds which satisfy the hypothesis of theorem [ theorem1 ] .
[ theorem2 ] let @xmath53 be a closed oriented connected 3-manifold which can be cut along a nonempty family of incompressible tori into a family @xmath54 of irreducible manifolds with boundary such that the component @xmath55 satisfies : * @xmath55 is the mapping torus of a diffeomorphism @xmath56 with pseudo - anosov monodromy on a surface @xmath57 with non - empty boundary .
then @xmath53 can be given infinitely many different tight contact structures @xmath58 , such that : * there exist disjoint legendrian knots @xmath19 , @xmath20 on @xmath59 , and a contact form @xmath60 on @xmath59 adapted to the pair @xmath19 and @xmath20 , such that @xmath61 has exponential homotopical growth rate .
it follows that every reeb flow on @xmath59 has positive topological entropy .
the contact manifolds covered by this theorem fall are among the examples of tight contact manifolds constructed in @xcite and coincide with the ones studied in @xcite .
in particular , the theorem above that on these contact 3-manifolds every reeb flow has positive topological entropy , a result which was proved by the author using different methods in @xcite .
it is important to mention that the methods developed in this paper and in @xcite are independent .
we believe that they are complementary .
one advantage of the methods of @xcite is that they give estimates for the topological entropy for @xmath62 reeb flows while the method we use here is only valid for @xmath44 reeb flows . on the other hand ,
there are examples of high - dimensional contact manifolds which are simply connected and on which every reeb flow has positive topological entropy : the unit tangent bundles of simply connected manifolds presented in @xcite and the new examples constructed in @xcite are of this type . on simply connected manifolds there is no hope of applying the methods of @xcite but the volume growth technique for estimating topological entropy developed here can still be used as we show in @xcite . in general
, we expect that detecting exponential volume growth of legendrian submanifolds should be the most robust method to estimate topological entropy of reeb flows in high dimensional contact manifolds .
the techniques developed in this article are fundamental for several projects being developed by the author with collaborators .
we list some of these projects . * in a joint project with pedro a.s .
salomo @xcite we combine ideas of this article with those of @xcite to study the forcing of topological entropy in contact 3-manifolds .
more precisely , we establish a criterion to determine if a transverse knot @xmath49 on a contact 3-manifold @xmath0 forces topological entropy , in the sense that every reeb flow on @xmath0 which has @xmath49 as a reeb orbit has positive topological entropy .
this criterion is then used to obtain a contact topological generalisation of a theorem of denvir and mackay @xcite that says that if a riemannian metric of the two - dimensional torus that has a simple closed contractible geodesic , then its geodesic flow has positive topological entropy . * in a joint project with leonardo macarini @xcite we use contact connected sums to obtain examples of contact manifolds of dimension @xmath63 on which every reeb flow has positive topological entropy , and which are not unit tangent bundles or quotients of unit tangent bundles . in order to prove this
we show that a higher - dimensional version of theorem [ theorem1 ] is true provided that the legendrian submanifolds in question are spheres .
* in a joint work with vincent colin and ko honda we combine the techniques developed in this paper with those of @xcite to use openbook decompositions to prove positivity of topological entropy for all reeb flows on many contact 3-manifolds .
this work aims at establishing that on `` most '' tight contact 3-manifolds every reeb flow has positive topological entropy . * in our proof of theorem [ theorem1 ]
we have showed that if , on a contact 3-manifold @xmath7 , there exists a contact form @xmath25 and a pair of disjoint legendrian knots @xmath19 and @xmath20 satisfying the hypothesis of theorem [ theorem1 ] then for any contact form @xmath2 on @xmath7 the length of @xmath64 grows exponentially as @xmath65 .
it is natural to ask if , under the same hypothesis , this growth property is shared by all legendrian knots @xmath66 belonging to the same legendrian isotopy class of @xmath19 . in a forthcoming paper @xcite we will show that this is in fact the case . in this same paper
we will also show that , under the hypothesis of theorem [ theorem1 ] , for any contact form @xmath2 on @xmath0 the topological entropy of @xmath67 is positive when restricted to @xmath68-limit @xmath69 of a legendrian knot @xmath66 , if @xmath66 is a generic element in the legendrian isotopy class of @xmath19 is the unit tangent bundle of a surface of genus @xmath70 is given in @xcite . ] .
the @xmath68-limit - limit set of a point , which is central in the theory of dynamical systems . ]
@xmath69 of @xmath66 is a compact set invariant by the the flow @xmath71 defined as : + @xmath72 + the positivity of the topological entropy of @xmath71 restricted to @xmath69 can be interpreted in the following way : asymptotically , the path of curves @xmath73 detects the positive topological entropy of @xmath74 .
we believe that this result helps to clarify the role that legendrian curves can have in the study of the global dynamics of reeb flows . we will give an intuitive idea of the proof of theorem [ theorem1 ] .
the basic idea , which is also used in @xcite , is to use the number of reeb chords of @xmath2 from a legendrian submanifold @xmath19 to other legendrian submanifolds to estimate the growth rate of the volume of @xmath75 .
in @xcite the authors study reeb chords from one fixed unit tangent fiber @xmath76 in the unit tangent bundle @xmath77 to the unit tangent fibers @xmath78 , where @xmath79 belongs to a set of full measure in @xmath26 . using lagrangian floer homology
they show that if the homology of the based loop space of @xmath26 is rich , then there exist numbers @xmath80 , @xmath81 , and @xmath82 , and a set @xmath83 of full measure in @xmath26 , such that for @xmath84 we have : @xmath85 where @xmath86 is the number of reeb chords of @xmath2 from @xmath76 to @xmath78 . using the canonical projection from @xmath77 to @xmath26 they estimate the area of the cylinder @xmath87\}$ ] via the counting functions @xmath86 .
the idea of using counting functions for such area estimates is due to gabriel paternain ; see @xcite .
the result is an inequality of the form : @xmath88 where @xmath89 is the measure induced by a riemannian metric @xmath27 on @xmath26 , and @xmath90 is volume of @xmath26 for the measure @xmath89 .
we point out , that in this last inequality , the fact that @xmath30 is a legendrian fibration is used in a crucial way .
most contact 3-manifolds do not have the structure of legendrian fibration .
in fact the only contact 3-manifolds with such structures are unit tangent bundles of surfaces and their coverings .
however , a sufficiently small neighbourhood of a legendrian knot @xmath20 on a contact 3-manifold @xmath7 , is always a legendrian fibration .
this is a consequence of the weinstein legendrian neighbourhood theorem , whose 3-dimensional version asserts that sufficiently small neighbourhoods of legendrian knots are always contactomorphic ; i.e there exists a normal form for small neighbourhoods of legendrian knots . in the hypotheses of theorem [ theorem1 ] we have a pair of disjoint legendrian knots @xmath91 in @xmath7 and a contact form @xmath25 , for which the strip legendrian contact homology @xmath92 has exponential homotopical growth rate .
we begin by choosing a small neighbourhood @xmath93 of @xmath20 which is contactomorphic to @xmath94 ; where @xmath95 and @xmath20 is identified with @xmath96 .
it is clear that for all @xmath97 the curve @xmath98 is legendrian in @xmath93 .
let @xmath43 be the exponential weight of the growth of @xmath92 .
using the invariance properties of the legendrian contact homology we show that given @xmath99 , if we choose @xmath93 to be sufficiently small , then there exist numbers @xmath80 and @xmath100 , and a subset @xmath83 of @xmath101 of full measure , such that for all @xmath102 and @xmath103 the number @xmath104 of reeb chords of @xmath2 from @xmath19 to @xmath105 satisfies : @xmath106 where @xmath45 is the function such that @xmath46 .
we can use the counting functions @xmath107 to estimate the area of the intersection @xmath108 between the cylinder @xmath109 and the neighbourhood @xmath93 ( see figure [ figure ] ) .
we first choose a metric @xmath110 on @xmath4 which restricts to the `` flat '' metric @xmath111 in the coodinates @xmath112 on @xmath93 . because @xmath93 has the structure of a legendrian fibration
, we can apply a local version of paternain s idea to obtain that for @xmath113 @xmath114 where the constant @xmath115 .
this local picture shows that the area @xmath116 ( of the intersection between @xmath109 and @xmath93 ) already grows exponentially fast and allows us to estimate the topological entropy of the reeb flow of @xmath10 by using yomdin s theorem .
it is likely that the local picture we used here can also be applied to estimate the intermediate and slow entropies which were studied in @xcite in contact manifolds which are not unit tangent bundles . in section [ section2 ]
we define the strip legendrian contact homology and show under which conditions it can be defined . in section [ sectiongrowth ] we define the exponential homotopical growth of the strip legendrian contact homology and prove theorem [ theorem1 ] . in section [ section4 ]
we prove theorem [ theorem2 ] . in section [ section5 ]
we propose some open questions which we believe are interesting for future developments .
* acknowledgements : * special thanks to my professors frdric bourgeois and chris wendl for their guidance and support during the development of this work , which is a part of the author s ph.d .
thesis that was developed under their supervision .
my special thanks to frdric for the guidance in the world of legendrian contact homology , without which this work would not have been possible .
i would also like to thank pedro salomo for helpful discussions and for explaining me the techniques used in @xcite which made it possible to avoid dealing with multiply covered pseudoholomorphic curves .
my gratitude to felix schlenk for his interest in this work and for suggesting many future directions of research .
my personal thanks to ana nechita , andr alves , hilda ribeiro and lucio alves for their unconditional personal support .
i warmly thank fnrs - belgium for the financial support which allowed me to focus on mathematical research .
the strip legendrian contact homology is defined in the spirit of the sft invariants introduced in @xcite . to define it we use pseudoholomorphic curves in symplectizations of contact manifolds .
pseudo - holomorphic curves were introduced in symplectic manifolds by gromov in @xcite and adapted to symplectizations and symplectic cobordisms by hofer @xcite ; see also @xcite as a general reference for pseudoholomorphic curves in symplectic cobordisms .
let @xmath7 be a contact manifold and @xmath2 a contact form on @xmath7 .
the symplectization of @xmath7 is the product @xmath117 with the symplectic form @xmath118 ( where @xmath119 denotes the @xmath120 coordinate in @xmath117 ) .
the 2-form @xmath121 restricts to a symplectic form on the vector bundle @xmath8 and it is well known that the set @xmath122 of @xmath121-compatible almost complex structures on the symplectic vector bundle @xmath8 is non - empty and contractible . notice that if @xmath4 is 3-dimensional , the set @xmath122 does not depend on the contact form @xmath2 on @xmath7 . for @xmath123 we can define an @xmath120-invariant almost complex structure @xmath124 on @xmath117 by demanding that @xmath125 we will denote by @xmath126 the set of almost complex structures in @xmath117 that are @xmath120-invariant , @xmath127-compatible and satisfy for some @xmath123 .
an exact symplectic cobordism is , roughly , an exact symplectic manifold @xmath128 that outside a compact subset is like the union of cylindrical ends of symplectizations .
we restrict our attention to exact symplectic cobordisms having only one positive end and one negative end .
let @xmath129 be an exact symplectic manifold without boundary , and let @xmath130 and @xmath131 be contact manifolds with contact forms @xmath132 and @xmath133 .
we say that @xmath129 is an exact symplectic cobordism from @xmath132 to @xmath133 if there exist subsets @xmath134 , @xmath135 and @xmath136 of @xmath137 and diffeomorphisms @xmath138 and @xmath139 \times y^-$ ] , such that : @xmath140 @xmath141 in such a cobordism , we say that an almost complex structure @xmath142 is cylindrical if @xmath143 where @xmath144 and @xmath145 are constants . for fixed @xmath146 and @xmath147 ,
we denote by @xmath148 the set of cylindrical almost complex structures in @xmath149 coinciding with @xmath150 on @xmath135 and @xmath151 on @xmath134 .
it is well known that @xmath148 is non - empty and contractible .
we will write @xmath152 if there exists an exact symplectic cobordism from @xmath153 to @xmath154 as above .
we remind the reader that @xmath155 and @xmath156 implies @xmath152 ; or in other words that the exact symplectic cobordism relation is transitive ; see @xcite for a detailed discussion on symplectic cobordisms with cylindrical ends .
notice that a symplectization is a particular case of an exact symplectic cobordism .
* remark : * we point out that in many references in the literature , a slightly different definition of cylindrical almost complex structures is used : instead of demanding that @xmath142 satisfies and , the stronger condition that @xmath142 coincides with @xmath157 is required .
we need to consider this more relaxed definition of cylindrical almost complex structures when we study the cobordism maps for strip legendrian contact homologies .
let @xmath132 , @xmath2 and @xmath133 be contact forms on @xmath7 such that @xmath158 , @xmath156 . for @xmath159
sufficiently small , it is easy to see that one also has @xmath160 and @xmath161 .
then , for each @xmath162 , we can construct an exact symplectic form @xmath163 on @xmath164 where : @xmath165 \times y , \\ \kappa_r = e^{s+r+2}\lambda^- \ \mbox{in } \ ( -\infty , -r-2 ] \times y,\end{aligned}\ ] ] and @xmath166 \to [ 1-\epsilon,1+\epsilon]$ ] satisfies @xmath167 , @xmath168 and @xmath169 . in @xmath170
we consider a compatible cylindrical almost complex structure @xmath171 ; but we require an extra condition on @xmath171 : @xmath172 \times y.\end{aligned}\ ] ] again we divide @xmath137 into regions : @xmath173 , @xmath174 \times y$ ] , @xmath175 \times y$ ] , @xmath176 \times y$ ] and @xmath177 \times y$ ] . the family of exact symplectic cobordisms with cylindrical almost complex structures @xmath178 is called a splitting family from @xmath132 to @xmath133 along @xmath2 .
let @xmath132 , @xmath2 and @xmath133 be contact forms in @xmath4 .
suppose now that @xmath179 and @xmath180 for functions @xmath181 and @xmath182 satisfying @xmath183 for some @xmath184 .
for each @xmath185 , let @xmath186 such that : @xmath187 \times y , \\
\chi_r(s ) = e^{s - r-2 } \zeta^+ \ \mbox{in } \ [ r+2,+ \infty ) \times y , \\
\chi_r(s ) = ( 1+\frac{s}{r}\epsilon)\zeta \ \mbox{in } \ [ -r , r ] \times y , \\
\chi_r(s ) = e^{s+r+2 } \ \mbox{in } \ ( - \infty , -r-2 ] \times y.\end{aligned}\ ] ] then @xmath188 is an exact symplectic cobordism with cylindrical ends from @xmath132 to @xmath133 . as @xmath189 the region where the symplectic form @xmath190 becomes similar to the symplectization of @xmath191 becomes arbitrarily large
. to gain an intuition about this construction
, one can initially think that in the limit as @xmath189 the sequence @xmath192 splits into two exact symplectic cobordisms , @xmath193 from @xmath132 to @xmath2 , followed by @xmath194 from @xmath2 to @xmath133 . actually , when one studies pseudoholomorphic curves in such cobordisms , the limiting object is more complicated then just the pair of two cobordims we mentioned ; levels of symplectizations have to be inserted above @xmath193 , between @xmath193 and @xmath194 , and below @xmath194 to complete the picture .
we refer again to the paper @xcite for a complete discussion of this topic .
let @xmath195 and @xmath196 be legendrian submanifolds in , respectively , @xmath197 and @xmath198 and let @xmath199 be an exact symplectic cobordism from @xmath132 to @xmath133 .
we call a lagrangian submanifold in @xmath149 a lagrangian cobordism if there exists legendrian submanifolds @xmath200 in @xmath197 and @xmath201 in @xmath198 , and @xmath202 such that : @xmath203 \times y)= ( ( -\infty ,- n ] \times \overline{\lambda}^-).\end{aligned}\ ] ] in this case we say that @xmath204 is a lagrangian cobordism from @xmath200 to @xmath201 . if such an @xmath204 is an exact lagrangian submanifold of @xmath205 , we call it an exact lagrangian cobordism from @xmath200 to @xmath201
. if we take a legendrian submanifold @xmath19 in @xmath198 then @xmath206 is an exact lagrangian submanifold in the symplectization of @xmath207 .
it is also an exact lagrangian cobordism in @xmath188 from @xmath208 to itself .
let @xmath209 be a closed riemann surface , possibly with boundary , and @xmath210 be a finite set .
we define @xmath211 .
let @xmath2 be a contact form in @xmath7 and @xmath212 .
a finite energy pseudoholomorphic curve in the symplectization @xmath213 with boundary in a lagrangian submanifold @xmath204 is a map @xmath214 that satisfies @xmath215 and @xmath216 where @xmath217 with @xmath218 ; q ' \geq 0\}$ ] .
the quantity @xmath219 is called the hofer energy and was introduced in @xcite .
the operator @xmath220 above is called the cauchy - riemann operator for the almost complex structure @xmath124 . for us ,
particularly important will be the case where @xmath221 is biholomorphic to @xmath222,i_0)$ ] denotes the complex structure in @xmath223 .
] , @xmath224 for disjoint legendrian knots @xmath19 and @xmath20 , and @xmath225 satisfies @xmath226 and @xmath227 . in this case @xmath225 is called a pseudoholomorphic strip . by using a biholomorphism @xmath228,i_0)$ ] satisfying @xmath229 ( where @xmath230 is the northern hemisphere of @xmath231 ) and @xmath232 ( where @xmath233 is the southern hemisphere of @xmath231 ) we can also view pseudoholomorphic strips as maps having as domain the closed disc with two punctures on the boundary . in this paper ,
when studying pseudoholomorphic curves in symplectizations we will always assume that @xmath234 for some legendrian link @xmath235 in @xmath7 . for an exact symplectic cobordism @xmath236 from @xmath132 to @xmath133 , and @xmath237 where @xmath238 and @xmath239 , a finite energy pseudoholomorphic curve with boundary in a lagrangian submanifold @xmath204 is again a map @xmath240 that satisfies @xmath215 and @xmath241 where : + @xmath242 , + @xmath243 , + @xmath244 .
+ these energies were also introduced in @xcite . in this paper ,
when studying pseudoholomorphic curves in exact symplectic cobordisms we will always assume that @xmath204 is an exact symplectic cobordism from a legendrian link @xmath200 in @xmath245 to a legendrian link @xmath201 in @xmath246 .
in splitting symplectic cobordisms we use a slightly modified version of energy .
instead of demanding @xmath247 we demand : @xmath248 where : + @xmath249 , + @xmath250 , + @xmath251 , + and @xmath252 and @xmath253 are as above .
the elements of the set @xmath210 are called punctures of the pseudoholomorphic @xmath225 .
we first divide @xmath254 in two classes : we call the elements of @xmath255 boundary punctures and the elements in @xmath256 interior punctures . the work of hofer @xcite , hofer et al . @xcite and abbas @xcite allows us do classify the punctures in four different types
. we will describe the behaviour of punctures for pseudoholomorphic curves in exact symplectic cobordisms , since the case of symplectizations is contained in this one as a particular case . before presenting this classification
we introduce some notation : we let @xmath257 be the ball of radius @xmath258 centered at the puncture @xmath259 , and denote by @xmath260 the set defined as the closure @xmath261 of the intersection of the boundary of @xmath257 with the interior of @xmath57 .
notice that @xmath260 is a circle or an interval , depending on whether @xmath259 is an interior or a boundary puncture .
we can describe the types of punctures as follows : * @xmath262 is a positive boundary puncture if @xmath263 and @xmath264 , and there exists a sequence @xmath265 and a reeb chord @xmath266 of @xmath267 from @xmath200 to itself , such that @xmath268 converges in @xmath269 to @xmath266 as @xmath270 ; * @xmath262 is a negative boundary puncture if @xmath263 and @xmath271 , and there exists a sequence @xmath265 and reeb chord @xmath272 of @xmath273 from @xmath201 to itself , such that @xmath268 converges in @xmath269 to @xmath272 as @xmath270 ; * @xmath262 is a positive interior puncture if @xmath274 and @xmath264 , and there exists a sequence @xmath265 and reeb orbit @xmath275 of @xmath267 , such that @xmath268 converges in @xmath269 to @xmath275 as @xmath270 ; * @xmath262 is a negative interior puncture if @xmath274 and @xmath271 , and there exists a sequence @xmath265 and reeb orbit @xmath276 of @xmath273 , such that @xmath268 converges in @xmath269 to @xmath276 as @xmath270 .
the results in @xcite imply that these are indeed the only possibilities we need to consider for the behaviour of the @xmath225 near punctures .
intuitively , we have that at the punctures , the pseudoholomorphic curve @xmath225 detects reeb chords and reeb orbits . for a boundary ( interior )
puncture @xmath259 , if there is a subsequence @xmath277 such that @xmath268 converges to a given reeb chord @xmath21 ( orbit @xmath278 ) , we will say that @xmath225 is asymptotic to this reeb chord @xmath21 ( orbit @xmath278 ) at the puncture @xmath259 .
if a pseudoholomorphic curve is asymptotic to a transverse reeb chord or a non - degenerate reeb orbit at a puncture , more can be said about its asymptotic behaviour in neighbourhoods of this puncture . in order to describe this behaviour for a boundary puncture @xmath259 , we take a closed neighbourhood @xmath279 of @xmath259 that admits a holomorphic chart @xmath280 \ \subset \mathbb{c}$ ] , such that @xmath281 and @xmath282 $ ] . in coordinates
@xmath283 $ ] we have @xmath284 when @xmath285 tends to the puncture @xmath259 . with this notation , it is shown in @xcite , that if @xmath259 is a positive boundary puncture on which @xmath225 is asymptotic to a transverse reeb chord @xmath266 of @xmath267 from @xmath195 to and @xmath286 denote connected components of @xmath200 . ]
@xmath286 , then @xmath287 satisfies : * @xmath288 converges in @xmath289 to the reeb chord @xmath266 , uniformly in @xmath290 and exponentially in @xmath291 .
similarly , if @xmath259 is a negative boundary puncture on which @xmath225 is asymptotic to a transverse reeb chord @xmath272 of @xmath273 from @xmath196 to and @xmath292 denote connected components of @xmath201 . ]
@xmath292 , then @xmath287 satisfies : * @xmath288 converges in @xmath289 to the reeb chord @xmath272 , uniformly in @xmath290 and exponentially in @xmath291 .
we discuss now the case where @xmath259 is an interior puncture .
take a neighbourhood @xmath293 of @xmath259 that admits a holomorphic chart @xmath294 .
using polar coordinates @xmath295 we can write @xmath296 as @xmath297 . with this notation , it is shown in @xcite that if @xmath259 is a positive interior puncture at which @xmath225 is asymptotic to a non - degenerate reeb orbit @xmath275 of @xmath267 , then @xmath298 satisfies * @xmath288 converges in @xmath289 to a reeb orbit @xmath275 of @xmath267 , exponentially in @xmath291 and uniformly in @xmath290 . similarly ,
if @xmath259 is a negative interior puncture at which @xmath225 is asymptotic to a non - degenerate reeb orbit @xmath276 of @xmath273 , then @xmath287 satisfies * @xmath288 converges in @xmath289 to a reeb orbit @xmath276 of @xmath299 as @xmath300 , exponentially in @xmath291 and uniformly in @xmath290 .
_ remark : the exponential rate of convergence of pseudoholomorphic curves near punctures to reeb orbits and reeb chords is of crucial importance for the fredholm theory ( see @xcite ) that gives the dimension of the space of pseudoholomorphic curves with fixed asymptotic data , and is a consequence of asymptotic formulas obtained in @xcite .
_ the discussion above can be summarised by saying that near punctures the finite energy pseudoholomorphic curves detect reeb orbits and reeb chords .
it is exactly this behavior that makes these objects useful for the study of dynamics of reeb vector fields .
* fact : as a consequence of the exactness of the symplectic cobordisms and the lagrangian submanifolds considered above we obtain that the energy @xmath219 of @xmath225 satisfies @xmath301 where @xmath302 is the sum of the action of the reeb orbits and reeb chords detected by the punctures of @xmath225 counted with multiplicity ; see for example @xcite .
* we are now ready to define the strip legendrian contact homology .
first , we introduce the following notation : for a given contact form @xmath2 we denote by @xmath303 the set of reeb chords of @xmath304 starting at @xmath19 and ending at @xmath20 .
suppose that there is a contact form @xmath25 associated to @xmath7 , and disjoint legendrian knots @xmath19 and @xmath20 in @xmath7 , such that the reeb flow of @xmath25 : * \(a ) has no contractible reeb orbits , * \(b ) has no reeb chords from @xmath19 to itself that vanish in @xmath305 , * \(c ) has no reeb chords from @xmath20 to itself that vanish in @xmath306 , * \(d ) and all reeb chords in @xmath307 are transverse and do not intersect reeb orbits of @xmath25 .
we will call such a contact form _ adapted _ to the pair @xmath18 . in order to assign a @xmath308-grading to the reeb chords in @xmath307
, we use the conley - zehnder index .
we recall the definition of conley - zehnder index for reeb chords as presented in @xcite .
for the definition , we first fix , once and for all , orientations for @xmath19 and @xmath20 .
then , for each reeb chord @xmath309 , let @xmath310 be a nowhere vanishing section of the vector bundle @xmath311 that : * is tangent to @xmath19 on the initial point of @xmath21 and coincides with the orientation we fixed for @xmath19 at this initial point , * is tangent to @xmath20 on the final point of @xmath21 and coincides with the orientation we fixed for @xmath20 at this final point . the section @xmath310 induces a ( unique up to homotopy ) symplectic trivialisation of @xmath312 , which we also denote by @xmath310 . using the reeb flow @xmath313 we define a path of lagrangian subspaces @xmath314 of @xmath315 .
we consider the parametrisation @xmath316
\to y$ ] of the reeb chord @xmath21 given by the reeb flow .
letting @xmath317 denote the linearisation of the reeb flow , we define @xmath318 to be the unique lagragian subspace of @xmath319 that contains @xmath320 , where @xmath321 is a vector tangent to @xmath19 and giving the orientation we chose for @xmath19 . after arriving at the endpoint @xmath322 , we complete @xmath314 to obtain a lagrangian loop by making a continuous left - rotation of @xmath323 ( among legendrian subspaces of @xmath324 ) till it meets the tangent space to @xmath20 . with this completion and using our trivialisation @xmath310 , we associate to @xmath314 a path of lagragian subspaces of the standard symplectic plane .
the conley - zehnder index @xmath325 is defined to be the maslov index of this path .
it is clear from the constructions above that , because we fixed the orientations of @xmath19 and @xmath20 , the parity of @xmath325 is independent of the trivialisation @xmath326 .
this allows us to define , for each @xmath309 , its @xmath308-grading by @xmath327 .
we call chords with grading @xmath328 even chords , and chords with grading @xmath329 odd chords . let @xmath330 be the @xmath308 vector - space generated by @xmath307 .
we denote by @xmath331 the subspace of @xmath330 generated by odd chords , and @xmath332 the subspace generated by even chords . given two reeb chords @xmath333 and @xmath334 , and an almost complex structure @xmath335 , we know from the previous section that it makes sense to consider the moduli space @xmath336 whose elements are equivalence classes of finite energy pseudoholomorphic strips @xmath337 satisfying : * @xmath329 is a positive boundary puncture , and @xmath225 is asymptotic to @xmath333 at @xmath329 , * @xmath338 is a negative boundary puncture , and @xmath225 is asymptotic to @xmath334 at @xmath338 , * @xmath339 , * @xmath340 . two pseudoholomorphic strips @xmath225 and @xmath341 satisfying these four conditions represent the same element in @xmath342 if there exists a biholomorphism @xmath343 of the disc that fixes @xmath344 such that @xmath345 .
it follows from abbas asymptotic analysis @xcite explained in section [ pseudohol ] that all the elements of the moduli space @xmath336 are somewhere injective pseudoholomorphic curves .
it is well known that the linearization @xmath346 at any element @xmath336 is a fredholm map ( we remark that this property is valid for more general moduli spaces of curves with prescribed asymptotic behaviour ) .
moreover , dragnev @xcite and abbas @xcite showed that for a generic set @xmath347 all the elements in @xmath336 are transverse in the sense that the linearization @xmath346 of the cauchy - riemann operator @xmath220 at the elements of @xmath348 is surjective , this being valid for all reeb chords @xmath333 and @xmath334 from @xmath19 to @xmath20 .
thus , in the case where @xmath349 one can use the implicit function theorem , and obtain that any connected component of the moduli space @xmath336 is a finite dimensional manifold , and its dimension is given by the fredholm index @xmath350 of @xmath346 computed at any element of this connected component of @xmath336 .
we let @xmath351 be the moduli space of pseudoholomorphic strips with fredholm index @xmath352 .
it follows from the formula in @xcite for the fredholm index @xmath350 of the linearised @xmath346 operator over a strip in @xmath336 , that @xmath350 has the same parity as @xmath353 .
notice also that , as @xmath354 , @xmath336 can only be non - empty if @xmath355 .
because of the @xmath120-invariance of the almost complex structure @xmath124 there is an @xmath120-action on the spaces @xmath336 , and we let @xmath356 .
we are now ready to define a differential @xmath357 in @xmath330 .
let @xmath309 and @xmath358 .
we define : @xmath359 c'\ ] ] where @xmath360 is the cardinality of the moduli space @xmath361 of pseudoholomorphic strips of fredholm index @xmath329 modulo the @xmath120-action .
the differential is extended to @xmath330 by linearity . to complete the construction of the strip legendrian contact homology
, we must prove that @xmath357 is well - defined and that @xmath362 . before proceeding with the proofs of these results
we discuss the intuition behind the definition of this homology .
the strip legendrian contact homology can be seen as a relative version of the cylindrical contact homology ( see @xcite ) . for cylindrical contact homology to be well - defined for a contact form , this contact form has to have some special properties ; for example , for a hypertight contact form ( i.e. one that does not have contractible periodic orbits ) cylindrical contact homology is well - defined . as we will see later , the non - existence of contractible reeb orbits precludes the `` bubbling '' of pseudoholomorphic planes .
this , together with sft - compactness , implies that if its asymptotic orbits are in a primitive homotopy class , a sequence of pseudoholomorphic cylinders of fredholm index 2 can only break in a pseudoholomorphic building of 2 levels , each containing a cylinder of fredholm index 1 .
this implies that only such buildings can appear in the boundary of the compactified moduli space of pseudoholomorphic cylinders of fredholm index @xmath363 .
this description of the compactified moduli spaces of pseudoholomorphic cylinders of index @xmath363 , is the crucial step that allows us to define cylindrical contact homology with coefficients in @xmath308 .
the strip legendrian contact homology is the relative analogue of the cylindrical contact homology .
this time the differential involves pseudoholomorphic strips with boundary conditions on lagrangian cylinders over legendrian submanifolds .
for such a theory to be well - defined we have to preclude not only `` bubbling '' of planes but also of pseudoholomorphic half - planes . the conditions ( b ) and ( c ) above serve exactly to make impossible such `` bubbling '' phenomena , and the condition ( d ) is a non - degeneracy condition . under these hypothesis
it is possible to define the strip legendrian contact homology , and to carry out this construction one uses results on the analytical properties of pseudoholomorphic strips and discs .
for these results we refer to : @xcite for the necessary results on the asymptotic behaviour of punctures , @xcite for the necessary results on fredholm theory , @xcite for the necessary transversality results , @xcite for ( essentially ) the necessary techniques to perform gluing .
we now proceed to prove : [ lemma1 ] for @xmath364 , and @xmath357 defined above we have : * \(1 ) @xmath357 is well defined , * \(2 ) @xmath357 decreases the action of reeb chords , * \(3 ) for each @xmath309 , @xmath365 is a finite sum , * \(4 ) @xmath366 and @xmath367 _ proof : _ in order for @xmath357 to be well - defined we have to prove that the moduli space @xmath368 is finite for every @xmath21 and @xmath369 . because @xmath370 , it follows that @xmath368 is a @xmath328-dimensional manifold .
if we show that it is compact then it has to be a finite set . to obtain the compactness
we will apply the `` bubbling of '' analysis for pseudoholomorphic curves in symplectizations of @xcite and the sft - compactness results of @xcite .
let @xmath371 be a sequence of elements of @xmath368 . because of the assumptions we made on the contact form @xmath25 , the sequence @xmath371 can not have interior bubbling points : an interior bubbling point would imply the existence of a finite energy plane and thus of a contractible periodic orbit of @xmath372 which contradicts ( a ) above .
boundary bubbling points are also forbidden : they would give rise to either a pseudoholomorphic disc with boundary in @xmath206 , a pseudoholomorphic disc with boundary on @xmath373 , a pseudoholomorphic disc with only one puncture asymptotic to a reeb chord from @xmath19 to itself , or a a pseudoholomorphic disc with only one puncture asymptotic to a reeb chord from @xmath20 to itself .
the first two possibilities are impossible because @xmath206 and @xmath373 are exact lagrangian submanifolds and the later two because they would contradict conditions ( b ) and ( c ) satisfied by @xmath25 . combining this information with the sft - compactness results of @xcite we have that @xmath371 converges in the sft sense to a pseudoholomorphic building @xmath225 with k - levels @xmath374 , with all levels @xmath374 being pseudoholomorphic strips with non - vanishing hofer energy that satisfy : *
@xmath329 is a positive boundary puncture , and @xmath374 is asymptotic to @xmath375 at @xmath329 , * @xmath338 is a negative boundary puncture , and @xmath225 is asymptotic to @xmath376 at @xmath338 , * @xmath377 , where @xmath233 is the northern hemisphere , * @xmath378 , where @xmath233 is the southern hemisphere , where @xmath379 and @xmath380 . because every @xmath374 is somewhere injective and has non - zero hofer energy , we have that the indices @xmath381 satisfy @xmath382 .
this implies that @xmath383 . on the other hand as
@xmath225 is the limit of a sequence of pseudoholomorphic strips of fredholm index 1 , it has to satisfy @xmath384 . we thus conclude that @xmath385 , and @xmath386 , which implies the desired compactness .
this proves that @xmath360 is finite for every @xmath387 , and thus that @xmath357 is well defined . to verify item ( 2 )
, we remark that given @xmath309 , the number @xmath360 can only be non - zero for reeb chords @xmath369 that satisfy @xmath388 .
this implies that @xmath357 decreases the action of reeb chords . by the non - degeneracy condition ( d ) one obtains that the set of reeb chords with action smaller then @xmath22 is finite , and so @xmath360 is non - zero only for a finite number of @xmath369
this implies that @xmath365 is a finite sum , proving item ( 3 ) .
item ( 4 ) follows easily from the fact that the fredholm index of a strip connecting two chords @xmath21 and @xmath369 has the same parity as @xmath389 , as this implies @xmath368 can only be non - empty if @xmath21 and @xmath369 have different parity .
in @xcite it is proved that the moduli spaces @xmath390 admit a compactification .
the compactfied moduli space is composed not only of pseudoholomorphic curves , but also of pseudoholomorphic buildings .
we will denote this compactification of @xmath390 by @xmath391 .
[ lemma2 ] for @xmath358 , and @xmath357 as defined before we have @xmath392 .
_ proof : _ the lemma will be a consequence of the description we will give of the compactified moduli space @xmath393 of pseudoholomorphic strips with fredholm index @xmath363 . because of regularity of @xmath124 , it will follow that for all @xmath394 , @xmath393 is either empty , or the finite union of disjoint circles and closed intervals .
we summarise this in the following claim .
_ compactness claim : suppose @xmath393 is non - empty .
then , each connected component @xmath395 of @xmath393 is either a circle or a closed interval .
moreover , when @xmath395 is homeomorphic to a closed interval , its boundary is composed by pseudoholomorphic buildings @xmath225 with 2 levels @xmath396 and @xmath397 satisfying : _ _ @xmath398 and @xmath399 for some @xmath400_. before proving this claim we use it to prove lemma [ lemma2 ] .
for this , we write @xmath401 it is clear that the lemma will follow if we can prove that @xmath402 is even . on one hand , it follows from our definition of @xmath357 that @xmath402 counts the number of 2-level pseudoholomorphic buildings whose levels @xmath396 and @xmath397 satisfy @xmath398 and @xmath399 for some @xmath400 .
this together with the compactness claim implies that the number of boundary components of @xmath403 is smaller or equal to @xmath402 . on the other hand ,
because of the regularity of @xmath370 , we can apply the gluing theorem for fredholm regular pseudoholomorphic strips : it implies that if @xmath225 is 2-level pseudoholomorphic building whose levels @xmath396 and @xmath397 satisfy @xmath398 and @xmath404 for some @xmath400 , then @xmath225 represents exactly one point in the boundary of @xmath393 .
we thus have that @xmath402 is bigger or equal to the number of boundary points of @xmath403 .
summarising , the combination of the compactness claim and the gluing theorem allows us to conclude that the number @xmath402 is exactly the number of boundary components of the moduli space @xmath393 . because @xmath393 is a finite union of disjoint intervals and circles , this number is even
this finishes the proof of the lemma modulo the compactness claim .
_ proof of compactness claim : _ suppose @xmath393 is non - empty and let @xmath395 be one of its connected components .
it follows from the regularity of @xmath124 that the interior @xmath405 of @xmath395 is a 1-dimensional manifold .
we have now two possibilities , either @xmath395 is compact or not . if @xmath405 is compact then it is a circle .
if that is not the case , let @xmath371 be a sequence of elements of @xmath405 converging to the boundary of @xmath395 .
reasoning as we did in the proof of the lemma [ lemma1 ] , we obtain that no `` bubbling '' can occur .
thus the sft - compactness theorem of @xcite implies that @xmath371 converges to a pseudoholomorphic building @xmath225 with k - levels @xmath374 , such that all levels @xmath374 are pseudoholomorphic strips satisfying : * @xmath329 is a positive boundary puncture , and @xmath374 is asymptotic to @xmath375 at @xmath329 , * @xmath338 is a negative boundary puncture , and @xmath225 is asymptotic to @xmath376 at @xmath338 , * @xmath406 where @xmath233 is the northern hemisphere , * @xmath378 where @xmath233 is the southern hemisphere , where @xmath379 and @xmath380 . again , because every @xmath374 is somewhere injective , we have that the fredholm indices of the levels satisfy @xmath407 , and it follows that @xmath383 . on the other hand as
@xmath225 is the limit of a sequence of pseudoholomorphic strips of fredholm index 2 , it has to satisfy @xmath408 .
we have then two possibilities : either @xmath385 and @xmath409 ; or @xmath410 which forces @xmath411 , for @xmath412 and @xmath413 for some @xmath414 .
the first case is ruled out because we assumed that @xmath371 converges to the boundary of @xmath395 .
we have thus obtained that all the elements on the boundary of @xmath395 are 2-level pseudoholomorphic buildings with the properties claimed , and an analysis identical to the one done in lemma [ lemma1 ] shows that the moduli space of such buildings is a compact finite set .
we conclude that the boundary of @xmath395 is a compact 0-dimensional manifold .
the gluing theorem gives a description of a small neighbourhood @xmath279 of a 2-level pseudoholomorphic building @xmath225 appearing in the boundary of a connected component @xmath395 of @xmath393 .
the neighbourhood @xmath279 admits a homeomorphism to the interval @xmath415 , that takes @xmath328 to the 2-level building @xmath225 and all other values to pseudoholomorphic strips in the interior of @xmath395 . summing up
, we have shown that the connected component @xmath395 of the moduli space @xmath393 has the structure of a one - dimensional manifold with boundary , i.e a closed interval .
this finishes the proof of the compactness claim .
we denote by @xmath92 the homology associated to the chain - complex @xmath416 .
the free homotopy classes of paths starting at @xmath19 and ending at @xmath20 generate subcomplexes of @xmath92 . to formalize this
we denote by @xmath417 the set of homotopy classes of paths starting at @xmath19 and ending at @xmath20 . for our contact form @xmath25 and an element @xmath418
we denote by @xmath419 the set of reeb chords from @xmath19 to @xmath20 that belong to @xmath420 .
it is clear that for all @xmath421 , the terms @xmath422 $ ] appearing in the differential @xmath423 c'$ ] , can only be non - zero if @xmath424 .
this implies that the vector spaces @xmath425 are subcomplexes of the chain complex @xmath426 and that : @xmath427 where @xmath428 denotes the homology of the chain - complex @xmath429 .
symplectic cobordisms play a crucial role in sft because they induce maps between the sft invariants . under suitable conditions this is also true for the strip legendrian contact homology .
[ propcobordism ] let @xmath130 and and @xmath131 be contact manifolds of the same dimension , @xmath132 be a contact form on @xmath130 adapted to a pair of legendrian curves @xmath195 and @xmath286 in @xmath130 and @xmath133 be a contact form on @xmath131 adapted to a pair of legendrian curves @xmath196 and @xmath292 in @xmath130 .
let @xmath430 be an exact symplectic cobordism from @xmath431 to @xmath432 for some constant @xmath433 , @xmath434 be an exact lagrangian cobordism from @xmath195 to @xmath196 , and @xmath435 be an exact lagrangian cobordism from @xmath286 to @xmath292 .
then these cobordisms induce a map @xmath436 from @xmath437 to @xmath438 . _
proof : _ taking @xmath439 and @xmath440 , we can define the homologies @xmath437 and @xmath438 . the idea to define the cobordism map is to choose @xmath441 and define @xmath436 by counting pseudoholomorphic strips @xmath442 with fredholm index @xmath328 , having @xmath329 as a positive boundary puncture asymptotic to a reeb chord @xmath443 , @xmath338 as a negative boundary puncture asymptotic to a reeb chord @xmath444 , and with boundary conditions @xmath445 and @xmath446 . in order for this approach to work
we must first choose @xmath447 that satisfy certain regularity assumptions .
for a pair @xmath448 and @xmath449 , we introduce the moduli space @xmath450 whose elements are equivalence classes of pseudoholomorphic strips @xmath442 with fredholm index @xmath328 , having @xmath329 as a positive boundary puncture asymptotic to a reeb chord @xmath443 , @xmath338 as a negative boundary puncture asymptotic to a reeb chord @xmath444 , and with boundary conditions @xmath451 and @xmath452 .
two such pseudoholomorphic strips @xmath225 and @xmath341 represent the same element in @xmath450 if one can be obtained from the other via a biholomorphic reparametrization of the disk that fixes @xmath344 .
we let @xmath453 be the set of pseudoholomorphic strips of fredholm index @xmath352 . as all the strips in these moduli spaces
are somewhere injective curves , it follows from @xcite that for a generic set @xmath454 , all the pseudoholomorphic strips in @xmath455 , for all choices of @xmath448 and @xmath449 , are fredholm regular .
we assume from now on that @xmath456 .
initially , the map @xmath436 is obtained by counting elements in the moduli space @xmath457 , and is defined from @xmath458 to @xmath459 .
more precisely @xmath460 to see that it actually induces a map on the homology level one has to check that @xmath461 .
the proof of this fact is a combination of compactness and gluing , and is identical to similar statements for lagrangian floer homology or cylindrical contact homology ; see for example @xcite .
a compactness argument identical to the one in lemma [ lemma1 ] shows that @xmath436 is well defined and is a finite sum . because of the regularity for all pseudoholomorphic curves involved in the definitions of the maps @xmath462 , @xmath463 and @xmath436 , it is possible to perform gluing for the pseudoholomorphic curves involved in these definitions .
more precisely , the map @xmath464 counts the number of 2-level pseudoholomorphic buildings @xmath225 whose levels @xmath465 satisfy @xmath466 and @xmath467 for some @xmath468 .
analogously , @xmath469 counts the number of 2-level pseudoholomorphic buildings @xmath225 whose levels @xmath470 satisfy @xmath471 and @xmath472 for some @xmath473 and @xmath444 .
the gluing theorem implies that each 2-level building that appears in the definition of the map @xmath461 is one point in the boundary of the moduli space @xmath474 .
because of the exactness of @xmath430 , of the lagrangians @xmath204 and @xmath475 , and using that @xmath132 is adapted to the pair of legendrians @xmath476 , and @xmath133 is adapted to the pair of legendrians @xmath477 , we have that a sequence of elements in @xmath478 can only break in 2-level buildings that appear in the definition of the maps @xmath479 and @xmath480 . the complete proof of this compactness fact uses just the additivity of the fredholm index for pseudoholomorphic buildings and the fredholm regularity of the pseudoholomorphic buildings involved , and is identical to a similar argument we used in lemma [ lemma2 ] .
this combination of compactness and gluing implies that the moduli space @xmath474 is a one - dimensional manifold whose boundary is composed exactly of the 2-level buildings that appear in the definition of the map @xmath481 .
summarising , we obtain for each @xmath482 : @xmath483 c^-\ ] ] where @xmath484 is the number of pseudoholomorphic buildings appearing on the boundary of @xmath474 .
as @xmath474 is a 1-dimensional manifold , the number of its boundary components is even which implies that @xmath485=0 $ ] , and finishes the proof of the proposition . in the case where @xmath430 is the symplectization of a contact manifold with an @xmath120-invariant regular almost complex structure and the lagrangians @xmath204 and @xmath475 are just the lagrangian cylinders over @xmath19 and @xmath486 , respectively ,
then the induced cobordism map is the identity .
the reason for this is that in this situation the only pseudoholomorphic strips with fredholm index @xmath328 are the trivial strips over reeb chords .
we now restrict our attention to a special situation .
we assume * \(a ) that @xmath132 and @xmath487 are contact forms on the same contact manifold @xmath7 and are of the form @xmath488 and @xmath489 for a contact form @xmath25 on @xmath7 and a constant @xmath490 , * \(b ) that @xmath195 and @xmath196 are the same legendrian curve @xmath19 on @xmath7 , * \(c ) that @xmath286 and @xmath491 are the same legendrian curve @xmath492 on @xmath7 , * \(d ) that @xmath493 and there exists a homotopy @xmath494\}$ ] of exact symplectic cobordisms from @xmath25 to @xmath495 which is constant outside a compact set @xmath51 of @xmath496 , such that @xmath497 and @xmath498 is the symplectization of @xmath499 , * \(e ) that there exists a continuous homotopy of exact lagrangian cobordisms @xmath500 from @xmath19 to @xmath19 which is constant outside the compact set @xmath51 , and such that @xmath501 and @xmath502 , * \(f ) and that there exists a continuous homotopy of exact lagrangian cobordisms @xmath503 from @xmath20 to @xmath20 which is constant outside the compact set @xmath51 , and such that @xmath504 and @xmath505 .
notice that in this case the lagrangian cobordisms @xmath204 and @xmath475 are cylinders .
it is clear that such a cobordism actually induces a map that respects the homotopy classes of the chords .
therefore we can define , for each @xmath418 , a map @xmath436 from @xmath506 to @xmath506 . in order to define the cobordism maps for these subcomplexes @xmath507 the assumptions on the regularity of the almost complex structure @xmath447 are slightly weaker .
fix @xmath418 and @xmath508 . for each pair
@xmath21 and @xmath369 of reeb chords in @xmath419 and almost complex structure @xmath509 , we consider the moduli space @xmath510 of pseudoholomorphic strips with fredholm index @xmath352 .
as all the strips in these moduli spaces are somewhere injective curves , there exists a dense set @xmath511 , such that if @xmath512 all the pseudoholomorphic strips in all the moduli spaces @xmath513 , for every @xmath21 and @xmath369 , are fredholm regular .
applying the same reasoning as the one in the proof of proposition [ propcobordism ] one shows that for @xmath512 there exists a map @xmath436 from @xmath506 to @xmath506 .
notice that @xmath514 might contain elements that are not in @xmath515 .
we consider an exact symplectic cobordism @xmath430 and exact lagrangian cobordisms @xmath516 satisfying ( a ) , ( b ) , ( c ) , ( d ) , ( e ) and ( f ) .
let @xmath517 , @xmath518 and @xmath475 be the homotopies given in ( d ) , ( e ) and ( f ) .
we define for a fixed choice of @xmath508 and @xmath512 the space @xmath519 of smooth homotopies @xmath520,\ ] ] of almost complex structures such that @xmath521 , @xmath522 , and @xmath523 is compatible with @xmath524 for every @xmath525 $ ] . for reeb chords @xmath387 we define the moduli space @xmath526 \mbox { and } \widetilde{w}\in \mathcal{m}^k(c , c';j_t;l_t,\widehat{l}_t)\}.\ ] ] by the results in @xcite , we know that there is a generic subset @xmath527 such that all elements of @xmath528 are fredholm regular for every choice of @xmath529 and every number @xmath352 .
this implies that @xmath528 is a @xmath530-dimensional manifold with boundary . the crucial condition that makes
this valid is again the fact that all the pseudoholomorphic curves that make part of this moduli space are somewhere injective .
notice that @xmath528 is not necessarily compact since sequences of elements in @xmath528 might break .
however , as shown in @xcite , it can be compactfied with the addition of pseudoholomorphic buildings .
[ propchainhom ] let @xmath531 be an exact symplectic cobordism from @xmath25 to @xmath495 where @xmath25 is a contact form on @xmath7 and @xmath532 is a constant , and that @xmath25 is adapted to a pair of legendrians @xmath19 and @xmath20 .
assume also that there exists in @xmath533 exact lagrangian cobordisms @xmath534 from @xmath19 to @xmath19 , and @xmath435 from @xmath20 to @xmath20 , and that the cobordisms @xmath533 , @xmath204 and @xmath475 satisfy the conditions ( a ) , ( b ) , ( c ) , ( d ) , ( e ) and ( f ) .
then , if @xmath335 , we have that for all @xmath512 the map @xmath436 from @xmath506 to itself is chain homotopic to the identity . _
proof : _ the proof is a standard argument in sft , and we direct the reader to the original source @xcite for an exposition of this argument for general sft invariants and @xcite where the very similar case of cylindrical contact homology is treated .
we first take an almost complex structure @xmath535 and choose an almost complex structure @xmath536 compatible with @xmath537 . the map @xmath436 will count pseudoholomorphic strips in @xmath538 satisfying boundary conditions as the ones used in proposition [ propcobordism ] . for the deformation
@xmath524 we can take a homotopy @xmath523 of almost complex structures in @xmath539 .
the crucial ingredient of the proof is the description of the compactification @xmath540 of the moduli space @xmath541 .
we want to understand the boundary of @xmath542 .
take a sequence @xmath543 of elements in @xmath541 converging to the boundary of @xmath540 .
there are three possibilities : @xmath544 goes to @xmath328 , @xmath544 goes to @xmath329 or the limit @xmath545 of @xmath544 belongs to @xmath546 . in all three possibilities
we know that the conditions ( a ) , ( b ) , ( c ) and ( d ) satisfied by @xmath25 , @xmath19 and @xmath20 prevent any `` bubbling '' in the sequence . if @xmath547 regularity of @xmath447 and @xmath124 imply that the @xmath371 must converge to an element @xmath225 of @xmath548 .
if @xmath549 it follows from the regularity of @xmath124 that @xmath371 must converge to @xmath225 of @xmath550 . if @xmath551 the sequence @xmath371 converges to a pseudoholomorphic building @xmath225 whose levels @xmath374 are all pseudoholomorphic strips
the fact that for @xmath552 all the strips appearing have fredholm index @xmath553 gives us more information about this pseudoholomorphic building . because @xmath554 is in the boundary of @xmath540 and @xmath555 , we know that @xmath225 has at least 2-levels .
it also has fredholm index @xmath328 since all @xmath371 have fredholm index @xmath328 .
of all levels of @xmath225 there is precisely one in the cobordism @xmath556 , and all the others are in symplectizations of @xmath25 above and below the cobordism .
the levels in the symplectizations have fredholm index @xmath557 and the one on @xmath556 has fredholm index @xmath553 . combining this with the fact that @xmath225 has fredholm index @xmath328
we conclude that @xmath225 has exactly 2 levels : one with fredholm index @xmath338 in @xmath556 and one with fredholm index @xmath329 in the symplectization of @xmath25 .
there are then two possible cases : either the building @xmath225 has the top level @xmath558 and the lower level @xmath559 for some @xmath400 ; or the building @xmath225 has the top level @xmath560 and the lower level @xmath561 for some @xmath400 .
on the other hand , the gluing theorem implies that : if @xmath225 is a 2-level building whose top level @xmath558 and whose lower level @xmath559 , then @xmath562 is an element of the boundary of @xmath540 .
the same is valid for @xmath562 if the building @xmath225 has the top level @xmath560 and the lower level @xmath561 . finally , if @xmath563 then the regularity of the homotopy @xmath523 implies that @xmath564 is in the boundary of @xmath540 .
the same reasoning implies that @xmath565 is in the boundary of @xmath540 if @xmath566 .
we have thus obtained a complete description of the boundary of @xmath540 .
we now define a map @xmath567 that counts finite energy fredholm index @xmath338 pseudoholomorphic strips in the homotopy @xmath517 , with one boundary component in @xmath518 and one in @xmath568 . in order to define @xmath49 precisely we first define for @xmath387 the set @xmath569 |
\mathcal{m}^{-1}(c , c';j_{t_\infty},l_{t_\infty},\widehat{l}_{t_\infty } ) \neq \emptyset \}$ ] . from the regularity of the homotopy @xmath523
we know that @xmath570 is always a finite set .
we can now define for @xmath309 @xmath571 because of the regularity of the homotopy @xmath523 , we know that the moduli space @xmath572 is finite for every @xmath573 $ ] . for action reasons , we also know that for a fixed @xmath309 the term @xmath574 can only be non - zero for a finite number of chords @xmath575 .
these considerations imply that @xmath576 as defined above is well - defined and is a finite sum .
we then obtain @xmath49 by extending it linearly to @xmath577 .
consider the map @xmath578 from @xmath577 to itself .
it follows from our discussion so far , that the pseudoholomorphic curves which are counted in the definition of @xmath579 , are exactly the ones that appear in the boundary of the moduli spaces @xmath540 . as the compactified moduli space
@xmath540 is a finite union of compact intervals it has an even number of boundary components .
it then follows that @xmath580 .
as this is valid for all reeb chords @xmath309 we have proved that @xmath436 is chain - homotopic to the identity , as claimed .
one very important consequence of proposition [ propchainhom ] is that the strip legendrian contact homology @xmath506 does not depend on the regular almost complex structure @xmath581 used to define @xmath357 , something which is obvious from the definition of @xmath428 .
let @xmath7 be a contact manifold . given a contact form @xmath25 on @xmath7 adapted to a pair of disjoint legendrian knots @xmath19 and
@xmath20 we define the exponential homotopical growth of the strip legendrian contact homology @xmath92 with respect to the action .
we then use it to estimate the growth of the number of reeb chords from @xmath19 to @xmath66 for other contact forms associated to @xmath7 , if @xmath66 is a legendrian curve sufficiently close to @xmath20 .
we start defining for each number @xmath582 the subset @xmath583 of @xmath417 of homotopy classes @xmath420 satisfying : * all the chords in @xmath419 have action @xmath584 , * @xmath585 . [ defigrowth ]
let @xmath25 be a contact form on @xmath7 adapted to a pair of disjoint legendrian knots @xmath19 and @xmath20 .
we say that @xmath92 has exponential homotopical growth with exponential weight @xmath43 if there exist real numbers @xmath100 and @xmath586 such that : @xmath587 for all @xmath588 .
let @xmath19 and @xmath66 be legendrian submanifolds of a contact manifold @xmath7 and @xmath2 be a contact form on @xmath7 : we say that @xmath66 is @xmath589-transverse if all the reeb chords in @xmath590 are transverse .
in this case we will denote by @xmath591 the number of reeb chords in @xmath590 with action @xmath584 .
we will start by presenting a result which shows how to estimate the growth of @xmath591 using the strip legendrian contact homology .
we consider the space of @xmath592 of @xmath62 diffeomorphisms of @xmath4 .
we fix in this space a metric which generates the canonical topology in @xmath592 . from now on , when we say that two diffeomorphisms of @xmath4 are @xmath593-close , we mean close with respect to this fixed metric .
likewise , we fix a metric on the space @xmath594 of @xmath595-smooth loops in @xmath4 that generates the canonical topology in this space . by the weinstein tubular neighbourhood theorem for legendrian submanifolds ( @xcite ) ,
every legendrian knot @xmath20 in a contact 3-manifold @xmath7 has a tubular neighbourhood @xmath596 such that @xmath597 admits a contactomorphism to the local model @xmath598 , where @xmath599 and @xmath600 belongs to the 2-dimensional unit disk @xmath101 , which takes @xmath20 to the core circle @xmath601 of @xmath602 .
we denote by @xmath105 the legendrian submanifolds obtained by fixing the coordinate @xmath259 of @xmath101 . as the @xmath105 form a fibration of @xmath93 by legendrian knots we will refer to them as legendrian fibers of @xmath597 .
we choose the neighbourhood @xmath93 to be small enough so that all legendrian fibers @xmath603 are @xmath593-close to @xmath20 in the @xmath595 topology .
we will also assume that @xmath93 is disjoint from @xmath19 .
we now show how to construct an appropriate exact lagrangian cobordism from @xmath20 to the legendrian fibers @xmath105 .
it is clear that @xmath105 is legendrian isotopic to @xmath20 .
this isotopy can be explicitly obtained by taking a path of legendrian fibers which starts at @xmath105 and finishes at @xmath20 .
we take for each @xmath604 a smooth path @xmath605 ( @xmath606 ) such that @xmath607 for @xmath608 and @xmath609 for @xmath610 .
we then take a path of diffeomorphisms @xmath611 , which equals the identity outside a small neighbourhood of @xmath20 that does not intersect @xmath19 and such that @xmath612 , @xmath613 for @xmath614 and @xmath615 is independent of @xmath290 for @xmath616 .
it is clear that there exists a continuous function @xmath617 which is positive for @xmath184 and satisfies @xmath618 , such that if we take the neighbourhood @xmath93 to be small enough so that all legendrian fibers @xmath105 are @xmath593-close to @xmath20 , we can construct the isotopies @xmath615 satisfying @xmath619 and @xmath620 for @xmath621 , for all @xmath622 and @xmath623 .
in other words , we have an uniform control on both the distance of @xmath624 to the identity map @xmath625 and the speed of change of @xmath615 in time , provided that we restrict our attention to a sufficiently small neighbourhood of @xmath20 .
we will use the maps @xmath624 to construct an exact lagrangian cobordism from @xmath20 to the legendrian fiber @xmath105 . for this
we define the smooth family of contact forms @xmath626 .
notice that the reeb vector fields of these contact forms have the same dynamics as the reeb vector field @xmath372 .
however for the contact form @xmath627 the curve @xmath105 is represented by @xmath20 .
more precisely , we have a bijection between the sets of reeb chords @xmath628 and @xmath629 which preserves the actions of the elements , since the diffeomorphism @xmath615 takes elements of @xmath628 to @xmath629 with the same action .
this follows from the relation @xmath630 between the reeb flows of @xmath627 and @xmath25 .
let now @xmath631 be an increasing function satisfying : @xmath632\end{aligned}\ ] ] if we choose @xmath593 ( and consequently the neighbourhood @xmath93 ) sufficiently small the constant @xmath617 will be as small as we like .
a direct computation then shows that if @xmath617 is sufficiently small the 2-form @xmath633 is an exact symplectic form on the manifold @xmath117 .
this defines an exact symplectic cobordism from @xmath627 to @xmath634 , with the property that @xmath635 is an exact lagrangian submanifold of @xmath636 .
performing an essentially identical construction , we can produce an exact symplectic cobordism @xmath637 from @xmath638 to @xmath627 , such that @xmath635 is an exact lagrangian submanifold of @xmath637 . by concatenating ( @xcite ) the symplectic cobordisms constructed above
, we have established that given @xmath639 , there exists @xmath184 such that for every legendrian fiber @xmath105 in @xmath93 , we can construct an exact symplectic cobordism from @xmath640 to @xmath634 , that coincides with a piece of the symplectization of @xmath627 in @xmath641 \times y$ ] , and such that @xmath635 and @xmath642 are exact lagrangian submanifolds of this exact symplectic cobordism .
[ mainproposition ] let @xmath7 be a contact 3-dimensional manifold and @xmath19 and @xmath20 be two disjoint legendrian knots on @xmath7 .
suppose that @xmath25 is a contact form on @xmath7 adapted to the pair @xmath18 , and that the strip contact homology @xmath92 has exponential homotopical growth with exponential weight @xmath43 .
let @xmath2 be another contact form associated to @xmath7 , and denote by @xmath643 to be the positive function such that @xmath644 .
then , given @xmath639 there exists a tubular neighbourhood @xmath93 of @xmath20 such that for any every legendrian fiber @xmath105 in @xmath93 which is @xmath589 transverse the numbers @xmath104 satisfy @xmath645 for all @xmath113 , where @xmath646 is the constant in definition [ defigrowth ] .
remark : notice that we do not ask the contact form @xmath2 to be non - degenerate .
_ proof : _ we divide the proof into four steps .
_ step 1 : _ reduction to the non - degenerate case .
we first show that it suffices to prove the estimate for non - degenerate contact forms .
the general case follows from this one by an approximation procedure .
let @xmath647 be a natural number . as @xmath105 is @xmath589 transverse
, it is possible to make a @xmath44 small perturbation of the contact form @xmath2 to a non - degenerate contact form @xmath648 on @xmath7 such that : * @xmath649 , for all @xmath650 , * and @xmath105 is @xmath651 transverse .
we demand that the perturbations @xmath648 are taken to be close enough to @xmath2 to guarantee that there are exact symplectic cobordisms from @xmath2 to @xmath652 and from @xmath648 to @xmath653 .
supposing now that the proposition is true for all @xmath648 , we obtain by the conditions satisfied by the @xmath648 that if @xmath113 and @xmath654 it follows that : @xmath655 where @xmath656 is the positive function such that @xmath657 . letting @xmath658 we have that @xmath659 , which implies the desired inequality for @xmath2 .
_ step 2 : we construct a sequence of different symplectic cobordisms .
_ first , we construct a symplectic cobordism from @xmath660 to @xmath2 . for any @xmath661 we pick a function @xmath662 such that @xmath663 , @xmath664 for @xmath665 and @xmath666 for @xmath614 .
since @xmath667 is a positive constant , the reeb flow of @xmath668 is just a reparametrization of the reeb flow of @xmath2 .
the exact symplectic manifold @xmath669 is the desired symplectic cobordism . by an analogous construction
we obtain a symplectic cobordism from @xmath2 to @xmath670 for a sufficiently small constant @xmath671 .
notice that in both constructions @xmath635 and @xmath642 are exact lagrangian submanifolds .
wee can apply the same idea to obtain an exact symplectic cobordism from @xmath672 to @xmath673 , and an exact symplectic cobordism from @xmath673 to @xmath674 .
again , the cylinders @xmath635 and @xmath642 are exact lagrangian submanifolds in both these cobordisms .
we enumerate the exact symplectic cobordisms we have constructed so far , recalling that all of them are diffeomorphic to @xmath496 . * by the construction before the proposition [ mainproposition ] , we have an exact simplectic cobordism @xmath675 from @xmath676 to @xmath677 .
* by the construction in this step we have an exact simplectic cobordism @xmath678 from @xmath679 to @xmath680 .
* again , by the construction in this step we have an exact simplectic cobordism @xmath681 from @xmath682 to @xmath683 . * finally , by the construction made before the proposition , we know that there is an exact simplectic cobordism @xmath684 from @xmath674 to @xmath685 .
moreover such exact symplectic cobordisms are constructed in such a way that for all of them , the cylinders @xmath635 and @xmath642 are exact lagrangian submanifolds .
_ step 3 : the exact lagrangian cobordism and the chain map on the strip legendrian contact homology of the ends . _ by performing a concatenation of the four exact symplectic cobordisms above ( see @xcite for the definition of the concatenation of symplectic cobordisms ) we can produce an exact symplectic cobordism @xmath686 such that : * @xmath687 in @xmath688 and @xmath689 in @xmath690 , * @xmath691 in @xmath692\times y$ ] , * @xmath635 and @xmath642 are exact lagrangian submanifolds in @xmath686 .
following section [ splitting ] we produce the splitting family of symplectic cobordisms @xmath693 for @xmath162 .
it is clear from our construction that for all @xmath694 the symplectic cobordims @xmath693 can be homotoped through exact symplectic cobordisms to the symplectization of @xmath25 , in such a way that @xmath642 and @xmath635 are exact lagrangian submanifolds at every stage of the homotopy . as a consequence , we can apply proposition [ propchainhom ] to obtain that for regular almost complex structures @xmath695 in the cobordisms @xmath693 we get an induced isomorphism @xmath436 from @xmath506 to itself .
_ step 4 : proof of the theorem for @xmath2 non - degenerate .
_ we first pick for @xmath693 a splitting almost complex structure @xmath447 as in section [ splitting ] and take @xmath696 .
we claim that for such an almost complex structure , there exist reeb chords @xmath697 such that @xmath698 is non - empty .
we argue by contradiction : if no such strip exists we obtain that @xmath512 and it follows from proposition [ propchainhom ] that @xmath693 induces an isomorphism on @xmath506 . because @xmath699 is empty for all reeb chords @xmath697 , we conclude that the cobordism map @xmath436 is the zero map . on the other hand , as @xmath700 and @xmath436 is an isomorphism it can not be the zero map . we have thus reached a contradiction and therefore concluded that there must exist reeb chords @xmath697 such that @xmath698 is non - empty .
we send @xmath701 and take a sequence @xmath702 . because there is a global bound on the energy of all elements of @xmath703 ,
the sft - compactness results of @xcite imply that @xmath704 has a subsequence that converges to a pseudoholomorphic building @xmath225 .
because of the stretching the neck process , we have that one of the levels of this building lives in an exact symplectic cobordism from @xmath705 to @xmath706 .
we will see that for topological reasons one of the punctures of this level has to detect a reeb chord @xmath707 with action smaller than @xmath708 .
let @xmath709 for @xmath710 be the levels of the pseudoholomorphic building @xmath225 . because the topology of our pseudoholomorphic curve does not change on the breaking we must have the following picture . * the upper level @xmath396 is composed of one pseudoholomorphic disc , with one positive puncture , which is asymptotic to a reeb chord @xmath711 , and several negative boundary and interior punctures .
all of the negative punctures detect contractible reeb orbits , reeb chords from @xmath19 to itself that represent the trivial element of @xmath305 , or reeb chords from @xmath20 that represent the trivial element of @xmath306 , with the exception of one negative boundary puncture that detects a reeb chord @xmath333 which belongs : either to @xmath712 in case this level lives in the symplectization of @xmath25 , or to @xmath713 in case this level lives in a cobordism from @xmath25 to @xmath705 . * on every other level @xmath709 there is a special curve which has one positive puncture , which is asymptotic to a reeb chord @xmath714 from @xmath19 to @xmath20 in the homotopy class @xmath420 and possibly several interior and boundary negative punctures . of the negative boundary punctures there is one that is asymptotic to a reeb orbit @xmath715 from @xmath19 to @xmath20 in the homotopy class @xmath420 , and all other negative punctures detect contractible reeb orbits , reeb chords from @xmath19 to itself that represent the trivial element of @xmath305 , or reeb chords from @xmath20 that represent the trivial element of @xmath306 . as a consequence
we obtain that the level @xmath709 living in the exact symplectic cobordism from @xmath705 to @xmath706 contains a pseudoholomorphic curve with one positive puncture asymptotic to a reeb chord @xmath707 . the bound on the action of @xmath716 follows from the fact that all punctures in the pseudoholomorphic building @xmath225 detect reeb orbits and chords with action smaller than the action of @xmath717 , that is @xmath718 . as a consequence of this existence result for all our homotopy classes @xmath696
, we obtain that @xmath719 .
because of the action preserving bijection between the sets @xmath713 and @xmath720 it follows that @xmath721 . as @xmath722 can be chosen arbitrarily small , we obtain the promised estimate for @xmath2 , @xmath20 and @xmath105 . using the results of the previous subsection
, we prove theorem [ theorem1 ] . for this
, we first study reeb chords from a fixed legendrian knot @xmath19 to generic legendrian fibers @xmath603 of a tubular neighbourhood @xmath723 of a legendrian knot @xmath20 .
in the disc @xmath101 we will consider the lebesgue measure , which is obtained from restricting the lebesgue on @xmath724 to @xmath101 .
therefore , when we say for almost every @xmath604 we mean almost every with respect to this measure . [ generic ] let @xmath2 be a contact form associated to a contact 3-manifold @xmath7 , and @xmath19 and @xmath20 be disjoint legendrian knots on @xmath7 .
we consider a tubular neighbourhood @xmath723 of @xmath20 such that @xmath725 admits a contactomorphism to the local model @xmath598 , where @xmath599 and @xmath726 , which takes @xmath20 to the core circle @xmath601 of @xmath602 .
then , for almost every @xmath604 the legendrian knot @xmath603 is @xmath589 transverse . _
proof : _ taking a parametrization @xmath727 , we use the reeb flow of @xmath2 to define a map @xmath728 by : @xmath729 .
the set @xmath730 is an open subset of @xmath731 . using the coordinates @xmath732 on @xmath723 given by the contactmorphism from @xmath725 to @xmath598 considered in the statement of the lemma , we define the map @xmath733 to be the projection to the @xmath259 coordinate .
composing @xmath734 with @xmath735 to obtain a smooth map : @xmath736 .
a point @xmath737 is a critical point of @xmath738 if , and only if , the closed curve @xmath739 obtained by restricting @xmath740 to @xmath741 is tangent to a legendrian fiber @xmath603 at @xmath742 . to see this
, note that by the definition of @xmath740 we have @xmath743 and @xmath744 .
this implies that @xmath745 is always non - zero , because the vector field @xmath746 is never tangent to the legendrian fibers @xmath603 , and that @xmath747 if , and only if , @xmath748 is tangent to a legendrian fiber .
it follows that the regular values of @xmath749 in @xmath101 are in bijection with the set of legendrian fibers @xmath603 that are @xmath589 transverse .
applying sard s theorem to @xmath749 we conclude that almost every element of @xmath101 is a regular value of @xmath738 , from which the lemma follows .
we are now ready to prove the main result of this section .
[ theorem1 ] let @xmath41 be a contact 3-manifold , and @xmath25 be a hypertight contact form on @xmath7 adapted to the pair of disjoint legendrian knots @xmath18 .
assume that the strip legendrian contact homology @xmath42 has exponential homotopical growth rate with exponential weight @xmath43 .
then , the reeb flow of any @xmath44 contact form @xmath2 on @xmath7 has positive topological entropy .
moreover , if we denote by @xmath45 the positive function such that @xmath46 we have @xmath750 _ proof : _ for @xmath639 , we take a tubular neighbourhood @xmath93 as in proposition [ mainproposition ] .
it follows from lemma [ generic ] that there exists a set @xmath751 of full measure in @xmath101 such that for every @xmath752 the legendrian @xmath105 is @xmath589-transverse .
this combined with proposition [ mainproposition ] implies that the number @xmath753 of reeb chords of @xmath304 from @xmath19 to the legendrian fiber @xmath603 satisfies : @xmath754 now choose a riemannian metric on the manifold @xmath4 which restricts in @xmath93 to the euclidean metric @xmath755 , for the coordinates @xmath756 on @xmath93 used in proposition [ mainproposition ] .
this riemannian metric induces a measure of area @xmath757 for all surfaces @xmath758 immersed in @xmath4 .
we want to estimate the area @xmath759 of the immersed cylinder @xmath760\}$ ] .
this is the image of the map @xmath761 \to y$ ] defined by @xmath762 .
defining @xmath763 we have : @xmath764 where the last area is taken with multiplicities and with respect to the lebesgue measure in @xmath101 .
the inequality on the right side is true because the measure of area induced by our riemannian metric on @xmath93 coincides with the measure of area induced by the euclidean metric @xmath755 . from the estimate for the counting function @xmath104 for almost every @xmath726
, we get : @xmath765 it follows that @xmath766 combining this inequality with the fubini type estimate proved in appendix [ appendixa ] we have : @xmath767 where @xmath768 is the length of the curve @xmath769 with respect to the riemannian metric we chose .
we now invoke yomdin s theorem , that implies that @xmath770 .
since in proposition [ mainproposition ] the constant @xmath639 can be taken arbitrarily small , we obtain inequality .
in this section we prove the following theorem : let @xmath53 be a closed oriented connected 3-manifold which can be cut along a nonempty family of incompressible tori into a family @xmath54 of irreducible manifolds with boundary such that the component @xmath55 satisfies : * @xmath55 is the mapping torus of a diffeomorphism @xmath56 with pseudo - anosov monodromy on a surface @xmath57 with non - empty boundary .
then @xmath53 can be given infinitely many different tight contact structures @xmath58 , such that : * there exist disjoint legendrian knots @xmath19 , @xmath20 on @xmath59 , and a contact form @xmath60 on @xmath59 adapted to the pair @xmath19 and @xmath20 , such that @xmath61 has exponential homotopical growth rate .
it follows that every reeb flow on @xmath59 has positive topological entropy .
this result establishes the existence of many contact 3-manifolds which possess a contact form and a pair of legendrian knots that satisfy the hypothesis of theorem [ theorem1 ] .
the contact manifolds covered in theorem [ theorem2 ] are among the ones constructed by colin and honda @xcite .
we start with some preliminary notions .
let @xmath57 be a compact surface with non - empty boundary that has negative euler characteristic .
let @xmath771 be the number of boundary components of @xmath57 , and let @xmath772 for @xmath773 be the boundary components of @xmath57 .
for each @xmath774 let @xmath775 be a tubular neighbourhood of @xmath772 , which admits coordinates @xmath776\times s^1 $ ] identifying @xmath772 with @xmath777 .
it follows from ( * ? ? ?
* lemma 4.1 ) that there exists a symplectic form @xmath778 on @xmath57 and a liouville is a liouville vector field on @xmath779 if it satisfies @xmath780 where @xmath781 is the lie derivative of @xmath778 by @xmath782 . ]
vector field @xmath782 for @xmath778 such that : * the vector field @xmath782 is morse - smale , * there exist a riemannian metric @xmath27 on @xmath57 and a function @xmath783 satisfying @xmath784 and @xmath785 , such that @xmath782 is the gradient vector field of @xmath786 with respect to @xmath27 , * @xmath782 points outwards along all boundary components of @xmath57 , * the one form @xmath787 satisfies @xmath788 for the coordinates @xmath789 on the neighbourhood @xmath775 for a positive function @xmath790 that satisfies @xmath791 . notice that it follows from these properties that all singularities of @xmath782 have index @xmath328 or @xmath329 . in a pseudo - anosov mapping class
choose a diffeomorphism @xmath792 that preserves @xmath778 and which coincides with the identity map on all neighbourhoods @xmath775 .
we now present a well - known recipe to produce a contact form on the the mapping torus @xmath793 of the map @xmath794 .
we first choose a smooth non - decreasing function @xmath795 $ ] which satisfies @xmath796 for @xmath797 and @xmath798 for @xmath799 .
for @xmath800 we define @xmath801 .
fixing @xmath802 , we define a 1-form @xmath803 on @xmath804 by @xmath805 this defines a smooth 1-form on @xmath804 , and a simple computation shows that the 1-form @xmath803 is a contact form for @xmath593 small enough . for @xmath806 $ ] , the reeb vector field @xmath807 is equal to @xmath808 , where @xmath809 is the unique vector tangent to @xmath57 that satisfies @xmath810 . consider the diffeomorphism @xmath811 defined by @xmath812 .
the mapping torus @xmath793 is defined by @xmath813 and we denote by @xmath814 the associated covering map . because @xmath815 , there exists a unique contact form @xmath17 on @xmath793 such that @xmath816 .
notice that in the neighbourhoods @xmath817 with coordinates @xmath818 of the connected components of @xmath819 , we have the formula @xmath820 for @xmath17 .
this implies that @xmath821 is tangent to @xmath819 .
the reeb vector field @xmath821 on @xmath793 is transverse to the surfaces @xmath822 for @xmath823 .
this implies that for any @xmath824 the surface @xmath822 is a global surface of section for the reeb flow of @xmath17 .
moreover , by our expression for @xmath807 the first return map of the reeb flow of @xmath825 on any of these sections is isotopic to @xmath794 . in the mapping torus
@xmath793 consider the surface @xmath826 .
as mentioned above this surface is a copy of @xmath57 inside @xmath793 , and we denote by @xmath827 the embedding associated to @xmath828 .
we begin with the following lemma .
[ lemmacurve ] there exists an embedded curve @xmath829 contained in the interior of @xmath57 such that @xmath830 , and which is not homotopic to @xmath831 .
moreover , given @xmath99 , we can choose @xmath832 such that @xmath833 for all @xmath834 $ ] with @xmath835 . ] .
_ proof : _ as @xmath782 is a liouville vector field on @xmath779 we know that @xmath836 contracts the area form @xmath778 on @xmath57 .
this implies that @xmath836 has no singularities of source type .
notice that at every point of @xmath57 the vector @xmath836 belongs to the kernel of @xmath837 .
since @xmath836 is a gradient morse - smale vector field , @xmath836 has no closed orbits .
we know from morse theory that the flow of @xmath836 can be used to produce a deformation retraction from @xmath57 to the set @xmath838 defined as the union of the singularities of the flow of @xmath836 and the unstable manifolds of its saddle singularities . because @xmath838 has the same homotopy type as @xmath57 it must contain a simple closed piecewise smooth curve @xmath278 which is contained in the interior of @xmath57 and that can not be homotoped to @xmath831 . because @xmath278 consists of trajectories of @xmath836 , we know that vectors tangent to @xmath278 belong to the kernel of @xmath837 .
we orient @xmath278 using a piecewise smooth parametrisation which we denote also by @xmath839 .
then @xmath840 for all @xmath834 $ ] where again @xmath841 .
fix a point @xmath842 which is not a vertex point .
we can then smoothen @xmath278 in small closed neighbourhoods of the vertices of @xmath278 which are all disjoint from @xmath843 and produce a smooth embedded curve @xmath844 on @xmath845 . as
@xmath844 coincides with @xmath278 in a neighbourhood of @xmath843 , we pick the orientation in @xmath844 which coincides with the orientation chosen for @xmath278 in the region where the two curves coincide . we choose a parametrisation @xmath846 of @xmath844 which induces the same orientation as the one we just defined . given @xmath847
, we can ensure , by doing the smoothing in a sufficiently small neighbourhood of the vertices , that @xmath848 and that for all @xmath834 $ ] we have @xmath849 .
using that near the point @xmath843 the curve @xmath844 is tangent to @xmath850 , we find , for sufficiently small @xmath639 , a @xmath44 small perturbation @xmath851 of the parametrised curve @xmath844 supported in a neighbourhood @xmath852 of @xmath843 such that , for the small region @xmath853 bounded by @xmath851 and @xmath844 , and oriented such that @xmath854 , we have @xmath855 now stokes theorem and the fact that @xmath856 imply that @xmath857 .
this perturbation @xmath851 , which aims at correcting the change in the integral after the smoothing of the corners , can be made explicitly if one uses darboux coordinates in a neighbourhood of @xmath843 , and can be taken so small enough that for all @xmath834 $ ] we have @xmath858 .
taking @xmath829 to be the parametrised curve @xmath851 we have shown that @xmath832 has all the properties stated in the lemma .
let @xmath859 and take the parametrised curve @xmath832 obtained in lemma [ lemmacurve ] . because @xmath860 we have that the curve @xmath832 is the lagrangian projection of a legendrian curve @xmath76 in @xmath861 . to make this precise we first define the function @xmath862 .
we then define the curve @xmath863 by @xmath864 . as @xmath859 , and @xmath865
we conclude that image of @xmath76 is contained in the set @xmath866 \times s $ ] in which the contact form @xmath17 is given by @xmath867 . using this identity
we compute that @xmath868 at every @xmath834 $ ] which implies that @xmath76 is a legendrian curve of @xmath17 . from our construction
it is clear that @xmath76 is a graph over the curve @xmath832 which is embedded in @xmath869 .
we will denote by @xmath19 the legendrian knot @xmath870 in @xmath861 .
we are now ready to prove the following lemma .
[ lemmachords ] if @xmath871 is a legendrian curve in @xmath872 which is sufficiently close to @xmath19 in the @xmath269 topology , then there exists no reeb chord @xmath21 of @xmath17 from @xmath871 to itself such that @xmath873 $ ] is the trivial element in @xmath874 . _
proof : _ if a legendrian @xmath871 is sufficiently close to @xmath19 in the @xmath269 topology then there exists a knot @xmath875 in @xmath876 which is a connected component of @xmath877 , which is also contained in @xmath878 \times s$ ] , and which is also a graph over an embedded curve @xmath879 in @xmath57 .
assume there existed a reeb chord @xmath21 in @xmath871 of @xmath17 from @xmath19 to itself such that @xmath873 $ ] is the trivial element in @xmath874 .
then , the fact the @xmath873 $ ] is the trivial element in @xmath874 would imply that @xmath21 lifts to a reeb chord of @xmath803 from @xmath875 to itself in @xmath876 .
since the reeb vector field of @xmath803 is @xmath808 and @xmath875 is a graph over an embedded curve @xmath879 in @xmath57 , it is clear that there is no reeb chord of @xmath880 from @xmath875 to itself .
this contradiction proves the lemma .
we let @xmath881 be a legendrian knot in @xmath882 obtained as a @xmath44 small perturbation of @xmath76 which is disjoint from @xmath76 and is contained in @xmath878 \times s$ ] , and denote by @xmath20 the legendrian curve @xmath883 in @xmath884 .
choosing the perturbation generically , we can choose @xmath881 so that : * @xmath881 is also the graph over an embedded curve @xmath885 in @xmath57 , * all the reeb chords of @xmath17 from @xmath19 to @xmath20 are transverse .
there is another genericity condition that we want to impose on the reeb chords from @xmath19 to @xmath20 .
we want to guarantee that no reeb orbits of our reeb flow intersect reeb chords from @xmath19 to @xmath20 .
to guarantee this we have to perturb the contact form @xmath17 in the interior of @xmath793 . [ adapted ] there
exists a contact form @xmath886 on @xmath872 which is obtained by perturbing @xmath17 in the interior of @xmath793 and that satisfies : * @xmath886 has no contractible reeb orbits , * all reeb chords of @xmath886 from @xmath19 to itself represent non - trivial elements in @xmath887 , * all reeb chords of @xmath886 from @xmath20 to itself represent non - trivial elements in @xmath888 , * the reeb orbits of @xmath886 do not intersect the reeb chords of @xmath886 from @xmath889 to @xmath20 , * there exist no reeb chords of @xmath890 from @xmath76 to itself , or from @xmath881 to itself . _ proof : _ if @xmath891 is a sufficiently small perturbation of @xmath892 in the interior of @xmath793 then we can guarantee that the @xmath893 component of the vector field @xmath894 is bigger than @xmath895 .
in order to prove the second and third statements we first pick @xmath202 such that the curves @xmath896 and @xmath875 are contained in @xmath897 \times s$ ] .
it follows that any reeb chord of @xmath898 from @xmath896 to itself has action @xmath899 . indeed , for any point @xmath900 the trajectory @xmath901 stays outside @xmath897 \times s$ ] for @xmath902 .
the same is true for @xmath875 .
this implies that any reeb chord of @xmath891 from @xmath19 to itself that represents the trivial element in @xmath887 must have action @xmath899 , the same being true for @xmath20 .
we claim that if @xmath891 is sufficiently close to @xmath17 , then there exists no reeb chord of @xmath891 from @xmath19 to itself that represent the trivial element in @xmath887 , the same being true for @xmath20 . indeed ,
otherwise we could take a sequence of contact forms @xmath903
converging to @xmath904 all of them having reeb chords @xmath905 from @xmath19 to itself that represent the trivial element in @xmath887 with action @xmath899 . since the action of the reeb chords @xmath905 is bounded by @xmath906 , the arzel - ascoli theorem implies that @xmath905 has a convergent subsequence that converges to a reeb chord @xmath21 of @xmath17 from @xmath19 to itself that must be trivial in @xmath887 .
as this contradicts lemma [ lemmachords ] , we have proved our claim for @xmath19 . the claim for @xmath20
is proved in the same way .
the last claim of the lemma is proved by an approximation argument identical to the one used to prove the second and third claims .
we pick @xmath886 to be sufficiently close to @xmath17 , so that there exists no reeb chord of @xmath886 from @xmath19 to itself that represent the trivial element in @xmath887 , and no reeb chord of @xmath886 from @xmath20 to itself that represent the trivial element in @xmath907 .
moreover by choosing our perturbation generically we can choose @xmath886 such that : * the reeb orbits of @xmath886 do not intersect the reeb chords of @xmath886 from @xmath889 to @xmath20 , * all the reeb chords of @xmath886 from @xmath19 to @xmath20 are transverse . to complete the proof of the lemma
we must now show that @xmath886 has no contractible reeb orbits .
but since the @xmath908 coordinate of the vector field @xmath909 is always positive , we know that the flow of @xmath909 has no periodic orbits .
it follows that @xmath886 has no contractible reeb orbits .
we now construct another covering @xmath910 of the mapping torus @xmath793 .
we first define @xmath911 because @xmath828 is a global surface of section for the reeb flow of @xmath912 we conclude that @xmath913 is indeed a covering map . using the homotopy lifting theorem
we know that there exists a unique diffeomorphism @xmath914 that is the identity when restricted to @xmath915 and fits in the commutative diagram \(m ) [ matrix of math nodes , row sep=4em , column sep=3em , minimum width=3em ] s & s + ( s , h ) + ; ( m-1 - 1 ) edge node [ left ] @xmath916 ( m-2 - 1 ) edge node [ above ] @xmath917 ( m-1 - 2 ) ( m-1 - 2 ) edge node [ below right ] @xmath913 ( m-2 - 1 ) ; it is clear that the pullback of @xmath918 by @xmath913 is the vector field @xmath908 , and as observed before this pullback is of the form @xmath919 where @xmath920 is tangent to @xmath57 and @xmath921 is a positive function . because @xmath921 is invariant by the map @xmath922 we know that there exists a constant @xmath923 such that @xmath924 .
from the fact @xmath914 is the identity when restricted to @xmath915 we know that @xmath917 is isotopic to the identity . from the expression defining @xmath913 and
the fact that @xmath917 fits in the above diagram we know that @xmath925 , and therefore that @xmath917 conjugates the flows of @xmath908 and @xmath919 .
this property of @xmath917 and the fact @xmath924 imply that there exist constants @xmath926 and @xmath927 such that @xmath928 where @xmath929 is the projection onto the first coordinate .
we now proceed to construct contact manifolds containing @xmath930 .
let @xmath53 be a closed connected oriented 3-manifold which can be cut along a non - empty family of incompressible tori into a family @xmath54 of irreducible manifolds with boundary , such that the component @xmath55 is diffeomorphic to @xmath793
. then one can construct hypertight contact forms on @xmath53 which match with @xmath17 in the component @xmath55 .
more precisely , we have the following result due to colin and honda , and vaugon : [ colinhonda ] ( @xcite ) let @xmath53 be a closed connected oriented 3-manifold which can be cut along a non - empty family of incompressible tori into a family @xmath54 of irreducible manifolds with boundary , such that the component @xmath55 is diffeomorphic to @xmath793
. then , there exists an infinite family @xmath931 of non - diffeomorphic contact structures on @xmath53 such that * for each @xmath932 there exists a hypertight contact form @xmath60 on @xmath59 which coincides with @xmath886 on the component @xmath55 .
we briefly recall the construction of the contact forms @xmath60 , and refer the reader to @xcite for the details . for @xmath933 ,
we apply ( * ? ? ?
* theorem 1.3 ) to obtain a hypertight contact form @xmath934 on @xmath935 which is compatible with the orientation of @xmath935 , and whose reeb vector field @xmath936 is tangent to the boundary of @xmath935 . on the special piece @xmath55 we consider the contact form @xmath937 equal to @xmath886 constructed above .
let @xmath938 be the family of incompressible tori along which we cut @xmath53 to obtain the pieces @xmath935 .
then the contact forms @xmath934 give a hypertight contact form on each component of @xmath939 , where @xmath940 is a small open neighborhood of @xmath941 .
this gives a contact form @xmath942 on @xmath939 .
using an interpolation process ( see ( * ? ? ?
* section 7 ) ) , one can construct contact forms on the neighborhoods @xmath943 which coincide with @xmath942 on @xmath944 .
the interpolation process is not unique and can be done in many different ways as to produce an infinite family of distinct contact forms @xmath945 on @xmath53 that extend @xmath942 , and which are associated to contact structures @xmath946 that are all non - diffeomorphic . the contact topological invariant used to distinguish the contact structures @xmath58 is the _ giroux torsion _ ( see ( * ? ? ?
* section 7 ) ) . the legendrian curves @xmath19 and @xmath20 in @xmath872 are obviously also legendrian curves of @xmath59 . because the reeb flow of @xmath60 leaves the region @xmath793 invariant , we know that all reeb chords of @xmath60 from @xmath19 to @xmath20 are contained in @xmath793 and are therefore reeb chords of @xmath17 .
we conclude that @xmath20 is @xmath947-transverse .
[ propadapted ] all reeb chords of @xmath60 from @xmath19 to itself represent non - trivial elements in @xmath948 , and likewise all reeb chords of @xmath60 from @xmath20 to itself represent non - trivial elements in @xmath949 .
_ proof : _ we will show that lemma [ adapted ] implies the proposition .
we present the proof of the statement for the legendrian curve @xmath19 .
the proof for @xmath20 is identical . by contradiction
suppose there is a smooth map @xmath950 such that @xmath951 is the concatenation of a reeb chord @xmath21 of @xmath60 from @xmath19 to itself with a path @xmath952 . by genericity
, we can suppose that the map @xmath953 is transverse to @xmath954 .
this implies that @xmath955 is a one - dimensional submanifold of @xmath956 .
because @xmath957 is disjoint from @xmath954 , we conclude that @xmath955 is a finite collection @xmath958 of circles embedded in @xmath959 . each circle
@xmath960 is the boundary of a disk in @xmath961 , which we denote by @xmath962 .
it is clear that we can choose a subset @xmath963 that satisfies * for each @xmath964 there exists a unique @xmath965 such that @xmath966 .
in particular it follows that if @xmath967 are distinct elements of @xmath968 , the disks @xmath969 and @xmath970 are disjoint . from the definition of the set @xmath968 , we know that @xmath971 is contained in @xmath55 . for each @xmath972
the circle @xmath973 is a contractible curve in @xmath53 contained in @xmath954 . as @xmath954 is an incompressible torus in @xmath53 , we conclude that the @xmath973 are also contractible in @xmath954 . using this we define for each @xmath972 a map : * @xmath974 which satisfies @xmath975 for every @xmath976 . combining this with the fact that @xmath971 is contained in @xmath55
, we define a continuous map @xmath977 by : * @xmath978 if @xmath979 , * @xmath980 if @xmath981 .
it is clear that the existence of the map @xmath982 implies that the reeb chord @xmath21 represents the trivial element in @xmath983 , in contradiction to lemma [ lemmachords ] .
we have the following important corollary of proposition [ propadapted ] .
the contact form @xmath60 on @xmath59 is adapted to the disjoint legendrian knots @xmath889 and @xmath20 .
_ proof : _ we have to show that the triple @xmath984 satisfies conditions ( a ) , ( b ) , ( c ) and ( d ) from section [ sectionstrip ] . condition ( a ) follows from proposition [ colinhonda ] which says that the contact form @xmath60 is hypertight . conditions ( b ) , ( c ) and ( d ) follow from proposition [ propadapted ] . to study the growth rate of @xmath985 we will consider some special relative homotopy classes of paths from @xmath19 to @xmath20 .
[ defrelnielsen ] let @xmath333 and @xmath334 be reeb chords from @xmath19 to @xmath20 .
we say that @xmath333 and @xmath334 are in the same * relative nielsen class * if , and only if , there exists a smooth strip @xmath986\times[0,1 ] \to \omega(s , h ) $ ] such that : * @xmath987)$ ] is a path in @xmath19 and @xmath988)$ ] is a path in @xmath20 , * @xmath989 \times { 0})= c_1 $ ] and @xmath989 \times { 1})= c_2 $ ] .
relative nielsen classes are just homotopy classes of paths from @xmath19 to @xmath20 in the mapping torus @xmath990 .
our first step is to prove that the relative nielsen classes generate a partition of @xmath991 in subcomplexes because they can be regarded as elements in the set @xmath417 of homotopy classes of paths from @xmath19 to @xmath20 in @xmath53 .
[ relnielsen ] let @xmath333 and @xmath334 be reeb chords from @xmath19 to @xmath20 , and @xmath986\times[0,1 ] \to m$ ] such that : * @xmath992)$ ] is a path in @xmath19 and @xmath993)$ ] is a path in @xmath20 , * @xmath989 \times \{0\})=$ ] is a parametrisation of @xmath333 and @xmath989 \times \{1\})$ ] is a parametrisation of @xmath994 .
then , there exists a strip @xmath995\times[0,1 ] \to \omega(s , h)$ ] such that @xmath996\times[0,1 ] ) ) = \mathfrak{e}(\partial([0,1]\times[0,1]))$ ] . _
proof : _ the proof is very similar to the one of proposition [ propadapted ] above , so we will only give an outline . by genericity we can assume that the map of @xmath997 is transverse to @xmath998 .
the set @xmath999 consists of a finite collection of circles @xmath1000, ... ,@xmath1001 . for each @xmath1002
the circle @xmath1003 is contractible .
the assumption that @xmath998 is incompressible implies that the circles @xmath1004, ... ,@xmath1005 are also contractible in @xmath998 .
each circle @xmath1006 bounds a disk @xmath1007 in @xmath101 .
we can choose a subset @xmath1008 such that : * for every @xmath1009 there exists @xmath1010 such that@xmath1011 , * if @xmath1012 are both elements in @xmath1013 then @xmath1007 and @xmath1014 do not intersect . by a cut and paste procedure identical to the one used in proposition [ propadapted ] we can cut off the discs @xmath1015 for each @xmath1010 , and replace them by other discs to obtain the map @xmath1016\times[0,1 ] \to \omega(s , h)$ ] which coincides with @xmath997 outside the disks @xmath1007 .
the map @xmath1017 is the desired one .
recall from section [ sectionstrip ] that the differential @xmath1018 of the strip legendrian contact homology @xmath985 counts pseudoholomorphic strips @xmath1019 \to \mathbb{r } \times m$ ] in the symplectization of @xmath60 with fredholm index @xmath329 satisfying the boundary conditions : * @xmath1020 , * @xmath1021 .
it follows from lemma [ relnielsen ] that the relative nielsen classes can be seen as elements in @xmath417 .
more precisely , denoting by @xmath1022 the set of relative nielsen classes , we have a map @xmath1023 , defined as follows : given a relative nielsen class @xmath420 , we pick a reeb chord @xmath1024 and define @xmath1025 to be the class of @xmath873 \in \sigma_{\lambda \to \widehat{\lambda}}$ ] .
it is easy to see that @xmath1026 is well defined and the lemma [ relnielsen ] implies that @xmath1026 is injective . since @xmath60 coincides with @xmath886 on @xmath55 we know that @xmath55 is invariant by the reeb flow of @xmath60 .
it follows that all the reeb chords in @xmath1027 are contained in the component @xmath1028 , and therefore belong to elements in @xmath417 which are in the image of our map @xmath1026 .
we can therefore write @xmath985 as a direct sum : @xmath1029 we will use the coverings @xmath910 and @xmath1030 to obtain information about the relative nielsen classes .
the curve @xmath76 constructed in section [ sectionlegendrian ] is a lift of @xmath19 for the covering @xmath1030 .
we then fix in @xmath804 the lift @xmath1031 of @xmath19 for the covering @xmath910 .
we want to describe the set @xmath1032 of lifts of @xmath20 to @xmath804 via @xmath913 .
we begin by fixing the lift @xmath881 of @xmath20 for @xmath916 constructed in section [ sectionlegendrian ] .
we know that @xmath1033 it follows that @xmath1034 we define @xmath1035 . given a reeb chord @xmath21 from @xmath19 to @xmath20 we define by @xmath1036 the unique lift of @xmath21 via the covering map @xmath913 which has its starting point in @xmath1037 .
it is not difficult to see that if @xmath333 and @xmath334 are reeb chords from @xmath19 to @xmath20 that are in the same relative nielsen class , then @xmath1038 and @xmath1039 have to have endpoints in the same lift @xmath1040 of @xmath20 .
we will see , however , that this condition is far from sufficient to guarantee that @xmath333 and @xmath334 are in the same relative nielsen class .
we claim that every @xmath1040 is a graph over an embedded curve @xmath1041 of @xmath57 .
if this was not the case , it would follow that there exists a trajectory of the vector field @xmath908 which starts and ends in @xmath1040 .
applying the map @xmath1042 to this trajectory we would obtain a trajectory of the reeb flow of @xmath803 that starts and ends in @xmath76 , something that lemma [ adapted ] tells us does not exist . from lemma [ adapted ] we also know that @xmath1037 is a graph over an embedded curve @xmath1043 in @xmath57 . [ above ] we claim that there exists @xmath1044 such that if @xmath1045 then the curve @xmath1040 is above @xmath1046 in the sense that for all @xmath1047 and @xmath1048 we have @xmath1049 to see this let @xmath202 be such that @xmath1050 \times s$ ] and @xmath1051 \times s$ ] .
we then know that @xmath1052 \times s$ ] . combining this
with we see that there exists @xmath1044 such that if @xmath1045 then @xmath1053 .
let @xmath1054 be the projection on the second coordinate .
the curves @xmath1043 and @xmath1055 are embedded curves in @xmath57 . because @xmath1056 we know that if @xmath1057 the reeb chords from @xmath76 to @xmath1040 are in one - to - one correspondence with the intersection points of @xmath1043 and @xmath1041 . by the same reason we know that the transversality of all the reeb chords from @xmath19 to @xmath20 implies the transversality of @xmath1043 and @xmath1041 for every @xmath1057 .
we now proceed with the following characterization of the relative nielsen classes .
[ nielsen ] let @xmath333 and @xmath334 be reeb chords in @xmath1058 with @xmath1059 and @xmath1060 .
then @xmath333 and @xmath334 are in the same relative nielsen class if , and only if , @xmath1061 and @xmath1062 have end points in the same @xmath1040 , and there exists a map @xmath1063 \times [ 0,1 ] \to s$ ] , such that : @xmath1064 \times \{0\})= p_1 , \\
v([0,1 ] \times \{1\})= p_2 , \\
v(\{0\ } \times [ 0,1 ] ) \subset \eta_0 , \\
v(\{1\ } \times [ 0,1 ] ) \subset \widehat{\eta}_n.\end{aligned}\ ] ] _ proof : _ suppose @xmath333 and @xmath334 are in the same relative nielsen class .
we take the map @xmath1065 \times [ 0,1 ] \to \omega(s , h)$ ] given in definition [ defrelnielsen ] , and consider its lift @xmath1066 \times [ 0,1 ] \to \mathbb{r } \times s$ ] to @xmath876 via @xmath913 , that satisfies @xmath1067 \times \{0\})= \widetilde{c}_1 $ ] and @xmath1067 \times \{1\})= \widetilde{c}_2 $ ] .
it is easy to see that taking @xmath1068 gives a strip in @xmath57 satisfying the conditions in the statement of the proposition .
this proves one implication . to prove the reverse implication
take a map @xmath1069 \times [ 0,1 ] \to s$ ] satisfying the conditions in the statement of the proposition . by taking the path
@xmath1070)\subset \eta_0 $ ] there exists a unique function @xmath1071 \to \mathbb{r}$ ] such that the path @xmath1072 defined by @xmath1073 is a path in @xmath1037 .
analogously there exists a function @xmath1074 \to \mathbb{r}$ ] such that @xmath1075 is a path in @xmath1040 .
take @xmath1076 \times [ 0,1 ] \to \mathbb{r}$ ] to be an homotopy between @xmath110 and @xmath1077 that is , @xmath1078 and @xmath1079 .
now we can define the strip @xmath1080 in @xmath804 , and considering @xmath1081 we obtain a strip in @xmath793 which satisfies the conditions of the definition of relative nielsen classes for @xmath333 and @xmath334 .
this finishes the second implication and the proof of the proposition .
proposition [ nielsen ] gives a complete description of the relative nielsen classes .
it also shows how to identify different relative nielsen classes of reeb chords by looking at properties of intersection points of the curves @xmath1041 and @xmath1043 .
this is the crucial link that will allow us to use the pseudo - anosov monodromy of @xmath1082 to estimate the growth of the number of relative nielsen classes . among the relative nielsen classes ,
the subset of relative nielsen classes with an odd number of chords will be of special importance to us .
we will call them fundamental relative nielsen classes and denote their set by @xmath1083 . by the discussion above we can partition the set @xmath1083 in subsets @xmath1084 defined by the following rule : @xmath1085 if , and only if , for every reeb chord @xmath1086 , the lift @xmath1036 has its endpoint in @xmath1087 .
our next step will be to estimate the cardinality of @xmath1084 .
we start by noticing that @xmath1088 .
since @xmath917 is isotopic to the identity we conclude that the curve @xmath1089 is isotopic to @xmath1090 in the space of embedded curves in @xmath57 .
[ remarkgrowth ] recall from ( * ? ? ?
* expos 1 ) that for isotopy classes @xmath1091 and @xmath1092 of embedded closed curves in @xmath57 we define @xmath1093 to be the minimal number of intersection points of curves @xmath1091 and @xmath1092 .
because @xmath794 has pseudo - anosov monodromy and the curves @xmath832 and @xmath885 are not homotopic to boundary components of @xmath57 , it follows from ( * ? ? ?
* expos 1 and expos 11 ) that there exists real numbers @xmath1094 , @xmath1095 and @xmath1096 such that @xmath1097,[h^{-n}(\widehat{\eta } ) ] ) \geq e^{\overline{a}n+\overline{b}},\ ] ] for every @xmath1098 , where @xmath1099 $ ] denotes the isotopy class of curves in @xmath57 isotopic to @xmath832 , and @xmath1100 $ ] denotes the isotopy class of curves in @xmath57 isotopic to @xmath1100 $ ] .
since @xmath1043 is isotopic to @xmath832 , and @xmath1041 is isotopic to @xmath1090 , @xmath1101,[\widehat{\eta}_n ] ) = \mathfrak{int}([\eta],[h^{-n}(\widehat{\eta})]).\ ] ] we are now ready to prove the main result of this section . there exist numbers @xmath1094 , @xmath1095 and @xmath1096 such that @xmath1102 for every @xmath1103 . _ proof : _ we endow @xmath57 with a hyperbolic metric @xmath27 having @xmath831 as a geodesic boundary . since @xmath1043 is a simple closed curve , ( * ? ? ?
* lemma 2.6 ) implies that there is a homeomorphism @xmath1104 homotopic to the identity for which @xmath1105 is a geodesic of the hyperbolic metric @xmath27 .
as @xmath1106 is an embedded closed curve in @xmath57 it admits an isotopy to an embedded hyperbolic geodesic @xmath278 .
the number of intersection points of @xmath278 and @xmath1107 is equal to the number @xmath1108,[\psi(\widehat{\eta}_n)])=\mathfrak{int}([\eta_0],[\widehat{\eta}_n])=\mathfrak{int}([\eta],[h^{-n}(\widehat{\eta})])$ ] , because pairs of hyperbolic geodesics realise the minimal number of intersection points in their isotopy classes .
we denote by @xmath1109 the set of the intersection points of @xmath278 and @xmath1105 .
there exists a subset @xmath1110 of the poincar disc @xmath1111 which is the universal cover of @xmath1112 .
let @xmath1113 the covering map .
given an embedded closed curve @xmath1092 in @xmath57 which is not homologous to @xmath1105 , let @xmath1114 be a lift of @xmath1092 in @xmath1110 , and take a closed subinterval @xmath1115 of @xmath1114 such that @xmath1116 and that covers every point @xmath1117 of @xmath278 exactly once , in the sense that the intersection of @xmath1118 and @xmath1115 has one element .
we call @xmath1115 a fundamental interval of @xmath1092 . from now on
we suppose @xmath1098 so that @xmath278 and @xmath1106 are not homologous to @xmath1105 .
consider a lift @xmath1119 in @xmath101 , and take a fundamental interval @xmath1120 of @xmath1119 . because of the convexity of the hyperbolic metric , we know that @xmath1119 can not intersect a lift of @xmath1105 more than once .
therefore , @xmath1120 intersects exactly @xmath1121 different lifts @xmath1122 of @xmath1105 .
denote by @xmath1123 an isotopy between @xmath278 and @xmath1106 where @xmath1124 $ ] , and @xmath1125 and @xmath1126 . using the isotopy @xmath1123
, we can construct a path @xmath1127 of fundamental intervals of the curves @xmath1123 .
this generates an isotopy of @xmath1120 to a fundamental interval @xmath1128 of @xmath1106 through fundamental intervals of @xmath1123 .
it follows from the properties of fundamental intervals that @xmath1129 is disjoint from @xmath1105 for all @xmath573 $ ] .
we then conclude that @xmath1128 must also intersect the same @xmath1130 different lifts @xmath1122 of @xmath1105 that are intersected by @xmath1120 , though it may intersect also others lifts of @xmath1106 .
the set @xmath1131 of intersection points of @xmath1043 and @xmath1041 is in bijective correspondence with the set @xmath1132 of intersection points of @xmath1105 and @xmath1106 . because of the properties of fundamental intervals
, there also exists a bijection between the set @xmath1132 and the set @xmath1133 of intersection points of @xmath1128 with the geodesics @xmath1122 .
there exists then a bijection @xmath1134 . as observed above
, the set @xmath1131 is in bijective correspondence with the set of reeb chords of @xmath1135 from @xmath76 to @xmath1040 .
taking now @xmath1136 , we claim that there is a strip @xmath782 satisfying the four conditions of proposition [ nielsen ] if , and only if , @xmath1137 and @xmath1138 lie in the same @xmath1139 . to prove one direction of the claim notice that if there exists such a strip @xmath782 then we can take a lift @xmath1140 of @xmath1141 in the universal cover @xmath1110 . by the boundary conditions that are satisfied by @xmath1140 combined with the fact that @xmath1105 and @xmath1106 are embedded curves in @xmath57
, it follows that @xmath1137 and @xmath1138 have to lie in the same @xmath1139 . for the reverse direction of the claim we remark
that if @xmath1137 and @xmath1138 lie in the same @xmath1139 we can construct a strip @xmath1140 satisfying @xmath1142 \times \{0\})= \varphi(p_1)$ ] , @xmath1142 \times \{1\})= \varphi(p_2)$ ] , @xmath1143 ) \subset \kappa_j$ ] and @xmath1144 ) \subset \mathfrak{i}_n$ ] , where @xmath1145 denotes the lift of @xmath1106 that contains @xmath1128 .
taking @xmath1146 we obtain the desired strip satisfying the conditions of proposition [ nielsen ] . combining the last paragraph and proposition [ nielsen ]
we conclude that to each @xmath1139 is associated a different relative nielsen class @xmath1147 in @xmath1148 , and that the intersection points between @xmath1128 and @xmath1139 are in bijective correspondence with the reeb chords in @xmath1149 .
it follows that there are at least @xmath1130 different relative nielsen classes in @xmath1148 .
we must now prove that @xmath1147 is a fundamental relative nielsen class .
we know that @xmath1150 intersects @xmath1139 an odd number of times , and we have an isotopy @xmath1127 between @xmath1120 and @xmath1128 such that @xmath1151 never intersects @xmath1139 . since @xmath1150 and @xmath1128 are both transversal to @xmath1139 , we conclude that @xmath1128 also has to intersect @xmath1139 an odd number of times . combining this with the observation in the last paragraph
we conclude that @xmath1147 is a fundamental relative nielsen class .
thus , there are at least @xmath1130 different fundamental relative nielsen classes in @xmath1148 . taking the numbers @xmath1094 , @xmath1095 in remark [ remarkgrowth ] , and defining @xmath1152 for the numbers @xmath1044 in remark [ above ] and @xmath1096 in remark [ remarkgrowth ]
, it follows that if @xmath1103 @xmath1153 since @xmath1154 , we have proved the theorem .
we have now all the ingredients needed to establish the exponential homotopical growth rate of @xmath985 .
let @xmath202 be a number such that @xmath1155 \times s$ ] and @xmath1156 \times s$ ] . from the definition of the map
@xmath922 it follows that for every integer @xmath16
@xmath1157 \times s.\ ] ] using formula we conclude that @xmath1158 \times s,\ ] ] where @xmath1159 and @xmath927 . for @xmath1160 , we know that every reeb chord in @xmath1161 lifts to a trajectory of @xmath908 in @xmath876 that starts in @xmath1037 and ends in @xmath1162 .
it follows from that all reeb chords in @xmath1161 have action @xmath1163 .
the homotopy class @xmath1161 contains an odd number of reeb chords .
the computation @xmath1164 implies that the numbers @xmath1165 and @xmath1166 have the same parity . therefore @xmath1167 .
our discussion in this paragraph shows that @xmath1168 for @xmath1169 as defined in section [ sectiongrowth ] . combining the last two paragraphs we obtain : @xmath1170 if @xmath1171 is sufficiently large .
it is an elementary exercise to see that there exist numbers @xmath43 , @xmath1172 , and @xmath646 such that @xmath1173 for all @xmath113 .
we have thus established the following theorem .
[ theoremgrowth ] the strip legendrian contact homology @xmath1174 has exponential homotopical growth rate with exponential weight @xmath1175 .
we now proceed to deduce theorem [ theorem2 ] from theorem [ theoremgrowth ] .
+ _ proof of theorem 2 : _ for each @xmath1176 we have proved that @xmath61 has exponential homotopical growth rate with exponential weight @xmath1175 . using that the contact manifolds @xmath59 are all non - diffeomorphic by proposition [ colinhonda ] , we establish the theorem .
combining theorem [ theorem1 ] and theorem [ theorem2 ] , we have that for every contact form @xmath2 associated to @xmath59 , the reeb flow of @xmath2 has positive topological entropy .
we mention that the methods developed in the present paper can be used to obtain estimates for the topological entropy in other families of contact manifolds .
one other class of examples is obtained by applying the foulon - hasselblatt integral surgery ( introduced in @xcite ) on the legendrian lift of a separating geodesic of a hyperbolic surface ; we refer the reader to @xcite for the precise definition of the foulon - hasselblatt surgery . by the analysis in @xcite , we know that for most integral surgeries the resulting contact 3-manifold @xmath1177 is not a seifert fibre space , but an `` exotic '' graph manifold . in the author s ph.d .
thesis @xcite the following theorem is proved : let @xmath1178 be the contact manifold associated with the hypertight contact form @xmath1179 obtained via the integral foulon - hasselblat surgery on the legendrian @xmath1180 , where @xmath1181 is a separating geodesic in the closed hyperbolic surface @xmath57 and @xmath1182 is its legendrian lift . then there exist disjoint legendrian curves @xmath19 and @xmath66 on @xmath1183 such that @xmath1179 is adapted to the pair @xmath1184 and @xmath1185 has exponential homotopical growth rate . in a recent work @xcite dahinden has obtained an extension of the results of @xcite to positive contactomorphisms .
he showed that if the homology of the based loop space of a manifold @xmath26 is rich then every positive contactomorphism in @xmath33 has positive topological entropy .
we believe that a similar result should also hold for the contact manifolds covered by theorem [ theorem2 ] .
in @xcite the author showed that if a contact 3-manifold @xmath0 admits a contact form @xmath25 whose reeb flow is anosov , then every reeb flow on @xmath0 has positive topological entropy .
a question we believe is interesting is : does there exist a pair of legendrian knots @xmath19 and @xmath20 in @xmath0 such that , @xmath25 is adapted to @xmath19 and @xmath20 and @xmath92 has exponential growth ?
it would be extremely interesting to know if there exist overtwisted contact 3-manifolds on which every reeb flow has positive topological entropy .
unfortunately , there seems to be no available technology that could help to answer this question . in this appendix
we prove a fubini type estimate which is necessary in the proof of theorem [ theorem1 ] .
we begin by fixing some notation .
let @xmath53 be a closed oriented 3-manifold , @xmath1186 be a smooth vector field without singularities on @xmath53 , @xmath1187 be an embedded knot in @xmath53 , and @xmath27 be a riemannian metric on @xmath53 .
we denote by @xmath1188 the flow of the vector field @xmath1186 . for a vector @xmath782 tangent to @xmath53 we let @xmath1189 be the norm of @xmath782 with respect to @xmath27 .
the riemannian metric @xmath27 provides us with a way to measure area for any surface immersed on @xmath53 .
we fix a parametrisation @xmath1190 of the knot @xmath1187 .. ] define a map @xmath1191 \times s^1 \to m$ ] by the formula @xmath1192 we denote by @xmath1193 the area with respect to @xmath27 of the immersed surface @xmath1194 \times s^1)$ ] .
we will denote by @xmath1195 the pullback metric @xmath1196 of @xmath27 to @xmath1197 \times s^1 $ ] via the map @xmath1198 .
associated to the riemannian metric @xmath1195 to @xmath1197 \times s^1 $ ] we have an area form @xmath1199 on @xmath1197 \times s^1 $ ] . because @xmath1195 is the pullback metric @xmath1196 we have the following equality : @xmath1200 \times s^1 } \mu_{\widehat{g } } = \int_{[0,c ] \times s^1 } |\mu_{\widehat{g}}(\partial_t , \partial_\theta)| dt d\theta.\ ] ] for a vector @xmath782 tangent to @xmath1197 \times s^1 $ ] we denote by @xmath1201 its norm with respect to the metric @xmath1195 .
it is a classical fact that @xmath1202 .
it follows from the definitions of @xmath1198 and @xmath1195 that @xmath1203 where @xmath1204 denotes the differential of @xmath1198 .
since @xmath53 is compact there exists a constant @xmath1205 such that @xmath1206 on all points of @xmath53 . combining this with and we obtain : @xmath1207 \times s^1 } |\mu_{\widehat{g}}(\partial_t , \partial_\theta)| dt d\theta \leq \int_{[0,c ] \times s^1 } |\partial_t |_{\widehat{g } } |\partial_\theta |_{\widehat{g } } d\theta dt \leq \mathrm{k } \int_0^c
\big ( \int_{s^1 } |\partial_\theta |_{\widehat{g } } d\theta \big ) dt.\ ] ] it is immediate to see that for each @xmath1208 $ ] , the integral @xmath1209 equals the length @xmath1210 of the curve @xmath1211 with respect to @xmath27 . combining this with and we get : @xmath1212 for the statement of the next lemma we keep the notation introduced so far in the appendix . _ proof : _ we argue by contradiction , assuming that @xmath1215 .
then there would be @xmath1216 and @xmath1217 such that : @xmath1218 for all @xmath1219 . integrating both sides of the for @xmath290 between @xmath328 and @xmath1171 and invoking
, we would conclude that : @xmath1220 for all @xmath1221 .
it is immediate that the right side of becomes a lot smaller than @xmath1222 for sufficiently large @xmath1171 .
this would imply that @xmath1223 for large @xmath1171 and would lead to a contradiction , since we assumed @xmath1224 for @xmath1171 sufficiently large . t. ekholm .
rational sft , linearized legendrian contact homology , and lagrangian floer cohomology , in _ perspectives in analysis , geometry , and topology _ , progress in mathematics 296 , 109146 , birkhauser , 2012 . | in this paper we study the growth rate of a version of legendrian contact homology , which we call strip legendrian contact homology , in 3-dimensional contact manifolds and its relation to the topological entropy of reeb flows .
we show that : if for a pair of legendrian knots in a contact 3-manifold @xmath0 the strip legendrian contact homology is defined and has exponential homotopical growth with respect to the action , then every reeb flow on @xmath0 has positive topological entropy .
this has the following dynamical consequence : for all reeb flows ( even degenerate ones ) on @xmath0 the number of hyperbolic periodic orbits grows exponentially with respect to the period .
we show that for an infinite family of 3-manifolds , infinitely many different contact structures exist that possess a pair of legendrian knots for which the strip legendrian contact homology has exponential growth rate . |
actinides play an important role in astrophysics in different ways .
first , the use of @xmath5 , @xmath6 and @xmath7 to estimate astrophysical ages has a long history , a milestone of which is the much celebrated piece of work of @xcite . for long , the field of nucleo - cosmochronology that emerged from this paper has been aiming at the determination of the age @xmath8 of the nuclides from abundances in the material making up the bulk of the solar system .
the astrophysical importance of th and u was enhanced further with the observation of th and u in some very metal - poor stars @xcite .
these measurements raised the hope of a possible nuclear - based evaluation of the age of individual stars other than the sun .
additional recent observational advances have triggered substantial interest in other actinides that are shorter - lived than th and u. this comes about following the measurement of the galactic cosmic ray ( gcr ) abundances of the @xmath9 elements , including the actinides , with unprecedented resolution , using the trek detector @xcite .
further significant progress is expected by the determination of the gcr abundances of the actinides th , u , pu and cm both relative to each other and relative to the pt - group of elements .
precise abundance measurements of this type would yield an estimate of the time elapsed between the nucleosynthesis of the gcr actinides and their acceleration to gcr energies ( the gcr actinide propagation time after acceleration is very short , i.e of the order of 2 my ) .
hence , they would help determining whether gcrs were accelerated out of fresh ejecta of the astrophysical r - element sources ( supernovae or neutron - star mergers ) , superbubble material , or old , well - mixed galactic material . also attempts to measure the @xmath3pu content of the local interstellar medium should be mentioned here .
such measurements may allow for conclusions on the frequency of the astrophysical events producing actinides . at present
, this can be done through the analysis of dust grains of identified interstellar origin recovered in deep - sea reservoirs ( sediments and femn crusts ; e.g. * ? ? ?
* ; * ? ? ?
* ) and by the recovery of @xmath10fe incorporated in fossil biogenic samples @xcite .
marine sediments were also analysed for a @xmath3pu signal associated with measurements of @xmath10fe in a ferromanganese crust @xcite , which has been interpreted as product of a supernova close to the solar system about 2.2my ago @xcite . for all these reasons ,
there is an obvious need to provide detailed theoretical estimates of the possible stellar production of actinides with half - lives in excess of typically @xmath11 y. most importantly , the uncertainties in these predicted abundances should be evaluated as well . while the actinides are clearly produced by the rapid neutron - capture process ( or r - process ) of stellar nucleosynthesis , the site(s ) of this nucleosynthesis process
have not been unambiguously identified yet @xcite .
two possible astrophysical sites of r - process production are discussed : supernovae and neutron - star mergers . in the case of supernova explosions a variety of scenarios for r - element creation by primary and secondary processes
were proposed .
neutron - rich jets in rare , magnetohydrodynamic explosions of rapidly rotating stars ( e.g. * ? ? ?
* ) and neutron production by neutrino reactions in the helium layer of compact , metal - poor exploding stars @xcite were considered as being potentially responsible for an early enrichment of the young galaxy with r - process matter .
however , supernovae could be a major or even dominant source of r - process elements only if the so - called neutrino - driven wind , a low - mass baryonic outflow from newly formed neutron stars ( e.g. * ? ? ?
? * ; * ? ? ?
* ) , were able to provide an r - process viable environment @xcite .
self - consistent models , however , do not only yield wind entropies that are too low to enable a strong r - process that could make lanthanides and actinides ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , but state - of - the - art supernova models with high - fidelity neutrino transport also yield proton - rich conditions instead of neutron excess in the wind ejecta @xcite . recently , growing attention has been paid to mergers of binary neutron stars ( ns - ns ) and neutron star - black hole ( ns - bh ) systems as possible r - process sites @xcite , following the confirmation by hydrodynamic simulations that non - negligible amounts of matter , typically about @xmath12 to several @xmath13@xmath14 , can be ejected ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in contrast to the supernova case , investigations with growing sophistication have so far supported ns merger ejecta as viable sites for strong r - processing ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* for steps towards including neutrino effects ) .
in particular , comprehensive nucleosynthesis calculations @xcite show that the combined contributions from both the dynamical ( prompt ) ejecta expelled during ns - ns or ns - bh mergers , and the neutrino and viscously driven outflows generated during the post - merger remnant evolution of relic nss or bh - torus systems can lead to the production of r - process elements from @xmath15 up to thorium and uranium with an abundance distribution that reproduces extremely well the solar distribution , as well as the elemental distribution observed in low - metallicity stars @xcite .
despite a roughly 1000 times lower ns merger rate compared to the core - collapse supernova rate , ns - ns / bh mergers could still explain the total amount of r - material in the galaxy because the amount of ejected r - process - rich matter per ns merger is roughly 1000 times higher than the potential contribution from the neutrino - driven wind of a supernova .
based on observed binary pulsars and population synthesis calculations , the ns - ns merger rate is currently estimated to be within the plausible range of 3 to 190 myr@xmath16 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
initially there were concerns that the low rate of compact star mergers ( with their correspondingly larger ejecta masses per event ) in addition to the long time delay of binary mergers after the preceding , iron - producing supernovae would be incompatible with the galactic enrichment history as deduced from observations of ( ultra-)metal - poor stars @xcite .
however , more recent models of the chemical evolution of the milky way show a much more promising situation . taking into account a short - lived binary component ( with inspiral times of less than 100 myr ) , incomplete galactic mixing and/or the sub - halo merger history of the milky way
, these new studies conclude that double compact star mergers might indeed be the major producers of r - process elements and might even be responsible for the enrichment of metal - poor stars @xcite .
although these investigations , which are based on largely different modeling approaches , ranging from traditional , highly simplified box models to hydrodynamical simulations , do not agree in all quantitative details of the picture , the majority of them suggests that compact binary mergers as r - process sites are better compatible with the considerable event - to - event scatter of the r - process abundances in metal - poor stars . if ns - ns / bh mergers are indeed the main cosmic sources of heavy r - process matter , the rarity of such events would leave little hope for discovering on earth larger amounts of radioisotopes of cosmic origin like @xmath3pu .
in fact , a recent measurement found a very low @xmath3pu abundance in deep - ocean reservoirs , about two orders of magnitude lower than expected from continuous production by frequent sources like supernovae @xcite . in combination with abundance measurements for the early solar system this experimental result was therefore interpreted as strong evidence for the origin of @xmath3pu from compact binary mergers @xcite . however , more experimental and observational confirmation with better statistics and based on alternative reservoirs is highly desirable to consolidate the picture suggested by the existing measurements , which are subject to considerable uncertainties associated with limited statistics and our incomplete knowledge of the probability with which supernova - made nuclei make their way to earth and finally end up in the investigated sample material .
such caveats certainly justify ongoing efforts to directly identify supernovae as cosmic sources of r - process matter or , alternatively , to derive increasingly stronger bounds to the supernova origin of this matter by searching for r - nuclei in the envelopes of supernova companion stars and for radioisotopes in dated terrestrial sediments .
it should be kept in mind that without any unambiguous identification of an r - process source by the in - situ detection of r - nuclei our understanding of the astrophysical origin of this nucleosynthetic component is based on complex indirect arguments and on theoretical models with their natural limitations .
such limitations are associated with numerous simplifications that still enter the modeling of supernovae as well as compact binary mergers , with constraints set by the finite numerical resolution and by the omission of physics that could play a role like , for example , acoustic waves due to ns vibration @xcite , fallback in supernovae and associated re - ejection of matter @xcite , strong neutron - star magnetic fields ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , a rigorous treatment of neutrino oscillations , or non - standard weak - interaction physics . with the goal to assist future experimental searches we provide here state - of - the - art estimates of the possible production of the actinides .
in order to circumvent the uncertainties of current core - collapse supernova and neutrino - wind models , we adopt an optimistic point of view here and assume that the r - process taking place in type - ii supernovae is capable of producing elements up to the heaviest actinides and in addition that each such event leads to an r - abundance distribution similar to the one found in the solar system . the striking similarity between the solar distribution of r - element abundances in the @xmath17 range and the corresponding abundance pattern observed in ultra - metal - poor stars like
cs 22892 - 052 @xcite led to the conclusion that every astrophysical event producing r - elements gives rise to a solar system - like r - abundance distribution , at least for elements above ba .
such observations therefore tend to lend support to our assumption , although recent observations also indicate that star - to - star variations in the r - process content of metal - poor globular clusters may be a common , although not ubiquitous , phenomenon @xcite .
section [ sect_nass ] describes the analytical @xmath18-driven wind model considered in the present work , while our fitting procedure used to construct an optimal reproduction of the solar system r - abundance distribution is outlined in sect .
[ sect_fit ] .
the different nuclear inputs used in the present study are detailed in sect .
[ sect_nuc ] .
finally , our fit to the solar system r - abundances , the astrophysical ranges required for this fit as well as our predictions concerning the actinide production are presented in sect .
[ sect_res ] .
conclusions are drawn in sect .
[ sect_conc ] .
several wind models of analytical or semi - analytical nature exist . they differ in their level of physical sophistication and in their way to parametrize the wind characteristics . in all cases
, the wind is assumed to be spherically symmetric , which appears to be a reasonable first approximation even in two - dimensional simulations , at least at late enough times after core bounce @xcite .
in addition , the wind is generally treated as a stationary flow , meaning no explicit time dependence of any physical quantity at a given radial position @xmath19 , so that @xmath20 , let @xmath21 be the velocity , temperature , density , internal energy , pressure , entropy , or composition .
the validity of this approximation is discussed in @xcite , where it is concluded that stationarity may be reasonably assured , even if some caution is warranted .
newtonian , post - newtonian , and relativistic descriptions of spherically symmetric , stationary neutrino - driven winds originating from the surface of proto - neutron stars ( pns ) have been developed along the lines of a long experience with previous mathematical treatments of the solar wind and of accretion flows onto black holes .
the wind model adopted here corresponds to the analytical newtonian , adiabatic and steady - state wind model , referred to in the following as nass , derived by @xcite .
it provides a simple , fully analytical description of the dynamics of the wind outflow at relatively late times or sufficiently far away from the pns surface .
the nass wind model relies on the general assumptions listed above , in particular those of a newtonian pns gravitational potential and of an adiabatic expansion .
in addition , all the elementary @xmath22 and e@xmath23/e@xmath24 weak interaction processes are assumed to be frozen out , nuclear @xmath25-decays not to affect @xmath26 or @xmath27 , and possible deviations from nuclear equilibrium with regard to strong and electromagnetic interactions to have no influence on the thermodynamical properties of the wind . under such simplifying assumptions
, the nass model can not predict any time variations of the entropy @xmath27 , electron fraction @xmath28 , or of the wind mass loss rate @xmath29 , which are thus treated as constant input parameters .
the basic nass wind dynamics in the regime under consideration , and in particular for high enough entropies , is well approximated @xcite by @xmath30 where @xmath31 is the pns mass , @xmath19 the radius , @xmath32 the velocity , @xmath33 the temperature . the total energy per unit mass @xmath34 may be obtained by setting a boundary condition and is usually expressed as a function of the wind energy : @xmath35 , where @xmath36 and @xmath37 equals the local adiabatic sound speed @xcite . for @xmath38 ,
the solutions correspond to a sonic wind , whereas for @xmath39 they are of subsonic wind type and usually referred to as breeze solutions .
more details about the wind model can be found in @xcite .
the entropy is dominated by photons , electrons and positrons in the wind so that the so - called ` radiation entropy ' @xmath40 is given by @xmath41\ \ \
{ \rm with}\ \ \
s_{\rm rad}^{0 } = \frac{11\pi^2}{45 \rho n_{\rm a } } \bigl ( \frac{kt}{\hbar c } \bigr ) ^3 , \label{eq : wind_srad}\ ] ] where @xmath42 in the high-@xmath33 limit when electrons and positrons are highly relativistic , and decreases with @xmath33 for high @xmath43-values . in the present study
, it is approximated by @xmath44 , where @xmath45 is the temperature expressed in @xmath46k @xcite . since the nucleosynthesis in the wind
is followed by starting at a temperature of @xmath47 , the total entropy is essentially equal to the radiation entropy ( within a few percents ) . note that the ` radiation entropy ' is measured here per baryon in units of the boltzmann constant @xmath48 and includes both the photon and lepton contributions , but not the baryonic ones . for the initial conditions adopted here , the latter amounts to typically 9 @xmath48/baryon . in the wind model of @xcite , the radial evolution of the velocity , @xmath49 , and density , @xmath50 , for specified total energy ( @xmath34 ) per unit mass of outflow material ( or , equivalent , for given @xmath51 ) , given entropy and given mass - loss rate ( @xmath52 ) is determined by the energy equation ( based on bernoulli s equation , see eq .
[ eq : wind_motion ] ) and the continuity equation @xmath53 .
the temperature then follows from the assumption of constant entropy , which couples temperature and density .
varying @xmath51 ( or @xmath34 ) , @xmath54 or @xmath55 therefore also implies different values of the expansion time scale .
within the nass wind model , the velocity at small radii ( and high temperatures ) is found to vary as @xmath56 and consequently significantly increases with the radiative entropy for given @xmath55 and @xmath51 .
the nass equations describe the outflow dynamics ( wind and breeze solutions ) as functions of radius and as determined by @xmath55 , @xmath34 ( or , equivalent , @xmath51 ) , and @xmath54 .
these parameters as well as the electron fraction , @xmath57 , are considered to be independent of each other , which implies that the nass model does not invoke any physical mechanism by which these characteristic parameters of the flow are coupled . in reality , the physical processes that drive the mass outflow from the pns surface will determine the values of these quantities and may lead to relations between them .
the nass model therefore provides a more general description of outflows from new - born nss than , e.g. , the usual neutrino - driven wind models mentioned above , where neutrino heating near the pns surface sets the conditions in the ejecta .
this more general modeling approach is an advantage and requirement of our study , which intends to explore also the uncertainty limits of predictions of actinide nucleosynthesis associated with still unsettled aspects of the wind physics ( like , e.g. , the impact of acoustic waves from ns vibrations or strong magnetic fields ) .
because of this freedom of the nass model , however , we have to define suitable ranges of the parameters of the model , which allow us to scan the range of possibilities .
naturally , this can happen only by the investigation of a limited number of cases , which we define as `` ranges i iv '' .
the four ranges of outflow conditions considered in the present study are defined as follows , aiming at a good reproduction of the solar r - process abundances from the first to the third abundance peak . in range
i , we include 150 trajectories with entropies @xmath58 binned in steps of 50 , electron fractions @xmath59 in steps of 0.02 , mass outflow rates @xmath60 , 0.06 and @xmath61s , and @xmath62 . in range ii , a set of breeze solutions is considered , i.e. entropies @xmath63 in steps of 50 , electron fractions @xmath59 in steps of 0.02 , a single outflow rate of @xmath64s , and @xmath62 .
this restricted set of 40 trajectories will be shown in sect .
[ sect_res ] to still allow for a good fit to the solar system r - abundances and will consequently be used for a relatively detailed sensitivity analysis concerning nuclear uncertainties . in range
iii , wind solutions are considered , i.e. @xmath38 , with entropies @xmath58 in steps of 50 , electron fractions @xmath59 in steps of 0.02 , and an outflow rate @xmath64s .
the astrophysical conditions thus considered are associated with a wide range of expansion time scales @xmath65 . if we define @xmath65 as the time required for the temperature to drop from @xmath47 to @xmath66 , @xmath65 ranges between 120 and 950 ms for an entropy of @xmath67 and between 18 and 30 ms for @xmath68 at conditions of range i. in range ii ( iii ) , @xmath69 ( 31 ) ms for the largest entropies of @xmath70 , but increases up to 7.3 ( 7.2 ) s for @xmath71 .
in contrast to the usual behavior of neutrino - driven winds , where higher entropies are associated with increasing expansion time scales @xcite , the conditions defined by ranges i iii exhibit inverse relations , namely greater entropies for shorter expansion time scales .
this appeals to some unknown mechanism that can provide such conditions ( which is not implausible in view of the fact that ordinary neutrino - driven winds are not found to produce heavy r - process material ) .
the standard behavior of neutrino - driven winds can be reproduced by our nass model by making use of the relation @xmath72 and by modifying the outflow rate such that the effect of higher entropies on the wind velocity is compensated . in order to study a case
more compatible with neutrino - driven winds and breezes , we therefore consider an additional set of breeze solutions , range iv , with the following entropies @xmath73 and outflow rates @xmath55 ( in @xmath74s ) : ( 100 , 0.6 ) , ( 125 , 0.17 ) , ( 150 , 0.06 ) , ( 175 , 0.015 ) , and ( 200 , 0.006 ) . with @xmath62 ,
the corresponding expansion time scales @xmath65 of these five events are 72 , 86 , 106 , 209 , and 296 ms , respectively , which clearly increase for increasing entropies . for each of these combinations of entropy and outflow rate , 15 events for electron fractions varying within @xmath75 in steps of 0.02
are considered .
range iv therefore includes 75 events in total .
in a similar way to that developed in the multi - event canonical s- or r - process models @xcite , it is possible to define a superposition of a large number of @xmath18-driven wind components ( `` events '' ) that correspond to different thermodynamic conditions .
each event is characterized by a given entropy @xmath73 , electron fraction @xmath57 , outflow rate @xmath76 , and wind energy scaling factor @xmath51 .
the combination of r - process events that provides the best fit to the solar abundances can then be derived with the aid of an iterative inversion procedure that has been applied to astronomical inverse problems @xcite but also to parametric r - process calculations @xcite .
the solar abundance @xmath77 of a nuclide @xmath78 is approximated by the weighted superposition of the abundances @xmath79 resulting from all astrophysical events according to @xmath80 where @xmath81 represents the statistical weight of the event @xmath82 .
the recursion relation @xmath83 is used in order to obtain an `` improved '' @xmath84th estimate of @xmath81 from the @xmath19th iteration @xmath85 .
@xmath86 is defined by @xmath87 the iteration procedure starts with a uniform distribution of initial weights @xmath88 ( i.e. all events have initially the same weight ) and converges after several iterations to a `` best - fit '' abundance curve @xmath89 .
the corresponding weight profile @xmath81 allows us to identify the events that contribute most significantly to the synthesis of each fitted element .
two different sets of solar system r - abundance distributions are considered in the present study .
they differ by the way the s - process contribution to the solar system is estimated .
@xcite used the multi - event canonical s - process model and included a detailed analysis of observational , astrophysical and nuclear physics uncertainties , while @xcite estimated the s - process abundances on the basis of a galactic chemical evolution model with agb yields based on parametrized @xmath90c profiles to generate the s - process irradiation . due to the more complex approach
, uncertainties are not estimated in the latter case and the contribution from the weak s - component in massive stars responsible for the solar production of @xmath91 nuclei is not included either .
the corresponding solar r - abundance distributions can differ significantly for s - dominant nuclei , especially in the pb region , as shown in fig .
[ fig_sol ] .
the nucleosynthesis is followed with a reaction network including all 5000 species from protons up to @xmath92 lying between the valley of @xmath25-stability and the neutron - drip line .
all charged - particle fusion reactions on light and medium - mass elements that play a role when the nuclear statistical equilibrium freezes out are included in addition to radiative neutron captures and photodisintegrations .
the reaction rates on light species are taken from the netgen library , which includes all the latest compilations of experimentally determined reaction rates @xcite . by default
, experimentally unknown reactions are estimated with the talys code @xcite on the basis of the hfb-21 nuclear masses @xcite , the hfb plus combinatorial nuclear level densities @xcite and the qrpa e1 strength functions @xcite . fission and @xmath25-decay processes are also included , i.e. neutron - induced fission , spontaneous fission , @xmath25-delayed fission , as well as @xmath25-delayed neutron emission @xcite .
the @xmath25-decay processes are taken from the updated version of the gross theory @xcite based on the hfb-21 @xmath93-values @xcite , when not available experimentally , whereas all fission processes are estimated on the basis of the hfb-14 fission paths @xcite and the full calculation of the corresponding barrier penetration @xcite .
the fission fragment distribution is taken from the spy model as described in @xcite .
this nuclear physics set represents our standard input . due to the large uncertainties still affecting the properties of the neutrinos , in particular their luminosities and temperatures , no neutrino interactions on nuclei are included in the present calculations .
( neutrino interactions with free nucleons in the close vicinity of the neutron star are considered to be important in setting the dynamical and thermodynamic conditions and the neutron - to - proton ratio in the wind outflow , which is inherently accounted for by our parametric wind model . ) to estimate the sensitivity of the abundance calculations with respect to the still uncertain nuclear physics predictions far away from the valley of @xmath25-stability , different nuclear ingredients are also considered in addition to the above - described standard set .
these include * hfb-31 : reactions rates determined on the basis of the hartree - fock - bogoliubov ( hfb ) hfb-31 mass model @xcite ; all other inputs remaining identical ; * d1 m : reactions rates determined on the basis of the d1 m gogny - hfb mass model @xcite ; all other inputs remaining identical ; * frdm : reactions rates determined with the 2012 version of the frdm mass model @xcite , the back - shifted fermi gas model for nuclear level densities @xcite and lorentzian - type @xmath94 strength function @xcite ; @xmath25-decay rates are taken from the random phase approximation and frdm - based @xmath93-values @xcite ; fission probabilities are based on the @xcite fission barriers and the fragment yields on the gef model @xcite .
this nuclear physics set essentially includes so - called phenomenological macroscopic approaches in contrast to the microscopic input characterizing the standard set ; * fis : reaction and @xmath25-decay rates are identical to the default models , but fission probabilities are based on the @xcite fission barriers and the fragment yields on the gef model @xcite ; * beta : @xmath25-decay rates from the gross theory are replaced by the tamm - dancoff approximation @xcite ; all other inputs remaining identical to the standard set .
the solar system abundances have been fitted assuming different ranges of astrophysical conditions for the @xmath18-driven wind and different nuclear physics inputs , as described in the previous sections .
one example ( case 1 ; table [ tab1 ] ) of a fit corresponding to the range i of astrophysical conditions and the standard nuclear physics input , is shown in fig . [ fig_afit ] .
it can be seen that the solar distribution ( in this case , taken from * ? ? ?
* ) is rather well fitted , although deviations can be observed , in particular in the rare - earth region .
the corresponding distributions of entropies and electron fractions are illustrated for this standard case in fig .
[ fig_histo ] . for illustration ,
additional fits with range ii of astrophysical conditions and the nuclear physics inputs corresponding to d1 m and frdm ( cases 7 and 11 , respectively ; table [ tab1 ] ) are given in fig .
[ fig_afit2 ] .
it will also be noticed that the restricted range ii is already sufficient to provide the conditions needed to reproduce the solar system distribution fairly well .
the statistical weights of entropies and electron fractions needed to obtain these fits are obviously dependent on the adopted range of astrophysical conditions as well as the nuclear physics inputs , as shown in fig .
[ fig_histo2 ] . considering different combinations for the solar r - abundance distributions , ranges of thermodynamic conditions and nuclear physics inputs ,
quite some different predictions for the production of actinides can be derived as summarized in table [ tab1 ] .
in particular , it can be seen that wind solutions ( @xmath38 ) in the range iii of thermodynamic conditions lead to a significantly lower production of some actinides compared to the breeze solutions ( @xmath62 in range ii ) .
the specific conditions defined in range iv lead to an actinide production quite similar to the one obtained in range ii .
nuclear physics inputs , in particular nuclear mass models ( and the corresponding rates ) as well as @xmath25-decay rates , have also a non - negligible impact on the abundance predictions , as already well established for decades . uncorrelated lower and upper limits to the production of the long - lived actinides
can be deduced and are also given in table [ tab1 ] .
the uncertainties for the @xmath0th production are found to be considerably larger than those for the production of @xmath3pu and @xmath4 cm . for comparison ,
the upper and lower limits estimated within the multi - event canonical model @xcite on the basis on very different astrophysical conditions and nuclear physics inputs are also given in table [ tab1 ] .
as a function of the radiative entropy @xmath73 ( left panel ) and @xmath95 as a function of electron fraction @xmath57 ( right panel ) responsible for the abundance fit shown in fig .
[ fig_afit ] for range i conditions . ]
but for events with range ii conditions and the nuclear physics inputs corresponding to d1 m and frdm .
] but for the statistical weights corresponding to the abundance fit shown in fig .
[ fig_afit2 ] with d1 m mass model and range ii conditions . ] for completeness , table [ tab1 ] also gives the estimated age of the ultra - metal - poor star cs 31082 - 001 .
indeed , accurate observations of heavy r - elements have been used to estimate the age of ultra - metal - poor stars on the basis of the fundamental assumption that the r - process is universal @xcite . in particular ,
if th and u lines could be observed accurately and simultaneously in metal - poor stars , a relatively reliable age estimate could be derived from the expression @xmath96 where @xmath97 is the characteristic @xmath98-decay timescale of @xmath99u and similarly @xmath100 for @xmath0th .
note that the th abundance in the r - process ratio of eq .
( [ eq2 ] ) corresponds to the @xmath0th value after decay of its shorter - lived progenitors @xmath101u and @xmath3pu , and the u abundance corresponds to the @xmath99u contribution only .
th and u lines have been observed in particular in the cs 31082 - 001 star with a @xmath102 ( * ? ? ? * and private communication ) .
the 0.09 dex observational error gives rise to a @xmath103 gyr uncertainty on the estimate of @xmath104 .
for each simulation , the estimated age of cs 31082 - 001 is given in table [ tab1 ] .
the ages obtained are reasonably consistent with the wmap estimate of 13.8 gyr @xcite . [
cols="^,^,^,^,^,^,^,^,^,^,^,^,^",options="header " , ] [ tab2 ]
we derived updated estimates of the production of the radioactive actinides of @xmath0th , @xmath1u , @xmath2np , @xmath3pu , and @xmath4 cm and of the corresponding uncertainty ranges based on the assumption that the neutrino - driven wind of the proto - neutron star in core collapse supernovae is able to provide r - process viable conditions and can well reproduce the universal r - process abundance pattern observed in the sun and found in metal - poor stars .
since current state - of - the - art hydrodynamical models of supernova explosions and of the neutrino - driven wind do not yield the dynamic and thermodynamic conditions needed for the production of heavy r - process material , we base our study on the simple , analytical nass ( newtonian , adiabatic , steady - state ) wind model and take into account the mass - weighted superposition of a large number of wind components with different nucleosynthesis - relevant characteristics of entropy , expansion time scale , and neutron excess .
the set of chosen conditions is constrained by the ability of the integrated wind material to match the solar r - abundance distribution . in order to explore the sensitivity of our estimates for the actinide production to uncertain nuclear physics
, we tested the influence of six different sets of nuclear - rate ingredients .
the consideration of neutrino - driven winds of supernovae as possible r - process site , despite the unfavorable conditions provided by current hydrodynamical models , and the combination of wind components under the mentioned constraint , can be justified by the fact that also the state - of - the - art models still suffer from major uncertainties , e.g. with respect to neutrino opacities in correlated nuclear matter , the effects of potentially strong magnetic fields inside and around the nascent ns , or the incomplete exploration of neutrino - oscillation effects and possible non - standard weak interaction physics in the pns environment . on the other hand , despite considerable modeling progress and generally more promising properties of the ejecta dynamics , also ns - ns and/or ns - bh mergers are by far not established as the main sources of heavy r - process nuclei , because in this case too , major uncertainties still affect the weak - interaction sector as well as the description of strong magnetic field effects .
further observational efforts for a positive confirmation of the one or the other or of both possible sources are therefore needed .
this includes searches of electromagnetic transients associated with the radiation emission of ns - ns / bh merger ejecta heated by the radioactive decay of r - process nuclei ( see for example , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) as well as the exploration of cosmic and terrestrial reservoirs for signatures of freshly produced r - nuclei . our work is intended to facilitate the observational and experimental analyses of the latter kind and their interpretation .
we find a considerable spread of the results of actinide yields with the largest uncertainty ( a factor of @xmath105100 ) for @xmath0th and the lowest uncertainty for @xmath3pu ( factor of @xmath10512 ) and for @xmath4 cm ( factor of @xmath1057 ) , with supersonically expanding wind solutions producing considerably less actinide material than more slowly , subsonically expanding breeze outflows . for investigations of terrestrial material
like recently performed by @xcite , it is interesting to know the isotope - production ratio of @xmath3pu compared to the long - lived radioactive nucleus @xmath10fe .
current supernova models predict a yield of @xmath10fe of @xmath106 per massive - star death or @xmath107 per supernova @xcite .
since this theoretical value seems to exceed estimates based on cosmic - rays near the earth by about a factor of 2 @xcite , we consider here a @xmath10fe output per exploding massive star of @xmath108 . with this number
we obtain a theoretical range for the supernova - produced @xmath3pu/@xmath10fe isotope ratio between roughly @xmath109 and @xmath110 , using the yields of table [ tab2 ] . with experimentally determined limits of less than @xmath111 for their crust and sediment samples
, @xcite set a bound still more than a factor of 10 below our lower limit .
therefore this bound seems to exclude a recent insemination of the earth by r - process material from a frequent source like supernovae , in particular from a nearby supernova @xmath1052.2my in the past
. however , the exclusion may not be as convincing as suggested by the two orders of magnitude discrepancy advocated by @xcite .
not only the probability with which @xmath3pu from a possible supernova origin ends up in the investigated material samples is uncertain , also the extreme sensitivity of the actinide production to the model variations tested in our work should be taken as a warning .
we thank robert bollig for useful discussions and shawn bishop for inspiring conversations as well as comments on the manuscript .
we are grateful to the referee , stan woosley , for constructive questions that helped us to improve the presentation .
sg acknowledges financial support from frs - fnrs ( belgium ) , htj from deutsche forschungsgemeinschaft through the cluster of excellence `` origin and structure of the universe '' ( exc-153 ) .
ade p.a.r .
2015 , arxiv:1502.01589 arcones a. & janka h .- t .
, 2011 , a&a , 526 , a160 argast , j. et al . , 2004 , a&a , 416 , 997 arnould m. , goriely s. , takahashi k. , 2007 , phys .
, 450 , 97 banerjee p. , haxton w.c . , qian y .- z .
, 2011 , phys .
lett . , 106 , 201104 bauswein a. , goriely s. , janka h .-
, 2013 , apj , 773 , 78 bishop s. & egli r. , 2011 , icarus , 212 , 960 bisterzo s. , travaglio c. , gallino r. , wiescher m. , kppeler f. , 2014 , apj , 787 , 10 bouquelle v. , cerf n. , arnould m. , tachibana t. , goriely s. , 1996 , a&a , 305 , 1005 cayrel r. , hill v. , beers t.c .
et al . , 2001 , nature , 409 , 691 cowan j.j . , mcwilliam a. , sneden c. , burris d.l .
, 1997 , apj , 480,246 dominik , m. , belczynski , k. , fryer , c. , holz , d.e . , berti , e. , bulik , t. , mandel , i. & oshaughnessy , r. 2012 , apj , 759 , 52 eichler d. , livio m. , piran t. , schramm d.n . , 1989 ,
nature , 340 , 126 fischer t. , whitehouse s.c .
, mezzacappa a. , thielemann f .- k . , liebendrfer m. , 2010 , a&a , 517 , a80 fitoussi c. , raisbeck g. m. , knie k. , korschinek g. , faestermann t. et al . , 2008 ,
101 , 121101 fowler w.a . &
hoyle f. , 1960 , ann .
phys . , 10 , 280 freiburghaus c. , rosswog s. , thielemann f .- k . , 1999 ,
apj , 525 , l121 fryer c.l . ,
herwig f. , hungerford a. , timmes f.x . , 2006 ,
apjl , l131 goriely s. , arnould m. , 1996 , a&a , 312 , 327 goriely s. , 1997 , a&a , 327 , 845 goriely s. , 1999 , a&a , 342 , 881 goriely s. , clerbaux b. , 1999b , a&a , 346 , 798 goriely s. , arnould m. , 2001 , a&a , 379 , 1113 goriely s. , khan e. , samyn m. , 2004 , nucl .
a739 , 331 goriely s. , samyn m. , pearson j. m. , 2007 , phys .
c , 75 , 064312 goriely s. , hilaire s. , koning a.j .
, 2008 , a&a , 487 , 767 goriely s. , hilaire s. , koning a.j . , 2008b , phys .
c , 78 , 064307 goriely s. , hilaire s. , koning a.j . , sin m. , capote r. , 2009 , phys . rev .
c , 79 , 024612 goriely s. , hilaire s. , girod m , pru s. , 2009b , phys .
rev . lett . , 102 , 242501 goriely s. , chamel n. , pearson j. m. , 2010 , phys
c , 82 , 035804 goriely s. , bauswein a. , janka h .-
, 2011 , apjl , 738 , l32 goriely s. , sida j .-
lematre j .- f . , panebianco s. , dubray n. , hilaire s. , bauswein a. , janka h .-
, 2013 , phys .
lett . , 111 , 242502 goriely s. , 2015 , european physical journal , a51 , 172 goriely s. , bauswein a. , just o. , pllumbi e. , janka h .- th . , 2015 , mnras , 452 , 3894 goriely s. , chamel n. , pearson j. m. , 2016 , phys . rev .
c , 93 , 034337 hoffman r.d . ,
woosley s.e . , qian y .- z . , 1997 , apj , 482 , 951 honda s. , aoki w. , ishimaru y. , wanajo s. , 2007 , apj , 666 , 1189 hotokezaka k. , piran t. , paul m. , 2015 , nature physics , 11 , 1042 hdepohl l. , mller b. , janka h .-
, marek a. , raffelt g.g .
, 2010 , phys .
, 104 , 251101 ishimaru y. , wanajo s. , prantzos n. , 2015 , apj , 804 , l35 janka h .- t . , 2012 , ann .
science , 62 , 407 just o. , bauswein a. , ardevol pulpillo r. , goriely s. , janka h .-
t . , 2015 , mnras , 448 , 541 kasen d. , fernndez r. & metzger b.d . , 2015 , mnras , 50 , 1777 kim c. , kalogera v. & lorimer d.r . , 2010 , new astron .
, 54 , 148 klapdor h. v. , metzinger j. , oda t. , at .
data nucl .
data tables , 1984 , 31 , 81 knie k. , korschinek g. , faestermann t. , wallner c. , scholten j. , hillebrandt w. , phys .
, 83 , 18 knie k. , korschinek g. , faestermann t. , dorfi e.a . , rugel g. , wallner c. , phys .
lett . , 93 , 171103 komiya y. , et al . , 2014 , apj , 783 , 132 koning a.j . , hilaire s. , goriely s. , 2008 , nucl .
phys . , a810 , 13 koning a.j .
, rochman d. , nuclear data sheets , 2012 , 113 , 2841 kopecky j , uhl m. , 1990 , phys .
c , 41 , 1941 korobkin o. , rosswog s. , arcones a. , winteler c. , 2012 , mnras , 426 , 1940 k. kyutoku , k. ioka , 2016 , eprint arxiv:1603.00467 lattimer j.m . &
schramm d.n . , 1976 , apj , 210 , 549 lattimer j.m .
, mackie f. , ravenhall d.g . , schramm d.n . , 1977 ,
apj , 213 , 225 lucy l.b .
, 1974 , aj , 79 , 745 ludwig p. , bishop s. , egli r. , chernenko v. , deneva b. , et al . , 2016 , proc .
, accepted martin d. , perego a. , arcones a. , thielemann f .- k . , korobkin o. , rosswog s. , 2015 , apj , 813 , 2 matteucci f. , et al . , 2014 , mnras , 438 , 2177 mennekens n. & vanbeveren d. , 2014 , a&a , 564 , a134 metzger b.d . , thompson t.a .
, quataert e. , 2007 , apj , 659 , 561 metzger b.d . , martinez - pinedo g. , darbha s. , et al . , 2010 , mnras , 406 , 2650 mirizzi a. , tamborra i. , janka h .-
, saviano n. , scholberg k. , bollig r. , hdepohl l. , chakraborty s. , 2016 , la rivista del nuovo cimento , 39 , 12 mller p. , pfeiffer b. , kratz k .-
l . , 2003 ,
c , 67 , 055802 mller p. , sierk a.j .
, ichikawa t. , sagawa h. , 2016 , adndt , submitted myers w.d .
, swiatecki w.j . , 1999 ,
c , 60 , 014606 suzuki t.k .
& nagataki s. , apj , 628 , 914 paul m. , valenta a. , ahmad i. , et al . , 2001 , apj , 558 , l133 perego a. , rosswog s. , cabezn r.m . , korobkin o. , kppeli r. , arcones a. , liebendrfer m. , 2014 , mnras , 443 , 3134 pruet j. , woosley s.e .
, buras r. , et al . , 2005 , apj , 623 , 325 qian y .- z . &
woosley s.e . , 1996 , apj , 471 , 331 radice d. , galeazzi f. , lippuner j. , roberts l.f . ,
, rezzolla l. , 2016 , eprint arxiv:1601.02426 raisbeck g. , tran t. , lunney d. , gillard c. , goriely s. , waelbroeck c. , yiou f. , 2007 , nuclear instruments and methods in physics research b , 259 , 673 roberts l.f . , woosley s.e . ,
hoffman r.d . , 2010 ,
apj , 722 , 954 roberts l.f . , kasen d. , lee w.h . , ramirez - ruiz e. , 2011 , apjl , 736 , l21 roberts l.f . , et al . , eprint arxiv:1601.07942 roederer i. u. , cowan j. j. , karakas a. i. , kratz k .-
, lugaro m. , simmerer j. , farouqi k. , sneden c. , 2010 , apj , 724 , 975 roederer i.u . , 2011 , apjl , 732 , l17 roederer i.u . , et al . , 2012 ,
apjs , 203 , 27 rosswog s. , liebendrfer m. , thielemann f .- k .
, davies m.b . ,
benz w. , piran t. , 1999 , a&a , 341 , 499 schmidt k .- h . , jurado b. , phys .
procedia , 2012 , 31 , 147 sekiguchi y. , kiuchi k. , kyutoku k. , shibata m. , 2015 , phys . rev .
d , 91 , 064059 s. , cooke r. , ramirez - ruiz e. , madau p. , mayer l. , guedes j. , 2014 , arxiv:1407.3796 sneden c. , mcwilliam a. , preston g.w .
, et al . , 1996 , apj , 467 , 819 sneden c. , et al . , 2003 , apj , 591 , 936 sneden c. , cowan j. j. , gallino r. , 2008 , ara & a , 46 , 241 sneden c. , lawler j. e. , cowan j. j. , ivans i. i. , den hartog e. a. , 2009 , apjs , 182 , 80 sukhbold t. , ertl t. , woosley , s.e . , brown j.m .
, janka h .-
, 2015 , e - print arxiv:1510.04643 tachibana t. , yamada m. , yoshida y. , 1990 , prog .
phys . , 84 , 641 takahashi k. , witti j. , janka h .- th . , 1994 ,
a&a , 286 , 857 takahashi k. , janka h .- th . , 1997 , in _ origin of matter and evolution of galaxies _ , eds .
t. kajino et al . , ( singapore : world scientific ) , p. 213
thompson t.a . , 2003 , apjl , 585 , l33 thompson t.a . , burrows a. , meyer b.s .
2001 , apj , 562 , 887 f. , quataert e. , hopkins p. f. , keres d. , faucher - giguere c .- a .
, 2015 , mnras , 447 , 140 vangioni e. , goriely s. , daigne f. , franois p. , belczynski k. , 2015 , mnras , 455 , 17 vlasov a.d . , metzger b.d . ,
thompson t.a . , 2014 , mnras , 444 , 3537 wallner a. , faestermann t. , gerstmann u. , knie k. , korschine , g. , lierse c. , rugel g. , 2004 , new astronomy reviews 48 , 145 wallner a. , et al .
, 2015 , nature communications , 6 , 5956 wanajo s. , janka h .-
, mller b. , 2011 , apjl , 726 , l15 wanajo s. , et al . , 2014 , apj , 789 , l39 wehmeyer b. , pignatari m. , thielemann f .- k .
, 2015 , mnras , 452 , 1970 westphal a.j . ,
price p.b . , weaver b.a . ,
& afanasiev v.g . , 1998 , nature , 396 , 50 winteler c. , kppeli r. , perego a. , arcones a. , vasset n. , nishimura n. , liebendrfer m. , thielemann f .- k .
, 2012 , apjl ,
l22 witti j. , janka h .- th .
, takahashi k. , 1994 , a&a , 286 , 841 witze a. , 2013 , nature , 12797 woosley s.e . , wilson j.r . ,
mathews g.j .
, hoffman r.d . ,
meyer b.s . , 1994 ,
apj , 433 , 229 xu y , goriely s. , jorissen a. , chen g.l . , arnould m. , 2013 , a&a , 549 , a106 | long - lived radioactive nuclei play an important role as nucleo - cosmochronometers and as cosmic tracers of nucleosynthetic source activity .
in particular nuclei in the actinide region like thorium , uranium , and plutonium can testify to the enrichment of an environment by the still enigmatic astrophysical sources that are responsible for the production of neutron - rich nuclei by the rapid neutron - capture process ( r - process ) .
supernovae and merging neutron - star ( ns ) or ns - black hole binaries are considered as most likely sources of the r - nuclei . but
arguments in favour of one or the other or both are indirect and make use of assumptions ; they are based on theoretical models with remaining simplifications and shortcomings .
an unambiguous observational determination of a production event is still missing . in order to facilitate searches in this direction , e.g. by looking for radioactive tracers in stellar envelopes , the interstellar medium or terrestrial reservoirs , we provide
improved theoretical estimates and corresponding uncertainty ranges for the actinide production ( @xmath0th , @xmath1u , @xmath2np , @xmath3pu , and @xmath4 cm ) in neutrino - driven winds of core - collapse supernovae . since state - of - the - art supernova models do not yield r - process viable conditions
but still lack , for example , the effects of strong magnetic fields we base our investigation on a simple analytical , newtonian , adiabatic and steady - state wind model and consider the superposition of a large number of contributing components , whose nucleosynthesis - relevant parameters ( mass weight , entropy , expansion time scale , and neutron excess ) are constrained by the assumption that the integrated wind nucleosynthesis closely reproduces the solar system distribution of r - process elements .
we also test the influence of uncertain nuclear physics .
[ firstpage ] nuclear reactions , nucleosynthesis , abundances supernovae : general |
for many years now , aoki @xcite and collaborators @xcite have been advocating the existence of a parity - flavor breaking phase in qcd with wilson fermions as a means of explaining why the pion mass in this model approaches small values as the wilson parameter @xmath4 approaches , for every value of inverse square coupling @xmath3 , a critical value @xmath7 .
this in spite of no - go theorems @xcite that forbid such a phase in the continuum limit .
indeed analytic arguments have been presented to support the existence of such a phase at @xmath8 . for
finite and , in particular , larger values of @xmath3 where current lattice simulations are undertaken , such evidence is lacking @xcite .
although the picture advocated by aoki may explain the smallness of the pion masses as @xmath4 approaches @xmath7 it also explicitly states that these pions are not the goldstone modes of spontaneous chiral symmetry breaking .
this presents a problem in that it is not then clear that any of the soft pion and other theorems associated with this phenomenon will be respected on the lattice . in other words
this would not be the expected simulation of true qcd .
in fact the large @xmath9 analytic analysis which indicates the existence of this phase at @xmath8 also shows the non - vanishing of the @xmath10 scattering length which is contrary to the the expected spontaneous chiral symmetry breaking of qcd .
this is of course unimportant at @xmath8 but is very important , if true , at the values of @xmath11 where current simulations are performed .
the alternative picture where the explicit chiral symmetry breaking wilson term causes the ( otherwise goldstone ) pions to acquire a small mass proportional to the lattice spacing does not have such a problem . indeed , that all these extra effects would disappear as the lattice spacing is made smaller with the approach to the continuum limit was formally demonstrated some time ago @xcite .
several models exhibit a parity - flavor breaking phase . in the nambu - johna - lassinio @xcite model with wilson fermions
this phase was numerically and , in the large @xmath9 approximation , analytically @xcite confirmed for values of @xmath3 up to a specific cutoff value .
this phase disappears for larger values of @xmath3 .
the schwinger model with two flavors of wilson fermions exhibits this phase at strong coupling and also loses it at weak coupling @xcite .
whereas the picture advocated by aoki does not anticipate such a quenching effect for the phase in qcd , recent phenomenological arguments by creutz @xcite tend to show a preference for quenching of this phase if it exists .
thus it becomes necessary to explore this important feature by explicit simulations of qcd with wilson fermions on volumes larger than those already studied @xcite .
following arguments presented by aoki and aoki and gocksch @xcite , it is necessary , in order to investigate the presence of the parity - flavor - breaking phase in simulations , to introduce first an explicit breaking term into the action and then extrapolate the measured order parameter as this term tends to zero . since the extrapolation is to be done , in principle , after the ` infinite volume ' limit
is taken , such simulations must be done for larger volumes and any order parameter extrapolation be studied as a function of this increasing volume . in such a situation a typical behaviour of the data at finite volumes that one might expect
is shown in fig .
it is expected that the functional dependence of the ( measured ) order parameter on @xmath5 be such that for any @xmath12 volume this order parameter vanishes at @xmath13 .
the variation of this dependence with increasing volume is crucial to the initial determination of the existence of a broken phase or its absence .
the existence of a broken phase in the infinite volume limit is signaled by a flattening of this dependence for larger values of @xmath5 and a sharper drop to zero as @xmath5 approaches zero .
thus it is clear that a significant volume dependence of the order parameter at smaller values of @xmath5 is a necessary indicative factor for this phase . if , on the other hand , the approach to zero is not varying significantly as the volume increases , the infinite volume limit will not sustain a broken phase .
one may use the slightly varying values of the order parameter there to extrapolate these to @xmath14 @xmath15 . if this step is then followed by a linear extrapolation to @xmath16 a non - zero value for the order parameter at @xmath16 may be obtained . at this stage
it is tempting to conclude that a broken phase exists in that limit
. this may be the wrong conclusion if this is not accompanied by a significant increase in the value of the order parameter at smaller values of @xmath5 . ]
we report here on simulations done with two flavors of wilson fermions at @xmath17 , and @xmath18 on volumes of @xmath19 , and @xmath2 for a variety of values of @xmath4 ranging from less than the appropriate @xmath7 to values greater than @xmath7 .
the choice of these three values of @xmath3 was determined as follows .
the value @xmath20 represents current simulations on larger lattices where spectrum and matrix element calculations are being done ; that at @xmath21 represents a lower value below which relevance to continuum physics is not expected , and the last value at @xmath22 is to extend the search to a much larger value of @xmath3 in case the parity - flavor breaking phase were to be confirmed at the two smaller values .
we introduce into the qcd action a term of the form @xmath23 where @xmath24 is a @xmath25 matrix representing the third element of the generators of flavor su(2 ) algebra . upon integrating the fermionic variables
this is reflected in the simulation by the product of two determinants : @xmath26 where @xmath27 is given by a simple modification of the wilson matrix @xmath28 as : @xmath29 as pointed out by aoki , we also have here @xmath30 and : @xmath31 simulations were done for the parameter @xmath5 taking values ranging from 0.001 to 0.3 . for the volume dependence we concentrate on the smaller values of @xmath5 and in particular @xmath32 and @xmath33 for all three volumes considered and mostly for values of @xmath4 greater than @xmath7 .
the order parameter we compute is the expectation value of the operator @xmath34 . with our notation
this is given as @xmath35
for the three values of @xmath3 considered , simulations were performed , as mentioned above , at various values of @xmath4 both below and above @xmath7 .
we shall present the data and results for each value of @xmath3 considered separately . in all cases these simulations
were also done at various values of the external parameter @xmath5 . for each @xmath4 the results of the compuations on the three volumes @xmath36 , @xmath37 , and @xmath38 were compared at the two values of @xmath39 , and @xmath40 .
the variation of the order parameter with @xmath41 is then used to obtain an ` infinite volume ' limit for all values of @xmath5 used .
the choice of @xmath41 is indicated here by the naive dimension of the order parameter . following this
, the order parameter at these values of @xmath5 were fitted to : @xmath42 a separate fit to the pure quadratic polynomial @xmath43 was also done . the initial aim in this case is to detect the possible existence of any non - zero constant @xmath44 at @xmath16 as the limit of the order parameter at that point .
this is of particular interest for comparing results at values of @xmath4 above @xmath7 with those below @xmath7 .
it is useful to point out here that in the presence of a parity - flavor breaking phase the order parameter is expected to vary with @xmath5 as : @xmath45 this being the behaviour of the root of the cubic equation determining the position of the minimum of the quartic effective potential . in the absence of such a phase the same behaviour follows with @xmath46 .
as the quartic potential becomes quadratic the leading behaviour becomes : @xmath47 this should , when compared to the data , be also a useful tool in determining which situation one is in .
the value of @xmath7 at this value of @xmath3 is known to be about @xmath48 .
we consequently performed simulations well below that value at @xmath49 and well above it at @xmath50 and intermediate values in between .
we present in table [ table - one ] the results of these simulations .
.parameters and measured order parameter @xmath51 for the case of @xmath21 on lattices of volume @xmath36 , for @xmath52 , @xmath53 , and @xmath38 . [ cols="^,^,^,^,^",options="header " , ] [ table - three ] we show in fig . [ fig8 ] the variation of the computed order parameter with @xmath4 over the range used . no sharp change is indicated as the value of @xmath54 is crossed .
we concentrate here on the data at the values of @xmath55 above @xmath7 . using the same procedure as above
essentially the same conclusion follows .
[ fig9]a and b show that no significant change in the evaluation of the order parameter at the larger volumes exists for @xmath56 at @xmath40 and as seen by the overlapping histograms for @xmath57 at @xmath40 .
furthermore , quadratic fits in @xmath5 at , for example , @xmath58 and @xmath59 again have a leading ` constant ' that is consistent with zero as shown in figs .
[ fig10]a and b , respectively .
therefore , one is led again to the absence of any signal for a parity - flavor breaking phase at this value of @xmath3 .
finally the obvious leading linear dependence of the fit indicates again a dominant quadratic effective potential at this value of @xmath3 as well .
it is clear from the discussion above that qcd with two flavors of wilson fermions does not exhibit a parity - flavor breaking phase at @xmath60 as postulated by aoki and collaborators . since it has been demonstrated that at @xmath8 such a phase may exist in a large @xmath9 ( color ) limit , it is also clear that if this phase does extend beyound @xmath8 , it must pinch out in a manner similar to that in the njl model at @xmath61 . in either case
this phase would not be relevant for the discussion of the approach to the chiral limit in qcd and the ensuing goldstone nature of the pions for @xmath62 .
in fact , all indications are such that , as shown formally sometime ago , this is simply related to the approach to zero lattice spacing and infinite volume .
i wish to thank urs heller for providing a modified qcd code for performing the simulations reported in this paper .
all these simulations were done on the scri ibm compute cluster .
this research was supported by the u.s .
department of energy through contract nos .
de - fg05 - 92er40742 and de - fc05 - 85er250000 .
9 s. aoki , phys .
d30 , 2653 ( 1984 ) s. aoki , phys .
d33 , 2339 ( 1984 ) s. aoki , phys .
letters 57 , 3136 ( 1986 ) s. aoki , nucl .
b314 , 79 , ( 1989 ) s. aoki , on the phase structure of qcd with wilson fermions , uthep-318 , august ( 1995 ) s. aoki and a. gocksch .
letters b231 , 449 ( 1989 ) s. aoki and a. gocksch .
letters b243 , 409 ( 1990 ) s. aoki and a. gocksch .
d45 , 3845 ( 1992 ) c. vafa and e. witten , nuclear physics b234 , 173 ( 1984 ) c. vafa and e. witten , phys .
letters 53 , 535 ( 1984 ) l.h .
karsten and j. smit , nucl .
b183 , 103 ( 1981 ) e. seiler and i. stamatescu , phys .
d25 , 2177 ( 1982 ) m. bochicchio , l. maiani , g. martinelli , g. rossi and m. testa , nuclear physics b262 , 331 ( 1985 ) k. m. bitar and p. vranas , phys .
rev d50 , 3406 ( 1994 ) k. m. bitar and p. vranas , nuclear physic b , proc .
34 , 661 ( 1994 ) s. aoki , s. boettcher and a. gocksch , phys .
letters b331 , 157 ( 1994 ) i. horvath , the phase structure of the schwinger model on the lattice with wilson fermions in the hartree - fock approximation , fsu - scri-95 - 101 ; to appear in phys .
d ( 1996 ) , m. creutz , bnl-62123 , may 1995 p. lepage and p.g .
mackenzie , phys .
d48 , 2250 ( 1993 ) | we present data testing the existence of a parity - flavor breaking phase in simulations of qcd with two flavors of light wilson fermions .
this is done by explicit simulations on lattice sizes of @xmath0 , @xmath1 and @xmath2 for a variety of values of @xmath3 and @xmath4 as well as the coefficient , @xmath5 , of an explicit breaking term included in the action .
we find that at @xmath6 equal to or greater than 5.0 extrapolation in the parameter @xmath5 as well as in the lattice volume show no indication of a phase where parity and flavor are spontaneously broken in the limit of zero @xmath5 .
makecaption#1#2 10@ tempboxa tempboxa > _ _ _ _ _ _ # 1 : # 2 _ _ _ _ _ _ tempboxa = -.5truein = -.75truein = 100000 = 10000 |
the system of three identical bosons having a large two - body scattering length has been the subject of intense investigations in recent years . as shown by efimov in a sequence of papers @xcite ,
the three - body spectrum consists of an infinite number of states that accumulate to zero with the ratio between two consecutive states being @xmath4 .
in other words , in the limit @xmath5 the two - body system shows a continuous scale invariance that is broken in the @xmath6-wave three - body sector of a bosonic system .
the residual symmetry is the discrete scale invariance ( dsi ) ; namely , the physics is invariant under the rescaling @xmath7 , where the constant is usually written @xmath8 , with @xmath9 an universal number that characterises a system of three - identical bosons .
the dsi constrains the form of the observables to be log - periodic functions of the control parameters .
one example is the atom - dimer scattering length which has the general form @xmath10\ , , \label{eq : a_ad}\ ] ] where @xmath11 are universal constants @xcite . for atom - dimer collisions below the dimer breakup threshold
, dsi imposes the following universal form for the effective range function @xmath12\ , , \label{eq : cotdelta}\ ] ] with @xmath13 the atom - dimer phase - shift and @xmath14 universal functions of the dimensionless variable @xmath15 , where @xmath16 , being @xmath17 the incident energy in the atom - dimer center of mass frame , and @xmath18 the boson mass .
it is well known that for @xmath19 the effective range function satisfies the limit @xmath20 , which implies that at @xmath21 the constants @xmath11 and @xmath22 are related by simple trigonometric relations .
a parametrization of the universal constants and functions can be found in ref .
@xcite .
the dsi also constrains the form of the @xmath1-matrix for collisions above the dimer threshold , leading to the following peculiar form for the recombination rate at threshold @xcite @xmath23}\frac{\hbar a^4}{m}\ , , \label{eq : recomb}\ ] ] that , using the large value of the factor @xmath24 , can be approximated by @xmath25 a^4\hbar / m \ , , \label{eq : recombsine}\ ] ] where @xmath26 @xcite . in this work we study in detail the universal behavior of @xmath27 by solving the schrdinger equation for a family of attractive two - body gaussian potentials describing the he - he system .
these potentials are constructed to reproduce the two - body binding energy @xmath28 , the two - body scattering length @xmath0 and the effective range @xmath29 of the lm2m2 potential , widely used in the literature @xcite .
the variation of the potential strength produces different values of the scattering length @xmath0 allowing a comparison of the recombination rate given by eq.([eq : recomb ] ) .
the model includes a three - body interaction necessary to tune the trimer energy of the lm2m2 potential .
this procedure is equivalent , up to range corrections , to the implementation of effective field theory ( eft ) at leading order ( lo ) .
this strategy has been used before in the study of the atom - dimer effective range function @xmath30 @xcite .
it has been found that the zero - range formula of eq.([eq : cotdelta ] ) has to be modified in order to describe results obtained through the solution of the schrdinger equation using finite - range potentials .
the modification consists in a shift on the variable @xmath31 , and in the replacement of @xmath30 by @xmath32 , where @xmath33 is related to the two - body binding energy through @xmath34 . in the present analysis
we find that the same type of modification has to be made in order to describe the numerical results .
the universal character of eq.([eq : recomb ] ) allows its application to very different systems .
here we extend the study on atomic systems to describe a nuclear system : the neutron - neutron - proton recombination rate close to threshold .
this study is twofold ; from one side , we would like to confirm that systems whose energies and sizes differ by several order of magnitude are still described by the same universal equation . on the other hand
, very low energy recombination in nuclear systems can be achieved in stars , and the present study can therefore be applied to those systems in which the three - body structure is dominant , e.g. , neutron - neutron - core systems .
as an example we can mention the recently performed studies in recombination of @xmath35h @xcite and low energy @xmath36c collisions @xcite .
finally , we remark that in cold - atom physics the search for universal behavior is a very active sector of research . at present
, there is an intense experimental activity to study efimov physics in trapped ultracold gases . in these systems
the recombination rate has a major importance , being the main loss mechanism .
three different values of the two - body scattering length have relevance : @xmath37 , which characterizes the threshold where the trimer disappears into the three - atom continuum , @xmath38 , which characterizes the threshold where the trimer disappears into the atom - dimer continuum , and @xmath39 , which characterizes a minimum in the recombination rate .
discrepancies are found between the zero - range theoretical predictions and experimental determinations of these three quantities @xcite . to this respect ,
we would like to see if the quantities obtained from the calculations using finite range potentials are in better agreement with the experimental predictions .
the paper is organized as follows . in the next section
the two - body and three - body potential models are introduced and the results for the first two energy levels are discussed .
the main results of this work are given in section iii , which is divided into four subsections dedicated to analyze the elastic and breakup cross sections , the recombination and dissociation rates , comparison of the results to experimental data and nucleon - deuteron scattering above threshold .
the conclusions are given in the last section .
in this section we study universal aspects of a three - boson system by taking the three - helium system as reference . at the two - body level , we consider one of the most commonly used he - he potentials , i.e. , the lm2m2 interaction @xcite , which is taken as the reference interaction . in particular , in order to explore the @xmath40 plane ( @xmath41^{1/2}]$ ] and @xmath42 the energy level ) , we modify this potential as : @xmath43 examples of this strategy exist in the literature @xcite . for @xmath44
the interaction is close to the unitary limit ( @xmath5 ) .
for @xmath45 the values predicted by the lm2m2 are recovered : the scattering length @xmath46 @xmath47 , the two - body energy @xmath28=-1.303 mk , and the effective range @xmath48 @xmath47 , with the mass parameter @xmath49 .
following ref .
@xcite we define an attractive two - body gaussian ( tbg ) potential @xmath50 with range @xmath51 @xmath47 and strength @xmath52 fixed to reproduce the values of @xmath0 given by @xmath53 . for example with the strength @xmath54 k , corresponding to @xmath45 , the lm2m2 low - energy data are closely reproduced , @xmath55 mk , @xmath56 @xmath47 , and @xmath57 @xmath47 the use of the tbg potential in the three - atom system produces a ground state binding energy appreciable deeper than the one calculated with @xmath53 . for example , at @xmath45 the lm2m2 helium - trimer ground - state - binding energy is @xmath58 mk , whereas the one obtained using the two - body - soft - core potential in eq .
( [ eq : twobp ] ) is @xmath59 mk . to solve this discrepancy
we introduce a repulsive hypercentral - three - body ( h3b ) interaction @xmath60 with the strength @xmath61 tuned to reproduce the trimer energy @xmath62 obtained with @xmath53 for all the explored values of @xmath63 . here
@xmath64 is the hyperradius of three identical particles and @xmath65 gives the range of the three - body force .
following ref @xcite we use @xmath66 .
it should be noticed that the description of the three - boson systems using a two- plus three - body interaction constructed to reproduce the low energy data is equivalent , up to finite range corrections , to a description based on eft at lo ( see ref .
@xcite and references therein ) .
varying @xmath63 from the unitary limit to @xmath67 we obtain a set of values for the ground state binding energy @xmath62 and first excited state @xmath68 using the tbg and tbg+h3b potentials in a broad range of @xmath0 .
the results can be compared to the predictions given by the efimov s binding energy equations @xmath69}\frac{\hbar^2\kappa_*^2}{m } \ , , \label{eq : uni3}\ ] ] where @xmath70 and the function @xmath71 can be found in @xcite .
fixing @xmath72 , the three - body parameters @xmath3 is determined by calculating @xmath68 at the unitary limit ; we obtain @xmath73 @xmath74 and @xmath75 for the tbg and tbg+h3b , respectively . .
the dashed line is the universal prediction of the efimov law eq.([eq : uni31 ] ) without shift ( @xmath76 ) , while the solid line is the translated universal curve .
the circles and squares are the calculations using the tbg and tbg+h3b potentials respectively.,title="fig : " ] .
the dashed line is the universal prediction of the efimov law eq.([eq : uni31 ] ) without shift ( @xmath76 ) , while the solid line is the translated universal curve .
the circles and squares are the calculations using the tbg and tbg+h3b potentials respectively.,title="fig : " ] it has been shown in ref .
@xcite that in order to be in accord with the numerical results obtained solving a finite - range potential , the universal relation eq .
( [ eq : uni3 ] ) must be modified in the following way @xmath77}/\cos\xi - \gamma_n\ , , \label{eq : uni31}\ ] ] where the finite - range nature of the interaction has been taken into account by the substitution @xmath78 , and by the shift @xmath79 . in fig .
[ fig : plotnew ] we collect our numerical results for the ratios @xmath80 and @xmath81 as a function of @xmath31 for the tbg potential ( circles ) and of the tbg+h3b potential ( squares ) . in the upper panel of fig.[fig : plotnew ] we report the calculations for the ground state .
the dashed line corresponds to eq .
( [ eq : uni31 ] ) without shift , while the solid line , which fits our numerical results , has @xmath82 . in the lower panel of fig .
[ fig : plotnew ] we report the calculations for the excited state . as
above , the dashed line corresponds to eq .
( [ eq : uni31 ] ) without shift , while the solid has a finite shift , @xmath83 .
a more accurate analysis of the numerical results reveals that the shift depends on @xmath3 @xcite ; in first approximation , for the excited state , we can write @xmath84 with @xmath85 . from eq.([eq : uni31 ] ) we can extract @xmath86 , the scattering length at which the excited state @xmath68 disappears into the atom - dimer continuum .
this happens when @xmath87 or the angle @xmath88 . using the very accurate result , @xmath89 , given in ref .
@xcite , we obtain the relation @xmath90 which in our case gives @xmath91 and @xmath92 for the tbg and tbg+h3b potentials respectively .
moreover , dividing the above relation by @xmath3 , we can write @xmath93 , where we have introduced the universal zero - range scattering length @xmath94 . from this relation
we can interpret the shift , in units of @xmath95 , as the difference between the zero - range and the finite - range predictions for @xmath38 .
in addition , the relation applied to the @xmath96 branch can be extended to the different @xmath97-branches .
in ref . @xcite the universal character of the atom - dimer scattering below the breakup threshold has been discussed in terms of the potential model introduced in the previous section .
it has been shown that when finite - range interactions are used , the universal formula of the atom - dimer scattering length given in eq.([eq : a_ad ] ) can be modified as @xmath98\ , , \label{eq : ab_ad}\ ] ] where @xmath33 is defined from the relation @xmath34 , and @xmath99 is the shift for the atom - dimer scattering moreover , the effective range function of eq.([eq : cotdelta ] ) can be adapted to describe finite - range interactions too . in particular , it is given by @xmath100\ , , \label{eq : abcotdelta}\ ] ] with the effective - range shift @xmath101 @xcite .
this modified equation agrees with the numerical calculations of ref .
@xcite , which have been performed by using the potential models of the previous section and the hyperspherical harmonic ( hh ) method in conjunction with the kohn variational principle @xcite , for a wide range of values of @xmath31 varying from @xmath102 to @xmath103 . in this range
the effective range function presents different patterns as a function of the energy . for the lowest values of @xmath31 it was almost linear whereas when increasing @xmath31 a pole structure appeared ( see fig.4 of ref .
in addition , it was shown that around the value @xmath104 the structure of the effective range function coincides with the one describing neutron - deuteron scattering . in this way , a confirmation of the universal character of eqs.([eq : ab_ad ] ) and ( [ eq : abcotdelta ] ) has been done for systems with very different typical lengths . the fact that @xmath105 can be understood noticing that , for the values of @xmath0 considered , @xmath68 is the only excited state of the three - boson system .
probably , for larger values of @xmath0 , when a second excited state appears , a different shift has to be considered .
the analysis of atom - dimer scattering for larger values of @xmath0 is a numerical difficult task and remains outside the scope of the present work . from eq.([eq :
ab_ad ] ) it is possible to extract the value of @xmath106 by equating the argument of the tangent to @xmath107 . using the value of @xmath108 given in ref .
@xcite , the results are @xmath109 and @xmath110 for the tbg and tbg+h3b potentials respectively , in complete agreement with those obtained using eq.([eq : uni31 ] ) . these values can be used to evaluate the ratio @xmath111 that appears frequently in the literature from measurements on trapped ultracold atoms @xcite .
the universal zero - range theory predicts @xmath112 whereas using the values @xmath113 and @xmath114 for the tbg and tbg+h3b potentials respectively , given in ref .
@xcite , we obtain @xmath115 in both cases . an analysis of the present result in comparison to those given by different experimental groups is given in the section iii c. the universal character of eqs.([eq : ab_ad ] ) and ( [ eq : abcotdelta ] ) has been deeply studied in ref .
@xcite in a large range of @xmath0 values . here
we extend the analysis of atom - dimer scattering to energies above the breakup threshold .
in particular , we study the universal form predicted by petrov @xcite for the recombination rate at threshold . to this aim
we make use of the adiabatic expansion method as discussed in refs .
@xcite for energies below the dimer breakup threshold and recently extended to energies above that threshold @xcite .
as it is well known , the adiabatic expansion method is a very powerful method used to describe bound states @xcite .
however , the extension to describe scattering states encountered some difficulties , in particular in the case of atom - dimer elastic scattering .
the problem arose from the difference between the set of coordinates in which the process has a natural asymptotic description ( the usual jacobi coordinates ) and the expansion in terms of hyperradial functions which produced a low rate of convergence . in order to circumvent the problem ,
two integral relations have been derived in ref .
@xcite and already applied several times in the literature @xcite .
essentially the method establishes that the scattering matrix is obtained from the following two matrices @xmath116 where the indexes @xmath117 label the ingoing and outgoing channels ( either elastic or inelastic ) , @xmath118 is the solution of the adiabatic equations at a given energy @xmath42 , and @xmath119 are the ingoing and outgoing solutions of the free schrdinger equation @xmath120 ( see ref .
@xcite for details ) . the @xmath121-matrix is then given by the product @xmath122 .
the integral relations have a short - range character , and the problem mentioned above of the mismatch of the coordinates has not consequences once the internal part of the scattering wave function is properly described . for energies above the dimer breakup threshold
, the unitarity of the @xmath121-matrix implies that given an incoming channel , for instance channel 1 ( 1 + 2 channel ) , we have that @xmath123 which means that computation of the elastic term @xmath124 amounts to computation of the infinite summation of the @xmath125 terms ( @xmath126 ) corresponding to the breakup channels . the complex value of @xmath124 can be written in terms of a complex phase - shift @xmath13 as : @xmath127 the value of @xmath128 gives the probability of elastic atom - dimer scattering , and @xmath129 is usually referred to as the inelasticity parameter ( for example denoted by @xmath130 in @xcite ) .
obviously , the closer the inelasticity to 1 the more elastic the reaction . in fact , for energies below the breakup threshold the phase - shift is real and @xmath131 . for two different values of @xmath31 ; in the upper panel @xmath132 , and in the lower panel @xmath133 .
the dimer threshold corresponds to @xmath134 .
the ( red ) circles are the calculations below the dimer threshold of ref .
the ( green ) triangles are the present calculations .
the solid line is eq.([eq : abcotdelta ] ) whereas the dashed line is an effective range parametrization ( see text).,title="fig : " ] for two different values of @xmath31 ; in the upper panel @xmath132 , and in the lower panel @xmath133 .
the dimer threshold corresponds to @xmath134 .
the ( red ) circles are the calculations below the dimer threshold of ref .
the ( green ) triangles are the present calculations .
the solid line is eq.([eq : abcotdelta ] ) whereas the dashed line is an effective range parametrization ( see text).,title="fig : " ] the convergence pattern of the method has been studied in ref .
@xcite , where the inelasticity ( @xmath129 ) and the real part of the phase - shift ( re(@xmath13 ) ) have been calculated at selected energies for increasing values of @xmath135 ( grand - angular quantum number associated to the last adiabatic term included in the expansion of the scattering wave function ) .
the conclusion was that the use of the integral relations produces a pattern of convergence similar to a bound - state calculation .
for example , a @xmath135 value of around 12 is enough to get a rather well converged inelasticity , while re(@xmath13 ) requires a few more adiabatic terms in order to reach convergence . to make contact with the results in ref .
@xcite , in fig .
[ fig : redelta ] we show the effective range function of the tbg potential corresponding to @xmath136 ( upper panel ) , and to @xmath137 ( lower panel ) for a wide range of energies . for energies below the dimer threshold
the circles ( red online ) are the results of ref .
@xcite , whereas for energies above the dimer threshold , our results obtained using re(@xmath13 ) in the definition of the effective range function are given by the triangles ( green online ) .
the solid line has been obtained applying eq.([eq : abcotdelta ] ) , and the dashed line is the effective range parametrization of the low energy points ; in the upper panel , the effective - range is @xmath138 , with @xmath139 used as a parameter . in the lower panel ,
a pole - structure dominates the low - energy behaviour of the effective - range function , which is better parametrized by @xmath140 ; in addition to the effective range we must introduce the shape parameter @xmath141 and the momentum of the pole @xmath142 . from the figure we can observe that the universal form of eq.([eq : abcotdelta ] ) does not describe the points above the dimer threshold . on the other hand , the effective range parametrization remains close to the computed values in a larger range .
the zoom shows the behavior close to both thresholds , i.e. , @xmath143 for the elastic case and @xmath144 for the breakup case .
the circles correspond to actual calculations , while the curves are interpolations . ] limiting the discussion to the @xmath145 channel , the determination of the elastic matrix element @xmath124 permits the computation of the elastic ( @xmath146 ) and breakup ( @xmath147 ) cross sections , which are given by the well - known expressions : @xmath148 with @xmath16 and @xmath149 , the total energy of the process . following the method described in @xcite , we have computed @xmath146 and @xmath147 for the tbg+h3b potential ( @xmath45 case ) .
the results are given by the solid and dashed curves in fig .
[ fig : sigmas ] , respectively , where the cross sections are shown for incident energies @xmath17 up to @xmath150 mk . at this energy
both cross sections have a similar size . in the figure the corresponding behavior at threshold
are zoomed . in the case of the elastic scattering ,
when @xmath17 approaches 0 the cross section reaches the constant value of @xmath151 . in the case of the breakup cross section
the threshold corresponds to @xmath152 ( or @xmath144 ) , and we have that for @xmath153 the breakup cross section @xmath147 behaves as @xmath154 ( or , in other words , @xmath155 ) . the breakup cross section described in the previous subsection is directly related to the dissociation rate , @xmath156 , for the @xmath157 process .
in particular , for three identical bosons with angular momentum and parity @xmath158 it takes the form @xcite : @xmath159 where @xmath160 is the atom - dimer reduced mass and @xmath18 is the mass of the atom . also , making use of the detailed balance principle , it is possible to relate @xmath147 to the cross section corresponding to the inverse process @xmath161 .
this permits to obtain the recombination rate , @xmath162 , for such process in terms of the inelasticity parameter .
in particular , again for three identical bosons with angular momentum and parity @xmath158 , @xmath162 takes the form @xcite : @xmath163 where @xmath164 and @xmath165 . and
the dissociation rate @xmath156 for three helium atoms with the tbg+h3b potential ( @xmath45 ) as a function of the three - body energy @xmath42 . ]
the @xmath162 behavior at low energies results proportional to @xmath166 as demonstrated in ref .
@xcite ( @xmath167 is the smallest grand - angular quantum number associated with the continuum adiabatic channels ) . in the present case
@xmath168 and @xmath162 is almost constant as @xmath169 . in the case of @xmath156 ,
its low energy behavior follows the @xmath170 rule derived in ref .
these behaviors can be seen in fig.[fig : recom ] in which @xmath162 and @xmath156 are displayed as a function of the three - body energy @xmath42 for the tbg+h3b calculation ( @xmath45 case ) .
this low - energy behavior can also be deduced from the behavior of @xmath147 close to threshold shown in the zoom of fig .
[ fig : sigmas ] , and in particular from the fact that when the total energy goes to zero we have that @xmath171 . at threshold for different values of the product @xmath31 .
the circles ( red online ) are the results using the tbg potential whereas the squares ( green online ) are the results using the tbg+h3b potential .
the points have been fitted with eq.([eq : abrecom ] ) obtaining @xmath172 . ]
the recombination rate at threshold can be defined as @xmath173 . in ref .
@xcite petrov derived a log - periodic function for @xmath27 given in eq.([eq : recomb ] ) , whose simplified version is given by eq .
( [ eq : recombsine ] ) . in order to use this formula to describe our numerical results , which are obtained using finite - range potentials , we introduce the following modification in the definition of @xmath162 : @xmath174}\frac{\hbar a_b^4}{m } \label{eq : abrecom}\ ] ] with the simplified form @xmath175 ( \hbar a_b^4/m ) \ , , \label{eq : abrecombsine}\ ] ] where , as in the effective range function , we have replaced @xmath0 by @xmath33 and we have introduced the shift @xmath176 in the variable @xmath31 .
our results are shown in fig .
[ fig : recomth ] as circles ( tbg potential ) and squares ( tbg+h3b potential ) .
the dashed line represents the universal function of eq.([eq : recomb ] ) with the value @xmath177 from ref .
it is interesting to see that the calculated points organize in a curve shifted with respect to the universal curve .
the solid line represents eq.([eq : abrecom ] ) , with the same value of @xmath178 and @xmath179 . by equating the argument of the sine to zero we can extract @xmath180 , the value of @xmath0 at which the recombination rate has a minimum .
we obtain the relation @xmath181 which in our case results @xmath182 and @xmath183 for the tbg and tbg+h3b potentials respectively . with the values @xmath184 and @xmath185 for the tbg and tbg+h3b potentials respectively , given in ref .
@xcite , we obtain @xmath186 and @xmath187 to be compared to the theoretical prediction of -4.9 . moreover , defining @xmath188 and dividing the above relation by @xmath3 , we can write @xmath189 , where we have introduced the universal zero - range scattering length @xmath190 . in the present case
we obtain @xmath191 . as in the case of atom - dimer scattering
, we can interpret the shift , in units of @xmath95 as the difference between the zero - range and the finite - range predictions for @xmath39 .
moreover , the relation applied above to the @xmath96 branch can be extended to the other branches . at threshold as a function of @xmath81 .
the circles ( red online ) are the results using the tbg potential whereas the squares ( green online ) are the results using the tbg+h3b potential .
the dashed line corresponds to the zero - range theory , while the solid line is the translation of the zero - range curve . ] finally , in fig .
[ fig : alphaeratio ] we report our results in the form of a phillips plot , as has usually been done in the literature to study the correlation between different three - body observables ( see for instance refs .
in addition to our numerical data , circles for tbg potential and squares for tbg+h3b potential , we report the zero - range calculation without ( dashed line ) and with ( solid line ) translation .
we observe that the translated curve fits quite well the numerical data .
in fact , fig . [
fig : alphaeratio ] can be understood as the combination of fig .
[ fig : plotnew ] , lower panel , and fig .
[ fig : recomth ] , where the parameter @xmath31 has been eliminated .
we have shown that the finite - range effects manifest as a translation ( the shifts ) in the parameter @xmath31 , see for instance eqs .
( [ eq : uni31 ] ) , ( [ eq : ab_ad ] ) , ( [ eq : abcotdelta ] ) , ( [ eq : abrecom ] ) , and ( [ eq : abrecombsine ] ) .
now , the smallness of the shifts justifies the fact that the curve resulting from the elimination of the parameter @xmath31 is itself a translated curve , at least at the first order . if the shifts @xmath192 and @xmath176 were the same , the numerical calculations would had fallen over the zero - range curve without translation .
this has been for instance demonstrated in ref .
@xcite in the case of the atom - dimer scattering length as a function of the excited three - body energy ( cfr .
fig . 2 of ref .
@xcite ) . in the following
we compare our numerical results using the tbg+h3b to selected results given in the literature by different groups .
we concentrate in the results given for @xmath37 , @xmath38 and @xmath39 extracted from resonances observed in trapped ultracold atoms .
the value of @xmath37 is extracted from a maximum in the recombination rate in the negative sector of @xmath0 whereas @xmath38 corresponds to a maximum in an atom - dimer resonance , and @xmath39 to a recombination minimum in the positive sector of @xmath0 . in table
[ tab : expres ] we collect selected experimental results of these quantities to be compared to the results obtained using the tbg+h3b potential given in the last column . as the numerical results , the measured ratios do not completely agree with the zero - range universal theory .
this disagreement can be traced back to the shift introduced in the zero - range theory produced by the finite range nature of the potential .
this shift which it is not universal produces differences between the different atomic species .
however , in all cases the ratio @xmath193 , where @xmath194 is the natural - atomic length given by the van der waals ( vdw ) part of atomic potentials , introduces some common behavior as was recently justified @xcite .
motivated by this fact we extract the values of @xmath195 and @xmath196 from the measured experiments . to this end , we can define @xmath197 from the experimental value of @xmath198 to determine @xmath199 and @xmath200 and then calculate the two lengths , which using eq .
( [ eq : astar ] ) and ( [ eq : apiu ] ) can be expressed as @xmath201 the results in units of @xmath194 are given in the last two rows of the table .
as general trends , we can observe that the absolute value of the two lengths is of order of 5@xmath202 ( here potassium is the exception with a rather bigger value of @xmath196 ) , and that @xmath203 is slightly bigger than @xmath204 , but the values can be both positives and negatives .
the results of the present work agree in order of magnitude with the experimental data , while the sign of of both @xmath196 and @xmath195 are probably related to the details of the specific feshbach resonances . in the presented work
, the resonance has been explored by changing the strength of the potential , see eq .
( [ eq : mtbp ] ) , which is a way to simulate a broad resonance but not a narrow one , and the values of @xmath196 and @xmath195 we obtain are positives .
though this analysis is not conclusive the results show that the shifted formulae can be adapted to describe the experimental results after the determination of @xmath195 and @xmath196 .
even if several efforts have been produced to include finite - range corrections in the interpretation of experimental data @xcite , our results suggest that new experimental as well as theoretical efforts are in order to clarify this kind of corrections .
c|c c c c c & @xmath205li@xcite & @xmath205li@xcite & @xmath206k@xcite & @xmath207cs@xcite & present work @xmath208 & 32.5 & 32.5 & 64.5 & 101.0 & 5.1 @xmath209 & -241(8 ) & -280(12 ) & -690(40 ) & -872(22 ) & -48.1 @xmath210 & & & & & -975 @xmath211 & 426(20 ) & 196(4 ) & 930(40 ) & 367(13 ) & 16.0 @xmath212 & 88(4 ) & & 224(7 ) & 210 & @xmath213 & 1402(100 ) & 1130(120 ) & 5650(900 ) & & 135 @xmath214 & -1.77 & -0.7 & -1.35 & -0.42 & -0.32 @xmath215 & -2.7 & & -3.08 & -4.2 & -7.2 @xmath216 & -0.17 & -0.25 & -0.12 & & -0.35 @xmath217 & -7.4 & -8.6 & -10.7 & -8.6 & -9.4 @xmath218 & -5.5 & 2.9 & -3.4 & 5.3 & 4.1 @xmath219 & -9.0 & 4.9 & -38.5 & @xmath220 & 6.2 to test the universal description of the recombination rate encoded in the threshold behavior of the @xmath121-matrix we study a system which differs by order of magnitude on length and energy scales with respect to three - helium atoms . in particular , we shall consider the case of neutron - deuteron ( @xmath221 ) scattering . an initial analysis for energies below the deuteron breakup threshold has been done in ref .
@xcite where it has been shown that the universal formula of eq.([eq : abcotdelta ] ) describes this process quantitatively with @xmath222 and @xmath223fm .
these values have been estimated from the pole energy of @xmath224 kev and the doublet @xmath221 scattering length given in ref . @xcite . in this reference calculations of @xmath221 scattering
have been done using the @xmath6-wave spin dependent potential of malfliet and tjon ( mt ) i - iii model defined as @xmath225 where @xmath6 and @xmath226 are the singlet and triplet spin states , @xmath227 is given in fm , and the potential in mev . with @xmath228=41.47 mev @xmath229 the triplet potential leads to a binding energy for the deuteron @xmath230 mev .
the predictions for the singlet and triplet scattering lengths are @xmath231fm and @xmath232fm respectively , very close to the experimental results .
the scattering lengths are substantially bigger than the effective range of the system @xmath233fm allowing an analysis in terms of the universal formulae . and the dissociation rate @xmath156 for the @xmath234 state in the nucleon - deuteron system as a function of the energy . ] using the potentials above and the procedure described in @xcite we have calculated the corresponding recombination and the dissociation rates , which are shown in fig .
[ fig : recomnd ] as a function of the three - body energy @xmath42 .
the figure shows the @xmath235 state which is the only one having a @xmath168 contribution , being then the most important one at low energies .
when the threshold is approached ( @xmath153 ) the rates have the same behavior as in the atomic case , i.e. , @xmath162 goes to a constant and @xmath156 goes to zero as @xmath236 ( as shown by the dashed line in the right part of the figure ) . at threshold
we obtain @xmath237 @xmath238/s .
it is possible to analyze the behavior of @xmath162 at threshold in the context of the universal function of eq.([eq : abrecom ] ) .
however , it should be noticed that we can not simply replace the value of the triplet scattering length , since both channels , the spin triplet and the spin singlet , contribute to the process . to determine the value of the scattering length to be used ,
let us have a look into the @xmath235 scattering wave function @xmath239 , which is usually expanded in terms of a complete set of adiabatic functions as @xcite @xmath240 where @xmath241 are the hyperradial and hyperspherical coordinates and the @xmath242 functions are in fact the eigenfunctions of the angular part of the schrdinger ( or faddeev ) equations .
the angular eigenfunctions are typically expanded in terms of the hyperspherical harmonics , such that for our case with total angular momentum @xmath145 , @xmath242 is written as : @xmath243,\ ] ] where @xmath244 are hh functions , @xmath245 are the coefficients in the expansion , and we have made explicit the contributions from the possible spin states @xmath246 of two nucleons , which after coupled to the spin of the third nucleon give the total angular momentum of @xmath247 .
the @xmath96 term describes asymptotically the elastic channel and @xmath248 is proportional to the deuteron wave function .
the infinitely many remaining adiabatic channels describe the breakup process @xcite . among them , the only one that asymptotically contains contributions from hyperspherical harmonics with grand - angular momentum @xmath249 is the lowest adiabatic breakup channel ( the @xmath250 term ) .
this is the channel responsible for the behavior of the reaction rates at threshold . in fig.[fig :
psind ] the coefficients of the @xmath251 wave function are shown as a function of the hyperradius .
all of the coefficients with @xmath252 go to zero as @xmath253 , but for @xmath249 the two coefficients @xmath254 and @xmath255 go to @xmath247 and @xmath256 , respectively . in the figure the jacobi coordinates have been depicted indicating by @xmath257 the @xmath258 relative distance and by @xmath259 the relative distance of the third nucleon to the @xmath258 center of mass , so @xmath246 is the @xmath258 spin .
the shadow box shows explicitly the two surviving terms when @xmath260 increases . .
the picture show the jacobi set of coordinates used whereas the shadow box indicates the two coefficients surviving at long distances . ]
this analysis implies that @xmath162 receives contribution from the singlet and triplet channel properly weighted .
therefore in the universal formula the scattering length @xmath0 has to be considered as an effective scattering length @xmath261 fm , obtained from the average @xmath262 .
moreover , the value of @xmath33 results @xmath263fm if in the average @xmath264 is replaced by @xmath265 . with the values for @xmath0 ,
@xmath33 and the product @xmath266 ( obtained from the @xmath221 value of 0.578 multiplied to the ratio of @xmath267 ) , we obtain from eq.([eq : abrecom ] ) the recombination rate value of @xmath268 @xmath238/s , very close to the computed value .
this qualitative analysis can be used as an starting point in the spin analysis of the universal aspects of the recombination rate in light nuclear systems .
in the present paper we have investigated atom - dimer scattering above the breakup threshold using a set of gaussian potentials constructed to reproduce the low - energy atom - atom system .
we have introduced a three - body force in order to reproduce the trimer binding energies given by the lm2m2 potential multiplied by a constant @xmath63 varied from @xmath269 to @xmath270 in order to cover a large part of the @xmath40 plane .
we have studied the total elastic and the total breakup cross sections , as well as the recombination and dissociation rates , @xmath162 and @xmath156 , which are directly related to the latter .
we have payed special attention to the behavior of the system close to threshold , and we have investigated the universal behavior of @xmath162 for different values of the two - body scattering length @xmath0 .
our aim here was to analyze the differences between the universal zero - range theory which , for @xmath162 is given by eq.([eq : recomb ] ) , and more realistic finite range calculations .
this study started already in ref .
@xcite , where this kind of differences have been studied for the first excited level of the three - atom system and in the effective range function . in that reference
it was found that finite - range results organize in a curve shifted with respect to the zero - range theory .
here we have extended the analysis to the ground state energy and to the recombination rate .
our main results are given in fig .
[ fig : recomth ] where we can observe that the computed values for @xmath162 lie on a shifted curve in the variable @xmath31 .
additional range corrections come from the replacement of @xmath0 by @xmath33 in the @xmath271 factor .
this type of correction has been already introduced in the study of @xmath272 and @xmath273 as well as in the binding energies @xmath62 and @xmath68 which have been divided by @xmath28 instead of @xmath274 , as indicated by the zero - range theory .
the fact that the finite - range results lie on a shifted curve in the variable @xmath31 for the binding energies and the low - energy scattering quantities strongly support this particular type of correction .
following this observation , we have proposed eqs.([eq : ab_ad ] ) , ( [ eq : abcotdelta ] ) , and ( [ eq : abrecom ] ) as modifications to the zero - range universal formulae where we have introduced the shifts @xmath275 and @xmath176 . from the calculations we get that the shift of the first excited @xmath276 whereas @xmath176 is slightly larger . we have argued that in the range of @xmath0 studied the three - boson system has only one excited state making the shift of the same order . in the case of the recombination rate
one more channel is open and , in terms of adiabatic potentials , at least two channels have to be considered ( see for example section 6 of ref .
@xcite ) . as a consequence
a larger amount of potential energy appears increasing the shift .
this intuitive arguments can be stated formally ; work along this line is in progress .
most of the work presented here has been focussed in the necessity of adapting the universal formulae obtained in a zero - range theory to the physical case of finite - range interactions . accordingly , the modifications introduced in the three - helium system model similar modifications to be considered in the analysis of the experiments on efimov resonance in ultracold gases .
the analysis of the experimental results given in section iii.c was directed to see if there is experimental evidence of the lengths @xmath195 and @xmath196 introduced in the parametrization of the shift . we have observed some common behavior of the different species when the lengths are given in units of @xmath194 .
this fact encourages this kind of analysis and suggests that more theoretical as well experimental efforts can be done in this direction . in the last part of this study we have focused on the study of nucleon - deuteron scattering closely above the breakup threshold . using the spin dependent mt i
- iii potential we have shown that the @xmath162 and @xmath156 rates have a behavior close to threshold similar to the one observed in the three - atom systems .
we have applied the universal formula to this system using a spin average scattering length and we have seen that it was possible to describe the computed value of @xmath162 . this qualitative analysis can be considered as a first step in the study of universal aspects of nucleon - deuteron scattering considering the full spin dependence of the system .
e.g. acknowledges financial support provided by dgi of mineco ( spain ) under contract no .
fis2011 - 23565 .
32ifxundefined [ 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 10.1016/0370 - 2693(70)90349 - 7 [ * * , ( ) ] @noop * * , ( ) , link:\doibase 10.1016/j.physrep.2006.03.001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.83.1566 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.83.1751 [ * * , ( ) ] @noop `` , '' ( ) , link:\doibase 10.1063/1.460139 [ * * , ( ) ] http://arxiv.org/abs/1212.3457 [ ( ) ] http://arxiv.org/abs/1301.1905 [ ( ) ] link:\doibase 10.1016/j.physletb.2008.10.030 [ * * , ( ) ] link:\doibase 10.1038/nphys1334 [ * * , ( ) ] link:\doibase 10.1007/s00601 - 011 - 0260 - 7 [ * * , ( ) ] http://arxiv.org/abs/1302.0281 [ ( ) ] link:\doibase 10.1103/physrevlett.108.210406 [ * * , ( ) ] link:\doibase 10.1103/physreva.54.394 [ * * , ( ) ] link:\doibase 10.1103/physreva.64.042514 [ * * , ( ) ] link:\doibase 10.1103/physreva.86.042513 [ * * , ( ) ] link:\doibase 10.1007/s00601 - 011 - 0226 - 9 [ * * , ( ) ] \doibase doi : 10.1016/s0375 - 9474(98)00650 - 2 [ * * , ( ) ] link:\doibase 10.1016/s0375 - 9474(97)81832 - 5 [ * * , ( ) ] link:\doibase 10.1007/s00601 - 010 - 0151 - 3 [ * * , ( ) ] link:\doibase 10.1103/physreva.83.022705 [ * * , ( ) ] link:\doibase 10.1103/physreva.86.052709 [ * * , ( ) ] \doibase doi : 10.1016/s0370 - 1573(00)00107 - 1 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.103.090402 [ * * , ( ) ] link:\doibase 10.1103/physrevc.86.044330 [ * * , ( ) ] link:\doibase 10.1103/physrevc.42.1838 [ * * , ( ) ] link:\doibase 10.1103/physrevc.51.2356 [ * * , ( ) ] link:\doibase 10.1103/physrevc.64.024002 [ * * , ( ) ] link:\doibase 10.1103/physreva.78.062701 [ * * , ( ) ] link:\doibase 10.1103/physreva.65.010705 [ * * , ( ) ] link:\doibase 10.1103/physrevc.39.1261 [ * * , ( ) ] | we investigate universal behavior in the recombination rate of three bosons close to threshold . using the he - he system as a reference , we solve the three - body schrdinger equation above the dimer threshold for different potentials having large values of the two - body scattering length @xmath0 .
to this aim we use the hyperspherical adiabatic expansion and we extract the @xmath1-matrix through the integral relations recently derived . the results are compared to the universal form , @xmath2 $ ] , for different values of @xmath0 and selected values of the three - body parameter @xmath3 . a good agreement with the universal formula
is obtained after introducing a particular type of finite - range corrections , which have been recently proposed by two of the authors in a. kievsky and m. gattobigio , phys .
rev .
a * 87 * , 052719 ( 2013 ) .
furthermore , we analyze the validity of the above formula in the description of a very different system : neutron - neutron - proton recombination .
our analysis confirms the universal character of the process in systems of very different scales having a large two - body scattering length . |
to each simple lie algebra @xmath2 one can relate a toda field theory ( tft ) in @xmath0 dimensions .
it allows the lax representation : @xmath3=0\ ] ] where @xmath4 and @xmath5 are first order ordinary differential operators : @xmath6 whose potentials take values in @xmath2 . here
@xmath7 - the cartan subalgebra of @xmath8 , @xmath9 is its dual @xmath10-component vector , @xmath11 , and @xmath12 by @xmath13 we denote the set of admissible roots of @xmath2 , i.e. @xmath14 where @xmath15 are the simple roots of @xmath16 and @xmath17 is the minimal root of @xmath18 .
the corresponding tft is known as the affine tft .
the dynkin graph that corresponds to the set of admissible roots of @xmath19 is called extended dynkin diagrams ( edd ) .
the equations of motion are of the form : @xmath20 where @xmath21 are the minimal positive integers for which @xmath22 . the operators @xmath4 and @xmath5 are invariant with respect to the reduction group @xmath23 where @xmath24 is the coxeter number of @xmath18 .
this reduction group is generated by two elements satisfying @xmath25 which allow realizations both as elements in @xmath26 and in @xmath27 .
the invariance condition has the form @xcite : @xmath28 where @xmath29 and @xmath30 . here
@xmath31 are automorphisms of finite order of @xmath2 , i.e. @xmath32 while @xmath33 are conformal mappings of the complex @xmath34-plane
. the algebraic constraints ( [ eq:3.1 ] ) are automatically compatible with the evolution .
a number of nontrivial reductions of nonlinear evolution equations can be found in @xcite .
[ lem:1 ] let @xmath2 be a simple lie algebra from one of the classical series @xmath35 , @xmath36 , @xmath37 or @xmath38 and let @xmath24 be its coxeter number and @xmath39 the dimension of the typical representation . then the characteristic equation of @xmath40 taken in the typical representation has the form : @xmath41 [ rem:2 ] the constant @xmath42 for @xmath43 ; @xmath44 for @xmath45 and @xmath46 for @xmath47 .
solving the inverse scattering problem in the last two cases requires special treatment of the subspaces related to @xmath48 .
the present paper is an extension of a conference report of one of us @xcite . in section 2
we outline the spectral properties of the lax operator . in section 3
we generalize the theory of rhf proposed in @xcite to the atft in @xmath49 dimensions . in section 4
we show how the involution @xmath50 of the complexified hamiltonian dynamics ( defined in section 3 below ) can be related to an automorphism @xmath51 of the corresponding edd .
this allows one to construct the rhf of atft related to any kac - moody algebra .
we present here ( a non - exhaustive ) list of examples of rhf of atft related to several of the classical series of the kac - moody algebras . in deriving them
we make use of the method proposed in @xcite for constructing @xmath52-symmetries of the edd ; of course our construction needs only involutions ( @xmath53-reductions ) of edd .
the reduction conditions ( [ eq:3.1 ] ) lead to rather special properties of the operator @xmath4 . along with @xmath4
we will use also the equivalent system : @xmath54 = 0,\ ] ] where @xmath55 . combining the ideas of @xcite with the symmetries of the potential ( [ eq:3.1 ] )
we can construct a set of @xmath56 fundamental analytic solutions ( fas ) @xmath57 of ( [ eq : l - t ] ) and prove that : 1 .
the continuous spectrum @xmath58 of @xmath4 fills up @xmath56 rays @xmath59 passing through the origin : @xmath60 ; 2 .
@xmath57 is a fas of ( [ eq : l - t ] ) analytic with respect to @xmath61 in the sector @xmath62 satisfying @xmath63 ; 3 . to each @xmath59 one relates a subalgebra @xmath64 such that @xmath65 for @xmath66 and @xmath67 .
the symmetry ensures that each of the subalgebras @xmath68 is a direct sum of @xmath69-subalgebras ; 4 . on @xmath58 the fas @xmath57
satisfy @xmath70 where @xmath71 is the subgroup with lie algebra @xmath72 and @xmath73 is determined by the dispersion law of the nlee : @xmath74 ; 5 .
the fas of ( [ eq : l - t ] ) satisfy : @xmath75 where @xmath76 is equivalent to the coxeter automorphism @xcite : @xmath77 obviously @xmath78 and @xmath79 ; 6 .
the fas @xmath57 satisfy one of the following two involutions : @xmath80 where @xmath81 , @xmath82 is conveniently chosen weyl group element , or @xmath83 these relations lead to the following constraints for the sewing functions @xmath84 and the minimal set of scattering data : @xmath85 if @xmath4 has no discrete eigenvalues the minimal set of scattering data is provided by the coefficients of @xmath86 and @xmath87 .
all the other sewing functions @xmath84 can be determined from them by applying ( [ eq : g_nu1 ] ) , ( [ eq : g_nu2 ] ) or ( [ eq : g_nu1 ] ) , ( [ eq : g_nu3 ] ) .
the lax representations of the atft models widely discussed in the literature ( see e.g. @xcite and the references therein ) are related mostly to the normal real form of the lie algebra @xmath18 , see @xcite .
our aim here is to : 1 .
generalize the atft to complex - valued fields @xmath88 , and 2 .
describe the family of rhf of these atft models .
we also provide a tool to construct new inequivalent rhf s of the atft .
this tool is a natural generalization of the one in @xcite to @xmath0-dimensional systems .
indeed , the atft related to the algebra @xmath89 can be written down as an infinite - dimensional hamiltonian system as follows : @xmath90 where @xmath91 and @xmath92 are the canonical momenta and coordinates satisfying canonical poisson brackets : @xmath93 next we introduce an involution @xmath94 acting on the phase space @xmath95 as follows : @xmath96 here @xmath97 , @xmath98 and the hamiltonian @xmath99 are functionals on @xmath100 depending analytically on the fields @xmath101 and @xmath102 .
the complexification of the atft is rather straightforward .
the resulting complex atft ( catft ) can be written down as standard hamiltonian system with twice as many fields @xmath103 , @xmath104 , @xmath105 : @xmath106 @xmath107 the densities of the corresponding hamiltonian and symplectic form equal @xmath108 the family of rhf then are obtained from the catft by imposing an invariance condition with respect to the involution @xmath109 where by @xmath110 we denote the complex conjugation . the involution @xmath111 splits the phase space @xmath112 into a direct sum @xmath113 where @xmath114 the phase space of the rhf is @xmath115 . by @xmath116 and @xmath117
we denote the eigensubspaces of @xmath94 , i.e. @xmath118 for any @xmath119 . thus to each involution
@xmath94 satisfying 1 ) - 3 ) one can relate a rhf of the atft . due to the condition 3 ) @xmath120 must preserve the system of admissible roots of @xmath2 ; such involutions can be constructed from the @xmath53-symmetries of the extended dynkin diagrams of @xmath18 studied in @xcite .
in fact one can relate atft models not only to the algebra @xmath89 @xcite , but also to each kac - moody algebra @xcite ; for the theory of kac - moody algebras see @xcite .
the construction is as follows .
let @xmath121 be the set of admissible roots of @xmath2 .
then the corresponding atft model is given by : @xmath122 in what follows we will need also the integer coefficients @xmath123 that provide the decomposition of the @xmath17 over the simple roots of @xmath2 : @xmath124 just like for the @xmath89 case one can consider the complexification : @xmath125 in order to construct the rhf of the atft we will make use of specifically constructed involutions @xmath120 of the phase space @xmath126 . an important fact in our construction will be that to each involution @xmath120 there exist a dual involution ( involutive automorphism ) @xmath127 of @xmath2 .
property 3 ) above along with the definition ( [ eq : h - g ] ) of @xmath128 implies that @xmath127 preserves the set of admissible roots @xmath13 . for brevity below we will skip the dependence of @xmath129 and @xmath92 on @xmath130 and @xmath131 .
indeed , the condition 3 ) above requires that : @xmath132 and therefore we must have @xmath133 .
the relation ( [ eq:*1 ] ) defines uniquely the relation between @xmath120 and @xmath127 .
using @xmath134 we can split the root space @xmath135 into direct sum @xmath136 of two eigensubspaces of @xmath127 .
taking the average of the roots @xmath137 with respect to @xmath127 we get : @xmath138 by construction the set @xmath139 will be a set of admissible roots for some kac - moody algebra with rank @xmath140 .
graphically each set of admissible roots can be represented by an extended dynkin diagrams .
therefore one can relate an automorphism @xmath127 to each @xmath53 symmetry of the extended dynkin diagram .
the splitting of @xmath135 naturally leads to the splittings of the fields : @xmath141 where @xmath142 and @xmath143 . if we also introduce : @xmath144 [ rem : a ] in enumerating the vectors @xmath145 and @xmath146 we take into account only those values of @xmath147 which lead to non - vanishing results in the r.h.sides of ( [ eq:*2 ] ) and ( [ eq:*4 ] ) . then applying the ideas of @xcite we obtain the following result for the rhf of the atft related to @xmath2 : @xmath148 where @xmath149 are the minimal positive integers for which @xmath150 we can consider particular cases of the models ( [ eq:*h - g ] ) by imposing on them the constraints @xmath151 and @xmath152 .
this leads to an atft related to a kac - moody algebra @xmath153 of rank @xmath154 whose system of admissible roots @xmath155 is obtained from @xmath156 by averaging with respect to the involution @xmath120 . for each of the examples listed in the next section we provide both sets of admissible roots .
the rhf of atft are more general integrable systems than the models described in @xcite which involve only the fields @xmath157 , @xmath158 invariant with respect to @xmath120 .
in this section we provide several examples of rhf of atft related to kac - moody algebras from the series @xmath159 and @xmath160 where the height @xmath161 can be either 1 or 2 .
we show that the involutions @xmath120 in all these cases are dual to @xmath53-symmetries of the extended dynkin diagrams derived in @xcite . the dynamical variables for the corresponding atft systems related to the type @xmath35 kac - moody algebras are provided as @xmath162-component vector - valued functions @xmath163 and @xmath164 restricted by the condition @xmath165 this is related to the fact that the root systems of type @xmath35 kac - moody algebras are conveniently embedded as @xmath166-dimensional subspace of @xmath167 orthogonal to the vector @xmath168 .
below we list several examples which illustrate the procedure outlined above and display new types of atft .
[ exa:1 ] we choose @xmath169 and fix up the involution @xmath120 by : @xmath170 the coordinates in @xmath171 are given by : @xmath172 where @xmath173 , i.e. , @xmath174 , @xmath175 .
this involution induces an involution @xmath127 of the kac - moody algebra @xmath176 which acts on the root space as follows , ( see fig .
[ fig:1 ] ) : @xmath177 the involution @xmath127 splits the root space @xmath178 into a direct sum of its eigensubspaces : @xmath179 with @xmath180 and @xmath181 .
the restriction of @xmath156 onto @xmath182 leads to the admissible root system @xmath183 of @xmath184 : @xmath185 the subspace @xmath186 is spanned by @xmath187 then the densities @xmath188 , @xmath189 for the rhf of aftf equal : @xmath190 where @xmath191 .
if we put @xmath192 and @xmath193 we get the reduced atft related to the kac - moody algebra @xmath194 @xcite .
( 300,140 ) ( 111,83)@xmath195 ( 142,96)@xmath196 ( 142,45)@xmath197 ( 222,27)@xmath198 ( 109,5)@xmath199 [ exa:2 ] choose @xmath200 and fix up the involution @xmath120 by : @xmath201 here the coordinates in @xmath171 are given by : @xmath202
i.e. , @xmath174 and @xmath203 .
this involution induces an involution @xmath127 of the kac - moody algebra @xmath204 which acts on the root space as follows ( see fig .
[ fig:2 ] ) : @xmath205 the involution @xmath127 splits the root space @xmath206 into a direct sum of its eigensubspaces : @xmath207 with @xmath180 and @xmath208 .
the restriction of @xmath156 onto @xmath182 leads to the admissible root system @xmath183 of @xmath209 : @xmath210 the subspace @xmath186 is spanned by @xmath211 then the densities @xmath212 , @xmath213 for the rhf of aftf equal : @xmath214 where @xmath191 .
if we put @xmath215 and @xmath216 we get the reduced atft related to the kac - moody algebra @xmath209 @xcite .
( 300,100 ) ( 120,80)@xmath217 ( 120,46)@xmath218 ( 218,27)@xmath198 ( 120,5)@xmath219 [ exa:3 ] now choose @xmath220 and fix up the involution @xmath120 by : @xmath221 the coordinates in @xmath171 : @xmath222 i.e. , @xmath223 and @xmath224 .
this involution induces an involution @xmath127 of the kac - moody algebra @xmath225 which acts on the root space as follows ( see fig .
[ fig:3 ] ) : @xmath226 the involution @xmath127 splits the root space @xmath206 into a direct sum of its eigensubspaces : @xmath227 with @xmath228 and @xmath229 .
the restriction of @xmath156 onto @xmath182 leads to the admissible root system @xmath230 of @xmath231 : @xmath232 the subspace @xmath186 is spanned by @xmath233 then the densities @xmath234 , @xmath235 for the rhf of aftf equal : @xmath236 where @xmath237 . if we put @xmath238 and @xmath216 we get the reduced atft related to the kac - moody algebra @xmath231 @xcite .
( 300,120 ) ( 111,83)@xmath195 ( 142,96)@xmath239 ( 142,45)@xmath240 ( 222,27)@xmath198 ( 109,5)@xmath241 [ exa:4 ] let us choose now @xmath242 and fix up the involution @xmath120 acting on the coordinates in the phase space by : @xmath243 then introducing on @xmath244 new coordinates by @xmath245 i.e. @xmath246 and @xmath247 .
this involution induces an involution @xmath127 of the kac - moody algebra @xmath248 which acts on the root space as follows ( see fig .
[ fig:4 ] ) : @xmath249 the involution @xmath127 splits the root space @xmath167 into a direct sum of its eigensubspaces : @xmath250 with @xmath180 and @xmath251 .
the restriction of @xmath156 onto @xmath182 leads to the admissible root system @xmath230 of @xmath252 : @xmath253 the subspace @xmath186 is spanned by the only nontrivial vector @xmath254 the rhf is described by the following densities @xmath255 , @xmath256 : @xmath257 ( 400,120 ) ( 20,55)@xmath258 ( 192,67)(-1,3)5 ( 192,49)(-1,-3)5 ( 192,67)(195,58)(192,49 ) [ exa:5 ] next we choose @xmath259 and fix up the involution @xmath120 acting on the coordinates in the phase space by : @xmath260 then introducing on @xmath244 new coordinates by @xmath261 i.e. @xmath262 and @xmath263 .
this involution induces an involution @xmath127 of the kac - moody algebra @xmath264 which acts on the root space as follows ( see fig .
[ fig:5 ] ) : @xmath265 the involution @xmath127 splits the root space @xmath266 into a direct sum of its eigensubspaces : @xmath267 with @xmath180 and @xmath268 .
the restriction of @xmath156 onto @xmath182 leads to the admissible root system @xmath230 of @xmath209 : @xmath269 the subspace @xmath186 is spanned by the vectors @xmath270 the rhf is described by the following densities @xmath271 , @xmath272 : @xmath273 ( 400,120 ) ( 20,55)@xmath274 ( 192,67)(-1,3)5 ( 192,49)(-1,-3)5 ( 192,67)(195,58)(192,49 ) ( 20,67)(1,3)5 ( 20,49)(1,-3)5 ( 20,67)(17,58)(20,49 ) [ exa:6 ] next we choose @xmath275 and fix up the involution @xmath120 acting on the coordinates in the phase space by : @xmath276 then introducing on @xmath244 new coordinates by @xmath277 i.e. @xmath246 and @xmath278 .
this involution induces an involution @xmath127 of the kac - moody algebra @xmath279 which acts on the root space as follows ( see fig . [ fig:6 ] ) : @xmath280 the involution @xmath127 splits the root space @xmath206 into a direct sum of its eigensubspaces : @xmath227 with @xmath180 and @xmath229 .
the restriction of @xmath156 onto @xmath182 leads to the admissible root system @xmath230 of @xmath281 : @xmath282 the subspace @xmath186 is spanned by the vectors @xmath283 the rhf is described by the following densities @xmath284 , @xmath285 : @xmath286 ( 400,120 ) ( 5,55)@xmath287 ( 73,68)(-3,-1)10 ( 154,68)(3,-1)10 ( 73,68)(113.5,81.5)(154,68 ) ( 40,68)(-3,-1)10 ( 187,68)(3,-1)10 ( 40,68)(113.5,92.5)(187,68 ) ( 197,88)(1,0)7 ( 30,88)(-1,0)7 ( 30,88)(1,0)167 ( 30,28.5)(-1,0)7 ( 197,28.5)(1,0)7 ( 30,28.5)(1,0)167 [ exa:7 ] finally we choose @xmath288 and fix up the involution @xmath120 acting on the coordinates in the phase space by : @xmath289 then introducing on @xmath244 new coordinates by @xmath290 i.e. @xmath246 and @xmath291 .
this involution induces an involution @xmath127 of the kac - moody algebra @xmath292 which acts on the root space as follows ( see fig .
[ fig:7 ] ) : @xmath293 the involution @xmath127 splits the root space @xmath294 into a direct sum of its eigensubspaces : @xmath207 with @xmath180 and @xmath295 .
the restriction of @xmath156 onto @xmath182 leads to the admissible root system @xmath183 of @xmath252 : @xmath296 the subspace @xmath186 is spanned by the vectors @xmath297 the rhf is described by the following densities @xmath298 , @xmath299 : @xmath300 ( 400,120 ) ( 20,50)@xmath301 ( 88,63)(-3,-1)10 ( 124,63)(3,-1)10 ( 88,63)(106,69)(124,63 ) ( 55,63)(-3,-1)10 ( 157,63)(3,-1)10 ( 55,63)(106,80)(157,63 ) ( 167,83)(1,0)7 ( 45,83)(-1,0)7 ( 45,83)(1,0)122 ( 45,23.5)(-1,0)7 ( 167,23.5)(1,0)7 ( 45,23.5)(1,0)122
we outlined the generalization of the notion of rhf to infinite dimensional hamiltonian systems using as a basis the atft type models .
we established that the special properties of these models allow us to relate the construction of the rhf to the study of @xmath53 symmetries of the extended dynkin diagrams .
the general construction is illustrated by several examples of such models related to the type @xmath302 and @xmath303 kac - moody algebras .
the atft models obtained here are generalizations of the atft in @xcite .
indeed , they contain two types of fields @xmath304 and @xmath305 which have different properties with respect to the involution @xmath120 , namely @xmath306 .
the @xmath53-reductions analyzed in @xcite contain only one type of fields invariant with respect to @xmath120 .
some additional problems are natural extensions to the results presented here .
the first one is the complete classification of all rhf of atft , including the cases related to the exceptional kac - moody algebras .
next is the derivation of their hamiltonian properties based on the classical @xmath307-matrix approach , see @xcite .
the last and technically more involved problem is to solve the inverse scattering problem for the lax operator and thus prove the complete integrability of all these models . in @xcite
it was shown that the dynamics of different rhf for the finite - dimensional toda systems may be qualitatively different ; for example some of the non - compact trajectories may become compact .
similar qualitative changes may be expected also in the infinite - dimensional cases .
another interesting problem is related to the fact that the integrable systems possess a hierarchy of hamiltonian structures , for review of the infinite - dimensional cases see e.g. @xcite and the references therein ; for the ( finite - dimensional ) toda chains see @xcite .
it is an open problem to construct the rhf for atft using some of its higher hamiltonian structures .
finally , this method provides a tool for systematic construction and classification of the rhf of a wide class of finite and infinite - dimensional hamiltonian systems , which need not be necessarily integrable .
therefore to each hamiltonian system one can associate a family of its rhf . to do
this one has to construct the corresponding involutions of the poisson brackets preserving the hamiltonian .
one of us ( ggg ) thanks the organizing committee of the european research conference on `` symmetries and integrability of difference equations '' ( side vi ) for the scholarship and for the warm hospitality in helsinki and the expert council for young scientists of the bulgarian academy of sciences for financial support .
we also acknowledge support by the national science foundation of bulgaria , contract no .
f-1410 .
gerdjikov v s , kyuldjiev a , marmo g and vilasi g , complexifications and real forms of hamiltonian structures , _
european j. phys . _ * 29b * ( 2002 ) , 177182 ; real hamiltonian forms of hamiltonian systems , _
european phys .
j. b. _ * 38 * ( 2004 ) 635649 .
gerdjikov v s , @xmath308reductions and new integrable versions of derivative nonlinear schrdinger equations , in nonlinear evolution equations : integrability and spectral methods , editors : fordy a p , degasperis a , lakshmanan m , manchester university press , 1981 , 367379 .
gerdjikov v s , analytic and algebraic aspects of toda field theories and their real hamiltonian forms , in bilinear integrable systems : from classical to quantum , continuous to discrete , editors : faddeev f , van moerbeke p and lambert f , nato science series ii : mathematics , physics and chemistry * 201 * , springer , dordrecht , 2005 , in press .
khastgir s p and sasaki r , instability of solitons in imaginary coupling affine toda field theory , _ prog .
* 95 * ( 1996 ) , 485501 ; non - canonical folding of dynkin diagrams and reduction of affine toda theories , _ prog .
phys . _ * 95 * ( 1996 ) , 503518 .
olive d , turok n and underwood j w r , solitons and the energy - momentum tensor for affine toda theory , _ nucl .
phys . _ * b401 * ( 1993 ) , 663 ; affine toda solitons and vertex operators , _ nucl . phys . _
* b409 * ( 1993 ) , 509 .
evans j , complex toda theories and twisted reality conditions .
* b 390 * ( 1993 ) , 225 ; + evans , j and madsen j o , nucl .
b * 536 * ( 1999 ) , 657703 ; erratum - ibid .
* 547 * ( 1999 ) 665 ; + evans j and madsen j , real form of non - abelian toda theories and their @xmath311-algebras . _
* b 384 * ( 1996 ) , 131 .
sklyanin e k , quantum version of the method of inverse scattering problem . _
lomi seminars _ * 95 * ( 1980 ) , 55128 ( in russian ) , translated in j. sov
. math . * 19 * , no . 5 ( 1982 ) , 1546 1596 ; kulish , p p quantum difference nonlinear schrodinger equation .
phys . _ * 5 * ( 1981 ) , 191197 .
gerdjikov v s , evstatiev e g and ivanov r i , the complex toda chains and the simple lie algebras solutions and large time asymptotics .
_ j. phys .
* 31 * ( 1998 ) , 82218232 ; the complex toda chains and the simple lie algebras solutions and large time asymptotics ii . _
a _ * 33 * ( 2000 ) , 9751006 . | a family of real hamiltonian forms ( rhf ) for the special class of affine @xmath0-dimensional toda field theories is constructed .
thus the method , proposed in @xcite for systems with finite number of degrees of freedom is generalized to infinite - dimensional hamiltonian systems .
we show that each of these rhf is related to a special @xmath1-symmetry of the system of roots for the relevant kac - moody algebra .
a number of explicit nontrivial examples of rhf of atft are presented .
[ firstpage ] |
zero - bias anomalies in transport properties are one of the most intriguing features of the low - temperature physics in nanostructures .
the canonical example is the zero - bias peak in the conductance of interacting quantum dots coupled to metallic contacts , which is a clear manifestation of the kondo effect @xcite arising from the dynamical screening of the unpaired electron spin in the quantum dot by the itinerant electrons of the leads .
another example is the andreev bound state arising from electron and hole scattering at a normal - superconductor interface.@xcite -0.01 in -0.05 in recently , a new type of zero - bias anomaly has emerged in connection with the appearance of majorana bound states in zeeman split nanowires with spin - orbit interaction in close proximity to an s - wave superconductor.@xcite it is theoretically well established that these `` topological '' superconducting wires sustain chargeless zero - energy end states with peculiar features such as braiding statistics , possibly relevant for topological quantum computation.@xcite experimentally , however , there is still controversy as to what the observed zero - bias peak really means : kondo effect , andreev bound states and disorder effects are some of the possibilities.@xcite franz summarizes and discusses these issues in ref . .
here we propose a direct way to probe the majorana end mode arising in a topological superconducting nanowire by measuring the two - terminal conductance @xmath2 through a dot side - coupled to the wire , figs .
[ fig1](a ) and [ fig1](b ) . using an exact recursive green s function approach ,
we calculate the local density of states ( ldos ) of the dot and wire , and show that the majorana end mode of the wire leaks into the dot @xcite thus giving rise to a majorana resonance in the dot , figs .
[ fig1](c ) and [ fig1](d ) .
surprisingly , we find that this dot majorana mode is pinned to the fermi level @xmath0 of the leads even when the gate controlled dot level @xmath7 is far off resonance @xmath8 . based on the results above , we suggest three experimental ways of probing the majorana end mode in the wire via the leaked / pinned majorana mode in the dot : ( i ) with the dot kept off resonance [ @xmath9 one can measure @xmath2 vs @xmath10 , the wire - dot coupling @xmath10 can be controlled by an external gate , to see the emergence of the @xmath5 peak in @xmath2 as the majorana end mode `` leaks '' into the dot , fig .
[ fig1](e ) ( cf .
@xmath11 and @xmath12 , see also fig .
[ fig2 ] ) ; ( ii ) alternatively , one can measure @xmath2 vs @xmath6 over a range in which @xmath13 runs from far below to far above the fermi - level of the leads where we find @xmath2 to be essentially a _ plateau _ at @xmath5 , figs .
[ fig1](f ) and [ fig1](g ) ; ( iii ) yet another possibility is to drive the wire through a non - topological / topological phase transition , e.g. , electrically via the spin - orbit coupling , temperature or the chemical potential @xmath14 of the wire ( fig .
[ fig3 ] ) , while measuring the conductance of the dot ; the presence / absence of the majorana end mode in the wire would alter drastically the conductance of the dot , see circles ( black ) and stars ( green ) in fig .
[ fig1](g ) .
the above pinning of the dot majorana resonance at @xmath0 is similar to that of kondo.@xcite however , the kondo resonance only occurs for @xmath7 _ below _
@xmath0 [ cf . figs .
[ fig1](h ) and [ fig1](i ) ] and yields a conductance peak at @xmath15 ( per spin ) instead .
even though there is no kondo effect in our system ( spinless dot ) , we conjecture that this symmetry of the dot - majorana resonance with respect to @xmath7 above and below @xmath0 could be used to distinguish majorana - related peaks from those arising from the usual kondo effect whenever this effect is relevant.@xcite moreover , this majorana resonance in the dot follows quite simply by viewing the dot as an additional site ( though with no pairing gap ) of the kitaev chain.@xcite we emphasize that this unique pinning occurs only when the dot is coupled to a majorana mode a half - fermion state .
when the dot is coupled to usual fermionic modes ( bound , e.g. , due to disorder , or not ) in the wire , its energy level will simply split and broaden as we discuss later on .
a spin full version of our model with a hubbard @xmath16 interaction in the dot yields similar results.@xcite the paper is organized as follows . in secs .
[ hamiltonian ] and [ green_funcions ] we present the hamiltonian that describes our system and introduce the majorana greens functions that we use to calculate the relevant physical quantities , respectively . in sec .
[ results ] we present our numerical results and discussions . finally , we summarize our main findings in sec
. [ remarks ] .
we consider a single - level spinless quantum dot coupled to two metallic leads and to a kitaev chain,@xcite fig .
[ fig1](a ) . to realize a single - level dot ( ` spinless ' dot regime ) , we consider a dot with gate controlled zeeman - split levels @xmath17 ( @xmath18 ) and @xmath19 , with @xmath20 the zeeman energy . by varying @xmath6 such that @xmath21 we can maintain the dot either empty [ i.e. , both spin - split levels above the fermi level @xmath22 ( taken as zero ) of the leads ] or singly occupied [ i.e. , only one spin - split dot level , e.g. , @xmath23 below @xmath22 ] .
this is the relevant spinless regime in our setup.@xcite typically [ e.g. , fig .
[ fig1](g ) ] , we vary @xmath25 , assuming a realistic zeeman energy to attain topological superconductivity , i.e. , @xmath26 [ see rainis et al . in ref .
this picture also holds true in the presence of a hubbard @xmath16 term in the dot citemapping - u ) . in this spinless regime ,
our hamiltonian is @xmath27 , with @xmath28 describing the chain @xmath29,\ ] ] @xmath30 is the number of chain sites , @xmath31 ( @xmath32 ) creates ( annihilates ) a spinless electron in the @xmath33th site and @xmath34 is an arbitrary phase .
the parameters @xmath35 and @xmath36 denote the inter - site hopping and the superconductor pairing amplitude of the kitaev model , respectively ; its chemical potential is @xmath14 . the single - level dot hamiltonian @xmath37 is @xmath38 @xmath39 ( @xmath40 ) creates ( annihilates ) a spinless electron in the dot with energy @xmath41 and @xmath42 denotes the free electron source ( _ s _ ) and drain ( _ d _ ) leads @xmath43 where @xmath44 ( @xmath45 ) creates ( annihilates ) a spinless electron with wavevector @xmath46 in the leads , whose fermi level is @xmath0 .
the coupling between the qd and the first site of the chain and between the qd and the leads are , respectively , @xmath47 and @xmath48 the quantity @xmath49 is the tunneling between the qd and the source and drain leads and @xmath10 is the hopping amplitude between the qd and the kitaev chain .
our model and approach are similar to those of ref . and go beyond low - energy effective hamiltonians.@xcite let us introduce the majorana fermions @xmath50 , @xmath51 , via @xmath52 and @xmath53 , @xmath54 ( @xmath55 is the dot).@xcite the @xmath56 s obey @xmath57_+ = 2\delta_{\alpha\alpha^\prime } \delta_{jj^\prime}$ ] and @xmath58 .
we now define the majorana retarded green s function @xmath59_+\rangle e^{i{\varepsilon}(\tau)}d\tau,\end{aligned}\ ] ] where @xmath60 denotes either a thermodynamic average or a ground state expectation value at zero temperature ; @xmath61 is the heaviside function and @xmath62 , with @xmath63 . we can express the electron green s function as @xmath64\end{aligned}\ ] ] and determine the electronic ldos @xmath65 , @xmath66\right].\ ] ] in ( [ ldos ] ) we have introduced the majorana ldos @xmath67 and @xmath68 . using the equation of motion for the green s functions ,
we obtain a set of coupled matrix equations , e.g. , for @xmath55 ( dot ) @xmath69 where @xmath70 is [ see eq .
( [ gf-0 ] ) ] @xmath71 @xmath72^{-1}{\bf m}_{jj}({\varepsilon})$ ] and @xmath73^{-1 } { \bf i}$ ] .
here @xmath74 $ ] is the dot level broadening ( leads ) , with @xmath75 , @xmath76 and @xmath77 the @xmath78 identity matrix .
finally , @xmath79 with @xmath80 } $ ] , @xmath81 , and @xmath82 and @xmath83)/2 for all @xmath84 .
the quantity @xmath85 is an effective coupling matrix , see fig .
[ fig1](a ) . in the wide band limit and
assuming a constant @xmath86 , we obtain @xmath87 and @xmath88 , with the broadening @xmath89 and @xmath90 being the dos of the leads . similarly to ( [ g11 ] ) , we find for the first site ( @xmath91 ) of the chain @xmath92 with @xmath93^{-1}\bar{\bf m}_{11 } ( { \varepsilon})$ ] .
we can then recursively obtain the majorana matrix at any site .
following realistic simulations @xcite and experiments,@xcite here we assume @xmath94 , the dot level broadening @xmath95 and set @xmath96 ( we also set @xmath97 ) . in fig . [ fig1](b ) we show the ldos as a function of the energy @xmath98 for a site in the middle and on the edge of the chain , @xmath99 [ dashed ( red ) curve ] and @xmath100 [ solid ( black ) curve ] , respectively , for @xmath101 ( decoupled chain ) and @xmath102 . note that @xmath103 is fully gapped , while @xmath100 exhibits a midgap zero - energy peak , corresponding to the end majorana state of the chain .
figures [ fig1](c ) and [ fig1](d ) show the ldos of the dot @xmath104 and of the first chain site @xmath12 as functions of @xmath105 for @xmath106 and three different values of @xmath10 . for clarity ,
the curves are offset vertically .
for @xmath101 [ long dashed ( black ) line ] we see just the usual single particle peak of width @xmath107 centered at @xmath108 .
observe that there is essentially no density of states at @xmath109 , since the dot level is far below the fermi level of the leads .
as we increase @xmath10 to @xmath110 [ fine solid ( red ) line ] , however , we observe the emergence of a sharp peak at @xmath109 in addition to the peak at @xmath111 . for @xmath112 [ dashed ( blue ) line in fig .
[ fig1](c ) ] , the single - particle peak in @xmath113 slightly moves to lower energies while its zero - energy peak increases to @xmath114 ( in units of @xmath115 ) . as the this peak appears in @xmath116 for increasing @xmath10 s , the majorana central peak in fig .
[ fig1](d ) decreases .
we can still see a peak in @xmath12 for @xmath112 , dashed ( blue ) line in fig .
[ fig1](d ) , but much weaker than its @xmath101 value .
we further show @xmath117 and @xmath118 , @xmath119 $ ] , vs @xmath10 in fig .
[ fig1](e ) clearly showing the wire majorana leakage into the dot . in fig .
[ fig1](f ) we display a color map of the electronic ldos @xmath104 vs @xmath98 and @xmath120 for the wire in the topological phase ( @xmath121 and @xmath122 ) with @xmath123 . at @xmath124
we see three bright regions that correspond to the three peaks of @xmath104 vs. @xmath98 ref . .
in contrast , by fixing @xmath109 and varying @xmath120 , we see that the zero - energy peak remains essentially unchanged over the range of @xmath120 shown .
more strikingly , this central peak is pinned at @xmath125 for @xmath126 and @xmath127 .
the pinning for @xmath128 below @xmath96 is similar to that of the kondo resonance , which , however , is known to occur at @xmath115 , cf figs .
1(h ) and 1(i ) . here
again one can measure @xmath2 vs @xmath6 [ fig .
[ fig1](g ) ] : for the wire in its trivial phase ( @xmath129 ) , e.g. , @xmath130 [ circles ( black ) ] , @xmath2 exhibits a single peak , whose maximum corresponds to @xmath13 crossing the fermi level .
note that the peak is not at @xmath124 but slightly shifted .
this arises from the small real part of the self energy in the dot green s function . in the topological phase ( @xmath122 ) , e.g. @xmath123 and @xmath131 [ squares ( red ) and diamonds ( blue ) , respectively ] , we see an almost constant @xmath132 for @xmath120 up to @xmath133 . this conductance plateau is similar to that of kondo,@xcite except that here @xmath2 is half of it ( per spin ) and the plateau occurs even for @xmath134 .
the majorana ldos @xmath135 and @xmath136 shown in figs .
[ fig2](a ) and [ fig2](b ) , respectively , as functions of @xmath98 and @xmath120 [ same parameters as in fig .
[ fig1](f ) ] , display a zero - energy peak in @xmath135 and none in @xmath136 .
this shows that the pinned dot - majorana peak in fig .
[ fig1](f ) arises from majorana a only .
we note that the peaks in @xmath136 at @xmath137 ( for @xmath124 ) are affected by the dot - wire majorana coupling as compared to the @xmath138 case . for couplings to any ordinary fermionic wire modes , the dot ldos would obey @xmath139 and it would split and broaden . figures [ fig2](c ) and [ fig2](d ) show that the majorana ldos of the first chain site @xmath140 and @xmath141 have no zero - energy peaks , thus indicating that the wire end mode has indeed leaked into the dot .
we see two peaks in @xmath140 at @xmath142 [ see fig .
[ fig2](b ) ] resulting from the coupling @xmath143 between @xmath144 and @xmath145 , see fig .
[ fig1](a ) . a careful look at fig .
[ fig2](c ) reveals an enhancement of the zero - energy peaks for @xmath146 , as a result of the coupling between the dot majorana a and the majoranas of the chain via a finite @xmath147 .
the strength of this peak is much smaller than its magnitude without the dot .
figure [ fig3](a ) shows the conductance @xmath2 vs @xmath14 for several @xmath147 s [ same parameters as in figs .
[ fig1](f ) and [ fig1](g ) ] . for @xmath148 [ circles ( black ) ] and @xmath129 ( trivial phase ) ,
@xmath2 arises from the single - particle dot level at @xmath22 .
the effect of the chain is essentially to shift and broaden @xmath147 , so that the value @xmath15 is reached only for @xmath149 .
as @xmath14 varies across @xmath150 , the wire undergoes a trivial - to - topological transition and @xmath2 suddenly decreases to @xmath5 as the leaked dot majorana appears . for @xmath151 the asymptotic ( @xmath149 ) value of the @xmath2 is no longer @xmath15 as @xmath147 can not attain @xmath22 .
the squares ( red ) and diamonds ( blue ) in fig .
[ fig3](a ) show a tiny conductance for @xmath152 .
however , as @xmath153 becomes smaller than @xmath35 both curves rapidly go to @xmath5 . in fig .
[ fig3](b ) we fix @xmath148 and plot the conductance @xmath2 as a function of @xmath14 for distinct @xmath10 s . as @xmath10 increases , @xmath2 remains pinned at @xmath5 in the topological regime , while it decreases in the trivial phase since the dot level shifts due to the chain self energy @xmath154 .
figures [ fig3](c ) and [ fig3](d ) show @xmath2 for @xmath155 and the same parameters as in figs .
[ fig3](a ) and [ fig3](b ) , respectively . for @xmath122
the @xmath2 is very sensitive to @xmath147 for a fixed @xmath112 [ fig .
[ fig3](c ) ] , and to @xmath10 for @xmath148 [ fig .
[ fig3](d ) ] , which contrasts with its practically constant value for @xmath156 , figs . [ fig3](a ) and [ fig3](b ) .
this is so because the wire acts as a third normal lead for @xmath157 and @xmath158 , so the source - drain @xmath2 , e.g. , for @xmath123 , reduces to @xmath159 , where @xmath160 is the broadening due to the chain.@xcite curiously , for @xmath161 and @xmath162 the @xmath2 curves are indistinguishable for @xmath155 and @xmath163 , being pinned at @xmath5 in the topological _ and _ trivial phases , cf .
squares in [ fig3](d ) and [ fig3](c ) .
therefore , the peak value @xmath164 , first found in ref . in a similar setup as ours but only for an on - resonance dot ( i.e. , @xmath165 ) , _ is not _ per se a ` smoking - gun ' evidence for a majorana end mode in conductance measurements , as we find that this peak value can appear even in the trivial phase of the wire .
one should vary , e.g. , @xmath166 and/or @xmath10 to tell these phases apart as we do in fig .
[ fig3 ] . finally
, the kinks in [ fig3](d ) [ e.g. , diamonds ( blue ) and stars ( green ) ] result from discontinuities in @xmath167 @xcite at @xmath168 .
we have used an exact recursive green s function approach to calculate the ldos and the two - terminal conductance @xmath2 through a quantum dot side - coupled to a kitaev wire .
interestingly , we found that the end majorana mode of the wire leaks into the quantum dot thus originating a resonance pinned to the fermi level of the leads @xmath0 .
in contrast to the usual kondo resonance arising only for @xmath128 below @xmath0 , this unique dot - majorana resonance appears pinned to @xmath0 even when the gate - controlled energy level @xmath13 is far above or below @xmath0 , provided that the wire is in its topological phase .
this leaked majorana dot mode provides a clear - cut way to probe the majorana mode of the wire via conductance measurements through the dot .
we acknowledge helpful discussions with f. m. souza .
j.c.e . also acknowledges valuable discussions with d. rainis .
this work was supported by the brazilian agencies cnpq , capes , fapesp , fapemig , and prp / usp within the re- search support center initiative ( nap q - nano ) .
revtex4 - 1 d. goldhaber - gordon , h. shtrikman , d. mahalu , d. abusch - magder , u. meirav , and m. a. kastner , nature * 391 * , 156 ( 1998 ) . s. m. cronenwett , t. h. oosterkamp , and l. p. kouwenhoven , science * 281 * , 540 ( 1998 ) .
a. a. golubov , a. brinkman , y. tanaka , i. i. mazin , and o. v. dolgov , phys .
* 103 * , 077003 ( 2009 ) .
r. m. lutchyn , j. d. sau , and s. das sarma , phys .
* 105 * , 077001 ( 2010 ) .
y. oreg , g. refael , and f. von oppen , phys .
lett . * 105 * , 177002 ( 2010 ) .
kitaev , ann.phys . * 303 * , 2 ( 2003 ) . c. nayak , s. h. simon , a. stern , m. freedman , and s. das sarma , rev .
phys . * 80 * , 1083 ( 2008 ) .
v. mourik , k. zuo , s. m. frolov , s. r. plissard , e. p. a. m. bakkers , and l. p. kouwenhoven , science * 336 * , 1003 ( 2012 ) .
m. t. deng , c. l. yu , g. y. huang , m. larsson , p. caroff , and h. q. xu , nano letters * 12 * , 6414 ( 2012 ) .
a. das , y. most , y. oreg , m. heiblum , and h. shtrikman , nat .
phys . * 8 * , 887 ( 2012 ) .
e. j. h. lee , x. jiang , r. aguado , g. katsaros , c. m. lieber , and s. de franceschi , phys .
lett . * 109 * , 186802 ( 2012 ) .
a. d. k. finck , d. j. van harlingen , p. k. mohseni , k. jung and x. li , phys .
* 110 * , 126406 ( 2013 ) .
k. t. law , p. a. lee , and t. k. ng , phys . rev . lett . *
103 * , 237001 ( 2009 ) .
d. i. pikulin , j. p. dahlhaus , m. wimmer , h. schomerus , and c. w. j. beenakker , new j. phys .
* 14 * , 125011 ( 2012 ) .
h. o. h. churchill , v. fatemi , k. grove - rasmussen , m. t. deng , p. caroff , h. q. xu , and c. m. marcus , phys . rev .
b * 87 * , 241401 ( 2013 ) .
m. franz , nat .
* 8 * , 149 ( 2013 ) .
interestingly , klinovaja and loss [ phys .
b * 86 * , 085408 ( 2012 ) ] have reported majorana modes leaking across a superconducting - normal interface .
see also d. chevallier , d. sticlet , p. simon , and c. bena , phys .
b * 85 * , 235307 ( 2012 ) .
a. c. hewson , the kondo problem to heavy fermions ( cambridge university press , 1993 ) .
when the kondo resonance coexists with the dot majorana mode , the total conductance through the dot should ideally attain @xmath169 [ @xmath15 ( due to kondo ) + @xmath5 ( majorana ) ] , as was found by lee et al [ phys .
b * 87 * , 241402 ( 2013 ) ] for a dot in the resonant - tunneling regime . see also cheng et al .
( arxiv:1308.4156 ) and golub et al .
. lett . * 107 * , 176802 ( 2011 ) ] for a discussion on the interplay between kondo and majorana induced couplings in a quantum dot .
a. y. kitaev , phys .- usp . * 44 * , 131 ( 2001 ) . for an interesting proposal for the realization of the kitaev chain
see : i. c. fulga , a. haim , a. r. akhmerov , and y. oreg , new j. phys . * 15 * , 045020 ( 2013 ) . we have also implemented a spin - full version of our system , considering a tight - binding nanowire + rashba spin - orbit interaction + proximity - induced superconductivity + a zeeman field ( a wire with these ingredients can be mapped onto the kitaev chain,@xcite ) which corroborates our findings . in addition , we have considered a hubbard @xmath16 term in the dot and verified that our results still hold , provided that only a single spin - split dot level lies below the fermi level of the leads ( i.e. , coulomb blockade is irrelevant in this regime ) .
these results will be described elsewhere .
interestingly , this regime ( empty or singly occupied dot ) should by itself prevent the kondo effect and andreev bound states in our setup .
andreev bound states , however , have a strong gate voltage dependence as reported by deacon _
et al_. [ phys .
lett . * 104 * , 076805 ( 2010 ) ] and appear ( mostly ) as doublets not pinned to @xmath22 [ see also lee _ et al _ , nat .
* 9 * , 79 ( 2014 ) ] .
alicea , phys .
b * 81 * , 125318 ( 2010 ) .
j. alicea , y. oreg , g. refael , f. von oppen , and m. p. a. fisher , nat . phys .
* 7 * , 412 ( 2011 ) .
d. rainis , l. trifunovic , j. klinovaja , and d. loss , phys .
b * 87 * , 024515 ( 2013 )
. d. e. liu , and h. u. baranger , phys .
b * 84 * , 201308 ( 2011 ) .
m. leijnse and k. flensberg , phys .
b * 84 * , 140501 ( 2011 ) .
j. alicea , rep .
* 75 * , 076501 ( 2012 ) .
e. prada , p. san - jose , and r. aguado , phys .
b * 86 * , 180503(r ) ( 2012 ) . the general expression for the conductance at @xmath170 is @xmath171 $ ] , with @xmath172 and @xmath173 . | we investigate quantum transport through a quantum dot connected to source and drain leads and side - coupled to a topological superconducting nanowire ( kitaev chain ) sustaining majorana end modes . using a recursive green s function approach ,
we determine the local density of states ( ldos ) of the system and find that the end majorana mode of the wire leaks into the dot thus emerging as a unique dot level _ pinned _ to the fermi energy @xmath0 of the leads . surprisingly , this resonance pinning , resembling in this sense a `` kondo resonance '' ,
occurs even when the gate - controlled dot level @xmath1 is far above or far below @xmath0 .
the calculated conductance @xmath2 of the dot exhibits an unambiguous signature for the majorana end mode of the wire : in essence , an off - resonance dot [ @xmath3 , which should have @xmath4 , shows instead a conductance @xmath5 over a wide range of @xmath6 , due to this pinned dot mode .
interestingly , this pinning effect only occurs when the dot level is coupled to a majorana mode ; ordinary fermionic modes ( e.g. , disorder ) in the wire simply split and broaden ( if a continuum ) the dot level .
we discuss experimental scenarios to probe majorana modes in wires via these leaked / pinned dot modes . |
we report on four measurements by the cdf experiment @xcite of processes involving heavy quark production . all use data recorded at center of mass energy 1.96 tev at the fermilab tevatron collider during run ii .
the structure of jets derives from the gluon emissions from the primary parton involved . in heavy quark jets ,
the quark decay must be modeled as well as the underlying event .
multi - gluon emission is difficult to calculate so is often approximated by parton shower models .
jet shape is known to depend upon whether the primary parton is a quark or a gluon , and it is also expected to depend upon flavor .
it is expected to depend as well upon the production mechanism ; for example , the @xmath2 and @xmath3 from gluon splitting are expected to be often in the same jet @xcite , producing a broader jet than does flavor creation .
measurement of jet shapes casts light on all these aspects of jet evolution and heavy flavor production . to address the question of whether the fraction of @xmath2-jets originating from gluon splitting , and its evolution , is well described by contemporary models ,
we define the jet shape , a measure of the fraction of total jet transverse momentum @xmath4 , inside a given radius in the space of rapidity @xmath5 and azimuthal angle @xmath6 .
this is @xmath7 , the fraction of total @xmath4 in cone @xmath8 carried by particles in subcone @xmath9 , averaged over an ensemble of jets , normalized to 1 , and defined such that @xmath10 and particles outside the cone are excluded .
the @xmath2-jet event selection begins with triggers based solely on calorimeter transverse energy @xmath11 .
trigger level 1 requires one calorimetric trigger tower above a threshold between 5 and 10 gev .
level 2 seeks clusters about the level 1 towers among adjacent towers above 1 gev and requires that at least one cluster exceed a threshold between 15 and 90 gev .
level 3 applies the run 1 cone algorithm @xcite and imposes @xmath12 greater than 20 to 100 gev .
the jet minimum @xmath4 requirement is set to ensure that only events for which the trigger is @xmath13 efficient are used .
offline the jets are reconstructed by the midpoint cone algorithm @xcite .
the selection of b - jets is then enhanced by requiring a secondary vertex .
because @xmath2 s tend to be found near the jet axis , the search for them uses a cone of radius @xmath14 in @xmath15 space .
tracks are ranked by reconstruction quality , including the contribution from the distance @xmath16 of closest approach to the primary vertex .
the algorithm next attempts to reconstruct a secondary vertex , beginning with the highest quality track .
if this succeeds , a cut is placed on the significance of the two - dimensional projection along the jet axis of the distance between the primary and secondary vertices .
general event quality cuts follow , including a rejection of multiple interactions ( there must be one and only one primary vertex with @xmath17 cm ) , a rejection of cosmics based on the significance of the missing @xmath18 , and a requirement of jet centrality : @xmath19 .
the jet @xmath20 is corrected to the hadron level by matching hadron and calorimeter level jets in monte carlo , a correction that increases the @xmath20 by 20% to 10% with no change in jet shape .
the requirement of the secondary vertex tag biases the measured jet shapes by demanding clean , well - defined tracks .
an @xmath9-dependent correction for this is needed .
there are additionally corrections for the presence of non-@xmath2 jets misidentified as @xmath2-jets , corrections for detector effects ( for example , the number of calorimeter towers scales with @xmath4 ) , and a correction for the presence of a second @xmath2 quark in the jet .
the data are fitted to two-@xmath2 and one-@xmath2 templates separately . the final function to be fitted
is given by @xmath21 , where @xmath22 is the unfolding factor @xmath23 , @xmath24 is the measured jet shape for the tagged sample , @xmath25 is the measured inclusive jet shape , @xmath26 is the purity of @xmath2 s in the tagged jet sample , and @xmath27 and @xmath28 correct for biases to the jet shape arising from the secondary vertex requirement .
the data are compared to predictions by pythia , with tune a @xcite , and herwig v 6.506 @xcite .
the principle systematics are associated with the choice of monte carlo model , the effect of the calorimeter model , the jet energy calibration , and assumptions about the @xmath29-quark content of the jets .
the results of the study @xcite , for integrated luminosity 300 pb@xmath30 , are shown in figure [ unfold2 ] , for 4 bins in @xmath4 . in each bin , the data are compared to the pythia prediction for inclusive jets as well as the pythia and herwig predictions for single @xmath2 production and for single @xmath2 production with the default fraction diminished by 20% .
one sees that the data are incompatible with the inclusive prediction : the jet shapes are evidently influenced by the presence of heavy quarks , and @xmath31-jets are broader .
one also sees that the data jets are broader than the pythia and herwig defaults ; the leading order models underestimate the fraction of @xmath2 s from gluon splitting and the fraction of jets with two @xmath2 s .
-jet shapes for four @xmath4 bins .
the data are shown as open squares where the error bars represent the statistical and total uncertainties .
the statistical uncertainties are smaller than the squares .
the data are compared to pythia tune a ( solid lines ) and herwig ( dashed lines ) .
the uppermost solid curve is the inclusive prediction ; the next solid curve , and the uppermost dashed curve , are the predictions with the default @xmath32 fraction ; and the lowest solid curve and lowest dashed curve are the predictions with the default 1@xmath2 fraction diminished by 20%.[unfold2],width=510 ]
the mechanism for producing heavy vector mesons in hadron collisions is not well understood .
cdf run i data on production cross sections for prompt @xmath34 and @xmath0 were one to two orders of magnitude above predictions by color singlet models .
subsequent theoretical efforts @xcite with adjusted production matrix elements match the cross section but predict increasing transverse polarization with production @xmath4 that is not confirmed by the data @xcite .
new approaches @xcite have been proposed .
the @xmath0 is a good testing ground for studying charmonium hadroproduction as there are no significant charmonium states above it to produce feeddown .
the goal of this study is a measurement of @xmath35 for @xmath36 gev / c .
the data selection begins with muons reconstructed in four layers of the central tracking chamber and matched to three to four hit tracks in the muon detector .
events that pass the dimuon trigger , which requires two opposite sign tracks , each with significant @xmath4 , are required to have three hits in the silicon vertex detector svx ii .
the @xmath0 mass and lifetime are fit with a lifetime function , and requirements are placed on the dimuon mass , @xmath37 gev / c@xmath38 , and rapidity @xmath39 .
cdf separates the signal from background , and the prompt from the @xmath2-decay @xmath40 s , with an unbinned maximum likelihood fit in candidate mass and proper decay length @xmath41 .
the mass separates the signal from the background .
the signal is modeled with the crystal ball function @xcite , a gaussian core with a low side tail .
the mass background is modeled with a first order polynomial . the prompt signal
is separated from the feeddown with a @xmath41-fit . the prompt signal
is given by a double gaussian centered on zero .
the long - lived signal is modeled by an exponential convoluted with a gaussian .
the lifetime background is modeled by a prompt double gaussian plus symmetric , positive-@xmath41 , and negative-@xmath41 long - lived components .
the likelihood function is given by @xmath42 , where the @xmath43 s are fractions of signal ( @xmath44 , from the total number of candidates in the fit ) , prompt ( @xmath45 ) , symmetric long - lived background ( sym ) , positive-@xmath41 long - lived background ( + ) , and negative-@xmath41 long - lived background(- ) .
the @xmath46 s are probability density functions for signal mass @xmath47 , linear mass background @xmath48 , prompt signal double gaussian proper time @xmath49 , an exponential convolved with a gaussian for @xmath2-decay signal @xmath50 , and lifetime backgrounds ( @xmath51 , @xmath52 , and @xmath53 ) .
the efficiency calculation is given by the product of six terms whose values range from 95.3% to 99.9% .
the acceptance calculation is complicated by the fact that acceptance depends upon polarization per @xmath4 bin , yet cdf polarization data on the @xmath0 are too weak for direct application .
the @xmath34 polarization , for which strong cdf data exist , are , however , expected to correlate highly with prompt @xmath0 polarization , so after confirming that the @xmath34 and @xmath0 polarization data are consistent , cdf applies the @xmath34 polarization values to the @xmath0 @xmath4 bins .
monte carlo @xmath0 samples are generated with fixed polarizations at the extreme values of 0 and -1 and with flat distributions in @xmath4 , @xmath54 , and @xmath6 .
these samples are simulated in the cdf detector and reconstructed .
we compute the product of geometrical acceptance and trigger efficiency , @xmath55 , where for @xmath56 generated events , @xmath57 survive geometric and reconstruction requirements and @xmath58 survive the trigger . @xmath59
ranges from 0.0053 to 0.2548 .
we interpolate the data to intermediate polarization values predicted from the @xmath34 data .
the principle systematic uncertainties on this measurement are associated with the luminosity , the reconstruction efficiency , the dimuon trigger efficiency , and the @xmath0 polarization .
the variation of the prescale during the course of collecting integrated luminosity 1.1 fb@xmath30 produces an effective integrated luminosity of 954 pb@xmath30 .
the resulting prompt @xmath0 differential cross section is shown in figure [ gdiff3 ] .
one sees that in all three cases , the run i result is confirmed with an order of magnitude more statistics .
the measured value of @xmath35 is @xmath60 nb . the ratio of the run ii to the run i result , for identical ranges in @xmath4 , is @xmath61 .
the theoretical prediction @xcite for this ratio , including the energy dependence of the parton distribution functions , is @xmath62 . for data from cdf run i ( triangles ) and run ii ( circles).[gdiff3],width=510 ]
measurements of @xmath2 quark production in hadronic collisions test perturbative qcd .
the cdf run i inclusive @xmath63 meson cross section measurement @xcite motivated theoretical developments beyond next - to - leading order @xcite .
continued comparisons , at higher @xmath4 and with increased statistics , may motivate further improvements .
this study considers events whose primary vertex has @xmath17 cm , whose missing @xmath11 is not significant , and whose kinematical variables lie in the range @xmath64 gev / c and @xmath65 , with jets reconstructed with the midpoint cone algorithm and heavy flavor jets tagged via their secondary vertices . the jet energy corrections are based on minimum bias events which measure the @xmath4 deposited in the calorimeter , as a function of the number of primary vertices , and indicate a value of @xmath66 gev / c per extra vertex .
monte carlo events from pythia 6.203 , with tune a and the cteq5l parton distribution function set , are used to correct the measured energy for effects of partially instrumented regions and calorimeter nonlinearities .
typical unfolding factors range from 1.6 to 2.1 .
the @xmath2 quark content of the jet is inferred from the shape of the invariant mass of all charged tracks attached to the secondary vertex , as shown in figure [ bmass1 ] .
the measurement is compared to the theoretical prediction described in @xcite . in the range 80 to 92 gev / c ,
compared to the result from the fit to monte carlo templates for @xmath2 and non-@xmath2 jets .
[ bmass1],width=510 ] the measured differential cross section , @xmath67 , is shown in figure [ bcross1 ] for integrated luminosity 300 pb@xmath30 . here
@xmath68 is the number of tagged jets in each @xmath4 bin , @xmath69 is the bin size , @xmath70 is the fraction of jets in the tagged sample ( this ranges from 0.35 at low @xmath4 to 0.14 at high ) , @xmath71 is the @xmath2-tagging efficiency , @xmath72 is the jet rapidity range , and @xmath73 is the luminosity . the measured cross section is consistent with the theoretical prediction throughout the full six orders of magnitude of range in @xmath4 .
it exhibits a strong dependence on scale @xmath74 , however , which suggests that higher orders may make large contributions .
the largest systematic uncertainties derive from the luminosity , the jet energy scale , the jet energy resolution , the unfolding process , the @xmath2-tagging efficiency , and the fraction of @xmath2-jets , and produce a total systematic uncertainty that ranges from @xmath75 at low @xmath4 to @xmath76 at high @xmath4 . -jet cross section ( filled circles ) as a function of @xmath77 compared with the next - to - leading order perturbative qcd prediction ( empty crosses ) .
the shaded band represents the systematic uncertainty on the data and the dashed band represents the uncertainty on the theory .
[ bcross1],width=510 ]
momentum conservation requires that the azimuthal angle @xmath6 between a @xmath2 and @xmath3 produced at lowest order must be @xmath78 .
higher order qcd processes produce additional partons in the final state , modifying the range of allowed azimuthal angle difference , @xmath79 .
order @xmath80 diagrams are thought to contribute the same magnitude to the cross section as order @xmath81 diagrams .
measuring the cross section as a function of @xmath79 provides information on the contributions of the leading order and next - to - leading order terms .
the event selection requires two central calorimeter towers with significant @xmath11 in association with two tracks in the central tracking chamber with significant @xmath4 .
the calorimeter towers are clustered , and the tracks are reconstructed with the silicon vertex trigger @xcite .
events having two clusters with significant @xmath11 associated with two tracks having significant displaced vertices are retained .
jets are reconstructed with the run i cone algorithm .
one or more high - quality primary vertices are required .
corrections are applied to the jet energy scale as well as to the energies of @xmath2-jets ( whose fragmentation , for example , is harder than that of light quarks ) .
the presence of two jets with significant @xmath11 , both @xmath2-tagged and with displaced secondary vertices , is confirmed .
the data are compared to predictions by pythia , with tune a ; herwig ; and mc@nlo @xcite with the herwig parton shower and the underlying event generated by jimmy 4.3 @xcite .
the systematics are dominated by the jet energy scale and range from 20 to 30% . for integrated luminosity 260 pb@xmath30 ,
all three monte carlos show similar agreement with data distibutions of @xmath82 production cross section versus leading jet @xmath11 and versus dijet invariant mass .
the cross section as a function of @xmath79 , however , is significantly better modeled by mc@nlo , as shown in figure [ cross1 ] .
the peak at large angles reflects flavor creation , and the excess at small angles is due to higher order diagrams and multiple interactions . the measured total cross section for @xmath83 , @xmath84 gev , and
@xmath85 gev is @xmath86 pb . jet cross section as a function of the dijet @xmath87 correlation .
data are compared to hadron level cross sections obtained using mc@nlo+jimmy , pythia , and herwig+jimmy .
the shaded area represents the total systematic uncertainty on the data.[cross1],width=510 ] d. acosta et al .
d _ 71 , 032001 ( 2005 ) .
s. frixione et al .
, hep - ph/9702287 . f. abe et al . , _ phys . rev .
d _ 45 , 1448 ( 1992 ) .
blazey et al . , hep - ex/0005012 ; d. acosta et al . , _ phys .
d _ 74 , 071103(r ) , 2006 .
t. sjstrand et al . , _ comp .
_ 135 , 238 ( 2001 ) ; r. field and r. group , arxiv : hep - ph/0510198 . g. corcella et al . , _ jhep _ 01 , ( 2001 ) 010 .
a. abulencia et al . , arxiv:0806.1699 [ hep - ex ] , 2008 .
bodwin , e. braaten , and g.p .
lepage , _ phys .
51 , 1125 ( 1995 ) ; e. braaten and s. fleming , _ phys .
_ 74 , 3 327 ( 1995 ) .
a. abulencia et al . , _ phys .
lett . _ 99 , 132001 ( 2007 ) .
baranov , _ phys .
d _ 66 , 114003 ( 2002 ) .
khoze et al . , _ eur .
j. c _ 39 , 163 ( 2005 ) .
j. gaiser , slac-0255 ( 1982 ) .
k. anikeev et al . , hep - ph/0201071 ( 2001 ) . f. abe et al . , _ phys .
_ 75 , 1451 ( 1995 )
. m. cacciari and p. nason , _ phys .
_ 89 , 122003 - 1 ( 2002 ) .
s. frixione and m. mangano , _ nucl .
_ b 483 , 321 ( 1997 ) .
a. bardi et al . , _ nucl .
instr . and
meth . _ a 485 , 178 ( 2002 ) . s. frixione and b. webber , _ jhep _ 0206 ( 2002 ) 029 .
butterworth , j.r .
forshaw , m.h .
seymour , _
_ c 72 , 637 ( 1996 ) . | results are presented from four cdf analyses involving heavy quark production in proton - antiproton collisions at center of mass energy 1.96 tev .
the shapes of b - jets are found to be broader than inclusive predictions and broader than both pythia and herwig defaults .
a measurement of the production cross section for @xmath0 is consistent with run 1 results and with theoretical predictions associated with parton distribution function energy dependence .
the inclusive b - jet production cross section is also consistent with theoretical predictions over six orders of magnitude .
the @xmath1 differential production cross section is compared to several theoretical models and found to be best described by mc@nlo + jimmy . |
unlike the case for stars , the relative size and separation between galaxies is small , suggesting that galaxy interactions and mergers could be a major process in driving galaxy formation .
observational attempts to determine the role of galaxy mergers in the local universe have been limited to studies of how star formation and black hole growth are induced by the merger process ( e.g. , barton et al .
2000 ; ellison et al .
it is however clear that ongoing major galaxy merging is rare in the nearby universe , and is likely not a dominant mode for the future evolution of galaxies ( e.g. , patton et al .
if galaxy mergers play a major role in galaxy formation they must have occurred earlier in the history of the universe .
attempts to trace the merger history in the early universe has been carried out using two primary methods .
the oldest method involves measuring the fraction of galaxies which are in physical pairs at various look - back times , or redshifts ( e.g. , patton et al . 2002 ; le fevre et al .
2000 ; lin et al . 2004
; lin et al . 2008 ; bluck et al . 2008 ) .
a newer method , with a considerable background history ( e.g. , holmberg 1941 ; vorontsov - velyaminov 1959 ; arp 1966 ) , uses the structures of galaxies to determine the merger history ( e.g. , conselice 2003 ; conselice et al . 2003 ; lotz et al .
it is , however , not yet known if the two methods are revealing the same merger history for galaxies , although at @xmath15 , the agreement between the two methods appears good ( e.g. , de propris et al .
2007 ) .
the merger history of galaxies is still largely only beginning to be measured with some certainty .
two issues that still need to be addressed are the reliability of the current methods for determining merger histories , and the relatively small areas thus far used to find mergers .
both of these issues limit our ability to accurately measure the merger history .
the reliability argument has been addressed in great detail in papers such as conselice ( 2003 ) , conselice et al .
( 2003 ) , conselice et al .
( 2005 ) , and lotz et al .
briefly , the merger history can only be measured using special techniques and data , such as complete redshift surveys and/or deep hubble space telescope ( hst ) imaging .
it turns out that when examining galaxies that fit a structural merger criteria , nearly all are ongoing major mergers , as judged by eye and through kinematics ( e.g. , conselice 2000a , b ; conselice 2003 ; conselice et al .
this hold for low and high redshift galaxies ( e.g. , conselice et al .
2003 ) , and is also seen when examining n - body models of the merging process ( conselice 2006 ; lotz et al
. 2008b ) . using structure to determine the merger history requires the use of hst imaging or adaptive optics ( ao ) . however , the small fields of view used in hst and ao surveys make it difficult to measure the merger history with a high certainty , simply due to the small number of galaxies that have been examined .
another issue is that it is difficult to measure the merger history at higher redshifts due to the fact that there is very little high resolution deep near - infrared imaging , which is needed to directly study the rest - frame optical structures of @xmath16 galaxies .
this is required , as the rest - frame optical appearance of a galaxy often differs significantly from the rest - frame ultraviolet morphology ( e.g. , windhorst et al .
2002 ; taylor - mager et al .
2007 ) . ideally in the future , using multiple band data , it is desirable to measure parameters and structures on stellar mass images ( e.g. , lanyon - foster et al .
2009 ) . for our merger measurements , we utilise the cas method ( e.g. , conselice 2003 ) for determining the presence of galaxy mergers in the galaxy population at @xmath17 .
we first re - evaluate the use of the asymmetry index ( conselice et al .
2000a ) , and the cas method itself for locating major mergers .
we furthermore use our derived merger fractions to characterise the major galaxy merger evolution for m@xmath1 galaxies at @xmath18 .
we discuss various ways in which this merger evolution can be parameterised , investigating power - law , exponential , and the combination of the two forms .
we conclude that the use of a @xmath12 prior in fitting the merger fraction is a significant factor in the determination of the power - law slope @xmath19 .
we show that this value of @xmath19 can vary between @xmath20 and @xmath21 depending on how the merger fraction is fit ( cf .
bluck et al .
2008 for ultra massive galaxies ) .
in this paper we use the two largest hst advanced camera for surveys ( acs ) imaging data sets - the extended groth strip ( egs ) and the cosmic evolution survey ( cosmos ) to determine , using @xmath22 massive galaxies , the detailed merger history at @xmath2 .
this is an important epoch for measuring the merger history , and there is still considerable uncertainty regarding the merging history during this epoch , which traces the last half of the universe . by using acs f814w band images we are also tracing the structures of these galaxies in the rest - frame optical out to @xmath23 and do not need to consider the sometimes considerable morphological @xmath24corrections when imaging galaxies in the ultraviolet ( e.g. , taylor - mager et al .
2007 ) .
our general conclusion is that the merger fraction increases with higher redshifts to @xmath25 .
we use the latest model based time - scales for galaxies in pairs to merge , and for galaxies to remain symmetric during mergers , to calculate the number of mergers a typical m@xmath1 galaxy will undergo at @xmath18 , as well as the galaxy merger rate .
we conclude that a typical m@xmath1 galaxy will undergo 1 - 2 major mergers at @xmath17 .
we also investigate how structural merger fractions compare with pair fractions , finding that at a given redshift , within the same population of galaxies , the cas merger fraction is 3 - 6 times higher than the pair fraction .
we show that the ratio of the time - scales for cas mergers and 20 kpc pairs is nearly the same as the merger - pair fraction ratio .
both methods therefore appear to trace the same merging population at different phases .
this is further evidence that we are indeed measuring correctly the merging properties of galaxies .
this paper is organised as follows : 2 includes a discussion of the data sources we use in this paper , and how we select our sample , including a description of our morphological and structural analyses , and the stellar masses we utilise , 3 is a discussion of our results , including a description of the merger history up to @xmath25 , and 4 is our summary and conclusions .
we use a standard cosmology of h@xmath26 km s@xmath27 mpc@xmath27 , and @xmath28 = 0.3 throughout .
the data we use in the paper originate from the extended groth strip ( egs ) survey ( davis et al .
2007 ) , and the cosmos survey ( scoville et al . 2007 ) .
the selection of galaxies which we analyse is simply those systems which have a stellar mass m@xmath29 @xmath30 . as the egs and
the cosmos fields have different data sets , and methods for galaxy detection , and for measuring stellar masses , we consider both data sets separately in what follows . in total
there are 21,902 m@xmath29 @xmath30 galaxies at @xmath2 in our sample , with 2,388 galaxies in the egs and 19,514 in the cosmos field .
the main data set for this paper is the cosmos field , which is by far the largest mosaic of hubble space telescope imaging using the advanced camera for surveys ( acs ) .
the cosmos acs coverage is 1.8 deg@xmath31 and is imaged in the f814w ( i ) band over this entire area with 1 orbit depth per pointing , for a total of 590 orbits .
the 50% completeness of the cosmos acs data is i@xmath32 ( scoville et al . 2007 ) .
the cosmos field also has extensive optical data , described in mobasher et al .
( 2007 ) , from which stellar masses and photometric redshifts are computed .
the photometric data in which these quantities are derived include optical data in the @xmath33 band from the cfht , @xmath34 imaging from the suprimecam on subaru , @xmath35 imaging from chft and @xmath36 imaging from flamingos taken at ctio and kitt peak ( mobasher et al . 2007 ) . the i@xmath37 acs imaging is also included in the analysis .
the seeing for this optical and nir ground based imaging is roughly 1@xmath38 , with depths that range from 21.5 ab in @xmath36 to @xmath39 in the @xmath40-band .
the depth in every band is in any case well matched to image the most massive galaxies at @xmath2 at a high s / n .
the acs imaging of the egs field covers a 10.1@xmath41 70.5 strip , for a total area of 0.2 deg@xmath31 .
this acs imaging is discussed in lotz et al .
( 2008 ) and conselice et al .
( 2008b ) , and is briefly described here .
the imaging consists of 63 tiles imaged in both the f606w ( v ) and f814w ( i ) bands .
the 5-@xmath42 depths reached in these images are v = 26.23 ( ab ) and i = 27.52 ( ab ) for point sources , and about two magnitudes brighter for extended objects .
our sample for the egs comes from those systems which have k - band data taken as part of the palomar observatory wide - field infrared survey ( powir ; bundy et al . 2006 ; conselice et al . 2007a , b ; 2008b ) .
the powir survey was designed to obtain deep k - band and j - band data over a significant ( @xmath431.5 deg@xmath44 ) area .
observations were carried out between september 2002 and october 2005 over a total of @xmath45 nights .
this survey covers the goods field north ( giavalisco et al .
2004 ; bundy et al . 2005 ) , the extended groth strip ( davis et al . 2007 ) , and three other fields that the deep2 team has observed with the deimos spectrograph ( davis et al .
the total area we cover in the k - band is 5524 arcmin@xmath31 = 1.53 deg@xmath31 , with half of this area imaged in the j - band .
each of our fields reach 5-@xmath42 depths between k@xmath46 for point sources , as measured in a 2 diameter aperture with the egs .
our k@xmath47-band data were acquired utilising the wirc camera on the palomar 5 meter telescope .
wirc has an effective field of view of @xmath48 , with a pixel scale of 0.25pixel@xmath27 .
our k@xmath47-band data were taken using 30 second integrations , with four exposures per pointing , while the j - band observations were taken with 120 second exposures per pointing .
typical total exposure times are one and two hours for both bands .
our reduction procedure follows standard methods for combining nir ground - based imaging , and is described in more detail in bundy et al .
the resulting seeing fwhm in the k@xmath47-band imaging ranges from 0.8to 1.2 , and is typically 1.0(e.g .
, bundy et al . 2006 ) .
other data sets within the egs we use consist of : optical imaging from the cfht over all fields , imaging from the advanced camera for surveys ( acs ) on hubble , and spectroscopy from the deimos spectrograph on the keck ii telescope ( davis et al .
a summary of these ancillary data sets are included in davis et al .
( 2007 ) and conselice et al .
( 2007b ) . the optical imaging of the egs is taken with the cfht 3.6-m telescope .
this optical data consists of imaging in the b , r and i bands taken with the cfh12k camera - a 12,288 @xmath41 8,192 pixel ccd mosaic with a pixel scale of 0.21 .
the integration times for these observations are 1 hour in @xmath49 and @xmath50 , and 2 hours in @xmath51 , per pointing , with a r - band 5-@xmath42 depth of @xmath52 , and similar depths at @xmath49 and @xmath51 .
the seeing for the optical imaging is roughly the same as that for the nir imaging , and we measure photometry using a 2 diameter aperture .
we utilise both spectroscopic and photometric redshifts for the galaxies we study in both the egs and cosmos fields .
the only field which has extensive available spectroscopy is the egs .
egs keck spectra were acquired with the deimos spectrograph as part of the deep2 redshift survey ( davis et al .
target selection for the deep2 spectroscopy was based on the optical properties of the galaxies detected in the cfht photometry , with the basic selection criteria @xmath53 .
deep2 spectroscopy was acquired through this magnitude limit , with no strong colour cuts applied to the selection .
about 10,000 redshifts are measured for galaxies within the egs .
the sampling rate for galaxies that meet the selection criteria is 60% .
this deimos spectroscopy was obtained using the 1200 line / mm grating , with a resolution r @xmath54 covering the wavelength range 6500 - 9100 .
redshifts were measured through an automatic method comparing templates to data , and we only utilise those redshifts measured when two or more lines were identified , providing very secure measurements .
roughly 70% of all targeted objects result in secure redshifts . as the cosmos field is the major contributor to the data used in this paper we give a description of the photometeric redshifts which we utilise .
the details of its photometric redshifts and the catalog we use is described in great detail in mobasher et al .
we give a short summary here .
the cosmos photometric redshifts are measured through a standard template fitting technique .
the templates used by mobasher et al .
( 2007 ) are galaxies of various spectral types from ellipticals to starbursts across optical rest - frame wavelengths .
the photometric redshifts we use were tested by comparing with 868 spectroscopic redshifts from the zcosmos survey , where the rms scatter in the agreement between photometric redshifts and spectroscopic redshifts is @xmath55 = 0.031 , where @xmath56 ( mobasher et al . 2007 ) .
less than 2.5% of the galaxies are outliers in the agreement between spectroscopic and photometric redshifts .
mobasher et al .
( 2007 ) furthermore test their method of measuring photometric redshifts by calculating values using three different codes , and find a good agreement with their own calculation .
we utilise our own photometric redshifts within the egs ( e.g. , bundy et al . 2006 ; conselice et al .
2007b , 2008b ) .
the determination of the egs photometric redshifts is done in a different way than in the cosmos field . within the egs ,
photometric redshifts are based on the optical+near infrared imaging , in the brijk bands , and are fit in two ways , depending on the brightness of a galaxy in the optical . for galaxies that meet the spectroscopic criteria , @xmath53 , we utilise a neural network photometric redshift technique to take advantage of the vast number of secure redshifts with similar photometric data .
most of the @xmath53 sources not targeted for spectroscopy should be within our redshift range of interest , at @xmath57 .
the neural network fitting is done through the use of the annz ( collister & lahav 2004 ) method and code .
to train the code , we use the @xmath54 redshifts in the egs , which has galaxies spanning our entire redshift range .
the training of the photometric redshift fitting was in fact only done using the egs field , whose galaxies are nearly completely selected based on a magnitude limit of @xmath53 .
we then use this training to calculate the photometric redshifts for galaxies with @xmath53 .
the agreement between our photometric redshifts and our annz spectroscopic redshifts is very good using this technique , with @xmath58 out to @xmath59 .
the agreement is even better for the m@xmath60 galaxies where we find @xmath61 across all of our four fields ( conselice et al .
2007b ) . for galaxies which are fainter than @xmath62 in the egs we utilise photometric redshifts using bayesian techniques , and the software from benitez ( 2000 ) .
for an object to have a photometric redshift we require that it be detected at the 3-@xmath42 level in all optical and near - infrared ( brijk ) bands , which in the r - band reaches @xmath52 .
we optimised our results , and correct for systematics , through the comparison with spectroscopic redshifts , resulting in a redshift accuracy of @xmath63 for @xmath64 systems .
these @xmath64 galaxies are , however , only a very small part of our sample
. furthermore , all of these systems are at @xmath65 .
since stellar masses are a critical aspect of this analysis , we go into some detail for how these are calculated .
the stellar masses for the cosmos survey are taken from mobasher et al .
these stellar masses are computed through the use of k - band imaging and the measured rest - frame @xmath66 colours of galaxies within cosmos .
mobasher et al .
( 2007 ) use this measured colour to obtain a mass to light ratio through the relation : @xmath67 which assumes a salpeter imf within a mass range of 0.1 to 100 .
the colours used in this analysis are not directly measured from the data , but are taken from the best fit templates of various types ( e , sa , sb , sc , i m and starburst ) ( see mobasher et al . 2007 ) . due to
limited k - band data over the cosmos field , the stellar masses of these galaxies are calculated via rest - frame v - band magnitudes with the equation : @xmath68 mobasher et al .
( 2007 ) discuss using v - band luminosities for measuring stellar masses rather than k - band , which is very shallow over the cosmos fields .
they find for a subset of their sources , which are detected in the k - band , an agreement with the v - band based stellar masses , these are however the reddest , and thus perhaps the most evolved galaxies , and therefore may not be representative . within the egs field
we match our k - band selected catalogs to the cfht optical data to obtain spectral energy distributions ( seds ) for all of our sources , resulting in measured brijk magnitudes . from these
we compute stellar masses based on the methods and results outlined in bundy , ellis , conselice ( 2005 ) and bundy et al .
all our stellar masses are furthermore normalised by the observed rest - frame k - band light , which is roughly at rest - frame @xmath69 , or redder , for most galaxies .
we also convert these stellar masses , which are calculated in equation ( 2 ) using a salpeter imf to a chabrier imf . the basic mass fitting method within the egs consists of fitting a grid of model seds constructed from bruzual & charlot ( 2003 ) ( bc03 ) stellar population synthesis models , with different star formation histories .
we use an exponentially declining model to characterise the star formation history , with various ages , metallicities and dust contents included .
these models are parameterised by an age , and an e - folding time for parameterising the star formation history , where sfr @xmath70 .
the values of @xmath71 are randomly selected from a range between 0.01 and 10 gyr , while the age of the onset of star formation ranges from 0 to 10 gyr .
the metallicity ranges from 0.0001 to 0.05 ( bc03 ) , and the dust content is parametrised by @xmath72 , the effective v - band optical depth for which we use values @xmath73 .
although we vary several parameters , the resulting stellar masses from our fits do not depend strongly on the various selection criteria used to characterise the age and the metallicity of the stellar population .
it is important to realise that these parameterisations are fairly simple , and it remains possible that stellar mass from older stars is missed under brighter , younger , populations . while the majority of our systems are passively evolving older stellar populations , it is possible that up to a factor of two in stellar mass is missed in any star bursting blue systems .
however , stellar masses measured through our technique are roughly the expected factor of 5 - 10 smaller than dynamical masses at @xmath23 using a sample of disk galaxies ( conselice et al .
2005b ) , demonstrating their inherent reliability .
we match magnitudes derived from these model star formation histories to the actual data to obtain a measurement of stellar mass using a bayesian approach .
we calculate the likely stellar mass , age , and absolute magnitudes for each galaxy at all star formation histories , and determine stellar masses based on this distribution .
distributions with larger ranges of stellar masses have larger resulting uncertainties . while parameters such as age , e - folding time , metallicity , etc . are not likely accurately fit through these calculations due to various degeneracies , the stellar mass is robust .
typical errors for our stellar masses are 0.2 dex from the width of the probability distributions . there are also uncertainties from the choice of the imf .
our stellar masses utilise the chabrier ( 2003 ) imf .
there are additional random uncertainties due to photometric errors .
the resulting stellar masses thus have a total random error of 0.2 - 0.3 dex , roughly a factor of two .
furthermore , there is the issue of whether or not our stellar masses are overestimated based on using the bruzual & charlot ( 2003 ) models .
it has recently been argued by maraston ( 2007 ) , among others , that a refined treatment of thermal - pulsating agb stars in the bc03 models result in stellar masses that can be too high by a factor of a few . while we consider an uncertainty of a factor of two in our stellar masses , it is worth investigating whether or not our sample is in the regime where the effects of a different treatment of tp - agb stars in e.g. , maraston ( 2007 ) will influence our mass measurements .
this has been investigated in maraston ( 2005 ) who have concluded that galaxy stellar masses computed with an improved treatment of tp - agb stars are roughly 50 - 60% lower . however , the effect of tp - agb stars is less important at our rest - frame wavelengths probed than at longer wavelengths , especially in the rest - frame ir .
the egs survey is k - selected , and the observed k - band is used as the flux in which stellar masses are computed .
the rest - frame wavelength probed with the observed k - band ranges from 0.7@xmath74 m to 1.5@xmath74 m where the effects of tp - agb stars are minimised .
the ages of our galaxies are also older than the ages where tp - agb stars have their most effect ( maraston 2005 ) . to test this , after our analysis was finished , we utilised the newer bruzual and charlot ( 2009 , in prep ) models , which include a new tp - agb star prescription , on our massive galaxy sample . from this
we find on average a @xmath43 0.07 dex smaller stellar mass using the newer models . at most
, the influence of tp - agb stars will decrease our stellar masses by 20% .
the effect of this would decrease the number of galaxies within our sample , particularly those close to the m@xmath75 boundary .
this systematic error is however much smaller than both the stellar mass error we assume ( 0.3 dex ) , and the cosmic variance uncertainties , and thus we conclude it is not a significant factor for our analysis .
we use the cas ( concentration , asymmetry , clumpiness ) parameters to measure the structures of our @xmath2 galaxies quantitatively .
we include in our analysis the measurement of the gini and 20 parameters ( e.g. , lotz et al .
the cas / gini/20 parameters are a non - parametric method for measuring the forms and structures of galaxies in resolved ccd images ( e.g. , conselice et al .
2000a ; bershady et al .
2000 ; conselice et al . 2002 ; conselice 2003 ; lotz et al .
the basic idea behind these parameters is that galaxies have light distributions that reveal their past and present formation modes ( conselice 2003 ) .
furthermore , well - known galaxy types in the nearby universe fall in well defined regions of the cas parameter space . for example
, the selection @xmath76 locates systems which are nearly all major galaxy mergers in the nearby universe ( e.g. , conselice et al .
2000b ; conselice 2003 ; hernandez - toledo et al .
2005 ; conselice 2006b ) .
in addition to the classic cas parameters , we also investigate the use of the similar gini and m@xmath77 parameters ( lotz et al .
we give a brief description of these parameters below .
the way we measure structural parameters on our hubble images varies slightly from what has been done earlier in the hubble deep field , and goods imaging ( e.g. , conselice et al .
2003a ; conselice et al .
2004 ) . the basic measurement procedure , after cutting out the galaxy into a smaller image , is to first measure the radius in which the parameters are computed .
the radius we use for all our indices is defined by the petrosian radii , which is the radius where the surface brightness at a given radius is 20% of the surface brightness within that radius ( e.g. , bershady et al .
2000 ; conselice 2003 ) .
we use circular apertures for our petrosian radii and quantitative parameter estimation .
we begin our estimates of the galaxy centre for the radius measurement at the centroid of the galaxy s light distribution . through modelling and various tests
, it can be shown that the resulting radii do not depending critically on the exact centre , although the cas and other parameters do ( conselice et al .
2000 ; lotz et al .
the exact petrosian radius we use to measure our parameters is @xmath78 where @xmath79 is the radius where the surface brightness is 20% of the surface brightness within that radius .
a very important issue , especially for faint galaxies , is how to account for background light and noise . for faint galaxies
there is a considerable amount of noise added due to the sky , which must be corrected . through various test , outlined in detail in conselice et al .
( 2009 , in prep ) , we conclude that the proper way to correct parameters for the background requires that the selected background area be close to the object of interest .
this is only an issue for faint galaxies , and for galaxies imaged on large mosaics which have a non - uniform weight map , and whose noise characteristics vary across the field . by using a background near each object
we alleviate these issues as the noise properties do not vary significantly over @xmath80 arcmin , where the galaxy and the background area are selected .
we review below how the cas and gini/20 parameters are measured . for more detail
see bershady et al .
( 2000 ) , conselice et al . ( 2000 ) , conselice ( 2003 ) , and lotz et al . (
2008 ) .
the asymmetry of a galaxy is measured by taking an original galaxy image and rotating it 180 degrees about its centre , and then subtracting the two images ( conselice 1997 ) .
there are corrections done for background , and radius ( explained in detail in conselice et al .
most importantly , the centre for rotation is decided by an iterative process which finds the location of the minimum asymmetry .
the formula for calculating the asymmetry is given by : @xmath81 where @xmath82 is the original image pixels , @xmath83 is the image after rotating by 180 .
the background subtraction using light from a blank sky area , called @xmath84 , are critical for this process , and must be minimised in the same way as the original galaxy itself .
a lower value of @xmath85 means that a galaxy has a higher degree of rotational symmetry which tends to be found in elliptical galaxies .
higher values of @xmath85 indicate an asymmetric light distribution , which are usually found in spiral galaxies , or in the more extreme case , merger candidates .
concentration is a measure of the intensity of light contained within a central region in comparison to a larger region in the outer - parts of a galaxy .
the exact definition is the ratio of two circular radii which contain 20% and 80% ( @xmath86 , @xmath87 ) of the total galaxy flux , @xmath88 this index is sometimes called c@xmath89 .
a higher value of @xmath90 indicates that a larger amount of light in a galaxy is contained within a central region .
this particular measurement of the concentration correlates well with the mass and halo properties of galaxies in the nearby universe ( e.g. , bershady et al .
2000 ; conselice 2003 ) .
the clumpiness ( @xmath91 ) parameter is used to describe the fraction of light in a galaxy which is contained in clumpy distributions .
clumpy galaxies have a relatively large amount of light at high spatial frequencies , whereas smooth systems , such as elliptical galaxies contain light at low spatial frequencies .
galaxies which are undergoing star formation tend to have very clumpy structures , and high @xmath91 values .
clumpiness can be measured in a number of ways , the most common method used , as described in conselice ( 2003 ) is , @xmath92,\ ] ] where , the original image @xmath93 is blurred to produce a secondary image , @xmath94 .
this blurred image is then subtracted from the original image leaving a residual map , containing only high frequency structures in the galaxy ( conselice 2003 ) . to quantify this ,
we normalise the summation of these residuals by the original galaxy s total light , and subtract from this the residual amount of sky after smoothing and subtracting it in the same way .
the size of the smoothing kernel @xmath42 is determined by the radius of the galaxy , and is @xmath95 ( conselice 2003 ) .
note that the centres of galaxies are removed when this procedure is carried out .
the gini coefficient is a statistical tool originally used in economics to determine the distribution of wealth within a population , with higher values indicating a very unequal distribution ( gini of 1 would mean all wealth / light is in one person / pixel ) , while a lower value indicates it is distributed more evenly amongst the population ( gini of 0 would mean everyone / every pixel has an equal share ) .
the value of g is defined by the lorentz curve of the galaxy s light distribution , which does not take into consideration spatial position .
each pixel is ordered by its brightness and counted as part of the cumulative distribution ( see lotz et al . 2004 , 2008 ) .
the 20 parameter is a similar parameter to the concentration in that it gives a value that indicates whether light is concentrated within an image ; it is however calculated slightly differently .
the total moment of light is calculated by summing the flux of each pixel multiplied by the square of its distance from the centre .
the centre is deemed to be the location where 20 is minimised ( lotz et al 2004 ) .
the value of 20 is the moment of the fluxes of the brightest 20% of light in a galaxy , which is then normalised by the total light moment for all pixels ( lotz et al .
2004 , 2008 ) .
this analysis is unique in several ways .
first , it is based on two major data sets - the cosmos and the egs surveys .
the total number of galaxies we study for our morphological analysis is also by far the largest ever published to date .
similar previous work using the goods fields , the hubble deep fields , and the hubble ultra deep field ( e.g. , conselice et al .
2003 ; conselice et al .
2003 , 2004 ; conselice et al . 2005 ; lotz et al .
2008 ) used smaller samples , and thus were able to check systematics in a way we can not due to our large sample of galaxies .
the large number of galaxies in our sample makes the type of checking done in previous work largely impossible ( cf .
jogee et al .
2008 for this type of analysis done on the gems data set ) .
the sample of galaxies in which we use from the egs is described in great detail in conselice et al .
( 2007b ) , and we refer all readers to this paper for characteristics of the sample .
the sample we use for the cosmos field is however not as well characterised , nor as well understood , or as accurately calibrated .
we can however , investigate some basic features of the cosmos data set to determine how it compares with the better calibrated ( for our purposes ) egs data set and sample , described in conselice et al .
( 2007b ) .
figure 1 shows how stellar masses are distributed with redshift in the cosmos field , demonstrating some clustering features which may bias the results , if the merger history is a strong function of environment .
the cosmos field is known to contain a mixture of environments from single galaxies to large groups ( scoville et al .
2007 ) , and our sample is taken from all of these environments . to 120 mm ) relation for the cosmos field .
the vertical lines show the redshift limits for the various bins we use within our analysis .
the horizontal line represents the cut at log m@xmath96 we use in some of our analysis .
note that the clustered redshift peaks are due to real over - densities .
the distribution within the egs for the same stellar mass range is similar ( conselice et al .
2007a).,width=340 ] [ sample - figure ] the other feature through which we investigate the sample is the cas diagrams of concentration - asymmetry and asymmetry - clumpiness ( conselice 2003 ) .
these diagrams reveal which type of galaxies are likely to be within the sample .
what we found in our initial analysis is that a significant number of systems within the cosmos field had concentration values near @xmath97 , which is typically found for stars .
visual inspection of these images reveals that this is indeed the case - many stars appear to be within the released cosmos galaxy catalogs . on the other hand , our star / galaxy separation in the egs ( see conselice et al .
2007 and references therein ) was very effective at removing stars and no such contamination is found in the @xmath98 diagrams .
we utilize the fact that stars have a well defined concentration and small measured radii to remove these systems from our cosmos cas catalog .
figure 2 shows the resulting concentration - asymmetry diagram at @xmath99 and @xmath100 for the combined cosmos and egs sample .
what this reveals is that there is a wide diversity of structures for galaxies at all redshifts , as shown using smaller samples in e.g. , conselice et al .
( 2003 , 2005 ) and lotz et al .
this includes galaxies which would be classified at @xmath12 as early types , late - types , and systems which are asymmetric enough to be ongoing major mergers ( conselice 2006a ) .
one of the most interesting things concerning figure 2 is the clear bimodality of the structural parameters for these systems . at both redshifts bins ,
there is a peak in the density contours near the mid / early - type range , and another peak well within the late - type range .
this bimodality is present within all redshifts , and is the same bimodality found using similar structural parameters by zamojski et al .
( 2007 ) within cosmos . at @xmath100
we find one peak at @xmath101 and @xmath102 , and the other peak at @xmath103 and @xmath104 .
these two values of @xmath90 and @xmath85 are equivalent to early - types and late - type galaxies , and is likely similar to the bimodality found within the colour - magnitude diagram ( e.g. , the red - sequence and blue cloud ) .
another interesting feature of figure 2 is that the peak location for the centres of the bimodality differs slightly with redshift , becoming less concentrated and more asymmetric at high redshift .
this change is @xmath105 = 0.05 and @xmath106 = 0.2 .
the peaks of the bimodality are also softer , suggesting a larger scatter about the bimodality seen at lower redshifts .
the slightly lower values are in principle due solely to redshift effects , or morphological k - corrections , both of which are roughly of the order of the changes seen here .
before we can effectively study the merger history for our sample , we must select a robust sample of mergers , as cleanly as possible ( e.g. , conselice et al .
there are several ways to do this .
the classical rest - frame optical cas definition for determining whether a system is undergoing a merger is given by ( conselice 2003 ) : @xmath107 that is , the asymmetry @xmath85 must be larger than 0.35 and the asymmetry must exceed the value of the clumpiness of the galaxy .
this selection will nearly always cleanly find galaxies in mergers as revealed through nearby samples of galaxies ( conselice 2003 ; de propris et al .
2007 ) , and through n - body simulations of the merger process ( conselice 2006 ) .
tests of how well this criteria does at higher redshifts are limited to either small samples ( conselice et al .
2008a ) , and to the gems survey which was based on f606w ( v ) band imaging of galaxies ( jogee et al .
these studies however reveal a few features that we can use to determine how well the criteria given in equation ( 6 ) does for finding galaxies undergoing mergers .
the first is that although the jogee et al .
( 2008 ) study of gems uses the f606w band , at @xmath108 , it samples rest - frame optical light , as we also do for nearly all our galaxy sample . since jogee et al .
( 2008 ) classify all their galaxies by eye ( which we do not ) , and use the exact same methodology and cas code as we do , we can use their analysis to determine what fraction of galaxies we are finding using criteria from eq . ( 6 ) which are actual merging galaxies .
jogee et al .
( 2008 ) find that at their lowest redshift , the contamination by non - merging galaxies , as defined by visual inspection , is @xmath109% . a significant fraction of this 30% however are irregular galaxies , some of which could be systems in some phase of a merger .
conselice et al .
( 2008a ) within the udf , found that over the entire redshift range of @xmath110 the contamination is lower , roughly 15% .
these differences are likely due to the difficultly of defining what a merger is based on visual appearances .
we also use the clumpiness - asymmetry diagrams of our combined sample and the concentration - asymmetry diagram discussed in 3.1 to get an idea for how well the merger criteria selects merging systems .
first , we note that in paper i ( conselice et al .
2008a ) the clumpiness - asymmetry relation found for normal galaxies in conselice ( 2003 ) does not do a good job in tracing the relation between asymmetry and clumpiness for normal , non - merging , galaxies . for systems at @xmath111 the clumpiness values are lower than their corresponding asymmetry .
this is due to either to the structures of galaxies being actually less clumpy , or in the way the clumpiness is defined by conselice ( 2003 ) , or because of decreased resolution and s / n , the clumpiness decreases .
we reevaluate the criteria for how the asymmetry and clumpiness values for normal galaxies change for galaxies at @xmath111 through the use of the hubble ultra deep field ( described in paper i ) .
the relation between @xmath91 and @xmath85 for all galaxies in the conselice et al .
( 2008a ) sample is shown in figure 3 .
the solid line in figure 3 is the best fit relation between the asymmetry and clumpiness indices for the normal galaxies - i.e. , those that are not merging or peculiar at @xmath112 .
this relation is given by @xmath113 which is roughly @xmath114 .
in figure 4 we show the relation between @xmath85 and @xmath91 for the cosmos+egs sample .
as can be seen , there is a general correlation between these two parameters , with a single peak , and a distribution around @xmath115 and @xmath116 .
figure 4 also shows that the average asymmetry of these galaxies slightly increases at higher redshifts , with the clumpiness index remaining roughly the same at the peak value . to 120 mm [ sample - figure ] to 140 mm from paper
i ( conselice et al .
shown are the three main types of galaxies classified within the udf ( ellipticals , disks and peculiars ) .
the solid line shows the best fit relation between the @xmath85 and the @xmath91 parameters for the normal galaxies ( disks and ellipticals ) .
the dashed line shows the 3-@xmath42 deviation for this fit .
as can be seen , the disk galaxies which have an asymmetry @xmath76 generally follow this relation and have a relatively high asymmetry value for their clumpiness .
, width=340 ] [ sample - figure ] to 120 mm [ sample - figure ] understanding the merger history of galaxies is important for deriving how the merger process drives the assembly of galaxies , as well as its possible role in the triggering of star formation and agn activity .
the merger fraction history at @xmath112 is however considered somewhat controversial , due to published values of the derived merger history differing ( e.g. , conselice et al .
2003a ; lotz et al .
2008 ; de ravel et al .
2008 ) . in this section
we examine the merger history for massive galaxies ( m@xmath1 ) using our combined egs and cosmos data sets .
one of the benefits of using the cas system for finding mergers is that it allows us to quantify the merger fraction , merger rates , and thus the number of mergers occurring in a galaxy population ( conselice et al .
2003a ; conselice 2006b ) .
the reason for this is that the cas method has been well calibrated in terms of the types of galaxies picked up with various selection criteria at different redshifts , as well as estimates for the time - scale sensitivity and the ability to pick up mergers with different mass ratios . in this section
we investigate the merger fractions based on standard techniques we have developed ( e.g. , conselice 2003 ; conselice 2006 ; bridge et al . 2007 ;
3.2 ) .
there are a few caveats to measuring the merger fraction which we must consider before using these values to determine how galaxies are evolving due to major mergers , and to measure the merger fraction at any redshift .
we discuss these issues in 3.3.2 .
we determine the final correction for asymmetries based on redshift and contamination effects , and use these to calculate the merger fraction evolution at @xmath2 , which we discuss in 3.3.3 . in
3.3.4 we use these merger fractions to parameterise the merger history .
we use these results in later subsections of 3 to determine the merger rate , and the average number of mergers galaxies of various masses undergo at different redshifts .
the basic merger fraction ( @xmath117 ) is calculated as the number of mergers selected within a given redshift bin and stellar mass limit ( @xmath118 ) , divided by the total number of galaxies within the same redshift and stellar mass bin ( @xmath119 ) , @xmath120 however , there are several issues that have to be considered before we can use equation ( 8) to derive the merger fraction .
one factor we have to account for are the effects of redshift on the measured cas parameters of our galaxies .
these effects include the lowering of surface brightness , and morphological @xmath24corrections , on the measurement of merger fractions using the cosmos and the egs data .
we address these two corrections separately and quantitatively .
we also discuss how to account for the contamination rate due to non - mergers within our sample .
some of these issues were addressed , at least in terms of redshift effects , by kampczyk et al .
( 2007 ) who examined how a volume limited sample of 1813 low redshift galaxies in the sloan digital sky survey would appear at higher redshifts in the cosmos field .
kampczyk et al .
( 2007 ) simulated galaxy mergers at low redshift to how they would appear at high redshift within cosmos .
they also investigated what fraction of normal galaxies would be misidentified as mergers due to chance superpositions , and other effects .
they found that only one in five of the merging galaxies in the sdss would still be identified as such if placed at @xmath23 , after imaged within cosmos .
kampczyk et al .
( 2007 ) also found that 1.7% of normal sdss galaxies would be identified as mergers due to chance superpositions .
they however conclude that most of the mergers they identify by eye in cosmos at @xmath23 are real .
we go beyond kampczyk in this paper in a number of ways .
first we use quantitative methods that do not rely upon subjective understanding of what a merger is . because we purely use the cas system in this paper , our approach is quantitative and repeatable , and not subject to classification bias .
we also quantify directly using simulations how much the parameters we measure are affected by redshift effects , not just the effects of signal to noise and resolution examined by kampczyk et al .
we also note that kampczyk et al .
( 2007 ) find that the mergers they simulate remain asymmetric , but that the gini / m@xmath77 method fails to locate 90% of their merging galaxies , similar to our findings in the hubble ultra deep field ( paper i ; conselice et al .
2008a ) .
since we use the f814w data to measure the cas values , and therefore merger fractions , of our galaxies , we are sampling different rest - frame wavelengths for galaxies at different redshifts .
we address this empirically in two different ways .
the first is that we can use imaging from the hubble ultra deep field , where we have bv@xmath121 imaging for all galaxies , to determine the morphological @xmath24correction present within our galaxies , as discussed in conselice et al .
( 2008a ) .
the other is that we can use the f606w ( v - band ) and f814w ( i - band ) data from the egs as another determination for how the morphological k - correction changes the resulting measured cas parameters . the f814w band at observed @xmath122 results in a range of probed rest - frame wavelengths from 0.7 @xmath74 m at @xmath4 , to 0.38 @xmath74 m at @xmath25 .
this spans roughly the entire optical wavelength range , where the cas parameters can slight change ( e.g. , conselice et al .
however , these resulting changes are not large ( e.g. , taylor - mager et al . 2007 ;
conselice et al .
as such , we investigate the changes using galaxies in the udf , to determine how much the measured asymmetry in the f814w band changes due to the rest - frame wavelength probed .
we base our measurement of the morphological @xmath24correction on how the measured cas parameters , particularly the asymmetry index , change with respect to @xmath123 . for systems at @xmath124 we utilise the results of conselice et al .
( 2008a ) who find a change of asymmetry parameter with rest - frame wavelength in the optical of @xmath125 @xmath74m@xmath27 at @xmath100 . for galaxies at @xmath99 conselice et al .
( 2008a ) find a change of @xmath126 @xmath74m@xmath27 .
the effective result of this is a maximum change in the asymmetry index of @xmath127 at the highest redshifts .
we perform this correction for each of our galaxies , such that we are measuring the cas parameter at the same wavelength , in this case at rest - frame @xmath123 .
we find similar results when comparing the v and i band cas values within the acs imaging of the egs .
another correction due to redshift is the fact that the surface brightness declines with @xmath128 , and thus the galaxies we image at @xmath23 , with the same effective surface brightness , will appear a factor of @xmath129 fainter in their measured surface brightness than at @xmath130 .
we correct for this effect through simulations , as described previously in conselice ( 2003 ) and conselice et al .
( 2003a ) .
these simulations involve nearby galaxies which are simulated to various redshifts , and their cas parameters are measured and compared with those at lower redshifts .
what we find is that up to @xmath23 , the cas parameters are not affected to a large degree , and the asymmetry index declines by @xmath131 up to @xmath23 .
we therefore use this as the maximum correction at @xmath23 and apply it to galaxies at @xmath132 where this effect is found to be important .
the other correction we perform is to correct the cas merger fraction measures for contamination from galaxies which are within the merger region of cas space , but which are not structurally undergoing a merger ( see also 3.2 & conselice et al .
the estimates of this correction must be done by eye , and the large sample in this paper makes this exercise too prohibitively large to carry out .
previous studies however have revealed what this fraction is from samples of @xmath133s of galaxies ( jogee et al .
2008 ; conselice et al .
conselice et al .
( 2008a ) find that the contamination rate is roughly 14@xmath13410% for galaxies observed in the @xmath135band ( conselice et al .
other results include an examination of the cas and morphological types for @xmath136 galaxies in the gems v@xmath137 band ( jogee et al .
2008 ) , which find a contamination rate of @xmath138% at @xmath139 which is rest - frame @xmath140 .
since the gems survey better matches the depth and rest - frame wavelength of our images , we use a contamination correction of @xmath141 = 25% .
this correction is potentially smaller than this due to the removal of ` irregular ' galaxies from jogee et al .
( 2008 ) , some of which are potentially merging systems .
the final measure of the asymmetry index to calculate the merger fraction is given by : @xmath142 where @xmath143 is the ( usually negative ) morphological k - correction , @xmath144 is the ( positive ) correction for redshift effects , and @xmath141 is the fraction of galaxies found through cas which are not actually merging systems , which we take as @xmath141 = 0.25 in this paper .
the first observation we derive from our cas values is the evolution of the merger fraction ( figure 5 ) for both the egs and cosmos fields separately .
error bars reflect uncertainties due to shot noise , and photometric redshift errors .
before we discuss the merger fraction evolution it is important to address the issue of whether our merger analysis , which finds galaxies with a high asymmetry , locates the same types of galaxies at different redshifts .
a legitimate concern is that at higher redshifts , galaxies can be asymmetric due to reasons other than major mergers , such as star formation and minor mergers .
certainly , we know there are differences between galaxies at various redshifts , as for example , the star formation rate at z = 1 is several times higher than it is at z = 0 .
however , several detailed analyses over the past few years have shown that galaxies with high asymmetries are almost always major mergers .
this includes calibrating how the asymmetry index works , not just at low redshifts ( conselice 2003 ) , but also at high redshift ( e.g. , conselice et al . 2005 , 2008a ) .
these studies have found that galaxies with evidence for merging , both kinematically ( e.g. , kassin et al . 2007 ) , and through visual inspection , have high asymmetries , and likewise the non - mergers are not asymmetric .
note that this is only true in optical light , as asymmetries for star forming galaxies in the rest - frame uv can be quite high , independent of any merging activity ( taylor - mager et al .
2007 ) .
it is also possible that interactions and minor mergers can produce large asymmetries .
however , for reasons explained in previous papers , such as conselice ( 2006b ) and de propris et al .
( 2007 ) , it is unlikely that very many asymmetric galaxies are produced through interactions or minor mergers . de propris et al .
( 2007 ) examined the asymmetry values for interacting , merging , and normal galaxies in the nearby universe .
they found that very few of the interacting galaxies are considered mergers in the cas system , and those that would be counted are in very close pairs which are about to merge . by far the bulk of the systems with high asymmetries
are found in the major merger category .
this was furthermore quantified through n - body simulations by conselice ( 2006 ) who found that only major mergers with ratios of @xmath145 would be counted as a major merger within the cas system . because the asymmetry index , and the other cas values , are luminosity weight , faint features , such as tidal debris ( e.g , kawata et al .
2006 ) do not produce large asymmetries .
we thus determine the merger fraction for galaxies of various masses using the criteria from equation ( 6 ) and by using the asymmetries calculated in equation ( 9 ) .
our final merger fraction values are tabulated in table 1 , and are separated into the egs and cosmos fields in figure 5 .
figure 6 shows the merger fraction evolution for our m@xmath1 sample , with the @xmath146 point from de propris et al .
( 2007 ) , and the @xmath16 merger fractions from conselice et al .
( 2008a ) .
.merger fractions for galaxies with m@xmath1 [ cols="^,^,^,^",options="header " , ] we note that the merger fractions we measure are in no sense the total galaxy merger fraction , that is the total number of galaxies which have undergone , or are undergoing , a merger at the given time .
most galaxies will have undergone a merger sometimes in their history , and the fraction of galaxies which have undergone a merger sometime in the past will be close to 100% .
all merger fractions we derive in this paper are computed using the cas parameters , which is only sensitive to a well defined time - range during the major merger process ( conselice 2006 ; lotz et al
. 2008b ) . in the case of cas mergers ,
this time - span is roughly 0.4 - 1 gyr ( conselice 2006 ; lotz et al .
as shown in conselice ( 2006 ) there are phases of a merger which will not be picked up by the cas technique .
a different technique will find a different merger fraction if it has a time sensitivity different from the cas system .
for example if galaxy pair methods has roughly a factor of two longer time - scale for finding a merger than the cas system , the resulting computed merger rate would be the same , as the pair method would find a factor of two more galaxies merging than the cas parameters .
we therefore expect , and find there to be , galaxies that by eye appear as a merger , but do not have a high asymmetry ( e.g. , conselice et al .
2008a ) .
there are a few obvious features of the merger fraction that deserve note , as seen in figure 5 and 6 .
the first is that at the redshifts where we have data for both the egs and cosmos , the two agree remarkably well , always within 1-@xmath42 ( see also table 1 ) .
there is also a general trend for the merger fractions , as measured in both fields , to decline at lower redshifts , going from @xmath147 at @xmath5 down to @xmath148 at @xmath108 in the egs field , with a similar trend within the cosmos field .
the lowest redshift point for the cosmos data at @xmath4 is slightly higher than the two nearest and higher redshift bins .
this is at least partially due to the higher resolution reached at this redshift , which can result in higher computed asymmetries ( e.g. , conselice et al .
2000a ) .
the other obvious and outstanding feature of our measured merger history is the fact that the merger fraction , based on our measurements , declines rapidly at @xmath8 , although it appears roughly constant at @xmath149 .
this result is unlikely solely due to cosmic variance , as both fields show this behaviour , and this result is relatively insensitive to the exact corrections we apply for redshifts and @xmath24corrections .
this result is , within our errors , significant at @xmath150-@xmath42 , based on the combined egs+cosmos sample .
we also fit these two redshift ranges in the egs+cosmos merger fraction evolution separately , as shown in figure 6 by the two dashed lines .
here we fit both the evolution at @xmath8 and @xmath149 as power - laws , in which we find very different slopes . at @xmath8
we find that the best fit power - law slope is @xmath151 with @xmath152 , going right through the de propris et al .
( 2007 ) @xmath146 value . at @xmath149
the slope is @xmath153 , which indicates very little evolution , and therefore must be the result of significant merging activity within this epoch .
we also see a turnover in the merger history at @xmath154 ( conselice et al .
2008a ) , which has also been seen for galaxies in pairs at similar redshifts ( ryan et al .
2008 ) .
furthermore , figure 7 compares our merger fractions with results from other merger fraction studies in figure 7 . specifically , we compare our merger history with the results of both previous pair and structural studies .
points on figure 7 include de ravel et al .
( 2008 ) who use the vvds to determine the pair fraction evolution for galaxies from @xmath155 to @xmath156 .
we plot on figure 7 the results from this study for galaxies which have a pair separation of @xmath157 20 h@xmath27 kpc , and a velocity difference @xmath158 500 within a magnitude range of m@xmath159 .
the @xmath160 factor accounts for the evolution of stellar populations , and is an attempt to match the photometric evolution of galaxies so as to obtain the same galaxies at different redshifts .
we also compare with recent deep2 results for the pair fraction evolution from lin et al .
( 2008 ) , who find similar pair fractions to de ravel et al .
the other pair fraction we compare with is kartaltepe et al .
( 2007 ) who measure pair fractions within cosmos within @xmath157 20 h@xmath27 kpc separation using photometric redshifts .
this paper finds 1749 galaxy pairs for systems brighter than m@xmath161 , roughly equivalent to the stellar mass limit we have used .
finally , we compare our merger fractions to those published by lotz et al .
( 2008a ) using the gini / m@xmath77 method for 3009 galaxies within the egs brighter than 0.4l@xmath162 .
to 120 mm through the use of the cas system for the egs ( solid circles ) and cosmos ( open circles ) fields .
the open stars are the derived merger fractions from the egs using the gini / m@xmath77 parameters from lotz et al .
the point at z = 0.05 originates from de propris et al .
the five solid lines show the evolution from @xmath146 as a power - law , with @xmath163 fixed to 0.009 , and with various values of the power - law index shown ; m = 1,2,3,4,5 .
, width=340 ] [ sample - figure ] to 140 mm to @xmath14 using structural parameters from the cas system .
the point at @xmath146 originates from the study of de propris et al .
the points between @xmath108 and @xmath5 are from the combined egs and cosmos fields , which are plotted individually on figure 5 .
the points from @xmath65 ( crosses ) are taken from the cas study of the hubble deep and ultra deep fields ( conselice et al .
we show several fits as well , with the solid straight line the best fit power - law , while the curved solid line is the best fit for the combined exponential / power - law parameterisation ( see text ) .
also shown as the two dashed blue lines are the best fit power - law parametrisation for the egs+cosmos points at @xmath164 and @xmath149 , fit separately .
this shows the rapid evolution in the merger fraction within @xmath2 .
the dotted line shows the previous best fit exponential / power - law relation for galaxies with stellar masses m@xmath1 described in conselice et al .
( 2008a ) . ,
width=340 ] [ sample - figure ] to 140 mm method ( lotz et al .
the blue symbols are the merger fractions derived from pair studies , either kinematic pairs as in lin et al .
( 2008 ) and de ravel et al .
( 2008 ) or photometric pairs from the cosmos fields ( kartaltepe et al . 2007 ) .
in general the morphologically defined mergers give a higher merger fraction than those derived through pairs .
, width=340 ] [ sample - figure ] in addition to systems selected by m@xmath1 , we also examine the evolution of the merger fraction for the extremely massive galaxies in our sample , those with m@xmath165 .
there are several issues however when measuring the merger fraction for these systems .
the first is simply that there are not nearly as many galaxies with these masses as for systems selected by m@xmath1 .
this can be seen in figure 8 , where at the lowest redshifts at @xmath166 there are not enough systems to measure the merger fraction , and very few at @xmath167 , with resulting large error bars . in any case
, what we find is that the merger fractions for m@xmath165 galaxies at @xmath3 overall are very similar to that for m@xmath1 systems .
it appears that at @xmath168 the merger fraction for these extremely massive galaxies is slightly lower than for the @xmath169 systems .
this changes at lower redshifts where the m@xmath165 galaxies appear to have slightly higher merger fractions , although we note that the error bars on these fractions do not rule out that these points are in fact as low , or even lower than , those for systems at @xmath169 .
there are a few popular ways to fit the merger fraction evolution .
the first is the traditional power - law format ( patton et al .
2002 ; conselice et al .
2003 ; bridge et al .
2007 ) which dates back to early work by zepf & koo ( 1989 ) on the evolution of the merger fraction .
this fitting format is given by , @xmath170 where @xmath171 is the merger fraction at a given redshift , @xmath163 is the merger fraction at @xmath12 , and @xmath19 is the power - law index for characterising the merger fraction evolution .
zepf & koo ( 1989 ) calculated a merger fraction evolution which increased with redshift up to @xmath172 , and obtained a slope of @xmath173 .
nearly all further studies have found values within this range .
however there is currently a debate over the true nature of the increase , with some studies finding a fast evolution ( conselice et al .
2003 ; kartaltepe et al .
2007 ) , while others have found a more modest evolution ( e.g. , lotz et al . 2008a
; lin et al . 2008 ) .
there are several reasons for this diversity in the parameterisation of the merger fraction evolution , which we discuss after demonstrating the various ways the merger fraction evolution can be fit within our own sample .
we explore below the power - law behaviour of the merger fraction evolution in a few ways .
the first method is to investigate the power - law fit to the merger fractions solely within the cosmos+egs data set ( figure 5 ) .
when we fit a power - law to this evolution using just the points from cosmos+egs at @xmath3 we find a best fit given by @xmath174 and @xmath11 .
that is , internal to itself , the cas merger fraction from @xmath4 to @xmath5 increases modestly with time .
note that the derived merger fraction at @xmath12 for this sample is 0.025@xmath175 which is about a factor of three larger than the @xmath12 merger fraction derived from de propris et al .
when we hold the @xmath15 point at a constant value of @xmath176 we find that the best - fit slope for cosmos+egs data increases to @xmath13 .
our derived evolution is similar to the results found by lotz et al .
( 2008a ) , whose merger fractions are similar to ours ( figure 7 ) .
our merger fraction evolution power - law slope , @xmath19 , is however smaller than the index @xmath19 measured by kartaltepe et al .
( 2007 ) using pair counts .
even though kartaltepe et al .
( 2007 ) find in general lower pair fractions compare to our merger fractions , the increase with time is more pronounced , and they derive an index of @xmath177 .
this is more than 2-@xmath42 away from our result , and this difference deserves some explanation . when we fit the kartaltepe et al .
( 2007 ) points with the same @xmath12 prior we use for the egs+cosmos data , we find that the best fit slope is @xmath1780.1 . when we remove the @xmath15 prior in this fit , we find that the slope changes to @xmath179 , although most of this is driven by the high redshift points from kartaltepe et al .
if we remove the highest redshift point ( figure 7 ) , the measured value of @xmath180 , which is very similar to our own value at the same redshift range . to 140 mm and m@xmath165 from @xmath181 . in general we find that the evolution at higher masses is similar to those presented in the other plots in this paper , although the higher mass systems have a higher uncertainty associated with their measured merger fractions .
, width=340 ] [ sample - figure ] overall , we find that the value of @xmath19 can vary significantly depending on what value of @xmath163 is chosen .
for example ,
if the @xmath12 merger fraction for log m@xmath182 galaxies is twice as high as the de propris et al .
value ( i.e. , @xmath183 ) then the best fit value of the power - law slope changes from @xmath13 to @xmath184 , thus raising the @xmath12 merger fraction lowers the value of the power - law slope . if we hold the @xmath15 merger fraction as 4% , then we find that the power - law slope falls to @xmath185 .
note that this does not appear to be the case for very massive galaxies with m@xmath165 ( bluck et al .
2008 ) .
since the @xmath15 merger fraction can be critically important for deriving the merger history if it is used as a prior , it is worth examining what the merger fraction at @xmath15 is , and how certain we know this number .
we also have some estimates of what the local merger fraction is from previous work such as patton et al .
( 2000 ) and de propris et al .
we use the millennium galaxy catalog ( mgc ) selection for mergers , which is based on systems which are at @xmath186 , and thus a complete comparison is not possible without measuring the stellar masses for these galaxies .
de propris et al . ( 2007 ) find that the merger fraction at @xmath187 , is @xmath188 , based on the cas asymmetry .
by holding the value of @xmath163 fixed to 0.009 , we fit a power - law slope of @xmath13 .
however , this is not a good fit to the data ( figure 6 ) , and tends to under - predict the merger fraction at @xmath189 and over - predict the fraction at @xmath190 .
one of the features of the derived merger fraction up to @xmath5 is that between various neighboring redshifts there are significant differences in the derived merger fraction ( figures 5 - 7 ) .
this is true for the combined egs+cosmos sample , the egs and cosmos samples alone , and is also found for the gini / m@xmath77 approach in lotz et al .
( 2008a ) , as well as the pair method derived in kartaltepe et al .
in fact , the slope of the power - law merger fraction evolution ranges from @xmath191 to @xmath192 between two neighboring individual merger fractions .
it is not yet clear what the origin of these differences are , whether the result of cosmic variance , or systematics .
however on average the merger fraction does increase slightly with time up to @xmath5 ( 3.3.3 ) .
we do a similar examination of the best power - law fit to the merger fraction evolution by including higher redshift merger fractions taken from a combined udf+hdf sample ( conselice et al .
2008a ) selected with the same stellar mass cut of m@xmath1 .
the resulting merger fraction evolution is shown in figure 6 , which is currently our best estimate of how the merger fraction varies with time for a mass selected sample .
when we fit the merger fraction evolution for m@xmath1 galaxies from @xmath12 to @xmath14 we find that the best fit using all points , and no priors , is : @xmath193 and @xmath194 , although this power - law fit is poor , and the @xmath195 is large .
if we hold the value of @xmath176 , we find that the best fit value of the power - law slope increases to @xmath196 , although this does not go through the highest redshift point , and results in a poor fit .
however , the @xmath12 extrapolation from the best fit to all the data predicts that the nearby merger fraction is roughly twice the best value we currently have .
this again demonstrates the importance of the value of @xmath163 within the power - law fit .
the value of @xmath163 determines to a large degree what the fitted slope of the power - law evolution will be , especially if the value is held constant during the fit .
another way to characterise the merger fraction evolution , which dates back to theoretical arguments based on the press - schechter formalism for merging ( carlberg 1990 ) , is a combined power - law exponential evolution .
this form appears to be a better fit to all of the redshift data than a simple power - law ( conselice 2006b ) .
this is however not the case for the merger fraction for m@xmath165 galaxies at @xmath18 , which can be fit by a power - law ( bluck et al .
the formula for the power - law / exponential evolution is given by : @xmath197 where the @xmath12 merger fraction is given by @xmath198 exp(@xmath199 ) .
we find in general that this combined exponential power - law fit is better than a simple power - law and is likely a better representation for how the merger fraction evolves with time .
we also find this to be the case for a power - law / exponential form , without the square on the @xmath200 term in the exponential .
however , none of these forms are very satisfying , given the large errors on their fits , and the three free parameters needed to create the fit . as discussed in paper ii ( conselice et al .
2009 , submitted ) there does not appear to be a simple way to parameterise the merger fraction .
we explored several fitting routines , and found that the best - fit two parameter model is an exponential / power - law of the form : @xmath201 which is designed such that @xmath202 , and is only a two parameter parameterisation , and fits the data as well as the three parameter model above . in any case
, we have found that the fitting with the exponential / power - law and the carlberg ( 1990 ) version of the exponential / power - law , gives an exponent on the power - law portion of @xmath194 .
this is the index predicted for how the number densities of galaxies decline with redshift , and is therefore the natural exponent on the power - law for a passive merger evolution , which appears to occur at @xmath8 .
one of the results we derive from our measured structural mergers is how the derived merger fractions , based on the cas and other methods ( such as gini / m@xmath77 ) , compare with derived pair fractions .
figure 7 shows this comparison , with the morphological methods ( this work & lotz et al .
2008a ) shown in red , and the pair methods ( kartaltepe et al . 2007 ; lin et al .
2008 ; de ravel et al .
2008 ) shown in blue .
what is immediately clear is that the structural methods , while agreeing quite well with each other , find a higher merger fraction than the pair methodology .
there are several possible reasons for this . while it is possible that one or both methods have systematics that result in inaccurately measured merger fractions , we first investigate what the time - scales for these methods are .
this is an important question as these fractions differing can be partially , or entirely , explained if the time - scales for merging for the two methodologies are different .
for example , if the time - scale for a 20 kpc pair to merge is half the time - scale sensitivity for an asymmetric galaxy , then we would expect the pair fraction to be half of the cas merger fraction if both methods are tracing the same merger process . that is , the merger fraction for a given sample scales as the time - scale sensitivity of the method used to find mergers .
since the merger rates for the two methods should give the same result , then the ratio of the pair fraction to its merger timescale should be equal to the structural merger fraction divided by its time - scale , or : @xmath203 time - scales ( @xmath71 ) within the merger process are notoriously difficult to measure , and have large uncertainties .
initially , measuring the merger rate from pairs involved dynamical friction calculations ( e.g. , patton et al .
2002 ; conselice 2006 ) , and typical time - scales for a 20 kpc pair to merge are 0.5 - 1 gyr with various assumptions .
the time - scale calculation for dynamical friction used by conselice ( 2006b ) , and in earlier studies are based on isothermal mass distributions and the time - scale can depend highly on the mass of the galaxies , and the characteristic velocity of the system ( e.g. , conselice 2006b , eq .
likewise , the time - scales for merging within the cas system for dark matter dominated galaxies was found to be similar to the dynamical friction time - scales derived from an isothermal profile , and a galaxy with mass of @xmath204 .
there are however problems with both of these calculations , which have already been alluded to above for the dynamical friction time - scale .
the measured time - scales for the cas method are found by conselice ( 2006 ) to vary between @xmath205 gyr , depending on viewing angle and the orbital parameters of the two galaxies in the pair . also , the simulations used in conselice ( 2006 ) are purely dark matter , and it is desirable to determine the cas time - scale for systems with stars , star formation , and dust .
new simulations were recently analysed by lotz et al .
( 2008b ) in terms of cas , gini / m@xmath77 , and pair selection for mergers .
lotz et al .
( 2008b ) furthermore utilise simulations that include star formation and dust , and are currently by far the most thorough investigation into merger time - scales using both morphology and the time - scales for mergers to occur within a given pair separation from 20 h@xmath27 kpc to 30 h@xmath27 kpc , and 50 h@xmath27 kpc .
lotz et al .
( 2008b ) find a variety of time - scales for their merger simulations depending on the type of merger and the type of galaxy .
for their highest resolution simulation ( sbcppx10 ) , they calculate a merger time - scale of @xmath206 for the cas methodology , and a time - scale of @xmath207 for galaxies to merge within a 20 h@xmath27 kpc pair .
this ratio of time scale , which we denote as @xmath208 , is equal to @xmath209 6.3 for this model .
figure 9 shows the histogram for the value of @xmath210 for all the models published in lotz et al .
( 2008b ) .
what we find is a general distribution , but with all the sbc models having a ratio @xmath211 .
the models shown by the blue hatched histogram in figure 8 are for the lotz et al .
` g ' models , which are less gas dominated , and have sub - parabolic orbits which lead to artifically shortened merger - times in pairs . as can be seen , while earlier analytical calculations showed that the time - scale for pair merging was similar to the morphological merger time - scale , the simulations by lotz et al .
demonstrate that the time - scales for pairs to merge are shorter than the analytical estimates .
this implies that if the pair method and the structural methods are measuring the same merger process , then the merger fractions derived from pairs should be lower than those derived from structure by an equivalent amount .
we denote this ratio as @xmath212 . for the cosmos field
we find that the value of @xmath213 using the data from kartaltepe et al .
( 2007 ) and this paper , give @xmath213 values from 1.5 - 3.5 , yet the pairs from kartaltepe et al .
( 2007 ) are not selected in the same way our cas mergers are , i.e. , with m@xmath1 . a better test of the merger criteria time - scale is to determine how the ratio of the cas merger fraction and the pair fraction change with redshift in the same sample and stellar mass selection .
while it is likely that even within the same sample , the cas and pair methods will find galaxies in different modes of evolution ( de propris et al .
2007 ) , it is still instructive to determine this ratio within a well defined sample .
we utilize the powir database from the egs to determine this ratio ( e.g. , conselice et al .
2007b , 2008b ) .
figure 10 shows this ratio for galaxies at separations of less than 20 kpc , 30 kpc , and 50 kpc .
the values for f@xmath214(cas)/f@xmath214(pair ) , with a pair defined as having a separation of @xmath215 kpc , range from 12.2 to 2.3 , with a average value of 6.2 , which perhaps coincidently , is close to the value of @xmath210 for the highest resolution model from lotz et al .
( 2008b ) .
this implies that there is no inconsistency between the pair method and structural methods for measuring merger fractions and rates . to 140 mm
kpc pairs ( @xmath210 ) .
the blue hatch histogram shows the results for galaxies in sub - parabolic orbits , while the solid yellow show the results for the more realistic sbc orbital models from lotz et al .
( 2008b ) . ,
width=340 ] [ sample - figure ] to 140 mm within the egs and the other powir fields ( conselice et al .
2007b , 2008b ) .
the blue squares are for those systems with separations of @xmath215 kpc , the green triangles are for those with separations of @xmath216 kpc , and the red circles are for systems with separations of @xmath217 kpc .
, width=340 ] [ sample - figure ] we are now in a position to measure the galaxy merger rate for m@xmath1 galaxies from @xmath12 to @xmath14 .
the merger rate is difficult to measure , however , and our attempt should be viewed as a preliminary full measure of this evolution up to @xmath14 .
measurements of the merger rate will improve as our understanding and knowledge of galaxy number densities , merger fractions , and merger times scales improve .
since the merger rate has a relatively high error associated with it , we first investigate the merger rate per galaxy , which is simply just the number of mergers a galaxy of a given mass will undergo as a function of time .
we , in fact , examine the inverse of this , which we call @xmath218 , which is the average amount of time a galaxy exists before undergoing a merger as a function of redshift , or @xmath219 where we have utilised the galaxy merger fraction , or rather the fraction of galaxies merging in a population , which is related to the merger fraction ( @xmath117 ) by : @xmath220 in equation ( 14 ) , we also use the time - scales for mergers ( @xmath221 ) , which is potentially the largest uncertainty when trying to derive evolution from the merger fraction ( 3.4 ) . the value for @xmath221 is measured to range from @xmath222 gyr based on n - body models from conselice ( 2006b ) and from lotz et al .
( 2008b ) .
the nominal value we use , based on the average cas time - scale using the lotz et al .
( 2008b ) sbcpp , sbcpr and sbcrr simulations is 1 gyr , with an uncertainty in this value of 0.3 gyr .
note that this is more than a factor of two larger than the time - scale used in conselice et al .
( 2008a ) , which utilised simulations from conselice ( 2006 ) to derive a merger time - scale of 0.34 gyr .
the time - scale for a pair to merge , as found by lotz et al .
( 2008b ) , is 0.2@xmath223 gyr for a 20 kpc pair , and @xmath224 gyr for merging systems within the gini - m@xmath77 system , which we also use to calculate the value of @xmath218 , as plotted in figure 11 .
as can be seen in figure 11 , the typical value for @xmath218 is around @xmath225 gyr at @xmath17 , demonstrating that @xmath17 galaxies with masses m@xmath1 only undergo , on average , a single merger during this time .
we can calculate the exact number of mergers that occur at @xmath18 by integrating the inverse of @xmath218 over redshift , @xmath226 where @xmath227 is the hubble time , and @xmath228^{-1/2}$ ] = @xmath229 . in this case
, we parameterise @xmath218 by : @xmath230 where we find a best fit of @xmath231 ) gyr , and @xmath232 . using equation ( 16 ) , and the parameterisation of @xmath218 in equation ( 17 ) we calculate the number of mergers a galaxy with m@xmath1 undergoes from @xmath14 to @xmath12 .
the total number of mergers depends strongly on the adopted value of the cas merger time - scale ( @xmath221 ) .
the range in the total number of mergers a galaxy undergoes at @xmath18 ranges from @xmath233 = 2.3 to 6.6 , depending on the time - scale used . by integrating the individual merger fractions
, we calculate that the total number of mergers a galaxy undergoes can be expressed as @xmath234 at @xmath18 .
most of the merging within these massive galaxies occurs at @xmath65 , independent of the value of the merger time - scale , as discussed in conselice et al .
( 2008a ) .
the total number of cumulative mergers a galaxy undergoes from @xmath14 to @xmath12 is shown in figure 12 for three different time - scales of cas sensitivity to the merger process - @xmath235 gyr .
also shown in figure 12 , as the dashed line , is the evolution in the total cumulative number of mergers for a population with a constant @xmath236 gyr .
although the total number of mergers depends strongly on the still uncertain time - scale for the merger process , it is clear that most merging for massive galaxies must occur at @xmath237 .
the horizontal line s intersection with the various models shows at which redshift a galaxy will on average have had a single major merger since @xmath14 .
this varies between @xmath238 to @xmath239 .
thus , independent of the time - scale used , on average , massive galaxies with log m@xmath182 undergo a merger between @xmath239 and @xmath238 . likewise , at @xmath17 ( the vertical solid line ) , a typical massive galaxy will undergo between 0.5 to 2 major mergers , depending on the time - scale sensitivity .
our best estimate for the amount of stellar mass galaxies with m@xmath1 grow by at @xmath156 to @xmath12 , based on the evolution of the mass function , is a factor of @xmath240 increase ( conselice et al .
2007 ) , demonstrating consistency between our findings and the actual evolution in stellar mass in galaxies .
another important physical quantity to obtain when studying galaxy mergers , is the galaxy merger rate ( @xmath241 ) , which is the number of galaxies merging , per unit time , per unit co - moving volume .
the galaxy merger rates for our sample are calculated through the merger rate equation , @xmath242 where n@xmath243 is the number density of galaxies within a given stellar mass range , and f@xmath243 ( eq .
15 ) is the galaxy merger fraction .
note that this is not the merger fraction , which is the number of mergers divided by the number of galaxies , which is roughly half the galaxy merger fraction ( conselice 2006 ) .
we obtain our galaxy number densities @xmath244 from conselice et al .
( 2007b ) for galaxies with stellar masses
m@xmath1 at @xmath112 , and from cole et al .
( 2001 ) for galaxies at @xmath15 . to 140 mm , which is the average time between a merger for galaxies with m@xmath1 at @xmath18 .
we include on this plot the measured pair fractions ( e.g. , kartaltepe et al . 2007 ; de ravel et al . 2008 and lin et al . 2008 ) .
the merger fractions by which the value of @xmath218 is calculated from galaxy structure , include those from this paper , and previous work by lotz et al .
( 2008a ) at @xmath2 , and conselice et al .
( 2008a ) for systems at @xmath65 .
the solid line shows the best fit power - law evolution , which we use to parameterise @xmath218 for measuring the cumulative merger history for m@xmath1 galaxies.,width=340 ] [ sample - figure ] to 140 mm , as a function of the time - scale for cas sensitivity to the merger process .
the three solid lines show the evolution of how many mergers have occurred for galaxies with m@xmath1 since @xmath14 using different values for the time - scale in which the cas system is sensitive to mergers ( see text ) .
the dashed line shows the evolution for mergers with a constant time - scale of @xmath236 gyr .
, width=340 ] [ sample - figure ] the resulting galaxy merging rate is shown in figure 13 .
the galaxy merger rate increases with time for the m@xmath1 galaxies , from @xmath12 to @xmath14 .
this slightly differs from the findings of bluck et al .
( 2009 ) who find that the merger rate for m@xmath165 galaxies is consistent with being constant from @xmath12 to 3 .
we try to parameterise this evolution of the galaxy merger rate in various ways , and find that a linear form of @xmath245 gives the best fit ( as shown by the solid line in figure 13 ) .
however , as can be seen in figure 13 , this linear fit predicts that the merger rate should drop quicker at @xmath246 than what is actually seen .
this is potentially the result of either the merger rate dropping linearly from @xmath247 until @xmath248 , and then dropping very quickly and remaining relatively constant at @xmath249 .
it also could be the result of the time - scale for the pair measurements at @xmath15 to be incorrect .
if the time - scale were 0.4 gyr , this would drop the merger rate by a factor of two for the @xmath15 point , although this would still not alleviate the problem of matching with the rate of linear decrease seen at higher redshifts .
the merger rates at @xmath250 appear to be constant for m@xmath1 galaxies .
the merger rate therefore decreases nearly linearly with redshift at @xmath251 , whereas the merger fraction is relatively well fit as a power - law decrease over the same epoch .
while the galaxy number densities for m@xmath1 systems remains relatively constant at @xmath17 , they decline quickly at higher redshifts ( e.g. , conselice 2007 ) .
this is matched , and then some , by the increased merger fraction , which produces a linear decline with decreasing redshift .
it is possible , as for the case of very massive galaxies discussed in bluck et al .
( 2009 ) for the number density decline to match the increased merger fraction , producing a relatively flat evolution of the merger rate .
the time integral of the merger rate gives us the number of mergers which have occurred per unit volume since @xmath14 .
the result of this calculation for the number of mergers which have occurred in a 10 mpc@xmath252 co - moving cube is shown in figure 14 .
the solid lines show the evolution for the same time - scales as in figure 12 .
clearly , the number of mergers in this volume , which contains roughly six galaxies today , increases rapidly at higher redshifts , but levels off at @xmath17 .
most of the merging in the universe , as found in previous work ( e.g. , conselice et al .
2003 ; lotz et al .
2008a ) , is at higher redshifts . as discussed in bertone & conselice ( 2009 ,
submitted ) , this is in disagreement with cold dark matter semi - analytical models , such that we are finding a higher merger rate at very high redshifts , compared with the simulations . to 140 mm ) , defined as the number of galaxies undergoing a merger per unit time , per unit co - moving volume , as a function of redshift .
shown here are only systems with m@xmath1 , as measured by the cas system from this paper at @xmath3 ; conselice et al .
( 2008a ) for @xmath253 galaxies ( blue circles ) , while the @xmath146 is the rate derived from de propris et al .
( 2007).,width=340 ] [ sample - figure ] to 140 mm ) , from @xmath14 to @xmath12 over a volume of ( 10 mpc)@xmath252 .
this integral gives the total integrated number of mergers which have occurred within this volume since @xmath14 .
as for the integral of @xmath254 ( figure 12 ) , most of this merging occurs within a co - moving volume at higher redshifts , with roughly 20 - 60 mergers occurring at @xmath18.,width=340 ] [ sample - figure ]
we have carried out an analysis of the structural cas parameters for a sample of @xmath255 20,000 galaxies with stellar masses m@xmath1 within the egs and cosmos fields , between @xmath4 and @xmath5 .
we explore in this paper the merger fraction history , including various parameterisations , comparison of structural and pair merger fractions , the merger rate and role of mergers in galaxy formation , as well as systematics which are possibly playing a role in the derived merger fraction .
the primary results from this paper are : \ii .
we compare our derived merger fractions to previously published merger fractions from lotz et al .
( 2008a ) and pair studies from kartaltepe et al .
( 2007 ) , lin et al . ( 2008 ) and de ravel et al .
our results are comparable with the gini / m@xmath77 derived merger fractions from lotz et al .
( 2008a ) , but we find our merger fractions are between 3 - 6 times higher than those derived from pairs , even within the same stellar mass selection .
we argue this in some detail through the use of a m@xmath165 sample of galaxies in the egs .
we also show that this ratio of structural mergers and pair fractions is however predicted in the latest n - body models of galaxy mergers from lotz et al .
( 2008b ) and is due to differing merger time - scales .
we investigate various methods , including the use of priors , for parameterising the merger history through the cas method .
we show that the power - law formalism , whereby the merger fraction is parameterised by @xmath256 , is a poor fit to the merger fractions at @xmath17 .
we further show that the value of the power - law slope , @xmath19 , can vary depending on whether a prior is used to set the local @xmath12 merger fraction .
if we fit only our @xmath257 merger fractions from cosmos+egs we fit a merger fraction evolution of @xmath258 . however , by using a prior of @xmath259 from de propris et al .
( 2007 ) , the slope of the power - law fit is @xmath13 \iv . we find that a combined power - law exponential is a better fit to the merger fraction at @xmath18 than just a simple power - law .
we fit this form to the merger fraction using a @xmath12 prior , and without , finding a similar form that predicts a turnover in the merger fraction for m@xmath260 galaxies at z = 2 .
v. we calculate the merger rate by using the latest measures of merger time - scales from merger simulations , and number densities for galaxies with m@xmath260 from conselice et al .
( 2007b ) .
although the errors on measuring the merger rate are large , resulting from uncertainties in the merger fraction , the merger time - scale , and the galaxy density , it is the ultimate quantity , and can reveal the role of mergers in forming galaxies .
we find that the merger rate is roughly constant up to @xmath9 , and increases at @xmath65 by a factor of 10 .
we also show that the merger rate for m@xmath260 galaxies can be parameterised up to @xmath261 as a linear increasing function .
we thank the staff at the palomar and keck observatories for invaluable assistance in collecting the data used for the extended groth strip survey . funding to support this effort came from a national science foundation astronomy & astrophysics fellowship , and grants from the science technology facilities council ( stfc ) .
support for the acs imaging of the egs in go program 10134 was provided by nasa through nasa grant hst - g0 - 10134.13-a from the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, under nasa contract nas 5 - 26555 .
99 arp , h. 1966 , apjs , 14 , 1 barton , e.j . ,
geller , m.j . ,
kenyon , s.j .
2000 , apj , 530 , 660 benitez , n. 2000 , apj , 536 , 571 bershady , m.a . ,
jangren , j.a . ,
conselice , c.j . 2000 , aj , 119 , 2645 bridge , c. , et al .
2007 , apj , 659 , 931 bruzual , g. , charlot , s. 2003 , mnras , 344 , 1000 bluck , a. , conselice , c.j . ,
bouwens , r.j . , daddi , e. , dickinson , m. , papovich , c. , yan , h. 2008 , arxiv:0812.0926 bundy , k. , fukugita , m. , ellis , r. , kodama , t. , conselice , c.j .
2004 , apj , 601 , 123l bundy , k. , et al .
2006 , apj , 651 , 120 bundy , k. , ellis , r.s . , conselice , c.j .
2005 , apj , 625 , 621 carlberg , r. 1990 , apj , 359 , 1l cassata , p. , et al .
2005 , mnras , 357 , 903 cole , s. , et al .
2001 , mnras , 326 , 255 conselice , c.j . ,
bershady , m.a . ,
jangren , a. 2000a , apj , 529 , 886 conselice , c.j . ,
bershady , m.a . ,
gallagher , j.s .
2000b , a&a , 354 , 21l conselice , c.j . , gallagher , j.s . ,
calzetti , d. , homeier , n. , kinney , a. 2000c , aj , 119 , 79 conselice , c.j .
2003 , apjs , 147 , 1 conselice , c.j . ,
bershady , m.a . ,
dickinson , m. , papovich , c. 2003a , aj , 126 , 1183 conselice , c.j . ,
chapman , s.c . ,
windhorst , r.a .
2003b , apj , 596 , 5l conselice , c.j . , gallagher , j.s . ,
wyse , r.f.g .
2002 , aj , 123 , 2246 conselice , c.j . ,
2004 , apj , 600 , 139l conselice , c.j . ,
blackburne , j. , papovich , c. 2005a , apj , 620 , 564 conselice , c.j .
, bundy , k. , ellis , r. , brichmann , j. , vogt , n. , phillips , a. 2005b , apj , 628 , 160 conselice , c.j .
2006a , mnras , 373 , 1389 conselice , c.j .
2006b , apj , 638 , 686 conselice , c.j .
, et al . 2007a , apj , 660 , 55l conselice , c.j .
, et al . 2007b , mnras , 381 , 962 conselice , c.j . ,
rajgor , s. , myers , r. 2008a , mnras , 386 , 909 ( paper i ) conselice , c.j . ,
bundy , k. , u. , v. , eisenhardt , p. , lotz , j. , newman , j. 2008b , mnras , 383 , 1366 davis , m. , et al .
2003 , spie , 4834 , 161 davis , m. , et al .
2007 , apj , 660 , 1l de propris , r. , conselice , c.j . , driver , s.p . , liske , j. , pattton , d. , graham , a. , allen , p. 2007
, apj , 666 , 212 de ravel , l. , et al .
2008 , arxiv:0807.2578 ellison , s.l . ,
patton , d.r . , simard , l. , mcconnachie , a.w .
2008 , aj , 135 , 1877 giavalisco , m. , et al .
2004 , apj , 600 , 93l hernandez - toledo , h.m .
, avila - reese , v. , conselice , c.j . , puerari , i. 2005 , aj , 129 , 682 holmberg , e. 1941 , apj , 94 , 385 kampczyk , p. , et al .
2007 , 172 , 329 kassin , s. , et al .
2007 , apj , 660 , 35l kawata , d. , mulchaey , j.s . ,
gibson , b.k .
, sanchez - blazquez , p. 2006 , apj , 648 , 969 kartaltepe , j.s . , et al . 2007 ,
apjs , 172 , 320 lanyon - foster , m.m .
, conselice , c.j . ,
merrifield , m.r .
2007 , 380 , 571 le fevre , o. , et al .
2000 , mnras , 3111 , 565 lin , l. , et al .
2004 , apj , 617 , 9l lin , l. , et al .
2008 , apj , 681 , 232 lotz , j.m . , et al .
2008a , apj , 672 , 177l lotz , j.m . , et al .
2008b , arxiv:0805.1246 maraston , c. 2005 , mnras , 362 , 799 mobasher , b. , et al .
2007 , apjs , 172 , 117 papovich , c. , dickinson , m. , giavalisco , m. , conselice , c.j . ,
ferguson , h.c .
2005 , apj , 631 , 101 patton d.r . ,
carlberg , r.g .
, marzke , r.o . , pritchet , c.j .
da costa , l.n . ,
pellegrini , p.s .
2000 , apj , 536 , 153 patton d.r . ,
2002 , apj , 565 , 208 ravindranath , s. , et al .
2004 , apj , 604 , 9l ravindranath , s. , et al .
2006 , apj , 652 , 963 ryan , r.e . ,
jr . , cohen , s.h .
, windhorst , r.a . ,
silk , j. 2008 , apj , 678 , 751 stanford , s.a . ,
dickinson , m.e . ,
postman , m. , ferguson , h.c . ,
lucas , r.a .
, conselice , c.j . ,
budavari , t. , somerville , r. 2004 , aj , 127 , 131 taylor - mager , v. , conselice , c. , windhorst , r. , jansen , r. 2007 , apj , 659 , 162 vorontsov - velyaminov , b.a .
1959 , `` atlas and catalog of interacting galaxies '' , 1959 , sternberg institute , moscow : moscow state university windhorst , r. , et al .
2002 , apjs , 143 , 113 zamojski , m.a . ,
2007 , apjs , 172 , 468 zepf , s.e . ,
koo , d.c .
1989 , apj , 337 , 34 | utilizing deep hubble space telescope imaging from the two largest field galaxy surveys , the extended groth strip ( egs ) and the cosmos survey , we examine the structural properties , and derive the merger history for @xmath0 galaxies with m@xmath1 at @xmath2 .
we examine the structural cas parameters of these galaxies , deriving merger fractions , at @xmath3 , based on the asymmetry and clumpiness values of these systems .
we find that the merger fraction between @xmath4 and @xmath5 increases from roughly @xmath6 to @xmath7 .
we furthermore detect , at a high significance , an abrupt drop in the merger fraction at @xmath8 , which appears relatively constant from @xmath9 to @xmath5 .
we explore several fitting formalisms for parameterising the merger fraction , and compare our results to other structural studies and pair methods within the deep2 , vvds , and cosmos fields .
we also examine the basic features of these galaxies , including our selection for mergers , and the inherent error budget and systematics associated with finding mergers through structure .
we find that for galaxies selected by m@xmath1 , the merger fraction can be parameterised by @xmath10 with the power - law slope @xmath11 . by using the best available @xmath12 prior the slope increases to @xmath13 , showing how critical the measurement of local merger properties are for deriving the evolution of the merger fraction .
we furthermore show that the merger fraction derived through structure is roughly a factor of 3 - 6 higher than pair fractions .
based on the latest cosmological simulations of mergers we show that this ratio is predicted , and that both methods are likely tracing the merger fraction and rate properly .
we calculate , utilising merger time scales from simulations , and previously published merger fractions within the hubble deep and ultra deep fields , that the merger rate of galaxies with m@xmath1 increases linearly between @xmath9 and @xmath14 .
finally , we show that a typical galaxy with a stellar mass of m@xmath1 undergoes between 1 - 2 major mergers at @xmath2 .
[ firstpage ] galaxies : evolution , formation , structure , morphology , classification |
heavy quarkonia , either @xmath10 or @xmath11 , have provided good laboratories for the study of the strong interaction .
new , large data samples at cleo / cesr and bes / bepc have renewed the interest in heavy quarkonia@xcite .
although copiously produced in electric dipole ( e1 ) transitions from the @xmath2 and @xmath12 , the @xmath13 ( @xmath14 ) and @xmath0 ( @xmath15 ) are largely unexplored .
the dominant hadronic transitions among the heavy quarkonia involve di - pion emission , characterized by yan@xcite as the emission of two soft gluons which then hadronize as a di - pion system .
these have been studied for transitions among the quarkonia @xmath16 states , but have not been observed in other quarkonia transitions such as @xmath17 or the @xmath13 decays , which are the subject of this work .
new interest in @xmath13 decays has also been generated by the cleo observation@xcite of a large branching fraction for the decay @xmath18 .
this is the only presently known hadronic decay of the @xmath19-wave @xmath20 states and the only hadronic bottomonium transition that is not through @xmath21 .
we have investigated another hadronic transition of the @xmath13 , namely @xmath22 . as shown in fig .
[ fig : signal_and_ts ] , this search starts with the e1 transition @xmath23 , followed by the signal process @xmath22 , and the resulting @xmath0 decay ( again via an e1 transition ) as @xmath24 with @xmath5 .
thus the final state has two photons , two low - momentum ( `` soft '' ) pions and two high - momentum leptons . in this article
we ( i ) establish this @xmath13 decay and , with reasonable assumptions , ( ii ) estimate the partial width @xmath25(2p)@xmath1(1p)@xmath26 . the main background to our signal , also shown in fig . [ fig : signal_and_ts ] , has @xmath27 , followed by an e1 cascade through the @xmath0 states to the @xmath28 .
this background process , which we will denote as `` @xmath29 '' , has the same number of pions , leptons and photons , with similar kinematics .
while this means we need stringent selection criteria to define the signal , it also provides a known process with a nearly identical final state against which to test our analysis procedures .
the data were collected at the cornell electron storage ring using the cleo iii@xcite detector configuration .
the components most critical for this analysis were the cs - i electromagnetic calorimeter and the charged particle tracking system , each covering @xmath3093% of the @xmath31 solid angle .
consisting of 7800 crystals , the calorimeter was originally installed in the cleo ii configuration@xcite , with some reshaping and restacking for cleo iii to allow more complete solid angle coverage .
the shower energy resolution , @xmath32 , is 4% at 100 mev and 2% at 1 gev in the barrel region , defined as @xmath33 , with @xmath34 the dip angle with respect to the beam axis .
complemented at small radius by a 4-layer double - sided silicon vertex detector , a new drift chamber@xcite was installed for cleo iii ; its endplate design minimizes material , enhancing the resolution of the endcap electromagnetic calorimeter , extending the solid angle coverage to @xmath35 .
the signal was searched for in 1.39 fb@xmath36 of data accumulated at the center of mass energy corresponding to the @xmath2 resonance , consisting of @xmath37 resonance decays@xcite .
we also used 8.6 fb@xmath36 of data taken at @xmath38 gev ( `` high - energy continuum '' ) and 0.78 fb@xmath36 of data taken at the @xmath12 , or roughly @xmath39 decays@xcite , to study and evaluate backgrounds .
we used large monte carlo simulations based on geant3.211/11@xcite to estimate our efficiencies and tune our selection criteria .
in addition to the signal process , we simulated : ( i ) the main background process , `` @xmath29 '' , as described above and in fig .
[ fig : signal_and_ts ] ; ( ii ) @xmath27 with @xmath40 , a process with higher statistics and similar pion kinematics , to help confirm our efficiency determinations ; ( iii ) @xmath27 with @xmath41 , which could mimic our signal multiplicity if two photons were missed ; ( iv ) `` generic '' @xmath2 monte carlo , which uses all known properties and modes of @xmath2 decay , but for which the backgrounds ( i ) through ( iii ) are tagged and not analyzed ; ( v ) @xmath42 continuum processes at the @xmath2 center of mass energy ; ( vii ) for the charged pion decay channel , @xmath43 with @xmath18 and @xmath44 , which has the same initial photon transition as our signal but would have an additional photon in the decay of the @xmath45 resulting from the @xmath46 decay to @xmath47 ; and ( viii ) for the neutral pion decay channel , @xmath48 with @xmath49 ; we used @xmath50 , which is the present 90% cl upper limit for this decay . in our studies we assumed that there were no @xmath51-wave contributions to the decays , only @xmath52-wave , so that @xmath53 .
this assumption is supported by : ( a ) the maximum available energy , @xmath54 , for the nine possible decays is @xmath55 = 130 mev , making it difficult to have the extra kinetic energy associated with two units of angular momentum ; ( b ) previous studies of @xmath2@xmath56@xcite and @xmath57@xcite , systems with substantially more @xmath54 , indicate no angular momentum between the final state onium and the dipion system , although the former result is also consistent@xcite with a few percent of @xmath51-wave ; and ( c ) , the average ( weighted by the observed distribution of di - pion invariant mass , @xmath58 ) of the @xmath51-wave between the two pions in @xmath59@xcite is less than 10% . as shown in table [ tab : allowed ] , the entry and exit branching fractions@xcite strongly disfavor our observation of @xmath60 .
we also had to discriminate against this possible mode in order to suppress our dominant background source , `` @xmath29 '' , in that there is overlap in the energies of the e1 transition photon for the @xmath60 signal process and that of the dominant @xmath61 mode of that background .
therefore , we assumed that the transitions with @xmath62 or @xmath63 dominate . to estimate the relative abundance of these two transitions and , later , to calculate the partial width @xmath64 , we needed the _ full _ widths @xmath65 and @xmath66 .
we calculated these using the theoretical e1 partial widths for these two states@xcite and their experimental e1 branching fractions@xcite to @xmath67(1s ) and @xmath67(2s ) , where in the latter we took into account the new cleo iii value@xcite of @xmath68 .
our results , also listed in table [ tab : allowed ] , are @xmath69 kev and @xmath70 kev . given @xmath71 from this table we expected the @xmath62 transition to dominate @xmath72 by roughly a factor of 2.3 .
.the three di - pion transitions between @xmath73 and @xmath0 that leave the orbital angular momentum unchanged ( @xmath52-wave ) .
the third column is the mass difference .
columns four and five are the branching fractions for the entrance and exit e1 transitions : @xmath74(3s)@xmath75 and @xmath76(1s ) ) .
the e1 transition from @xmath77 is unobserved , with a limit of 6% on its branching fraction at 90% c.l .. [ cols="^,^,^,^,^,^,^ " , ] the individual contributions to eqn .
[ eqn : defining ] are shown in table [ tab : contributions ] , along with the value of @xmath64 obtained , its statistical uncertainty , and its individual ( cleo - based ) systematic uncertainty . taking the _ statistical _ average of the three gives @xmath78 } = ( 0.84 \pm 0.22)$ ] kev .
a more complete average takes into account the individual systematic uncertainties ; this weighted average is @xmath79 kev . following the pdg@xcite prescription for evaluating the consistency of results
being averaged , we obtain an uncertainty scale factor of 1.09 , which is close to unity .
de - convolving the statistical uncertainty and adding in a separate term for the 22% `` common '' uncertainties that are not based on measurements in this analysis yields our final result of @xmath80 with the uncertainties being statistical , systematics from our analyses , and systematics from outside sources .
this result for @xmath22 can be compared to values derived from the pdg@xcite of @xmath81 ) kev for a process with somewhat less @xmath54 and @xmath82 kev for a process with considerably more @xmath54 .
our result is consistent with the theoretical expectations of kuang and yan@xcite , who have calculated @xmath83 kev . in summary , we have searched the cleo iii data at the @xmath2 resonance for the decay @xmath22 using four different approaches .
the combined probability that the signal process is absent is small , leading to the conclusion that the null hypothesis can not be substantiated . under the assumption of no @xmath51-wave contributions
we obtain a partial width for each of the @xmath62 and @xmath72 transitions of @xmath84 .
we gratefully acknowledge the effort of the cesr staff in providing us with excellent luminosity and running conditions .
this work was supported by the national science foundation and the u.s .
department of energy . | we have searched for the di - pion transition @xmath0(2p)@xmath1(1p ) in the cleo iii sample of @xmath2 decays in the exclusive decay chain : @xmath3(2p ) , @xmath0(2p)@xmath1(1p ) , @xmath0(1p)@xmath4(1s ) , @xmath5 .
our studies include both @xmath6 and @xmath7 , each analyzed both in fully reconstructed events and in events with one pion undetected .
we show that the null hypothesis is not substantiated . under reasonable assumptions ,
we find the partial decay width to be @xmath8(2p)@xmath1(1p)@xmath9 with the uncertainties being statistical , internal cleo systematics , and common systematics from outside sources . |
pr - based intermetallic compounds have evolved as a topic of current interest among the condensed matter physicists as some of these compounds exhibit interesting physical properties .
while in ce - compounds the relative strengths of the rkky and kondo interactions decide the ground state properties , in case of pr - compounds the ground state depends critically on the crystal electric field ( cef ) level scheme .
for example , in pros@xmath2sb@xmath3 a small cef splitting energy of 0.7 mev and quadrupolar excitations lead to unconventional heavy - fermion superconductivity [ 1 - 3 ] .
pr@xmath0rh@xmath4ge@xmath5 exhibits heavy fermion behaviour in which low lying crystal field excitations are responsible for the mass enhancement instead of the the usual kondo effect @xcite .
we have also investigated prrh@xmath0si@xmath0 in view of the unusual superconducting and magnetic properties of cerh@xmath0si@xmath0 and ybrh@xmath0si@xmath0 .
we found an antiferromagnetic ordering at 68 k in prrh@xmath0si@xmath0 which is anomalously high compared to the expected de - gennes - scaled transition temperature of 5.4 k @xcite .
further , we decided to investigate prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 in view of the interesting features of cepd@xmath0si@xmath0 and cept@xmath0si@xmath0 .
while cepd@xmath0si@xmath0 is a heavy - fermion antiferromagnet system which exhibits pressure induced superconductivity @xcite , cept@xmath0si@xmath0 is a kondo lattice non - fermi liquid system that does not order down to 60 mk @xcite .
a preliminary magnetization study on prpt@xmath0si@xmath0 reports it to be paramagnetic down to 1.8 k @xcite .
we report here our results of magnetization , electrical resistivity , and heat capacity studies of prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 .
polycrystalline samples of prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 and their la - analogs were prepared by standard arc - melting on a water cooled copper hearth under an inert argon atmosphere starting with high purity ( 99.99% and above ) elements in stoichiometric ratio . to ensure a proper mixing of the constituents ,
arc melted ingots were flipped and remelted several times .
weight loss during the melting process was less than 0.5% .
the samples were annealed at 1000 @xmath6c for one week to improve the sample quality .
thereafter , the samples were characterized using powder x - ray diffraction and scanning electron microscopy ( sem ) equipped with energy dispersive x - ray analysis ( edax ) .
magnetization measurements were performed using a commercial squid magnetometer .
the heat capacity was measured using relaxation method in a physical property measurement system ( ppms - quantum design ) .
the electrical resistivity was measured by four probe ac technique using the ac transport option of ppms .
while the compounds prpd@xmath0si@xmath0 , lapd@xmath0si@xmath0 crystallize in thcr@xmath0si@xmath0-type tetragonal structure ( space - group _ i4/mmm _ ) , prpt@xmath0si@xmath0 and lapt@xmath0si@xmath0 form in cabe@xmath0ge@xmath0-type primitive tetragonal structure ( space group _ p4/nmm _ ) .
the lattice parameters and unit cell volumes for these compounds are listed in table 1 .
the lattice parameters obtained for prpt@xmath0si@xmath0 , lapt@xmath0si@xmath0 , and lapd@xmath0si@xmath0 are close to the values reported in references @xcite and @xcite .
though all the peaks in x - ray diffraction pattern of prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 are well indexed , the scanning electron micrographs show the presence of impurity phase(s ) which we estimated to be less than 3% in prpd@xmath0si@xmath0 and about 6% in prpt@xmath0si@xmath0 .
the edax analysis confirms the desired stoichiometry of 1:2:2 .
.[tab : table1 ] lattice parameters and unit cell volumes of tetragonal compounds rt@xmath0si@xmath0 ( r = la , pr and t = pd , pt ) . [ cols="<,^,^,^,^",options="header " , ]
the magnetic susceptibility data of prpd@xmath0si@xmath0 are shown in figure 1 .
magnetic susceptibility follows the curie - weiss behavior above 50 k. fitting the inverse susceptibility data to the expression 1/@xmath7 = ( @xmath8)/@xmath9 , we obtained an effective moment @xmath10 = 3.59 @xmath11 ( theoretically expected value for pr@xmath12 ions is 3.58 @xmath11 ) and a weiss temperature @xmath13 = - 4.2 k. the low temperature susceptibility data ( shown in the inset of figure 1 ) show a peak at 3.5 k. the position of the peak shifts towards lower temperatures with increasing field , thereby confirming the antiferromagnetic nature of the transition at 3.5 k. the isothermal magnetization data at 2 k ( figure 2 ) exhibits a very smooth metamagnetic - type transition at 4.8 t. the critical field for the metamagnetic transition is determined from the @xmath14 _ vs. _ @xmath15 plot ( inset of figure 2 ) .
no saturation is observed upto b = 6.0 t. the magnetic field dependence of magnetization of prpd@xmath0si@xmath0 at 2 and 10 k. the inset shows the derivative of magnetization data at 2 k.,width=377 ] the magnetic contribution to the specific heat of prpd@xmath0si@xmath0 as a function of temperature in the temperature range 0.5
52 k. the solid line represents the fit for equally degenerate two level schottky anomaly with a separation of 50 k. the inset shows the temperature dependence of magnetic entropy.,width=377 ] figure 3 shows the magnetic contribution to the specific heat of prpd@xmath0si@xmath0 which we obtained after subtracting the lattice contribution assuming it to be roughly equal to that of the nonmagnetic analog lapd@xmath0si@xmath0 .
the specific heat data of prpd@xmath0si@xmath0 exhibits a sharp @xmath16-type peak at 3 k which confirms the intrinsic nature of magnetic order in this compound .
the magnetic entropy reaches a value close to @xmath17 (= 11.52 j / mole - k ) at 7.5 k , suggesting a quasi - quartet ground state .
we also observe a pronounced schottky type anomaly with a broad maximum centered around 20 k which we attribute to the crystal field effect .
the position of schottky peak suggests that the excited states lie about 50 k above the low lying cef states .
the peak height of the schottky anomaly is consistent with equal degeneracy between the low lying states and the excited states , suggesting the presence of four cef levels close to 50 k. in a tetragonal symmetry one expects the cef levels of pr@xmath12 to split into five singlets and two doublets .
thus in prpd@xmath0si@xmath0 , the temperature dependence of the entropy indicate a separation into four low lying levels ( either two doublets , or one doublet and two singlets , or four singlets ) separated by less than 10 k , four further levels around 50 k , and an upper singlet at much higher energy . since in rpd@xmath0si@xmath0
the cef schemes seem to be determined by the higher order terms in the cef hamiltonian @xcite , a simple preliminary guess of the cef sheme of prpd@xmath0si@xmath0 can not be given .
temperature dependence of electrical resistivity of prpd@xmath1si@xmath1 in the temperature range 2 300 k. the upper inset shows the low temperature resistivity data at different fields .
the lower inset shows the magnetoresistance normalized to @xmath18 at b = 0.,width=415 ] in the paramagnetic regime the electrical resistivity decreases almost linearly with decreasing temperature ( figure 4 ) and merge into a constant value of 5.5 @xmath19-cm below 10 k. the resistivity drops rapidly below the ordering temperature due to reduction of spin disorder scattering .
the resistivity at 2 k , where @xmath18 is still decreasing with t , is 3.4 @xmath19-cm leading to a lower bound of @xmath20 15 for the residual resistivity ratio .
this evidences a good quality of our polycrystalline sample .
the upper inset of figure 4 shows the effect of a magnetic field on the resistivity .
the resistivity anomaly related to the magnetic order smoothens out for b @xmath21 2 t. the magnetoresistance @xmath22/@xmath23 = [ @xmath23(@xmath15)-@xmath23(0)]/@xmath23(0 ) is shown in the lower inset of figure 4 . in the ordered state ( at 2 k ) the magnetoresistance initially increases with increasing field , peaks at 6 t and decreases for higher fields .
such a behavior of the magnetoresistance is expected for an antiferromagnetic state . for t @xmath24 t@xmath25
, an increasing magnetic field is first weakening the af - state , leading to an increase of spin scattering , but above the metamagnetic transition the formation of the field aligned state results in a decrease of spin scattering . in the paramagnetic state
the magnetoresistane is always negative , since alignment of the moments in the external field reduces the spin scattering .
thus , magnetic susceptibility , magnetization , specific heat and magnetoresistance data provide conclusive evidence for an antiferromagnetic state in prpd@xmath1si@xmath1 below 3.0 k.
figure 5 shows the magnetic susceptibility data of prpt@xmath0si@xmath0 .
no anomaly is observed in susceptibility data down to 2 k implying the absence of magnetic ordering in this compound . the paramagnetic susceptibility displays a curie - weiss character . from a fit of the inverse susceptibility data above 50 k to the expression 1/@xmath7
= ( @xmath8)/@xmath9 we found the effective moment to be @xmath10 = 3.46 @xmath11 which is very close to the value of 3.58 @xmath11 expected for pr@xmath12 ions .
the curie - weiss temperature @xmath13 = + 18.8 k indicates dominant ferromagnetic exchange in this compound .
the isothermal magnetization curve is linear with field at 2 and 20 k which is consistent with a paramagnetic state down to 2 k in this compound .
inverse magnetic susceptibility plot of prpt@xmath0si@xmath0 at a field of 1.0 t. the inset shows magnetization as a function of field at two different temperatures of 2 and 20 k.,width=377 ] temperature dependence of magnetic contribution to the specific heat of prpt@xmath0si@xmath0 in the temperature range ( 2 50 k ) .
the inset shows the temperature dependence of magnetic entropy.,width=377 ] the magnetic contribution to the specific heat of prpt@xmath0si@xmath0 ( figure 6 ) was obtained by subtracting the specific heat data of lapt@xmath0si@xmath0 from that of prpt@xmath0si@xmath0 .
no signature of magnetic or superconducting transition is observed in the specific heat data , which is consistent with the paramagnetic ground state inferred from the magnetic susceptibility data . from the low temperature specific heat data we estimate a @xmath26 value of @xmath20 3 mj / mole - k@xmath27 .
such a low magnitude of specific heat at low t rules out the possibility of magnetic ordering even at lower temperature .
a broad schottky type anomaly is observed in the magnetic part of the specific heat above 20 k. the upturn in the magnetic contribution ( c@xmath28 ) to the specific heat around 10 k could be reproduced with a first excited doublet separated from the singlet ground state by approximately 50 k. for the case of a singlet ground state and a doublet as first excited state one would expect a peak height of 6.2 j / mole - k in c@xmath28 .
however , the experimentally observed value is 9.2 j / mole - k , clearly indicating that further higher lying states are also contributing .
the large separation between the singlet ground state and the first excited state is responsible for the absence of magnetic order in prpt@xmath0si@xmath0 .
usually the cef schemes of homologous rt@xmath1x@xmath1 compounds with different t- ( or x- ) elements of one row in the periodic table are quite similar , because neither the effective ionic charge at the ligands nor their distance to the r - atoms changes significantly .
thus , in the present case , the strong differences between the cef schemes of prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 are very likely related to the loss of a mirror plane at the pr - site upon changing the crystal structure from the thcr@xmath1si@xmath1 to the cabe@xmath1ge@xmath1 type , although both structure types are closely related .
this difference in the structures is likely also responsible for the change from dominant antiferromagnetic exchange in the pd - based compound to dominant ferromagnetic exchange in the pt - based compound .
we have investigated magnetic properties of prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 . from a detailed measurements of magnetization , specific heat , electrical resistivity and magnetoresistance we have established antiferromagnetic ordering in prpd@xmath0si@xmath0 below 3 k. below t@xmath25 this compound also exhibits field induced metamagnetic transition and large magnetoresistance ( 20% ) at a magnetic field of 6 tesla . in contrast ,
no magnetic order is observed in prpt@xmath0si@xmath0 down to 2 k. we attribute this paramagnetic ground state in prpt@xmath0si@xmath0 to the cef scheme with a singlet ground state and well separated first excited state located around 50 k. the pronounced differences between prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 are attributed to the loss of a mirror plane upon changing the crystal structure from the thcr@xmath1si@xmath1 to the cabe@xmath1ge@xmath1 type .
further investigations are needed to determine the exact magnetic structure of prpd@xmath0si@xmath0 and the crystal field level schemes in these two compounds . | we have investigated the two rare - earth intermetallic compounds prpd@xmath0si@xmath0 and prpt@xmath0si@xmath0 by means of magnetization , electrical resistivity and heat capacity measurements . while prpd@xmath0si@xmath0 exhibits an antiferromagnetic ordering at 3 k , no magnetic ordering is observed in prpt@xmath0si@xmath0 down to 2 k. the different magnetic behaviors of these two compounnds are due to different crystalline electric field ( cef ) level schemes .
the specific heat data suggest a quasi - quartet ground state in prpd@xmath1si@xmath1 in contrast to a nonmagnetic singlet ground state in prpt@xmath1si@xmath1 .
this difference is attributed to the loss of a mirror plane upon changing the crystal structure from the thcr@xmath1si@xmath1 type ( prpd@xmath1si@xmath1 ) to the cabe@xmath1ge@xmath1 type ( prpt@xmath1si@xmath1 ) .
further on , a large magnetoresistance is also observed in the magnetically ordered state of prpd@xmath1si@xmath1 . |
every living organism obtains the energy needed for its survival from the outside world .
this energy can be in the form of sunlight or food ; but in both cases it is unstable and can not be utilized directly , so several energy - conversion steps are necessary .
one of the most widely used intermediate forms for energy storage is the electrochemical proton gradient across lipid membranes , such as the inner mitochondrial membranes or plasma membranes in bacteria . to achieve and maintain this proton gradient
, nature employs several different types of electron- or light - driven systems , where the energy of high - energetic electrons or absorbed photons is used for the energetically - uphill proton transfer from the negative ( @xmath0 ) to the positive ( @xmath1 ) sides of the membrane .
here we discuss two mechanisms of energy conversion from the highly unstable electronic form of energy to the proton gradient , namely , proton pumps and redox loops @xcite .
both mechanisms rely on the electrostatic interaction between electrons and protons , although the specific details of the proton pumps and the redox loops look very different .
for example , in a proton pump , such as _ cytochrome c oxidase _ , electrons move mainly along the membrane , whereas protons move across the membrane , which results in an accumulation of the positive charge on the @xmath1-side and in the generation of a proton - motive force ( pmf ) @xcite . in the redox - loop mechanism of pmf generation , taking place in the nitrate respiratory chain of _
e. coli _ bacterium , the _ neutral shuttle _ , carrying _ both _ protons and electrons , crosses the membrane .
here , the charge accumulation occurs when electrons cross the membrane , just before embarking on the shuttle , and right after unloading from the shuttle @xcite .
it should be noted that the proton pump operating in the _ cytochrome c oxidase _ has no essential mechanically - moving parts , whereas the redox - loop mechanism is impossible without the molecular shuttle diffusing between the negative and the positive sides of the lipid membrane . in general ,
the treatment of the electron and proton transfer events is extremely difficult because the total number of the occupation states increases exponentially with the number of the electron- and proton - binding sites , when all of them are electrostatically coupled . in the present work , however , we show that both above - mentioned mechanisms of the transmembrane proton translocation can be described with a similar mathematical model , taking into account the coulomb interaction between _ one _ electron- and _ one _ proton - binding sites only , and neglecting electrostatic couplings to other sites .
it is necessary to have at least three redox sites and three proton - binding sites in order to obtain a proton pumping effect and suppress a reverse flow of protons from the @xmath1-side to the @xmath0-side of the membrane . in the absence of strong coulomb interaction between all sites
, there is no need to introduce a complete set of electron and proton occupation states ( as was done in our previous works , refs .
@xcite ) , which grows exponentially with the number of sites . instead
, we now divide the whole system into clusters of strongly coupled sites .
these clusters are described by their own set of occupation states , and the total number of the states in the system is equal to the sum ( not the product ! ) of the states in the clusters .
the clusters are weakly coupled by electron tunneling terms and by proton transfer amplitudes , so that transitions between the clusters can be considered within perturbation theory .
while in this work we present quite simple models , similar approaches can be applied to much more complicated biological systems , such as photosystem ii and the whole respiratory chain in the inner mitochondrial membrane @xcite .
the quantum yield for the two models analyzed in this paper can be about 1 .
why such a high quantum yield ?
this can be explained from the fact that , in order to be transferred through the system , an electron needs to loose its energy .
this can not be done via the environment because the reorganization energy is not large enough .
consequently , electron transport occurs with the assistance of protons gaining this energy and being transferred to the positive side of the membrane .
thus , the transfer of a single electron is accompanied by the transfer of a single proton and the corresponding currents are equal , which results in an almost perfect quantum yield .
we consider a physical model describing an electron - coupled translocation of protons from the negative ( @xmath0 ) to the positive ( @xmath1 ) side of a membrane .
the model consists of an interaction site , @xmath2 , containing a single electron level with energy @xmath3 and a single proton energy level characterized by the energy @xmath4 .
we also introduce two electron sites , @xmath5 and @xmath6 , coupled to the electron site @xmath7 , and two proton sites , @xmath8 and @xmath9 , coupled to the proton site @xmath10 ( fig . 1 ) .
the electron site @xmath5 is coupled to the electron source @xmath11 , and the site @xmath6 is connected to the electron drain @xmath12 .
the proton site @xmath8 is coupled to the proton reservoir @xmath0 ( the negative side of the membrane ) , and the site @xmath9 is coupled to the positive side of the membrane ( proton reservoir @xmath1 ) .
the coulomb interaction between an electron and a proton , both located on the central site @xmath13 , is described by the energy @xmath14 , so that the hamiltonian of the site @xmath13 has the form @xmath15 where @xmath16 is the electron population of the site q , and @xmath17 is the proton population of this site .
electrons are described by the fermi - operators @xmath18 , and protons are characterized by the fermi - operators @xmath19 with @xmath20 and @xmath21 and with the corresponding populations @xmath22 the contribution of the electron sites @xmath23 and the proton sites @xmath24 to the total hamiltonian of the system is described by the term @xmath25 where @xmath26 are the energy levels of the electron sites @xmath5 and @xmath6 , and @xmath27 are the energies of the proton - binding sites @xmath8 and @xmath9 .
the strongly - interacting electron and proton sites @xmath7 and @xmath10 form a single ( interaction ) cluster , whereas the sites @xmath23 and @xmath24 separately form other four ( peripheral ) clusters .
the cluster @xmath13 can be characterized by the vacuum ( empty ) state and by three additional occupation states , or , equivalently , by the average electron and proton populations , @xmath28 and @xmath29 , complemented by the correlation function , @xmath30 the other electron and proton clusters are described by the corresponding average occupations , @xmath31 and @xmath32 . for six
electron and proton - binding sites we should have @xmath33 occupation states .
however , with the cluster approach , the system can be completely described by only seven functions : @xmath34 ( for the interaction cluster ) , and @xmath35 ( for the peripheral clusters ) .
previously , we applied a similar approach to analyze quantum transport problems in nanomechanical systems @xcite .
the electron tunneling hamiltonian between the site @xmath13 and the sites @xmath5 and @xmath6 is given by @xmath36 whereas the @xmath8-@xmath13 and @xmath9-@xmath13 proton transitions are described by the term @xmath37 here @xmath38 are the electron tunneling coefficients , and @xmath39 are the proton transfer amplitudes . in the case of a movable interaction site , e.g. ,
when the electron and proton sites @xmath13 are located on the shuttle ( quinone / quinol ) , the amplitudes @xmath40 and @xmath39 depend on the position @xmath41 of the shuttle .
the @xmath11-lead serves as a source of electrons , and the @xmath12-lead works as an electron drain .
the coupling to these leads is characterized by the hamiltonian @xmath42 the proton transitions between the @xmath0-side of the membrane and the site @xmath8 and between the @xmath1-side of the membrane and the site @xmath9 are described by the hamiltonian @xmath43 here @xmath44 are fermi operators of the electron reservoirs @xmath11 and @xmath12 , and @xmath45 are the fermi operators of protons in the reservoirs @xmath0 and @xmath1 .
the electron reservoirs @xmath11 and @xmath12 have the hamiltonian @xmath46 and are characterized by the fermi distributions @xmath47 with the corresponding electrochemical potentials @xmath48 and @xmath49 . for the proton reservoirs @xmath0 and @xmath1 we have the hamiltonian @xmath50 with the fermi distributions @xmath51 and @xmath52 and the proton electrochemical potentials @xmath53 and @xmath54 .
the interaction of the electron - proton system with the protein environment , which is described as a sum of independent oscillators @xcite , is characterized by the hamiltonian @xmath55 where @xmath56 is the population of the electron site @xmath57 ( @xmath58 ) , @xmath59 is the population of the proton site @xmath60 ( @xmath61 ) .
the constants @xmath62 determine the electron coupling to the environment , and the parameters @xmath63 describe the proton - environment interaction . with the unitary transformation , @xmath64 , \label{ut}\end{aligned}\ ] ] the environment hamiltonian can be rewritten as @xmath65 whereas the hamiltonians @xmath66 and @xmath67 acquire the stochastic phase factors : @xmath68 and @xmath69 with the phases @xmath70 @xmath71 and @xmath72 @xmath73 for simplicity , we assume that there are no phase shifts for the electron transitions between the electron source @xmath11 and the site @xmath5 , and the electron drain @xmath12 and the site @xmath6 , so that @xmath74 and @xmath75 , with the same assumption for the @xmath0-@xmath8 and @xmath1-@xmath9 proton transitions , @xmath76 and @xmath77 the time evolution of the electron operators @xmath78 is determined by the heisenberg equations : @xmath79 and @xmath80 for the proton populations @xmath81 , we derive the similar set of heisenberg equations , @xmath82 this set should be complemented by the equation for the proton population of the interaction site , @xmath83 as well as by the equations for the operators of electron and proton reservoirs , @xmath84 @xmath85 it follows from eq .
( [ csd ] ) that the electron operator @xmath86 can be represented as @xmath87_{+}\rangle \ , a_l ( t_1 ) \,\theta(t - t_1 ) , \label{cs}\ ] ] where @xmath88 is the free variable of the @xmath11-lead , and @xmath89 is the heaviside step function .
similar expressions take place for the electron operator @xmath90 and for operators @xmath45 of the proton reservoirs . for the weak coupling between the reservoir @xmath11 and the electron site @xmath5
we obtain @xmath91_{+}\rangle \,\theta(t - t_1 ) .
\label{alcs}\ ] ] thus , contribution of the @xmath11-lead to the evolution of the average electron population @xmath92 ( see eq .
( [ nlrh ] ) ) is determined by the expression @xmath93 the correlator @xmath94 is proportional to the fermi distribution function , @xmath95 of electrons in the reservoir @xmath11 , @xmath96 where the fermi function , @xmath97^{-1},\ ] ] is characterized by the electrochemical potential @xmath48 and temperature @xmath98 .
we assume that the site @xmath5 is weakly - coupled to the reservoir @xmath11 and to the site @xmath13 , thus , we can use free - evolving operators , @xmath99 to calculate the corresponding correlation functions in eq .
( [ alcs2 ] ) , e.g. , @xmath100 introducing the energy - independent rate constant , @xmath101 [ gammas ] we calculate the contribution of the @xmath11-lead to the time evolution of the population @xmath92 , @xmath102 .
\label{alcs3}\ ] ] the same analysis can be applied for a calculation of contributions of the electron lead @xmath12 and the proton leads @xmath0 and @xmath1 to the corresponding populations @xmath103 and @xmath104 the proton transfer rates between the sites @xmath8 and @xmath105 and the negative and positive sides of the membrane , respectively , are determined by the coefficients @xmath106 and @xmath107 where , e.g. , @xmath108 to calculate a contribution of the @xmath5-@xmath13 tunneling to the evolution of the populations @xmath92 and @xmath109 , we start with the amplitude @xmath110 , which obeys the equation @xmath111 in the case of weak @xmath5-@xmath13 and @xmath6-@xmath13 tunnel couplings , the formal solution of eq .
( [ aq1 ] ) can be written in the form @xmath112_{+}\rangle \,\ { \delta_l^ * \,e^{-i\xi_{l}(t_1)}\ , a_l(t_1 ) + \delta_r^*\ , e^{-i\xi_{r}(t_1 ) } \,a_r(t_1)\ } , \label{aq2}\end{aligned}\ ] ] where @xmath113 is the free operator of the site @xmath13 , obeying the equation ( [ aq1 ] ) with the tunneling terms neglected ( @xmath114 .
taking into account the formula , @xmath115_{+}\rangle\ , \theta(t - t_1 ) , \label{alaq1}\end{aligned}\ ] ] which is similar to eq .
( [ alcs ] ) , we obtain @xmath116 dropping the label @xmath117 , we assume that the time evolution of the operators @xmath110 in eq .
( [ alaq2 ] ) is calculated with the free - evolution formula , @xmath118 \,n_q(t_1)\ , a_q(t_1 ) .
\label{aq3}\ ] ] for the free - evolving proton operator of the interaction site we obtain a similar expression @xmath119 \,n_q(t_1)\ , b_q(t_1 ) .
\label{bq1}\ ] ] the influence of the environment on the electron tunneling between the sites @xmath5 and @xmath13 , and between the sites @xmath6 and @xmath13 , is determined by the correlators @xmath120 and @xmath121 , where @xmath122 the reorganization energy , @xmath123 , is defined as @xcite @xmath124 the electron reorganization energy @xmath125 , and the proton reorganization energies @xmath126 and @xmath127 are defined in a similar way .
in particular , @xmath128 consequently , we derive the system of rate equations for the average populations of the electron sites , @xmath129 and for the average populations of the proton - binding sites , @xmath130 here @xmath131 ( @xmath132 ) and @xmath133 ( @xmath134 ) are the functions of the average electron and proton populations , respectively . in addition , due to a strong electron - proton coulomb interaction on the site @xmath13 ,
the kinetic terms @xmath131 and @xmath133 depend on the correlation function , @xmath135 of the electron and proton populations on the site q , @xmath136 where @xmath137 is the marcus rate for electron transfer between the site @xmath57 and the interaction site @xmath13 , @xmath138 the proton term @xmath133 is determined by the expression , similar to eq .
( [ phi1 ] ) , as @xmath139 where @xmath140 is the proton marcus rate for the transitions between the site @xmath60 and the proton - binding site @xmath13 , @xmath141 for the correlator , @xmath142 , of the electron ( @xmath143 ) and proton ( @xmath144 ) populations of the interaction site , we derive the following equation @xmath145 where @xmath146 electron currents @xmath147 and proton currents @xmath148 are determined by an increase of the number of particles , electrons or protons , in the corresponding reservoir .
in particular , a variation of the electron number in the drain lead gives a current @xmath149 , \label{id}\ ] ] whereas the proton current @xmath150 is given by @xmath151 .
\label{ip}\ ] ] here , @xmath152 are the electron ( @xmath153 ) and proton ( @xmath107 ) transfer rates between the electron site @xmath6 and the lead @xmath12 , and between the proton - binding site @xmath9 and the @xmath1-side of the membrane , respectively .
the multiplications of the particle currents introduced above by the electron or proton charges produce the standard electric currents .
it follows from eqs .
( [ nlrq],[nabq ] ) that , in the steady - state , we have the relations : @xmath154 so that @xmath155 @xmath156 and @xmath157 @xmath158 the productivity of the proton pump is determined by a quantum yield , @xmath159 and by the power - conversion efficiency @xmath160 , @xmath161 at the standard conditions , we have @xmath162 and @xmath163 therefore , @xmath164 if a quantum yield @xmath165 is of order one ( or 100% ) , the power - conversion efficiency @xmath160 may be as much as 0.35 ( or 35% ) . for the redox - loop mechanism of a proton translocation through the membrane , the electron and proton sites , labelled by the letter @xmath13 ,
are attached to the shuttle : a molecule diffusing between the @xmath0 and @xmath1 sides of the membrane ( see fig . 2 ) .
this brownian motion can be described by the one - dimensional overdamped langevin equation for the coordinate @xmath41 of the shuttle , @xmath166 we assume that the shuttle molecule moves along a line connecting the sites @xmath5 and @xmath8 , located at @xmath167 , and the sites @xmath6 and @xmath9 , both having the coordinate @xmath168 .
the borders of the membrane , at @xmath169 are schematically shown in fig . 2 .
( [ langevin ] ) , @xmath170 is the drag coefficient of the shuttle , and @xmath171 is the gaussian fluctuation force , which is characterized by the zero - mean value , @xmath172 , and the correlation function , @xmath173 proportional to the temperature @xmath98 of the environment .
the diffusion coefficient @xmath12 of the shuttle is also proportional to the temperature : @xmath174 the motion of the shuttle is restricted by the membrane walls , which are simulated by the confinement potential @xmath175 , @xmath176^{-1 } + \left [ \exp\left ( \frac{x + x_c}{l_c } \right ) + 1 \right]^{-1 } \right\ } , \label{uc}\ ] ] having the barrier height @xmath177 , the width @xmath178 and the steepness @xmath179 .
the potential barrier @xmath180 , @xmath181^{-1 } - \left [ \exp\left ( \frac{x + x_s}{l_s } \right ) + 1\right]^{-1 } \right\ } , \label{us}\ ] ] does not allow the shuttle with a non - zero charge @xmath182 ( in units of @xmath183 ) to cross the lipid interior of the membrane .
this barrier is determined by the height @xmath184 , the steepness @xmath185 , and the width @xmath186 .
we solve the rate equations ( [ nlrq],[nabq ] ) for the electron ( @xmath78 ) and proton ( @xmath81 ) populations jointly with the equation ( [ k1 ] ) for the electron - proton correlation function on the site @xmath13 , @xmath187 .
our approach can describe two mechanisms of the redox - linked proton translocation across the membrane : ( i ) the static interaction site @xmath13 and ( ii ) the situation when the site @xmath13 diffuses between the sides of the membrane .
the mechanism ( i ) roughly corresponds to the proton pump operating in _ cytochrome c oxidase _ ( cco ) @xcite , whereas the design ( ii ) can be attributed to the redox loop mechanism , which is responsible for electron and proton transfers in the inner membrane of bacteria @xcite . here , we consider the mechanism ( i ) , where the interaction site @xmath13 does not change its position ( see fig
we assume that protons are transferred across the membrane , from the negatively charged side @xmath0 , with an electrochemical potential @xmath53 , to the positively charged side @xmath1 , having an electrochemical potential @xmath54 .
all potentials and energies are measured in mev .
the difference of electrochemical potentials , @xmath188 is determined by the following expression @xmath189 where @xmath190 is the transmembrane voltage , @xmath6 and @xmath191 are the gas and faraday constants , respectively , @xmath98 is the temperature ( in kelvins , @xmath192 ) , and the concentration gradient @xmath193 is about @xmath194 .
the coefficient @xmath195 is about 60 mev at room temperature , @xmath196k .
it follows from eq .
( [ dmu ] ) that the potentials of the @xmath0 and @xmath1 sides of the membrane can be written as @xmath197 where @xmath198 at the standard conditions , when @xmath199 , @xmath200 mev , for the electrochemical potential @xmath201 we have : @xmath202 mev .
thus , the total proton gradient across the membrane , @xmath203 , is about 210 mev . as in the cco proton pump @xcite
, we assume that the proton - binding sites @xmath204 , and @xmath9 are located approximately on the line connecting the @xmath0 and @xmath1 sides of the membrane with the following coordinates : @xmath205 the coordinates of the sites are counted from the middle of the membrane in a direction towards the @xmath1-side and are measured in units of the membrane width @xmath206 with @xmath207 nm .
protons are delivered from the @xmath0-side to the site @xmath8 by the so - called @xmath12-pathway crossing about a half of the membrane .
we also note that the @xmath9-site is located next to the @xmath1-side ( see fig . 1 ) .
an influence of the transmembrane voltage @xmath190 on the energy levels of the proton sites is described by the formulas @xmath208 for the proton energy levels , @xmath209 , and @xmath210 , at the voltage @xmath211 , we assume the following values ( in mev ) : @xmath212 , and @xmath213 unless otherwise specified .
this means that at the standard conditions , the proton begins its journey at the @xmath0-side with the potential @xmath214 mev and jumps to the @xmath8-site having a lower energy ( @xmath215 mev ) .
however , the next proton - binding site @xmath10 has a much higher energy ( @xmath216 mev ) , so that the proton transfer can not occur without a mediation of the electron component .
the electron site @xmath7 is electrostatically coupled to the proton - binding site @xmath10 with the coulomb energy @xmath14 .
thus , in the presence of an electron on the site @xmath7 the energy of the @xmath13-proton decreases to the level @xmath217 mev , provided that @xmath218 mev . now the proton can move from site @xmath8 to site @xmath13 , since @xmath219 .
depopulation of the electron site @xmath13 returns the energy level of the @xmath13-proton to its original value @xmath220 mev , which is higher than the energy level of the next - in - line @xmath9 site , @xmath221 mev , and is _ much higher _ than the energy level of the @xmath8-site .
we assume that the backward proton transfer ( from @xmath10 to @xmath8 site ) is described by the _ inverted region _ of the marcus formula , so that the probability of such transfer is low , compared to the probability of the proton transfer from the site @xmath10 to the site @xmath9 .
no additional gate mechanism is necessary here . for the sake of simplicity , we assume that three electron - binding sites @xmath222 as well as the source and drain leads are positioned on a line , which is parallel to the surface of the membrane ( see fig
thus , the transmembrane gradient @xmath190 has no effect on electron transport from the electron source @xmath11 to the drain @xmath12 . for the potentials of the electron reservoirs , we choose the following form @xmath223 with @xmath224 mev and with the electron voltage gradient @xmath225 mev , unless otherwise indicated .
the electron voltage gradient @xmath226 roughly corresponds to the drop of the redox potential along the electron transfer chain in the cytochrome c oxidase @xcite .
we assume that the electron pathway includes the source reservoir ( @xmath227 mev ) , the site @xmath5 ( @xmath228 mev ) , the interaction site @xmath7 ( @xmath229 mev ) , the site @xmath6 ( @xmath230 mev ) , and the electron drain reservoir having the potential @xmath231 mev .
we assume that the electron and proton transfer between the active sites , @xmath5-@xmath13 , @xmath6-@xmath13 and @xmath8-@xmath13 , @xmath9-@xmath13 , are quite fast , with amplitudes @xmath232/ps and @xmath233/ps , whereas the transitions to and out the electron and proton reservoirs are characterized by much slower rates : @xmath234/ns , and @xmath235/ns .
the responses of the environment to the electron and proton transitions are described by the corresponding reorganization energies : @xmath236 and @xmath237 respectively . here
, for the standard case , we assume that @xmath238 mev and @xmath239 mev .
this set of parameters provides an efficient operation of the redox - linked proton pump . in fig .
3 , we show the steady - state proton current @xmath150 as a function of the transmembrane voltage gradient @xmath190 , at three different values of the electron voltage : @xmath240 mev .
we use here the standard set of other parameters ( see the previous subsection ) , where @xmath241 k and @xmath242 mev . the proton current @xmath150 is equal to the number of protons pumped _ energetically uphill _ ( at @xmath243 ) , from the negative side @xmath0 to the positive side @xmath1 of the membrane , _ per one microsecond_. at the difference @xmath225 mev of source and drain redox potentials , the system pumps more than 200 protons per one microsecond against the transmembrane voltage gradient @xmath244 mev . according to eq .
( [ munp ] ) , this voltage corresponds to the proton electrochemical gradient @xmath245 mev , which is usually applied to the internal membrane of mitochondria and the plasma membranes of bacteria .
the number of pumped protons goes down as the proton voltage @xmath190 increases , and goes up with increasing the electron voltage difference @xmath226 .
the proton current saturates at @xmath246 mev .
it is evident from fig . 3 that at high enough electron voltages ( @xmath247 mev ) , the pump is able to translocate more than 100 protons per microsecond against the proton gradient @xmath190 , exceeding 250 mev ( @xmath248 mev ) .
the quantum yield @xmath165 is about one ( with a power - conversion efficiency @xmath249% ) in the whole region of electron and proton voltages : 500 mev @xmath250 mev , 0 @xmath251 mev .
figure 4 shows the pumping proton current , @xmath150 ( i.e. , the number of protons translocated from the negative to the positive side of the membrane per one microsecond ) versus the temperature @xmath98 measured in kelvins .
the graphs are presented at three values of the electron and proton reorganization energy : @xmath252 mev .
we assume here that @xmath253 , with the electron voltage @xmath225 mev and the proton gradient @xmath244 mev .
it is of interest that at @xmath254 mev the pumping current has a pronounced maximum near the room temperature , 200 k @xmath255 k , although the quantum yield is higher , @xmath256 , at lower temperatures .
the performance of the pump deteriorates at higher reorganization energies when the coupling to the environment increases .
increasing the reorganization energy leads to increasing the probability for an electron to be transferred through the system , losing all its excess energy to the environment without transferring this energy to protons .
such a probability is further increased at large temperatures leading to the observed decrease of the quantum yield .
the energy transfer from the electron to the proton component occurs on the interaction site @xmath257 , which has one electron ( @xmath3 ) and one proton ( @xmath258 ) energy levels ( see eq . ( [ eaqb ] ) ) .
the electron on the site @xmath7 is electrostatically coupled to the proton , which populates the site @xmath10 , with the coulomb energy @xmath14 .
it follows from fig .
5 that the proton pumping current @xmath150 exhibits a resonant behavior as a function of the charging energy @xmath14 and the position of the proton energy level @xmath259 .
the dependence of the pumping current on the electron energy @xmath3 has a resonant character as well .
here we assume that @xmath225 mev , @xmath244 mev , @xmath242 mev , and @xmath260 k. the energetically - uphill proton current has a pronounced maximum ( @xmath261s ) at the coulomb energy @xmath262 mev and the proton energy @xmath263 mev , provided that the electron energy @xmath229 mev .
it is important that the proton pump is robust to the variations of the coulomb energy @xmath14 and the proton energy @xmath259 in the range @xmath264 mev from the resonant values .
the quantum yield @xmath165 is very close to one in the central region of fig . 5 , so that the power - conversion efficiency @xmath160 is about of 35% .
figures 3 , 4 and 5 clearly demonstrate that , at standard physiological conditions , the static redox - linked proton pump ( cco - pump " ) efficiently converts the energy of electrons to the more stable energetic form of the proton electrochemical gradient across the membrane . in many biological systems , electrons and protons can be transferred across a membrane by means of a molecular shuttle diffusing inside of the membrane , from one side to another . here
we show that the mathematical model described in section ii can be successfully applied for a description of the redox loop mechanism , which utilizes the brownian motion of the shuttle @xmath13 carrying both electron , @xmath7 , and proton , @xmath10 , sites ( see fig . 2 ) .
as in the previous case , we have to solve here a system of master equations for the electron ( @xmath265 ) and proton ( @xmath266 ) populations , eqs .
( [ nlrq],[nabq ] ) , and for the correlation function @xmath267 of electron and proton populations on the site @xmath13 , eq .
( [ k1 ] ) . however , these master equations should be complemented by the langevin equation , eq .
( [ langevin ] ) , for the time - dependent shuttle position @xmath41 .
we note that the electron tunneling between the sites @xmath5-@xmath13 , @xmath13-@xmath6 , as well as the proton transfer rates between the sites @xmath8-@xmath13 and @xmath13-@xmath9 , depend on the position @xmath41 of the shuttle .
we assume that the electron site @xmath5 is located near the negative ( @xmath0 ) side of the membrane , at @xmath268 , where @xmath269 nm . the other electron site @xmath6 is near the @xmath1-side of the membrane , at @xmath270 .
the reservoir @xmath11 , connected to the site @xmath5 , serves as a source of electrons , and the reservoir @xmath12 , coupled to the site @xmath6 , serves as an electron drain ( see fig.2 ) .
the tunneling amplitudes @xmath40 are determined by the amplitudes @xmath271 , and by the electron tunneling length @xmath272 : @xmath273 the proton - binding site @xmath8 is located at the end of the @xmath0-side proton pathway , whereas the site @xmath9 terminates a pathway , which goes into the @xmath1-side of the membrane . for the @xmath41-dependencies of the proton transfer
amplitudes @xmath274 and @xmath275 , we choose the following relations : @xmath276^{-2 } , \nonumber\\ \delta_b(x ) = \delta_{b0}\times \left [ \exp\left(\frac{x_0-x}{l_p}\right ) + 1 \right]^{-2 } , \label{deltaab}\end{aligned}\ ] ] where @xmath277 is the proton transfer length .
it should be noted that our model produces the same results when the proton amplitudes are given by the expressions similar to eqs .
( [ deltalr ] ) . for the transfer parameters , we choose the following values : @xmath278 mev , @xmath279 mev , and @xmath280 nm ,
@xmath281 nm . couplings to the electron and proton reservoirs are described by the rates @xmath282ns and @xmath283ns .
the system is robust to significant variations of the transfer parameters .
the confinement potential @xmath175 is determined by the height @xmath284 mev , the steepness @xmath285 nm , and the half - width @xmath286 nm .
the potential barrier @xmath180 , preventing the charged shuttle from entering into the membrane , is characterized by the height @xmath287 mev , the width @xmath288 nm , and the steepness @xmath289 nm . accordingly , the electron and proton populations of the shuttle are almost completely compensated , @xmath290 , so that the potential @xmath180 gives a negligible contribution to the energies of electrons and protons .
however , we have to take into account the fact that in the presence of the voltage gradient , @xmath291 mev , the electron ( @xmath3 ) and proton ( @xmath4 ) energies on the moving shuttle depend on the shuttle position @xmath41 : @xmath292 with @xmath293 mev , and @xmath294 mev , where for the charging energy @xmath14 of the shuttle we have : @xmath295 mev .
thus , electrons move from the source reservoir , having the electrochemical potential @xmath296 mev , to the @xmath297site ( with the energy @xmath298 mev ) , and , thereafter , to the shuttle . on the opposite side of the membrane , the electron , populating the shuttle , jumps to the site @xmath6 ( @xmath299 mev ) and , finally , to the drain reservoir ( @xmath300 ev ) .
the total drop of the redox potential in this electron - transport chain can be estimated as @xmath301 mev .
protons move from the @xmath0-side of the membrane ( @xmath214 mev ) to the site @xmath8 , having a lower energy @xmath302 mev .
the energy level @xmath303 mev of the proton on the shuttle , located near the @xmath0-side of the membrane ( at @xmath268 ) , is much higher than @xmath304 , if the shuttle contains no electrons .
however , the shuttle populated with a single electron is more attractive for protons , since in this case the effective energy of the proton , @xmath305 mev , is less than the energy of the proton - binding site @xmath8 .
the shuttle , carrying one electron and one proton , diffuses to the opposite side of the membrane ( @xmath270 ) , where the electron , with energy @xmath306 mev , is able to tunnel to the site @xmath6 , having a slightly higher energy @xmath307 mev . in the absence of an electron
, the energy of the proton on the shuttle ( at @xmath270 ) increases to the level @xmath308 mev , which exceeds the energy of the proton on the site @xmath9 : @xmath309 mev .
consequently , the proton moves from the shuttle to the site @xmath9 and , thereafter , to the @xmath1-side of the membrane characterized by the electrochemical potential @xmath310 mev .
thus , this redox loop mechanism translocates protons across the membrane against the proton electrochemical gradient @xmath311 mev , and against the transmembrane potential @xmath312 mev . figure 6 exhibits the electron and proton populations of the shuttle , @xmath313 and @xmath314 , correlated with the shuttle s position @xmath315 at @xmath316 k , @xmath244 mev , and at @xmath317 mev . in this figure
, we also show the time dependencies of the number of electrons , @xmath318 , transferred to the drain reservoir , and the number of protons , @xmath319 , translocated to the positive side of the membrane .
the shuttle diffuses between the membrane walls located at @xmath320 ( @xmath269 nm ) with an average crossing time @xmath321s .
this time - scale is closely related to the diffusion time , @xmath322 obtained at @xmath323 nm , for the diffusion coefficient of the quinone molecule @xmath324m@xmath325sec . at @xmath326 , the shuttle , located at @xmath327 ,
is loaded with one electron and one proton taken from the negative side of the membrane ( see fig . 2 ) . when @xmath328s , the shuttle reaches the positive side ( @xmath329 nm ) and unloads the electron to the the site @xmath6 ( and later to the drain lead @xmath12 ) and the proton to the site @xmath9 , coupled to the @xmath1-side of the membrane .
consequently , the population @xmath330 of the @xmath1-side grows .
the empty shuttle diffuses back , to the @xmath0-side , completing the cycle , and the process starts again . in twenty microseconds , the shuttle performs four complete trips and translocates about four electrons and four protons across the membrane .
the numbers of electrons and protons , @xmath331 and @xmath330 , respectively , transferred across the membrane in one millisecond , are shown in fig .
7 as functions of the transmembrane proton voltage @xmath332 the electrochemical gradient of protons , @xmath333 , is proportional to @xmath190 : @xmath334 mev ( at @xmath260 k ) .
the results in fig .
7 are averaged over ten realizations .
the system is able to translocate more than 120 protons per ms against the high transmembrane voltage , @xmath335 mev , that corresponds to the electrochemical gradient @xmath336 mev .
it follows from fig .
8 that the translocation mechanism works efficiently in a wide range of temperatures , 250 k @xmath337 k. in this range , the system pumps more than 120 protons per millisecond with a quantum yield exceeding 90% and with a power - conversion efficiency @xmath160 higher than 40% . with increasing temperature
, the shuttle performs more trips between the sides of the membrane , thus , carrying more electrons and protons .
this increases the proton current ( i.e. , the number of protons translocated per unit time ) .
we note that the proton population of the shuttle occurs only after loading the shuttle with an electron . at very high temperatures , @xmath338 k
, the shuttle moves quite fast , and protons have less chances to jump on the shuttle .
consequently , the gap between electron and proton currents grows with the temperature , thus deteriorating the performance of the pump .
two different mechanisms of energetically - uphill proton translocation across a biomembrane are described by the same physical model .
this model includes three redox sites @xmath339 and three proton binding sites @xmath340 attached to the source ( @xmath11 ) and drain ( @xmath12 ) electron reservoirs , as well as to the proton reservoirs on the positive and negative sides of the membrane .
we have shown that it is the strong coulomb interaction between the electron site @xmath7 and the proton site @xmath10 , which plays the most prominent role in the process of energy transformation from electrons to protons . in this case
, the whole electron - proton transport chain can be divided into weakly coupled clusters of sites , so that the total number of occupation states is equal to the sum ( not to the product ) of occupation states in each cluster . at physiological conditions ,
our model demonstrates a proton pumping effect with a quantum yield near 100% and a power - conversion efficiency of order of 35% , for both the static proton pump , related to the _ cytochrome c oxidase _ , as well as for the redox - loop mechanism , where electrons and protons are translocated by the diffusing molecular shuttle .
* acknowledgements .
* this work was supported in part by the laboratory of physical sciences , national security agency , army research office , national science foundation grant no .
0726909 , jsps - rfbr contract no .
09 - 02 - 92114 , grant - in - aid for scientific research ( s ) , mext kakenhi on quantum cybernetics , and funding program for innovative r&d on s&t ( first ) .
was partially supported by the nsf nirt , grant no .
ecs-0609146 and by the psc - cuny award no .
41 - 613 . )
thereafter , high - energy electrons , @xmath341 , tunnel _ energetically - downhill _ ( through the yellow path ) to the sites @xmath342 and , finally , to the drain @xmath12 . low - energy protons , @xmath343 ,
move _ energetically - uphill _ ( in blue ) from the negative ( @xmath0 ) side of the membrane to the sites @xmath344 and , eventually , reach the positive ( @xmath1 ) side of the membrane .
, width=529,height=604 ] , is placed on the molecular shuttle ( shown in green ) , which diffuses along the line connecting the negative and positive sides of the membrane . from the source reservoir @xmath11
, an electron @xmath341 jumps to the site @xmath5 and , thereafter , to the shuttle , located at @xmath345 the shuttle also accepts a proton @xmath343 transferred from the @xmath0-side of the membrane via the site @xmath8 .
the loaded shuttle moves randomly toward the positive side ( @xmath1 ) of the membrane , where ( at @xmath346 ) the electron is subsequently transferred from the site @xmath7 to the site @xmath6 and to the drain reservoir @xmath12 , and the proton jumps from the site @xmath10 to the site @xmath9 and , finally , to the positive ( @xmath1 ) side of the membrane .
we note that , in this design , the electron site @xmath5 and the proton site @xmath8 are located near the @xmath0-side of the membrane ( shown by the horizontal blue dashed line ) , and the electron site @xmath6 and the proton site @xmath9 are placed near the @xmath1-side . , width=415,height=491 ] at room temperature , @xmath260 k , and three different electron potentials : @xmath347 and @xmath348 mev .
the proton current is almost constant for low values of @xmath190 , and decreases for increasing @xmath190 . ,
width=793,height=453 ] mev , transmembrane proton voltage @xmath244 mev , and three different reorganization energies : @xmath349 and @xmath350 mev .
the proton current and quantum yield both decrease , for increasing @xmath351.,width=793,height=453 ] s , see the color bar on the right side ) on the charging energy @xmath14 , and on the energy @xmath259 of the central proton site for @xmath225 mev , @xmath352 mev , and @xmath260 k. , width=699,height=453 ] is the location of the shuttle , @xmath143 and @xmath144 are the electron and proton populations of the shuttle , respectively , @xmath331 is the number of electrons transferred from the electron source @xmath11 to the electron drain @xmath12 , and @xmath330 is the number of protons translocated from the negative ( @xmath0 ) to the positive ( @xmath1 ) side of the membrane .
it can be seen from this figure that the loading / unloading of the shuttle with electrons and protons , as well as the electron and proton transfer across the membrane , are clearly correlated with the spatial motion of the shuttle . ,
width=831,height=453 ] , and protons , @xmath330 , translocated across the membrane in one millisecond , versus the transmembrane proton voltage @xmath190 at room temperature , @xmath260 k , and at @xmath353 mev .
clearly , it is much harder to transfer protons against the higher transmembrane voltages .
, width=680,height=377 ] , and protons , @xmath330 , transferred across the membrane by the diffusing shuttle , at @xmath244 mev and @xmath353 mev .
we also present here the quantum yield , @xmath165 , and the power - conversion efficiency , @xmath160 , of the process as functions of the temperature . at higher temperatures ,
the shuttle moves faster and carries more electrons and more protons .
however , if the temperature is too high , the shuttle has not enough time to be loaded with electrons and protons , and sometimes travels empty . as a result of this , the electron and proton currents decrease at high temperatures , thus decreasing the efficiency of the pump .
, width=831,height=453 ] | we present two models for electron - driven uphill proton transport across lipid membranes , with the electron energy converted to the proton gradient via the electrostatic interaction . in the first model , associated with the _ cytochrome c oxidase _ complex in the inner mitochondria membranes , the electrostatic coupling to the site occupied by an electron lowers the energy level of the proton - binding site , making the proton transfer possible . in the second model , roughly describing the redox loop in a nitrate respiration of _
e. coli _ bacteria , an electron displaces a proton from the negative side of the membrane to a shuttle , which subsequently diffuses across the membrane and unloads the proton to its positive side .
we show that both models can be described by the same approach , which can be significantly simplified if the system is separated into several clusters , with strong coulomb interaction inside each cluster and weak transfer couplings between them .
we derive and solve the equations of motion for the electron and proton creation / annihilation operators , taking into account the appropriate coulomb terms , tunnel couplings , and the interaction with the environment .
for the second model , these equations of motion are solved jointly with a langevin - type equation for the shuttle position .
we obtain expressions for the electron and proton currents and determine their dependence on the electron and proton voltage build - ups , on - site charging energies , reorganization energies , temperature , and other system parameters .
we show that the quantum yield in our models can be up to 100% and the power - conversion efficiency can reach 35% . |
given two differentiable functions @xmath0 , the cauchy mean value theorem ( mvt ) states that for any interval @xmath1\subset { { \mathbb r}}$ ] , where @xmath2 , there exists a point @xmath3 in @xmath4 such that @xmath5\,g(c ) = [ g(b)-g(a)]\,f(c),\ ] ] where @xmath6 and @xmath7 .
a particular situation is the lagrange mvt when @xmath8 is the identity function , in which case reads as @xmath9 the problem to be investigated in this note can be formulated as follows .
[ the - problem ] find all pairs @xmath10 of differentiable functions
@xmath11 satisfying the following equation @xmath12 \ , g(\alpha a+\beta b ) = [ g(b)-g(a ) ] \ , f(\alpha a+\beta b)\ ] ] for all @xmath13 , @xmath2 , where @xmath14 , @xmath7 , @xmath15 are fixed and @xmath16 .
for the case of the lagrange mvt with @xmath17 , this problem was considered first by haruki @xcite and independently by aczl @xcite , proving that the quadratic functions are the only solutions to .
this problem can serve as a starting point for various functional equations @xcite .
more general functional equations have been considered even in the abstract setting of groups by several authors including kannappan @xcite , ebanks @xcite , fechner - gselmann @xcite . on the other hand ,
the result of aczl and haruki has been generalized for higher order taylor expansion by sablik @xcite . for the more general case of the cauchy mvt
much less is known .
we mention aumann @xcite illustrating the geometrical significance of this equation and the recent contribution of ples @xcite providing the solution of a related equation under additional assumptions . in this note
we provide a different approach to the cauchy mvt .
as it will turn out , the most challenging situation corresponds to @xmath17 in which case our main result is the following : [ thm : sym.cauchy.main ] assume that @xmath18 are three times differentiable functions with derivatives @xmath19 , @xmath20 such that @xmath21\,g\bigl(\frac{a+b}{2}\bigr ) = [ g(b)-g(a)]\,f\bigl(\frac{a+b}{2}\bigr ) \ ] ] for all @xmath22 .
then one of the following possibilities holds : a. @xmath23 are linearly dependent on @xmath24 ; b. @xmath25 , @xmath26 ; c. there exists a non - zero real number @xmath27 such that @xmath28 d. there exists a non - zero real number @xmath27 such that @xmath29 the paper is organized as follows . in section 2
we consider the problem first for the known case of the lagrange mean value theorem as an illustration of our method . in section 3
we provide a preliminary result that will allow to pass local information to a global one about the pairs of differentiable functions @xmath10 satisfying . in sections 4 , 5 we consider the asymmetric ( @xmath30 ) and symmetric ( @xmath31 ) cases , respectively .
section 6 is for final remarks .
here we also provide a partial result to an open problem by sahoo and riedel which corresponds to a more general version of .
note that every @xmath32 can be written uniquely as @xmath33 for some @xmath34 with @xmath16 .
it is easy to check that holds for all @xmath22 with fixed @xmath35 if @xmath36 is a linear function , and with @xmath37 if @xmath36 is a quadratic function .
we claim that the converse of this statement is also true .
as mentioned earlier , there are various proofs of the latter in the literature , see for example @xcite , @xcite , @xcite .
nevertheless , we give here a short and self - contained argument mainly to illustrate our approach to the more general case of the cauchy mvt .
[ lagrange ] let @xmath38 be fixed and @xmath39 .
assume that @xmath40 is a continuously differentiable function with @xmath41 such that @xmath42 then the following statements hold : 1 .
if @xmath35 then @xmath36 is a linear function ; 2 . if @xmath37 then @xmath36 is a quadratic function . let us denote by @xmath43 and @xmath44 . then reads as @xmath45 from this equation it is apparent that @xmath14 is differentiable as a linear combination of two differentiable functions and thus @xmath36 is twice differentiable . by induction , it follows that @xmath36 is infinitely differentiable . differentiating with respect to @xmath46
, we obtain the relation @xmath47 again , we differentiate with respect to @xmath46 and find that @xmath48 since @xmath49 is continuous , letting @xmath50 , we obtain @xmath51 if @xmath35 , this implies that @xmath52 identically .
therefore @xmath53 is constant and thus @xmath36 is a linear function , proving the first statement .
if @xmath37 , then the equation reads as @xmath54 and twice differentiation with respect to @xmath46 leads to @xmath55 now letting @xmath50 , we get @xmath56 for all @xmath26 , so @xmath53 is linear and @xmath36 is a quadratic function , proving the second statement .
let us introduce the sets @xmath57 and also their complements @xmath58 and @xmath59 .
observe that if @xmath60 is empty , i.e. @xmath61 is constant on @xmath24 , then holds for trivial reasons ( both sides are identically zero ) for arbitrary differentiable function @xmath36 .
of course , we can change the roles of @xmath61 and @xmath36 and claim : if @xmath36 is constant then holds for any differentiable function @xmath61 .
assume therefore that @xmath62 .
then there is a sequence of mutually disjoint open intervals @xmath63 , @xmath64 , such that @xmath65 [ prop : cauchy.gen.u_g<>empty ] if @xmath62 but @xmath66 , then @xmath67 , i.e. @xmath68 on @xmath24 and thus @xmath36 is constant . by assumption , there is a non - empty interval @xmath69 such that @xmath70 on @xmath71 , but @xmath72 for all @xmath73 $ ]
. then with the change of variables @xmath74 , @xmath75 , yields @xmath76 denoting by @xmath77 for @xmath73 $ ] and @xmath78 , we get @xmath79 if @xmath80 lies within the semi strip @xmath81 . then
, for @xmath82 choosing @xmath78 such that @xmath83 , we have @xmath84 so @xmath85 is a constant , say @xmath86 for @xmath82 .
however , by , we have @xmath87 and thus @xmath85 is the same constant for all @xmath88
. therefore , @xmath89 for all @xmath90 .
proposition [ prop : cauchy.gen.u_g<>empty ] shows that the condition @xmath91 holds only if at least one of the sets @xmath92 and @xmath60 is empty .
then we have the simple cases described as in the beginning of the section . [ prop : x ] let @xmath10 be a solution of the problem [ the - problem ] satisfying @xmath93 and consider the representation . if @xmath94 are linearly dependent as functions on @xmath95 for every @xmath96 , then @xmath94 are linearly dependent on @xmath24 . for @xmath97 with @xmath98 ,
consider the intervals @xmath99 , @xmath100 with @xmath101 and assume that @xmath94 are linearly dependent on @xmath102 and @xmath103 .
then it follows that there are constants @xmath104 , @xmath105 , @xmath106 , @xmath107 such that @xmath108 with the change of variables @xmath74 , @xmath75 , yields @xmath109\ , g(x ) = [ g(x+\alpha h)-g(x-\beta h)]\ , f(x)\ ] ] for all @xmath26 and @xmath78 . since @xmath110 if @xmath111 by and @xmath70 for @xmath111 , we have @xmath112 , \quad x\in i_{\sigma_2 } , h>0.\end{aligned}\ ] ] if at the same time @xmath113 , then @xmath114 by . inserting this value into
, we obtain @xmath115 for @xmath116 put @xmath77 , then @xmath117 , and means that @xmath80 lies within the parallelogram @xmath118 . since @xmath119
, guarantees that @xmath120 , and implies @xmath121 therefore , at any point of @xmath122 , we have @xmath123 but @xmath124 by , so @xmath125 and thus @xmath126 so far our analysis says nothing about @xmath106 , @xmath127 in , but since @xmath128 were arbitrary , together with , imply @xmath129 on the other hand , by changing the roles of @xmath36 and @xmath61 in the above analysis , we come to the conclusion that @xmath130 by there is a point @xmath131 so @xmath132 and these coefficients are not zero . but then implies @xmath133 and implies @xmath134 ; therefore , @xmath135 and @xmath136 .
the latter means that @xmath137 by trivial reasons ( all these values are zeros ) so with and these identities are valid on the entire @xmath138 .
in particular , it follows that @xmath94 are linearly dependent on @xmath24 .
in this section we consider the asymmetric case , i.e. , in we take @xmath139 the following proposition describes all pairs @xmath10 of two times continuously differentiable functions satisfying under the assumption on @xmath140 in the intervals where @xmath7 does not vanish .
[ cauchy1 ] let @xmath10 be a solution of the problem [ the - problem ] with @xmath140 satisfying and @xmath141 , @xmath142 , be an interval where the derivative @xmath143 does not vanish . if @xmath144 are two times continuously differentiable on @xmath145 , then @xmath94 are linearly dependent on @xmath145 . with the change of variables @xmath74
, @xmath75 , yields @xmath146\ , g(x ) = [ g(x+\alpha h)-g(x-\beta h)]\ , f(x)\ ] ] if @xmath147 and @xmath78 such that @xmath148 , @xmath149 .
the latter condition tells that holds if @xmath150 lies within the open triangle @xmath151 by differentiating both sides of with respect to @xmath46 twice , we obtain the following relation in @xmath152 @xmath153\,g(x)= [ \alpha^2 g(x+\alpha h)-\beta^2 g'(x-\beta h)]\,f(x).\end{aligned}\ ] ] all the functions are continiuous so holds on the closure @xmath154 as well , in particular , on the interval @xmath155 .
therefore , with @xmath156 by , we get @xmath157 for all @xmath158
. we can divide both sides by @xmath159 and conclude that @xmath160 on @xmath145 .
this implies that @xmath161 for some constant @xmath162 , and @xmath163 , @xmath147 .
after integration we get @xmath164 , @xmath147 .
the following theorem is the main result of this section .
[ thcauchy1 ] let @xmath10 be a solution of the problem [ the - problem ] with @xmath140 satisfying . if @xmath144 are two times continuously differentiable on @xmath24 , then @xmath94 are linearly dependent on @xmath24 , i.e. there exist constants @xmath165 such that not all of them zeroes and @xmath166 consider the following cases : case 1 : @xmath167 . in this case
@xmath61 is a constant on @xmath24 and holds for any differentiable function @xmath36 .
hence holds , for example , with @xmath168 , @xmath169 , @xmath170 and thus @xmath94 are linearly dependent on @xmath24 .
case 2 : @xmath171 but @xmath172 . in this case proposition [ prop : cauchy.gen.u_g<>empty ]
yields that @xmath36 is a constant on @xmath24 and holds for any differentiable function @xmath61
. hence holds , for example , with @xmath173 , @xmath174 , @xmath175 and thus @xmath94 are again linearly dependent on @xmath24 .
case 3 : @xmath176 . in this case
propositions [ cauchy1 ] and [ prop : x ] immediately imply that @xmath94 are linearly dependent on @xmath24 .
in this section we consider the problem of describing all pairs @xmath10 of smooth functions for which the mean value in is taken at the midpoint of the interval .
our first result gives a necessary ( and also sufficient in case @xmath177 are not linearly dependent ) condition on such pairs in the intervals where @xmath7 does not vanish .
[ cauchy2 ] assume that @xmath18 are three times differentiable functions with derivatives @xmath19 , @xmath20 .
let @xmath178 be such an interval that @xmath179 for all @xmath147 and holds for all @xmath180 . then there exist constants @xmath181 and @xmath182 such that @xmath183 moreover , if holds with @xmath184 , then holds if and only if @xmath185 for all @xmath186 such that @xmath187 .
with the change of variables @xmath188 , @xmath189 , we can rewrite as @xmath190g(x)= [ g(x+h)-g(x - h)]f(x)\ ] ] for all @xmath191 with the property that @xmath192 . by differentiating this equality three times with respect to @xmath46
, we get @xmath193\,g(x ) = [ g''(x+h)+g''(x - h)]\,f(x).\ ] ] setting @xmath194 , we obtain @xmath195 and thus @xmath196 for some constant @xmath197
. then @xmath198 , @xmath199 , and integration over @xmath200 with any @xmath201 yields .
now assume holds with a nonzero constant @xmath197
. then we have @xmath202 and @xmath203\frac{f(x)}{g(x ) } = & \bigg(\int_{x - h}^{x+h } g(t ) dt\bigg ) \bigg(a + k\int_{x_0}^x \frac{du}{g^2(u)}\bigg)\\\label{fct_comp2 } = & a\int_{x - h}^{x+h } g(t ) dt + k\bigg(\int_{x - h}^{x+h } g(t ) dt\bigg ) \bigg(\int_{x_{0}}^{x}\frac{du}{g^{2}(u ) } \bigg).\end{aligned}\ ] ] by comparing and , it is easy to see that is equivalent to .
the following example illustrates that there are non - trivial functions satisfying ( and hence ) on @xmath24 .
consider @xmath204 on @xmath205 and let @xmath206 .
the integral condition reads as the following identity @xmath207 a direct computation gives @xmath208 , and consequently , @xmath209 we invite the interested reader to verify directly that the pair @xmath10 in satisfies the relation , giving a non trivial example of such pairs .
now we assume that @xmath210 and analyze the property for all @xmath211 such that @xmath212 . differentiating it with respect to @xmath46
, we obtain @xmath213\int_{x_0}^x\frac{du}{g^2(u)}.\ ] ] differentiation two more times with respect to @xmath46 gives @xmath214\int_{x_0}^x\frac{du}{g^2(u)},\ ] ] for all @xmath147 and @xmath215 such that @xmath212 . setting @xmath216 in these two equations ,
we obtain @xmath217\int_{x_0}^x\frac{du}{g^2(u)},\end{aligned}\ ] ] and @xmath218\int_{x_0}^x\frac{du}{g^2(u)},\end{aligned}\ ] ] for all @xmath147 with @xmath219 .
since @xmath220 and @xmath221 has no zeros in @xmath145 , both sides of do not vanish . by comparing and
, we get @xmath222 for all @xmath147 such that @xmath219 . denoting by @xmath223 and @xmath224 ,
yields the second order differential equation @xmath225 , whose general real - valued solution ( depending on the sign of @xmath226 ) , has the following form @xmath227 where @xmath228 , @xmath229 are real constants .
hence @xmath61 has one of the following forms @xmath230 where @xmath231 are real constants .
[ rem : cauchy.sym ] altogether , we come to the following conclusion : on every interval @xmath232 on which @xmath233 , either @xmath94 are linearly dependent , or @xmath61 and thus also @xmath36 , cf . , has one of the forms described in . in the sequel , we call a function @xmath61 ( resp .
the pair @xmath10 ) to be of _ quadratic , exponential _ or _
trigonometric type on @xmath145 _ if @xmath61 has ( resp . both of @xmath36 and @xmath61 have ) the form , or , respectively .
consider the set @xmath60 and its representation , cf .
, . the following lemma plays a crucial role in the analysis of the equation . [
lem : cauchy.sym.f(q)=0 ] let @xmath234 be such that @xmath235 and @xmath236
. then @xmath94 are linearly dependent on @xmath237 .
@xmath238 by definition so by remark [ rem : cauchy.sym ] , it is sufficient to consider the following cases .
case 1 : @xmath61 is of quadratic type on @xmath71 .
then @xmath36 is also of quadratic type on @xmath71 , and since @xmath239 , we have @xmath240 . thus @xmath94 are linearly dependent on @xmath71 .
case 2 : @xmath61 is of either exponential or trigonometric type on @xmath71 .
first suppose that @xmath61 is of exponential type on @xmath71 .
then so is @xmath36 and since the set of functions satisfying is invariant with respect to the addition of constant functions , we can assume , without loss of generality , that @xmath241 for some @xmath242 .
hence there are real constants @xmath243 , @xmath244 such that @xmath245 , @xmath246 .
since @xmath247 , we get @xmath248 and thus @xmath249 .
the same argument for @xmath61 explains that @xmath250 for some real @xmath251 , and consequently @xmath36 and @xmath61 are multiples of the same function @xmath252 .
if @xmath61 is of trigonometric type , then by the same way as above , we can conclude that @xmath36 and @xmath61 are multiples of the same function @xmath253 , implying that @xmath94 are linearly dependent on @xmath237 .
[ [ proof-of-theoremthmsym.cauchy.main ] ] proof of theorem [ thm : sym.cauchy.main ] + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + consider the set @xmath60 defined in .
if @xmath254 , then @xmath255 on @xmath24 , and thus @xmath61 is identically constant on @xmath24 . in this case
@xmath36 can be an arbitrary differentiable function on @xmath24 and thus @xmath23 are linearly dependent on @xmath24 . if @xmath256 , then it follows ( cf .
remark [ rem : cauchy.sym ] ) that either @xmath23 are linearly dependent or @xmath61 has one of the forms on the whole of @xmath24 .
moreover , we get the same conclusion if @xmath257 ( cf .
proposition [ prop : cauchy.gen.u_g<>empty ] ) .
next , let us assume that @xmath258 and @xmath60 is a proper subset of @xmath24 .
consider the representation .
it is clear ( cf .
remark [ rem : cauchy.sym ] ) that the index set @xmath259 can be split into disjoint subsets as @xmath260 , where @xmath261 _ claim 1_. if @xmath262 , then @xmath263 .
assume @xmath264 is a proper subest of @xmath259 .
then there exists @xmath265 such that @xmath266 . since @xmath262
, there is @xmath267 and @xmath268 such that @xmath269 on @xmath270 .
consider all @xmath211 such that @xmath271 and @xmath270 .
using for @xmath272 and @xmath273 , and recalling that @xmath179 on @xmath102 , we get @xmath274 therefore , @xmath275 and thus @xmath276 for all @xmath211 such that @xmath271 and @xmath270 . from this
it follows that @xmath36 and @xmath61 are in linear relationship on @xmath103 , that is , @xmath277 , which leads to a contradiction .
_ claim 2_. if @xmath278 , then only one of the index sets @xmath279 , @xmath280 , @xmath281 is non - empty .
let @xmath282 and @xmath283 . since @xmath60 is proper subset of @xmath24 , one of @xmath284 is finite .
we can assume @xmath235 .
then @xmath238 , and lemma [ lem : cauchy.sym.f(q)=0 ] yields @xmath285 . hence using for @xmath286 and @xmath287 we get @xmath288 so the graph of @xmath61 is symmetric with respect to the vertical line @xmath289 . if @xmath290 or @xmath291 , then @xmath292 since the functions of quadratic type has exactly one and the functions of exponential type has at most one critical point . therefore , if @xmath290 , then @xmath293 , @xmath26 and @xmath294 .
similarly , it follows from that if @xmath291 , then @xmath295 .
next , assume @xmath296 .
then @xmath297 and let @xmath298 .
since @xmath61 is of trigonometric type on @xmath299 , we must have @xmath300 .
so @xmath301 and it follows as in the proof of lemma [ lem : cauchy.sym.f(q)=0 ] that there are real constants @xmath243 , @xmath244 such that @xmath302 using we obtain that holds on the whole of @xmath24 .
since @xmath62 , at least one of @xmath264 , @xmath279 , @xmath280 , @xmath281 is non - empty . if @xmath262 , then claim 1 and proposition [ prop : x ] imply that @xmath94 are linearly dependent on @xmath24 .
if @xmath278 , then claim 2 yields that one of the possibilities @xmath303 @xmath304 holds .
as a consequence of our main result we can give a partial answer to following still open question of sahoo and riedel ( cf .
* section 2.7 ) for an equivalent formulation ) .
we provide a partial result to this problem under certain assumptions on the unknown functions .
first let us make the change of variables @xmath308 , @xmath309 and write equivalently as @xmath310\,\phi(s ) = [ g(s+t)-g(s - t)]\,\psi(s ) , \quad s , t\in{{\mathbb r}}.\ ] ] a. there exist constants @xmath315 such that for all @xmath316 , we have @xmath317 and @xmath318=0 $ ] ; b. there exist constants @xmath319 such that for all @xmath316 , we have @xmath320 , @xmath321 and @xmath322 c. there exists @xmath323 and constants @xmath319 such that for all @xmath316 , we have @xmath324 , @xmath325 and @xmath326 d. there exists @xmath323 and constants @xmath319 such that for all @xmath316 , we have @xmath327 , @xmath328 and @xmath329\,\phi(s ) = [ b_1\cos(\mu s)-b_2\sin(\mu s)]\,\psi(s ) . \ ] ] let @xmath330 be the derivatives of @xmath144 , respectively and the sets @xmath60 , @xmath92 ( resp .
@xmath331 , @xmath332 ) be defined as in section 3 . without loss of generality , assume that @xmath333 does not vanish on @xmath24 . by differentiating with respect to @xmath334 and setting @xmath335 in the resulting equation , we get @xmath336 for any @xmath337 and @xmath338 , by and , we have @xmath339\,\frac{\psi(s)}{\phi(s)}\\ & = [ g(s+t)-g(s - t)]\,\frac{f(s)}{g(s)},\end{aligned}\ ] ] and thus @xmath340\ , g(s ) = [ g(s+t)-g(s - t)]\,f(s ) , \quad s\in u_g , t\in{{\mathbb r}}.\ ] ] on the other hand , observe that we have @xmath341 by since @xmath313 on @xmath24 . so holds for all @xmath342 .
therefore , theorem can be applied to and the four characterizations follows immediately .
it is likely that the methods of this paper work for related equations when we replace the linear mean @xmath343 by the @xmath344-mean @xmath345 , for @xmath346 . here
@xmath347 is defined for all values of @xmath348 .
for @xmath349 the corresponding mean is defined by @xmath350 . moreover , for @xmath351 we can define @xmath352 and @xmath353 .
we intend to investigate this issue in a subsequent paper .
the essence of our approach is to reduce a functional equation to an ode . for this strategy
we need certain smoothness assumptions
. it would be interesting to provide an alternative way that will not require this additional smoothness assumptions .
1 j. aczl , _ a mean value property of the derivative of quadratic polynomials - without mean values and derivatives .
_ math . mag .
* 58 * ( 1985 ) , no .
1 , 4245 .
g. aumann , _ ber eine elementararen zusammenhang zwischen mittelwerten , streckenrechnung und kegelschnitten .
_ tohoku math . j.
* 42 * ( 1936 ) , 3237 .
b. ebanks , _ generalized cauchy difference equations ii .
* 136 * ( 2008 ) , no .
11 , 39113919 .
w. fechner , e. gselmann , _ general and alien solutions of a functional equation and of a functional inequality .
debrecen * 80 * ( 2012 ) , no . 1 - 2 , 143154
. s. haruki , _ a property of quadratic polynomials .
monthly * 86 * ( 1979 ) , no . 7 , 577579
kannappan , _
rudin s problem on groups and a generalization of mean value theorem . _ aequationes math . * 65 * ( 2003 ) , no . 1 - 2 , 8292 .
z. ples , _ on the equality of quasi - arithmetic and lagrangian means .
_ j. math .
* 382 * ( 2011 ) , 8696 .
m. sablik , _
taylor s theorem and functional equations . _ aequationes math .
* 60 * ( 2000 ) , no .
3 , 258267 . p.k .
sahoo , t. riedel , _ mean value theorems and functional equations . _
world scieintific publishing co. inc . , river edge , nj , 1998 . | the aim of this note is to characterize all pairs of sufficiently smooth functions for which the mean value in the cauchy mean value theorem is taken at a point which has a well - determined position in the interval . as an application of this result
, a partial answer is given to a question posed by sahoo and riedel .
146477 . ] |
the derivative of non - integer order has been an interesting research topic for several centuries .
the idea was motivated by the question , `` what does it mean by @xmath5 , if @xmath6 ? '' , asked by lhospital in 1695 in his letters to leibniz @xcite . since then , the mathematicians tried to answer this question for centuries in several points of view .
the outcomes are many folds .
various types of fractional derivatives were introduced : riemann - liouville , caputo , hadamard , erdlyi - kober , grnwald - letnikov , marchaud and riesz are just a few to name @xcite .
most of the fractional derivatives are defined via fractional integrals @xcite . due to the same reason , those fractional derivatives inherit some non - local behaviors , which lead them to many interesting applications including memory effects and future dependence @xcite . among the inconsistencies of the existing fractional derivatives
are : 1 .
most of the fractional derivatives except caputo - type derivatives , do not satisfy @xmath7 , if @xmath8 is not a natural number .
2 . all fractional derivatives do not obey the familiar product rule for two functions : @xmath9 3 .
all fractional derivatives do not obey the familiar quotient rule for two functions : @xmath10 4 .
all fractional derivatives do not obey the chain rule : @xmath11 5 .
fractional derivatives do not have a corresponding rolle s theorem .
fractional derivatives do not have a corresponding mean value theorem
all fractional derivatives do not obey : @xmath12 , in general .
the caputo definition assumes that the function @xmath3 is differentiable . to overcome some of these and other difficulties , khalil et al .
@xcite , came up with an interesting idea that extends the familiar limit definition of the derivative of a function given by the following .
@xcite let @xmath13 and @xmath14 .
then the `` conformable fractional derivative '' of @xmath3 of order @xmath8 is defined by , @xmath15 for @xmath16 .
if @xmath3 is @xmath4differentiable in some @xmath17 , and @xmath18 exists , then define @xmath19 as a consequence of the above definition , the authors in @xcite , showed that the @xmath4derivative in ( [ def1 ] ) , obeys the product rule , quotient rule and has results similar to the rolle s theorem and the mean value theorem in classical calculus .
the purpose of this work is to further generalize the results obtained in @xcite and introduce a new fractional derivative as the most natural extension of the familiar limit definition of the derivative of a function @xmath3 at a point .
in this section , we give the main definition of the paper and obtain several results that are close resemblance of the results found in classical calculus .
we prove that the new fractional derivative satisfies the product rule , quotient rule , chain rule and have results which are natural extensions of the rolle s theorem and the mean value theorem . to this end , we start with the following definition , which is a generalization of the limit definition of the derivative .
let @xmath13 and @xmath14 .
then the _ fractional derivative _ of @xmath3 of order @xmath8 is defined by , @xmath20 for @xmath16 .
if @xmath3 is @xmath4differentiable in some @xmath17 , and @xmath21 exists , then define @xmath22 the first result is the generalization of theorem 2.1 of @xcite . if a function @xmath13 is @xmath4differentiable at @xmath23 $ ] , then @xmath3 is continuous at @xmath24 . since @xmath25 , we have @xmath26 = \lim_{\epsilon \rightarrow 0 } \frac{f(ae^{\epsilon a^{-\alpha}})-f(a)}{\epsilon } .\lim_{\epsilon \rightarrow 0}\epsilon.\ ] ] let @xmath27 . then , @xmath28 = \mathcal{d}^\alpha ( f)(a ) \cdot 0 = 0 $ ] , which , in turn , implies that @xmath29 .
this completes the proof .
the following is the main result of this paper .
[ thm3 ] let @xmath30 $ ] and @xmath31 be @xmath4differentiable at a point @xmath32 . then , 1 . @xmath33 = a\mathcal{d}^\alpha
( f ) + b\mathcal{d}^\alpha ( g)$ ] , for all @xmath34 .
2 . @xmath35 for all @xmath36 .
@xmath37 for all constant functions , @xmath38 4 .
5 . @xmath40 .
@xmath41 , @xmath3 @xmath42 .
if , in addition , @xmath3 is differentiable , then @xmath43 part ( 1 ) and ( 3 ) follow directly from the definition .
we shall prove ( 2 ) , ( 4 ) , ( 6 ) and ( 7 ) , since the proofs are different from what appear in @xcite .
now , for fixed @xmath30 $ ] , @xmath36 and @xmath32 , we have @xmath44 this completes the proof of ( 2 ) .
then , we shall prove ( 4 ) . to this end , since @xmath31 are @xmath4differentiable at @xmath45 note that , @xmath46 + f(t)\lim_{\epsilon \rightarrow 0 } \frac{g(te^{\epsilon t^{-\alpha}})-g(t)}{\epsilon } \nonumber \\ & = \mathcal{d}^\alpha \big(f\big)(t)\lim_{\epsilon \rightarrow 0 } g(te^{\epsilon t^{-\alpha } } ) + f(t)\mathcal{d}^\alpha \big(g\big)(t ) .
\nonumber\end{aligned}\ ] ] since @xmath47 is continuous at @xmath48 , @xmath49 this completes the proof of ( 4 ) .
the proof of ( 5 ) is similar and is left for the reader to verify .
next , we prove ( 6 ) in two different approaches .
first , suppose @xmath50 is @xmath4differentiable at a point @xmath51 and @xmath52 is _ differentiable _ at a point @xmath53 .
let @xmath54 , and @xmath55 .
since @xmath56 , where @xmath57 , we have @xmath58 where @xmath59 as @xmath60 .
thus , @xmath61 is a continuous function of @xmath62 if we define @xmath61 to be @xmath63 when @xmath64 .
now , if @xmath65 is an increment in @xmath48 and @xmath66 and @xmath67 ( with the possibility of both being equal to @xmath63 ) are corresponding increments in @xmath68 and @xmath69 , respectively .
then we may write , using equation ( [ eq2 ] ) , @xmath70 where @xmath71 as @xmath72 , and @xmath73 \delta u \nonumber \\ & = \big[d^1 f(b ) + \epsilon_1 \big ] \cdot \big[\mathcal{d}^\alpha g(a ) + \epsilon_2 \big ] \delta t \nonumber \end{aligned}\ ] ] where both @xmath59 and @xmath71 as @xmath72 .
taking @xmath74 , we now have , @xmath75 \cdot \big[\mathcal{d}^\alpha g(a ) + \epsilon_2 \big ] \\ & = d^1 f(b)\mathcal{d}^\alpha g(a ) = f^\prime(g(a))\mathcal{d}^\alpha g(a).\end{aligned}\ ] ] it can be seen that equation [ eq3 ] , can not be written in the form of equation [ eq2 ] .
this makes it necessitates to use the @xmath4derivative in equation [ eq3 ] .
now , we prove the result following a standard limit - approach . to this end , in one hand , if the function @xmath47 is constant in a neighborhood containing @xmath24 , then @xmath76 . on the other hand ,
assume that the function @xmath47 is non - constant in the neighborhood of @xmath24 .
in this case , we can find an @xmath77 such that , @xmath78 for any @xmath79 . now , since @xmath47 is continuous at @xmath24 , for @xmath80 sufficiently small , we have @xmath81 for @xmath82 .
this establishes the chain rule . to prove part ( 7 )
, we use a similar line of argument as in equation ( [ eq1 ] ) .
thus , taking @xmath83 , we have @xmath84 since , by the assumption , @xmath3 is differentiable at @xmath32 .
this completes the proof of the theorem .
the result similar to ( 6 ) does not appear in @xcite .
the argument used in equation ( [ eq1 ] ) was / will be utilized in several occasions in the paper .
the following theorem lists @xmath4fractional derivative of several familiar functions .
the results are identical to that of conformable fractional derivative discussed in @xcite , which will be shown to be a special case of @xmath4fractional derivative defined in ( [ df ] ) .
[ thm4 ] let @xmath85 and @xmath30 $ ] .
then , we have the following results . 1 . @xmath86 .
2 . @xmath87 3 .
@xmath88 4 .
@xmath89 5 .
@xmath90 6 .
@xmath91 it is easy to see from part ( 6 ) of theorem [ thm3 ] that we have rather unusual results given in the next theorem .
[ thm5 ] let @xmath30 $ ] and @xmath92 .
then , 1 .
@xmath93 2 .
@xmath94 3 .
@xmath95 theorem [ thm5 ] suggests that there is a pseudo - invariant space corresponding to the @xmath4fractional derivative on which the @xmath96 , @xmath97 and the exponential function , @xmath98 behave as they are having familiar classical derivatives . due to the fourier theory , this in turn ,
implies that any function which possesses a fourier expansion behaves as having classical derivatives .
this would be an interesting topic for further research .
in @xcite , the author introduced fractional integrals and derivatives , which generalize the riemann - liouville and the hadamard fractional integrals and derivatives to a single form . in the next subsection , we show that the results obtained by khalil et at .
@xcite are special cases of the more general extension discussed in this paper .
let us first define a truncated exponential function given by , @xmath99 with the help of this definition , we will define another fractional derivative given below .
let @xmath13 and @xmath14 .
then the _ fractional derivative _ of @xmath3 of order @xmath8 is defined by , @xmath100 for @xmath16 .
if @xmath3 is @xmath4differentiable in some @xmath17 , and @xmath101 exists , then define @xmath102 now , it is easy to see that @xmath103 and @xmath104 where the expression on the right - hand - side of ( [ ndef1 ] ) is the conformable fractional derivative defined in ( [ def1 ] ) in @xcite , while the one on ( [ ndef2 ] ) is the @xmath4fractional derivative defined in this paper . furthermore , if @xmath2 the equation ( [ ndef1 ] ) becomes the classical definition of the first derivative of a function @xmath3 at a point @xmath48 . while , noting that @xmath105 , we may also show that the equation ( [ ndef2 ] ) is equivalent to the classical definition of the first derivative of a function @xmath3 .
these observations further suggest that there are corresponding results similar to the rolle s theorem and the mean value theorem .
next , we give those two results .
let @xmath51 and @xmath106 \rightarrow \mathbb{r}$ ] be a function with the properties that 1 .
@xmath3 is continuous on @xmath107 $ ] , 2 .
@xmath3 is @xmath4differentiable on @xmath108 for some @xmath109 , 3 .
@xmath110 .
then , there exists @xmath111 , such that @xmath112 we prove this using contradiction . since @xmath3 is continuous on @xmath107 $ ] and @xmath110 , there is @xmath113 , at which the function has a local extrema .
then , @xmath114 but , the two limits have opposite signs .
hence , @xmath115 let @xmath51 and @xmath106 \rightarrow \mathbb{r}$ ] be a function with the properties that 1 .
@xmath3 is continuous on @xmath107 $ ] , 2 .
@xmath3 is @xmath4differentiable on @xmath108 for some @xmath109 , then , there exists @xmath111 , such that @xmath116 consider the function , @xmath117 then , the function @xmath47 satisfies the conditions of the fractional rolle s theorem .
hence , there exists @xmath111 , such that @xmath118 .
using the fact that @xmath119 , the result follows . as a result of the fractional mean value theorem
, we also have the following proposition .
let @xmath106 \rightarrow \mathbb{r}$ ] be @xmath4differentiable for some @xmath109 .
suppose that @xmath120 is bounded on @xmath107 $ ] .
then , if 1 .
@xmath51 , or 2 .
@xmath120 is continuous at either @xmath24 or @xmath121 , the function @xmath3 is uniformly continuous on @xmath107 $ ] , and hence @xmath3 is bounded .
figure 1 depicts the graphs of the @xmath4fractional derivatives of the power function @xmath122 for @xmath123 and @xmath124 .
it is also interesting to note that the function @xmath125 has a constant derivative equal to @xmath126 , and not equal to @xmath127 as in the case of the riemann - liouville derivative .
this also suggests that the new derivative should have a rather different interpretation in the geometrical sense , which has yet to be discovered .
further , it can be seen that a function could be @xmath4differential at a point but not differentiable in the classical sense .
( cf . page 67 of @xcite ) .
for example , the function @xmath128 is everywhere @xmath129differentiable and has the derivative of @xmath130 while it is not differentiable at @xmath131 .
the similar results are there for the riemann - liouville and the caputo derivatives .
next , we consider the possibility of @xmath132 $ ] , for some @xmath133 . we have the following definition .
[ def5 ] let @xmath132 $ ] , for some @xmath133 and @xmath3 be an @xmath134differentiable at @xmath92 .
then the @xmath4fractional derivative of @xmath3 is defined by @xmath135 if the limit exists . as a direct consequence of definition [ def5 ] and part ( 7 ) of theorem [ thm3 ]
, we can show that @xmath136 , where @xmath132 $ ] and @xmath3 is @xmath137differentiable at @xmath32 .
now , the question whether the fractional derivative @xmath138 has a corresponding @xmath4fractional integral will be answered next . for simplicity
, we shall only consider the class of continuous functions here .
the results can easily be extended to more general settings and will be discussed in forthcoming papers .
as in the works of @xcite , it is interesting to note that , in spite of the variation of the definitions of the fractional derivatives , we can still adopt the same definition of the fractional integral here due to the fact that we obtained similar results in theorem [ thm4 ] as of the results ( 1)(6 ) and ( i)(iii ) in @xcite .
so , we have the following definition .
let @xmath139 and @xmath140 .
also , let @xmath3 be a function defined on @xmath141 $ ] and @xmath142 .
then , the _ @xmath4fractional integral _ of @xmath3 is defined by , @xmath143 if the riemann improper integral exists .
the proof is a direct consequence of the fundamental theorem of calculus . since @xmath3 is continuous , @xmath145 is clearly differentiable .
therefore , using part ( 7 ) of theorem [ thm3 ] , we have @xmath147 this completes the proof . the applications of this @xmath4integral and derivative will be discussed in a forthcoming paper .
several interesting applications of a similar fractional derivative defined in ( [ def1 ] ) and the corresponding fractional integral appear in @xcite , the major reference of this paper . 1 . what is the physical meaning or geometric interpretation of the @xmath4derivative ?
2 . is there a similarity between the @xmath4derivative and the _ gteaux derivative _ ?
one of the limitations of this version of the fractional derivative is that it assumes that the variable @xmath92 .
so the question is whether we can relax this condition on a special class of functions , if so , what is it ? | we introduce a new fractional derivative which obeys classical properties including : linearity , product rule , quotient rule , power rule , chain rule , vanishing derivatives for constant functions , the rolle s theorem and the mean value theorem .
the definition , @xmath0 is the most natural generalization that uses the limit approach . for @xmath1
, it generalizes the classical calculus properties of polynomials . furthermore ,
if @xmath2 , the definition is equivalent to the classical definition of the first order derivative of the function @xmath3 . furthermore , it is noted that there are @xmath4differentiable functions which are not differentiable . |
in spectroscopic studies ionization potential ( ip ) , electron affinity ( ea ) , excitation energy ( ee ) and double ip ( dip ) are important in many aspects . for example , in all kind of electron spectroscopy , including photo - electron and auger the ip , dips always play major roles .
specially , the knowledge of core ip provides insights into the choice and tunability of the ionizing beam in all kinds of electron scattering experiments like @xmath1 , @xmath2 , @xmath3 , double auger decay etc .
these experiments are not only important to understand the physics but also to the development of new fields like quantum computing thorough the idea of entanglement involved in the process .
hence accurate theoretical determination of ip for core as well as valance electrons may give some impetus to the physics underlying in those experiments . open - shell coupled cluster ( oscc ) method is used since several years to compute the ip , ea @xcite and dips @xcite - usually employing complete model space ( cms ) @xcite . in 1980
s coupled cluster ( cc ) methods have been generalized to handle the incomplete model space ( ims ) @xcite in a size - extensive manner . in this paper
we have tried to demonstrate the results arising from the algebraic structure of the expansion of the wave operator @xmath4 .
we have used both the ordinary and normal ordered cluster expansion of the wave operator @xmath0 .
normally the operator @xmath5 corresponds to core - correlation .
lindgren @xcite has shown how the normal ordered cluster expansion of the wave operator @xmath0 can include the correlation effects . here
we will show in a sophisticated way how the relaxation effects , the many - body effect of different kind can also be taken into account in the calculation of ionization potential within the algebraic structure of the semi - normal ordered cluster expansion of the wave operator @xmath0 .
the form of the cluster expansion takes care the relaxation effect in a compact and efficient manner and it can be easily demonstrated how the coupling of different many body effects like relaxation and correlation effects come into the picture automatically . in an earlier publication chaudhuri _
@xcite have explored this techniques and applied to molecules . in this paper
we have calculated ionization potentials ( ips ) of different states of ne , ar and kr . ips are calculated for core and valance orbitals of those closed shell atoms using two different approaches .
section [ theory ] deals with the theoretical formulation of the two different methods of calculation using the @xmath6 and @xmath7 theories respectively , where @xmath8 denotes normal ordering .
section [ results ] presents the results obtained from our calculation and a detailed discussion about the physical aspects of the two theories . in section [ concl ]
we have made the concluding remarks and the important findings of our calculations .
it is a well established fact that the orbital relaxation effects are more important than the correlation effects in the core hole ionization ( chi ) process . since in chi
the relaxation effects are stronger than the correlation effects , perturbative treatment will not be a efficient one because in higher orders it will be very cumbersome and that is the reason to choose the non - perturbative approach like cc .
let us start from the @xmath9-electron dirac fock ( df ) determinant @xmath10 .
the @xmath11 ionized determinants @xmath12 are obtained from the df reference state where one electron is ionized from the orbital @xmath13 . here
the labels @xmath14 denote holes and @xmath15 denote particles . according to thouless@xcite , the orbital relaxation corrections can be introduced through an exponential transformation involving one body operator of the form : @xmath16 where @xmath17 provided @xmath18 the amplitudes @xmath19 can be uniquely determined and orbital relaxation effects may be induced through the above exponential transformation . like eq.([eq1 ] ) , @xmath20 , the combination of the determinantal states @xmath21 also transforms in a similar way and we have the new function having modified determinant with the associated modified orbital set for each determinant . we write it as @xmath22 with @xmath23 where @xmath24 s are the coefficients of @xmath25 and the @xmath26 s are the projectors onto the @xmath27 .
now our aim is to define an expansion which is analogous to eq.([eq4 ] ) .
to do so , we introduce a valence universal wave operator @xmath0 @xcite for both neutral and the ( n-1 ) electron ionized model space states , satisfying the fock - space bloch equations @xcite@xmath28 where @xmath29 and @xmath30 are the projectors onto @xmath31 and @xmath32 , respectively .
we invoke a _
normal ordered _ exponential ansatz @xcite for @xmath0 as @xmath33 that satisfies @xmath34 the cluster operators @xmath5 and @xmath35 appearing in eq.([eq7 ] ) are defined as @xmath36 and @xmath37 here @xmath38 denotes normal ordering with respect to @xmath31 . due to the normal
ordering , @xmath35 acting on @xmath27 will give zero unless the determinantal state has a hole in @xmath13 .
for convenience , we express the cluster operator @xmath35 in terms of correlation ( @xmath39 ) and relaxation ( @xmath40 ) operator as @xmath41 where @xmath42 and @xmath43 figures ( [ fig-1]a)-([fig-1]b ) represent the hugenholtz diagrams corresponding to the terms given in eq.([eq12 ] ) and figure ( [ fig-2]b ) represents eq.([eq13 ] ) for @xmath44 .
since @xmath45 and higher powers of @xmath35 contains more than one valence destruction operator @xmath46 and contraction among @xmath35 operators is not allowed , eq.([eq7 ] ) reduces to @xmath47 it is evident from eq.([eq14 ] ) that @xmath48 can not induce relaxation corrections via thouless - type exponentiation .
let us now consider expansion of @xmath0 where @xmath0 is an ordinary exponential ansatz , i.e. , @xmath49 substituting eq.([eq11 ] ) in eq.([eq15 ] ) , we get @xmath50.\label{eq16}\ ] ] this expansion looks similar to the normal ordered expansion with the extra property that the contractions like @xmath51 involving the orbital @xmath13 is allowed .
it can easily shown that in the above expression ( eq.[eq16 ] ) , terms like @xmath52 , @xmath53 etc .
do not contribute because these terms contain more than one valence destruction operator @xmath46 .
introducing an operator @xmath54 through the relation @xmath55 we can rewrite eq.([eq16 ] ) in normal order as @xmath56 eq.([eq18 ] ) clearly indicates that the ordinary exponential ansatz @xmath6 is capable of inducing the orbital relaxation effects through thouless - type exponentiation .
the relaxation effects arising from the contraction of terms like @xmath51 are shown in figure [ fig-2 ] where in ( [ fig-2]a ) @xmath57 and in figure ( [ fig-2]b ) @xmath58 . to compute the energy difference
, we insert eq.([eq14])/eq.([eq18 ] ) into eq.([eq6 ] ) and proceed hierarchically from @xmath59 to @xmath60 . for @xmath59 , eq.([eq6 ] ) reduces to closed shell cc equation for core - cluster amplitudes @xmath5 . for @xmath60 , we premultiply eq.([eq6 ] ) by @xmath61 , introduce a connected dressed operator @xmath62 and obtain the bloch equation for energy difference @xmath63 where @xmath64 hugenholtz type diagrams to show the relaxation effects ] diagrammatic representation of the cluster operator @xmath35 ]
the results obtained by our calculation of ips of ne , ar and kr using two different approaches are presented in table [ ip - table ] .
figures [ ip - ne ] , [ ip - ar ] and [ ip - kr ] show the relative error for all the states of ne , ar and kr respectively , where as the figure [ ip-1s ] shows the relative error in ip for @xmath66 states explicitly of the systems studied . from table
[ ip - table ] and figures [ ip - ne ] , [ ip - ar ] and [ ip - kr ] it is clear that the agreement of ips obtained by the @xmath6 theory is better than @xmath7 theory for the core states .
this nature can be interpreted in terms of the different many - body effects , namely relaxation and correlation effects . for deep lying cores ,
the relaxation effects are stronger than the correlation effects , which is taken care of by the @xmath6 theory as compared to the @xmath7 theory ( fig.[fig-1 ] , [ fig-2 ] ) . where as for high - lying core orbitals the relaxation effects are comparatively less than the deep core orbitals like @xmath67 and @xmath68 . and thus the results obtained by the @xmath7 theory performs better than the @xmath6 theory for high lying states like @xmath69 and @xmath70 .
these can be interpreted in a better way if we look at the thouless s
treatment @xcite .
when the relaxation is very strong , a perturbative treatment is very cumbersome in higher orders and also inefficient .
the widely used normal ordered cluster expansion approach is not compact enough to include the relaxation effects in an adequate manner in the theories to calculate ip , ea , ee and dip . to overcome this problem
we have explored the generalized thouless s treatment and have calculated the results by using the @xmath6 theory .
.[ip - table]ionization potentials ( in ev ) of ne , ar and kr atoms .
entries within parentheses are from ucc(3 ) . [ cols= " < , > , > , > , > , > , > " , ] in table [ ip - table ]
the results in the parentheses are obtained from ucc(3 ) and it is clear that for these kind of problem the introduction of the unitary excitation operator does not improve the results much .
the details and the results using ucc(3 ) @xcite will be communicated shortly .
if we minutely observe the results for kr we can see that the relaxation and correlation effects are canceled out for almost all the states . in other words , these two effects are compensated by each other .
this becomes more clear if we see the koopman s energy for this system given in table [ ip - table ] . besides the experimental values , ips from @xmath71-scf
are also listed in table [ ip - table ] .
the @xmath71-scf ips are computed using grasp programme @xcite . in this procedure
, the electron binding energy corresponds to the @xmath13th orbital is obtained from @xmath72 where @xmath73 and @xmath74 are the scf energies of @xmath9 and @xmath11-electron systems respectively .
the present calculation shows that @xmath71-scf estimates the core ip more accurately than the valence ip .
this is in accordance with our expectation .
note that the differential electron correlation which is important for valence ip can not be incorporated through @xmath71-scf procedure .
since the orbital relaxation effect can be introduced through @xmath71-scf procedure , this yield reasonably accurate core ips .
% error for different states in computing ip of ne using different methods ] % error for different states in computing ip of ar using different methods ] % error for different states in computing ip of kr using different methods ] % error for different @xmath66 states in computing ip of ne , ar and kr using different methods ]
in this paper we have presented the calculation of ionization potentials for core and valance orbitals of some rare gas atoms , namely ne , ar and kr . we have explored both the @xmath6 and @xmath7 theory to obtain the results .
along with the normal cluster expansion ansatz we have also obtained the results by using unitary cluster operator theory ( ucc ) for both the formalisms . the close comparison between the results obtained by the two different approaches are explored and discussed in details .
the figures containing the computational errors for different theories help us to understand the trends of the obtained results .
this calculation clearly demonstrate the importance of different many - body effects , namely differential electron correlation and orbital relaxation effects in determining ionization potential of closed shell atoms .
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ * acknowledgments * : one of the author ( cs ) acknowledge brns for project no .
the computation was carried out on our group s xeon pc cluster . _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 10 r. chaudhuri , s. guha , d. sinha and d. mukherjee , lec .
notes in chemistry , * 52 * , ed .
u. kaldor , springer verlag , berlin , 1989 .
s. pal , m. rittby , r. j. bartlett , d. sinha and d. mukherjee , j. chem .
phys * 88 * , 4357 ( 1988 ) s. pal , m. rittby , r. j. bartlett , d. sinha and d. mukherjee , chem .
. lett . * 137 * , 273 ( 1987 ) .
d. sinha , s. k. mukhopadhyay and d. mukherjee , chem .
. lett . * 125 * , 213 ( 1986 ) .
d. mukherjee and s. pal , advn .
chem . * 20 * , 292 ( 1989 ) and the references therein .
i. lindgren and d. mukherjee , phys . rep . *
151 * , 93 ( 1987 ) and the references therein .
u. kaldor , in .
symp . * 20 * , 445 ( 1986 ) .
u. kaldor , j. chem . phys .
* 87 * , 467 ( 1987 ) .
i. lindgren , int .
s12 * , 33 ( 1978 ) .
d. mukhopadhyay , r. chaudhuri and d. mukherjee , chem .
. lett . * 172 * , 515 ( 1990 ) .
d. j. thouless , nucl . phys . * 21 * , 225 ( 1960 ) .
w. kutzelnigg , j. chem .
phys * 77 * , 3081 ( 1982 ) .
http://www.webelements.com c. sur , r. k. chaudhuri , b. k. sahoo , b. p. das and d. mukherjee , ( to be communicated ) and the references therein . k. g. dyall _ et al _ , comp .
phys . comm .
* 55 * , 425 ( 1989 ) . | in this paper we have applied the cluster - expansion ansatz for the wave operator @xmath0 which incorporates the orbital relaxation and correlation effects in an efficient manner .
we have used both ordinary and normal ordered cluster operator ( @xmath0 ) to compute the ionization potential for different states of ne , ar , and kr . the result in our calculation clearly shows how the relaxation and correlation effects play the role in determining the ionization potential , one of the interesting aspects in theoretical spectroscopy .
* pacs number(s ) .
* : 31.10.+z , 31.15.ar , 32.10.hq , 31.15.dv |
the first generation of stars are believed to have had a significant impact on the cosmic evolution through their radiative , mechanical and chemical feedback ( e.g. , * ? ? ?
stellar extreme - ultraviolet ( euv ) photons ( @xmath5 ev ) ionize the circumstellar gas to form an hii region , thereby initiating the process of the reionization of the intergalactic medium ( igm ) . even the single supernova ( sn ) of a massive ( @xmath6 ) first - star is able to totally evacuate the gas and suppresses subsequent star formation in the same minihalo @xcite although some of the materials blown out by the sn of a less - massive ( @xmath7 ) progenitor may fall back again and commence another episode of star formation @xcite .
the ejected gas by a sn contributes to the metal enrichment of the igm .
the extent of such feedback depends crucially on the initial mass function ( imf ) , as well as the star formation rate ( sfr ) , of first stars .
stars formed from the metal - free primordial gas , or population iii ( pop iii ) stars , have been classified into two sub - populations depending on whether the gas is still in an undisturbed pristine state with a low electron fraction @xmath8 ( pop iii.1 ) , or the gas has already been affected by the radiative or kinematic feedback from pre - existing stars ( pop iii.2 ) , following the nomenclature proposed by oshea et al .
despite with the same primordial composition , the thermal evolution of the gas is fairly different in those cases , resulting in different characteristic stellar masses . in the standard @xmath9cdm cosmology ,
pop iii.1 stars are predicted to be formed in minihalos of @xmath10 @xcite . during its protostellar collapse ,
cooling is entirely by @xmath11 , which is produced by the following electron - catalyzed reaction ( so - called the @xmath12 channel ; * ? ? ?
* ; * ? ? ?
* ) : @xmath13 @xmath14 with the high value of the minimum temperature @xmath15 attained at the density @xmath16 , the fragmentation mass is as large as @xmath17 @xcite , which is set by the jeans mass at this epoch . in this case ,
the final mass of formed stars is set by the radiative feedback from the growing protostar , and its value is a few 10 - a few 100 @xmath18 , much higher than the present - day counterpart @xcite .
on the other hand , pop iii.2 stars are formed from the metal - free gas affected either by the radiative or kinematic feedback from earlier generation of stars , typically in atomic - cooling halos with virial temperatures @xmath19 @xcite .
an example of pop iii.2 stars is those formed in the relic hii regions of defunct pop iii stars @xcite . in this case ,
a larger amount of @xmath11 is produced ( @xmath20 ) via the @xmath12 channel ( eqs .
[ eq : h-1 ] and [ eq : h-2 ] ) than in the pop iii.1 case ( @xmath8 ) ( e.g. , * ? ? ?
once the temperature plummets below @xmath21 by this @xmath11 cooling , deuterium is rapidly converted into hd via the exothermic reaction , @xmath22 this cooling by hd lowers the temperature further to @xmath23 ( e.g. , * ? ? ?
the characteristic mass of stars formed in such an environment is predicted to be @xmath24 @xcite .
that is , pop iii.2 stars may be typically less massive than preceding pop iii.1 stars although they are more massive than ordinary present - day stars . with high effective temperatures ( @xmath25 ) @xcite ,
massive pop iii stars emit a copious amount of far - uv ( fuv ) photons in the lyman - werner ( lw ) bands ( @xmath26 ev ) .
the fuv radiation field photodissociates @xmath11 , thereby prohibiting efficient cooling in the primordial gas .
even a single massive pop iii star emits enough fuv photons to suppress the subsequent star formation in the same halo @xcite .
in addition , the gas in a minihalo is fairly vulnerable to the external irradiation of fuv photons and thus the number of pop iii.1 stars formed can be severely regulated by the fuv background ( e.g. , * ? ? ?
on the other hand , pop iii.2 star formation in more massive halos still goes on by atomic cooling even under the fuv background and is not strongly affected .
( haiman et al .
1997 ; * ? ? ?
this indicates the possibility that pop iii.2 stars can be the major population of primordial stars and significant sources of radiative and chemical feedback to subsequent star formation @xcite .
moreover , the pop iii.2 star - formation epoch may have lasted until redshifts as low as @xmath27 @xcite .
if so , gamma - ray bursts ( grbs ) or sne of pop iii.2 stars are detectable with future facilities , which allow us to peer into the nature of primordial stars @xcite .
later , however , it was found that hd formation is strongly suppressed in the presence of even a weak fuv radiation field ( yoshida et al .
2007b ; * ? ? ?
* ) since the slight photodissociation of @xmath11 causes the significant reduction of the hd fraction . without hd cooling , the thermal evolution of a pop iii.2 star - forming gas becomes identical to the pop iii.1 case , resulting in pop iii.2 stars having similar masses to pop iii.1 stars .
however , this is not yet the conclusive result since the previous studies have not properly taken account of the following effects , which favors hd formation / cooling . in the galaxy formation epoch , shocks occur ubiquitously , for example , associating with the virialization of halos or sn explosions . in a shocked gas ,
the enhanced ionization degree enables efficient h@xmath28/hd formation and cooling @xcite . for a strong shock ,
the post - shock gas is compressed isobarically with its cooling , i.e. , the temperature decreases with increasing the density ( e.g. , * ? ? ?
on the other hand , for the free - falling cloud in a relic hii region , rapid cooling makes the temperature to plummet almost isochorically , i.e. , the temperature decreases at a constant density ( yoshida et al .
2007b ; * ? ?
* ; * ? ? ?
* ) . with the higher column density ,
the fuv radiation field is more effectively shielded for the post - shock flow than in the relic hii region , resulting in the lower fuv flux in the former case . nonetheless , in the previous studies ( e.g. , * ? ? ?
* ) , the post - shock flows are treated without the proper account of the isobaric nature of evolution .
associated with star - formation activities , not only the fuv background , but also the cosmic ray ( cr ) background is expected to be generated by way of the cr acceleration in sn remnants .
cr irradiation promotes @xmath11 and hd formation / cooling by enhancing the ionization degree @xcite .
thus , cr irradiation is able to counteract with the negative feedback of fuv radiation .
however , only either fuv or cr irradiation has been considered in the previous studies @xcite and their combined effects have not been fully understood . in this paper , we study the conditions for hd cooling to become important in pop iii.2 star formation under the presence of both fuv and cr backgrounds .
we calculate the temperature and chemical evolution of ( i ) a cloud compressed by plane - parallel steady shocks @xcite , and of ( ii ) a free - falling cloud in a relic hii region .
we find that the critical value for fuv intensity above which hd cooling is quenched is about an order of magnitude higher in the former than in the latter case .
also the critical fuv intensity increases with increasing cr intensity .
we also estimate the background fuv and cr intensities from theoretically predicted values for the sfr in the high - redshift universe . by comparing the background fuv intensity with the critical value ,
we conclude that the hd mode of pop iii.2 star formation is fairly common at the galaxy formation epoch .
the rest of this paper is organized as follows . in [ sec : model ] , we describe the numerical model and initial conditions for our calculation . in [ sec : result ] , we present the results for the free - falling clouds in relic hii regions and for the clouds compressed either by the virialization shock or by the sn blast wave .
we also present the critical fuv intensity for hd cooling as a function of the cr intensity . in [ sec : estimate ] , we estimate the background fuv and cr intensities in high - redshift universe , and discuss whether the hd cooling condition is satisfied at these epochs .
after discussing the implications and uncertainties of our study in [ sec : discussion ] , we summarize our findings in [ sec : conclusion ] .
in this section , we present our model for calculating the thermal and chemical evolution of primordial clouds . here , we first describe the method of calculation for the free - falling cloud in a relic hii region and then for the shock - compressed gas in structure formation or an sn explosion .
we use the one - zone model , where we focus on the evolution of the central core for the free - falling cloud and of the coolest gas layer for the shock - compressed gas before the self - gravity becomes effective by using the plane - parallel and steady shock approximation . when a pop iii star dies and collapses directly to a black hole ( bh ) , the gas in the relic hii region surrounding the star begins to recombine and cool to form a subsequent star @xcite . here
, we assume that the formed bh has no influence on the evolution of the gas in the relic hii region for simplicity . at first
, the baryon mass in the relic hii region is too small for the gas to contract by self - gravity , but the gas loses pressure - support by efficient radiative cooling and contracts being attracted by the gravitational potential of the dm halo .
after the temperature decreases so that the jeans mass becomes lower than the hii region mass , the gas begins to contract in the almost free - fall way by self - gravity .
owing to its high pressure , the gas in the hii region is accelerated monotonically with radius up to a supersonic velocity and expands into the interstellar medium ( whalen et al .
2004 ; kitayama et al .
. however , our model does not take into account the initial velocity profile within the hii region for simplicity .
the following prescription is the same as the previous studies ( johnson & bromm 2006 ; yoshida , omukai & hernquist 2007 ; wolcott - green & haiman 2011 ) .
an isolated cloud collapsing at the rate close to free - fall by its self - gravity develops a core - envelope structure @xcite , consisting of a nearly jeans - scale core with the constant density and an envelope where the density decreases radially as @xmath29 . in our one - zone model
, we follow the evolution of the physical quantities at the core center @xcite .
since collapse proceeds roughly at the free - fall rate , the density evolves as @xmath30 where @xmath31 is the free - fall time and @xmath32 the gravitational constant .
temperature evolution follows the energy equation : @xmath33 where @xmath34 is the internal energy per unit mass , @xmath35 the pressure , @xmath36 the net cooling rate per unit volume . to supplement the equations above
, the ideal - gas equation of state @xmath37 is used , where @xmath38 is the boltzmann constant , @xmath39 the proton mass and @xmath40 the mean molecular weight . using the ratio of specific heat @xmath41 , specific
energy is related to pressure as @xmath42 . throughout this paper
, we take @xmath43 and @xmath44 neglecting molecular contribution .
considering that the size of the contracting core is half the jeans length @xmath45 , we calculate the column density of the core as @xmath46 @xcite . in this case , the initial size of this region is estimated as @xmath47 kpc , which is an overestimation compared to the typical size of an hii region @xmath48 pc ( whalen et al .
2004 ; kitayama et al .
. however , at the beginning of collapse , owing to the low gas density and h@xmath28 fraction , the h@xmath28 column density will be too low for effectively shielding the gas from the fuv filed , and the above overestimation will not matter for estimating the shielding factor . in our model of free - fall collapse , the collapse equation and the energy equation ( eqs .
[ eq : drho_dt ] , [ eq : energy ] ) are decoupled and feedback from thermal pressure is not included for simplicity . in the runaway - collapse phase , an isothermal cloud collapses approximately at a free - fall rate @xcite .
although the pressure gradient is not negligible , the deviation of the collapse time from free fall is only by a factor of 1.58 @xcite . for a non - isothermal case ,
the deviation is represented as a monotonically increasing function of the effective adiabatic index @xmath41 of the cloud in equations 7 - 9 of @xcite . for example , the collapse time becomes @xmath49 times longer than the free - fall time for @xmath50 .
if we use this generalized collapse time for the collapse equation , the collapse equation 4 is coupled with the energy equation 5 . however , we find that the effective adiabatic index becomes smaller than 1.2 for most of the evolutionary trajectory , and the deviation from free - fall may be at most by a factor of a few .
thus , feedback from thermal pressure may change our results little . here , we consider that an overdense region collapsing within an hii region is large enough . then , since the overdense region is strongly bounded by the gravitational field of dm and gas , it contracts by the gravitational force which exceeds the pressure force from its less - dense surroundings .
therefore , the compression of the overdense region by external pressure may be neglected , and it will collapse roughly at a free - fall rate . although our formulation does not include dm gravity , it becomes significant at the beginning of gas evolution .
then , we recalculate the evolution of the gas in relic hii regions taking account of dm gravity by modifying @xmath51 in @xmath52 of eq . to @xmath53 , but we find little change in our results . here , @xmath54 is the mean dm mass density of a virialized halo at @xmath55 . after the passage of a blast wave , which is generated in structure formation or an sn explosion , the post - shock gas has a very high temperature due to shock heating and it cools and contracts with the radiative cooling time . at first , self - gravity is not effective in the shocked gas layer .
we focus on the evolution of the most cooled gas layer , assuming that the post - shock gas layer is plane - parallel and the flow is steady @xcite .
the initial condition of a shock - compressed gas is characterized by two parameters : the pre - shock density @xmath56 and the velocity of the shock front @xmath57 . in the strong shock limit ,
the physical quantities just behind the shock front are given by the rankine - hugoniot relation : @xmath58 @xmath59 @xmath60 @xmath61 where the subscripts 0 and 1 refer to the pre- and post - shock quantities , respectively .
behind the shock front , the flow evolution is described by @xmath62 @xmath63 along with the energy equation ( [ eq : energy ] ) .
the column density of the post - shock gas can be estimated from @xmath64 where @xmath65 is the number density in the pre - shock gas and @xmath66 is the time since the shock occurs .
rapid cooling in the post - shock layer leads to the formation of a dense gas shell , which eventually becomes gravitationally unstable and fragments . for the sn shock , we assume that fragmentation occurs when the column density of the post - shock layer @xmath67 exceeds that in the jeans length @xmath68 @xcite : @xmath69 for the structure - formation shock , the mass of the post - shock layer is doubled compared to the sn case , since in one dimension , two flows
collide to form the shocked region and two shock fronts propagate in opposite directions to each other . in this case
, fragmentation occurs under the condition : @xmath70 in addition to them , for fragmentation , the growth time of a perturbation , which is approximately given by the local free - fall time @xmath52 , must be shorter than the contraction time of the layer , i.e. , the cooling time @xmath71 ( yamada & nishi 1998 ) .
we have confirmed this condition is always satisfied at the fragmentation epoch determined by the column - density conditions given above . after fragmentation , each fragment or clump begins to contract by its self - gravity .
we also follow its evolution by the same one - zone model as in sec .
[ subsec : free - fall_form ] . the net cooling rate @xmath36 in eq .
consists of the following terms : @xmath72 where @xmath73 , @xmath74 and @xmath75 are the radiative cooling rates by h lyman-@xmath76 line emission , @xmath11-line emission and hd - line emission , respectively , and @xmath77 is the cr heating rate .
we adopt the cooling rates @xmath73 from @xcite , @xmath74 from @xcite , and @xmath75 from @xcite . since each cr ionization associates with heating of @xmath78 @xcite , @xmath79 where @xmath80 is the hydrogen ionization rate by crs , and @xmath81 is the number density of hydrogen atoms . at densities we consider ( @xmath82 ) ,
the contribution from chemical cooling and heating is small and is neglected in this paper @xcite .
our chemical network includes following 11 species : h , @xmath11 , @xmath34 , @xmath83 , @xmath84 , @xmath12 , d , hd , @xmath85 , @xmath86 and @xmath87 .
the deuterium fraction is set to @xmath88 , and the he chemistry is neglected since it is thermally inert in the temperature range we consider .
chemical reactions include 18 hydrogen collisional reactions in table 1 of @xcite , 18 deuterium collisional reactions in table 1 of @xcite , photo - dissociation of h@xmath28 , hd , and h@xmath89 , and cr ionization .
for collisional reactions , we use the updated rate coefficients by @xcite .
the photo - dissociation rate coefficients are taken from @xcite and @xcite , and the @xmath11 and hd shielding factors are from @xcite . for the shielding factors ,
we take account of the doppler shift due to the bulk motion of the cloud as in @xcite for the free - falling case and in @xcite for the shock - compressed case .
the stellar radiation field is assumed to have the diluted black - body spectrum with the temperature of @xmath90 k , resembling that from a pop ii star cluster , and its intensity is parameterized by @xmath91 , where @xmath92 is the intensity at the lyman limit .
the cr ionization rate of neutral hydrogen @xmath80 is treated as a free parameter , since the cr intensity at the high - redshift universe is highly uncertain .
however , the cr intensity may be related with the star - forming activity , and the sfr at @xmath93 is theoretically estimated to be no more than the present - day value ( e.g. , tornatore et al .
2007 ; trenti & stiavelli 2009 ; johnson et al . 2013 ) .
we thus consider the cr intensity in the range of @xmath94 , which are smaller than the galactic value @xmath95 @xcite . in the relic hii region of a defunct pop iii star ,
a primordial cloud begins to collapse from a highly ionized state .
the temperature and density in the relic hii region are almost uniform with @xmath96 and @xmath97 ( kitayama et al .
2004 ; whalen et al . 2004
; abel , wise & bryan 2007 ; yoshida et al .
we thus adopt @xmath98 and @xmath99 as the fiducial initial values , and also study the cases with higher and lower densities @xmath100 to see the dependence on the initial density .
we set the initial chemical abundances as @xmath101 , @xmath102 , @xmath103 , and @xmath104 for other species , where @xmath105 is the fractional abundance of the @xmath106-th species to the hydrogen nuclei .
we consider two kinds of shocks which occur in the virialization of halos and sn explosions . here
, we need to specify the shock velocity @xmath57 , pre - shock density @xmath65 , and initial chemical abundance @xmath107 for each case . for the virialization shock , the mean gas number density within the virialized halo is estimated as @xcite @xmath108 however , according to the recent studies on first galaxy formation , the shock front does not sustain at the virial radius and its location shrinks inward owing to efficient ly-@xmath76 emission @xcite .
protogalaxies accrete gas via cold flows in dense filaments , with the infall velocity of @xmath109 .
such cold flows penetrate deep inside the halo and collide each other to form virialization shocks with pre - shock densities @xmath110 , much higher than in eq . .
considering this effect , we adopt @xmath111 as the fiducial values for the pre - shock gas .
the shocked gas behind a sn blast wave cools and is compressed significantly after the sn remnant enters the pressure - driven expansion phase @xcite .
the shock velocity early in this phase is @xmath112 ( machida et al .
2005 ; nagakura et al .
2009 ; chiaki et al .
the pre - shock density is comparable to the density in relic hii regions @xmath97 , since core - collapse sne explode within hii regions which their progenitors produce .
hence , we adopt @xmath113 as the fiducial values for the sn shock .
@xcite and @xcite studied the chemical composition in the primordial gas just behind a shock front .
we adopt their results as the initial composition in the shock - compressed gas , which are summarized in table [ tab : shock_comp ] for the virialization and sn shocks .
the fractions of species not presented in table [ tab : shock_comp ] are set to zero , except for the electron fraction @xmath103 .
.initial chemical composition in the shock - compressed gas [ cols="^,^,^,^,^",options="header " , ] [ tab : shock_comp ]
here , we present the results for the free - falling cloud in a relic hii region .
we also find the critical fuv intensity above which hd cooling is suppressed for the different values of the cr intensity .
first , we see the results without cr irradiation to extract the fuv effect alone . fig .
[ fig : h2_nt_cr0 ] shows the temperature evolution , and fig .
[ fig : h2_nt_cr0_chem ] shows the fractions of ( a ) @xmath11 , ( b ) electron , and ( c ) hd , for five different fuv intensities : @xmath114 , and @xmath115 , both as a function of the number density @xmath116 .
the temperature first drops isochorically to @xmath117 via efficient ly-@xmath76 cooling in all the cases in fig .
[ fig : h2_nt_cr0 ] , and the subsequent behavior depends on the fuv intensities .
the same results are obtained in johnson & bromm ( 2006 ) , yoshida , omukai & hernquist ( 2007 ) , and wolcott - green & haiman ( 2011 ) . in the absence of fuv irradiation ( @xmath118 ) , abundant @xmath11 with the fraction @xmath20 forms via the h@xmath119 channel ( eqs .
[ eq : h-1 ] and [ eq : h-2 ] ) owing to the high initial ionization degree ( solid line in fig .
[ fig : h2_nt_cr0_chem]a ) .
the cooling by this enhanced h@xmath28 makes the temperature to fall below @xmath120 , the minimum value attainable in the primordial pristine gas , where the ionization degree is only @xmath8 @xcite .
once the temperature reaches @xmath121 , the exothermic reaction ( eq . [ eq : hd ] ) converts most of the deuterium into hd ( solid line in fig .
[ fig : h2_nt_cr0_chem]c ) .
hd cooling further drops the temperature to the minimum value @xmath122 at density @xmath123 , which is the critical density for hd to reach local thermodynamic equilibrium . toward higher density , the temperature increases gradually by compressional heating .
at the temperature minimum , the parent cloud is considered to fragment into clumps with the jeans mass of @xmath124 ( yoshida et al .
2007b ; * ? ? ?
we call hereafter this mode of star formation as _ hd - pop iii star formation _ , in contrast to
_ h@xmath28-pop iii star formation _ , where h@xmath28 is the sole coolant during proto - stellar collapse . with fuv irradiation
, the temperature becomes higher owing to the photodissociation of the coolants , h@xmath28 and hd ( fig .
[ fig : h2_nt_cr0_chem]a , c ) . with @xmath125 as low as 0.01 , the evolutionary track is similar to that of @xmath118 , although with slightly higher temperature . with intensity @xmath126 , the @xmath11 fraction is reduced to @xmath8 , and the temperature does not reach the regime of efficient hd formation / cooling ( short - dashed line in fig .
[ fig : h2_nt_cr0_chem]a , c ) . with the even higher fuv intensity @xmath127
, photodissociation strongly reduces the h@xmath28 fraction at low densities ( dotted and dot - dashed lines in fig .
[ fig : h2_nt_cr0_chem]a ) . the clouds contract isothermally at @xmath117 by ly-@xmath76 cooling until @xmath128 for @xmath129 ( @xmath130 for @xmath131 , respectively ) , where the h@xmath28 self - shielding against fuv radiation becomes effective .
thereafter , the temperature drops first almost vertically to @xmath132 k and then gradually to @xmath133 by @xmath11 cooling . in the cases with @xmath134 , without hd cooling
, the gas cools only to the minimum temperature @xmath135 by h@xmath28 cooling at @xmath136 as in the case of pop iii.1 star formation .
namely , star formation proceeds in the h@xmath28-pop iii mode . in this mode , the clouds fragment into rather massive clumps of @xmath137 .
note that , in the high density regime ( @xmath138 ) , all the evolutionary tracks converge to either of the two tracks , one by hd cooling for @xmath139 or the other by @xmath11 cooling for @xmath134 . .
different lines correspond to the different fuv intensities : @xmath140 , and @xmath115 from bottom to top .
the thin dashed lines indicate the constant jeans masses . ] , ( b ) @xmath34 , and ( c ) hd in the same clouds as in fig . [
fig : h2_nt_cr0 ] . in each panel ,
different lines correspond to the different fuv intensities : @xmath141 ( solid ) , @xmath142 ( long - dashed ) , @xmath143 ( short - dashed ) , @xmath144 ( dotted ) , and @xmath115 ( dot - dashed ) . ]
next , we see the results with cr irradiation on cloud evolution , as shown in figs . [
fig : h2_nt_j01 ] and [ fig : h2_nt_j01_chem ] , for the cases with fuv intensity @xmath126 .
the different lines in the figures show different values of the cr intensity : @xmath145 ( i.e. , @xmath146 ) , @xmath147 and @xmath148 . recall that , with @xmath149 , hd cooling is suppressed without cr irradiation . as seen in fig .
[ fig : h2_nt_j01 ] , with higher @xmath80 , the temperature becomes lower .
this is because the elevated ionization degree by cr ionization causes more efficient h@xmath28 formation and cooling ( see fig .
[ fig : h2_nt_j01_chem ] ) , which outweighs the effect of cr ionization heating . for @xmath150
, the temperature falls below 150 k , where the hd fraction increases significantly , owing to the enhanced @xmath11 cooling ( solid and long - dashed lines in fig .
[ fig : h2_nt_j01_chem]c ) . by hd cooling
, the temperature further drops to @xmath122 , and hd - pop iii star formation is realized in these cases . in this way , cr ionization can compensate the negative feedback of fuv photodissociation . , and the initial density to @xmath99 .
different lines correspond to the different cr intensities : @xmath151 , and @xmath148 .
the thin dashed lines indicate the constant jeans masses . ] , ( b ) @xmath34 , and ( c ) hd fractions for the same clouds as in fig .
[ fig : h2_nt_j01 ] . in each panel ,
the different lines correspond to the different values of cr intensity : @xmath152 ( dot - dashed ) , @xmath153 ( dotted ) , @xmath154 ( short - dashed ) , @xmath155 ( long - dashed ) , and @xmath148 ( solid ) . ] since in our model for the free - falling cloud , the collapse equation ( eq . [ eq : drho_dt ] ) is decoupled from the energy equation ( eq . [ eq : energy ] ) , the collapse time becomes the same for the same initial density and is obtained as @xmath156 myr for @xmath157 .
the hubble time is estimated as @xmath158 and this is roughly equal to the galaxy merger time .
we find that the collapse timescale estimated above is smaller than the hubble time for @xmath159 .
moreover , in the early universe we consider , the halo does not have an angular momentum large enough to form a rotation - supported disk which is comparable to the protogalaxy scale ( abel et al .
thus , the gas cloud in a relic hii region may remain isolated without being affected by the protogalaxy rotation and merger , and star formation there proceeds within the local hubble time .
we have learned that with the fuv intensity exceeding a critical value @xmath160 , hd cooling is totally suppressed even in initially ionized clouds .
in addition , the value of the critical fuv intensity is elevated under cr irradiation . in fig .
[ fig : h2_zeta_jcrit ] , the critical fuv intensity @xmath160 is plotted as a function of @xmath80 for initial densities @xmath161 ( dashed ) , 0.3 ( solid ) and 3 ( dot - dashed)@xmath162 . in finding the value of @xmath160
, we judge hd cooling is effective if the following two conditions are satisfied : ( i ) the minimum temperature at @xmath163 is less than 100 k , and ( ii ) the hd - cooling rate exceeds a half of the total ( i.e. the sum of compressional and cr ) heating rate at some moment .
these two conditions are imposed to discriminate whether the high - density temperature tracks are either those by h@xmath28 cooling or by hd cooling . in the fiducial case of @xmath164 ( solid line ) ,
the critical intensity is @xmath165 for the negligible cr intensity ( @xmath166 ) , in accord with previous studies ( yoshida et al .
2007b ; * ? ? ?
the critical intensity @xmath160 increases with the cr strength @xmath80 , since strong enough cr irradiation enables efficient hd cooling even under an intense fuv field . in the weak cr regime @xmath167 , the electrons present initially , rather than those produced by cr ionization , work as catalysts for the h@xmath28 formation reaction .
abundant h@xmath28 is formed if the self - shielding against the fuv field becomes effective before the ionization degree is reduced significantly by recombination .
since the self - shielding is more effective for larger column densities and thus for larger @xmath168 , more h@xmath28 is formed under the same fuv intensity .
this results in higher @xmath160 for higher @xmath168 . on the other hand , in the strong cr regime @xmath80 @xmath169 ,
cr ionization provides the necessary electrons for abundant h@xmath28 formation .
moreover , with a fuv field as high as @xmath160 , the shielding against the fuv field becomes effective and @xmath11 formation occurs only after the density is significantly increased from the initial value .
therefore , the evolutionary track and critical fuv intensity @xmath160 do not depend on the initial density . for hd cooling as a function of the cr intensity @xmath80 .
each curve corresponds to the different initial densities : @xmath161 ( dashed ) , 0.3 ( solid ) and 3 ( dot - dashed)@xmath162 . below @xmath160 ,
hd dominates the cooling in the course of cloud collapse .
the shaded region shows the background fuv and cr intensities estimated from eqs . and with @xmath3 and @xmath170 .
the blue thin - solid line shows the evolution of the background intensities evaluated using the pop ii sfr @xmath171 of johnson et al .
the blue points refer to values at @xmath172 and @xmath173 from left to right . ] here , we present the results for a shock - compressed gas in the two cases of the shock velocity : ( i ) @xmath174 ( virialization shock ) and ( ii ) @xmath175 ( sn shock ) , respectively .
first , to see the fuv effect on the gas compressed by the virialization shock ( @xmath176 ) , the results are shown for the cases without cr irradiation in figs .
[ fig : shock20_nt_cr0 ] and [ fig : shock20_nt_cr0_chem ] . just behind the shock front
, the gas is heated to @xmath19 and then cools immediately to @xmath117 by ly-@xmath76 cooling . with the negligible or weak fuv field ( @xmath177 ) , h@xmath28 forms abundantly owing to the high ionization degree ( @xmath178 ) in the post - shock gas ( solid and long - dashed lines in fig .
[ fig : shock20_nt_cr0_chem]a , b ) . as in the relic hii region
, the gas enters the hd formation regime ( @xmath179 k ) by h@xmath28 cooling ( solid and long - dashed lines in fig .
[ fig : shock20_nt_cr0_chem]c ) , and the temperature further decreases to @xmath23 by hd cooling .
as long as the cooling time is shorter than the free - fall time , the post - shock gas is compressed almost isobarically ( e.g. , * ? ? ?
* ) . however , below @xmath122 , hd cooling is not efficient anymore , and the cooling time exceeds the free - fall time .
when the sufficient mass for gravitational instability accumulates in the dense layer , it fragments into clumps of approximately the jeans mass @xmath1 , which then continue collapsing by hd cooling in the free - fall manner .
this late - time temperature evolution is along the hd - cooling track , like the free - falling clouds in relic hii regions ( @xmath141 in fig .
[ fig : h2_nt_cr0 ] ) . for @xmath126 , the isobaric contraction proceeds in the same way until @xmath23 as with the weaker fuv case .
however , during the dense gas layer stays at @xmath23 until the jeans instability sets in , the hd fraction is significantly reduced by electron recombination and h@xmath28/hd photodissociation ( short - dashed lines in fig . [ fig : shock20_nt_cr0_chem]c ) . as a result , hd cooling does not become important in the subsequent free - fall phase and temperature evolution eventually converges to the h@xmath28-cooling track . for @xmath127 , strong fuv radiation completely quenches hd cooling and
massive clumps of @xmath180 will be formed .
especially , in the case of @xmath181 , fragmentation takes place early at @xmath182 .
the subsequent free - fall evolution is similar to that of the free - falling clouds in relic hii regions with @xmath127 ( fig .
[ fig : h2_nt_cr0 ] ) . and @xmath174 .
different lines correspond to the different fuv intensities : @xmath140 , and @xmath115 . ] , ( b ) @xmath34 , and ( c ) hd in the same clouds as in fig .
[ fig : shock20_nt_cr0 ] .
different lines correspond to the different fuv intensities : @xmath141 ( solid ) , @xmath142 ( long - dashed ) , @xmath143 ( short - dashed ) , @xmath144 ( dotted ) , and @xmath115 ( dot - dashed ) . ] next , we see the cr effect for the case with fuv intensity @xmath183 . the results are shown in figs .
[ fig : shock20_nt_j1 ] and [ fig : shock20_nt_j1_chem ] . recall that
, with @xmath184 , hd cooling is suppressed in the absence of cr irradiation . as in the case of the relic hii region ,
more @xmath11 is formed by cr ionization ( fig .
[ fig : shock20_nt_j1_chem]a , b ) and the temperature becomes lower by the enhanced h@xmath28 cooling . with the cr strength exceeding @xmath185
, most deuterium is converted into hd ( long - dashed and solid lines in fig .
[ fig : shock20_nt_j1_chem]c ) , and temperature falls below @xmath186 k. thermal evolution finally converges to the hd - cooling track in the free - falling phase . .
the initial density and velocity are the fiducial values : @xmath187 and @xmath174 .
the different lines correspond to the different cr intensities : @xmath188 , and @xmath148 . ] , ( b ) @xmath34 , and ( c ) hd in the same clouds as in fig .
[ fig : shock20_nt_j1 ] .
the different lines correspond to the different cr intensities : @xmath152 ( dot - dashed ) , @xmath154 ( short - dashed ) , @xmath155 ( long - dashed ) , and @xmath148 ( solid ) . ] in our model for the shock - compressed gas , the contraction equations ( eqs .
[ eq : rho_v ] , [ eq : rho_v^2 ] ) are coupled with the energy equation ( eq . [ eq : energy ] ) before the self gravity becomes effective , so that the timescale of the isobaric contraction phase differs among the lines .
however , the evolution time after self - gravity becomes effective is almost the same and the time to reach @xmath189 is @xmath190 myr for the structure - formation shock , which is by an order of magnitude smaller than the local hubble time @xmath158 for @xmath159 .
thus , the shock - compressed gas in structure formation may be unaffected by the protogalaxy rotation and merger , and star formation there proceeds within the local hubble time . in fig .
[ fig : shock100_nt_cr0 ] , we show the fuv effect on the shock - compressed gas in an sn explosion ( @xmath191 ) for cases without cr irradiation .
we find that , despite the different initial conditions , the evolutionary tracks are similar to those in the virialization shock with the same fuv intensity ( see fig . [
fig : shock20_nt_cr0 ] ) .
this is because , although the initial density @xmath192 is ten times lower in the sn case , the post - shock temperature is 25 times higher and thus the pressure in the isobaric evolution phase differs only by a factor of 2.5 ( see fig .
[ fig : shock20_nt_cr0 ] ) .
we also find that the chemical compositions vary in a similar way as in the case of virialization shock ( see fig .
[ fig : shock20_nt_cr0_chem ] ) , except that the ionization degree reaches @xmath47 due to the very high temperature @xmath25 behind the sn shock front . and @xmath193 .
the different lines correspond to the different fuv intensities : @xmath140 , and @xmath115 . ]
the same trend holds also in the cases with cr irradiation , and the evolutionary tracks are similar to those in the virialization shock with the same cr intensity ( see figs .
[ fig : shock20_nt_j1 ] and [ fig : shock100_nt_j1 ] ) . .
the fuv intensity is fixed to @xmath184 .
the different lines correspond to the different cr intensities : @xmath188 , and @xmath148 . ] for the sn - shock case , the evolution time after self - gravity becomes effective is almost the same and the time to reach @xmath189 is calculated as @xmath194 myr , which is much longer than the lifetime of an sn remnant @xmath47 myr .
thus , the sn shock will dissolve into the ism before it collects enough materials to trigger fragmentation and star formation in the post - shock region .
this result is consistent with chiaki et al .
( 2013 ) , where they find that the sn - shock - compressed gas can collapse within the lifetime of the sn remnant for the ism density @xmath195 and @xmath196 erg . however , we discuss the evolution of the sn - shock - compressed gas for comparison with that of structure formation . as in [ subsec : h2_zeta_jcrit ] , the critical fuv intensity for hd cooling @xmath160 is calculated for a shock - compressed gas and is presented in fig .
[ fig : sh_zeta_jcrit ] as a function of the cr intensity @xmath80 .
the solid and dashed lines are for the virialization shock in the fiducial case @xmath111 and in the high density case @xmath197 , respectively , and the dot - dashed line indicates the sn shock case @xmath113 . from fig .
[ fig : sh_zeta_jcrit ] , we see that for @xmath198 , the critical value @xmath160 becomes higher for a higher initial density @xmath168 or higher shock velocity @xmath57 .
the reason is the same as in the case of a relic hii region : in the weak cr regime , abundant h@xmath28/hd formation is caused by the high post - shock ionization degree , and thus the prompt fuv shielding before significant recombination is needed .
this can be realized more easily with a higher initial density @xmath168 or higher shock velocity @xmath57 , since the column density at @xmath199 k , below which recombination proceeds significantly , is higher in these cases . on the other hand , for @xmath200 , the critical value @xmath160 is solely determined by the cr strength @xmath80 and becomes independent of the initial condition @xmath201 .
again the reason is the same as in the relic - hii - region case : the electrons from cr ionization , rather than those present initially , now work as catalysts for h@xmath28 formation and thus the h@xmath28 formation rate does not depend on the initial parameters . the comparison between figs .
[ fig : h2_zeta_jcrit ] and [ fig : sh_zeta_jcrit ] tells us that in the weak cr regime , @xmath202 , the critical intensity is @xmath2 times higher for the shock - compressed gas ( @xmath203 ) than in the relic hii region ( @xmath204 ) . in this case , for abundant h@xmath28/hd formation , h@xmath28 should be promptly shielded against the fuv field before the electrons present initially recombine significantly .
this condition can be more easily satisfied in the shock - compressed gas than in the relic hii region .
the reason is that the shock - compressed gas evolves isobarically ( see fig .
[ fig : shock20_nt_cr0 ] ) rather than isochorically as in the relic hii region ( see fig . [
fig : h2_nt_cr0 ] ) , which leads to the larger column density in the shock - compressed gas .
in addition , shocks tend to occur in denser environments ( e.g. , cold - accretion flows ) than relic hii regions . consequently , in the shock - compressed gas , h@xmath28 is shielded from the fuv field more effectively , and the critical fuv intensity becomes higher than that in the relic hii region . for hd cooling as a function of the cr intensity @xmath80 in shock - compressed clouds ( see also fig . [
fig : h2_zeta_jcrit ] ) .
the solid and dashed lines correspond to the virialization shock with the fiducial case , @xmath111 , and the high density case , @xmath197 , respectively .
the dot - dashed line shows the sn - shock case , @xmath113 . ]
we have seen that pop iii.2 star formation proceeds through the hd - cooling track , i.e. , in the hd - pop iii star formation mode , under a weak fuv field or strong enough cr irradiation . in this section
, we estimate the intensities of the fuv and cr backgrounds as a function of redshift .
then , by comparing them with the critical value @xmath160 obtained in [ sec : result ] , we discuss whether hd - pop iii star formation is realized at the epoch of galaxy formation ( @xmath205 ) . at this epoch ,
metal enrichment in an average galaxy may have proceeded to the level for pop ii star formation to occur ( e.g. , wise et al .
2012 ; johnson et al .
we thus consider massive pop ii stars and their sn remnants as the plausible sources of the fuv and cr backgrounds , respectively .
then , the intensity of the fuv background is estimated as @xcite @xmath206 where @xmath207 is the planck constant , @xmath208 the speed of light , @xmath209 the average frequency of the lw band ( @xmath210 ev ) , @xmath211 the lw band width ( @xmath212 ev ) , @xmath213 the number of fuv photons emitted per stellar baryon , and @xmath214 the stellar mass density at @xmath215 .
johnson et al . ( 2013 ) evaluated @xmath214 by the relation @xmath216 , where @xmath171 is the pop ii sfr density , and @xmath217 the lifetime of a massive star .
then , eq . can be rewritten as @xmath218 following @xcite , the energy density of the background crs @xmath219 is estimated from @xmath220 @xmath221 where @xmath222 is the sn energy , @xmath223 the efficiency of @xmath224 converted to cr acceleration , @xmath225 the number of sne per unit mass of stars formed , and @xmath226 the hubble time at @xmath215 .
@xmath225 is obtained as @xmath227 assuming that massive stars with @xmath228 explode as sne for the salpeter imf .
the cr ionization rate of a hydrogen atom @xmath80 is related with @xmath229 as @xcite : @xmath230 combining eqs . and , @xmath231
the high-@xmath215 sfr density has been evaluated theoretically by some authors ( e.g. , tornatore et al .
2007 ; trenti & stiavelli 2009 ; johnson et al . 2013 ) .
according to johnson et al .
( 2013 ) , @xmath171 increases almost monotonically from @xmath232 to @xmath233 as @xmath215 decreases from @xmath234 to @xmath173 .
the blue thin - solid lines in figs . [ fig : h2_zeta_jcrit ] and [ fig : sh_zeta_jcrit ] show the evolution of @xmath235 and @xmath236 evaluated from eqs . , and @xmath171 of johnson et al .
the background intensities move from left to right along the line , and the blue points indicate the values at @xmath172 and @xmath173 , respectively .
although the high-@xmath215 sfr density is still uncertain , most studies concluded @xmath237 for @xmath238 ( e.g. , tornatore et al . 2007 ;
trenti & stiavelli 2009 ; johnson et al .
2013 ) . the shaded regions in figs .
[ fig : h2_zeta_jcrit ] and [ fig : sh_zeta_jcrit ] indicate the range of @xmath235 and @xmath236 corresponding to this parameter range .
first , we examine the case of free - falling clouds in relic hii regions ( fig .
[ fig : h2_zeta_jcrit ] ) . for initial densities
lower than the fiducial case @xmath164 ( black solid line ) , the critical fuv intensity for hd cooling @xmath160 is below the background level @xmath239 at @xmath240 , and hd cooling is suppressed in pop iii.2 star formation . as the background cr intensity @xmath80 increases with cosmic star formation , the critical intensity @xmath160 rises and eventually exceeds the background level @xmath239 at @xmath241 . below this redshift ,
hd cooling becomes efficient and hd - pop iii star formation occurs if some primordial environments still survive .
although in a very dense hii region ( dot - dashed line ) , the critical intensity @xmath160 is above the background level @xmath239 even at @xmath240 , such dense conditions appear to be hardly realized ( kitayama et al .
2004 ; whalen et al .
2004 , yoshida et al .
2007a , abel2007 ) .
we conclude that , in relic hii regions , hd - pop iii star formation is possible only at @xmath241 , where the intensity of the cr background is modest . on the other hand , for the shock - compressed gas ( fig .
[ fig : sh_zeta_jcrit ] ) , the critical value @xmath160 exceeds the background level @xmath239 for most of the redshift we consider ( @xmath3 ) , and thus hd - pop iii star formation occurs . note that the pre - shock densities we adopt in fig .
[ fig : sh_zeta_jcrit ] are rather low , and our results may give the lower bound of the critical fuv intensity @xmath160 .
in fact , in the mass assembly process of first galaxies , cold - streaming shocks occur at the regions as dense as @xmath242 ( wise & abel 2007 ; wise et al .
in such a dense shock environment , hd - pop iii star formation would be more common than the cases we have studied . in conclusion , the hd - pop iii mode is the main mode of pop iii.2 star formation in the cloud compressed either by a virialization or sn shock .
so far , we have considered that the background fuv photons are uniformly distributed in the universe . in reality , they have some spatial fluctuations caused by local stellar sources , like star - forming galaxies ( dijkstra et al . 2008 ; ahn et al . 2009 ; agarwal et al . 2012 ; johnson et al . 2013 ) . in the primordial clouds irradiated with the strong fuv field as @xmath243
, hd formation is suppressed even in the shock - compressed gas unless the initial density is high ( @xmath244 ) or the cr intensity is strong ( @xmath245 : see fig .
however , according to johnson et al .
( 2013 ) , only @xmath246 percent of the dense primordial cloud with @xmath247 is exposed to such a strong fuv field as @xmath243 .
thus , our analysis with the uniform background fuv field is enough to discuss the major character of primordial star formation and the spatial fluctuations of it do not affect our main conclusions .
we also assume that crs accelerated in sn remnants escape freely from their host galaxies and fill the universe homogeneously .
however , in reality , they may be trapped in the host galaxy by the interaction with the magnetic field , and propagate diffusively through the ism of the galaxy @xcite .
on the way of propagation , crs also lose their kinetic energy by ionizing the neutral ism .
for the efficient cr escape into the igm , the cr escape time by diffusion @xmath248 should be shorter than both the ionization loss time @xmath249 and the age of the system @xmath250 , which is roughly equal to the age of the universe . here , @xmath248 and @xmath249 depends on the kinetic energy of a cr particle @xmath251 and the redshift @xmath215 . @xcite
estimated the above timescales for crs with @xmath252 mev and discussed the cr escape efficiency during the cosmic structure formation . here , we extend their arguments by examining the @xmath251 dependence of the timescales . following @xcite
, we assume that the energy spectrum of the cr number density has a power - law shape @xmath253 with @xmath254 ev .
we find that the inequality @xmath255 holds for all the cr energy and redshift we consider ( @xmath3 ) . on the other hand , the inequality @xmath256 holds only for high energy crs with @xmath257 ev at @xmath3 .
thus , the cr background is composed of crs with @xmath257 ev and its intensity varies with the spectrum index @xmath258 . in the standard fermi acceleration theory ,
the spectrum index is given as @xmath259 ( e.g. , * ? ? ?
* ) , and the background intensity is reduced only by 10 % due to the confinement of crs with @xmath260 ev . on the other hand ,
if we adopt a steeper slope as @xmath261 in @xcite , it is reduced by an order of magnitude . even in this case , if a star - forming pristine gas resides in the same galaxy hosting sne , it is exposed to the cr intensity in the original or even the enhanced level , because of the efficient confinement within the galaxy . for more quantitive and definitive discussions ,
a detailed study of cr propagation in the ism of high-@xmath215 galaxies is needed and is beyond the scope of this paper . at the epoch of galaxy formation
, high - mass x - ray binaries and miniquasars may be the plausible sources of high energy photons extending to the x - ray band @xcite .
since the photoionization cross section of neutral atoms generally decreases with frequency in the power - law way , the igm becomes optically thin for x - ray photons with @xmath262 kev and the x - ray background develops at this epoch ( haiman et al . 2000 ) .
the injection of x - ray photons enhances the cooling efficiency of primordial clouds by indirectly promoting h@xmath28 and hd formation through the photoionization of neutral atoms ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
nonetheless , we did not include this x - ray feedback , and here we briefly discuss how it changes our results .
following @xcite , we assume that the x - ray background has a power - law spectrum @xmath263 where @xmath264 and @xmath265 is the intensity at 1 kev .
then , using eqs .
( 17 - 19 ) in @xcite , the ionization rate of h is calculated as @xmath266 for the hydrogen column density @xmath267 .
since the background field derives from the stellar activity , the x - ray intensity is related with the cr energy density as @xmath268 where we suppose that cr sources are pop ii clusters with the salpeter imf and the mass range @xmath269 @xcite erg as the energy of a pop ii sn instead of @xmath270 erg in @xcite . ] .
substituting eq . into eq .
and eq . into eq .
, the x - ray ionization rate is related with the cr ionization rate as @xmath271 thus , we find that , at each redshift , x - ray feedback changes the ionization rate in pristine clouds not by an order of magnitude , but by only up to a factor of two .
note that this is the upper limit of the x - ray effect since x - ray is more effectively shielded than crs .
therefore , our main conclusion still remains valid even if we consider x - ray feedback .
as we have seen above , pop iii star formation at the epoch of galaxy formation is affected significantly by the injection of crs .
nonetheless , some previous studies calculated the pop iii sfr by considering the regulation process of star formation from fuv feedback alone , and the positive feedback from crs is neglected ( e.g. , tornatore et al . 2007 ; trenti & stiavelli 2009 ; de souza et al . 2011 ; johnson et al . 2013 ) .
future studies should take all these feedback into consideration .
we must admit that our treatment of the plane - parallel and steady shock is too simplistic to capture possible complex phenomena induced by e.g. , the turbulence , rotation , thermal instability , and magnetic field .
for example , @xcite studied the effect of magnetic fields on the thermal evolution of a shock - compressed gas , using the two - dimensional mhd simulation .
they found that after the short isobaric contraction , the magnetic pressure is amplified to be comparable to the ram pressure of the pre - shock materials , and thereafter the shock - compressed gas cools almost isochorically .
this implies that the shielding of h@xmath28 against a fuv field becomes less effective and the resultant h@xmath28 fraction is lowered in the presence of magnetic fields . in this case , we expect that more massive clumps may form after the gas becomes gravitationally unstable and fragments , compared to the isobaric contraction case without magnetic fields . however , their calculation did not follow the evolution until the self - gravity of the shocked layer becomes important .
they did not consider the gas in the primordial composition as well .
thus , the above effects on our results should be investigated through the 3d mhd simulations in the future .
another example is that our assumption of the time - independent shock velocity becomes invalid at least in the sn - shock , according to the one - dimensional numerical calculations @xcite .
they showed that the sn - shock velocity becomes lower than @xmath175 at @xmath272 yr after the explosion .
moreover , at @xmath273 yr , the ram pressure of the ambient medium decreases below the pressure in the shell so that the shell re - expands and decreases its temperature and density in an almost adiabatic manner .
although in this case , h@xmath28 is produced efficiently to @xmath20 and the temperature decreases to @xmath274 k by h@xmath28 cooling , hd is produced only to @xmath275 , which is @xmath2 times less than our results , in the adiabatic expansion phase owing to the low gas density . in this case
, the temperature decrease below @xmath276 k is not caused by hd cooling but by the adiabatic expansion of the shell @xcite . however ,
if the sn shock can collect enough materials to trigger gas fragmentation and contraction by self - gravity before dissolving into the ism , hd formation becomes effective and hd cooling becomes dominated in the free - falling phase , as we can see from fig . 6 in @xcite and fig . 6 in @xcite
thus , whether this realistic evolution of the sn shock front significantly changes our results is not clear , and more detailed numerical calculations are needed for more concrete discussion .
hd molecules form abundantly in the primordial gas with enhanced initial ionization and play a dominant role in gas cooling . at the epoch of galaxy formation ( @xmath205 ) , shocks occur ubiquitously associated with the mass assembly as well as the sn explosions . in such a shock - compressed gas ,
the ionization degree jumps up owing to collisional ionization .
other examples include the cloud in the relic hii region of a defunct pop iii star and that with modest cr irradiation . however , even a low level of the fuv background suppresses hd formation / cooling via @xmath11 photodissociation in cases without cr irradiation . in this paper , we have examined the conditions for efficient hd cooling in primordial star formation by calculating the thermal and chemical evolution of a gas cloud under both fuv and cr irradiation .
we have obtained the critical fuv intensity @xmath160 for hd cooling as a function of the cr intensity @xmath80 , and compared it with the estimated background level @xmath239 .
we have considered both cases of the free - falling cloud in a relic hii region and a shock - compressed gas . at @xmath277
, the background cr intensity is estimated as @xmath278 of the galactic value . in this case
, the critical fuv intensity @xmath160 in relic hii regions is below the background level @xmath239 and hd cooling is suppressed there .
however , as the background cr intensity increases with the cosmic star formation , the critical intensity @xmath160 is also elevated and eventually exceeds the background level @xmath239 at @xmath4 . below this redshift , hd cooling becomes efficient and hd - pop iii star formation proceeds if some primordial environments still survive . on the other hand ,
the critical fuv intensity @xmath160 for the shock - compressed gas is @xmath2 times higher than that for the relic hii region even in the weak cr regime as @xmath278 of the galactic value .
this is because the shock - compressed gas evolves isobarically ( i.e. , gas cools with its density increasing ) , while the gas in a relic hii region evolves isochorically ( i.e. , the gas cools at a constant density ) due to the initial rapid cooling .
moreover , shocks tend to occur in denser environments ( e.g. , cold - accretion flows ) than relic hii regions . as a result
, the shock - compressed gas is shielded from the fuv field more effectively , and has higher critical intensity .
thus , for the shock - compressed gas , the critical value @xmath160 exceeds the background level @xmath239 and hd - pop iii star formation proceeds for most of the redshift we consider ( @xmath3 ) .
our result suggests that hd - pop iii stars can be more common than previously considered and could be even the dominant population of pop iii stars .
iii stars are born in clumps as massive as @xmath279 , but the final stellar mass is set to @xmath1 by the stellar radiative feedback onto the accretion flow @xcite . on the other hand , hd - pop iii stars form in clumps of @xmath1 , and the final mass is typically set to a few @xmath280 by the stellar feedback @xcite .
these previous results may suggest that the star formation efficiency , i.e. , the mass ratio of formed stars to the parent clouds , of hd - pop iii stars is several times higher than that of h2-pop iii stars .
they also imply the possibility that hd - pop iii stars could be the majority of pop iii.2 stars . then , hd - pop iii stars of a few 10 @xmath281 can be important sources of radiative feedback at the epoch of galaxy formation . moreover , according to the observations of the abundance pattens in the metal - poor halo stars in the galaxy , the imprints of a pair - instability sn have not been discovered yet @xcite . since stars with @xmath282
end their lives as core - collapse sne @xcite , hd - pop iii stars might be the major contributor to the metal enrichment in the ism / igm at that epoch , if there were no pisne in the universe .
recent discoveries of the almost pristine gas with metallicity @xmath283 in the damped ly @xmath76 systems at @xmath284 and @xmath285 @xcite imply that metal enrichment in the igm has proceeded quite inhomogeneously .
theoretical models also show the inhomogeneous nature of metal enrichment in the igm .
they predict that pop iii star formation can continue up to @xmath27 or even lower , and evaluate the ratio of the pop iii sfr to the total sfr as @xmath286 ( @xmath8 ) at @xmath287 ( @xmath288 ) ( tornatore et al .
2007 ; johnson et al . 2013 ) .
this indicates that pop iii stars can form to some extent even at @xmath241 , and our study show that they are typically formed as hd - pop iii stars with a few @xmath280 .
grbs and sne from these pop iii progenitors at @xmath4 will be detected with the current or future facilities and may provide us with information on the mass , imf and sfr of these stars ( nakauchi et al . 2012
; whalen et al . 2013 ; tanaka et al . 2013 ) . moreover , if the hd absorption lines are detected in the spectrum of a grb afterglow which lacks metal absorption , hd is confirmed as the dominant coolant in that primordial gas @xcite .
we thank the anonymous referee for helpful comments and improving the quality of this paper .
we also thank k. kashiyama , t. nakamura y. suwa , and k. tanaka for fruitful discussions and comments .
this work is supported in part by the grant - in - aid from the ministry of education , culture , sports , science and technology ( mext ) of japan ( 23 - 838 ki ; 25287040 ko ) .
haiman , z. , rees , m. j. , & loeb , a. 1996 , apj , 467 , 522 haiman , z. , thoul , a. a. , & loeb , a. 1996 , apj , 464 , 523 haiman , z. , rees , m. j. , & loeb , a. 1997 , apj , 476 , 458 haiman , z. , abel , t. , & rees , m. j. 2000 , apj , 534 , 11 hayakawa , s. , nishimura , s. , & takayanagi , t. 1961 , pasj , 13 , 184 | hd dominates the cooling of primordial clouds with enhanced ionization , e.g. shock - heated clouds in structure formation or supernova remnants , relic hii regions of pop iii stars , and clouds with cosmic - ray ( cr ) irradiation .
there , the temperature decreases to several 10 k and the characteristic stellar mass decreases to @xmath0 , in contrast with first stars formed from undisturbed pristine clouds ( @xmath1 ) .
however , without cr irradiation , even weak far ultra - violet ( fuv ) irradiation suppresses hd formation / cooling . here
, we examine conditions for hd cooling in primordial clouds including both fuv and cr feedback . at the beginning of collapse
, the shock - compressed gas cools with its density increasing , while the relic hii region gas cools at a constant density .
moreover , shocks tend to occur in denser environments than hii regions .
owing to the higher column density and the more effective shielding , the critical fuv intensity for hd cooling in a shock - compressed gas becomes @xmath2 times higher than in relic hii regions .
consequently , in the shock - compressed gas , the critical fuv intensity exceeds the background level for most of the redshift we consider ( @xmath3 ) , while in relic hii regions , hd cooling becomes effective after the cr intensity increases enough at @xmath4 .
our result suggests that less massive ( @xmath0 ) pop iii stars may be more common than previously considered and could be the dominant population of pop iii stars .
stars : formation - stars : population iii - dark ages , reionization , first stars - early universe . |
in the few hundred central parsecs of the galaxy ( hereafter gc ) clouds are denser ( @xmath1 ) , more turbulent ( @xmath2 km s@xmath3 ) , and hotter than the clouds of the galactic disk .
there is a widespread warm gas component with temperatures of 100 - 200 k first known by nh@xmath4 observations ( see e.g. @xcite ) and now studied in detail with h@xmath0 pure - rotational lines observations by iso ( @xcite , hereafter rf01 ) .
the heating of the warm gas component over large regions where the dust temperatures are much lower than those of the gas is a puzzle .
the discrepancy between the dust and the gas temperature is usually considered to imply a mechanical , rather than radiative , heating mechanism ( @xcite , @xcite ) . with the purpose of investigating the heating mechanisms of the gc molecular clouds
, we have studied a sample of 18 clouds distributed all along the cmz ( central molecular zone ) at millimeter and infrared wavelengths using the iso and the iram-30 m telescopes .
iso has allowed us to study the thermal balance of the gc clouds by observing the major coolants of the gas with temperatures of a few hundred kelvin like h@xmath0 , oi , or cii . in the following
we will review what we have learned from infrared ( ir ) and mm wavelength studies in the field of the large scale study of the gc ism .
we have observed the j=1 - 0 and j=2 - 1 lines of @xmath5 and @xmath6 with the iram-30 m telescope .
the j=2 - 1 to j=1 - 0 line ratios are compatible with cold ( 20 k ) and dense gas ( 10@xmath7 ) or warmer ( @xmath8 100 k ) but less dense gas ( 10@xmath9 ) . however , the column densities derived in both cases are rather similar . in general ,
if one considers a mixture of cold and warm gas the total column densities traced by co will be similar to those derived for gas at 20 k , which vary from source to source but are in the range of @xmath10 ( rf01 ) we have observed several @xmath11 pure - rotational lines ( from the s(0 ) to s(5 ) lines ) with the sws spectrometer on board iso .
the s(3 ) line is strongly absorbed by the 9.7 band of the silicates and has only been detected in the sources with the most intense s(1 ) emission .
the visual extinction derived from the @xmath11 data is @xmath12 mag . after correcting for extinction one
finds that the excitation temperature derived from the s(0 ) and s(1 ) lines ( @xmath13 ) is between 130 and 200 k while that derived from the s(4 ) and s(5 ) lines when detected is 500 - 700 k. there is not a clear dependence of @xmath13 on the distance to the galactic center . extrapolating the populations in the j=2 and j=3 levels ( as derived from the s(0 ) and
s(1 ) lines ) to the j=0 and j=1 levels at the temperature @xmath13 one finds that the total warm @xmath11 column density varies from source to source but is typically of 1 - 2 @xmath14 .
the column density of gas at @xmath15 k is less than @xmath16 of the column density at @xmath17 k. on average , the warm @xmath11 column densities are about @xmath18 of the total @xmath11 column densities derived from co. for a few clouds the fraction of warm gas is as high as 77@xmath19 or even @xmath20 .
this implies that for these clouds almost all of the co emission should arise from warm gas ( see rf01 ) .
comparing with the observations of @xcite one finds relatively high abundances of a few 10@xmath21 in both the warm and the cold gas .
there are indirect arguments that points both to shocks and to photo - dissociation regions ( pdrs ) as the heating mechanism of the warm gas ( see rf01 for a complete discussion ) .
direct comparison of the @xmath11 data with pdrs and shocks models ( see fig .
[ figexcith2 ] ) indicate that the s(4 ) and s(5 ) lines trace the densest gas in the gc clouds ( 10@xmath22 ) heated in pdrs , shocks , or both . nevertheless , to explain the large column densities of gas at @xmath17 k traced by the s(0 ) and s(1 ) lines several less dense pdrs ( with @xmath23 and @xmath24 ) or low velocity shocks ( @xmath25 ) in the line of sight are required .
the curvature of the the population diagrams is in agreement with the temperature gradient expected in a pdr ( fig .
[ figexcith2]b ) but probably also with a composition of shocks with different velocities . in summary , it is difficult to know if the @xmath11 emission arises in pdrs or shocked regions .
with both the lws and sws spectrometers on board iso we have detected fine structure lines of ions like [ n ii ] 122 , [ siii ] 33 and 18 or [ ne ii ] 12 in most sources .
we have even detected [ n iii ] 57 , [ o iii ] 88 and 52 or [ ne iii ] 15.6 in the few sources located closer to the galactic center .
all these lines of ions with high excitational potentials must arise in h ii regions .
otherwise , one needs to invoke j - shocks with velocities 100 and there is no hint of such strong shocks from the line profiles or the radial velocities of any observed line .
we derive effective temperatures for the ionizing radiation of @xmath26 k , typical of an o7 star ( see @xcite , 2000 ) . we have observed the h35@xmath27 and h@xmath28 recombination lines in all the sources of our sample with the iram-30 m telescope .
we have not detected any of these lines in any source ( @xcite , 2000 ) .
assuming line widths as large as 100 and an electron temperature ( @xmath29 ) of 8000 k we can set a conservative _ upper limit _ to the number of lyman continuum photons emitted per second by the excitation source of @xmath30 .
this rate is typical of a star with a spectral type b0 or later and effective temperature @xmath31 k. this implies that the number of lyman continuum photons derived from the radio recombination lines is not the total number of those photons emitted by the ionizing source .
the apparent inconsistency between the radio recombination lines and the fine structure lines disappears if one considers that the hii regions are more extended than the 30-m telescope beam , e.g. , that the ionizing stars are surrounded by cavities with sizes of @xmath32 pc ( @xcite , 2000 ) .
figure [ figdustall ] shows the dust continuum spectra toward a representative sample of sources .
any other spectrum is very similar to one of those in fig .
[ figdustall ] .
the wavelength of the maximum of emission varies from @xmath33 for m+0.76 - 0.05 and other sources with galactic longitude @xmath34 to @xmath35 for the sources located in the sgr c region and even to @xmath36 for a few sources located close to the radio arc . to explain the dust emission one needs at least two grey bodies .
figure [ figfitdust ] displays the dust emission of the sources with coldest ( m+0.76 - 0.05 ) and warmest dust ( m+0.21 - 0.12 ) together with the predictions of a model with two grey bodies at different temperatures . for the predictions
we have used the following expression : @xmath37\ ] ] where @xmath38 represents the planck function and the opacity , @xmath39 , is given by : @xmath40 where @xmath41 is the visual extinction in magnitudes and @xmath42 expressed in .
this model assumes that the warmer component is being extinguished by the colder one .
taking @xmath43 ( the lws beam ) we obtain good fits for the values of the parameters listed table [ tabfit ] .
lllll + source & @xmath44 & @xmath45 & @xmath46 & @xmath47 + + m+0.76 - 0.05 & 49 & 1.5 & 16 & 10 + & 7.5 & 1 & 30 & 1 + m+0.21 - 0.12 & 50 & 1.5 & 16 & 6 + & 2.2 & 1 & 47 & 1 + + to explain the emission at large @xmath42 s it is necessary a grey body with temperature @xmath48 k. the visual extinction associated to this component would be 3050 mag . to explain the large fluxes observed at large @xmath42 s , filling factors ( @xmath47 ) @xmath49 are needed .
the temperature of the cold component does not vary much from source to source .
in addition to the cold dust component a warmer one is also required .
this component would fill the beam and its temperature varies from source to source from the @xmath12 k of m+0.75 - 0.05 , m+3.06 + 0.34 and m+1.56 - 0.30 ( @xcite ) to the @xmath50 k of m+0.21 - 0.12 and m+0.35 - 0.06 .
the visual extinction associated to this component would be @xmath8 2 - 8 mag , that is , 5 - 10 @xmath19 of the extinction due to the cold dust component . clearly , a heating mechanism is needed to raise the dust temperature from @xmath48 k to 30 k or even 45 k. for the standard dust - to - gas ratio the extinction caused by the warm dust component is equivalent to a @xmath11 column density of 2 - 8 @xmath51 .
these equivalent column densities are only a factor of 2 lower than the column densities of warm gas as derived from the @xmath11 pure - rotational lines .
the simplest explanation is that both the warm gas and dust arise in low - density pdrs like those modeled by @xcite . in the external layers of the pdr
the gas is heated via photoelectric effect in the grain surfaces without heating the dust to high temperatures .
for instance , in their standard " model ( @xmath24 and @xmath23 ) , the gas can reach temperatures of 100200 k in the first 3 mag of visual extinction into the cloud where on average the dust temperature would be @xmath52 k ( see fig . 3 of @xcite ) .
those values for the pdr parameters are quite reasonable for the gc ism at large scale based on average densities and the far - ir continuum ( see discussion in @xcite ) .
pdr models with @xmath53 and @xmath54 can also explain the @xmath11 excitation temperatures traced by the s(0 ) and s(1 ) lines and the large scale emission of the @xmath11 @xmath55 s(1 ) line ( @xcite ) .
in this scenario , the warm dust component would be located at the gc while the cold dust emission would arise both from the gc clouds and from clouds in the line of sight ( typically @xmath56 mag of visual extinction are due to material located between us and the gc region ) .
iso has , for the first time , measured directly the column densities of warm molecular gas in the gc . on average ,
the warm @xmath11 represents a fraction of @xmath18 of the total @xmath11 column densities derived from @xmath6 and @xmath5 observations .
iso has also detected fine structure lines of ions with high ionization potential that should arise from hii regions .
the comparison of the fine structure lines with hydrogen recombination lines at mm wavelengths shows evidences for an extended ionized component with high effective radiation temperature .
this suggest the presence of a pdr in the interface between the hii region and the molecular material that should contribute to the heating of the warm gas and the dust component with temperatures of 30 - 45 k. there are many evidences that shocks play an important role in the physics and chemistry of the gc ism .
the high abundances of molecules which are easily photodissociated like , sio ( @xcite ) and c@xmath0h@xmath57oh ( @xcite ) suggest that they should be sputtered from the grains by shocks .
nevertheless , iso points out that the effect of uv radiation on the gc ism ( and in particular on the heating of the molecular gas ) can not be ruled out with the simple argument that the dust temperature is lower than that of the gas .
first will help to understand the energetics of the gc ism .
for instance , the high spectral and spatial resolution that the hifi instrument will achieve , will be of great interest to study the interplay between the ionized material and the warm molecular gas .
njr - f acknowledges _ consejera de educacin cultura de la comunidad de madrid _ for a pre - doctoral fellowship .
this work has been partially supported by the spanish cicyt and the european commission under grant numbers esp-1291-e and 1fd1997 - 1442 . | the _ infrared space observatory _ ( iso ) has shown the complexity of the galactic center ( gc ) interstellar medium ( ism ) detecting , not only large column densities of warm molecular gas ( h@xmath0 ) , but the emission of neutral atoms and ions of low ionization potential ( cii , oi , siii , ... ) that should arise in shocked or photon - dominated regions ( pdrs ) .
in addition , iso has also detected emission from ions like siii , neii , arii , or nii ( in some clouds we have even detected neiii and oiii ) that should arise from hii regions that were previously unsuspected due to the non - detection of hydrogen recombination lines .
here we review some iso results on the large scale study of the gc ism and in particular , on the heating mechanisms of the clouds .
although , shocks should play an important role on the physics and chemistry of the gc ism , iso shows that the effect of radiation on the heating of the gas can not be ruled out with the simple argument that the dust temperature is lower than that of the gas . |
in recent years , dynamical quantum phase transition ( dqpt ) has extended our understanding of phase transitions and universality greatly @xcite , which provides us a new perspective on exploring the behavior of phase transitions far from equilibrium . as a simple but important paradigm of nonequilibrium processes , the quantum quench attracted intensive studies . to describe the dynamics of a quantum system which
is pushed out equilibrium by a sudden change of the hamiltonian , an important quantity is the loschmidt echo , which measures the overlap of the initial quantum state and the time - evolved state after the quench @xcite .
many theoretical works have demonstrated that the loschidmt echo plays an important role in characterizing the nonequilibrium dynamic signature of a quantum phase transition @xcite .
after mapping the loschmidt amplitude to a boundary partition function , the singularity of dynamical free energy density in thermodynamic limit can be found at critical times @xmath3 , which are similar to the well - known fisher zeros @xcite .
this singularity is found to have relationship with the dynamics of order parameter @xcite . up to now , the dqpt has been explored in a series of models which are known to exhibit quantum phase transition , such as transverse field ising model ( tfim ) @xcite , anisotropic xy model @xcite , hubbard and falicov - kimball models @xcite and two - band topological systems @xcite , etc . thanks to the developments of quantum simulation techniques , dqpt has been realized by ions simulations of tfim @xcite . besides , by observing the appearance , movement , and annihilation of vortices in reciprocal space , dynamical topological order parameter has also been recognized in optical lattices systems @xcite .
according to the theory of dqpt , the appearance of a series of zero points in the loschmidt echo at critical times @xmath3 can be viewed as a dynamic signature of quantum phase transitions . while most theoretical studies of dqpt and loschmidt echo focus on the traditional quantum systems driven by competing quantum terms @xcite , less attention is paid on the dynamics and loschmidt echo in a quantum disorder system which exhibits the localization - delocalization transition @xcite .
a natural but interesting question is whether the loschmit echo can still be used to characterize the dqpt of a quantum disorder system ? and
if yes , whether we can observe zero points of the loschmidt echo by studying the quench dynamics of the quantum disorder system ? aiming to answer these questions , in this work we shall study the quench dynamics in a one - dimensional ( 1d ) incommensurate lattice , which is effectively described by the aubry - andr ( aa ) model @xcite .
it is known that all the eigenstates of the aa model are either extended or localized and there exists a transition from an extended state to a localized state with the change of the strength of incommensurate potential @xcite . the localization transition in the 1d incommensurate lattice has been experimentally verified in a bichromatic optical lattice by observing the expansion dynamics of a bose - einstein condensate initially trapped in the center of optical lattice @xcite , which exhibits different diffusion behaviors for the extended or localized phase @xcite .
different from previous works on the expansion dynamics , we study the quench dynamics with the initial state being an eigenstate of the initial hamiltonian . after performing a sudden quench of the strength of incommensurate potential @xmath0 , the behaviors of loschmidt echo are found to be quite different depending on whether the initial and final hamiltonians locate in the same phase regime or not .
the paper is organized as follows . in section
ii , we introduce the model and study the quench dynamics in the limiting cases of quenching between @xmath1 ( @xmath4 ) and @xmath4 ( @xmath1 ) . in these two limiting cases , we can give analytical expressions of the loschmidt echo and demonstrate that there are a series of zeros of loschmidt echo , which is also consistent with our numerical results . in section iii , we study the general quench process , for which no analytical results are available and we thus study the evolution of loschmidt echo with the help of numerical methods . by analyzing the statistical behavior of the values of loschmidt echo in a long time ,
we demonstrate that loschmidt echo will oscillate and take a finite value in the case of quenching in the same phase , while loschmidt echo can approach zero in the case of quenching in different phases .
a brief summary is given in section iv .
we investigate the aa model with hamiltonian @xmath5 where @xmath6 denotes the creation ( annihilation ) operator of fermions , @xmath7 is hopping amplitude , @xmath8 is an irrational number and @xmath0 is the strength of the incommensurate potential .
the incommensurate potential can be viewed as a kind of quasi - random disorder , which drives the system undergoing a delocalization - localization transition at @xmath9 . when @xmath10 , all the eigenstates are extended , whereas all the eigenstates are localized , when @xmath11 @xcite . for convenience
we take @xmath12 as the energy unit and fix @xmath13 . while conventional studies of dynamical properties in disordered systems focus on the diffusion of a wave packet , in this work we consider the quench dynamics of the incommensurate system described by the aa model . by preparing the system as an eigenstate of the hamiltonian @xmath14 and then performing a sudden quench to the final hamiltonian @xmath15
, we consider the behavior of loschmidt amplitude ( return amplitude ) @xmath16 and loschmidt echo ( return probability ) @xmath17 where @xmath18 stands for the eigenstate of the initial hamiltonian , and @xmath19 ( @xmath20 ) represents the strength of the incommensurate potential corresponding to the initial ( final ) state before ( after ) the quench .
it is known that the loschmidt echo plays an important role in the theory of dqpts .
the behavior of loschmidt echo approaching zero at some times @xmath21 in the thermodynamic limit can be viewed as a signature of the occurrence of the dqpt , which has been demonstrated by studying various models exhibiting quantum phase transitions .
however , it is still not clear whether the loschmidt echo approaching zero can be viewed as a signature for the localization - delocalization transition , which shall be clarified in this work . in order to get an intuitive understanding ,
we first consider two limiting cases of quench processes , i.e. , quench processes between states with @xmath22 @xmath23 and @xmath24 @xmath25 , which can be calculated analytically , whereas the general quench processes between arbitrary @xmath19 and @xmath20 are studied with the help of numerical calculations .. s and the size of the system @xmath26 .
the initial state is fixed to be the ground state of the initial hamiltonian with @xmath22 .
( a ) @xmath27 versus @xmath21 .
( b ) @xmath27 versus the rescaled time @xmath28 .
( c ) evolution of dynamic free energy " @xmath29 for various large @xmath20s .
the black dashed - dotted curve corresponds to the analytical result @xmath30 .
( d ) evolution of loschmidt echo for various @xmath20s . for @xmath31
, @xmath32 will approach zero after some time intervals.,title="fig : " ] + in the first case , we fix @xmath22 and consider the periodic boundary condition , i.e. , the system is initially prepared in a plane wave state , which is the eigenstate of the hamiltonian ( [ eq1 ] ) with @xmath22 : @xmath33 where the wave vector @xmath34 $ ] ( @xmath35 ) lies in the first brillouin zone ( bz ) and @xmath36 represents the lattice spacing .
the corresponding eigenvalue is @xmath37 performing a sudden quench of @xmath0 from the initial value @xmath19 to the final value @xmath20 , the return amplitude can be written as @xmath38 where @xmath39 and @xmath40 denote the @xmath41 eigenevalue and eigenstate of final hamiltonian respectively .
now use the fact that when in the limit of @xmath42 , the energy is determined by the diagonal terms and the eigenstates are localized in a single site , we can simplify eq .
( [ eq6 ] ) and obtain @xmath43 for an irrational number @xmath8 , the phase @xmath44 ( @xmath45 ) modulus @xmath46 is distributed randomly between @xmath47 and @xmath48 when we sum over @xmath49 to the large @xmath50 limit .
so we can approximately replace the summation by the integration @xmath51 where @xmath52 is the zero - order bessel function . from the properties of bessel function , we know that @xmath53 has a series of zeros @xmath54 with @xmath55 .
these zeros indicate the loschmit amplitude and echo reach zero at times @xmath56 according to the theory of dqpt , the occurrence of a series of zeros in the loschmit amplitude can be viewed as the signature of dqpt as these zeros correspond to divergences of the boundary partition function .
although the analytical result is obtained in the limit of @xmath42 , the above results are expected to hold true as long as @xmath20 is large enough ( see fig . [
fig1](a)-(c ) ) .
since the transition time @xmath57 is inversely proportional to @xmath20 , the loschmidt echo will oscillate more rapidly as @xmath20 is increasing .
if we rescale the time @xmath58 , the evolution of loschmidt echo will display a similar behavior for quench processes with different @xmath20 . to see it clearly
, we display the numerical results of the evolution of loschmidt echo as a function of @xmath21 and @xmath28 in fig . [ fig1](a ) and fig .
[ fig1](b ) , respectively . here
the initial strength is set at @xmath22 .
it is clear that the loschmit echoes @xmath32 for @xmath59 , @xmath60 and @xmath61 oscillate with different frequencies , but they almost completely overlap to the analytical result @xmath62 and are indistinguishable after rescaling the time as shown in fig . [ fig1](b ) .
when @xmath63 , the loschmit echo obviously deviates @xmath62 , indicating the analytical result obtained in the limit @xmath42 is no longer a good approximation . to see the zeros of @xmath32 more clearly , we can use the dynamical free energy " which is defined as @xmath64 @xcite , where @xmath29 will be divergent at the dynamical phase transition time @xmath65 .
the evolution of @xmath29 for different @xmath20s are shown in fig . [ fig1](c ) .
@xmath32 exhibits obvious peaks around @xmath65 and the behavior gets more close to the limiting case with the increasing of @xmath20 . in fig .
[ fig1](d ) , we display @xmath32 versus @xmath28 for various @xmath20 with @xmath63 , @xmath66 , @xmath67 , @xmath68 , @xmath69 and @xmath70 from bottom to top .
it can be seen that @xmath32 exhibits different behaviors for @xmath31 and @xmath71 . for @xmath71 ,
@xmath32 oscillates with its average decreasing with the increasing of @xmath20 .
we do not find any zero point of @xmath32 even in a long time , which is obviously different from cases with @xmath31 . as a comparison , for the case of @xmath72 the loschmit echo @xmath32 has an obvious decay and reaches nearby zero at about @xmath73 . with the increase of @xmath20
, @xmath32 decays more quickly and gets more closed to the limiting case described by eq .
( [ eq7 ] ) . for large @xmath20s .
the black dashed - dotted curve corresponds to the analytical result @xmath74 .
( b ) evolution of loschmidt echo with various @xmath20s and the size of the system @xmath26 .
the initial state is fixed to be the ground state of the initial hamiltonian with @xmath75 .
, title="fig : " ] + next we consider the quench processes from a very large @xmath19 to @xmath76 . in the limit of @xmath77 ,
the initial state is chosen as an eigenstate of the system , which is localized in one site , e.g. , the site @xmath49 .
substituting eq .
( [ eq4 ] ) and ( [ eq5 ] ) into eq .
( [ eq2 ] ) we get @xmath78 in the large n limit , we can replace the summation by the integration , which leads to @xmath79 from this expression , it is clear that the zeros of loschmit echo occur at @xmath80 which are half of the zeros of the zero - order bessel function @xmath53 .
when @xmath20 deviates a little from the limit case of @xmath76 , the analytical result eq.([eq9 ] ) is expected to be still a good approximation .
different from eq .
( [ eq8 ] ) , the transition time @xmath57 is independent of @xmath20 . furthermore , we find that @xmath57 is also not sensitive to the initial value @xmath19 as long as @xmath19 is large enough because the only information of the initial hamiltonian we have used is the localized wave function . in fig .
[ fig2 ] , we show the numerical results for quenching processes with the initial state prepared in a localized state , which is taken to be the ground state of the initial system with @xmath81 . from fig .
[ fig2 ] ( a ) , we see that systems with @xmath82 , @xmath83 , @xmath84 and @xmath85 display similar behaviors to the limit case with @xmath76 , for which the divergent points of @xmath29 occur at @xmath65
. the more close to @xmath76 , the curves of numerical results are more close to the analytical result @xmath74 , which are not sensitive to the values of @xmath20 . in fig .
[ fig2](b ) , we display @xmath32 versus @xmath21 for various @xmath20 with @xmath86 , @xmath68 , @xmath67 , @xmath87 , @xmath88 and @xmath89 from bottom to top .
similar to the previous case displayed in fig .
[ fig1](d ) , @xmath32 exhibits quite different behavior for @xmath31 and @xmath71 . for @xmath71
, @xmath32 will approach zero at some given times . on the other hand ,
when @xmath31 , @xmath32 never approaches zero in the evolution process .
in the above section , starting from the initial state prepared in the limit case with @xmath22 ( or @xmath77 ) , we have shown that the loschmidt echo can reach nearby zero in the evolution process if the incommensurate strength @xmath20 after the quench is larger ( or less ) than the critical value @xmath90 , which is also the localization - to - delocalization transition point of the aa model .
now we consider the general cases that @xmath19 and @xmath20 are neither close to zero nor the infinity limit .
although no analytical solution can be found for the general case , we can still explore whether the presence or absence of the zeros of loschmidt echo can still serve as a characteristic signature of dynamic quantum phase by numerically analyzing the evolution of the loschmidt echo . in fig .
[ fig3 ] , we show the evolution of loschmidt echo for various @xmath20s with @xmath91 in ( a ) and ( b ) , and @xmath92 in ( c ) and ( d ) , respectively . if both @xmath19 and @xmath20 locate in the same regime , i.e. , both in the regime of @xmath93 or @xmath94 , @xmath32 oscillates and has a positive lower bound , which never approaches zero during the evolution process , as shown in fig .
[ fig3 ] ( a ) and ( c ) .
however , if @xmath19 and @xmath20 locate in different regimes , @xmath32 shall approach zero after some time intervals , as shown in fig .
[ fig3 ] ( b ) and ( d ) . .
the initial state is chosen to be the ground state of the initial hamiltonian with ( a ) ( b ) @xmath91 and ( c ) ( d ) @xmath92 .
different @xmath20s are shown by different colors .
loschmidt echo can reach nearby zero only if @xmath20 passes through the critical point @xmath95.,title="fig : " ] + to give a quantitative description on how loschmidt echo approaches zero , we define a cutoff of small value @xmath96 close to zero . at a given large length range of time @xmath97
, we measure the length of time interval which fulfills @xmath98 in @xmath99 $ ] . denoting this length as @xmath100 , which is a function of @xmath96 when
fixing @xmath97 , it can be viewed as the lebesgue measure @xmath101 @xcite . for convenience
we use a normalized function @xmath102 . in fig .
[ fig4](a ) we show @xmath103 as a function of @xmath20 for different @xmath96s with @xmath91 fixed in the extended regime . here the initial state is chosen as the ground state of the system .
it can be seen that the behavior of @xmath104 is quite different for @xmath71 and @xmath31 .
for @xmath71 , @xmath104 is always zero for @xmath105 , @xmath106 , @xmath107 and @xmath108 .
however , there is a sharp increasing as @xmath20 passes through the transition point @xmath90 , and @xmath103 takes a finite value when @xmath31 .
despite the fact that the value of @xmath103 in the regime of @xmath31 depends on the cutoff value @xmath96 , we note that the sharp change behaviors around the transition point are similar for different cutoffs . as a function of @xmath20 for @xmath26 , @xmath109 and @xmath91 .
( a ) different @xmath96s are shown by different colors .
there is a sharp increasing around @xmath110 .
it is shown that @xmath111 for @xmath71 and @xmath112 for @xmath31 .
( b ) different choice of initial state with @xmath113 standing for the label of eigenstates of @xmath14 . a clear boundary located at @xmath110 can be seen . here
@xmath114.,title="fig : " ] + although the initial state is taken to be the ground state of @xmath14 in the above calculations , we would like to indicate that our conclusion is independent of the choice of the initial eigenstates . to see this clearly , in fig .
[ fig4](b ) we show @xmath104 as a function of @xmath20 by choosing different eigenstates of @xmath14 as the initial state with @xmath91 and @xmath114 .
we can see that there exists a clear boundary at @xmath110 . for @xmath71 , @xmath49 is close to zero in the whole region .
a sharp increase can be found around the transition point @xmath90 for all the initial eigenstates , and @xmath103 takes a finite value when @xmath31 . and @xmath21 for the system with @xmath26 .
the valley only occurs at the critical point @xmath95 .
( b ) the cross section of @xmath95 for different sizes of systems.,title="fig : " ] + finally , we consider the special case of @xmath20 being very close to @xmath19 .
in such a case , the loschmidt echo can be represented as @xmath115 where @xmath116 is a very small value . in terms of the above definition
, a sharp decay of the loschmidt echo around the critical point has been taken as the signature of quantum phase transition @xcite .
the quench process can be viewed as from @xmath117 to @xmath118 , so the initial and final hamiltonian are quite similar except around the critical point @xmath95 . in fig .
[ fig5](a ) , we fix @xmath119 and show the loschmidt echo as a function of @xmath0 and @xmath21 . a deep valley can be found at @xmath95 , as the localization - delocalization transition enhances the decay of loschmidt echo .
while in the region apart from the critical point , the loschmidt echo oscillates near @xmath120 and does not decay in a long time .
the cross section of fig .
[ fig5](a ) at @xmath95 is shown in fig . [
fig5](b ) . as a comparison
, we also provide results for systems with different sizes .
it is clear that the loschmidt echo decays in an oscillating way and can always reach nearby zero in quite a long time interval , which is consistent with our conclusions .
in summary , we have studied the quench dynamics of the aa model by preparing the initial state as an eigenstate of the initial hamiltonian @xmath14 and then performing a sudden quench to the final hamiltonian @xmath15 . for the quench process between two limiting cases , i.e. , with @xmath22 and @xmath121 or @xmath122 and @xmath123
, we obtain the analytical expression of the loschmidt echo , which suggests the existence of a series of zero points at critical times @xmath3 . by comparing with the numerical results , we find the analytical results are still good approximations as long as the quench parameters deviate these limits not far away .
for the general quench processes , we study the statistical behavior of loschmidt echo numerically and demonstrate that loschmidt echo would oscillate but never decay to zero in a long time if @xmath19 and @xmath20 are located in the same phase ; however , loschmidt echo would decay and reach nearby zero if @xmath19 and @xmath20 are located in different phases .
our results suggest that the occurrence of zero points in the loschmidt echo can give a dynamical signature of localization - delocalization transition in the 1d incommensurate lattice .
the work is supported by the national key research and development program of china ( 2016yfa0300600 ) , nsfc under grants no .
11425419 , no . 11374354 and no .
11174360 , and the strategic priority research program ( b ) of the chinese academy of sciences ( no .
xdb07020000 ) . k. machida and m. fujita , phys .
b * 34 * , 7367 ( 1986 ) .
t. geisel r. ketzmerick and g. petschel , phys .
lett . * 66*,1651 ( 1991 ) .
t. roscilde , phys .
a * 77 * , 063605 ( 2008 ) .
g. roux , t. barthel , i. p. mcculloch , c. kollath , u. schollwck , and t. giamarchi , phys .
a * 78 * , 023628 ( 2008 ) ; x. deng , r. citro , a. minguzzi , and e. orignac , phys .
a * 78 * , 013625 ( 2008 ) .
m. albert and p. leboeuf , phys .
rev . a * 81*,013614 ( 2010 ) .
x. cai , s. chen , and y. wang , phys .
a. * 81 * , 023626 ( 2010 ) ; k. he , i. i. satija , c. w. clark , a. m. rey , and m. rigol , phys .
a * 85 * , 013617 ( 2012 ) .
n. lo gullo and l. dellanna , phys .
a * 92 * , 063619 ( 2015 ) .
p. zanardi and n. paunkovic , phys .
e * 74 * , 031123 ( 2006 ) ; w. l. you , y. w. li , and s. j. gu , phys .
e * 76 * , 022101 ( 2007 ) ; h. q. zhou and j. p. barjaktarevic , j. phys . a : math
. theor . * 41 * , 412001 ( 2008 ) . | we investigate the quench dynamics of a one - dimensional incommensurate lattice described by the aubry - andr model by a sudden change of the strength of incommensurate potential @xmath0 and unveil that the dynamical signature of localization - delocalization transition can be characterized by the occurrence of zero points in the loschmit echo . for the quench process with quenching taking place between two limits of @xmath1 and @xmath2 ,
we give analytical expressions of the loschmidt echo , which indicate the existence of a series of zero points in the loschmidt echo . for a general quench process
, we calculate the loschmidt echo numerically and analyze its statistical behavior .
our results show that if both the initial and post - quench hamiltonian are in extended phase or localized phase , loschmidt echo will always be greater than a positive number ; however if they locate in different phases , loschmidt echo can reach nearby zero at some time intervals . |
over half of gev gamma - ray sources detected by the egret experiment have not yet been identified as known astronomical objects ( hartman et al .
1999 ) , and understanding their origin is one of the most important issues in high - energy astrophysics .
many of unidentified sources located in the galactic plane are believed to be associated with either of pulsars , supernova remnants , or massive stars as suggested by statistically significant correlation between unidentified egret sources and tracers of these objects ( see , e.g. , romero 2001 for a review ) .
another galactic population of unidentified egret sources at intermediate galactic latitude ( @xmath6 ) has been reported by gehrels et al .
( 2000 ) , which are apparently associated with the gould belt , and they might be off - beam gamma - ray pulsars ( harding & zhang 2001 ) .
there are also unidentified sources at even higher latitude ( @xmath7 ) , suggesting the extragalactic origin .
variable sources among them are likely undetected blazars , while some stable sources might be dynamically forming or merging clusters of galaxies ( totani & kitayama 2000 ; waxman & loeb 2000 ; kawasaki & totani 2002 ) .
one of the unidentified egret sources , 3eg j1835 + 5918 has been paid particular attention in recent years , since it is the brightest unidentified source outside the galactic plane ( @xmath8 , @xmath9 ) .
its gamma - ray properties , i.e. , steady flux and hard spectrum are consistent well with other gamma - ray pulsars observed by egret ( reimer et al .
2001 ) , rather than blazars .
no strong radio counterpart also argues against the blazar origin .
intensive multi - wavelength studies of this source ( mirabal et al .
2000 ; mirabal & halpern 2001 ; reimer et al .
2001 ) have revealed about a dozen of x - ray sources in or around the error circle of 3eg j1835 + 5918 , and most of them are coronal emission from stars or quasars without any special or blazar - like characteristics , which are not likely the counterpart of 3eg j1835 + 5918 .
then the only one unidentified x - ray source , rx j1836.2 + 5925 , remains as the most likely counterpart .
it has very soft x - ray spectrum ( @xmath10k ) and has no optical counterpart down to @xmath11 ( mirabal & halpern 2001 ) giving a high x - ray / optical ratio .
this and the lack of strong radio emission are reminiscent of the characteristics of isolated neutron stars or radio - quiet pulsars ( e.g. , brazier & johnston 1999 ; neuhuser & trmper 1999 ) , and this is why both mirabal & halpern ( 2001 ) and reimer et al .
( 2001 ) concluded that the most likely identification of 3eg j1835 + 5918 / rx j1836.2 + 5925 is an isolated radio - quiet pulsar , which is similar to the proto - type object geminga .
however , with the x - ray flux of 26@xmath12 ( mirabal & halpern 2001 ) , the optical upper limit ( @xmath13 at @xmath14 ) corresponds to the x - ray / optical flux ratio of @xmath15 300 , and it is not large enough compared with the flux ratio expected for isolated pulsars or neutron stars , although all x - ray sources other than isolated neutron stars and low - mass x - ray binaries have @xmath16 ( stocke et al .
1991 ) . here
we report a deep optical imaging observation of 3eg j1835 + 5918 / rx j1836.2 + 5925 by the 8.2 m subaru telescope , to examine the proposed identification and the origin of this enigmatic source .
the observation was made on july 17 , 2001 , by the subaru / focas whose field of view is a circle with @xmath17 diameter .
we took images in @xmath18 and @xmath19 bands , with the total exposure time of 2400 and 6000 sec , respectively .
the 5@xmath20 peak level corresponds to @xmath21 and @xmath22 .
figures [ fig : b-30x30 ] and [ fig : b-10x10 ] show the b band images centered on the position of rx j1836.2 + 5925 determined by mirabal & halpern ( 2001 ) with the error radius of @xmath23 ( j2000 , @xmath24 , @xmath25 ) .
the galactic extinction to this direction is @xmath26 and @xmath27 ( schlegel , finkbeiner , & davis 1998 ) giving upper bounds on extinction in any galactic objects .
the seeing size in the @xmath18 band is @xmath28 fwhm . on the other hand ,
the quality of @xmath19 band image is not as good as the @xmath18 band and no object is found in the error circle ; we then set a rough upper limit of @xmath29 including systematic uncertainty of photometry . in the @xmath18 band image we detected two extended sources which are connected with each other ( labeled as c and d in the images ) .
their peak are 5.1 and 5.8@xmath20 of the sky fluctuation level , respectively .
we found two point - like sources with peak levels higher than 3@xmath20 levels , which are labeled as a and b in the images .
the results of photometry as well as the source locations for the above four objects are summarized in table [ table : photometry ] . in the following of this paper
, we will refer to the sources c and d as one source ( c+d ) .
we checked that the noise fluctuation is symmetric above and below the zero point of the background , and we did the same detection procedure for the inverted image to estimate the number of spurious objects .
this indicates that sources a and b are not firm detections ; the probabilities that they are spurious detection are estimated to be @xmath30 and 30% , respectively , from the counts of spurious objects in the inverted frame .
an even fainter , extended source can be seen in the bottom of the error circle , whose peak level is less than 3@xmath20 , and we do not consider this source in this letter . if 3eg j1835 + 5918 is a pulsar with high proper motion , it might have moved to outside of the error circle determined by rosat data .
therefore examination only within the rosat error circle is not sufficent .
our observation is 3.5yr later than the rosat hri observation ( mirabal & halpern 2001 ) , and assuming a distance of 200pc and transverse velocity of 500km / s , the proper motion becomes @xmath31 .
therefore we examined the outer region @xmath32 beyond the rosat error circle , but we found no significant sources like a d .
first we consider the possibility that c+d is an interacting galaxy .
the @xmath18-band galaxy count at @xmath2 is about @xmath33 ( e.g. , totani & yoshii 2000 ) , and hence the number of galaxies expected in the error circle of rx j1836.2 + 5925 becomes @xmath340.4 , indicating that the appearance of a galaxy by chance is not surprising at all .
the x - ray spectrum of rx j1836.2 + 5925 is very soft , and this argues against the possibility that the x - ray emission is coming from nuclear activity in extended faint galaxies . as mentioned earlier , the gamma - ray properties and lack of strong radio emission also disfavor the agn origin for 3eg j1835 + 5918 .
the morphology of c and d suggests that they are merging or interacting galaxies , and steady radiation of gamma - rays may be possible by particle acceleration from merging shocks .
however , assuming typical values of mass / light ratio of galaxies , merging velocity ( @xmath34 a few hundreds km / s ) , and interaction time scale ( @xmath35yr ) , we estimated the maximum luminosity possible from merging , as the merging kinetic energy divided by the interaction time scale .
it turns out to be of order @xmath36 , and this is more than four orders of magnitude weaker than the gamma - ray flux of 3eg j1835 + 5918 , safely excluding this possibility .
therefore , if the source c+d is a galaxy , it is not a counterpart of 3eg j1835 + 5918 .
another possibility which should be considered here is that the extended object c+d is a pulsar wind nebula .
the optical @xmath37 flux is more than 5 orders of magnitude lower than the gamma - ray flux , and hence than the spin - down luminosity of the pulsar .
the size of the extended objects , @xmath38 , seems small as an entire size of typical synchrotron nebulae , but observations of other pulsars or pulsar candidates have often revealed extended cores or knots with similar size around central pulsars , e.g. , optical knots in the crab pulsar ( hester et al .
1995 ) or chandra observations of several crab - like young supernova remnants ( slane et al . 2000 ; hughes et al . 2001 ) .
nonthermal x - ray synchrotron nebula emission is known to be correlated to the spin - down luminosity as @xmath39 ( e.g. , seward & wang 1988 ; becker & trmper 1997 ) , and the compact ( but extended ) x - ray cores have a flux of about 10% of the total nebula flux ( slane et al . 2000 ; hughes et al . 2001 ) .
the optical flux of the object c+d seems small compared with these x - ray fluxes .
on the other hand , the optical knots of the crab pulsar have luminosity about seven orders of magnitude lower than the spin - down luminosity , and hence the object c+d may be a similar knot embedded in a nebula which is much more extended and undetected .
since the object is located at intermediate galactic latitude ( @xmath40 ) , the pulsar is likely to have a high velocity .
then it is possible that the extended nebular is a bow shock where the ram pressure from interstellar matter and the pulsar wind pressure are in equilibrium : @xmath41 where @xmath42 is the spin - down luminosity assumed to be the same as the wind luminosity , @xmath43 the nucleon density of ambient medium , @xmath44 the proton mass , @xmath45 the pulsar velocity , @xmath46 the distance to the pulsar , @xmath47 the ratio of gamma - ray luminosity to spin - down luminosity , @xmath48 the observed gamma - ray flux ( @xmath49 ) , @xmath50 the radius of the bow shock , and @xmath51 .
this equation suggests a reasonable density of ambient matter as @xmath52 although it might be higher than expected for the mid galactic latitude region .
we must await further multi - band optical observation to clarify whether the extended object is a galaxy or a synchrotron nebula . in the following we assume that c+d is not related to rx j1836.2 + 5925/3eg j1835 + 5918 .
the flux of possible point sources a and b sets an upper limit of @xmath53 for point sources , which is much stronger than previous limit of @xmath13 .
this translates to the lower limit of @xmath3 on rx j1836.2 + 5925 .
if we assume the reyleigh - jeans spectrum ( @xmath54 ) as expected for thermal emission from an isolated neutron star , the limit on the x - ray / v ratio becomes @xmath55 5300 .
( however , it should also be noted that the optical spectrum of isolated neutron stars is often not entirely the black - body , as is the case for the geminga . )
this limit on x - ray / optical flux ratio further strengthens the argument for the isolated neutron - star identification for rx j1836.2 + 5925 .
star counts in such a faint magnitude are not well constrained , but a typical theoretical model predicts about 1000015000 stars per square degree down to @xmath56 to the direction of rx j1836.2 + 5925 , with an extreme assumption that ancient white dwarfs contribute 100% of the dark halo of our galaxy ( robin , reyl , & crze 2000 ; a. robin , private communication ) .
then the expected star count in the @xmath23-radius error circle is about 0.03 stars , and hence a random coincidence of a star in the error circle of rx j1836.2 + 5925 seems to be unlikely . on the other hand
, we can not reject a possibility that the point - like objects a and/or b are small galaxies which could not be resolved .
keeping these caveats in mind , we assume that one of the two is a real isolated pulsar and the physical counterpart of rx j1836.2 + 5925 , and then discuss implications in this case . to examine the physical connection between rx j1836.2 + 5925 and 3eg j1835 + 5918 , we should also estimate the chance probability of finding an isolated neutron star within the error circle of 3eg j1835 + 5918 ( @xmath57 radius ) .
based on the estimations of pulsar birth rate ( e.g. , narayan & ostriker 1990 ) and supernova rate ( e.g. , totani , sato , & yoshii 1996 ) , about @xmath58 neutron stars are typically expected to exist in our galaxy .
thermal emission can be observed by either cooling neutron star surfaces or those heated by accretion of interstellar matters .
neuhuser & trmper ( 1999 ) compiled observational constraints on the x - ray counts of such neutron stars detectable by the rosat all sky survey , and compared them with theoretical expectations .
since neutron stars rapidly cool down after @xmath59yr according to the standard cooling theory of neutron stars ( friedman & pandharipande 1981 ) , the expected count of cooling neutron stars is much smaller than that of the accreting neutron stars .
the former is @xmath34 110 sr@xmath60 for rosat x - ray flux of @xmath61 , and the expected number in the 12@xmath62-radius error circle of 3eg j1835 + 5918 is found to be sufficiently small as @xmath63@xmath64 .
the estimate for accreting neutron stars is rather uncertain and it could be much higher than that for cooling neutron stars , but the observational upper bound by rosat is @xmath65 1001000 sr@xmath60 again at @xmath66 , and the expected count in the egret error circle should be less than @xmath34 0.04 .
it should also be noted that 3eg j1835 + 5918 is at intermediate galactic latitude .
then it seems unlikely that it is heated by accretion , since the ambient gas density should not be high and the neutron star should have high - velocity proper motion .
( the bondi - hoyle accretion rate scales as @xmath67 . ) now we discuss the broad - band spectral energy distribution of this source , assuming that 3eg j1835 + 5918 and rx j1836.2 + 5925 are physically associated and it is a gamma - ray pulsar . assuming the thermal flux of @xmath1 from the neutron star surface , we can derive thermal temperature and distance , by scaling from the proto - type gamma - ray pulsar geminga .
the thermal x - ray spectrum of geminga can be fitted with @xmath68k ( halpern & wang 1997 ) , and its optical @xmath18-band flux ( @xmath69 , mignani , caraveo , & bignami 1998 ) is about a factor of 5.8 higher than the extrapolation from the x - ray flux .
such a trend , i.e. , optical flux that is a factor of several higher than that extrapolated from x - ray temperature , has been known also for another isolated neutron star , rx j185635@xmath703754 ( walter & matthews 1997 ) . assuming this factor also for rx j1836.2 + 5925
, we obtain the black - body temperature of @xmath71k from the x - ray / optical flux ratio , assuming @xmath72 . if the effective emission area of thermal radiation is the same , the optical flux in the rayleigh - jeans region scales as @xmath73 . therefore , the distance to rx j1836.2 + 5925 can be inferred as @xmath74pc , from the known distance of geminga ( @xmath75pc , caraveo et al .
1996).^{1/2}$ ] = 4.6 km . ] the inferred black - body spectrum of rx j1836.2 + 5925/3eg j1835 + 5918 is shown in terms of absolute @xmath76 luminosity by the solid line in fig .
[ fig : sed ] , and the observed gamma - ray , x - ray and optical luminosities or upper limits are also shown by open circles or solid lines .
the error bar of the soft x - ray data point ( 0.2kev ) is showing the flux change when different column densities of @xmath77 or @xmath78 are assumed , with the same temperature .
the upper bound on hard x - ray flux ( @xmath791kev ) is from asca observation ( mirabal et al . 2000 ) . for comparison ,
the black - body spectrum and observed luminosities of geminga in optical , soft and hard x - ray , and gamma - ray bands are shown by dashed lines or crosses .
a striking similarity of the spectral energy distribution between geminga and rx j1836.2 + 5925/3eg j1835 + 5918 can be seen .
the no detection of nonthermal x - rays from rx j1836.2 + 5925 is marginally consistent with the asca upper bound , if the nonthermal x - ray luminosity is similar to that of geminga . assuming the distance of 300pc and an age of the pulsar to be @xmath80yr , which is similar to that of geminga
, the galactic latitude of @xmath81 implies a reasonable transverse proper motion of @xmath82 km / s
. to summarize , we found an extended source having a peculiar shape ( c+d ) at @xmath83 within the error circle of rx j1836.2 + 5925 , which has been suspected as an isolated pulsar associated with 3eg j1835 + 5918 .
this source might be a pulsar nebular or knot whose luminosity is about 56 orders of magnitude lower than the gamma - ray luminosity .
however , it is statistically well possible that a galaxy appeared by chance as the source c+d within the examined error circle .
there are two fainter possible point sources ( a , b ) at @xmath1 , although they are not firm detections because of low signal - to - noise .
it is interesting that if either of a or b is a real counter part of rx j1836.2 + 5925 and thermal emission from a neutron star , its optical , x , and gamma - ray luminosities are very similar to those of the proto - type gamma - ray pulsar geminga .
the pulsar identification of geminga came from the detection of pulsation in x - ray flux , and hence it can be imagined that the ultimate identification of the enigmatic source 3eg j1835 + 5918 will also be brought by discovery of x - ray pulsation . on the other hand ,
further deep optical study is also very important to verify whether the extended source c+d is a galaxy or not , to confirm the possible point sources a and/or b with higher signal - to - noise ratio , and to measure proper motion and parallax for distance determination .
we would like to thank l. eyer and a. robin for useful information for star counts .
tt has been financially supported in part by the postdoctoral fellowship of the japan society for the promotion of science ( jsps ) for research abroad .
wk is supported in part by jsps research fellowship . | we report a deep optical imaging observation by the subaru telescope for a very soft x - ray source rx j1836.2 + 5925 , which has been suspected to be an isolated neutron star associated with the brightest as - yet unidentified egret source outside the galactic plane , 3eg j1835 + 5918 .
an extended source having a complex , bipolar shape is found at @xmath0 , and this might be an extended pulsar nebular whose flux is about 5 - 6 orders of magnitude lower than gamma - ray flux , although finding a galaxy of this magnitude by chance in the error circle is of order unity .
we have found two even fainter , possibly point sources at @xmath1 , although their detections are not firm because of low signal - to - noise .
if the extended object of @xmath2 is a galaxy and not related to 3eg j1835 + 5918 , a lower limit on x - ray / optical flux ratio is set as @xmath3 , giving a further strong support of the neutron - star identification of 3eg j1835 + 5918 .
interestingly , if either of the two sources at @xmath1 is the real counterpart of rx j1836.2 + 5925 and thermal emission from the surface of an isolated neutron star , the temperature and distance to the source become @xmath4k and @xmath5pc , respectively , showing a striking similarity of its spectral energy distribution to the proto - type radio - quiet gamma - ray pulsar geminga .
no detection of nonthermal hard x - ray emission is consistent with the asca upper limit , if the nonthermal flux of 3eg j1835 + 5918/rx j1836.2 + 5925 is at a similar level with that of gemiga . |
the numerical calculations have been performed on the hitachi sr8000 at lrz ( munich ) , on the cray t3e at nic ( jlich ) and zib ( berlin ) , as well as on the ape1000 and quadrics at desy ( zeuthen ) .
we thank all institutions for their support .
this work has been supported in part by the eu integrated infrastructure initiative hadron physics ( i3hp ) and by the dfg ( forschergruppe gitter - hadronen - phnomenologie ) .
a. a k acknowledges support by `` berliner programm zur frderung der chancengleichheit fr frauen in forschung und lehre '' . 9 c.r .
allton et al . , phys .
d65 ( 2002 ) 054502 .
s. capitani et al . , nucl .
79 ( 1999 ) 548 . m. gckeler et al . , nucl .
suppl . 119
( 2003 ) 32 .
a. ali khan et al . , nucl .
b689 ( 2004 ) 175 .
hemmert , m. procura and w. weise , phys .
d68 ( 2003 ) 075009 .
s. sasaki , k. orginos , s. ohta and t. blum , phys .
d68 ( 2003 ) 054509 .
beane and m.j .
savage , hep - ph/0404131 . | we present preliminary results for the axial and tensor charge of the nucleon obtained from simulations with @xmath0 clover fermions .
a comparison with chiral perturbation theory is attempted .
lattice qcd offers the possibility to investigate the internal structure of the nucleon without additional model assumptions . in particular , one can study how the mass , the axial charge and further properties of the nucleon change when the quark mass is varied or the nucleon is squeezed into a finite spatial box . even if we could simulate at the physical quark mass and in a huge volume
, we would like to examine the dependence on the quark mass and on the volume , because it contains important information on low - energy qcd .
this information can be extracted by comparing the outcome of lattice qcd simulations with parameterisations provided by chiral effective field theories ( cheft ) .
this comparison with cheft , if successful , leads to two important achievements : firstly , it gives us control over the chiral extrapolation and the thermodynamic limit of the simulation results and secondly it allows us to determine phenomenologically relevant coupling constants in cheft . in the following ,
we shall present preliminary results from simulations with @xmath1 non - perturbatively improved wilson quarks and the standard plaquette action for the gauge fields obtained by the qcdsf and ukqcd @xcite collaborations .
previous quenched qcdsf results can be found in ref .
@xcite and for a review see ref . @xcite . for the time being we do not yet perform an extrapolation to the continuum limit and neglect lattice artefacts .
originally , cheft was invented to describe the quark mass dependence of low - energy quantities by means of effective hadron fields .
however , for sufficiently large volumes and sufficiently small pion masses , finite size effects are dominated by pions `` propagating around the world '' , and therefore cheft can also be used to calculate the volume dependence : the effective lagrangian is independent of the volume .
our paradigm is the analysis of the nucleon mass @xcite , where we could describe the pion mass dependence and the finite size effects up to surprisingly large masses by formulae derived from relativistic baryon chiral perturbation theory .
we are trying to perform a similar analysis also for the axial and the tensor charge of the nucleon , and in the following we shall report on the status of these attempts .
whenever we need numbers in physical units , we shall set the scale using @xmath2 .
compared with the case of the nucleon mass there are several additional difficulties when dealing with such matrix elements of composite operators : * the cheft calculations are not as advanced as for the nucleon mass . * in order to suppress lattice artefacts , the operators should be improved . however , the required improvement coefficients are not yet known non - perturbatively . in the following , we shall present results from the unimproved operators .
* the operators have to be renormalised .
we shall use non - perturbative renormalisation factors computed by means of the rome - southampton method with a linear chiral extrapolation at fixed bare gauge coupling . *
the quark - line disconnected contributions to the matrix elements are hard to evaluate .
therefore we shall consider only flavour - nonsinglet quantities , where these contributions cancel . in order to compute the axial form factor of the nucleon @xmath3 on the lattice we use the flavour - nonsinglet axial vector current @xmath4 and the form factor decomposition of its proton matrix element .
using the so - called small scale expansion , which takes @xmath5 degrees of freedom explicitly into account , the authors of ref .
@xcite have studied the mass dependence of the nucleon axial charge @xmath6 to @xmath7 and compared with quenched data . in fig .
[ fig.ga ] we perform a similar comparison with our presently available dynamical data for @xmath8 choosing some reasonable values for the parameters .
the curve reproduces the weak mass dependence of the simulation results as well as the physical point , at least approximately , but the chiral logarithm dominates only for very small pion masses .
what about the finite size effects for @xmath8 ? there are indications that they could be quite large ( for a quenched study see , e.g. , ref .
@xcite ) , and they might not yet be negligible for the results in fig .
[ fig.ga ] .
we have performed simulations on three different volumes at the same bare gauge coupling and quark mass , leading to @xmath9 on the largest lattice , and found a clear volume dependence .
first calculations within cheft have appeared @xcite . to @xmath10 in heavy baryon chiral perturbation theory ( without explicit @xmath5s ) one finds @xmath11 \displaystyle \quad { } - \frac{g_a^0 m_\pi^2}{4 \pi^2 f_\pi^2 } \left ( 1 + { \mbox{\small $ \frac{2}{3}$}}(g_a^0)^2 \right ) \sum_{\vec{n } } { } ^\prime \frac{k_1(l |\vec{n}| m_\pi)}{l |\vec{n}| m_\pi } \ , , \end{array}\ ] ] where @xmath12 is the value of @xmath8 in the chiral limit and @xmath13 is the pion decay constant . in fig . [ fig.gafs ] we plot our data together with the above expression for the finite size effect .
not even the sign is reproduced correctly .
of course , one may blame the large pion mass in connection with the low order of chiral perturbation theory for this discrepancy , and further investigations are clearly needed .
the tensor charge of the nucleon is the lowest moment of the transversity distribution @xmath14 . for flavour @xmath15
it is given by the proton matrix element @xmath16 with @xmath17 .
again , we restrict ourselves to the flavour nonsinglet combination @xmath18 and plot it in fig .
[ fig.combi ] together with @xmath8 data , because in the non - relativistic limit one expects that both quantities agree .
hence this comparison gives us some information on how relativistic the quarks in our simulations are and the data seem to indicate that the quarks are not very relativistic .
concluding we may say that the interpretation of our results is only at its beginning .
once we have analysed our configurations completely , we hope to gain more insight into the structure of the nucleon .
but further improvements are certainly desirable , in particular smaller quark masses and lattice spacings in the monte carlo simulations as well as more extensive calculations in cheft . |
we will study an infinite system of interacting brownian motions with , @xmath1 , denoting the position of the @xmath2-th brownian particle on the real line at time @xmath3 .
initially the positions are ordered as @xmath4 , with the convention that @xmath5 .
as indicated in the title , particle @xmath6 interacts through a steep , narrowly supported potential with its left neighbor , @xmath2 , only . in the limit of zero support , the singular limit studied in our contribution , this interaction amounts to brownian motion @xmath6
being reflected from brownian motion @xmath2 .
a mathematical definition will be given below .
under this dynamics the order is preserved , @xmath7 for all times @xmath8 . in previous work
we investigate the case of initial conditions with equal spacing , @xmath9 @xcite .
another natural initial condition is such to have the process space - time stationary , which is accomplished by assuming that @xmath10 is a poisson point process with uniform intensity which , without loss of generality , can be taken as @xmath11 .
then @xmath12 is again an intensity 1 poisson point process .
our interest are the fluctuations of @xmath13 for large @xmath3 and @xmath2 . to understand their properties one first has to find out how an initially small perturbation close to the origin propagates in time .
this path is known as characteristic . in our model , because of the one - sided collisions , the characteristic turns out to be a straight line with velocity 1 .
if @xmath14 , @xmath15 denoting integer part , then for @xmath16 only the randomness of the initial conditions plays a role and the fluctuations of @xmath13 will be gaussian asymptotically on the @xmath17 scale .
however close to the characteristic , i.e. , @xmath18 with @xmath19 , one observes non - gaussian fluctuations in the @xmath20 scale , the properties of which will be analysed in great detail in this contribution . with our methods we can handle the stochastic process in @xmath21 at fixed @xmath3 .
two - time properties along the characteristic are known to be difficult .
for example , a long - standing problem is to obtain the joint distribution of @xmath22 for large @xmath3 . at the time of writing johansson reports on asymptotic results for the model studied in this paper @xcite .
our results are closely linked to the one - dimensional kardar - parisi - zhang ( kpz ) equation @xcite with stationary initial data .
kpz is a stochastic pde for a height function @xmath23 and reads @xmath24 with @xmath25 space - time white noise . as written the equation is only formal , but a precise mathematical meaning has been given @xcite .
as random initial data @xmath26 we choose the statistics of two - sided brownian motion with constant drift @xmath27 .
the dynamics is stationary in the sense that @xmath28 is again two - sided brownian motion with drift @xmath27 @xcite .
very recently , borodin _ et al . _
@xcite succeeded in writing down reasonably concise formulas for the distribution of @xmath29 , confirming the prior replica computation @xcite . through an intricate asymptotic analysis
they establish ( theorem 2.17 of @xcite ) that in distribution , for fixed @xmath21 , @xmath30 where @xmath31 denotes the airy process corresponding to stationary initial data . in spirit one should think of @xmath13 as @xmath29 with @xmath32 being a continuum version of the discrete particle label @xmath2 .
more precisely , as one of our results we will establish in theorem [ thmasymp0 ] that @xmath33 which is the immediate analogue of .
in fact , convergence is proved in the sense of finite - dimensional distribution , not only for the one - point distribution .
similar results have been obtained earlier for the stationary png model @xcite and for the stationary tasep @xcite . for the latter ,
the full stochastic process in @xmath21 has been worked out @xcite .
the expression we obtain for the joint distributions of @xmath31 , see definition [ defastat ] , is new and differs from the one in @xcite . for several reasons we believe that it is of interest to add a third model to the list of kpz type models with stationary initial conditions
obviously , the universality hypothesis is further strengthened .
more importantly our model provides a bridge to diffusion processes with one - sided interaction as discussed in @xcite . besides we also have to develop a method different from the previous ones . as in the case of png and tasep
, one can not study the stationary initial conditions directly .
one has to start from a step , in our case meaning that to the right of @xmath34 the poisson point process has density @xmath11 , while to the left it has density @xmath35 , @xmath36 .
surprisingly , as for png and tasep , by a burke type theorem the left - half system can be replaced by a boundary condition for @xmath37 and , in fact , only the right - half system with labels @xmath38 has to be considered . for png and tasep the limit
@xmath39 has been accomplished for fixed @xmath3 , while here we first take the limit @xmath0 at step size @xmath40 .
this leads us to a novel transition process , see theorem [ thmasymp ] . the stationary case , @xmath41 ,
is then reached through a careful analytic continuation .
+ * acknowledgments*. the work of p.l .
ferrari is supported by the german research foundation via the sfb 1060b04 project .
the work of h. spohn is supported by the fondation science mathmatiques de paris .
the work of t. weiss is supported by the german research foundation project .
to state our result we introduce the rescaled process @xmath42 and define the limit process @xmath43 .
[ defastat ] let @xmath44 be the projection operator on @xmath45 and @xmath46 the one on @xmath47 .
set @xmath48 and define @xmath49 as well as an operator @xmath50 with integral kernel @xmath51 further , define the functions @xmath52 with these definitions , set @xmath53 where @xmath54 denotes the inner product on @xmath55 .
then , the _ airy@xmath56 process _ , @xmath31 , is the process with @xmath57-point joint distributions at @xmath58 given by @xmath59 we can now state our main result .
[ thmasymp0 ] in the sense of finite - dimensional distributions , @xmath60 the joint distributions of the _ airy@xmath56 process _
were first obtained in @xcite , see definition 1.1 and theorem 1.2 therein . in definition [ defastat ] we state an alternative formula for the joint distributions of the airy@xmath61 process .
the main difference between the two formulas is that in @xcite the joint distributions are given in terms of a fredholm determinant on @xmath62 , while here we have a fredholm determinant on @xmath55 .
a similar twist was already visible in @xcite and has been generalized in @xcite .
since @xmath12 is a poisson point process , the process @xmath63 is a scaled poisson jump process up to a linear part and @xmath64 hence the limit process @xmath65 must also have the statistics of two - sided brownian motion , a property which is not so easily inferred from our formulas in definition [ defastat ] .
but we will provide a direct proof of this fact in section [ sectgaussianincr ] .
note that @xmath66 and @xmath67 are not independent . as already familiar from other models in the kpz universality class @xcite , the proof of theorem [ thmasymp0 ] proceeds via a sequence of approximating initial conditions .
firstly we consider the case where @xmath68 and assume that the particles on @xmath69 is a poisson process with intensity @xmath70 and on @xmath71 is a poisson process with intensity @xmath72 .
in other words ,
@xmath73 , @xmath1 , with @xmath74 as explained in lemma [ lemma7.1 ] , setting @xmath75 will induce a difference of order one as compared to the case considered in theorem [ thmasymp0 ] . in the scaling limits such differences are irrelevant .
thus it is enough to prove theorem [ thmasymp0 ] for the initial conditions with @xmath76 .
_ in the sequel @xmath13 always refers to the initial conditions , in such a way that the choice of the parameters @xmath77 can be inferred from the context_. we obtain the fixed time multi - point distributions of the system @xmath78 in terms of a fredholm determinant in the case @xmath79 .
the restriction to non - negative integers comes from burke s theorem by which the particles with @xmath80 can be replaced by choosing @xmath37 as a brownian motion with drift @xmath35 .
[ propkernel ] let @xmath81 . for any finite subset @xmath82 of @xmath83
, it holds @xmath84 where @xmath85 . the kernel @xmath86 is given by @xmath87 where @xmath88 and @xmath89 for any fixed @xmath90 .
notice that this result holds for @xmath91 only and not for the most interesting case @xmath92 .
the latter can be accessed through a careful analytic continuation of the formulas .
one of the novelty of this paper is to perform the analytic continuation _ after _ the scaling limit .
this allows us to discover a new process , called _ finite - step airy@xmath56 process _ , describing the large time limit close to stationarity ( actually , one still needs to take care of the random shift of @xmath93 , which is however irrelevant as it goes to zero after scaling in the large time limit ) . as before
, this process is defined through its finite - dimensional distributions .
[ defairystat]the finite - step airy@xmath56 process with parameter @xmath94 , @xmath95 , is the process with m - point joint distributions at @xmath58 given by @xmath96 where @xmath97 and the kernel @xmath98 is defined by @xmath99 here , @xmath100 is defined as in ( [ defv ] ) , and @xmath101 as mentioned already above , we are going to take the limit to stationarity after the long time limit .
however , in general , the limits @xmath0 and @xmath102 do not commute .
therefore we have to consider @xmath103 ( to be able to apply proposition [ propkernel ] ) , but vanishing with a tuned scaling exponent as @xmath0 , a critical scaling .
we set @xmath104 for @xmath94 . as will be proven with this choice the limit @xmath0 commutes with @xmath105 .
such considerations lead naturally to define the rescaled process as @xmath106 where the superscript of @xmath32 indicates @xmath107 and @xmath108 .
the second main result of our paper is the description of the joint distributions of the rescaled process in the long time limit .
[ thmasymp ] for every @xmath94 , the rescaled process converges to the _ finite - step airy@xmath56 process _
@xmath109 in the sense of finite - dimensional distributions .
consider the initial conditions stated in ( [ statmodel ] ) .
first we show that the system with infinitely many particles is well - defined . for that purpose
we use the skorokhod representation @xcite to define the reflected brownian motions .
this representation is the following : the process @xmath110 , driven by the brownian motion @xmath111 , starting from @xmath112 and being reflected at some continuous function @xmath113 with @xmath114 is defined as : @xmath115 let @xmath116 , @xmath1 , be independent standard brownian motions starting at @xmath34 and define the random variables @xmath117 with the convention @xmath118 and @xmath119 .
we will define the system @xmath120 as the limit of half - infinite systems @xmath121 as @xmath122 , where @xmath123}\{y_{k , n}(t)+\zeta_k\},\quad n\geq - m.\ ] ] notice that these processes indeed satisfy the skorokhod equation , @xmath124 for @xmath125 , while the leftmost process is simply @xmath126 thus as desired @xmath127 is a brownian motion starting from @xmath128 and reflected off by @xmath129 for @xmath125 .
the representation ( [ eq3a ] ) can be sees as a zero - temperature case of the oconnell - yor semi - discrete directed polymer @xcite with appropriate boundary conditions ( see discussion at the end of this section ) .
next we show strong converge of the system @xmath121 to the processes we are studying . [ propmconv ] let us define @xmath130 then , for any @xmath131 , @xmath132}|x_n^{(m)}(t)-x_n(t)|=0,\quad \mathrm{a.s.}\ ] ] as well as @xmath133}|x_n(t)|<\infty,\quad \mathrm{a.s.}.\ ] ] for the proof of this proposition we first need the following concentration inequality , which is proposition 2.1 of @xcite : [ concin ] for each @xmath131 there exists a constant @xmath134 such that for all , @xmath94 , @xmath135 let us define an auxiliary system of processes , which we will use later in proving the burke property , by @xmath136 and @xmath137 for @xmath125 .
this system differs from @xmath127 just in the drift of the leftmost particle , which of course influences all other particles as well ( the choice of the extra drift is because the system with infinite many particles in @xmath71 generates a drift @xmath35 ) .
this system of particles satisfies @xmath138}\{y_{k , n}(t)+\zeta_k\}\big\},\ ] ] with @xmath139 also , we have the inequalities @xmath140 consider the event @xmath141 we can deduce exponential decay of @xmath142 in @xmath143 from combining the exponential tails of @xmath144 and @xmath128 with proposition [ concin ] , using @xmath145 and elementary inequalities . in particular @xmath146
, so by borel - cantelli , @xmath147 occurs only finitely many times almost surely .
this means , that a.s .
there exists a @xmath148 , such that for all @xmath149 : @xmath150 consequently , @xmath151 for all @xmath149 and therefore @xmath152 moreover , applying , gives @xmath153 resulting in @xmath154 repeating the same argument , we see that for every @xmath155 $ ] there exists @xmath156 such that @xmath157 .
applying lemma [ lemmaxpath ] then gives @xmath158 for every @xmath155 $ ] .
this settles the convergence .
to see , which is equivalent to @xmath159}|x_m^{(m_t)}(t)|<\infty$ ] , we apply the bound @xmath160 [ lemmaxpath ] consider @xmath161 and @xmath57 , @xmath162 , @xmath163 such that @xmath164 then @xmath165 this is a straightforward generalization of lemma 3.2 @xcite .
we establish a useful property which will allow us to study our system of interacting brownian motions through a system with a left - most brownian particle .
[ propbm ] for each @xmath166 , the process @xmath167 is a standard brownian motion .
[ remarkburke ] proposition [ propbm ] allows us to restrict our attention to the half - infinite system .
in fact , conditioned on the path of @xmath168 , the systems of particles @xmath169 and @xmath170 are independent , as it is clear by the definition of the system .
then ( [ eq2.9 ] ) implies that the law of is the same as the one obtained replacing the infinitely many particles @xmath171 with a single brownian motion @xmath37 which has a drift @xmath35 .
this property will be used to derive our starting result , proposition [ propkernel ] .
first notice that @xmath172 is a brownian motion .
now assume @xmath173 is a brownian motion . by definition ,
@xmath174 which allows us to apply proposition [ propneil ] , i.e. , we have that @xmath175 is a brownian motion . since @xmath176 the proof is completed .
it is clear , that in the case @xmath92 the process is a brownian motion for @xmath177 , too , i.e. , the system is stationary in @xmath2 .
we also have stationarity in @xmath3 , in the sense that for each @xmath178 the random variables are independent and distributed exponentially with parameter @xmath35 .
the following result is a small modification of theorem 2 in @xcite .
[ propneil ] fix @xmath72 and let @xmath111 , @xmath179 be standard brownian motions , as well as @xmath180 , independent . define the process @xmath181 then @xmath182 is distributed as a standard brownian motion .
extend the processes @xmath111 , @xmath179 to two - sided brownian motions indexed by @xmath183 . defining @xmath184 and @xmath185
we can apply theorem 2 @xcite , i.e. , @xmath186 is a brownian motion .
now , @xmath187 so by lemma [ lembmexp ] it has exponential distribution with parameter @xmath35 .
as it is independent of the processes @xmath188 we can write @xmath189 . dividing the supremum into @xmath190 and @xmath191 we arrive at : @xmath192 which is up to a sign flip of @xmath193 .
[ lembmexp ] fix @xmath72 and let @xmath111 be a standard brownian motion
. then @xmath194 the random variable @xmath195 is distributed as a brownian motion with drift @xmath196 started at zero and being reflected ( upwards ) at zero , at time @xmath3 .
as @xmath0 , this converges to the stationary distribution of this process , which is the exponential distribution with parameter @xmath197 . from a stochastic analysis point of view , the system satisfies @xmath198 here @xmath199 , @xmath200 , are continuous non - decreasing processes increasing only when @xmath201 .
in fact , @xmath199 is twice the semimartingale local time at zero of @xmath202 .
notice that @xmath203 is a standard brownian motion independent of @xmath204 , but not equal to @xmath205 .
one can also view the system as a model for last passage percolation ( or zero - temperature semi - discrete directed polymer ) .
we assign random background weights on the set @xmath206 in the following way : * white noises @xmath207 on the lines @xmath208 for @xmath209 , * white noise @xmath210 plus a lebesgue measure of density @xmath35 on the line @xmath211 , and * dirac measures of magnitude @xmath212 on @xmath213 for @xmath209 .
[ lb]@xmath34 [ lb]@xmath11 [ lb]@xmath214 [ lb]@xmath215 [ lb]@xmath216 [ lb]@xmath217 [ lb]@xmath218 [ lb]@xmath219 [ lb]@xmath220 [ lb]@xmath3 ( thick black ) and the random background noise ( grey).,title="fig:",height=188 ] an up - right path in @xmath221 is characterized by its jumping points @xmath222 and it consists of line segments @xmath223\times \{n\}$ ] , see figure [ figdp ] .
the set of up - right paths can then be parameterized by @xmath224 the _ percolation time _ or _
weight _ of a path @xmath225 is the integral over the background weights along the path .
explicitly , we have : @xmath226 for @xmath227 , and for @xmath228 , @xmath229 the _ last passage percolation time _ is given by the supremum over all paths : @xmath230 the supremum is almost surely attained by a unique path @xmath231 , called the maximizer .
it exists because the supremum can be rewritten as a composition of a finite maximum and a supremum of a continuous function over a compact set .
uniqueness follows from elementary properties of the brownian measure .
most importantly , from the definition , we have @xmath232 we will use this interpretation in section [ sectancont ] , however , it also gives some connections to different works .
our model can be seen as the semi - continuous limit of a more widely studied discrete last passage percolation model ( see for example @xcite ) .
moreover , our last passage percolation model is the zero temperature limit of a directed polymer model , which has been studied thoroughly in the recent past @xcite . for later use
we also define a version without boundary weights : @xmath233
in order to prove proposition [ propkernel ] we first start by considering the transition probability for a finite number of reflecting brownian motions with drifts and arbitrary initial positions ( proposition [ propwarren ] ) . then we will set the drift of the first brownian motion to @xmath35 , see remark [ remarkburke ] , and we will randomize the initial positions ( proposition [ proprandomic ] ) .
proposition [ propwarren ] generalizes proposition 4.1 @xcite , which has been first shown in @xcite , to the case of non - zero drifts .
[ propwarren ] the transition probability density of @xmath234 one - sided reflected brownian motions with drift @xmath235 from @xmath236 to at time @xmath237 has a continuous version , which is given as follows .
@xmath238 where @xmath239,\ ] ] and @xmath240 with @xmath241 .
we follow the proof of proposition 8 in @xcite .
the strategy is to show that the transition density satisfies three equations , the backwards equation , boundary condition and initial condition , the latter one being contained in lemma [ lemincond ] .
these equations are then used for it s formula to prove that it indeed is the transition density .
we start with the backwards equation and boundary condition : @xmath242 to see , move the prefactor @xmath243 inside the integral in the th row of the determinant and notice that the differential operator transforms @xmath244 into @xmath245 .
consequently , @xmath246 implies the @xmath247-th being the negative of the @xmath248-th row .
can be obtained by the computation @xmath249 let be a @xmath250 function , whose support is compact and has a distance of at least some @xmath251 to the boundary of @xmath252 .
define a function @xmath253 as @xmath254 the previous identities and carry over to the function @xmath255 in the form of : @xmath256 our processes satisfy @xmath257 , where @xmath116 are independent brownian motions , @xmath258 and @xmath199 , @xmath200 , are continuous non - decreasing processes increasing only when @xmath259 .
in fact , @xmath199 is twice the semimartingale local time at zero of @xmath202 .
now fix some @xmath251 , @xmath131 , define a process @xmath260 for @xmath261 $ ] and apply it s formula : @xmath262 from the definition it follows that @xmath263 and @xmath264 because continuous functions of finite variation do not contribute to the quadratic variation . inserting the differentials , by the integrals with respect to @xmath265 integrals cancel , which results in : @xmath266 since the measure @xmath267 is supported on @xmath268 , where the spatial derivative of @xmath255 is zero ( see ( [ eqboundcond2 ] ) ) , the last term vanishes , too .
so @xmath269 is a local martingale and , being bounded , even a true martingale . in particular
its expectation is constant , i.e. : @xmath270={\mathbbm{e}}\left[f\left({\varepsilon},\vec{x}(t)\right)\right].\ ] ] applying lemma [ lemincond ] we can take the limit @xmath271 , leading to @xmath272.\ ] ] because of the assumptions we made on @xmath273 it is still possible that the distribution of @xmath274 has positive measure on the boundary .
we thus have to show that @xmath275 is normalized over the interiour of the weyl chamber : start by integrating over @xmath276 .
pull the prefactor indexed by @xmath277 as well as the integration inside the @xmath278 column of the determinant .
the @xmath279 entry is then given by : @xmath280 the contribution from @xmath281 is a constant multiple of the second column and thus cancels out .
the remaining terms are zero for @xmath282 , since all these functions @xmath283 have gaussian decay .
the only non - vanishing term comes from @xmath284 and returns exactly @xmath11 by an elementary residue calculation .
the determinant can thus be reduced to the index set @xmath285 . successively carrying out the integrations of the remaining variables in the same way
, we arrive at the claimed normalization .
this concludes the proof .
[ lemincond ] for fixed @xmath286 , the transition density @xmath275 as given by , satisfies @xmath287 for any @xmath250 function , whose support is compact and has a distance of at least some @xmath251 to the boundary of @xmath252 .
at first consider the contribution to the determinant in coming from the diagonal .
for @xmath288 the products in cancel out , so we are left with a simple gaussian density .
this contribution is thus given by the multidimensional heat kernel , which is well known to converge to the delta distribution .
the remaining task is to prove that for all other permutations the integral vanishes in the limit .
let @xmath289 be such a permutation .
its contribution is @xmath290 where we have extended the domain of @xmath273 to @xmath291 , being identically zero outside of @xmath292 .
we also omitted the prefactor since it is bounded for @xmath293 in the compact domain of @xmath273 .
there exist @xmath294 with @xmath295 .
let @xmath296 it is enough to restrict the area of integration to these two sets , since on the complement of @xmath297 , we have @xmath298 so we are not inside the support of @xmath273 .
we start with the contribution coming from @xmath299 .
notice that by @xmath300 all functions @xmath244 with @xmath301 can be written as iterated derivatives of @xmath302 and some exponential functions .
for each @xmath303 with @xmath304 we write @xmath305 in this way and then use partial integration to move the exponential factors and derivatives onto @xmath273 .
the result is @xmath306 for a new @xmath250 function @xmath307 , which has compact support and is therefore bounded , too .
we can bound the contribution by first integrating the variables @xmath308 with @xmath309 , @xmath303 , where we have a gaussian factor @xmath302 : @xmath310 @xmath311 consists of the yet to be integrated @xmath293-components that are contained in the set . in particular , @xmath311 is compact , so the functions @xmath305 , @xmath303 , are bounded uniformly in @xmath237 by lemma [ lemfkl ] .
the remaining integral gives : @xmath312 which converges to @xmath34 as @xmath313 by .
the contribution of @xmath314 can be bounded analogously with @xmath315 playing the role of @xmath316 .
the final convergence is then given by .
[ lemfkl ] for each @xmath251 we have @xmath317 in addition , for each @xmath318 the function @xmath319 is bounded uniformly in @xmath237 on compact sets .
let @xmath320 , and choose a @xmath321 which is positive . by a transformation of variable we have @xmath322 where @xmath323 denotes a bound on the fraction part of the integrand , which grows at most polynomial in @xmath324 .
convergence of the integral is ensured by the exponential term , so integrating and taking the limit @xmath313 gives .
to see , notice that by @xmath325 the integrand has no poles , so we can shift the contour to the right , such that @xmath321 is negative , and obtain the convergence analogously .
we are left to prove uniform boundedness of @xmath244 on compact sets for @xmath326 .
for @xmath327 we can use to get @xmath328 for @xmath329 . in the case
@xmath330 we shift the contour to negative @xmath321 , thus obtaining contributions from residua as well as from the remaining integral .
the latter can be bounded as before , while the residua are well - behaved functions , which converge uniformly on compact sets . to obtain a representation as a signed determinantal point process we have to introduce a new measure .
this measure @xmath331 coincides with @xmath332 on the sigma algebra which is generated by @xmath333 , @xmath334 , and the driving brownian motions @xmath335 , @xmath334 .
but under @xmath331 , @xmath336 is a random variable with an exponential distribution instead of being fixed at zero .
formally , , with @xmath337 giving rise to , so that @xmath332 is the result of conditioning @xmath331 on the event @xmath338 . this new measure satisfies a determinantal formula for the joint distribution at a fixed time .
[ proprandomic ] under the modified initial condition specified by @xmath331 , the joint density of the positions of the reflected brownian motions is given by @xmath339\,{\mathrm{d}}\vec{\xi } \end{aligned}\label{prnd}\ ] ] with @xmath340 for the related model , the totally asymmetric simple exclusion process , a formula similar to the one of proposition [ proprandomic ] also exists @xcite .
here we provide a direct proof of it .
the fixed time distribution can be obtained by integrating the transition density over the initial condition .
denote by @xmath341 the probability density of @xmath342 , i.e. , @xmath343 .
@xmath344 equals @xmath345\\ = & ( \lambda-\rho)\lambda^{n-1}e^{-{t}\rho^2/2+\rho\xi_0}\int_{w^n\cap\{\zeta_0>0\}}{\mathrm{d}}\vec{\zeta}\ , e^{\lambda\zeta_{n-1 } } \\ & \quad\times\det_{1\leq k ,
l\leq n}\left[\frac{1}{2\pi{{\rm i}}}\int_{{{\rm i}}{\mathbb{r}}+\mu}{\mathrm{d}}w_k\,e^{{t } w_k^2/2}e^{\xi_{n - l}w_k}e^{-\zeta_{n - k}w_k}w_k^{k - l}\right]\\ = & ( \lambda-\rho)\lambda^{n-1}e^{-{t}\rho^2/2+\rho\xi_0}\sum_{\sigma\in s_n}(-1)^{|\sigma|}\prod_{k=1}^n\int_{{{\rm i}}{\mathbb{r}}+\mu}\frac{{\mathrm{d}}w_k}{2\pi{{\rm i}}}\,e^{{t } w_k^2/2}e^{\xi_{n-\sigma(k)}w_k}w_k^{k-\sigma(k)}\\ & \quad\times\int^{\infty}_0d\zeta_0\dots\int^{\infty}_{\zeta_{n-2}}d\zeta_{n-1}\,e^{-\lambda\zeta_{n-1}}e^{-\zeta_{n-1}w_1}e^{-\zeta_{n-2}w_2}\dots e^{-\zeta_0w_n}\\ = & ( \lambda-\rho)\lambda^{n-1}e^{-{t}\rho^2/2+\rho\xi_0}\sum_{\sigma\in s_n}(-1)^{|\sigma|}\prod_{k=1}^n\int_{{{\rm i}}{\mathbb{r}}+\mu}\frac{{\mathrm{d}}w_k}{2\pi{{\rm i}}}\,\frac{e^{{t } w_k^2/2}e^{\xi_{n-\sigma(k)}w_k}w_k^{k-\sigma(k)}}{w_1+\dots+w_k+\lambda}. \end{aligned}\ ] ] since all @xmath346 are integrated over the same contour , we can replace @xmath346 by @xmath347 : @xmath348 we apply lemma [ lemdetident ] below to the sum and finally obtain @xmath349\\ & = ( \lambda-\rho)\lambda^{n-1}e^{-{t}\rho^2/2+\rho\xi_0}\det_{1\leq k , l\leq n}\left[\widetilde{f}_{k - l}(\xi_{n - l},{t})\right ] .
\end{aligned}\ ] ] [ lemdetident ] given @xmath350 , @xmath70 and @xmath351 , the following identity holds : @xmath352.\ ] ] we use induction on @xmath234 . for @xmath353
the statement is trivial . for arbitrary @xmath234 ,
rearrange the left hand side of as @xmath354 where we applied the induction hypothesis to the second sum .
further , @xmath355+\lambda\det_{1\leq k , l\leq n}\left[w_l^{k-1}\right]\right ) , \end{aligned}\ ] ] with @xmath356 being the kronecker delta . inserting the identity
@xmath357=(w_1+\dots+w_n)\det_{1\leq k ,
l\leq n}\left[w_l^{k-1}\right],\ ] ] we arrive at @xmath358=\det_{1\leq k ,
l\leq n}\bigg[\frac{w_l^k}{w_l+\lambda}\bigg].\ ] ] to show we introduce the variable @xmath359 and consider the factorization @xmath360=\prod_{i=1}^n(w_{n+1}-w_i)\det_{1\leq k , l\leq n}\left[w_l^{k-1}\right],\ ] ] which follows directly from the explicit formula for a vandermonde determinant . expanding the determinant on the left hand side along the @xmath361-th column
gives an explicit expression in terms of monomials in @xmath359 .
examining the coefficient of @xmath362 on the left and right hand side respectively provides .
we can rewrite the measure in proposition [ proprandomic ] in terms of a conditional ( see lemma 3.4 of @xcite reported here as lemma [ lemdetmeasure ] ) and obtain a fredholm determinant expression for the joint distribution of any subsets of particles position .
then it remains to relate the law under @xmath331 and @xmath332 , which is the law of the reflected brownian motions specified by the initial condition .
this is made using a _
shift argument _
, analogue to the one used for the polynuclear growth model with external sources @xcite or in the totally asymmetric simple exclusion process @xcite .
the proof is divided into two steps . in _
step 1 _ we determine the distribution under @xmath331 and in _ step 2 _ we extend this result via a shift argument to @xmath332 .
_ we consider the law of the process under @xmath331 for now .
the first part of the proof is identical to the proof of proposition 3.5 @xcite , so it is only sketched here . using repeatedly the identity @xmath363 relabeling @xmath364 , and introducing new variables @xmath365 for @xmath366 , we can write @xmath367=\int_{\mathcal d ' } \det_{1\leq k , l\leq n}\big[\widetilde{f}_{k-1}(\xi_l^{n},{t})\big]\prod_{2\leq l\leq k\leq n } { \mathrm{d}}\xi_l^k,\ ] ] where @xmath368 . using the antisymmetry of the determinant and encoding the constraint on the integration variables into indicator functions ,
we obtain that the measure is a marginal of @xmath369\det_{1\leq k , l\leq n}\big [ \widetilde{f}_{k-1}(\xi_l^{n},{t})\big]\\= { \mathrm{const}}&\cdot \prod_{n=1}^{n}\det_{1\leq i , j\leq n}\big[\tilde{\phi}_n(\xi_i^{n-1},\xi_j^n)\big]\det_{1\leq k , l\leq n}\big [ \widetilde{f}_{k-1}(\xi_l^{n},{t})\big ] \end{aligned}\ ] ] with @xmath370 and using the convention that @xmath371 always holds .
the measure has the appropiate form for applying lemma [ lemdetmeasure ] .
the composition of the @xmath372 functions can be evaluated explicitly as @xmath373 define @xmath374 for @xmath375 and some @xmath90 . in the case
@xmath376 the integrand has no poles in the region @xmath377 , which implies @xmath378 .
the straightforward recursion @xmath379 eventually leads to condition being satisfied .
the space @xmath380 is generated by @xmath381 so a basis for @xmath380 is given by @xmath382 choose functions @xmath383 as follows @xmath384 by residue calculating rules , @xmath383 is a polynomial of order @xmath385 for @xmath282 and a linear combination of @xmath11 and @xmath386 for @xmath284 , so these functions indeed generate @xmath380 . to show for @xmath387
, we decompose the scalar product as follows : @xmath388 since @xmath389 we are free to choose the sign of @xmath390 as necessary . for the first term , we choose @xmath391 and the path @xmath392 close enough to zero , such that always .
then , we can take the integral over @xmath293 inside and obtain @xmath393 for the second term , we choose @xmath251 to obtain . then again , we can take the integral over @xmath293 inside and arrive at the same expression up to a minus sign .
the net result of is a residue at @xmath394 , which is given by @xmath395 the case @xmath396 uses the same decomposition and requires the choice @xmath397 resp .
@xmath391 , finally leading to @xmath398 furthermore , both @xmath399 and @xmath400 are constants , so the kernel has a simple form ( compare with ) @xmath401 however , the relabeling @xmath364 included a index shift , so the kernel of our system is actually @xmath402 note that we are free to extend the summation over @xmath403 up to infinity , since the integral expression for @xmath404 vanishes for @xmath405 anyway .
taking the sum inside the integrals we can write @xmath406 with @xmath407 by choosing contours such that @xmath408 , we can use the formula for a geometric series , resulting in @xmath409 inserting this expression back into gives the kernel , which governs the multidimensional distributions of @xmath410 under the measure @xmath331 , namely @xmath411 _ step 2 . _ the distributions under @xmath332 and under @xmath331
can be related via the following _ shift argument_. introducing the shorthand @xmath412 we have @xmath413 combining the identity @xmath414 with finishes the proof .
[ lemdetmeasure ] assume we have a signed measure on @xmath415 given in the form , @xmath416_{1\leq i , j\leq n } \det[\psi_{n - i}^{n}(x_{j}^n)]_{1\leq i , j \leq n},\ ] ] where @xmath417 are some `` virtual '' variables and @xmath418 is a normalization constant .
if @xmath419 , then the correlation functions are determinantal . to write down the kernel we need to introduce some notations .
define @xmath420 where @xmath421 , and , for @xmath422 , @xmath423 then the functions @xmath424 are linearly independent and generate the @xmath2-dimensional space @xmath380 .
define a set of functions @xmath425 spanning @xmath380 defined by the orthogonality relations @xmath426 for @xmath427 .
further , if @xmath428 , for some @xmath429 , , then the kernel takes the simple form @xmath430
[ remlambda ] the change in variables @xmath431 reproduces the same system with new parameters @xmath432 and @xmath433 .
we can therefore restrict our considerations to @xmath434 without loss of generality .
fix @xmath434 from now on .
according to we use the scaled variables @xmath435 with @xmath94 . correspondingly , consider the rescaled ( and conjugated ) kernel @xmath436 which naturally decomposes into @xmath437 instead of integrals over airy functions can also be written as contour integrals : @xmath438 in the integral defining @xmath50 , the path for @xmath25 and @xmath439 do not have to intersect .
in addition , the gaussian part has a representation in terms of an integral over airy functions : @xmath440 in order to establish the asymptotics of the joint distributions , one needs both a pointwise limit of the kernel , as well as uniform bounds to ensure convergence of the fredholm determinant expansion .
the first time this approach was used is in @xcite .
these results are contained in the following propositions .
[ proppointw ] consider any @xmath441 in a bounded set and fixed @xmath442 .
then , the kernel converges as @xmath443 uniformly for @xmath444 ^ 2 $ ] .
[ corbound ] consider @xmath441 fixed . for any @xmath442 there
exists @xmath445 such that for @xmath446 the bound @xmath447 holds for all @xmath444 ^ 2 $ ] .
[ propk0bound ] for fixed @xmath448 and @xmath94 there exists @xmath449 such that the estimate @xmath450 holds for any @xmath446 and @xmath451 .
[ propphibound ] for fixed @xmath452 there exists @xmath449 and @xmath134 such that @xmath453 now we can prove the asymptotic theorem : the joint distributions of the rescaled process @xmath454 under the measure @xmath331 are given by the fredholm determinant with series expansion @xmath455 , \end{aligned}\ ] ] where @xmath456 and @xmath457 are given in ( [ eqscaling ] ) . using the change of variables and a conjugation
we obtain @xmath458 , \end{aligned}\ ] ] where the fraction inside the determinant is the new conjugation , which does not change the value of the determinant . using corollary [ corbound ] and propositions [ propk0bound ] , [ propphibound ] , we can bound the @xmath459-coefficient inside the determinant by @xmath460 assuming the @xmath461 are ordered .
the bounds @xmath462 lead to @xmath463 using the hadamard bound on the determinant , the integrand of is therefore bounded by @xmath464 which is integrable .
furthermore , @xmath465 which is summable , since the factorial grows like @xmath466 , i.e. , much faster than the nominator .
dominated convergence thus allows to interchange the limit @xmath467 with the integral and the infinite sum .
the pointwise convergence comes from proposition [ proppointw ] , thus @xmath468 it remains to show that the convergence carries over to the measure @xmath332 .
the identity @xmath469 leads to @xmath470 notice that in , @xmath222 appears only in the indicator function , so differentiation just results in one of the @xmath471 not being integrated but instead being set to @xmath222 . using the same bounds as before we can again show interchangeability of the limit @xmath467 with the remaining integrals and the infinite sum . before showing propositions [ proppointw ] and [ propk0bound ]
, we introduce some auxiliary functions and establish asymptotic results for them .
define @xmath472 [ lemalphalimit ] for fixed @xmath21 and @xmath442 the limits @xmath473 hold uniformly for @xmath474 $ ] .
let @xmath475 be the normalized @xmath2-th order hermite polynomial with respect to the weight @xmath476 , i.e. , satisfying @xmath477 there exist two explicit integral representations for these polynomials : @xmath478 from @xcite , p. 201
, one obtains the asymptotic behaviour @xmath479 uniformly for any bounded @xmath480 . introducing the shorthands @xmath481 , @xmath482 and applying the change of variables @xmath483 , we can write @xmath484 using stirling s approximation and taylor expansion in the exponents one can further analyze this as @xmath485 with with the error terms being uniform for @xmath474 $ ] .
the observation @xmath486 as @xmath467 , settles the convergence of @xmath487 .
using the second integral representation of the hermite polynomials one can rewrite @xmath488 , too : @xmath489 analyzing the prefactor as done before finishes the proof .
[ lemalphabound ] for fixed @xmath21 , there exist @xmath490 such that for all @xmath446 and @xmath491 the following bounds hold @xmath492 we start by analyzing @xmath488 . defining functions as @xmath493
we can write @xmath494 , leading to @xmath495 [ lb]@xmath496 [ lb]@xmath497 [ lb]@xmath498 [ lb]@xmath499 [ lb]@xmath500 used for obtaining the uniform bounds.,title="fig:",height=188 ] define a new parameter @xmath497 given by @xmath501 for some small , positive @xmath390 chosen in the following , and let @xmath502 .
as shown in figure [ figboundcontour ] , we change the contour @xmath392 to , with @xmath503 \}\\
\gamma_2(r)&=\{-1+ue^{{{\rm i}}\theta } , u\in[\omega/\cos\theta , r]\}\\ \gamma_3(r)&=\{-1+re^{{{\rm i}}u } , u\in[-\theta,\theta]\}. \end{aligned}\ ] ] since we will only estimate the absolute value of the integrals , the direction of integration does not matter .
if @xmath237 and @xmath504 are fixed , the integrand is dominated by the @xmath505 term for large @xmath506 .
thus the contribution coming from @xmath507 converges to @xmath34 as @xmath508 . with @xmath509
our choice for the contour of integration is now .
we start by analyzing @xmath510}{\mathrm{d}}u\ , e^{g(z_0+{{\rm i}}u)-g(z_0)},\ ] ] where @xmath511 .
let us consider the prefactor @xmath512 at first . since @xmath497 is small we can use taylor expansion , as well as , to obtain the bounds @xmath513 all error terms are to be understood uniformly in @xmath514 .
the @xmath515 term dominates both @xmath516 , if @xmath442 is chosen large enough , and @xmath517 , for @xmath390 being small .
this results in @xmath518 to show convergence of the integral part of we first bound the real part of the exponent : @xmath519\\ = & { t}\left(\frac{u^2}{2}-\frac{1}{2}\ln\left(1+\frac{u^2}{z_0 ^ 2}\right)\right)+{t}^{2/3}r\ln\left(1+\frac{u^2}{z_0 ^ 2}\right)\\ \leq&{t}\frac{u^2}{2}\left(1-\frac{1}{z_0 ^ 2}+\frac{u^2}{2z_0 ^ 4}\right)+{t}^{2/3}r\frac{u^2}{z_0 ^ 2}=:-\eta{t}^{2/3}u^2 . \end{aligned}\ ] ] @xmath520 satisfies : @xmath521 where we used @xmath522 . given any @xmath390 we can now choose both @xmath442 and @xmath445 large , such that the first term dominates .
consequently @xmath520 will be bounded from below by some positive constant @xmath523 . the integral contribution coming from @xmath524
can thus be bounded as @xmath525}{\mathrm{d}}u\,e^{g(z_0+iu)-g(z_0)}\right|\\ & \leq e^{-\frac{1}{2}s^{3/2}}\frac{{t}^{1/3}}{2\pi}\int_{\mathbb{r}}{\mathrm{d}}u\,e^{-\eta_0{t}^{2/3}u^2}=e^{-\frac{1}{2}s^{3/2}}\frac{1}{2\pi}\int_{\mathbb{r}}{\mathrm{d}}u\,e^{-\eta_0u^2}\\ & = e^{-\frac{1}{2}s^{3/2}}\frac{1}{2\sqrt{\pi\eta_0}}\leq e^{-s}. \end{aligned}\ ] ] finally we need a corresponding bound on the @xmath526 contribution to the integral . by symmetry
this case covers also the contour @xmath527 .
write @xmath528 with @xmath529 . from the previous estimates one
easily gets @xmath530 so the remaining task is to show boundedness of the integral part of . at first notice that the real part of the @xmath515 contribution in the exponent is negative ,
so we can omit it , avoiding the problem of large @xmath504 . by elementary calculus
, we have for all @xmath531 , @xmath532 that is , @xmath526 is a steep descent curve for @xmath517 .
we can therefore restrict the contour to a neighbourhood of the critical point @xmath533 , which we choose of magnitude @xmath534 , at the expense of an error of order @xmath535 : @xmath536 where @xmath537 .
taylor expanding these functions leads to @xmath538 for some positive constants @xmath539 , @xmath540 , by choosing @xmath534 and @xmath390 small enough . for large @xmath442 and @xmath445
, @xmath541 dominates over @xmath542 , so we can further estimate : @xmath543 this settles the uniform exponential bound on @xmath488 . combining and one obtains @xmath544 for some function @xmath545 . from the convergence of @xmath487 and @xmath488
it is clear that @xmath546 converges , too . since we already know that @xmath488 is uniformly bounded by a constant times @xmath547 , the exponential bound on @xmath487 follows .
regarding the first part of the kernel , we notice , that @xmath548 does not appear in our scaling , so we can use the formula ( for @xmath549 ) @xmath550 this is the same function as in @xcite , proof of proposition 5.1 , so the limit @xmath551 is not proven here .
the different parts of the remaining kernel can be rewritten as integrals over the previously defined functions @xmath552 and @xmath553 . for @xmath554 , choose the contours in such a way that @xmath555 is ensured .
@xmath556 also rewrite @xmath557 as follows : @xmath558 similarly , @xmath559 with @xmath560 the residuum satisfies the limit @xmath561 uniformly in @xmath219 .
the prefactor of the last part of the kernel is simply .
combining all these equations gives @xmath562 using the previous lemmas we can deduce compact convergence of the kernel . indeed ( omitting the @xmath21-dependence for greater clarity ) we can write : @xmath563}&\left|\int_0^\infty { \mathrm{d}}x\,\alpha_{t}(s_1+x)\beta_{t}(s_2+x)-\int_0^\infty { \mathrm{d}}x\,\alpha(s_1+x)\beta(s_2+x)\right|\\ \leq&\int_0^\infty { \mathrm{d}}x\,\sup_{s_1,s_2\in[-l , l]}\left|\alpha_{t}(s_1+x)\beta_{t}(s_2+x)-\alpha(s_1+x)\beta(s_2+x)\right| . \end{aligned}\ ] ] by lemma [ lemalphalimit ] the integrand converges to zero for every @xmath330 . using lemma [ lemalphabound ] we can bound it by @xmath564 , thus ensuring that goes to zero , i.e. , @xmath554 converges compactly .
in the same way we can show the convergence of @xmath557 and @xmath565 .
applying the limit in and inserting the expressions for @xmath552 and @xmath553 finishes the proof . since the convergence is uniform in @xmath219 we can deduce @xmath566 inserting this as well as the bounds from lemma [ lemalphabound ] into results in @xmath567
using the results from @xcite we can transform the formula for the multidimensional probability distribution of the finite - step airy@xmath61 process from the current form involving a fredholm determinant over the space into a path - integral style form , where the fredholm determinant is over the simpler space @xmath55 .
the result of @xcite can not be applied at the stage of finite time as one of the assumption is not satisfied .
[ proppathint ] for any parameters @xmath568 , @xmath569 , satisfying @xmath570 define the multiplication operator @xmath571 , with @xmath572 writing @xmath573 , the finite - dimensional distributions of the finite - step airy@xmath61 process are given by @xmath574 with @xmath575 where @xmath46 denotes the projection operator on @xmath47 .
[ remv ] the operator @xmath576 for @xmath577 is defined only on the range of @xmath578 and acts on it in the following way : @xmath579 in particular , we have also @xmath580 .
we will denote conjugations by the operator @xmath143 by a hat in the following way : @xmath581 applying the conjugation also in the determinant in , the identity we have to show is : @xmath582 this is done by applying theorem 1.1 @xcite .
it has three groups of assumptions we have to prove .
we merged them into two by choosing the multiplication operators of assumption 3 to be the identity .
[ [ assumption-1 ] ] assumption 1 + + + + + + + + + + + + 1 .
the operators @xmath583 , @xmath584 , @xmath585 and @xmath586 for @xmath577 preserve @xmath55 and are trace class in @xmath55 .
the operator @xmath587 preserves @xmath55 and is trace class in @xmath55 .
[ [ assumption-2 ] ] assumption 2 + + + + + + + + + + + + 1 .
right - invertibility : @xmath588 2 .
semigroup property : @xmath589 3 .
reversibility relation : @xmath590 the semigroup property is clear . to see the reversibility relation , start from the contour integral representation of @xmath591 and @xmath592 and
use the gaussian identity : @xmath593 this results in @xmath594 . on the other hand
we have @xmath595 so @xmath596 , which proves assumption 2 ( iii ) .
noticing remark [ remv ] , the right - invertibility follows immediately .
assumption 1 ( ii ) can be deduced from assumption 1 ( i ) as shown in remark 3.2 , @xcite . using the previous identities we
thus are left to show that the three operators @xmath583 , for @xmath577 , as well as @xmath597 and @xmath598 , for arbitrary @xmath599 , @xmath600 , are all @xmath601-bounded and trace class .
first notice that @xmath602 . using the shorthand @xmath603 and inserting this into the integral representation of
@xmath604 we have @xmath605 with the new operators @xmath606 introducing yet another operator , @xmath607 , we can write @xmath608 the hilbert - schmidt norm of the first factor is given by @xmath609 the asymptotic behaviour of the airy function and the inequalities imply that both integrals are finite . similarly , @xmath610 where we used @xmath611 as well . as a product of two hilbert - schmidt operators ,
@xmath583 is thus @xmath601-bounded and trace class .
we decompose the operator @xmath612 as @xmath613 where @xmath614 again , we bound the hilbert - schmidt norms , @xmath615 as well as @xmath616 the superexponential decay of the airy function implies that for every we can find @xmath617 such that @xmath618 .
this proves finiteness of the integrals
. regarding the last operator , start by decomposing it as @xmath619 for some function @xmath620 with @xmath601-norm @xmath11 .
next , notice that @xmath621 it is easy to see that @xmath622 , so @xmath623 is bounded on the area of integration .
but then the @xmath624 term ensures the decay , implying that the integral is finite .
furthermore , @xmath625 analyzing the asymptotic behaviour of @xmath626 we see that for large positive arguments , the first part decays exponentially with rate @xmath627 and the second part even superexponentially . @xmath628 thus gives convergence on the positive half - line .
for negative arguments , it is sufficient to see that @xmath626 does not grow faster than exponentially .
first of all let us show that the choice of @xmath68 is asymptotically irrelevant . denote by @xmath629 the rescaled process as in ( [ eqscaledprocess ] ) , where @xmath68 , and @xmath630 the rescaled process as in ( [ eqscaledprocessoriginal ] ) , where @xmath631 .
this corresponds to a finite shift of the system , which is therefore irrelevant in the large time limit .
[ lemma7.1 ] for any @xmath632 , @xmath633 and @xmath634 , it holds @xmath635 we can construct the processes @xmath636 and @xmath637 on the same probability space so that , for any @xmath1 , @xmath638 and with @xmath93 being independent of @xmath639 . after scaling we have @xmath640 . as @xmath641 converges to @xmath34 is distribution ,
the result follows .
we know from theorem [ thmasymp ] and proposition [ proppathint ] that : @xmath642 in this section we prove the main theorem [ thmasymp0 ] by extending this equation to @xmath643 .
the right hand side can actually be analytically continued for all @xmath644 ( see proposition [ propanalyt ] ) .
additionally we have to show that the left hand side is continuous at @xmath643 .
this proof relies mainly on proposition [ propexpoint ] , which gives a bound on the exit point of the maximizing path from the lower boundary in the last passage percolation model .
we adopt the point of view of last passage percolation discussed in section [ sectlpp ] .
the superscripts of @xmath32 , @xmath442 and @xmath645 indicate the choice of @xmath35 , while @xmath646 is always fixed at @xmath11 .
it is clear that for any path @xmath647 the weight @xmath648 is non - decreasing in @xmath35 .
but then the supremum is non - decreasing , too , and : @xmath649 for @xmath36 . we know that there exists a unique maximizing path .
we can therefore define @xmath650 , the exit point from the lower boundary specifically with @xmath651 .
we want to derive the inequality @xmath652 this can be seen as follows : @xmath653 note that @xmath654 maximizes @xmath655 and not necessarily @xmath648 .
in particular we have @xmath656 combining the last two equations results in . and imply that for the rescaled processes @xmath657 , see ( [ eqscaledprocess ] ) , we have @xmath658 for any @xmath251 it holds @xmath659 then , taking @xmath467 , we obtain @xmath660 using proposition [ propexpoint ] on the last term and proposition [ propanalyt ] on the other terms , we can now take the limit @xmath661 , resulting in @xmath662 continuity of in the @xmath663 finishes the proof .
[ propexpoint ] for any @xmath664 , @xmath665 by scaling of @xmath237 and @xmath553 , is equivalent to @xmath666 for any @xmath664 , which is the limit we are showing .
we introduce some new events : @xmath667 notice that if @xmath668 occurs , then @xmath669 resulting in @xmath670 .
we arrive at the inequality : @xmath671 we further define new random variables @xmath672 by theorem 7 @xcite , for any fixed @xmath664 , @xmath673 where @xmath674 has the gue tracy - widom distribution .
@xmath675 follows the distribution of the largest eigenvalue of a critically spiked gue matrix , as will be shown in lemma [ lemspike ] .
@xmath676 has the distribution of a standard normal random variable @xmath677 for any @xmath678 , @xmath679 . combining these definitions
, we have : @xmath680 fix @xmath681 , such that : @xmath682 using the independence of @xmath676 and @xmath675 , we obtain @xmath683 further , the inequality @xmath684 leads to @xmath685 inserting and into , we arrive at @xmath686 by and lemma [ lemspike ] we can take limits : @xmath687\\&=0 .
\end{aligned}\ ] ] [ lemspike ] let @xmath664 be fixed .
for any @xmath688 , as @xmath467 , the random variable @xmath689 converges in distribution , @xmath690 in addition , @xmath691 satisfies @xmath692 the family of processes @xmath693 indexed by @xmath694 and time parameter @xmath178 is precisely a marginal of warren s process with drifts , starting at zero , as defined in @xcite . in our case
only the first particle has a drift of @xmath11 , and all the others zero . by theorem
2 @xcite , the fixed time distribution of this process is given by the distribution of the largest eigenvalue of a spiked @xmath695 gue matrix , where the spikes are given by the drifts .
thus we can apply the results on spiked random matrices , more concretely we want to apply theorem 1.1 @xcite , with the potential @xmath696 .
since @xmath697 represents a @xmath695 gue matrix diffusion @xmath698 at time @xmath699 , it is distributed according to the density @xmath700 where @xmath701 is a @xmath695 matrix with a one at entry @xmath702 and zeros elsewhere . in order to apply the theorem
we need the density given in equation ( 1 ) @xcite , i.e. , consider the scaled quantity @xmath703 .
the size of the first - order spike is then : @xmath704 we are thus in the neighbourhood of the critical value @xmath705 . for @xmath706 ,
let @xmath707 with @xmath708 being the cumulative distribution function of the gue tracy - widom distribution , and @xmath709 as in , define : @xmath710 applying ( 28 ) @xcite , we have @xmath711 with @xmath712 where @xmath713 . since in our case @xmath714
, we can estimate : @xmath715 combining this with the usual bounds on the airy kernel and the airy function , we see that as @xmath716 , the scalar product in converges to zero and we are left with the limit of @xmath717 which is one . on the other hand , @xmath718 and @xmath719 from which the claim follows .
[ propanalyt ] the function @xmath720 can be extended analytically in the domain @xmath644 .
its value at @xmath643 is given by @xmath721 we use the identity @xmath722 and lemma [ leminv ] to factorize @xmath723 since the second factor is independent of @xmath534 , the remaining task is the analytic continuation of the first . using , decompose @xmath724 as @xmath725 now , @xmath726 the function @xmath727 is analytic in @xmath644 . using these two identities as well as @xmath728 , we can rearrange the inner product as follows : @xmath729 since @xmath730 is evidently analytic in @xmath644 , we are left to show convergence of the scalar product .
all involved functions are locally bounded , so to establish convergence it is enough to investigate their asymptotic behaviour .
@xmath730 may grow exponentially at arbitrary high rate , depending on @xmath731 and @xmath534 , for both large positive and large negative arguments .
we therefore need superexponential bounds on the function : @xmath732 for this purpose we first need an expansion of the operator @xmath733 : @xmath734 notice that all operators @xmath735 , @xmath736 and @xmath737 map superexponentially decaying functions onto superexponentially decaying functions .
moreover @xmath735 and @xmath736 generate superexponential decay for large negative resp . positive arguments .
the function @xmath738 decays superexponentially for large arguments but may grow exponentially for small ones .
since every part of the sum contains one projection @xmath739 , @xmath740 decays superexponentially on both sides .
examining @xmath741 , notice that the @xmath284 contribution in is equal to @xmath742 , which is cancelled out here .
all other contributions contain both @xmath743 and @xmath739 , which ensure superexponential decay . using the usual asymptotic bound on the airy function
, we see that the operator @xmath50 maps any function in its domain onto one which is decreasing superexponentially for large arguments . by previous arguments , functions in the image of @xmath744 decay on both sides , in particular @xmath745 .
now , in order to establish the finiteness of the scalar product , decompose the inverse operator as @xmath746 .
the contribution coming from the identity has just been settled . as inverse of a bounded operator , @xmath747 is also bounded .
because of the rapid decay , the functions @xmath740 , @xmath745 and @xmath741 are certainly in @xmath55 and thus mapped onto @xmath55 by this operator . finally , the image of an @xmath55-function under the operator @xmath744 is decaying superexponentially on both sides .
the expression is thus an analytic function in @xmath534 in the domain @xmath183 .
setting @xmath643 returns the value of @xmath748 . combining these results with finishes the proposition .
[ leminv ] the operator @xmath749 is invertible .
we employ the same strategy as in @xcite . for that purpose
we use the following equivalence @xmath750 let @xmath751 .
@xmath752 for any @xmath753 , where @xmath754 is the goe tracy - widom distribution function . for the last equality see @xcite .
the tails of the goe tracy - widom distribution have been studied in great detail in various publications , see for instance @xcite .
in this section we prove that the airy@xmath61 process has brownian increments for nonnegative arguments : without loss of generality we can assume that @xmath757 . denoting the partial derivative with respect to the @xmath316-th coordinate by @xmath758 , we have @xmath759 with @xmath760 with a small abuse of notations , in what follows we will write @xmath761 then , @xmath762 the crucial identity is : @xmath763 we therefore have to study the asymptotics of @xmath764 as @xmath765 .
now consider @xmath768 .
@xmath769 regarding the first term , notice that the multiple derivative of the fredholm determinant gives exactly the multipoint density of the airy@xmath770 process , which is known to decay exponentially for both large positive and negative arguments .
this exponential decay dominates over the linear growth of @xmath771 .
similarly , the @xmath772-fold derivative is smaller the @xmath772-point density of the airy@xmath770 process , so this contribution vanishes in the limit , too . continuing to @xmath773 , using @xmath774 ,
we first simplify the expression @xmath775 we introduce the shift operator @xmath82 , @xmath776 , which satisfies @xmath777 and @xmath778 , and consequently also @xmath779 using @xmath780 , we have @xmath781 now the dependence on the vector @xmath782 is only in the projection operators , while the dependence on @xmath783 is only in these two operators : @xmath784 for large @xmath783 , we have @xmath785 and @xmath786 ( both strong types of convergence from the superexponential airy decay ) .
so @xmath787 applying the expansion , we arrive at : @xmath788 writing out this scalar product and applying the fundamental theorem of calculus leads to : @xmath789 which is the desired gaussian density after setting @xmath790 as in . for large negative @xmath783
, we have @xmath791 and @xmath792 .
the rank one contribution is thus @xmath793 we have to be somewhat careful here , as the convergence is weak ( only pointwise ) and @xmath794 is not even @xmath601-integrable .
but the first factor decays superexponentially on both sides for finite @xmath783 and also in the limiting case @xmath795 , so one should be able to derive nice convergence properties . neglecting this rank one contribution
we are left with @xmath796 | we consider an infinite system of brownian motions which interact through a given brownian motion being reflected from its left neighbor .
earlier we studied this system for deterministic periodic initial configurations . in this contribution
we consider initial configurations distributed according to a poisson point process with constant intensity , which makes the process space - time stationary .
we prove convergence to the airy process for stationary the case . as a byproduct
we obtain a novel representation of the finite - dimensional distributions of this process .
our method differs from the one used for the tasep and the kpz equation by removing the initial step only after the limit @xmath0 .
this leads to a new universal cross - over process . |
crystal growth is often performed from melts contained in crucibles , e.g. by czochralski or bridgman technique .
chemical compatibility of the substance to be grown , the crucible material , and other constructional elements under the given temperature @xmath4 and atmosphere is an inevitable prerequisite for the establishment of stable growth conditions . in case of metal oxides meo@xmath5 (
@xmath6 is the valency ) usually crucibles made of another metal me@xmath7 with high melting point @xmath8 are used : platinum ( @xmath9c ) , iridium ( @xmath10c ) , and tungsten ( @xmath11c ) are typical examples .
( thermodynamic data are taken from the factsage databases @xcite throughout this article . )
unfortunately , the chemical stability of these refractory metals drops with rising @xmath8 .
the chemical stability of all components is governed by thermodynamic equilibria if the temperature is sufficiently high : this is usually true for @xmath12c . in such cases the behavior of systems containing only pure condensed phases can be predicted easily considering redox reactions @xmath13 for all metals me ( crucibles ) or metal oxides meo@xmath14 ( substances to be grown , insulating ceramics ) .
the description of the system by ( [ eq : redox ] ) for the me@xmath7 , me@xmath15 , is almost correct even if some of them are forming condensed phases with more then one metal ( e.g. melts of several oxides , intermediate compounds meme@xmath7o@xmath14 , or solid solutions ( me , me@xmath7)o@xmath14 ) , as the formation enthalpy of such phases from the meo@xmath14 is often small compared to @xmath16 for ( [ eq : redox ] ) . at equilibrium @xmath17 for ( [ eq : redox ] ) vanishes .
if moreover the vapor pressures of the metals and their oxides can be neglected compared to the oxygen partial pressure @xmath18 , one finds for the equilibrium constant @xmath19 of reaction ( [ eq : redox ] ) @xmath20 where @xmath21 is the standard free energy change of ( [ eq : redox ] ) . as @xmath22 and
@xmath23 change usually weakly with @xmath4 it is obvious from ( [ eq : ellingham ] ) that plots of @xmath24 vs. @xmath4 ( `` ellingham diagrams '' ) are straight lines for each oxidation reaction .
these lines separate phase regions where one oxidation state prevails and the whole graph represents a predominance phase diagram for the corresponding metal me and its oxides @xcite .
ellingham and other predominance diagrams can be calculated either manually from tabulated @xmath22 , @xmath23 @xcite or with software packages like or @xcite , respectively , and are discussed in more detail elsewhere @xcite .
showing stability regions for tungsten ( 2 condensed phases ) , tungsten(iv ) oxide ( 1 condensed phase ) , tungsten(vi ) oxide ( 3 condensed phases ) , and gas ( no condensed phases).,scaledwidth=47.0% ] fig .
[ fig : w_o2 ] shows an ellingham diagram for the system tungsten
oxygen where transitions between different oxidation states ( [ eq : redox ] ) are indicated by inclined lines and transformations between different pure phases ( e.g. triclinic @xmath25-wo@xmath2 , tetragonal @xmath26-wo@xmath2 @xcite , and liquid wo@xmath2 ) as vertical lines .
it will be shown in the following that these diagrams are powerful tools for the manipulation of oxidation states during high @xmath4 crystal growth processes .
it is an inherent result of ( [ eq : redox ] ) and ( [ eq : ellingham ] ) that @xmath18 may never be zero to keep any metal oxide thermodynamically stable . if @xmath18 is too low ( [ eq : redox ] ) is shifted to the left hand side resulting in an oxide with lower oxidation state . for sufficiently low @xmath18 , finally the metal ( valency @xmath27 )
is formed .
`` pure '' argon or nitrogen are often reported as `` inert '' gases to be used during crystal growth processes .
`` pure '' means usually 5n gases : actually such gas represents a mixture of 99.999% gas ( e.g. ar , n@xmath1 ) + 0.001% ( 10ppm ) impurity .
this impurity is not well defined , but could e.g. be air ( ca .
21% o@xmath1 + ca .
79% n@xmath1 ) . in this case
, 5n nitrogen with total pressure @xmath28bar can be described as a constant admixture of @xmath29bar o@xmath1 in nitrogen .
@xmath30bar is too high for the growth of ( la , sr)(al , ta)o@xmath2 ( lsat ) crystals that are dark - red colored in the as - grown state and can be bleached by annealing in h@xmath1 @xcite .
other oxides are unstable at elevated @xmath4 under too low @xmath18 , and typically minor amounts of oxygen are added to the gas : la@xmath3nd@xmath14gao@xmath2 from n@xmath1 + 1% o@xmath1 @xcite ; sm@xmath2ga@xmath31o@xmath32 or gd@xmath2ga@xmath31o@xmath32 from n@xmath1 with 0.9 or 1.5% o@xmath1 , respectively @xcite .
the influence of oxygen partial pressures from @xmath33bar to @xmath34bar on the color of melt grown srlagao@xmath35 and srlaalo@xmath35 was described by pajaczkowska et al .
@xcite . that are necessary for the stabilization of several metal oxides ( decomposition products for too low @xmath18 given in brackets ) .
dashed lines : equilibrium values @xmath36 that are created by several gases or gas mixtures ( @xmath28bar).,scaledwidth=47.0% ] obviously , by addition of o@xmath1 to the growth atmosphere only a constant ( independent of @xmath4 ) partial pressure @xmath37 can be maintained . in fig .
[ fig : t_vs_ox ] the level of the upper abscissa with @xmath38bar represents pure oxygen and the horizontal dashed lines represent air ( @xmath39bar ) or 5n argon with 10ppm air , respectively . as the reactions ( [ eq : redox ] ) are mainly exothermal , the @xmath18 that is required to stabilize a specific meo@xmath14 rises usually with @xmath4 .
[ fig : t_vs_ox ] demonstrates this for lu@xmath1o@xmath2 , al@xmath1o@xmath2 , ga@xmath1o@xmath2 , zno , and bi@xmath1o@xmath2 .
these oxides are stable for all @xmath36 above the corresponding lines and tend to decompose to the species given in brackets for lower oxygen partial pressures .
the stability of the oxides listed above requires growing @xmath18 .
the gaseous oxides co@xmath1 , co , and h@xmath1o decompose with @xmath4 like the reverse reaction of ( [ eq : redox ] ) @xmath40 and lead to the production of oxygen if @xmath4 rises .
the reaction partners of ( [ eq : co2])([eq : h2o ] ) and methane ch@xmath35 are combined e.g. by ( [ eq : ch4 ] ) .
[ fig : t_vs_ox ] shows that mixtures of co@xmath1 , co , and ch@xmath35 allow to adjust @xmath18 over more than 10 orders of magnitude .
moreover , @xmath36 for such atmospheres changes similarly to many meo@xmath14 .
an appropriate gas composition can provide a `` self adjusting '' oxygen partial pressure that keeps the desired oxidation state thermodynamically stable over an extended @xmath4 range .
gallium(iii ) oxide : : : tomm et al .
@xcite reported the first successful czochralski growth of @xmath25-ga@xmath1o@xmath2 .
the melting point @xmath41c @xcite is well beyond the stability limit of platinum , and hence ir crucibles had to be used .
already in the proximity of the ga@xmath1o@xmath2 stability limit ( at @xmath8 : @xmath42bar , fig .
[ fig : t_vs_ox ] ) the evaporation , under formation of suboxides gao and ga@xmath1o , becomes remarkable . at cooler parts ga(i)@xmath1o tends to sublimate and dissociate under formation of ga(iii)@xmath1o@xmath2 and ga metal .
the latter easily alloys with ir and destroys the crucible .
a sufficiently large oxygen concentration is known to suppress suboxide formation , but does unfortunately affect ir by oxidation especially in colder parts of the setup ( e.g. afterheater ) .
an atmosphere of 90% ar + 10% co@xmath1 supplies a @xmath36 that enables crystal growth without affecting ir crucibles .
ti : sapphire : : : under ambient conditions titanium forms the stable ti@xmath43o@xmath1 , but lasing in ti : sapphire is obtained by ti@xmath44 and is suppressed by ti@xmath44-ti@xmath43 pairs @xcite . usually , al@xmath1o@xmath2 crystals with titanium doping are grown in vacuum and the desired ti@xmath44 is obtained by subsequent annealing in a reducing atmosphere ( e.g. h@xmath1 or co ) .
one should keep in mind that the term `` vacuum '' means a state that is not well defined from the thermodynamic point of view
always some residual @xmath18 will be present depending on level of vacuum and composition of the residual gas .
uecker et al .
@xcite proposed an alternative technology based on ar / co mixtures with @xmath45 mbar that allow the adjustment of the oxygen partial pressure @xmath46 ( o atoms are in excess of o@xmath1 molecules under the growth conditions ) to the desired range @xmath47 .
iron(ii ) oxide : : : depending on @xmath18 , @xmath4 , and chemical environment , iron acquires usually the oxidation states fe@xmath48 ( fe metal , carbonyl fe(co)@xmath31 , carbide fe@xmath2c ) , fe@xmath49 ( wustite fe@xmath3o ) , fe@xmath44 ( hematite fe@xmath1o@xmath2 ) , fe@xmath43 ( strontium ferrate(iv ) srfeo@xmath2 @xcite ) , or fe@xmath50 ( potassium ferrate(vi ) k@xmath1feo@xmath35 @xcite ) . magnetite ( iron(ii , iii ) oxide fe@xmath2o@xmath35 )
has a large stability field ( @xmath51 ) under mildly oxidizing conditions , whereas the stability field of wustite is small and restricted to @xmath52bar that can hardly be maintained under the conditions of crystal growth from the melt in the technically available `` inert '' gas .
residual oxygen within ar ( 99.999% purity ) is sufficient to oxidize wustite to magnetite @xcite .
contrarly , the oxygen partial pressure @xmath36 of a gas mixture 85% ar , 10% co@xmath1 + 5% co is `` self adjusting '' over an extended range @xmath53 @xmath54k which allows the growth of wustite ( @xmath55c ) as well as olivine ( mg , fe@xmath49)sio@xmath35 ( @xmath56c ) crystals @xcite .
zinc oxide : : : the triple point of zno is @xmath57c , 1.06bar . fig . [
fig : t_vs_ox ] shows that the sensibility of zno with respect to reduction is intermediate compared to ga@xmath1o@xmath2 and bi@xmath1o@xmath2 .
fortunately , the @xmath36 supplied by co@xmath1 is sufficient to stabilize zno . in companion papers
was reported recently on the melt growth of zno boules from ir crucibles in a bridgman - like setup @xcite . during crystal growth @xmath4 gradients
must be established , hence compatibility of the materials must be guaranteed over a certain @xmath4 range .
[ fig : zn - ir - gas]a shows that any fixed mixture of ar and o@xmath1 will not stabilize iridium metal ir(s ) and solid or liquid zinc oxide ( zno(s ) , zno(liq ) ) for an extended @xmath4 range . for
@xmath58c ir would oxidize to iro@xmath1 .
in contrast , fig . [
fig : zn - ir - gas]b shows that co@xmath1/co mixtures with @xmath59% co@xmath1 , including pure co@xmath1 , can stabilize ir together with zno in both condensed phases over the entire @xmath4 range of importance for melt growth . for a total pressure @xmath60bar . a ) for gas mixtures of o@xmath1 and ar .
b ) for gas mixtures of co@xmath1 and co.,title="fig:",scaledwidth=42.0% ] a ) for a total pressure @xmath60bar .
a ) for gas mixtures of o@xmath1 and ar .
b ) for gas mixtures of co@xmath1 and co.,title="fig:",scaledwidth=42.0% ] b )
it is often reported that crystal growth experiments are performed in an `` inert atmosphere '' .
this term is vague , as each gas that is technically available , including noble gases , contains residual impurities resulting in a small and often unknown residual oxygen partial pressure .
often the residual gas is air , resulting in @xmath61bar . without giving a quantitative explanation , mateika and
rusche @xcite reported that the mass loss of an empty ir crucible heated to @xmath62c was in pure co@xmath1 by ca .
40% lower compared to an atmosphere of 98% n@xmath1 + 2% o@xmath1 .
a quantitative thermodynamic treatment shows that dynamic atmosphere compositions , where oxygen is produced by equilibrium chemical reactions between gases , can produce variable @xmath36 in such a way , that the specific oxidation state @xmath63 of a desired meo@xmath5 is stable over the whole @xmath4 range of the crystal growth process .
moreover , properly chosen atmospheres based on co@xmath1 allow the application of ir crucibles under highly oxidative conditions .
such conditions allow the growth of oxides like zno which would otherwise easily be reduced to the metal . | commercially available gases contain residual impurities leading to a background oxygen partial pressure of typically several @xmath0bar , independent of temperature .
this oxygen partial pressure is inappropriate for the growth of some single crystals where the desired oxidation state possesses a narrow stability field .
equilibrium thermodynamic calculations allow the determination of dynamic atmosphere compositions yielding such self adjusting and temperature dependent oxygen partial pressures , that crystals like zno , ga@xmath1o@xmath2 , or fe@xmath3o can be grown from the melt . , , , , , , and a1 phase equilibria , a2 growth from melt , b1 oxides 64.75.-g , 81.10.-h , 82.33.pt , 82.60.hc |
during the last few years impressive progress was made in the experimental investigation of ultracold atomic gases in optical lattices.@xcite it is now possible to measure not only density profiles and static correlation functions but even dynamical quantities such as spectral functions of the trapped particles , using bragg spectroscopy.@xcite thereby momentum resolved bosonic spectral functions were obtained in the case of condensed bosons@xcite and across the phase transition to a mott insulator.@xcite by contrast , only rather few calculations of bosonic spectral functions have been performed so far for the bose - hubbard model .
early studies were based on the strong - coupling approximation to the bose - hubbard model.@xcite later the weak - coupling limit at zero temperature was thoroughly analyzed with the functional renormalization group.@xcite other approaches include the variational cluster approximation,@xcite which was used to investigate systems in one@xcite and two@xcite dimensions , the quantum rotor approximation,@xcite which was recently applied to three - dimensional systems,@xcite and the linked - cluster expansion ( lce).@xcite in particular the lce , which was employed to approximately solve the equations of the bosonic dynamical mean - field theory ( b - dmft),@xcite allows one to study the bose - hubbard model for strong interactions and near the phase boundary between the superfluid and the mott - insulating phase .
the b - dmft treats local correlations in time exactly and includes spatial correlations on the mean - field level .
it is applicable for all values of the hubbard interaction , density and temperature .
the b - dmft was derived subsequent to the dmft for lattice fermions@xcite and was applied to various bosonic problems@xcite as well as to mixtures@xcite of bosons and fermions .
it has been extended and applied also to inhomogeneous situations@xcite and to bosonic systems in non - equilibrium@xcite . in this paper we solve the bose - hubbard model within the b - dmft framework on a simple cubic lattice in three dimensions using a continuous - time quantum monte carlo ( ct - qmc ) solver.@xcite this approach is known to give excellent results for the phase diagram and static properties@xcite of interacting bosons .
we compute the momentum resolved and the momentum integrated spectral functions and the dispersion relation of interacting bosons in , both , the superfluid and the mott insulating phase .
the following questions will be addressed and answered : ( i ) how well can strong - coupling approaches , which are known to capture the phase diagram of correlated bosons very well , describe dynamical properties such as spectral functions ? ( ii )
how does the presence of the superfluid influence the spectral properties of normal bosons ? ( iii )
how is the dispersion relation modified by the interaction ?
the paper is organized as follows : in section [ sect2 ] we introduce the bose - hubbard model and recapitulate the main steps of the b - dmft framework .
we also introduce an improved method for calculating the self - energy which makes use of two - particle green functions and discuss the method for the numerical analytic continuation .
in section [ sect3 ] we present the results for the momentum resolved spectral function in the limit of weak , intermediate , and strong interactions , respectively . finally , section [ sect4 ] concludes the paper with a summary .
we consider spinless bosons on a lattice described by the bose - hubbard hamiltonian @xmath0 where @xmath1 ( @xmath2 ) is a bosonic creation ( annihilation ) operator on a lattice site @xmath3 , @xmath4 is the chemical potential , @xmath5 is the local interaction strength , and @xmath6 is the hopping amplitude .
we assume nearest neighbour ( nn ) hopping , i.e. , @xmath7 if sites @xmath8 and @xmath9 are nn and @xmath10 otherwise .
our calculations were performed for a simple cubic lattice with coordination number @xmath11 and @xmath12 in arbitrary units . in the b - dmft
the lattice problem is replaced by a single - site ( `` impurity '' ) problem with self - consistency conditions .
a detailed derivation can be found in ref . .
the impurity action reads @xmath13 here @xmath14 is the inverse of the temperature @xmath15 , @xmath16 is a geometrical parameter depending on a lattice type , and @xmath17 is the imaginary ( matsubara ) time .
we use the nambu vector notation for operators @xmath18 and complex variables @xmath19 , i.e. , @xmath20 then the connected green functions are defined by @xmath21 where @xmath22 denotes the equilibrium average in the grand canonical ensemble , and @xmath23.@xcite the last two terms in eq . represent the coupling of the site to two types of external mean - fields : ( i ) the bose - einstein condensate ( bec ) which is represented by a static mean - field @xmath24 , and ( ii ) the dynamical mean - field of normal bosons represented by the matrix @xmath25 , the elements of which are hybridization functions .
the hybridization functions are related to the green functions @xmath26 for a lattice with a cavity at site @xmath10 ( i.e. , where site @xmath10 is removed ) by@xcite @xmath27 similarly , the condensate field @xmath24 for a lattice with a cavity at site @xmath10 reads @xmath28 the connection between the condensate field on a full lattice , @xmath29 , and the condensate field on a lattice with a cavity , @xmath24 , is expressed by @xmath30\mathbf{\phi}.\ ] ] once the mean fields @xmath24 and @xmath31 are known it is possible to solve the impurity problem and calculate the local green functions and local condensate fields .
this can be achieved by exact diagonalization@xcite , the lce approximation@xcite , or ct - qmc.@xcite in this paper we use the ct - qmc method to solve the impurity problem .
our code is based on the work of anders _ _ et al.__@xcite we employ an improved calculation of the self - energy via two - particle green functions , to be discussed in section [ imp_se ] and in the appendix .
tests of our code show numerical agreement with the results of ref . .
once the single - particle green functions have been obtained by solving the single - site problem , the local self - energy can be calculated from the dyson equation @xmath32^{-1},\ ] ] where @xmath33 are even matsubara frequencies for bosons ( @xmath34 ) . in the b - dmft
the self - energies are local , i.e. , momentum independent .
the momentum resolved green functions are then given by @xmath35^{-1 } , \end{aligned}\ ] ] where @xmath36 is the dispersion relation for noninteracting bosons . to close the self - consistency equations we calculate the local green functions from the @xmath37-integrated dyson equation @xmath38^{-1 } , \end{aligned}\ ] ] where @xmath39 is the noninteracting density of states ( dos ) .
the self - energies can be calculated by solving the single - impurity model with the ct - qmc method and then using eq . .
however , this calculation is subject to stochastic uncertainties at high positive and negative frequencies for the following reason : in the first step the impurity problem is solved with the ct - qmc from which one obtains the green function @xmath40 , which has the form @xmath41 for large @xmath42 . the numerical error of this result is almost independent of @xmath33 .
therefore relative error grows linearly with @xmath33 . in the second step
the green function is inverted and substituted into eq . .
the linear term cancels out , and only terms of the order of @xmath43 , @xmath44 and smaller remain .
this results in a progressively larger relative error , i.e. , the error increases quadratically with frequency @xmath33 ( see fig .
[ self_energy ] ) . to improve the accuracy of the calculation of the self - energies we adapt the method proposed by snoek and hofstetter,@xcite and use the two - particle green functions .
the strategy goes as follows : finding equations for the self - energies requires solving the equations of motion for the green functions in the bosonic single - impurity model .
this is similar to the method proposed for fermions.@xcite the final expression for the self - energy reads @xmath45 \mathbb{g}^{-1},\ ] ] where @xmath46 is the fourier transform of a matrix of two - particle , _ disconnected _ green functions @xmath47 given by @xmath48(\tau ) \hat{b}^{\dag}(0 ) & [ \hat{b}^{\dag } \hat{b } \hat{b}](\tau ) \hat{b}(0 ) \\ [ \hat{b}^{\dag } \hat{b}^{\dag } \hat{b}](\tau ) \hat{b}^{\dag}(0 ) & [ \hat{b}^{\dag } \hat{b}^{\dag } \hat{b}](\tau ) \hat{b}(0 ) \end{pmatrix}\bigg\rangle.\ ] ] details of this derivation are presented in appendix [ appa ] . in fig .
[ self_energy ] we see that the improved method significantly reduces the error , i.e. , it now grows only linearly in frequency .
the main goal of this paper is to calculate the momentum resolved spectral function @xmath49 and the momentum integrated spectral function @xmath50 , respectively , of the bose - hubbard model . from the ct - qmc impurity solver
we obtain the green functions and the self - energies in imaginary time or in matsubara frequencies . since these functions are analytic in the upper complex plane we can analytically continue them to the real axis .
this involves an inversion of the hilbert transform : @xmath51 or @xmath52 it is well - known that this is a numerically ill - posed problem , in particular for noisy data obtained with ct - qmc . to make the analytic continuation tractable we use the maximum entropy ( maxent )
method.@xcite we found that the maxent procedure sometimes fails to converge , especially in calculations of sharply peaked momentum resolved spectral functions @xmath49 . in these cases
we use the _ historic _ rather than the _ bryan _ version of maxent .
in the following we present results for spectral functions of the bose - hubbard model on a simple cubic lattice for weak , intermediate , and strong interaction strengths . for illustration
we indicate the sets of parameters at which the calculations for strong and intermediate coupling , respectively , were performed in the phase diagram ( fig .
[ phase_diag ] ) .
the diagram was calculated within the static fisher mean - field theory @xcite and the b - dmft,@xcite which was solved by ct - qmc@xcite and lce.@xcite our results in the weak interaction limit were obtained for @xmath53 , @xmath54 , and @xmath55 . for these parameters the average occupation per site is @xmath56 and the average number of condensed bosons per site is @xmath57 . since @xmath58 is small in comparison to the bandwidth @xmath59 , the weak - coupling bogoliubov approximation@xcite can be expected to be applicable .
the momentum resolved spectral function @xmath49 obtained within the b - dmft is presented in fig .
[ spectr_weak]a .
there are two bands : with positive energies for particle addition and with negative energies for particle removal .
most of the spectral weight is concentrated in the upper band .
the widths of the peaks for specific @xmath37-points represent the mean lifetime of the quasiparticles .
the weight of the lower band is orders of magnitude smaller , and therefore its exact shape and position is not determined reliably by maxent .
the dispersion relation @xmath60 , obtained from @xmath49 according to the definition @xmath61 is shown in fig .
[ spectr_weak]b .
the dispersion relation obtained from the bogoliubov approximation is found to be in very good agreement with the b - dmft result , except near the @xmath62 point , where the b - dmft dispersion does not go to zero .
this is attributed to the fact that the hugenholtz - pines theorem@xcite is not fulfilled in the b - dmft as already reported by anders _ _
et al.__@xcite calculations in the limit of strong interactions were performed for @xmath53 , and @xmath63 . in fig .
[ spectr_scl ] we present the spectral function @xmath49 for the mott insulating phase at @xmath64 and the superfluid phase at @xmath65 ( see fig .
[ phase_diag ] ) .
a striking difference between the spectra in the mott insulating and superfluid phases , respectively , is the number of bands and their width at specific @xmath37-points . in the mott insulating phase
there are two bands , separated by a gap whose width is approximately given by @xmath66 .
this value becomes exact in the atomic limit ( @xmath67 ) .
the factor @xmath68 corresponds to the sum of half - widths of the upper and lower bands .
the shape of this dispersion relation is almost the same as that for a noninteracting band . in the mott insulating phase with average occupation @xmath69 one particle
is frozen on every lattice site such that an extra particle or hole can move almost freely through the system .
since the energy for creating an excitation is large , the density of excitations is very low , and therefore one can neglect their interaction . as a result quasiparticle and quasihole excitations
behave almost like noninteracting particles , i.e. , have almost noninteracting dispersion relations .
the width of the peaks is therefore small , implying that the mean lifetime is large .
in fact , we checked that the width corresponds to the uncertainty of the analytic continuation rather than to the mean lifetime .
the bandwidth of particle excitations is wider than that of hole excitations by approximately @xmath70 . in the strong interaction limit
the bandwidths of the hole and particle excitations in the mott insulating phase with integer filling @xmath71 are @xmath72 and @xmath73 , respectively .
this is a quantum effect related to particle indistinguishability , which is simple to derive by starting with a state in which each site is occupied by @xmath74 particles and treating the hopping term in the hamiltonian as a small perturbation . in the superfluid phase
this is no longer valid . as the chemical potential increases , the energy for creating particle excitations decreases
therefore the interaction between the quasiparticles needs to be taken into account .
indeed the spectrum of the superfluid is significantly different from that in the mott insulating phase .
in particular , we find four rather than two bands .
only the positions of the three lowest bands are determined reliably by maxent .
the four bands arise from two processes : ( i ) the splitting of a single band due to the interaction , similar to the mott insulating phase ; ( ii ) the creation of bogoliubov quasiparticles due to the mixing of particle and hole excitations .
we checked that , in contrast to the mott insulating phase , the width of the peaks for small @xmath75 is robust with respect to maxent parameters as well as re - sampling and therefore represents the mean lifetime of quasiparticles and not the accuracy of analytic continuation . in fig .
[ disprel_scl ] we show the dispersion relations @xmath60 obtained within the b - dmft scheme according to eq . .
we compare them with the dispersion relations obtained from the self - energies calculated within the static fisher mean - field approximation.@xcite the results of both methods are in good agreement in the mott insulating phase ( fig .
[ disprel_scl ] , top panel ) . in the superfluid phase ( fig .
[ disprel_scl ] , bottom panel ) this comparison is presented for the three dominant bands .
the results are also in good agreement with the static fisher mean - field results . in both approaches
the high - energy ( negative ) band is similar to the band of a noninteracting hole .
the remaining two dispersions are linear for small values of @xmath76 in the fisher mean - field theory and correspond to massless bogoliubov quasiparticles . in the b - dmft
we also see the linear behaviour except for the vicinity of the @xmath62-point .
the dispersion relation around @xmath77 does not go to zero , since the hugenholtz - pines theorem is not obeyed in the b - dmft ( see also section [ non_lim ] ) .
the slight deviation of the ct - qmc result from that of the static fisher mean - field theory@xcite in the middle of the band around the r point is attributed to the finite resolution of maxent .
the most interesting regime is that of intermediate interactions where both the bogoliubov approximation and the static fisher mean - field theory@xcite are no longer valid .
we computed the spectral functions in this regime within b - dmft and compare our ct - qmc results with those obtained within the lce@xcite .
we used the same set of parameters as in ref .
, i.e. , each set corresponds to a point in a \{@xmath78 } parameter space for @xmath79 .
the selected points allow us to study the evolution of the spectral functions through the phase transition . here
we consider the phase transitions driven by the change of the interaction and by the change of the chemical potential ( fig [ phase_diag ] , left and right panel , respectively ) . in fig.[lce_ctqmc ] we present the b - dmft results and compare with the lce .
there is good agreement regarding the widths and the positions of the bands in the superfluid phase .
however , in the mott insulating phase the lce spectral functions develop a two - peak structure for both positive nad negative energies .
this is not supported by the ct - qmc results .
the origin of this feature in the lce may be an overfitting of the numerical analytic continuation , since in the lce the error is small but unknown .
we now focus on the momentum dependence of the spectral functions and the dispersion relations for the mott insulating and superfluid phases ( figs .
[ ak ] , [ disprel_int ] ) .
the most striking feature is the difference of the width of the peaks for specific @xmath37-points in these two phases .
deep in the mott insulating phase the width represents the uncertainty of the maxent fit .
this is not the case in the superfluid phase in which the width represents the mean lifetime of quasiparticles .
effectively , deep in the insulating phase we can describe the particles as almost free , whereas approaching the phase transition the quasiparticles obtain a mean lifetime . in the superfluid phase we observe high - energy excitations , which are shown in fig .
[ ak]c , resembling a structure found in the strong - coupling regime . within the accuracy of our method a more precise calculation of their position is not possible .
the dispersion relations shown in fig.[disprel_int ] are compared with the strong - coupling dispersions obtained for @xmath80 and @xmath81 . in the mott insulating phase
the gap obtained by the static fisher mean - field theory@xcite is much smaller than in the b - dmft .
thereby the phase transition is shifted to larger interactions , see fig [ phase_diag ] . in the superfluid phase
we observe that , in contrast to the weak and strong - coupling limits , the dispersion relation obtained at intermediate coupling within the b - dmft is not symmetric with respect to zero energy .
, which leads to a symmetric dispersion relation . ] in fig .
[ spectr_weak_u10 ] we present the results obtained in the dilute gas regime .
the parameters are @xmath53 , @xmath82 , and @xmath83 .
this corresponds to @xmath84 and @xmath85 . as in the previous cases ( superfluid phase )
we observe that the peaks in the spectral function are wide , which means that the mean lifetime of excitations is finite .
the negative energy band has a weight which is orders of magnitude smaller , and therefore its exact shape and position is again not determined reliably by our method of analytic continuation .
the results at the @xmath62-point are not included , since maxent failed to converge for very small @xmath37 .
although the particle density is low and the majority of the particles constitute the condensate , the interaction is not weak .
therefore one is not in the parameter range where the bogoliubov approximation is applicable . indeed , as seen in fig .
[ spectr_weak_u10]b , the dispersion relation is not reproduced by this approximation .
a similar discrepancy was reported in the experiment@xcite , where at high momenta the bogoliubov approximation was found to overestimate the measured excitation energy .
a further effect of the interaction is the appearance of an additional band at high energies , fig .
[ spectr_weak_u10]a .
this high energy band appears to repel the one with low energies . as a result
the low energy band is narrower .
this explains the discrepancy between the b - dmft and the bogoliubov results , since the high energy band is absent in the latter .
we presented a computational method which allows one to calculate spectral functions for bosonic systems described with the bose - hubbard hamiltonian .
this approach has the great advantage of being applicable for arbitrary values of @xmath86 , the ratio of the interaction strength and the hopping amplitude .
it reproduces well the results in the limiting cases of large and small values of the interaction .
results obtained by this method were shown for parameters where perturbative methods fail , thus providing new insights into the properties of correlated lattice bosons in previously inaccessible parameter ranges .
the results show that while in the insulating phase particle and hole excitations tend to behave like free particles , in the superfluid phase the system is described by bogoliubov quasiparticles with a finite mean lifetime . finally , we discussed an alternative method for calculating the local self - energy in the ct - qmc solver which employs two - particle green functions , and which produces accurate data at large matsubara frequencies .
this work was supported by the foundation for polish science ( fnp ) international ph.d .
projects programme co - financed by the eu european regional development fund .
kb acknowledges support by the foundation for polish science ( fnp ) through the team/2010 - 6/2 project , co - financed by the eu european regional development fund . this research was also supported in part by the deutsche forschungsgemeinschaft through trr 80 ( dv )
we consider the hamiltonian representation@xcite of the impurity model @xmath87 where @xmath88 and @xmath89 are vectors in nambu notation . here
@xmath90 ( @xmath91 ) annihilates ( creates ) a particle from the bath state @xmath92 , @xmath18 ( @xmath93 ) annihilates ( creates ) a particle on the impurity , @xmath94 is the nambu matrix of couplings between the impurity and a bath state @xmath92 , @xmath95 is the energy of the bath state @xmath92 , and the vector @xmath24 represents the condensate field to which the impurity is coupled .
the bath state energies @xmath95 and couplings @xmath94 are chosen such that @xmath96 where @xmath97 is the dynamical mean field from eq . . for this hamiltonian
we consider the disconnected green functions @xmath98 here the indices @xmath10 and @xmath92 refer to the impurity and bath states , respectively .
the relation between @xmath99 and the connected impurity green function @xmath100 defined in is given by @xmath101 the time evolution of the operators is governed by the hamiltonian @xmath102 ( for brevity we will drop the index henceforth ) , and we can calculate the derivatives of the green functions with respect to imaginary time . for the local green function
we have @xmath103(\tau ) \\ [ \hat{h},\hat{b}^\dag ] ( \tau ) \end{pmatrix } \begin{pmatrix } \hat{b}^\dag ( 0 ) & \hat{b } ( 0 ) \end{pmatrix}\right\rangle \\ & = & -\delta(\tau)\bbsigma_3+\mu \bbsigma_3 \mathbb{g}^d_{00}(\tau ) -u\bbsigma_3 \mathbb{f}(\tau)\\ & & -\bbsigma_3\sum_l \mathbb{v}_{l}\mathbb{g}^d_{l0}(\tau)- \kappa\bbsigma_3\mathbf{\psi}\mathbf{\phi}^\dagger , \end{array}\ ] ] where @xmath104 $ ] denotes the commutator , @xmath105 is the diagonal pauli matrix , @xmath106 is a nambu vector , and the nambu matrix @xmath107 is defined as @xmath108 a similar relation holds for @xmath109 : @xmath110(\tau ) \\ [ \hat{h},\hat{a}^\dag_l ] ( \tau ) \end{pmatrix } \begin{pmatrix } \hat{b}^\dag ( 0 ) & \hat{b } ( 0 ) \end{pmatrix}\right\rangle \\ & = -\epsilon_l\bbsigma_3\mathbb{g}^d_{l0}-\bbsigma_3 \mathbb{v}^\dagger_{l}\mathbb{g}^d_{00}. \end{array}\ ] ] to handle the imaginary time derivative we perform a fourier transform to matsubara frequencies and obtain @xmath111 @xmath112 using ( [ app_g_omn10 ] ) it is easy to find an expression for @xmath109 in terms of @xmath113 : @xmath114 we concentrate now on the case @xmath115 , i.e. , nonzero matsubara frequencies .
inserting into one obtains @xmath116^{-1}(\mathbb{1}+u\mathbb{f}(\i\omega_n ) ) .
\end{aligned}\ ] ] using one can write @xmath117 combining with the dyson equation we arrive at for non - zero matsubara frequencies ( for @xmath118 one has @xmath119 ) . by inserting into one finds in the case @xmath120 : @xmath121
starting from for @xmath120 and replacing the disconnected green function with the connected one leads to @xmath122 in the last step the self - consistency condition , i.e. , @xmath123 , is inserted into . after some regrouping one obtains @xmath124-\mathbf{\phi}\mathbf{\phi}^\dagger . \end{aligned}\ ] ] after the above transformations one finally finds @xmath125,\ ] ] which , in combination with the dyson equation , reduces to for @xmath120 .
45ifxundefined [ 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 * * , ( ) `` , '' ( , ) chap . , pp . @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) | we calculate the momentum dependent spectral function of the bose - hubbard model on a simple cubic lattice in three dimensions within the bosonic dynamical mean - field theory ( b - dmft ) .
the continuous - time quantum monte carlo method is used to solve the self - consistent b - dmft equations together with the maximum entropy method for the analytic continuation to real frequencies .
results for weak , intermediate , and strong interactions are presented . in the limit of weak and strong interactions
very good agreement with results obtained by perturbation theory is found .
by contrast , at intermediate interactions the results differ significantly , indicating that in this regime perturbative methods fail do describe the dynamics of interacting bosons . |
[ fig : phelix ] -wave pairing states ) . @xmath84 and @xmath85 are measured in the unit of @xmath86 while @xmath87 and @xmath88 in the unit of @xmath89 defined below eq .
( [ eq : d ] ) in the main text .
@xmath90 shows the direction of the moment at each site . ]
[ fig : pskx ] -wave like pairing states ) . here
the spin configuration is given in a graphical way , which is identical to fig .
[ fig : skx ] ( a ) .
the arrows , circles with x ( into the paper ) and filled circle ( out of the paper ) inside indicate the directions of the moments .
broken lines on right side show the magnetic unit cell .
we take the same units as in fig .
[ fig : phelix ] . ]
we gave three particular spin configurations shown in figs .
[ fig : helix ] ( a ) , [ fig : skx ] ( a ) , and [ fig : dhelix ] ( a ) . here we show the explicit values of the effective parameters calculated from eqs .
( [ eq : t ] ) and ( [ eq : d ] ) . figures [ fig : phelix ] , [ fig : pskx ] , and [ fig : pdhelix ] ( for coplanar spin helix , skyrmion crystal , and non - coplanar spin helix , respectively ) show following information ; ( i ) direction @xmath91 of the moment at each site in a unit cell , ( ii ) effective transfer integrals @xmath92 and in the unit of @xmath86 , and ( iii ) effective pair potentials @xmath93 in the unit of @xmath89 .
the directions of moments are given by @xmath94 in figs .
[ fig : phelix ] and [ fig : pdhelix ] , and in a graphical way in fig .
[ fig : pskx ] .
the unit cell is @xmath61 block for both spin helix configurations , and skyrmion crystal contains 8 atomic sites in a unit cell . @xmath86 and @xmath89 are defined just below eq .
( [ eq : d ] ) in the main text .
[ fig : pdhelix ] -wave like pairing states ) .
we take the same units as in fig .
[ fig : phelix ] . ]
in the main text , we discuss the realization of the @xmath2-wave like state by means of the skyrmion structure . here in supplemental material
we consider another non - coplanar configuration of magnetic moments .
it is shown in fig .
[ fig : dhelix ] ( a ) .
this can be seen as follows ; we have sequential spin helix stripes along @xmath60-direction , but the planes on which spins rotate are tilted along @xmath53-direction . describing in more detail ,
we start from the moment at the origin pointing along @xmath30-direction , we rotate it by @xmath59 within @xmath58-plane along @xmath60-direction .
now we have a row of moments extended in @xmath60-direction .
then we rotate the plane where the moments lie by @xmath59 around @xmath95-axis along @xmath53-direction .
after three successive rotations , the plane comes back to the original one , i.e. , @xmath58-plane as shown in fig .
[ fig : dhelix ] ( a ) right side .
it is expected that the former rotation gives real @xmath41 s in eq .
( [ eq : d ] ) while the latter imaginary ones .
the resulting @xmath40 and @xmath41 are explicitly given in fig .
[ fig : pdhelix ] .
in fact , @xmath96 and @xmath97 , which are consistent with the invariance of them under common rotation of two spins . on the other hand , along @xmath53-direction the moments have finite @xmath65 component , and hence complex @xmath98 results . actually , they have pure imaginary values while @xmath99 are not necessarily real . note
that any coplanar configurations of moments are transformed onto @xmath58-plane by a common rotation , and all @xmath40 and @xmath41 are essentially real . to obtain complex @xmath41 s ,
the non - coplanar configurations of moments are indispensable . while the transfer integrals and pair potentials in the effective model are not uniform with this configuration ( see fig .
[ fig : pdhelix ] ) , it has the same topological nature as chiral @xmath3-wave superconductors as we will discuss below in the completely same ways as we do in the main text for the skyrmion structure .
first , the system described by the hamiltonian in eq .
( [ eq : tb ] ) has an energy gap in the whole brillouin zone , which suggests complex value of pair potential .
figure [ fig : dhelix ] ( b ) shows the energy dispersion of quasiparticles with the periodic ( open ) boundary condition along @xmath60- ( @xmath53-)direction .
the magnitude of @xmath20 is chosen to be @xmath100 so that the energy of the bound state around a single moment is about zero energy ( see fig .
[ fig : idea ] ) .
there are two one - dimensional bands as chiral majorana edge channels localized at the open boundaries in the energy gap .
to confirm the emergence of edge modes , we also calculate current density along @xmath60-direction with site indices @xmath12 s along @xmath53-direction . the result is given in fig .
[ fig : dhelix ] ( c ) , which clearly shows the presence of the edge current .
it also shows that the signature of the topological quantum phase transition at @xmath101 .
namely , the system enters into a trivial phase for @xmath102 where the edge modes and current vanish .
[ fig : dhelix ] -wave pairing .
they have a non - coplanar structure .
the moments along @xmath60-direction rotate by @xmath59 within the planes in the spin space schematically shown on the right side .
this plane is rotated along the @xmath53-direction by @xmath59 around the @xmath103-axis indicated by the arrows .
therefore @xmath61 sites constitute a unit cell .
( b ) energy spectrum of quasiparticles with the periodic ( open ) boundary conditions along @xmath60-(@xmath53-)direction .
( c ) current density calculated with the configuration given in ( a ) with the width @xmath104 .
the other parameters are the same as in ( b ) .
there are finite chiral current density at the edges . ]
we have also confirmed that zero energy majorana bound state appears at the cores of a vortices , which is consistent with chiral @xmath3-wave superconductors .
we insert vortices into the pair potential of the original @xmath0-wave superconductor . to avoid the effect of the boundary which may cause in - gap states
, we impose the periodic boundary condition for both @xmath60- and @xmath53-directions . for the consistency of the phase ,
we introduce two vortices with winding number 1 and other two anti - vortices with winding number -1 .
the four - fold zero energy states appear in the induced gap corresponding to the four vortices in total , and the probability of one of their wave functions in the real space shows localized distribution around the vortices as shown in fig .
[ fig : vortex ] .
these results are the same as those in the case of the genuine chiral @xmath3-wave superconductor .
[ fig : vortex ] with periodic boundary condition , and @xmath75 .
the zero energy states are four - fold degenerate .
( b ) distribution of a wave function of one of the zero energy states .
it has peaks at the positions of vortex cores . ]
in addition we study the effect of a defect . here
we use `` defect '' in the meaning of taking away one magnetic moment from the suite of aligned magnetic moments . in the tight binding picture
, there is a lattice site for the original @xmath0-wave superconductor at the defect point . in the conventional @xmath0-wave superconductor
, the non - magnetic defect does not produce any in - gap state .
in sharp contrast , we find that new states appear inside the lower energy gap .
note that the energy is not necessarily zero .
this property is characteristic to unconventional superconductors .
@xcite we employ the rotation angle @xmath59 in the coplanar and non - coplanar spin helix structure for the sake of numerical calculation .
however , the properties discussed above obviously hold when we consider other angles as long as it is not equal to @xmath105 or @xmath106 .
as stated in the main text , we assume the configuration of magnetic moments are strictly fixed all through the calculations . here
we examine the influence of spin fluctuation .
it is shown that the results obtained above are qualitatively unaffected . to take into account the effect of spin fluctuation ,
we mimic it by modulating directions of the moments with random magnitude about @xmath107 from the static configuration ( fig .
[ fig : skx ] ( a ) ) . in fig .
[ fig : fluc ] the edge modes still connect occupied and unoccupied states even though they are slightly modified by the fluctuation . | unconventional superconductivity induced by the magnetic moments in a conventional spin - singlet @xmath0-wave superconductor is theoretically studied . by choosing the spin directions of these moments
, one can design spinless pairing states appearing within the @xmath0-wave superconducting energy gap .
it is found that the helix spins produce @xmath1-wave state while the skyrmion crystal configuration @xmath2-wave like state .
nodes in the energy gap and the zero energy flat band of majorana edge states exist in the former one , while the chiral majorana channels along edges of the sample and the zero energy majorana bound state at the core of the vortex appear in the latter case . _
introduction. _ unconventional superconducting states are one of the most important issues in current condensed matter physics .
@xcite although most of the superconductors show the conventional spin - singlet @xmath0-wave pairing , strongly correlated materials sometimes show unconventional pairing since the on - site pairing is suppressed by the repulsive interaction . however , the discovery of the unconventional pairings relies on serendipity to some degree , and their theoretical designs and artificial fabrications are highly desired .
especially , recent intensive interest in the topological superconductivity and consequent majorana fermions enhance the importance of this topics since majorana fermions are the leading candidate for the platform of the quantum computation .
@xcite a promising proposal for realization of a topological superconducting state is the combined system of semiconducting nanowire with @xmath0-wave superconductor under an external magnetic field .
the spin - orbit interaction and the magnetic field reduce the degrees of freedom of electrons concerning superconducting states , and effectively generate spinless @xmath3-wave superconductor .
@xcite as for one - dimensional system , signals suggesting majorana fermions at the ends of the nanowire have been observed in some experimental setups .
@xcite there are other routes for creating topological superconductors ; spin - singlet superconductor deposited on the topological insulator .
@xcite superfluid of cold atoms with laser - generated effective spin - orbit interaction .
@xcite , aligned quantum dots connected by @xmath0-wave superconductors , @xcite and magnetic moments in @xmath0-wave superconductors @xcite or nodal superconductors . @xcite the last ones are significantly distinct in that they do nt explicitly require spin - orbit interaction in the system . with respect to the cooperation between magnetic moment and superconductivity , it has been known that the bound states are created around the impurities with the energy inside the bulk superconducting gap ( not necessarily zero energy ) .
@xcite the modulation of the local density of states by a single magnetic impurity has been observed in the experiment .
@xcite the authors of refs considered the one - dimensional array of the magnetic impurities , and studied the possibility of kitaev state with the majorana bound states at the ends of the array .
the influence of magnetic moments on a superconductor by the proximity effect has been intensively studied albeit in different interests .
@xcite on the other hand , it has been recognized that the spin - orbit interaction at the interface results in rashba type interaction and hence non - collinear spin configuration is organized .
@xcite especially , the skyrmion crystal state is observed at the interface of fe and ir .
@xcite therefore , the magnetic proximity effect of non - collinear moments to a superconductor becomes a realistic and important issue .
( a ) schematic illustration for the formation of an effective lattice model from the bound states localized around magnetic moments on the surface of @xmath0-wave superconductor ( see eq .
( [ eq : wf ] ) ) .
( b ) energy levels of quasiparticles obtained by a tight binding model calculation with a single moment with @xmath4 in eq .
( [ eq : tb ] ) .
@xmath5 is the coupling constant between electrons and magnetic moment and @xmath6 is the magnitude of the spin moment .
the solid ( blue ) curve shows the analytical solution @xmath7 for the continuum model ( see the main text ) . ] in this paper , we propose a generic principle to design unconventional superconductivity in terms of non - collinear / non - coplanar configurations of magnetic moments on the surface of an @xmath0-wave superconductor .
we derive an effective model constituted from the bound states around magnetic moments . the effective pair potentials as well as transfer integrals in the effective model
depend on the directions of two neighboring moments .
we show that a @xmath1-wave pairing state with nodes in the energy gap is generated by a non - collinear helical spin configuration , and moreover , we design a topological @xmath2-wave like state by means of a non - coplanar skyrmion crystal configuration of moments , as evidenced by chiral majorana channels along the edges of the system and zero energy majorana bound states at the cores of vortices . _
model. _ figure [ fig : idea ] ( a ) shows a schematic illustration of the present model .
we analyze the following tight - binding hamiltonian describing double exchange model with the superconducting order parameter defined on a square lattice @xmath8 the first three terms describe a conventional spin - singlet @xmath0-wave superconductor with the transfer integral @xmath9 , the chemical potential @xmath10 , and the pairing potential @xmath11 . in addition ,
electrons couple with magnetic moments located at sites @xmath12 s with the strength @xmath5 through double exchange mechanism .
this model can describe the interface between a bulk @xmath0-wave superconductor and a magnetic material .
we assume that the pairing potential is not affected by magnetic moments , which are supposed to be solidly ordered .
below we construct unconventional superconducting states with some particular structures of magnetic moments .
we derive an effective model in the aim of choosing appropriate magnetic order for intended unconventional states before directly solving eq .
( [ eq : tb ] ) .
first , we start with the case of a single moment in a superconductor .
the bogoliubov - de gennes equations are given by @xmath13 where we set the origin at the site of moment , @xmath14 , and @xmath15 is the tight - binding dispersion . the numerical result of energy level is shown in fig . [ fig : idea ] ( b ) .
the dispersion can be approximated in the continuum limit as @xmath16 with @xmath17 .
we can find solutions with energy @xmath18/
\left[1+(\pi jsn_0/2)^2\right]$ ] , where @xmath19 is the density of states in the normal phase @xcite ( solid line in fig .
[ fig : idea ] ( b ) ) . with increasing the magnitude of @xmath20
, @xmath21 changes from @xmath22 to @xmath23 within the bulk superconducting energy gap .
the corresponding wave functions are real and asymptotically given for @xmath24 as @xmath25 , \nonumber \\ v_{\downarrow}(\bm{r } ) & \sim \frac{\sin(p_{\mathrm{f}}r-\delta_-)}{p_{\mathrm{f}}r } \exp\left [ -\frac{r}{\xi_0 } |\sin(\delta_+-\delta_-)|\right],\end{aligned}\ ] ] where we define some quantities ; @xmath26 , @xmath27 is the fermi momentum , @xmath28 is the fermi velocity , and @xmath29 . when the moments are aligned in a lattice , we expect that the bound state around each moment has overlap with neighboring bound states .
the overlap causes effective transfer integrals and pair potentials among the bound states .
the low energy properties , i.e. , in the bulk superconducting gap , can be described by an effective bdg lattice model constructed from these bound states .
one can find similar arguments in refs . .
_ design of @xmath3-wave superconducting states. _ we propose a principle to design the superconducting states appearing within the gap of the host spin - singlet @xmath0-wave superconductor .
it will be shown that configurations of magnetic moments play the essential role for the emergence of unconventional superconducting states . in the last section ,
we assume the magnetic moment parallel to @xmath30-direction . here
, we introduce a unitary transformation for the description of general directions of moments . the coupling term with magnetic moments in eq .
( [ eq : tb ] ) can be transformed as @xmath31 by the unitary matrix @xmath32 where @xmath33 and @xmath34 are the polar coordinates such that @xmath35 .
the wave functions for arbitrary spin directions are obtained by operating @xmath36 on eqs .
( [ eq : wf ] ) .
then , the electron operators are expressed for the low energy sector as @xmath37 , \nonumber \\
\psi_{\downarrow } & = \sum_{i } \left [ e^{i\phi_{i } } \sin\frac{\theta_{i}}{2 } u_{\uparrow}(\bm{r}-\bm{r}_{i } ) \alpha_{i } + \cos\frac{\theta_{i}}{2 } v_{\downarrow}^{*}(\bm{r } - \bm{r}_{i } ) \alpha_{i}^{\dagger } \right],\end{aligned}\ ] ] where @xmath38 is the annihilation operator of the bound state around the moment located at site @xmath12 . by substituting eqs .
( [ eq : me ] ) into the original hamiltonian eq .
( [ eq : tb ] ) , we obtain @xmath39,\end{aligned}\ ] ] where @xmath40 and @xmath41 are effective transfer integrals and pair potentials for the nearest neighbor sites @xmath42 in the present low energy hamiltonian .
we keep them up to the nearest neighboring sites . here
, we define @xmath43 which represents the spin as @xmath44 .
the effective transfer integrals and pair potentials are represented by @xmath45 with @xmath46 $ ] , @xmath47 $ ] , @xmath48 , and @xmath49 . based on these equations
, we can design various kinds of superconducting states .
note that @xmath40 and @xmath41 are invariant for the common rotation of both moments at sites @xmath12 and @xmath50 .
namely , these quantities depend only on the relative direction of the two moments .
the electron spin of the bound state is uniquely determined by the magnetic moment , and hence we have a spinless lattice model with controllable parameters depending on the configuration of moments . _ numerical studies. _ we design two - dimensional superconducting states by choosing appropriate configurations of moments designed from eqs .
( [ eq : t ] ) and ( [ eq : d ] ) ( see supplemental material [ sec : td ] ) .
we investigate properties of the system _ by directly solving the original tight - binding hamiltonian _
( eq . ( [ eq : tb ] ) ) .
the calculations are performed with transfer integral @xmath51 and on - site superconducting order parameter of the host conventional superconductor @xmath52 .
the magnetic moment is attached to each site .
we consider both periodic and open boundary conditions to see bulk properties and edge states of the system , respectively .
hereafter when we calculate the dispersion of andreev bound states , we use a cylindrical configuration with open boundaries along @xmath53-direction .
we consider following two cases ; @xmath54-wave pairing and @xmath55-wave like pairing .
\1 . _ nodal superconductor .
_ we can see from eqs .
( [ eq : t ] ) and ( [ eq : d ] ) that @xmath41 vanishes when the two neighboring moments point the same direction ( @xmath56 ) , while @xmath40 vanishes for the opposite direction ( @xmath57 ) .
therefore , non - collinear spin configurations are required to obtain the nontrivial states .
we know that one - dimensional helical or spiral spin structure generates @xmath3-wave superconducting states .
@xcite ( a ) configuration of moments producing effective @xmath1-wave pairing .
all spins lie in @xmath58-plane rotating by @xmath59 along both @xmath60- and @xmath53-directions , therefore @xmath61 block constitute a unit cell .
( b ) energy spectrum of quasiparticles with the configuration in ( a ) calculated by tight binding model ( eq . ( [ eq : tb ] ) ) with @xmath62 and @xmath63 .
the spectrum has two point nodes as expected .
( c ) energy spectrum with open boundaries .
one can see dispersionless andreev bound states connecting two nodal points .
] we can generalize this to two - dimensions .
we choose the spinor @xmath64 s so that spins rotate by @xmath59 around @xmath65-axis along both @xmath60- and @xmath53-directions as shown in fig .
[ fig : helix ] ( a ) and fig . [ fig : phelix ] in supplemental material [ sec : td ] ; all the moments lie in the @xmath58-plane .
the resulting state is expected to have real order parameter with @xmath3-wave pairing for @xmath60- and @xmath53-directions .
this can be named as @xmath1-wave pairing state .
the present superconducting state is not stabilized as a bulk phase because it is energetically disadvantageous compared with chiral @xmath3-wave pairing without nodal structures . however , in the model considered here , induced @xmath3-wave pairing is localized in the vicinity of the surface of the bulk superconductor , and is controlled by the configuration of moments . as a result , @xmath1-wave pairing state is realized by a single spiral structure of moments .
we have tested our expectation by numerical calculations solving eq .
( [ eq : tb ] ) .
figure [ fig : helix ] ( b ) shows the dispersion of the quasiparticles , where the induced energy gap has nodes on the line of @xmath66 .
we also calculate andreev bound states at the boundary of the system .
the resulting dispersions have flat bands , which are believed to be a hallmark of unconventional superconductors . @xcite it is noted that flat band of zero energy bound states
are realized starting from conventional @xmath0-wave superconducting pairing .
one can find in ref . a related work . \2 .
_ chiral @xmath3-wave superconductor .
_ next we attempt to generate fully gapped superconducting states . for this purpose
, the phase of pair potential along @xmath60- and @xmath53-direction should be different .
the case where the phase difference is equal to @xmath67 is well known as chiral @xmath3-wave superconductor .
we can always choose the phases of the pairing order parameter @xmath68 as real by appropriate gauge transformation once the moments lie in a plane .
then we need to consider non - coplanar spin configurations . here
we study the case of skyrmion crystal state ( fig .
[ fig : skx ] ( a ) ) recently observed experimentally .
@xcite in this case , moments obviously have a non - coplanar configuration .
the parameters in the effective model calculated by eqs .
( [ eq : t ] ) and ( [ eq : d ] ) are given in supplemental material [ sec : td ] ( fig .
[ fig : pskx ] ) .
we have confirmed the following properties of the system , and based on them we conclude that it has the same topological nature as chiral @xmath3-wave superconducting states while the transfer integrals and pair potentials are not uniform with this configuration .
the characteristic properties of chiral @xmath3-wave superconductors are ( i ) the full gap nature , ( ii ) the existence of edge modes and currents , and ( iii ) the emergence of zero energy states at the cores of vortices .
@xcite first we have confirmed that the system has an energy gap in the whole brillouin zone by the same calculation as fig .
[ fig : helix ] ( b ) , which indicates the complex value of the pair potential because the system is essentially spinless .
figure [ fig : skx ] ( b ) shows the energy dispersion of quasiparticles with the open boundaries .
one can see two linearly dispersing bands crossing at @xmath69 . which are localized at the edges of the system .
@xcite here the fermi surface in the normal state is hole - like , then chiral edge modes cross at @xmath69 not @xmath70 .
the fact that these modes yield edge current is manifested by directly calculating current density @xmath71 the result is shown in fig .
[ fig : skx ] ( c ) . at @xmath72
the clear signal appears indicating the topological quantum phase transition .
we also confirmed that zero energy majorana bound states appear at the cores of vortices ( fig .
[ fig : skx ] ( d ) ) .
we conclude that the resulting superconducting state is in the same topological phase as the chiral @xmath3-wave superconductor based on these observations . as another example of the non - coplanar spin configuration , we study the double spiral structure in supplemental material [ sec : dhelix ] , and found the similar @xmath2-wave like state . ( a ) configuration of skyrmion lattice producing an effective fully gapped superconducting phase ( see also fig .
[ fig : fluc ] in supplemental material [ sec : fluc ] ) .
( b ) energy spectrum of quasiparticles with the configuration in ( a ) calculated by tight binding model ( eq . ( [ eq : tb ] ) ) with @xmath62 and @xmath63 with open boundary condition along @xmath53-direction .
one can see linearly dispersing bands at @xmath73 which are localized at the boundaries of the system .
( c ) current density calculated with the configuration given in ( a ) with 20 unit cells along @xmath53-direction .
one can see finite current density at the edges of the system , which is a crucial evidence for topologically nontrivial phase .
( d ) energy levels of quasiparticles obtained by a tight binding model calculation with vortices .
the horizontal axis shows the indices of energy eigenvalues .
the system size is set @xmath74 with periodic boundary condition , and @xmath75 .
the zero energy states are four - fold degenerate . ]
_ discussion and conclusions. _ in this paper , we have proposed a new way of creating effective two - dimensional unconventional superconductivity by local moments on the conventional spin - singlet @xmath0-wave superconductor .
the non - collinear configurations of moments are essential to induce @xmath3-wave pairing .
there is a hierarchical structure in energy scale , i.e. , the original @xmath0-wave energy gap and that of the induced @xmath3-wave superconductivity .
andreev bound states by a topological origin appear within the latter energy gap .
even the chiral @xmath76-wave like pairing is realized by the non - coplanar configuration of moments , which shows the chiral majorana edge channel with linear dispersion and zero energy majorana bound states at the vortex cores .
moreover it indicates that we can create various kinds of superconducting states by choosing appropriate configurations of moments .
for example , by changing the distance between the moments , one can tune the magnitudes of effective transfer integrals and pair potentials .
then , the anisotropy @xmath77 can be controlled to obtain the rich topological phases discussed in ref . .
here we briefly discuss the effect of self - energy correction and spin fluctuation for legitimizing our approach and results .
the self - energy correction due to the dynamical quantum fluctuation of spins can be estimated as @xmath78 and @xmath79 where @xmath80 is the dimensionless coupling constant with @xmath81 being the exchange coupling between spins in the magnet , and @xmath82 the fermi energy of the superconductor .
this correction is tiny at small electron energy @xmath83 , and does not change the mini - gap structure . also we have confirmed numerically the robustness of the induced gap structure against the small ( static ) spin fluctuation as shown in fig . [
fig : fluc ] in supplemental material [ sec : fluc ] .
though our model will simulate the interface of bulk superconductors and magnetic materials , we end with an account of another experimental realization of these proposals . to create intended patterns of magnetic moments
, we can use atomic manipulation techniques using scanning tunneling microscopy .
@xcite the spin structure in organized magnetic impurities is also observed , @xcite although it is antiferromagnetic and can not be utilized for our proposal .
_ acknowledgment. _
s. n. was supported by grant - in - aid for jsps fellows .
this work was supported by grant - in - aid for scientific research ( s ) ( grant no .
24224009 ) ; the funding program for world - leading innovative rd on science and technology ( first program ) ; strategic international cooperative program ( joint research type ) from japan science and technology agency ; innovative areas `` topological quantum phenomena '' ( grant no .
22103005 ) from the ministry of education , culture , sports , science , and technology of japan .
50ifxundefined [ 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.63.239 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.43.1892 [ * * , ( ) ] link:\doibase 10.1143/jpsj.81.011013 [ * * , ( ) ] link:\doibase 10.1103/physrevb.61.10267 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.86.268 [ * * , ( ) ] \doibase doi : 10.1016/0550 - 3213(91)90407-o [ * * , ( ) ] link:\doibase 10.1103/revmodphys.80.1083 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.83.1057 [ * * , ( ) ] http://stacks.iop.org/0034-4885/75/i=7/a=076501 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.104.040502 [ * * , ( ) ] link:\doibase 10.1103/physrevb.81.125318 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.077001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.105.177002 [ * * , ( ) ] http://stacks.iop.org/0953-8984/25/i=23/a=233201 [ * * , ( ) ] link:\doibase 10.1126/science.1222360 [ * * , ( ) ] link:\doibase 10.1021/nl303758w [ * * , ( ) ] link:\doibase 10.1038/nphys2429 [ * * , ( ) ] link:\doibase 10.1038/nphys2479 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.100.096407 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.104.067001 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.103.020401 [ * * , ( ) ] link:\doibase 10.1038/ncomms1966 [ * * , ( ) ] link:\doibase 10.1103/physrevb.82.045127 [ * * , ( ) ] link:\doibase 10.1103/physrevb.84.195442 [ * * , ( ) ] link:\doibase 10.1103/physrevb.85.144505 [ * * , ( ) ] link:\doibase 10.1103/physrevb.88.020407 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.110.096403 [ * * , ( ) ] link:\doibase 10.1143/ptp.40.435 [ * * , ( ) ] http://www.jetp.ac.ru/cgi-bin/e/index/e/29/6/p1101?a=list [ * * , ( ) ] http://wulixb.iphy.ac.cn/en/abstract/abstract851.shtml [ * * , ( ) ] link:\doibase 10.1126/science.275.5307.1767 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.77.935 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.77.1321 [ * * , ( ) ] link:\doibase 10.1103/physrevb.79.054523 [ * * , ( ) ] http://dx.doi.org/10.1038/nnano.2013.29 [ * * , ( ) ] http://dx.doi.org/10.1038/nphys2045 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.78.373 [ * * , ( ) ] link:\doibase 10.1143/ptp.76.1237 [ * * , ( ) ] link:\doibase 10.1140/epjb / e2004 - 00066 - 4 [ * * , ( ) ] http://stacks.iop.org/0034-4885/63/i=10/a=202 [ * * , ( ) ] link:\doibase 10.1103/physrevb.83.224511 [ * * , ( ) ] link:\doibase 10.1103/physrevb.87.054501 [ * * , ( ) ] link:\doibase 10.1134/1.568223 [ * * , ( ) ] link:\doibase 10.1103/physrevb.64.054514 [ * * , ( ) ] link:\doibase 10.1103/physrevb.86.100504 [ * * , ( ) ] link:\doibase 10.1126/science.262.5131.218 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.92.056803 [ * * , ( ) ] link:\doibase 10.1126/science.1125398 [ * * , ( ) ] link:\doibase 10.1126/science.1214131 [ * * , ( ) ] link:\doibase 10.1143/jpsj.68.3054 [ * * , ( ) ] @column@grid * supplemental material for + `` two - dimensional @xmath3-wave superconducting states with magnetic moments + on a conventional @xmath0-wave superconductor '' * |
transit spectroscopy is currently the leading technique for studying exoplanet atmospheric composition . following the discovery of the first exoplanet atmosphere @xcite
, transit observations have enabled the characterization of a number of different exoplanets for atmospheric molecular species , clouds , and/or hazes @xcite .
recent observational results have shown that hot jupiter transit spectra demonstrate a complex continuum from clearsky to heavily clouded conditions @xcite , and that cloudiness remains a key factor into the regime of lower mass planets @xcite .
the field of exoplanet transit spectroscopy will be revolutionized with the anticipated launch of nasa s _ james webb space telescope _ ( _ jwst _ ) in 2018 @xcite . over the course of the design five year mission for _ jwst _ , the observatory is expected to provide in - depth observations of many tens of transiting exoplanets @xcite . some of these observations will probe planets in the poorly understood 24 earth mass regime @xcite .
excitingly , _
jwst _ may even be capable of characterizing temperate earth - sized planets , though the ability to conduct such studies will depend largely on how well systematic noise sources can be constrained @xcite . as the quality of transit spectrum observations continues to improve , so should models of exoplanet transits .
thus , certain processes initially thought to be of second - order importance should be revisited and possibly added to modeling tools .
for example , atmospheric refraction , which was initially shown to be unimportant for the case of certain hot jupiters @xcite , has recently been shown to be critically important for understanding some terrestrial exoplanet spectra and , possibly , gas giant transit spectra @xcite .
additionally , refraction can affect the shape of an exoplanet transit lightcurve .
beyond refraction , another process that has seen little study with regards to exoplanet transits is light multiple scattering .
@xcite used a plane - parallel monte carlo scattering model to determine the significance of a molecular rayleigh scattering `` glow '' around the limb of a transiting exoplanet . for the case of hd 209458b
, these authors found that the contribution of rayleigh scattering glow to the transit lightcurve is negligible .
similar conclusions were reached by @xcite , who used analytic arguments to deduce that multiple isotropic scatterings could have only a small impact on transit depths . since the study of @xcite ,
scattering opacity in exoplanet transits has largely been treated as being equivalent to absorption opacity ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
this equivalence can not always hold .
for example , a very strong forward and conservatively scattering aerosol would only weakly attenuate a beam following a straight - line path from a stellar disk , passing through an exoplanet atmosphere , and traveling towards a distant observer .
this issue was plainly demonstrated by , who used a monte carlo scattering model to study the transmission of a pencil beam through a cloud or haze layer with variable optical thickness and whose scatterers had asymmetry parameters of either 0 , 0.9 , or 0.98 ( their figure 3 ) . in the work that follows ,
we do not seek to outline specific definitions for `` hazes '' or `` clouds . '' in general , though , we use the former to refer to aerosols that form aloft and grow due to coagulation during the sedimentation process , thereby developing a distribution that may look something like titan s tholin haze , which has a number density profile described ( roughly ) by an exponential with a constant scale height in titan s upper atmosphere @xcite . for clouds , we envision an aerosol distribution formed by lifting a gas to its condensation point , thereby developing a distinct cloud base with an overlying cloud deck whose thickness is controlled by mixing processes . advances in exoplanet and brown dwarf cloud models
have been reviewed recently by @xcite .
the most popular tools include the eddysed model , where an upward diffusion of condensible vapor balances a downward sedimentation of aerosol particles , which has been successfully used to study both cloud and haze processes in transiting exoplanet atmospheres @xcite .
additionally , the sophisticated microphysical cloud treatments of @xcite continue to see development and application to transiting exoplanets @xcite .
thus , the question of the relative importance of scattering , refraction , and absorption remains largely unexplored for a wide portion of exoplanet parameter space .
such studies have been hindered by the lack of available simulation tools , and also by a lack of a single , coherent theory of exoplanet transit spectroscopy .
significant advances can be made by developing such a theory , and by building models that implement these ideas . in this paper
, we construct a theory of exoplanet transit spectroscopy that spans key fundamental integral relations through to its efficient , vectorized implementation .
simplifications of the theory cover the geometric ( i.e. , straight - line ) limit and cases that add refraction . in general , though , we emphasize the `` full physics '' scenario , where a three - dimensional monte carlo model is used to incorporate scattering effects .
below , we introduce the concept of the `` path distribution '' , which effectively separates the paths of photons ( or rays ) through a transiting exoplanet atmosphere from gaseous and/or aerosol absorption effects .
as the former tends to vary weakly in wavelength , while the latter can vary strongly with wavelength , our approach is computationally efficient . following the presentation of these ideas
, we validate our implementation of the theory against a number of trusted simulation tools .
we then give examples of the path distribution for a variety of different model atmospheres .
finally , we use our tools to present the first simulated transit spectra of hot jupiters with multiply - scattering clouds .
transit spectra typically probe low density regions of exoplanetary atmospheres . here ,
molecular absorption lines are relatively narrow , as the effects of pressure broadening are not dominant over doppler broadening ( as is the case in the deep atmosphere ) .
thus , the gas opacity can vary by orders of magnitude over narrow spectral ranges .
by contrast , gaseous refractive indexes as well as gas and aerosol scattering properties tend to vary smoothly ( and , sometimes , weakly ) in wavelength .
since refraction and scattering , which are more computationally intensive processes to simulate , influence the path of a photon ( or ray ) through an atmosphere , significant advances can be made by outlining a technique that separates the processes which influence photon trajectories from those which influence absorption .
we introduce the concept of the ray ( or photon ) atmospheric path distribution , @xmath0 , where @xmath1 is the impact parameter of the ray and @xmath2 is altitude in the planetary atmosphere .
we define the path distribution such that @xmath3 is the linear distance traversed by the ray at altitudes between @xmath2 and @xmath4 . while @xmath5 is dimensionless , it can be thought of as having units of km linear distance per km vertical distance .
the geometry of these parameters , for a simple case , is shown in figure [ fig : refract_geom ] .
the appendix contains additional discussion on the use of the path distribution . in transit spectroscopy ,
the essential quantity is the attenuation of a ray along its path , as this distinguishes the opaque portions of the atmosphere ( that block light from the stellar disk ) from the transparent portions . in 1-d atmospheric structure models , which are the variety most commonly used for exoplanets ,
the atmospheric opacity is only a function of altitude ( or pressure ) and wavelength . here ,
then , the optical depth along the path of a ray can be computed using the path distribution via , @xmath6 where @xmath7 is the atmospheric extinction ( in units of inverse distance ) , and the integral can be taken to infinity as the extinction is zero at very large altitudes .
since the layer vertical differential optical depth is defined as @xmath8 , this expression demonstrates that the path distribution can also be thought of as the linear optical depth encountered between @xmath9 and @xmath10 .
similar extensions exist for quantities like the column mass and number densities .
note that equation [ eqn : tau_b ] is critical , as the strongly wavelength - dependent extinction is separated from the ray path . using the standard definition of transmission , @xmath11 a simple transit spectrum in the pure absorbing limit ( i.e. , where all optical depth is taken as absorption optical depth ) can then be computed by considering the light transmitted through concentric annuli on the planetary disk , each at their own impact parameter , @xmath12 b db \ , \label{eqn : transit_int}\ ] ] where @xmath13 is the wavelength - dependent planetary radius , @xmath14 is the stellar radius , and @xmath15 is the transit depth . in practice ,
model planetary atmospheres are defined on an altitude ( or pressure ) grid , and equation [ eqn : transit_int ] is computed using a sum over a collection of impact parameters ( and assuming that the planetary disk is opaque below some fiducial radius , @xmath16 , which is either at the surface or deep in the atmosphere ) . in this case , the path distribution is a matrix , @xmath17 , where @xmath18 is the path traversed through atmospheric layer ` @xmath19 ' ( whose thickness is @xmath20 ) by a ray whose impact parameter is @xmath21 . at any given wavelength , the transmission is now a sum over @xmath22 atmospheric layers , @xmath23 where we have used the definition of the layer vertical differential optical depth , @xmath24 .
a similar expression was used by ( * ? ? ?
* their equations 5 and 7 ) , where these authors work in numerically - computed column number densities and molecular opacities , as compared to our dimensionless quantities .
note that equation [ eqn : trans_sum ] can be easily written in matrix notation as , @xmath25 where we have defined the absorptivity vector , @xmath26 .
the pure absorption transit spectrum derives from a sum over @xmath27 impact parameters , @xmath28 b_{i } \delta b_{i } \right ) \ ,
\label{eqn : transit_sum}\ ] ] where @xmath29 is the thickness of the impact parameter gridpoint . continuing with the matrix notation ,
if we define a vector of annulus areas as @xmath30 , then the transit spectrum can be written simply as , @xmath31 we briefly note that the transit spectra expressions given in equations [ eqn : transit_int ] , [ eqn : transit_sum ] , and [ eqn : transit_vec ] are in the pure absorption limit , and assume that all rays trace back to the stellar disk , and that the star has uniform surface brightness . while these are common assumptions when computing transit spectra for model atmospheres , we will now discuss the more general multiple scattering case .
the appendix contains brief details about computing the path distribution in the geometric limit or in the case that refraction is considered .
cases that include clouds , especially strongly forward scattering clouds , require three - dimensional radiative transfer treatments to compute accurate transit spectra , and are well suited to monte carlo models . here ,
the path distributions for a number of photons , @xmath32 , are used to derive an average transmission for a grid of impact parameters .
the individual path distributions are determined using a monte carlo approach , and only consider the scattering optical depths ( @xmath33 , where @xmath34 is the single - scattering albedo , and a sub - script ` s ' indicates `` scattering '' while ` e ' indicates `` extinction '' ) since only @xmath35 affects the path of rays through the atmosphere .
given the path distribution , @xmath36 , for photon ` @xmath37 ' , the transmission for this photon is , @xmath38 where @xmath39 is a vector of the wavelength - dependent layer differential absorption optical depths , and the transmission averaged over all photons is simply , @xmath40 execution of our monte carlo model follows a standard approach . for a given impact parameter , @xmath21 ,
a photon is directed into the atmosphere along the @xmath41 direction in figure [ fig : refract_geom ] , and then tracked through the atmosphere .
as this photon originated from the direction of the observer , we are performing a so - called `` backward '' monte carlo simulation .
the optical distance the photon travels , either initially or following a scattering event , is determined by randomly sampling the scattering optical depth distribution , with , @xmath42 where @xmath43 is a random number between zero and one , and @xmath44 is the probability that a photon scatters between @xmath45 and @xmath46 . as @xmath47 , we have , @xmath48 for whichever layer ( centered at @xmath49 ) the photon is currently in , the pathlength corresponding to @xmath50 is , @xmath51 where @xmath52 is the layer scattering extinction coefficient . given a pathlength , a photon location ( @xmath53 ) , and trajectory ( @xmath54 ) ,
the photon position is updated according to @xmath55 , @xmath56 , and @xmath57 .
depending on the size of @xmath58 , the photon either experiences a scattering event within layer @xmath19 , in which case @xmath59 is added to the path distribution for this photon ( i.e. , @xmath36 ) , or the photon exits the layer before a scattering event occurs .
for the latter , @xmath58 must be compared to the distance to the layer boundaries . given the trajectory of the photon , and its current radial position , @xmath60 , the altitude of the photon along its path is , @xmath61 thus , the quadratic equations ( in @xmath58 ) , @xmath62 give the distances to the layer boundaries . when a photon travels to a layer boundary before a scattering event
, @xmath63 is added to @xmath36 , the cumulative optical depth experienced by the photon us updated as @xmath64 , and the photon is passed to the appropriate layer ( either @xmath65 or @xmath66 ) .
this process of randomly generating @xmath50 and passing the photon through sequential layers is repeated until the photon either exits the atmosphere , or reaches the lower boundary of the model and is considered absorbed .
( note that , as with any transit spectrum model of gaseous worlds , the lower boundary must be set deep enough to not influence the simulated spectrum . ) for photons that exit the atmosphere , the position and trajectory at exit determine whether or not the photon will intersect the stellar disk along a straight - line trajectory .
when a scattering event occurs , a scattering angle must be sampled from the scattering phase function and the photon trajectory must then be updated . assuming that the scattering phase function , @xmath67 , is only a function of the cosine of a single angle , @xmath68 , then a randomly sampled scattering deflection angle is determined via , @xmath69 and an azimuthal scattering angle , @xmath70 , is sampled uniformly , with , @xmath71 as the scattering angles are referenced from the propagation direction , a transformation must be done to convert the initial propagation direction , @xmath72 , and the scattering angles into a new propagation direction , @xmath73 .
standard expressions exist for completing this transformation ( e.g. , * ? ? ?
* their equations 22 and 23 ) .
refraction can be included in the monte carlo simulation .
as the photon moves along @xmath58 , the path can be sub - divided , and a curvature applied to each smaller @xmath74 . at the photon height , the curvature , @xmath75 , is computed according to equation [ eqn : curvature ] , and a deflection angle for the photon is determined from the refractive portion of equation [ eqn : dtheta_ds ] ( i.e. , @xmath76 ) .
the change in trajectory is referenced from @xmath72 and occurs in either the local upward or downward direction ( depending on the sign of @xmath75 ) .
a new set of direction cosines is found by requiring , @xmath77 @xmath78 and @xmath79 where the first equation implies that the trajectory changes through @xmath80 , the second equation forces the trajectory change to be either locally up or down , and the final equation ensures that the new direction of travel is a unit vector .
the computational efficiency of the monte carlo approach to multiple scattering in transit spectra outlined above can be increased in several ways .
first , for a given impact parameter , the monte carlo routine need only be executed if the straight - line scattering optical depth is a substantial fraction of the absorption optical depth .
( we choose a conservative cutoff at @xmath81 . ) also , as suggested by , the number of photons used in the monte carlo simulation at a given impact parameter should be influenced by the straight - line transmission
these authors recommend the use of @xmath82 photons in transparent conditions and up to @xmath83 photons in opaque conditions . however , used a traditional monte carlo radiative transfer approach , where photons are lost to absorption while passing through the atmosphere , thus driving the need for very large numbers of photons in their simulations . as we have separated absorption from scattering in our model , ( nearly ) all of the photons we simulate in our monte carlo exit the atmosphere , so that we only require @xmath84@xmath85 photons .
even more dramatic efficiency gains can be made by noting that , since our monte carlo model only considers scattering optical depths , these simulations need only be recomputed if either the layer differential scattering optical depths or asymmetry parameters change significantly .
thus , the individual path distribution matrices for each of the photons in our simulations can be saved and reused over a wide range of wavelengths in a high - resolution spectral grid . of course
, running @xmath32 transmission calculations ( i.e. , using equation [ eqn : phottrans ] ) at every wavelength in a high - resolution grid becomes computationally unfeasible if the grid is large . here
, runtime reductions of factors of 10100 can be straightforwardly made by computing the analytic jacobian of equation [ eqn : avgtrans ] , @xmath86 , with , @xmath87 this jacobian can be used to rapidly adapt @xmath88 to changes in @xmath39 , and our tests indicate that this approach is accurate with up to 2550% variations in level - dependent absorption optical depths .
since a set of path distributions , transmissions , and jacobians apply to wavelengths where the atmospheric optical property profiles are similar , the monte carlo approach outlined above is best implemented within a spectral mapping model .
we note that the many efficiency gains outlined above stem from our separate treatments of scattering ( which influences the path distributions ) and absorption ( which influences the path - derived transmissions and jacobians ) .
so far as we know , this treatment is a novel approach to monte carlo radiative transfer .
additionally , our approach to multiple scattering can be applied beyond exoplanet transits , and , for example , could be used to study scattering effects in occultation observations .
the transit spectra expressions given in equations [ eqn : transit_int ] , [ eqn : transit_sum ] , and [ eqn : transit_vec ] are primarily useful in the geometric limit , since rays in refracting or scattering models are not guaranteed to trace back to the stellar disk . to efficiently model spectra in these latter cases
, we generalize the concept of the vector of annulus areas , @xmath89 .
key variables used in the discussion below are visualized in figure [ fig : integration_geom ] .
ultimately , a transit spectrum is determined by comparing intensities integrated over the range of solid angles influenced by the planet ( @xmath90 ) in the case when the planet is present versus when only the star is considered ( with corresponding intensities @xmath91 and @xmath92 , respectively ) .
thus , @xmath93 where @xmath94 is the range of solid angles corresponding to the stellar disk . given the large distance @xmath95 between earth and any exoplanetary system , so that @xmath96 , and taking the stellar intensity to be normalized such that , @xmath97 then , @xmath98 the first integral in the numerator of equation [ eqn : transit_general ] is simply a stellar limb darkening law integrated over the portion of the planetary disk that overlaps the stellar disk . using a polar coordinate system centered on the planetary disk , and
letting @xmath99 be the distance from the center of the star to the center of the planet , this integral is then , @xmath100 where @xmath101 is the cosine of the angle of incidence on the stellar disk at coordinates @xmath102 and @xmath103 , and @xmath104 @xmath105 however , as limb darkening is accounted for in standard transit observation data reduction procedures , one may wish to omit the limb darkening law from the first integral in the numerator of equation [ eqn : transit_general ] , leaving an analytic integral that can be performed by considering the geometry of two overlapping disks .
the second integral in the numerator of equation [ eqn : transit_general ] contains the details associated with absorption , scattering , and refraction in the planetary atmosphere .
we define @xmath106 as the background surface brightness mapped onto by an area element on the planet at @xmath102 and @xmath103 .
this function is zero for rays that do not map back to the stellar disk , and , in the geometric limit , is equal to @xmath107 for portions of the planetary disk that overlap the star . given this definition , we then have , @xmath108 where we have assumed that @xmath16 is set sufficiently deep so that @xmath109 for @xmath110 .
of course , it is straightforward to consider thermal emission from the planetary disk , which can be important at longer wavelengths , by including the planet - to - star flux ratio for the planetary nightside on the right - hand side of equation [ eqn : transit_general ] . for efficient implementation
, we define a grid on the atmospheric portion of the planetary disk in @xmath111 angular and @xmath27 radial points . an @xmath112 area _ matrix _ , with @xmath113 , need only be computed once , and the @xmath112 background surface brightness mapping matrix , @xmath114 , is computed alongside the path distributions and transmissions . given these , we have @xmath115 \ , \label{eqn : tdvec}\ ] ] where ` @xmath116 ' indicates the hadamard product , @xmath117 is a vector of ones with length @xmath111 , and , as before , @xmath118 is computed using either equation [ eqn : trans_matrix ] or equations [ eqn : avgtrans ] and [ eqn : jacobian ] . critically , this approach still efficiently isolates the parameters that vary rapidly in wavelength , @xmath118 , from the parameters influenced by geometry and ray paths the background surface brightness mapping matrix need only be computed once in the geometric limit , and 12 times when refraction is included . for the multiple scattering case , full monte carlo simulations need only be run at wavelength scales over which the atmospheric scattering properties vary , and transmissions need only be recomputed at wavelength scales over which their linear corrections via the analytic jacobians become inaccurate .
integration can also be made more efficient by noting the symmetry about the line connecting the planet center to the star center .
the matrix @xmath114 is straightforward to compute in the geometric limit , as rays trace straight lines back to either the stellar disk or not .
when refraction is included , as in the appendix , a ray tracing is performed once for each gridpoint in @xmath27 ray impact parameters .
then , given the exit location and direction for these rays , and the angular location , @xmath119 , for each area element , simple geometry indicates which if the @xmath111 gridpoints have rays that map back to the stellar disk ( and what the @xmath120 value is for each of these rays ) . in the multiple scattering case
, @xmath32 monte carlo simulations are run , and @xmath114 is built up by averaging over the photons in each instance . for each photon that exits the atmosphere ,
its trajectory leaving the top of the atmosphere is investigated to see if the stellar disk is intersected .
the geometry is then rotated through @xmath121 , and the intersection is reinvestigated .
thus , only @xmath32 monte carlo simulations are needed to build up the average @xmath112 background mapping matrix .
figure [ fig : simulation ] shows the surface brightness in a full monte carlo simulation prior to integration over solid angle .
a haze with vertical optical depth of @xmath122 is placed above the 1 @xmath123bar pressure level following a linear power - law in pressure .
the atmosphere is isothermal at 1,500 k , the stellar and planetary size are appropriate for hd 189733b , and refraction and rayleigh scattering are included ( at 0.8 @xmath123 m wavelength ) .
a secondary figure enhances the surface brightness off the stellar disk by a factor of 10 . for the portion of the planet that is not overlapping the stellar disk ,
the sub - figure with enhanced surface brightness shows a thin ring due to forward scattered light .
additional surface brightness structure in this ring is related to the scattering phase function the region of the planetary limb that is at the greatest angular separation from the stellar disk less strongly samples the haze forward scattering peak .
the `` path distribution '' formalism represents a framework that allows for the computation of transit spectra across a broad range of conditions .
most existing transit spectrum models assume that rays travel along straight - line trajectories , and are extinguished from the line - of - sight path according to the extinction optical depth .
equations [ eqn : pathgeom ] , [ eqn : trans_sum ] , and [ eqn : transit_sum ] apply in this simplified regime .
slightly more sophisticated transit spectrum tools incorporate the physics of refraction through a ray tracing scheme .
we outline a similar ray tracing approach to computing the path distribution for a refracting atmosphere in the appendix . for planets centered on their host stellar disk ,
only a single ray tracing need be performed .
this symmetry is broken as the planet moves away from the center of the disk , implying that different locations around the planetary disk may or may not refractively map back to the stellar disk . here ,
equation [ eqn : trans_sum ] gives the ( radially dependent ) transmission , and equation [ eqn : tdint ] or [ eqn : tdvec ] describes the angular integration required to obtain a transit spectrum .
extremely few models ( or modeling approaches ) exist that allow for the simulation of multiple scattering effects in transit spectra .
@xcite describes a monte carlo model that treats the stratified limb of a transiting exoplanet as a set of non - interacting slabs .
following these authors investigation into rayleigh scattering effects , this model has seen little use .
more recently , constructed a realistic , three - dimensional monte carlo for investigating forward scattering effects in occultation and transit observations .
this tool has not been used to predict exoplanet transit spectra that include the effects of forward scattering , though .
our own three - dimensional monte carlo approach ( detailed above ) , when paired with the path distribution formalism and with its use of analytic jacobians , enables the calculation of high - resolution , multiple scattering transit spectra with computational efficiency gains of factors of 10@xmath12410@xmath125 over previous three dimensional models .
to best enable the implementation of the theory outlined above , a stand - alone model , which we call scatran , has been made publicly available .
the software can be downloaded from https://github.com/tdrobinson/scatran , or can be found by searching github for `` scatran . ''
we validated our transit model , which we implement within the framework of the spectral mapping atmospheric radiative transfer ( smart ) model , using a number of techniques and data sources .
first , with refraction and scattering optical depth omitted , we verified that our ray tracing routine ( discussed in the appendix ) and the monte carlo approach return the path distribution in the geometric limit .
then , with refraction included and scattering optical depth omitted , we showed that the ray tracing and monte carlo routines are in agreement . finally , to check our monte carlo approach with scattering included , we made the simple switch to a plane - parallel geometry ( where height is only measured relative to the z - axis , not radially ) and validated the output radiances against results from the disort radiative transfer model @xcite for a wide range of conditions .
figure [ fig : mc_valid ] shows the results from two of these experiments , plotting the top - of - atmosphere intensity ( scaled by the incident flux ) as a function of observer zenith angle . clouds with a given scattering optical depth and asymmetry parameter were placed over a perfectly absorbing surface , and a henyey - greenstein phase function was assumed . to further validate our model , we performed model inter - comparisons .
first , using a standard earth atmospheric model @xcite , we compared our transit spectrum for an earth - sun twin system to the refracting model of @xcite .
note that the @xcite model has been extensively validated against solar occultation data for earth @xcite and observations of transmission through earth s atmosphere during a lunar eclipse @xcite .
figure [ fig : ref_valid ] shows the result of this inter - comparison , where agreement ( in effective transit height , equal to @xmath126 ) is to within the atmospheric model grid spacing .
agreement improves if we use the coarser integration path lengths adopted by misra et al .
( i.e. , 5 km ) .
second , and finally , we applied our transit spectrum tool to a standardized hot jupiter - like atmospheric model widely used for inter - comparison purposes .
this model has a planetary radius ( at the 10 bar pressure level ) of 1.16 @xmath127 , a planetary mass of 1.14 @xmath128 , a stellar radius of 0.78 @xmath129 , and atmospheric volume mixing ratios of h@xmath130 , he , and h@xmath130o of 0.85 , 0.15 , and @xmath131 , respectively .
the 126 model layers are placed evenly in log - pressure between 10 bar and 10@xmath132 bar .
the atmosphere is isothermal at 1500 k. figure [ fig : geom_valid ] compares our model , in the geometric limit , to output from the chimera retrieval suite @xcite for this standard case .
agreement is within the 20 ppm scatter seen in comparisons between other @xcite transit spectrum tools ( m. r. line , personal communication ) .
we note that the offset in the rayleigh scattering slopes between our model and the chimera calculation is due to differing approaches to computing rayleigh scattering opacities which results in optical depths that differ at the 5% level or less .
we use the formalism outlined above to , first , compute the path distribution for earth - like and hot jupiter - like atmospheres for a range of conditions . following these instructive examples ,
we explore how scattering influences the transit depth associated with a single aerosol layer over a wide range of parameter space .
finally , we demonstrate the effects of thin , scattering clouds on a typical hot jupiter transit spectrum . the aforementioned standard earth and hot jupiter atmospheric models can be used to demonstrate the path distribution for a range of conditions and assumptions .
to show the effects of clouds on the path distribution for the hot jupiter atmosphere , we use the `` cloud '' and `` haze '' differential optical depth profiles shown in figure [ fig : demo_dtau ] .
the former is based on the shape of the aerosol mixing ratio profiles for a condensational cloud from the model , while the latter follows a power - law in altitude with scale height equal to the pressure scale height .
both models are taken to have integrated vertical optical depth unity , and to be conservatively and moderately forward scattering ( @xmath133 ) .
a henyey - greenstein phase function is assumed for cloudy and hazy simulations .
figure [ fig : demo_geom ] demonstrates the path distribution in the geometric limit for the standard hot jupiter - like model atmosphere .
altitudes are measured relative to the 10 bar radius ( at @xmath134 ) .
this graph is interpreted by selecting an impact parameter , and using the color shading to interpret the path distribution ( which , as stated earlier , can be thought of as an enhancement of the vertical optical depth ) . for example , taking an impact parameter `` altitude '' ( @xmath135 ) of 3000 km , the path distribution is zero for all atmospheric layers below this height , since rays with this impact parameter never pass through these deeper atmospheric layers .
the path distribution is then largest for layers near an altitude of 3000 km , as rays pass through these layers either horizontally or nearly horizontally . finally ,
for atmospheric layers well above 3000 km , rays pass through on much less extreme slant paths , so the path distribution is smaller ( typically 515 , for this particular model atmosphere ) .
the path distribution considering refraction for our cloud - free earth atmosphere is shown in figure [ fig : demo_refract ] . for comparison , a case in the geometric limit is also presented . for elements of the path distribution where the impact parameter altitude is near the surface ,
the path distribution deviates substantially from the geometric case due to the refractive bending of ray paths . as an example , take an impact parameter altitude of 5 km . in the geometric case , rays with this impact parameter would never reach altitudes below 5 km .
but , due to refraction , these rays are bent `` downward '' to probe deeper atmospheric layers .
thus , the path distribution for the refracting case is now non - zero for atmospheric layers below 5 km .
furthermore , the atmospheric layer that sees the most enhancement of optical depth ( i.e. , has the largest path distribution ) is now slightly below 5 km , instead of being right at 5 km as in the geometric case .
figures [ fig : demo_ray ] and [ fig : demo_clouds ] show an average path distribution from our multiply - scattering monte carlo approach , with , @xmath136 we note that such an average path distribution is not used in our calculation of transit spectra , but is simply meant to indicate the characteristic path that scattered photons take through the atmosphere . in the figure [ fig : demo_ray ] we only consider molecular rayleigh scattering opacity ( at 0.55 @xmath123 m wavelength ) in our hot jupiter - like model atmosphere . for figure
[ fig : demo_clouds ] , we show the effects of our nominal haze and cloud models for the hot jupiter - like case .
while both the haze and cloud have the same integrated optical depth , the haze path distribution has a less pronounced transition into the aerosol - affected atmospheric layers since the condensate cloud is concentrated into a more narrow range of altitudes / pressures . note the use of a logarithmic color contour scale to emphasize the impacts of scattering .
the path distributions for the scattering cases are less straightforward to interpret . for the rayleigh scattering case ( figure [ fig : demo_ray ] ) , and taking an impact parameter altitude of 800 km , we see that the path distribution is non - zero for altitudes below 800 km ( but would be zero in the geometric case ) . here
, a small fraction of photons have been scattered down to these deeper layers , and the small values of the path distribution at depth come at the ( small ) expense of the path distribution aloft . selecting a deeper impact parameter altitude of 200 km
, we see that the path distribution is now greatest at large altitudes ( where the path distribution would have been roughly 10 in the geometric case ) , and is small at altitudes below about 400600 km . here
, photons with a 200 km impact parameter altitude are scattered at altitudes larger than about 400600 km , so that relatively few of these photons can probe much deeper than this . the haze case in figure [ fig : demo_clouds ]
can be interpreted similarly to the rayleigh scattering case , although the range of impact parameters that are influenced by the haze is greater than that of the rayleigh scattering case .
the condensate cloud case in figure [ fig : demo_clouds ] appears distinct due to the well - defined cloud top . here
, rays / photons with impact parameter altitudes larger than 1400 km never encounter the cloud , and the path distribution is the same as in the geometric case . at altitudes just below this , though , the photons encounter the cloud , and are scattered before reaching atmospheric layers below the cloud ( whose base is at about 1200 km ) .
for these photons , their path distribution is concentrated at altitudes above the cloud and in the cloud itself . as most exoplanet transit spectrum models do not include multiple scattering ,
a key exercise is to explore the range of conditions for which scattering is expected to influence transit depth . for this exploration
, we note that the transit depth attributed to a narrow annulus on the planetary disk ( i.e. , at a single ray impact parameter ) scales with @xmath137 .
thus , in the absence of limb darkening , with the atmosphere transparent at radii larger than @xmath138 , and with the planetary disk centered on the star , only three parameters will influence the annulus transit depth for a scattering scenario the layer scattering optical depth , the aerosol scattering phase function , and the angular size of the stellar disk from the orbital location of the planet .
figure [ fig : tdepth_var ] show the relative difference between a transit spectrum model in the geometric and pure absorption limits ( where scattering optical depth is treated as absorption optical depth ) and a multiple scattering model for a single annulus on the planetary disk .
results are given for three different values of the host star angular size , which is determined by the ratio of the stellar radius to the planetary orbital distance ( @xmath139 ) .
these values span a host star angular size like that for mercury and the sun ( @xmath140 au , so that @xmath141 ) to that of a hot jupiter - like planet orbiting at 0.05 au from a sun - like star ( @xmath142 ) .
also , this range of host star angular sizes spans those for the habitable zones of m and k dwarf stars @xcite .
color contours in figure [ fig : tdepth_var ] are given as a function of layer scattering slant optical depth and the scattering asymmetry parameter .
no other opacity source is added to the layer . for simplicity ,
a henyey - greenstein phase function is assumed , as using a multi - parameter phase function would introduce additional variables to our phase space exploration .
regardless , these results will still serve to indicate under which conditions scattering can become important .
the standard hot jupiter - like atmospheric model outlined in section [ sec : valid ] provides a useful test case for exploring some important aspects of scattering in exoplanet transit spectra , especially when considering that planets near to their host stars are those for which scattering can have the largest effects .
the parameter space for such an exploration is extremely large , since the wavelength - dependent asymmetry parameter and scattering optical depth , as well as the cloud / haze vertical distribution , will all influence the transit spectrum .
exploring all of this phase space is certainly beyond the scope of this manuscript , so we adopt a straightforward set of conditions . specifically , we investigate the impact of general , isolated aerosol layers , placed at different pressure levels in the model atmosphere . rather than adopt a specific cloud or haze optical depth profile ( e.g. , figure [ fig : demo_dtau ] ) , we simply distribute the aerosol optical depth uniformly over a single pressure scale height . scattering optical depths and asymmetry parameters
are assumed gray , the single - scattering albedo is taken to be unity ( i.e. , pure scattering clouds ) , and , as before , a henyey - greenstein scattering phase function is adopted .
figure [ fig : tdepth_thincloud_g0.95 ] shows transit spectra for strongly forward scattering cloud / haze particles ( @xmath143 ) over the 12 @xmath123 m wavelength range , which overlaps the accessible wavelength range for _
figure [ fig : tdepth_thincloud_g0.90 ] is similar , but for less strongly forward scattering cloud / haze particles ( @xmath144 ) .
single - layer clouds were placed at pressures of 10@xmath145 , 10@xmath146 , 10@xmath147 , and 10@xmath148 bar , and spanned a vertical range of @xmath149 .
the clearsky case is shown in black , while pure absorbing and scattering cases are shown in gray and purple , respectively .
for each cloud pressure , three scenarios with different slant scattering optical depths are shown , with @xmath50 equal to 1 , 10 , and 100 .
this range of scattering optical depths spans the limits from a relatively optically thin aerosol layer to a relatively optically thick layer .
the path distribution approach provides a coherent and computationally efficient method to computing transit spectra . furthermore , analyzing the path distribution can provide an understanding of which atmospheric layers can contribute information to rays emerging from the planetary disk at a given impact parameter .
for example , figure [ fig : demo_refract ] demonstrates how refraction bends rays to probe deeper atmospheric layers than would be encountered in the geometric limit .
however , since the path distribution is intrinsic to just the planetary atmosphere , the other important effect of refraction bending rays such that they do not strike the stellar disk is not represented .
this is a strength of the path distribution approach , since only a single distribution would need to be computed for identical atmospheric models for planets orbiting different host stars .
similarly , figure [ fig : demo_ray ] shows how , on average , rayleigh scattering will scatter a small fraction of photons at large impact parameters `` downward '' to probe deeper parts of the atmosphere .
conversely , rays at small impact parameters no longer probe the deep atmosphere , as they are scattered at larger altitudes .
of course , as rayleigh scattering is not strongly forward scattering , it is unlikely that any significant fraction of the scattered photons will still intersect the stellar disk .
figure [ fig : demo_clouds ] shows similar scattering effects to those in figure [ fig : demo_ray ] .
however , the impact parameter where ( roughly ) deeper atmospheric layers become ineffectively probed depends on the vertical structure of the aerosols and , more specifically , where the slant scattering optical depth unity occurs .
we note , again , that these figures display average path distributions for our monte carlo simulations , and that transit spectra that include scattering must be computed using using the formalism developed in section [ section : scattering ]
. ultimately , when considering scattering , the brightness of any given annulus ( at impact parameter @xmath1 ) on the planetary disk will depend on the transparency of atmospheric layers at radii larger than @xmath1 , the slant scattering optical depth of the layer at @xmath1 , the scattering phase function for the aerosols in this layer , and the angular size of the host star as seen from the planet .
for the situation where overlaying layers are transparent ( implying that we can , thus , see down to a cloud ) , figure [ fig : tdepth_var ] provides a rough guide to the circumstances where scattering can be important . here , one can use simple details from a forward model the angular size of the host star , the scattering optical thickness of a cloud layer , and cloud scattering asymmetry parameter to determine if scattering can safely be ignored in a simulation . in the case where the host star angular size is much smaller than would be at a distance of roughly 0.4 au from a sun - like star ( i.e. , @xmath141 )
, figure [ fig : tdepth_var ] shows that essentially any amount of scattering will prevent the path of a ray from tracing back to the stellar disk ( which is a very small `` target '' ) . with @xmath150 at roughly 15@xmath151 , which , for example , is appropriate for the habitable zones of late m dwarf stars , strongly forward scattering aerosols ( @xmath152 ) can have significant impacts on the transit depth due to an annulus for clouds whose slant optical depths are not very optically thick ( i.e. , @xmath153 less than 10 ) .
this effect becomes quite dramatic for hot jupiter - like conditions , where the host star is a relatively large `` target . '' here , even modestly forward scattering aerosols ( @xmath154 ) can cause substantial variations as compared to the commonly - used pure absorption assumption .
this dependence on the asymmetry parameter explains why both @xcite and @xcite found that multiple scattering was unimportant for their presented hot jupiter transit spectra .
as these authors considered only either rayleigh scattering or isotropic phase functions ( which both have @xmath155 ) , scattering could effectively be treated as absorption .
our adoption of the henyey - greenstein phase function is driven by the computational simplicity of this parameterization , and also the need to capture the process of forward scattering while minimizing the number of added parameters to our simulations .
nevertheless , is has been shown that the henyey - greenstein phase function , when compared to mie calculations , can underestimate the power in the forward scattering peak @xcite , especially at wavelengths comparable to the particle size .
thus , depending on the specifics of a given aerosol in an exoplanetary atmosphere , the results shown in figure [ fig : tdepth_var ] may underestimate the increased transmission due to forward scattered light at a given asymmetry parameter ( as compared to the pure absorption limit )
. figures [ fig : tdepth_thincloud_g0.95 ] and [ fig : tdepth_thincloud_g0.90 ] further explore the importance of scattering in hot jupiter transit spectra . for high - altitude clouds / hazes ,
these results show that differences between a pure absorption model versus a scattering model can approach ( or exceed ) 200 ppm , which is larger than the typical uncertainties achieved in current observations with _
@xcite or that are expected for _ jwst _ @xcite .
very optically thick clouds ( i.e. , with slant scattering optical depths approaching 100 ) approach the pure absorption limit , and the difference between a scattering model and a pure absorption model for a thin cloud ( i.e. , scattering optical depth less than about unity ) is small since the overall impact of the cloud on the transit spectrum is weak .
similarly , the impact of scattering for a deep cloud or haze layer is small , as few wavelength regions are sensitive to the presence of this aerosol layer . in general
, the impact of aerosol multiple scattering will depend on cloud configurations and optical properties , and this manuscript does not undertake the complex microphysical calculations needed to attempt predictions of cloud / haze distributions and compositions in exoplanet atmospheres our goal is to highlight the conditions where scattering becomes an important consideration .
nevertheless , the results in figures [ fig : tdepth_thincloud_g0.95 ] and [ fig : tdepth_thincloud_g0.90 ] indicate that forward scattering exoplanet clouds analogous to earth s cirrus clouds ( which are at higher altitudes and have optical depths less than roughly unity ) may warrant a multiple scattering approach .
additionally , any forward scattering cloud with a scale height comparable to ( or greater than ) the pressure scale height could require a multiple scattering model to properly simulate a transit spectrum , as transit depth variations due to multiple scattering effects in this clouds will be comparable to those of molecular features ( whose scale are typically set by the pressure scale height ) . differences between the scattering and pure absorption models for the hot jupiter cases will also depend on the asymmetry parameter and single - scattering albedo . for very strongly forward scattering aerosols ( i.e. , asymmetry parameters larger than our adopted value of 0.95 )
, the transit spectrum will approach the clearsky limit ( if the aerosols are weakly- or non - absorbing ) .
in essence , such clouds would be invisible , having no effect on the transit depth .
on the other hand , asymmetry parameters much smaller than roughly 0.80.9 will push the scattering transit spectra towards the pure absorption limit , since these scattered photons are increasingly unlikely to connect to the observer . similarly , as the single - scattering albedo is decreased ( away from our adopted value of unity ) , the scattering spectra will approach the pure absorption case .
the extent of the influence of the single - scattering albedo depends on the average number of scatterings the photons experience along their path .
for example , for our @xmath156 case , the photons are scattered of order ten times , implying a single - scattering albedo of less than about 0.9 would reduce the transmission to roughly 50% .
perhaps the most important message from the scattering studies above is the influence of the scattering asymmetry parameter on transit spectra of close - in exoplanets .
aerosols whose scattering properties vary from strongly forward scattering to weakly forward scattering could impart features in a transit spectrum , as the cloud ( or haze layer ) will be more transparent where the particles have a larger asymmetry parameter .
such signatures would appear along with absorption bands caused by particle vibrational modes and cloud base features @xcite .
additionally , as transit spectra of forward scattering clouds would be reproduced by relatively thinner clouds in the pure absorption limit , transit spectral retrievals using pure absorption models could return cloud thickness ( or number densities ) that are biased low , where retrieved cloud optical depths would , instead , represent an effective optical depth that incorporates information about the optical properties of the cloud particles and the stellar angular size . whether or not aerosol multiple scattering will be important for any given exoplanet transit spectrum will depend on a number of parameters , including the slant optical thickness of any haze / cloud structure and the scattering asymmetry parameter ( which will both , in general , depend on wavelength ) .
the planet must also be located near to its host star , which limits the range of relevant aerosols to those which form in warm and hot atmospheric conditions .
for example , asymmetry parameters for water droplets in the visible wavelength range are 0.80.9 @xcite , and span 0.740.94 for ice crystals ( where crystals with plate - like morphologies are the most strongly forward scattering ; * ? ? ?
thus , conditions may be appropriate for multiple scattering to influence the transit spectra of potentially habitable worlds around the coolest stars ( e.g. , trappist-1 * ? ? ?
looking beyond earth in the solar system , ( their figure 1 ) show that a variety of ices , dusts , droplets , and fractal hazes are both non - absorptive and forward scattering ( with @xmath157 ) below 23 @xmath123 m .
the wavelength range for these solar system cases is especially relevant to transiting exoplanets around later - type stars that may be investigated by _ hubble _ ,
_ jwst _ , the fast infrared exoplanet spectroscopy survey explorer concept ( finesse ; * ? ? ?
* ) , and/or the atmospheric remote - sensing infrared exoplanet large - survey concept ( ariel ; * ? ? ?
the composition , size distributions , and optical properties of aerosols in warm and hot exoplanet atmospheres are largely unknown . for the hottest exoplanets , metal and silicate condensates may be expected to form @xcite .
@xcite investigated the optical properties of a large variety of metallic and silicon - bearing aerosols , and showed that many of these species can be strongly forward scattering over certain wavelength ranges .
especially relevant cases are aluminum oxide , forsterite , enstatite , and pyroxine .
it is interesting to note that our results show that multiple scattering can be an important consideration for exoplanet transits when the angular size of the host star is relatively large , which is the opposite regime from where refraction has been shown to be important . for refraction
, a host star with small angular size implies that relatively little refractive bending is required to deflect a ray off the stellar disk , thereby setting a floor in the transit spectrum which is not captured in the commonly - used geometric , pure absorption limit . by comparison , only a single scattering ( with even a strong forward peak )
is required to deflect a ray from the stellar disk , which is then in agreement with the geometric , pure absorption limit . of course , the opposite of these statements is true for a star with large angular size . in future work
we plan to further explore the importance of refraction in transit observations of gas giants .
finally , since this manuscript primarily emphasizes the development of a light scattering theory for exoplanet transits and the impacts of vertically thin cloud layers , a number of future studies will be undertaken .
these will include scattering effects in vertically more - extended cloud structures , and could also investigate the role of the assumed ( or computed ) particle scattering phase function .
finally , the formalism outlined above can be used to study light scattering effects in time - resolved transit lightcurves .
we have detailed a new theory of exoplanet transit spectroscopy that includes the effects of light multiple scattering . by effectively separating the path that photons take through an exoplanet atmosphere from the absorption processes of that atmosphere , the technique discussed in this manuscript yields models that are both physically rigorous and computationally efficient .
this approach , which relies on the so - called `` path distribution '' defined herein , can be extended to other areas of study , including ( most straightforwardly ) stellar occultations by planetary atmospheres .
when applying our validated scattering model to isolated cloud layers , models show that multiple scattering is most important for cases where the exoplanet host star is large in angular size as seen from the world ( i.e. , when the orbital distance is less than 1020 times the stellar radius ) . in these cases ,
multiple scattering by aerosols with asymmetry parameters larger than 0.80.9 can have substantial effects on the transmission of the cloud layer .
for all cases , differences between a multiple scattering model and a geometric pure absorption diminish for clouds ( or hazes ) with slant scattering optical depths approaching 100 . in an exploratory case of a conservatively and forward ( @xmath1580.900.95 ) scattering cloud / haze layer in the atmosphere of a hot jupiter , differences in the transit depth from a multiple scattering model and a model in the pure absorption limit can exceed 200 ppm .
these differences are most pronounced when the cloud / haze is well above the level of gas absorption optical depth unity , and when the slant scattering optical depth is of order several to several tens
. future work will further explore the situations where scattering in more extended exoplanet cloud ( or haze ) layers is key to understanding transit observations .
tr gratefully acknowledges support from nasa through the sagan fellowship program executed by the nasa exoplanet science institute .
the results reported herein benefitted from collaborations and/or information exchange within nasa s nexus for exoplanet system science ( nexss ) research coordination network sponsored by nasa s science mission directorate .
certain essential tools used in this work were developed by the nasa astrobiology institute s virtual planetary laboratory , supported by nasa under cooperative agreement no .
the author thanks m. line and j. lustig - yaeger for sharing model outputs for inter - comparison purposes , and j. fortney and r. de kok for providing helpful feedback .
this manuscript is dedicated to the memory of prince , who passed away during the completion of this project``thank you for a funky time . ''
given the ray atmospheric path distribution as defined in section [ sec : theory ] , a number of key quantities can be derived .
first , the pathlength ( in km , for example ) traversed between any two altitudes , @xmath159 and @xmath160 , by a ray incident with impact parameter @xmath1 is simply , @xmath161 where the absolute value forces all distances to be measured non - negative .
then , if @xmath162 at the bottom of a terrestrial planetary atmosphere or at some reference pressure ( e.g. , 10 bar ) for a gaseous world , and @xmath163 is the effective top of the atmosphere , the total linear distance traverse through the entire atmosphere by the ray is , @xmath164 where we note that this integral would diverge if taken to infinite altitude .
the pathlength integral can also be written in terms of two pressure coordinates , @xmath165 and @xmath166 , assuming the ideal gas law and hydrostatic equilibrium , @xmath167 where @xmath168 is the universal gas constant , @xmath169 is the atmospheric temperature profile , @xmath170 is the acceleration due to gravity , @xmath37 is the atmospheric mean molecular weight , and @xmath171 is the pressure scale height . also , given the definition of the path distribution , and the differential definition of ( path ) column mass from the atmospheric mass density profile [ @xmath172 , @xmath173 the air mass encountered by a ray with impact parameter @xmath1 between two altitudes , @xmath159 and @xmath160 , is given by , @xmath174 similarly , a column ( path ) number density is , @xmath175 where @xmath176 is the number density profile .
in the geometric limit , rays pass straight through the planetary atmosphere . here , the path distribution can be determined analytically using geometric arguments . for an atmospheric layer centered at @xmath2 with width @xmath177 , and for a ray incident on the atmosphere with impact parameter @xmath1 ,
the path distribution is given by @xmath178 \ , & b \leq r - \delta h/2 \ . \\
\end{cases } \label{eqn : pathgeom}\ ] ] where we have defined @xmath179 for conciseness . note that this expression is only independent of @xmath177 in the limit that this value is small when compared to @xmath102 .
performing a first - order expansion in @xmath177 yields , @xmath180 which is in agreement with the linearized geometry discussed in @xcite ( their figure 1 ) .
the geometric path distribution need only be computed once for a model atmosphere .
thus , when paired with equation [ eqn : transit_vec ] , the geometric approach can be executed with great computational efficiency .
when considering refraction , rays no longer travel on straight - line paths through the atmosphere . in this case , the path distribution must be computed numerically . fortunately , as the refractive indexes for gases likely to be major atmospheric constituents vary weakly in wavelength , the path distribution need only be computed at a small number of spectral points when generating a transit spectrum .
refraction will deflect the trajectory of a ray upward or downward , depending on the sign of the local atmospheric refractive index profile . here
, the local curvature experienced by the ray , @xmath75 , is , @xmath181 where @xmath182 is the zenith angle for the ray trajectory , and @xmath183 is the atmospheric refractive index .
the zenith angle is determined via , @xmath184 using pathlength as the integration variable , and according to the geometry shown in figure [ fig : refract_geom ] , we have , @xmath185 @xmath186 \ , \label{eqn : dtheta_ds}\ ] ] @xmath187 and @xmath188 where @xmath70 is the polar angle measured from the @xmath189 direction , and @xmath190 is the so - called refraction integral ( which measures the deflection from the initial straight - line trajectory ) . for a ray with initial impact parameter @xmath21 , the path is determined by numerically integrating the expressions above using an algorithm like that described in @xcite . while the ray is passing through the atmospheric layer at @xmath49 , the increments of @xmath74 used in the path integration are divided by @xmath191 and added to @xmath17 . the path integration proceeds through all atmospheric layers until either the ray exits the atmosphere or the planetary surface is struck . when the ray exits the atmosphere , the radial coordinate and trajectory are stored for later use in determining which rays map back to the stellar disk . the straight - line trajectory
the ray follows after exiting the atmosphere is defined by the total refraction angle at exit , which is equal to the direction cosine of the ray along the x - axis in figure [ fig : refract_geom ] ( i.e. , @xmath192 ) . natexlab#1#1 , a. s. , & marley , m. s. 2001 , , 556 , 872 , j. k. , aigrain , s. , irwin , p. g. j. , et al .
2012 , arxiv e - prints , arxiv:1212.5020 , j. k. , aigrain , s. , irwin , p. g. j. , kendrew , s. , & fletcher , l. n. 2016 , , 458 , 2657 , n. , kalirai , j. , lunine , j. , clampin , m. , & lindler , d. 2015 , arxiv e - prints , arxiv:1507.02655 bean , j. l. , kempton , e. m .-
r . , & homeier , d. 2010 , nature , 468 , 669 beichman , c. , benneke , b. , knutson , h. , et al .
2014 , publications of the astronomical society of the pacific , 126 , 1134 benneke , b. , & seager , s. 2012 , the astrophysical journal , 753 , 100 , y. 2016 , , 456 , 4051 , y. , & kaltenegger , l. 2014 , , 791 , 7 btrmieux , y. , & kaltenegger , l. 2015 , monthly notices of the royal astronomical society , 451 , 1268 , o. 1998 , journal of atmospheric sciences , 55 , 128 brown , t. m. 2001 , the astrophysical journal , 553 , 1006 , j. , kocifaj , m. , salmeron , r. , & hubeny , i. 2015 , , 454 , 2 , a. , hubbard , w. b. , lunine , j. i. , & liebert , j. 2001 , reviews of modern physics , 73 , 719 , d. , brown , t. m. , noyes , r. w. , & gilliland , r. l. 2002 , , 568 , 377 , n. b. , greene , t. , angerhausen , d. , et al .
2015 , , 127 , 311 , p. a. , muirhead , p. s. , fortney , j. j. , et al .
2015 , , 814 , 154 de kok , r. , & stam , d. 2012 , icarus , 221 , 517 , d. , seager , s. , winn , j. , et al .
2009 , , 121 , 952 , p. , swain , m. r. , & green , r. o. 2012 , in , vol . 8442 , space telescopes and instrumentation 2012 : optical , infrared , and millimeter wave , 844241 fortney , j. j. 2005 , monthly notices of the royal astronomical society , 364 , 649 , j. , deming , d. , benneke , b. , et al .
2014 , , 513 , 526 , j. p. , mather , j. c. , clampin , m. , et al .
2006 , , 123 , 485 , m. , jehin , e. , lederer , s. m. , et al .
2016 , , 533 , 221 greene , t. p. , line , m. r. , montero , c. , et al .
2016 , the astrophysical journal , 817 , 17 gunson , m. r. , abbas , m. , abrams , m. , et al .
1996 , geophysical research letters , 23 , 2333 , c. , oevermann , m. , lttke , m. j. h. , klein , r. , & sedlmayr , e. 2001 , , 376 , 194 , l. g. , & greenstein , j. l. 1941 , , 93 , 70 hubbard , w. , fortney , j. , lunine , j. , et al .
2001 , the astrophysical journal , 560 , 413 hui , l. , & seager , s. 2002 , the astrophysical journal , 572 , 540 , f. w. , gunson , m. r. , toon , g. c. , et al .
2002 , , 41 , 6968 , p.
g. j. , teanby , n. a. , de kok , r. , et al .
2008 , , 109 , 1136 kaltenegger , l. , & traub , w. 2009 , the astrophysical journal , 698 , 519 , r. a. , mackenty , j. w. , oconnell , r. w. , & townsend , j. a. 2008 , in , vol . 7010 , space telescopes and instrumentation 2008 : optical , infrared , and millimeter , 70101e , d. m. , & tinetti , g. 2010 , , 407 , 2589 , h. a. , benneke , b. , deming , d. , & homeier , d. 2014 , , 505 , 66 , h. a. , dragomir , d. , kreidberg , l. , et al .
2014 , , 794 , 155 , a. 2004 , earth science reviews , 64 , 189 , r. k. , ramirez , r. , kasting , j. f. , et al .
2013 , , 765 , 131 , l. 2015 , , 127 , 1161 kreidberg , l. , bean , j. l. , dsert , j
, et al . 2014 , nature , 505 , 69 , g. , dobbs - dixon , i. , helling , c. , bognar , k. , & woitke , p. 2016
, , 594 , a48 , j .- m . , irwin , p. g. j. , fletcher , l. n. , heng , k. , & barstow , j. k. 2014 , , 789 , 14 , m. r. , knutson , h. , deming , d. , wilkins , a. , & desert , j .- m .
2013 , , 778 , 183 , m. r. , zhang , x. , vasisht , g. , et al .
2012 , , 749 , 93 line , m. r. , wolf , a. s. , zhang , x. , et al .
2013 , the astrophysical journal , 775 , 137 , j. i. , hubbard , w. b. , burrows , a. , wang , y .- p . , & garlow , k. 1989 , , 338 , 314 , a. , francis , p. n. , mcfarquhar , g. m. , & kinne , s. 1998 , journal of atmospheric sciences , 55 , 2874 , k. , & agol , e. 2002 , , 580 , l171 , m. s. , ackerman , a. s. , cuzzi , j. n. , & kitzmann , d. 2013 , in comparative climatology of terrestrial planets , ed . s. j. mackwell , a. a. simon - miller , j. w. harder , & m. a. bullock ( tucson : university of arizona press ) , 367391 , m. s. , gelino , c. , stephens , d. , lunine , j. i. , & freedman , r. 1999 , , 513 , 879 , r. a. , fenn , r. w. , selby , j. e. a. , volz , f. e. , & garing , j. s. 1972 , optical properties of the atmosphere ( third edition ) , tech .
rep . , air force cambridge research labs , v. s. , & crisp , d. 1996 , , 101 , 4595 , a. , meadows , v. , & crisp , d. 2014 , , 792 , 61 morley , c. v. , fortney , j. j. , kempton , e. m .-
, et al . 2013 , the astrophysical journal , 775 , 33 , c. v. , knutson , h. , line , m. , et al . 2016 , arxiv e - prints , arxiv:1610.07632 muoz , a. g. , osorio , m. z. , barrena , r. , et al .
2012 , the astrophysical journal , 755 , 103 pall , e. , osorio , m. r. z. , barrena , r. , montas - rodrguez , p. , & martn , e. l. 2009 , nature , 459 , 814 pont , f. , knutson , h. , gilliland , r. , moutou , c. , & charbonneau , d. 2008 , monthly notices of the royal astronomical society , 385 , 109 seager , s. , & sasselov , d. 2000 , the astrophysical journal , 537 , 916 , o. , & sari , r. 2010 , the astrophysical journal , 720 , 904 sing , d. , dsert , j .- m . , lecavelier des etangs , a. , et al .
2009 , astronomy and astrophysics , 505 , 891 , d. k. , fortney , j. j. , nikolov , n. , et al .
2016 , , 529 , 59 stamnes , k. , tsay , s .- c . , jayaweera , k. , wiscombe , w. , et al .
1988 , applied optics , 27 , 2502 , k. b. 2016 , , 817 , l16 , m. r. , vasisht , g. , & tinetti , g. 2008 , , 452 , 329 , g. , drossart , p. , eccleston , p. , et al .
2016 , in , vol .
9904 , society of photo - optical instrumentation engineers ( spie ) conference series , 99041x tomasko , m. , doose , l. , engel , s. , et al .
2008 , planetary and space science , 56 , 669 , d. 1996 , applied optics , 35 , 3270 , s. , cuzzi , j. n. , marley , m. , & fortney , j. 2014 , , 789 , l11 van der werf , s. 2008 , applied optics , 47 , 153 , h. r. , & sing , d. k. 2015 , , 573 , a122 , i. p. , tinetti , g. , rocchetto , m. , et al .
2015 , , 802 , 107 , r. , crisp , d. , & chen , l. 1990 , , 43 , 191 , a. n. 1977 , , 35 , 1 | exoplanet transit spectroscopy enables the characterization of distant worlds , and will yield key results for nasa s _ james webb space telescope_. however , transit spectra models are often simplified , omitting potentially important processes like refraction and multiple scattering .
while the former process has seen recent development , the effects of light multiple scattering on exoplanet transit spectra has received little attention . here , we develop a detailed theory of exoplanet transit spectroscopy that extends to the full refracting and multiple scattering case .
we explore the importance of scattering for planet - wide cloud layers , where the relevant parameters are the slant scattering optical depth , the scattering asymmetry parameter , and the angular size of the host star .
the latter determines the size of the `` target '' for a photon that is back - mapped from an observer .
we provide results that straightforwardly indicate the potential importance of multiple scattering for transit spectra .
when the orbital distance is smaller than 1020 times the stellar radius , multiple scattering effects for aerosols with asymmetry parameters larger than 0.80.9 can become significant .
we provide examples of the impacts of cloud / haze multiple scattering on transit spectra of a hot jupiter - like exoplanet . for cases with a forward and conservatively scattering cloud / haze ,
differences due to multiple scattering effects can exceed 200 ppm , but shrink to zero at wavelength ranges corresponding to strong gas absorption or when the slant optical depth of the cloud exceeds several tens .
we conclude with a discussion of types of aerosols for which multiple scattering in transit spectra may be important . |
the search for resonant structures in the excitation functions for various combinations of light @xmath1-cluster ( @xmath2=@xmath3 ) nuclei in the energy regime from the coulomb barrier up to regions with excitation energies of @xmath4=20@xmath550 mev remains a subject of contemporary debate @xcite .
these resonances have been interpreted in terms of nuclear molecules @xcite .
the question of how quasimolecular resonances may reflect continuous transitions from scattering states in the ion - ion potential to true cluster states in the compound systems is still unresolved @xcite . in many cases ,
these resonant structures have been associated with strongly - deformed shapes and with alpha - clustering phenomena @xcite , predicted from the nilsson - strutinsky approach , the cranked @xmath1-cluster model @xcite , or other mean - field calculations @xcite . in light
@xmath1-like nuclei clustering is observed as a general phenomenon at high excitation energy close to the @xmath1-decay thresholds @xcite .
this exotic behavior has been perfectly illustrated by the famous `` ikeda''-diagram for @xmath2=@xmath3 nuclei in 1968 @xcite , which has been recently extended by von oertzen @xcite for neutron - rich nuclei , as shown in the left panel of fig.1 .
clustering is a general phenomenon @xcite not only observed in light neutron - rich nuclei @xcite , but also in halo nuclei such as @xmath6li @xcite or @xmath7be , for instance @xcite . the problem of cluster formation has also been treated extensively for very heavy systems by r.g .
gupta @xcite and by zagrebaev and w. greiner @xcite where giant molecules and the collinear ternary fission may exist @xcite .
the real link between superdeformation ( sd ) , nuclear molecules and alpha clustering @xcite is of particular interest , since nuclear shapes with major - to - minor axis ratios of 2:1 have the typical ellipsoidal elongation for light nuclei ( with quadrupole deformation parameter @xmath8 @xmath9 0.6 ) .
furthermore , the structure of possible octupole - unstable 3:1 nuclear shapes ( hyperdeformation ( hd ) with @xmath8 @xmath9 1.0 ) has also been discussed for actinide nuclei @xcite in terms of clustering phenomena .
typical examples for possible relationship between quasimolecular bands and extremely deformed ( sd / hd ) shapes have been widely discussed in the literature for @xmath10 @xmath1-conjugate @xmath2=@xmath3 nuclei , such as @xmath11si @xcite , @xmath12s @xcite , @xmath13ar @xcite , @xmath14ca @xcite , @xmath15ti @xcite , @xmath16cr @xcite and @xmath17ni @xcite .
in fact , highly deformed shapes and sd rotational bands have been discovered in several light @xmath1-conjugate ( @xmath2=@xmath3 ) nuclei , such as @xmath13ar and @xmath14ca by using @xmath18-ray spectroscopy techniques @xcite . in particular , the extremely deformed rotational bands in @xmath13ar @xcite ( shown as crosses in fig .
2 ) might be comparable in shape to the quasimolecular bands observed in both @xmath0c+@xmath19 mg @xcite ( shown as open triangles in fig . 2 ) and @xmath20o+@xmath21ne @xcite ( shown as full rectangles ) reactions .
ternary clusterizations are also predicted theoretically , but were not found experimentally in @xmath13ar so far @xcite . on the other hand , ternary fission of @xmath17ni related to its hyperdeformed shapes was identified from out - of - plane angular correlations measured in the @xmath12s+@xmath19 mg reaction with the binary reaction spectrometer ( brs ) at the vivitron tandem facility of the iphc , strasbourg @xcite .
this finding @xcite is not limited to light @xmath2=@xmath3 compound nuclei , true ternary fission @xcite can also occur for very heavy @xcite and superheavy @xcite nuclei .
new partial ( high - energy ) level scheme of @xmath20o corresponding to @xmath18-ray transitions observed in the @xmath0c(@xmath19mg,@xmath21ne)@xmath20o@xmath22 @xmath1-transfer reactions .
this figure has been adapted from ref .
@xcite . ] new partial ( high - energy ) level scheme of @xmath20o corresponding to @xmath18-ray transitions observed in the @xmath0c(@xmath19mg,@xmath21ne)@xmath20o@xmath22 @xmath1-transfer reactions .
this figure has been adapted from ref .
@xcite . ]
there is a renewed interest in the spectroscopy of the @xmath20o nucleus at high excitation energy @xcite .
exclusive data were collected on @xmath20o in the inverse kinematics reaction @xmath19mg@xmath23c studied at e@xmath24(@xmath19 mg ) = 130 mev with the brs in coincidence with the euroball iv installed at the vivitron facility @xcite . from the @xmath1-transfer reactions ( both direct transfer and deep - inelastic orbiting collisions @xcite ) , new information has been deduced on branching ratios of the decay of the 3@xmath25 state of @xmath20o at 11.085 mev @xmath26 3 kev .
the high - energy level scheme of @xmath20o shown in fig .
3 indicates that this state does not @xmath1-decay because of its non - natural parity ( in contrast to the two neighbouring 4@xmath25 states at 10.36 mev and 11.10 mev ) , but it @xmath18 decays to the 2@xmath25 state at 6.92 mev ( 54.6 @xmath26 2 @xmath27 ) and to the 3@xmath28 state at 6.13 mev ( 45.4% ) . by considering all the four possible transition types of the decay of the 3@xmath25 state ( _ i.e. _ e@xmath29 and m@xmath30 for the 3@xmath25 @xmath31 3@xmath32 transition and , m@xmath29 and e@xmath30 for the 3@xmath25 @xmath31 2@xmath25 transition ) , our calculations yield the conclusion that @xmath33 ev , a value fifty times lower than known previously , which is an important result for the well studied @xmath20o nucleus @xcite .
clustering effects in the light neutron - rich oxygen isotopes @xmath34o will be discussed in section 3 . in the study of the bose - einstein condensation ( bec ) the @xmath1-particle states in light @xmath2=@xmath3
nuclei @xcite , are of great importance . at present , the search for an experimental signature of bec in @xmath20o is of highest priority .
a state with the structure of the hoyle `` state @xcite in @xmath0c coupled to an @xmath1 particle is predicted in @xmath20o at about 15.1 mev ( the 0@xmath35 state ) , the energy of which is @xmath9 700 kev above the 4@xmath1-particle breakup threshold @xcite . however ,
any state in @xmath20o equivalent to the so - called
hoyle '' state @xcite in @xmath0c is most certainly going to decay by particle emission with very small , probably un - measurable , @xmath18-decay branches , thus , very efficient particle - detection techniques will have to be used in the near future to search for them .
bec states are expected to decay by alpha emission to the hoyle " state and could be found among the resonances in @xmath1-particle inelastic scattering on @xmath0c decaying to that state ( an early attempt to excite these states by @xmath1 inelastic scattering was presented in ref .
@xcite ) , or could be observed in an @xmath1-particle transfer channel leading to the @xmath36be@xmath36be final state .
another possibility might be to perform coulomb excitation measurements with intense @xmath20o beams at intermediate energies .
overview of four rotational band structures observed in @xmath21o .
excitation energy systematics for the members of the rotational bands forming inversion doublets with k=0 are plotted as a function of j(j+1 ) .
the curves are drawn to guide the eye for the slopes .
the indicated slope parameters contain information on the moments of inertia .
square and triangle symbols correspond to cluster bands , whereas diamonds symbols correspond to molecular bands.this figure is adapted from @xcite ] overview of four rotational band structures observed in @xmath21o .
excitation energy systematics for the members of the rotational bands forming inversion doublets with k=0 are plotted as a function of j(j+1 ) .
the curves are drawn to guide the eye for the slopes .
the indicated slope parameters contain information on the moments of inertia .
square and triangle symbols correspond to cluster bands , whereas diamonds symbols correspond to molecular bands.this figure is adapted from @xcite ]
as discussed previously , clustering is a general phenomenon observed also in nuclei with extra neutrons as it is presented in an extended `` ikeda''-diagram @xcite proposed by von oertzen @xcite ( see the left panel of fig . 1 ) . with additional neutrons ,
specific molecular structures appear with binding effects based on covalent molecular neutron orbitals . in these diagrams
@xmath1-clusters and @xmath20o - clusters ( as shown by the middle panel of the diagram of fig . 1 )
are the main ingredients .
actually , the @xmath7c nucleus may play similar role in clusterization as the @xmath20o one since it has similar properties as a cluster : i ) it has closed neutron p - shells , ii ) first excited states are well above e@xmath22 = 6 mev , and iii ) it has high binding energies for @xmath1-particles .
a general picture of clustering and molecular configurations in light nuclei can be drawn from a detailed investigation of the light oxygen isotopes with a @xmath37 17 . here
we will only present recent results on the even - even oxygen isotopes : @xmath38o @xcite and @xmath21o @xcite .
but very striking cluster states have also been found in odd - even oxygen isotopes such as : @xmath39o @xcite and @xmath40o @xcite .
4 gives an overview of all bands in @xmath38o as a plot of excitation energies as a function of j(j+1 ) together with their respective moments of inertia . in the assignment of the bands both
the dependence of excitation energies on j(j+1 ) and the dependence of measured cross sections on 2j+1 @xcite were considered .
slope parameters obtained in a linear fit to the excitation energies @xcite indicate the moment of inertia of the rotational bands given in fig .
the intrinsic structure of the cluster bands is reflection asymmetric , the parity projection gives an energy splitting between the partner bands .
the assignment of the experimental molecular bands are supported by either generator - coordinate - method @xcite or antisymmetrized molecular dynamics ( amd ) calculations @xcite
. we can compare the bands of @xmath21o @xcite shown in fig . 5 with those of @xmath38o displayed in fig . 4 .
the first doublet ( k=0@xmath41 ) has a slightly larger moment of inertia ( smaller slope parameter ) in @xmath21o , which is consistent with its interpretation as @xmath7c@xmath42he or @xmath20c@xmath43he molecular structures ( they start well below the thresholds of 16.8 mev and 12.32 mev , respectively ) . the second band , for which the negative parity partner is yet to be determined , has a slope parameter slightly smaller than in @xmath38o .
this is consistent with the study of the bands in @xmath21o by furutachi et al .
@xcite , which clearly establishes parity inversion doublets predicted by amd calculations for the @xmath7c@xmath44he cluster and @xmath7c-2n-@xmath1 molecular structures .
the corresponding moments of inertia given in fig . 4 and fig .
5 are strongly suggesting large deformations for the cluster structures .
we may conclude that the reduction of the moments of inertia of the lowest bands of @xmath45o is consistent with the assumption that the strongly bound @xmath7c nucleus having equivalent properties to @xmath20o , has a similar role as @xmath20o in relevant , less neutron rich nuclei .
therefore , the ikeda - diagram @xcite and the `` extended ikeda - diagram '' consisting of @xmath20o cluster cores with covalently bound neutrons @xcite must be further extended to include also the @xmath7c cluster cores as illustrated in fig . 1 .
the connection of @xmath1-clustering , quasimolecular resonances , orbiting phenomena and extreme deformations ( sd , hd , ... ) has been discussed in this work by using @xmath18-ray spectroscopy of coincident binary fragments from either inelastic excitations and direct transfers ( with small energy damping and spin transfer ) or from orbiting ( fully damped ) processes @xcite in the @xmath19mg+@xmath0c reaction . from a careful analysis of the @xmath20o+@xmath21ne @xmath1-transfer exit - channel
( strongly populated by orbiting ) new information has been deduced on branching ratios of the decay of the 3@xmath25 state of @xmath20o at 11.089 mev .
this result is encouraging for a complete @xmath18-ray spectroscopy of the @xmath20o nucleus at high excitation energy .
in addition , we have presented new results on neutron - rich oxygen isotopes displaying very well defined molecular bands in agreement with amd predictions .
consequently , the extended ikeda diagram has been further extended for light neutron - rich nuclei by inclusion of the @xmath7c cluster , similarly to the @xmath20o one .
of particular interest is the quest for the 4@xmath1 states of @xmath20o near the @xmath36be+@xmath36be and @xmath0c+@xmath1 decay thresholds , which correspond to the so - called hoyle `` state . the search for extremely elongated configurations ( hd ) in rapidly rotating medium - mass nuclei , which has been pursued by @xmath18-ray spectroscopy measurements , will have to be performed in conjunction with charged - particle techniques in the near future since such states are most certainly going to decay by particle emission ( see @xcite ) .
+ we thank the vivitron staff and the euroball group of strasbourg for their excellent support .
c.b . would like to acknowledge christian caron ( springer ) for initiating the new series of three volumes of _ lecture notes in physics _ dedicated to ' ' clusters in nuclei " .
99 greiner w , park y jae , and scheid w 1995 _ nuclear molecules _ , ( singapore : world scientific ) beck c _ et al .
_ 1994 _ phys .
c _ * 49 * 2618 freer m 2007 _ rep .
phys . _ * 70 * 2149 horiuchi h 2010 clusters in nuclei - vol.1 , ed .
beck c , _ lecture notes in physics _ * 818 * 57 ; horiuchi h , this conference .
gupta r k 2010 clusters in nuclei - vol.1 , ed .
beck c , _ lecture notes in physics _ * 818 * 232 von oertzen w , freer m , and kanada - enyo y 2006 _ phys . rep . _
* 432 * 43 horiuchi h and ikeda k 1968 _ prog .
_ * 40 277 von oerzten w 2001 _ eur .
j. a _ * 11 403 von oerzten w and milin m , 2013 clusters in nuclei - vol.3 , ed .
beck c , _ lecture notes in physics _
( in press ) ; milin m , this conference kanada - enyo y and kimura m 2010 clusters in nuclei - vol.1 , ed .
beck c , _ lecture notes in physics _ * 818 * 129 ikeda k _ et al .
_ 2010 clusters in nuclei - vol.1 , ed .
c. beck , _ lecture notes in physics _ * 818 * 165 nakamura n and kondo y 2012 clusters in nuclei - vol.2 , ed .
beck c , _ lecture notes in physics _ * 848 * 67 zagrebaev v and greiner w 2010 clusters in nuclei - vol.1 , ed .
beck c , _ lecture notes in physics _ * 818 * 267 pyatkov y and kamanin d 2013 clusters in nuclei - vol.3 , ed .
beck c , _ lecture notes in physics _
( in press ) beck c 2004 _ nucl .
phys . _ * a 738 24 beck c 2004 _ int .
j. mod .
phys . _ * e13 9 cseh j _ et al .
_ 2009 _ phys .
c _ * 80 * 034320 taniguchi y _ et al .
_ 2009 _ phys .
c _ * 80 * , 044316 ichikawa t , kanada - enyo y , and meller p , 2011 _ phys .
c _ * 83 * 054319 jenkins d g _ et al .
_ 2012 _ phys .
c _ * 86 * 064308 jenkins d g 2013 clusters in nuclei - vol.3 , ed .
beck c , _ lecture notes in physics _
( in press ) ; jenskins d g this conference .
darai j , cseh j and jenkins d g 2012 _ phys .
c _ * 86 * 064309 kimura m and horiuchi h 2004 _ phys .
c _ * 69 * 051304 lnnroth t _
_ 2010 _ eur .
phys . j. a _ * 46 5 pandit d _
_ 2010 _ phys .
c _ * 81 * 061302 beck c _ et al .
_ 2008 _ aip conf .
proc . _ * 1098 * 207 svensson c e , _ et al .
_ phys . rev .
* 85 * 2693 sciani w _
_ 2009 _ phys .
c _ * 80 * 034319 beck c _ et al .
_ 2009 _ phys .
c _ * 80 * 034604 ideguchi e _ et al .
_ 2001 _ _ phys . rev .
_ _ * * 87 * * 222501 rousseau m _ et al .
_ 2002 _ phys .
c _ * 66 034612 taniguchi y _ et al .
_ 2007 _ phys .
c _ * 76 044317 norrby m _
et al . _ ( unpublished ) ; norrby m , private communication and norrby m , this conference .
oleary c d _
_ 2000 _ phys
c _ * 61 064314 fukada m _
_ 2009 _ phys .
c _ * 80 064613 salsac m d _ et al .
_ 2008 _ nucl .
phys . _ * a 801 * 1 nouicer m _
_ 1999 _ phys .
c _ * 60 041303(r ) rudolph d _
_ 1999 _ phys .
* 82 3763 beck c _ et al .
_ 2001 _ phys .
c _ * 63 014607 bhattacharya c _ et al .
_ 2002 _ phys .
c _ * 65 014611 cindro n _ et al .
_ 1979 _ phys .
lett . _ * 84b 55 mermaz m c _ et al .
_ 1984 _ nucl .
phys . _ * a 429 351 pocanic d and cindro n 1985 _ nucl .
_ * a 433 531 schimizu j _ et al .
_ 1982 _ phys .
lett . _ * 112b 323 gai m _
1984 _ phys .
c _ * 30 925 von oertzen w _
_ 2008 _ eur .
j. a _ * 36 * 279 zagrebaev v i _ et al _ 2010 _ phys .
c _ * 81 044608 pyatkov y f _ et al .
_ 2010 _ eur . phys .
j. a _ * 45 * 29 sanders s j , szanto de toledo a and beck c 1999 _ phys .
_ * 311 487 tohsaki a , _ et al .
_ 2001 _ phys .
* 87 192501 von oertzen w 2010 clusters in nuclei - vol.1 , ed .
beck c , _ lecture notes in physics _ * 818 * 102 yamada t _
_ 2012 clusters in nuclei - vol.2 , ed .
beck c , _ lecture notes in physics _ * 848 * 229 hoyle f 1954 _ astrophys .
j. suppl .
* 1 121 funaki y _ et al .
_ 2008 _ phys .
lett . _ * 101 082502 itoh m _ et al .
_ 2004 _ nucl .
phys . a _ * 738 268 von oertzen w
_ 2010 _ eur .
j. a _ * 43 17 von oertzen w
_ 2009 _ aip conf .
proc . _ * 1165 19 milin m _
_ 2009 _ eur .
j. a _ * 41 335 bohlen h g _
_ 2011 _ eur .
phys . j. a _ * 47 44 descouvemont p and baye d 1985 _ phys . rev .
c _ * 31 2274 furutachi n _ et al .
_ 2008 _ prog .
phys . _ * 119 403 wheldon c _ et al .
_ 2008 _ nucl .
a _ * 811 276 papka p and beck c 2012 clusters in nuclei - vol.2 ed .
beck c , _ lecture notes in physics _ * 848 * 299 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * | since the pioneering discovery of @xmath0c+@xmath0c molecular resonances half a century ago , a great deal of research work has been undertaken in the @xmath1-clustering study .
our knowledge on the physics of nuclear molecules has increased considerably and nuclear clustering remained one of the most fruitful domains of nuclear physics , facing some of the greatest challenges and opportunities in the years ahead . in this work , the occurrence of `` exotic '' shapes in light @xmath2=@xmath3 @xmath1-like nuclei is investigated .
various approaches of the superdeformed and hyperdeformed bands associated with quasimolecular resonant structures are presented .
clustering aspects are also briefly discussed for light nuclei with neutron excess trough very recent results on neutron - rich oxygen isotopes . |
black holes are the most exotic objects in physics and their connection with thermodynamics is even more surprising . just like other thermodynamical systems ,
black holes have physical temperature and entropy .
the analogy between the black hole thermodynamics and the four laws of thermodynamics was first proposed in 1970 s @xcite and the temperature ( @xmath0 ) and entropy @xmath1 are analogous of the surface gravity @xmath2 and area @xmath3 of the black hole event horizon respectively
. laws of black hole thermodynamics are studied in literature @xcite . in @xcite universal properties of black holes and the first law of black hole inner mechanics is discussed . in @xcite horizon entropy sums in a(ds ) spacetimes is studied . in @xcite authors have discussed the spin entropy of a rotating black hole .
the study of phase transition in black holes is a fascinating topic @xcite .
if a black hole has a cauchy horizon ( @xmath4 ) and an event horizon(@xmath5 ) then it is quite interesting to study different quantities like the product of areas of a black hole on these horizons .
products of thermal quantities of the rotating black holes @xcite and area products for stationary black hole horizons @xcite have been studied in literature .
calculations show that sometimes these products do not depend on the adm ( arnowitt - deser - misner ) mass parameter but only on the charge and angular momentum .
the relations that are independent of the black hole mass are of particular interest because these may turn out to be `` universal '' and hold for more general solutions with nontrivial surroundings too .
kiselev @xcite considered einstein s field equation surrounded by quintessential matter and proposed new solutions , dependent on state parameter @xmath6 of the matter surrounding black hole .
recently some dynamical aspects , i.e. collision between particles and their escape energies after collision around kiselev black hole @xcite have been studied . in this work
we consider the solution of reissner nordstrm(rn ) black hole surrounded by energy - matter , derived by kiselev and study the important thermodynamic features of black hole at both the horizons of the black hole .
we also consider solution of schwarzschild black hole surrounded by energy - matter and analyzed its different thermodynamic products .
furthermore , we have considered the charged dilaton black hole and computed its various thermodynamic products .
the plan of the work is as follows : in section ( ii ) , we discuss the basic aspects of rn black hole surrounded by radiation .
results show that the products of area and entropy calculated at @xmath7 are independent of mass of the black hole , while the other products are mass dependent . in subsections of (
ii ) , the first law of thermodynamics is obtained by the smarr formula approach , later the rest mass is written in terms of irreducible mass of the black hole , also the phase transition in heat capacity of the black hole is discussed . section ( iii )
consists of discussions on thermodynamic aspects of rn black hole surrounded by dust . in section ( iv ) , thermodynamics of the schwarzschild black hole surrounded by quintessence is studied . in section ( v ) we have computed the thermodynamic product relations for dilaton black hole .
all the work is concluded in the last section .
we set @xmath8 , throughout the calculations .
the spherically symmetric and static solutions for einstein s field equations , surrounded by energy - matter , as investigated by kiselev @xcite can be written as : @xmath9 where @xmath10 here @xmath11 and @xmath12 are the mass and electric charge , of the black hole respectively , @xmath13 is the normalization parameter and @xmath6 is the state parameter of the matter around black hole .
we consider the cases when rn black hole is surrounded by radiation ( @xmath14 ) and dust ( @xmath15 ) . for @xmath14 two horizons of the black hole
are obtained from : @xmath16i.e .
@xmath17 here @xmath18 denotes the normalization parameter for radiation case , with dimensions , @xmath19=l^2 $ ] , where @xmath20 denotes length , @xmath21 is the outer horizon named as event horizon @xmath5 and @xmath22 is the inner horizon known as cauchy horizon @xmath23 , @xmath24 are the null surfaces of infinite blue - shift and infinite red - shift respectively @xcite .
( [ m3 ] ) one can obtain , @xmath25 so product of horizons is independent of mass of the black hole but depends on electric charge and @xmath18 .
areas of both horizons of the black hole are : @xmath26 the corresponding semi - classical bakenstein - hawking entropy at @xmath24 is @xcite : @xmath27 hawking temperature of @xmath24 is determined by using the formula @xmath28 \nonumber\\ & = & \frac{r_{\pm}- { \cal m}}{2\pi(2{\cal m}r_{\pm } -q^2 + \sigma_r)}.\end{aligned}\ ] ] surface gravity is the force required to an observer at infinity , for holding a particle in place , which is equal to the acceleration at horizon due to gravity of a black hole @xcite : @xmath29 @xmath30 the komar energy of the black hole is defined as @xcite @xmath31 products of surface gravities and surface temperatures at @xmath24 are @xmath32 the komar energies at @xmath24 results in @xmath33 since these products ( except product of horizons @xmath34 ) are mass dependent , so are not universal quantities .
the products of areas and entropies at @xmath24 , @xmath35 are mass independent , so these are the universal parameters of the black hole .
the expression of the area of black hole is @xcite : @xmath37,\ ] ] from the area of both the horizons @xmath38,\ ] ] one can write the adm mass of the black hole as : @xmath39 according to the first law of thermodynamics , differential of mass of the black hole can be related with the change in its area and electric charge . since the effective surface tension at the horizon is proportional to the temperature of the black hole horizons , so we can write : @xmath40 where @xmath41 and @xmath42 are defined as : @xmath43 the effective surface tension can be rewritten as : @xmath44 so rn black hole surrounded by radiation satisfies the first law of thermodynamics , verified by smarr formula approach
. christodoulou and christodoulou and ruffini @xcite had shown that the mass of a black hole could be increased or decreased but the irreducible mass @xmath45 of a black hole can not be decreased .
in fact , most processes result in an increase in @xmath45 and during reversible process this quantity also does not change .
also , the surface area of a black hole has behavior @xcite @xmath46 so there exist a relation between area and irreducible mass .
the @xmath45 is proportional to the square root of the black hole s area .
since the rn - radiation space - time has regular event horizon and cauchy horizons so the irreducible mass of a black hole is proportional to the square root of its surface area @xcite@xmath47where @xmath48 and @xmath49 are irreducible masses defined on @xmath24 respectively .
product of the irreducible masses at @xmath50 is : @xmath51 this product is universal because it does not depend on mass of the black hole .
the expression for the rest mass of the rotating charged black hole given by christodoulou and ruffini in terms of its irreducible mass , angular momentum and charge is @xcite : @xmath52 setting @xmath53 , mass of the rn black hole surrounded by radiation becomes : @xmath54 another important measure to study the thermodynamic properties of a black hole is the heat capacity , its nature ( positivity or negativity ) reflects the stability or instability of a thermal system ( black hole ) .
a black hole with negative heat capacity is in unstable equilibrium state , i.e. it may decay to a hot flat space by emitting hawking radiations or it may grow without limit by absorbing radiations @xcite .
one can get heat capacity of a black hole by using @xmath56 where mass @xmath11 in terms of @xmath57 is : @xmath58 the partial derivatives of mass @xmath11 and temperature @xmath59 with respect to @xmath57 are : @xmath60 and from eq .
( [ m9 ] ) we have @xmath61 the expression for heat capacity @xmath62 for rn black hole surrounded by radiation at horizons becomes : @xmath63 note that there are two possible cases for heat capacity to be positive : + * case @xmath64 : * when both @xmath65 and @xmath66 are positive . +
* case @xmath67 : * when both @xmath65 and @xmath66 are negative .
+ since we are interested in positive @xmath68 only , so case-@xmath64 implies that heat capacity is positive if @xmath69 from case-@xmath67 we get @xmath70which is not possible , so we exclude this case .
+ heat capacity is negative if + * case @xmath71 * : @xmath72 and @xmath73 , + * case @xmath74 * : @xmath75 and @xmath76 .
+ from case-(@xmath71 ) implies that for : @xmath77 heat capacity is negative . from case-(@xmath74 ) we get negative capacity for @xmath78 we are interested in positive @xmath68 only .
the region where heat capacity is negative , corresponds to an instable region around black hole , whereas a region in which the heat capacity is positive , represents a stability region .
behavior of heat capacity given in eq .
( [ m21 ] ) is shown in fig .
( [ 1 m ] ) .
heat capacity is negative in the region @xmath79 and @xmath80 , while positive for @xmath81 . and
again goes to instable region , we chose @xmath82 and @xmath83,width=340 ] interestingly , the product of heat capacity on @xmath24 becomes @xmath84 } { \left[4(q^2-\sigma_r)-3{\cal m}^2\right]},\end{aligned}\ ] ] the product depends on mass parameter and charge parameter .
thus the product of specific heats is not universal .
metric of rn black hole surrounded by dust is same as in eq .
( [ m1 ] ) , @xmath85 defined in eq .
( [ m01 ] ) with @xmath15 and @xmath86 becomes : @xmath87 where @xmath88= l$ ] .
the horizons are @xmath89 area of the horizons @xmath24 is : @xmath90,\nonumber\\ & = & 4\pi [ ( 2{\cal m}+\sigma_d)r_{\pm}-q^2].\end{aligned}\ ] ] entropy of the horizons is @xmath91.\end{aligned}\ ] ] surface gravity and hawking temperature of horizons are respectively : @xmath92},\ ] ] and @xmath93},\nonumber\\ & = & \frac{1}{4 \pi}\big(\frac{r_{\pm}^2-q^2}{r_+^3}\big),\end{aligned}\ ] ] where we have used @xmath94 .
the komar energy becomes : @xmath95 product of surface gravities and temperatures at the horizons , are @xmath96 product of komar energies at the horizons is : @xmath97 note that all products are mass dependent , so these quantities are not universal .
products of areas and entropies at both horizons @xmath24 are : @xmath98 it is clear that area product and entropy product are universal entities .
area of both the horizons must be constant given by @xmath99.\end{aligned}\ ] ] using eq .
( [ m015 ] ) mass of the black hole or adm mass is expressed in terms of the areas of horizons as : @xmath100 differential of mass could be expressed in terms of physical invariants of the horizons , @xmath101 where @xmath102 we can rewrite effective surface tension as : @xmath103.\end{aligned}\ ] ] or @xmath104 so the first law of black hole thermodynamics is verified , for rn black hole surrounded by dust , using the smarr formula approach .
the expression for irreducible mass for rn black hole surrounded by dust is @xmath105 here @xmath48 and @xmath49 are irreducible masses defined on inner and outer horizons respectively .
area of @xmath24 , in terms of @xmath106 is : @xmath107 product of the irreducible mass at the horizons @xmath50 is : @xmath108 this product is independent of mass of the black hole .
mass of the black hole expressed in terms of its irreducible mass and charge is : @xmath109 mass of rn black hole surrounded by dust in terms of @xmath57 is : @xmath110 partial derivatives of mass @xmath11 and temperature @xmath59 with respect to @xmath57 are : @xmath111 and @xmath112where @xmath0 is given in eq .
( [ m09 ] ) .
the heat capacity @xmath113 at the horizons is : @xmath114 in this case the product formula for heat capacity is found to be @xmath115}{4q^4-q^2 ( 2{\cal m}+\sigma_d)^2}.\end{aligned}\ ] ] it is clear that the product formula does depend on mass parameter , so it is not universal in nature .
note that there are two possible cases for heat capacity , @xmath116 , to be positive : + * case @xmath64 : * when both @xmath117 and @xmath118 are positive . +
* case @xmath67 : * when both @xmath117 and @xmath118 are negative .
+ considering @xmath119 as a positive quantity ( for physically accepted region , @xmath68 ) from case-1 , we can say @xmath116 is positive for only @xmath120 i.e. @xmath121 while from case-@xmath67 we can say @xmath116 is negative for only @xmath122 i.e. @xmath123 heat capacity is negative if + * case @xmath71 * : @xmath124 and @xmath125 , + * case @xmath74 * : @xmath122 and @xmath126 .
+ case ( @xmath74 ) is not possible mathematically since @xmath127 , while in case ( @xmath71 ) heat capacity is negative in the region , where @xmath68 satisfies both the conditions @xmath128 and @xmath129 simultaneously .
we consider @xmath13 and @xmath12 both are positive in all the calculations .
behavior of the heat capacity given in eq .
( [ m23 ] ) is shown in fig .
( [ 2 m ] ) . ,
@xmath83 and @xmath130.,width=340 ]
metric for schwarzschild black hole surrounded by quintessence is same as defined in eq .
( [ m1 ] ) and @xmath85 defined in eq .
( [ m01 ] ) with @xmath131 , @xmath132 and @xmath133 becomes : @xmath134 where dimensions of @xmath135 are that of @xmath136 .
the horizons , @xmath57 , of the black hole are : @xmath137 product of the two horizons yield : @xmath138 and it is depending on mass of the black hole and @xmath135 .
areas of the horizons are : @xmath139.\end{aligned}\ ] ] entropy at the horizons @xmath24 is : @xmath140 hawking temperature of the horizons is : @xmath141,\end{aligned}\ ] ] and the surface gravity on the black hole horizons @xmath24 is given by : @xmath142.\ ] ] the komar energy is given by : @xmath143 product of surface gravities and temperatures of @xmath24 is : @xmath144 product of komar energies of the horizons is : @xmath145 respectively .
it is clear that all these products are depending on mass of the black hole , so these quantities are not universal .
the products of areas and entropies at @xmath24 are : @xmath146 again both products are not universal quantities
. writing area of both horizons of the black hole as : @xmath147.\ ] ] using eq .
( [ om15 ] ) mass of the black hole or adm mass is expressed in terms of the areas of horizons as : @xmath148 .
\end{aligned}\ ] ] differential of mass , expressed in terms of physical invariants of the horizons is : @xmath149 where @xmath150\nonumber\\ & = & \frac{1}{16 \pi \mathcal{a}_{\pm}}[8 \pi \mathcal{m}- \mathcal{a}_{\pm}\sigma_q ] \nonumber\\ & = & \frac{\kappa_{\pm}}{8\pi},\end{aligned}\ ] ] where we have used @xmath151 , and @xmath152 is defined in eq .
( [ om8 ] ) .
hence the first law of thermodynamics is satisfied by schwarzschild black hole surrounded by quintessence .
the irreducible mass of schwarzschild black hole surrounded by quintessence is @xmath153 here @xmath48 and @xmath49 are irreducible masses defined on inner and outer horizons respectively .
area of @xmath24 , in terms of @xmath106 is : @xmath107 product of the irreducible mass at the horizons @xmath50 is : @xmath154 this product is depending on mass of the black hole .
expression of mass given in eq .
( [ om16 ] ) , in terms of irreducible mass becomes : @xmath155 mass of schwarzschild black hole surrounded by quintessence in terms of @xmath57 is : @xmath156 the partial derivative of mass @xmath11 with respect to @xmath57 is : @xmath157 and using eq .
( [ om9 ] ) we get @xmath158.\ ] ] the expression for heat capacity @xmath113 at the horizon becomes : @xmath159 .a comparison of thermodynamical parameters for rn - radiation , rn - dust and schwarzschild - quintessence black holes .
[ cols="^,^,^,^",options="header " , ] heat capacity would be positive if : + * case @xmath64 : * @xmath160 and @xmath161 . +
* case @xmath67 : * @xmath162 and @xmath163 .
+ heat capacity is negative if + * case @xmath71 : * @xmath162 and @xmath161 , + * case @xmath74 : * @xmath160 and @xmath163 .
+ behavior of the heat capacity given in eq .
( [ sm23 ] ) is shown in fig .
( [ 3 m ] ) for @xmath164 and @xmath165 , heat capacity is negative for @xmath166 , positive for @xmath167 and @xmath168 , it is zero at @xmath169 and @xmath170 . , and @xmath130 .
heat capacity is positive for @xmath167 and @xmath168 , negative for @xmath166 and is zero at @xmath169 and @xmath171.,width=340 ] a comparison of all the parameters calculated for kiselev solutions is shown in table .
( [ 1table ] ) .
the action for charged black hole in string theory is @xcite : @xmath172,\]]where @xmath173 is the maxwell field , @xmath174 is the scalar field and @xmath71 is an arbitrary parameter specifying the strength of dilaton and the maxwell field s coupling .
we are going to derive the area product formula and entropy product formula for a spherically symmetric dilaton black hole @xcite whose metric can be written in schwarzschild - like coordinates as : @xmath175 where the function @xmath176 is defined by @xmath177 and @xmath178 in these equations , @xmath179 and @xmath180 are constants , which are related to mass and charge of the black hole as : @xmath181 where as usual @xmath11 is mass of the black hole and @xmath12 is electric charge of the black hole .
it may be noted that @xmath12 and @xmath71 are positive .
the horizons of the black hole are determined by the function @xmath182 which yields @xmath183,\label{hocdc}\end{aligned}\ ] ] and @xmath184 is defined by @xmath185 here @xmath179 and @xmath180 are called event horizon ( @xmath5 ) or outer horizon and cauchy horizon ( @xmath4 ) or inner horizon respectively , and @xmath186 or @xmath187 corresponding to the extreme charged dilaton black hole . +
* case i : * when @xmath188 or @xmath189 , the metric corresponds to rn black hole . + * case ii : * when @xmath190 or @xmath191 , the metric corresponds to gibbons - maeda - garfinkle - horowitz - strominger ( gmghs ) black hole .
the expressions for surface gravity of dilaton black hole at both the horizons ( @xmath24 ) are @xmath192 the black hole temperature or hawking temperature at @xmath193 are @xmath194 areas of the horizons ( @xmath193 ) are @xmath195 interestingly , the area of both the horizons go to zero at the extremal limit ( @xmath186 ) which is quite different from the well known rn and schwarzschild black hole .
the other characteristics of this spacetime is that there is a curvature singularity at @xmath196 .
now the entropies of both the horizons ( @xmath193 ) are @xmath197 finally , the komar energy is given by @xmath198 now we compute products of all the parameters given above : @xmath199 interestingly their products go to zero value and independent of mass thus are universal quantities .
all of the above thermodynamical quantities must satisfied the first law of thermodynamics : @xmath200 where @xmath201 and @xmath202 , we chose @xmath83 , @xmath203 and @xmath130.,width=340 ] the irreducible mass at @xmath7 for this black hole is @xmath204 their product yields @xmath205 the heat capacity for this dilaton black hole is calculated to be @xmath206\left[1-\frac{2}{1+n}\frac{q^2}{r_{+}}\right ] } { \left[1-\frac{2(1 + 2n)}{1+n}\frac{q^2}{r_{+}}\right]\left[1-\frac{2}{1+n}\frac{q^2}{r_{+}}\right]^{n}}.~\label{cv1}\end{aligned}\ ] ] due to curvature singularity at @xmath196 the heat capacity at the cauchy horizon diverges .
thus the product of heat capacity diverges .
note that for odd @xmath184 , heat capacity would be positive if @xmath207 or @xmath208 and heat capacity is negative for : @xmath209 or @xmath210 behavior of the heat capacity is shown in fig .
( [ d3 m ] ) . for @xmath83 and @xmath203 ,
given expression of heat capacity ( eq .
( [ cv1 ] ) ) diverges at @xmath211 and @xmath212 , positive for @xmath213 and negative for @xmath214 , @xmath215 , and @xmath216 ( for @xmath217 ) .
we have studied the thermodynamical properties on the inner and outer horizons of kiselev solutions ( rn black hole surrounded by energy - matter ( radiation and dust ) and schwarzschild black hole surrounded by quintessence ) and charged dilaton black hole .
we have studied some important parameters of black hole thermodynamics with reference to their event and cauchy horizons .
we derive the expressions for temperatures and heat capacities of all the black holes mentioned above .
it is observed that the product of surface gravities , surface temperature product and product of komar energies at the horizons are not universal quantities for the kiselev s solutions while products of areas and entropies at both the horizons are independent of mass of black hole ( except for schwarzschild black hole surrounded by quintessence ) .
for dilaton black hole these products are universal except the products of specific heat which has shown divergent properties due to the curvature singularity at the cauchy horizon .
thus the implication of these thermodynamical products may somehow give us further understanding of the microscopic nature of black hole entropy ( both exterior and interior ) in the black hole physics .
using the heat capacity expressions , stability regions of the black holes are also observed graphically . figs .
( [ 1m]-[d3 m ] ) show that the above mentioned black holes undergo a phase transition under certain conditions on @xmath68 . for rn black hole surrounded by radiation and dust and schwarzschild black hole surrounded by quintessence , we derived the first law of thermodynamics using the smarr formula approach .
it is observed that third law of thermodynamics which states that `` surface gravity , @xmath218 , of a black hole can not be reduced to zero in a finite sequence of processes '' holds for all the above mentioned black holes .
the derived expressions of @xmath218 show that it is zero for extreme black holes only .
99 d. christodoulou , _ phys .
lett . _ * 25 * , 1596 ( 1970 ) .
d. christodoulou and r. ruffini , _ phys
. rev . _ * d 4 * , 3552 ( 1971 ) .
r. penrose and r. m. floyd , _ nature _ * 229 * , 177 ( 1971 ) ; s. hawking , _ phys .
* 26 * , 1344 ( 1971 ) .
j. d. bekenstein , _ phys .
* d 7 * , 2333 ( 1973 ) ; j. d. bekenstein , _ phys .
rev . _ * d 9 * , 3292 ( 1974 ) . m. jamil , m. akbar , _ gen.rel.grav . _ * 43 * , 1061 ( 2011 ) ; m. jamil , i. hussain , m. u. farooq , _ astrophys .
space sci . _ * 335 * , 339 ( 2011 ) ; m. jamil , i. hussain , _ int .
phys . _ * 50 * , 465 ( 2011 ) ; m. akbar , a. a. siddiqui , _ phys .
* b 656 * , 217 ( 2007 ) .
a. castro and m. j. rodriguez , _ phys .
* d 86 * , 024008 ( 2012 ) .
j. wang , w. xu and x. meng , _ j. high energy phys .
_ * 01 * , 031 ( 2014 ) .
a. curir , _
nuovo cimento .
_ * b 51 * , 262 ( 1979 ) .
d. pavon , _ phys .
_ * d 43 * , 2495 ( 1990 ) ; s. w. hawking , d. n. page , _ commun .
phys . _ * 87 * , 577 ( 1983 ) .
e. t. newman , e. couch , k. chinnapared , a. exton , a. prakash , and r. torrence , _ j. math
_ * 6 * , 918 ( 1965 ) .
p. pradhan , _ eur .
j. _ * c 74 * , 2887 ( 2014 ) .
p. pradhan , arxiv : 1503.04514 [ gr - qc ] .
b. majeed , m. jamil , arxiv:1507.01547 [ hep - th ] .
m. cvetic , g. w. gibbons and c. n. pope , _ phys .
lett . _ * 106 * , 121301 ( 2011 ) .
m. visser , _ phys .
rev . _ * d 88 * , 044014 ( 2013 ) . v. v. kiselev , _ class .
grav . _ * 20 * , 1187 ( 2003 ) .
m. jamil , s. hussain , and b. majeed , _ eur .
j. _ * c 75 * , 24 ( 2015 ) .
s. chandrashekar , _ the mathematical theory of black holes _ , clarendon press , oxford ( 1983 )
. s. hawking , _ phys .
* 26 * , 1344 ( 1971 ) .
e. poisson , _ a relativist s toolkit : the mathematics of black hole mechanics _ , cambridge university press , ( 2007 ) .
a. komar , _ phys .
_ * 113 * , 934 ( 1959 ) .
l. smarr , _ phys .
lett . _ * 30 * , 71 ( 1973 ) ; l. smarr , _ phys .
* d 7 * , 289 ( 1973 ) .
j. m. bardeen , b. carter , s. w. hawking , _ commun .
phys . _ * 31 * , 161 ( 1973 ) .
d. j. gross , m. j. perry , and l. g. yaffe , _ phys .
* d 25 * , 330 ( 1982 ) .
d. garfinkle , g. t. horowitz , a. strominger , _ phys .
_ * d 43 * , 3140 ( 1991 ) . | * abstract : * we have studied some thermodynamics features of kiselev black hole and dilaton black hole . specifically we consider reissner nordstrm black hole surrounded by radiation and dust , and schwarzschild black hole surrounded by quintessence , as special cases of kiselev solution . we have calculated the products of black hole thermodynamics parameters , including surface gravities , surface temperatures , komar energies , areas , entropies , horizon radii and the irreducible masses , at the inner and outer horizons .
the products of surface gravities , surface temperature product and product of komar energies at the horizons are not universal quantities . for kiselev solutions products of areas and entropies at both the horizons are independent of mass of the black holes ( except for schwarzschild black hole surrounded by quintessence ) . for charged dilaton black hole ,
all the products vanish . using the smarr formula approach , the first law of thermodynamics is also verified for kiselev solutions .
the phase transitions in the heat capacities are also observed . |
the lotka - volterra equation is at the heart of population dynamics , but also possesses a famous economic interpretation
. introduced by richard goodwin @xcite in 1967 , the model in its modern form @xcite reduces to the planar oscillator on a subset @xmath0 of @xmath1 : [ eq : goodwin_model ] \ { ll d x_t & = x_t ((y_t)-)dt + d y_t & = y_t ((x_t ) -)dt . , where @xmath2 denotes the wage share of the working population and @xmath3 the employment rate , @xmath4 and @xmath5 are constant , and the following assumption is made on @xmath6 and @xmath7 .
[ ass : non - linear goodwin ] consider system .
1 . @xmath8 , @xmath9 , @xmath10 for all @xmath11 , @xmath12 and @xmath13 .
@xmath14 , @xmath15 for all @xmath16 , @xmath17 and @xmath18 .
lemma [ lem : goodwin system ] below asserts that assumption [ ass : non - linear goodwin ] is sufficient to have @xmath19 for any @xmath20 if @xmath21 .
this property preserves the above interpretation for @xmath22 and @xmath23 : the employment rate can not exceed one for obvious reasons , but the wage share can , depending on the chosen economic assumptions , see @xcite .
this distinctive feature of the economic version on its biological counterpart follows from a construction based on assumptions describing a closed capitalist economy .
it can be done in three steps : 1 . assume a leontief production function @xmath24 with full utilization of capital , i.e. , @xmath25 .
here , @xmath26 is the yearly output , @xmath27 the invested capital , @xmath28 a capital - to - output ratio , @xmath29 is the average productivity of workers and @xmath30 is the size of the labor class .
2 . the capital depreciates and receives investment , i.e. , @xmath31 , where @xmath32 is the depreciation rate and @xmath6 the investment function .
goodwin @xcite originally invokes say s law , i.e. , @xmath33 .
3 . assume a reserve army effect for wage negotiation of the form @xmath34 where @xmath35 represents the real wage of the total working population , and @xmath7 is the phillips curve .
defining @xmath36 allows to retrieve for @xmath37 .
the class - struggle model has been extensively studied because it allows to generate endogenous real business cycles affecting the production level @xmath26 , e.g. @xcite . on this matter ,
goodwin himself conceded that the model is `` starkly schematized and hence quite unrealistic '' @xcite .
it hardly connects with irregular observed trajectories , see @xcite .
the objective of this paper is thus to study the following perturbed version of by a standard brownian motion @xmath38 on a stochastic basis @xmath39 : [ eq : stochastic_goodwin_1 ] \ { ll d x_t & = x_t (((y_t)-+ ^2(y_t))dt + ( y_t)dw_t ) + d y_t & = y_t (((x_t ) - + ^2(y_t ) ) dt + ( y_t)dw_t ) . , where @xmath40 is a positive function of @xmath23 bounded by @xmath41 , and the filtration @xmath42 is generated by paths of @xmath43 .
the form of @xmath40 is discussed in remark [ rem : conditions ] after . a stronger condition , assumption [
ass : growth conditions ] , is assumed later on the behavior of @xmath40 to ensure that solutions of remain in @xmath0 .
the example of section [ sec : example ] will also illustrate how such condition can hold .
we modify the economic development ( i ) , ( ii ) and ( iii ) by introducing the perturbation on one assumption , namely we assume that for @xmath20 , [ eq : stochastic_productivity ] da_t : = a_t d_t = a_t ( dt - ( y_t)dw_t ) , a_00 , instead of @xmath44 . using it formula with in the previous reasoning retrieves .
productivity is one of the few exogenous parameters of the model , and one of those that were significantly invoked as influencial over business cycles , e.g. @xcite . without arguing for the pertinence of that particular assumption ,
we simply suggest here that a standard continuous perturbation in this crucial parameter seems a good starting point . to our knowledge , this is the first attempt to consider random noise in goodwin interpretation of the famous prey - predator model .
to stay in the spirit of the economic application , the present paper studies the cyclical behavior of the deterministic system and the stochastic version .
namely , our contribution are as follows , developed in the present order : * in section [ sec : goodwin ] , we fully characterize solutions of and the period of their orbits .
this generalizes standard results on lotka - volterra systems to bounded domains of existence . * in section
[ sec : stochastic ] , we provide existence conditions for regular solutions of .
we use the entropy of to estimate the deviation induced by .
we provide a definition of stochastic orbits for . the proof that solutions of draw stochastic orbits in finite time around a unique point is given in section [ sec : proof ] .
our contribution has to be put in contrast with numerous studies of random perturbations of the lotka - volterra system .
apart from the obviously different origin of perturbations in the model , attention was mainly given to systems like for its asymptotic behavior ( e.g. , @xcite ) , regularity , persistence and extinction of species ( e.g. , @xcite ) , and the addition of regimes , jumps or delay ( e.g. , @xcite ) .
here , we attempt to provide a relevant description of trajectories @xmath45 and indirectly @xmath26 , namely a cyclical behavior .
this is done using stochastic lyapunov techniques for recurrent domains as described in @xcite . by conveniently dividing the domain @xmath0
, we obtain that almost every trajectory `` cycles '' around a point in finite time .
the @xmath46-boundedness is out of reach with our method , but numerical simulations are presented in section [ sec : example ] , not only to provide expectation of cycles , but also allow to conjecture a limit cycle phenomenon for the expectation of @xmath45 .
it is somewhat unclear how assumption [ ass : non - linear goodwin ] and late assumption [ ass : growth conditions ] on @xmath7 , @xmath6 and @xmath40 , are relevant in these results .
we show below thatthey are sufficient to obtain existence of regular solutions to .
this actually emphasizes the role played by the entropy of the deterministic system in the well - posedness of the stochastic system and as a natural measure for perturbation .
according to assumption [ ass : non - linear goodwin ] , there exists only one non - hyperbolic equilibrium point to in @xmath0 given by @xmath47 . on the boundary of @xmath0 ,
there exists also an additional equilibrium @xmath48 which is eluded along the paper .
[ def : lyapunov ] let @xmath49 and @xmath50 be three functions defined by @xmath51 and @xmath52 [ lem : goodwin system ] let @xmath21 .
let assumption [ ass : non - linear goodwin ] hold .
then a solution @xmath45 to starting at @xmath53 at @xmath54 describes closed orbits given by the set of points @xmath55 , and @xmath56 for all @xmath57 .
[ proof : goodwin system ] it is well - known @xcite that @xmath50 is a lyapunov function and a constant of motion for system : @xmath58 and @xmath59 take non - negative values with @xmath60 , and @xmath61 .
additionally , under assumption [ ass : non - linear goodwin].(i)-(ii ) , @xmath62 so that for any @xmath21 , @xmath63 and the solution stays in @xmath0 .
the value of @xmath50 characterizing an orbit , it is in bijection with its period . the following generalizes @xcite .
[ th : periods_goodwin ] let @xmath64 be a solution to satisfying assumption [ ass : non - linear goodwin ] , with @xmath65 .
let @xmath66 , and @xmath67 the two solutions to equation @xmath68 .
define three functions @xmath69 by @xmath70 then @xmath71 for @xmath72 defined by [ eq : period_goodwin ] t(v_0):=_log()^(|x ) - dz .
let @xmath73 , @xmath74 and @xmath45 a solution to starting at @xmath75 .
according to lemma [ lem : goodwin system ] , @xmath76 implies @xmath77 .
then @xmath78=:i$ ]
. homogeneity of allows to set @xmath79 without loss of generality .
let @xmath80 .
for @xmath81 $ ] , @xmath82\x [ \wt{y } , \bar{y}]$ ] , with @xmath83 such that @xmath84 .
let @xmath85 for @xmath20 . then rewrites @xmath86 , @xmath87 and we get @xmath88 \;.\ ] ]
let @xmath89 and define @xmath90 , to rewrite again [ eq : second_order_system ] \
{ ll dz / dt & = u + du / dt & = (u+ ) [ ( e^z ) - ] . .
since @xmath91:=[\log(\underline{x } ) , \log(\bar{x})]$ ] and @xmath92 $ ] for @xmath81 $ ] , separation of variables in provides two quantities @xmath93 and @xmath94 : [ eq : period_function_1 ] f(u):=_0^u ds = _
^z[(e^s ) - ] ds = : g(z ) .
the function @xmath93 verifies @xmath95 , is increasing on @xmath96 $ ] and decreasing on @xmath97 $ ] with @xmath98 so that @xmath99 . coming back to @xmath100 we get @xmath101 implying that @xmath102 $ ] for @xmath103 $ ] .
we can write @xmath104 where @xmath105 and @xmath106 are the two restrictions of @xmath93 on @xmath1 and @xmath107 respectively . notice that if @xmath108 $ ] , then @xmath109 $ ] .
thus , @xmath110 is a strictly increasing function of @xmath111 taking its values in @xmath112 $ ] .
getting back to @xmath113 for @xmath94 , we have for @xmath114 @xmath115 since @xmath116 we have @xmath117}g(z ) = g(\log(\wt x ) ) = v_1(\underline{x } ) = v(\underline{x } , \wt y)$ ] , while minimums are given by @xmath118 .
this sums up with @xmath119)\subset [ 0,v_0]$ ] , so we can write on this interval @xmath120 which finally gives @xmath121 we apply the same method for the other half orbit , taking @xmath122 and @xmath123 , to reach the other half of expression , i.e. , @xmath124 .
[ rem : period ] a first order approximation of at @xmath125 provides a linear homogeneous system , which solution is trivially given by a linear combination of sines and cosines of @xmath126 .
it follows that @xmath127
we study a specific case of where the deterministic part cancels at a unique point in @xmath0 defined by @xmath128 where @xmath129 comes from the following . [
ass : unique_root ] there is a unique @xmath130 such that @xmath131 for a stochastic differential equation to have a unique global solution for any given initial value , functions @xmath7 and @xmath6 are generally required to satisfy linear growth and local lipschitz conditions , see @xcite .
we can however consider the following theorem of khasminskii @xcite , which is a reformulation of theorem 3.4 , theorem 3.5 and corollary 3.1 of @xcite to our context .
[ th : existence ] consider the following stochastic differential equation for @xmath132 taking values in @xmath133 : @xmath134 let @xmath135 be an increasing sequence of open sets , and @xmath136 a sequence of constants , verifying 1 .
@xmath137 for all @xmath138 , 2 .
3 . for any @xmath138
, functions @xmath140 and @xmath40 are lipschitz on @xmath141 and verify @xmath142 for any @xmath143 .
let @xmath144 and @xmath145 be such that , denoting @xmath146 the generator associated with , 1 .
@xmath147 on the set @xmath148 , 2 .
@xmath149 for any @xmath20 .
then for any @xmath150 , there exists a regular adapted solution to , unique up to null sets , with the markov property and verifying @xmath151 for all @xmath20 almost surely . to satisfy conditions ( a ) to ( e ) , we study under the additional sufficient growth conditions .
[ ass : growth conditions ] there exist two positive constants @xmath152 such that 1 .
@xmath153 for all @xmath154 , 2 .
@xmath155 for all @xmath156 .
[ rem : conditions ] assumption [ ass : growth conditions].(i ) involves both @xmath7 and @xmath40 to ensure that @xmath157 for all @xmath20 almost surely .
assumption [ ass : growth conditions].(ii ) holds for polynomial growth of @xmath6 , suiting the classical conditions of existence on @xmath1 for @xmath2 .
the dependence of @xmath40 could be generalized to @xmath22 in full generality , implying a stronger condition than ( ii ) .
we refrain from doing this easy extension , emphasizing the unavoidable dependence in @xmath23 . for @xmath158 , we recall the diffusion operator associated with by @xmath159 ( t , x , y ) . \end{array}\ ] ]
[ prop : existence ] let @xmath160 .
let assumptions [ ass : non - linear goodwin ] , [ ass : unique_root ] and [ ass : growth conditions ] hold .
then there exists a solution @xmath161 to staying in @xmath0 almost surely .
let us show that conditions ( a ) to ( e ) of theorem [ th : existence ] are fulfilled .
consider the sequence of sets @xmath135 defined by @xmath162 .
for any @xmath138 , @xmath141 is open and @xmath163 .
( a ) and ( b ) are satisfied with the limit @xmath164 . according to assumption [ ass : non - linear goodwin ] , one can always find @xmath165 big enough such that @xmath166 for any @xmath167 , and ensures the local lipschitz condition ( c ) .
now consider @xmath50 of definition [ def : lyapunov ] which is @xmath168 on @xmath0
. applying , @xmath169(\phi(y)-\alpha+\sigma^2(y ) ) + \l[\phi(y ) - \phi(\wt y ) \r](\kappa(x)-\gamma+\sigma^2 ( y ) ) \\ & + \left(\l [ \kappa(x ) - \kappa(\wt x)-x\kappa'(x ) \r ] + \l [ \phi(\wt y ) - \phi(y ) + y \phi'(y)\r ] \right)\sigma^2(y)/2 \;.\end{aligned}\ ] ] since @xmath170 and @xmath171 , @xmath172 + \l [ \phi(y ) - \phi(\wt y ) + y \phi'(y)\r ] \right)\sigma^2(y)/2 \;.\ ] ] assumption [ ass : growth conditions ] implies @xmath173 for two positive constants @xmath174 , checking condition ( d ) . from definition [ def : lyapunov ] ,
@xmath175 which implies that @xmath176 , the latter going to infinity with @xmath177 , recall .
similarly , @xmath178 goes to infinity as @xmath22 goes to @xmath179 or @xmath180 .
condition ( e ) is then satisfied , which allows to apply theorem [ th : existence ] .
[ rem : basic properties ] notice that @xmath181 and thus @xmath182 . following assumption [ ass : non - linear goodwin ] , at @xmath183
provides @xmath184 .
it is straightforward that has no equilibrium point in @xmath0 , nor on its boundary @xmath185 .
if the point @xmath48 cancels , we highlight that @xmath186 and by continuity , it holds on a small region @xmath187 .
recalling implies that @xmath45 will diverge from @xmath48 almost surely if @xmath21 .
a solution to can be pictured as a trajectory continuously jumping from an orbit of to another .
along this idea , @xmath50 provides an estimate on trajectories , and can be related via theorem [ th : periods_goodwin ] to the period @xmath72 .
[ prop : estimate ] let @xmath21 , @xmath66 , and @xmath188 be a regular solution to .
we first introduce a constant @xmath189 , the set @xmath190 and the stopping time @xmath191 .
we then introduce two finite constants @xmath192 and @xmath193 then for all @xmath194 @xmath195 fix @xmath194 .
now we define the @xmath196-measurable set @xmath197 where @xmath198 is a martingale defined by @xmath199 for @xmath54 and for @xmath200 by @xmath201 the process @xmath202 is not right - continuous at @xmath54 but still verifies @xmath203 for all @xmath204 .
the property holds by replacing @xmath205 by its representation .
doob s martingale inequality can then be applied : @xmath206 . at last , using it s formula , we have from : @xmath207 so that on @xmath208 \}$ ] @xmath209 almost surely . also , @xmath210 on that set . put in another way , @xmath211 on @xmath212 . according to bayes rule , @xmath213
we now introduce the main result of the paper .
we provide the following tailor - made definition for the cycling behavior of .
[ def : stochastic orbit ] let @xmath214 and @xmath215 .
let @xmath45 be a stochastic process starting at @xmath53 staying in @xmath216 almost surely .
we then introduce @xmath217 the angle between @xmath218^{\top}$ ] and @xmath219^{\top}$ ] .
let @xmath220 be a stopping time ( a stochastic period ) .
then , the process @xmath45 is said to orbit stochastically around @xmath221 in @xmath216 if @xmath222 almost surely .
[ th : finite period ] let @xmath223 and @xmath45 a solution to starting at @xmath53
. then @xmath45 orbits stochastically around @xmath183 in @xmath0 .
more precisely the system produces clockwise orbits inside @xmath0 .
the angle @xmath224 is only defined if @xmath225 .
this can be ensured by either proving that @xmath225 for all @xmath20 almost surely , or by defining @xmath226 as in definition [ def : stochastic orbit ] .
see also remark [ rem : out of point ] the proof of theorem [ th : finite period ] is removed to section [ sec : proof ] .
recall that the probability space is given by @xmath39 with the filtration generated only by @xmath43 is completed with null sets . our proof ,
although unwieldy , allows us to describe precisely the possible trajectories of solutions of .
it consists in defining subregions @xmath227 of the domain @xmath0 , see definition [ def : regions ] below illustrated by fig .
[ fig : rotation ] , and prove that the process exits from them in finite time by the appropriate frontier . according to theorem [ prop : existence ] any regular solution of is a markov process .
we then repeatedly change the initial condition of the system , as equivalent of a time translation and use definition [ def : exit times ] herefater .
we obtain recurrence properties via theorem 3.9 in @xcite .
since it is repeatedly used hereafter , we provide here a version suited to our context .
[ th : reccurent domain ] let @xmath228 be a regular solution of in @xmath0 , starting at @xmath229 , for some @xmath230 .
let @xmath231 verifying @xmath232 for all @xmath233 and @xmath234 where @xmath235 and @xmath236 . then @xmath45 leaves the region @xmath237 in finite time almost surely .
[ def : theta ] let @xmath238 be defined by @xmath239 as a concave decreasing function . for a solution @xmath45 to , we define @xmath240 the finite variation process verifying @xmath241 .
additionally , let @xmath242 .
[ def : regions ] we define eight sets @xmath243 such that @xmath244 and @xmath245 , by @xmath246 [ def : exit times ] let @xmath45 be a solution to starting at @xmath247 . for any @xmath248 , we define the stopping times @xmath249 . by @xmath250 .
since @xmath251 and @xmath252 , the graph illustrates the general case.,height=264 ] [ rem : out of point ] it seems rather clear that the point @xmath183 is not reached in finite time with a positive probability . in the following , the fact that @xmath253 for some small @xmath254 in a neighborhood of @xmath183 implies that @xmath183 is not a limit to almost every path of a solution to , recall remark [ rem : basic properties ] . to ease the reading of the proof of theorem [ th : finite period ] which follows from the following propositions [ prop : r1_r8 ] to [ prop : r8_r1 ] , we divide it in four quadrants around @xmath183 .
we first prove that the process cycles , even in infinite time , for some particular starting points .
[ prop : r1_r8 ] if @xmath255 , then @xmath256 . if @xmath257 , then @xmath258 .
this is a direct consequence of the absence of brownian motion in @xmath259 .
take @xmath255 .
then on @xmath260 $ ] , the process @xmath259 is non increasing almost surely , meaning that @xmath261 can not be reached without first crossing region @xmath262 .
the other side is identical .
[ rem : r1_r2_r3_r4 ] proposition [ prop : r1_r8 ] holds even if @xmath263 , for any @xmath264 involved .
it also implies that if @xmath265 , then @xmath266 almost surely for @xmath267 .
we ought to prove that for @xmath268 , the process reaches @xmath269 in finite time almost surely .
[ prop : r1_r2 ] if @xmath255 then @xmath270 .
let @xmath271\mapsto \sqrt{y}$ ] .
then @xmath272 for any @xmath23 such that @xmath273 .
moreover , @xmath274 theorem [ th : reccurent domain ] stipulates that @xmath275 leaves @xmath276 in finite time almost surely which is only possible via @xmath277 according to proposition [ prop : r1_r8 ] . reaching the boundary
is prevented by theorem [ prop : existence ] .
[ prop : leave r_2 and r_3 ] if @xmath278 then @xmath279 .
we follow the proof of proposition [ prop : r1_r2 ] with @xmath280 . [
prop : r_1 to r_3 ] if @xmath281 then @xmath282 .
_ let @xmath283 be a sequence of stopping times defined by @xmath284 and @xmath285 by construction if @xmath286 for some @xmath287 , then @xmath288 for all @xmath289 .
following propositions [ prop : r1_r8 ] , [ prop : r1_r2 ] and [ prop : leave r_2 and r_3 ] , @xmath290 for all @xmath291 almost surely , and @xmath292 .
we prove in step 2 that this implies @xmath293 providing that holds we immediately get @xmath294 .
_ step 2 . _
if @xmath295 , then for all @xmath138 , @xmath296 and according to proposition [ prop : leave r_2 and r_3 ] , @xmath275 does not converge to the set @xmath297 .
since @xmath298 is a positive decreasing process for @xmath299 , doob s martingale convergence theorem implies that @xmath298 converges pathwise in @xmath300 .
assume now that @xmath298 does not converge to @xmath179 with @xmath111 on @xmath301 .
then for any @xmath254 , and for almost every @xmath302 @xmath303 if the integral explodes to @xmath180 for some @xmath254 on some non null subset @xmath304 , then for almost every @xmath305 , @xmath306 and @xmath307 for almost every @xmath308 , implying that @xmath298 converges to @xmath179 on @xmath93 , a contradiction with @xmath304 , so that holds on @xmath216 .
we then consider the random time @xmath309 , being the first time such that [ eq : limite_borne ] _
0^t _ , n ds c_- , and @xmath310 the smallest index such that @xmath311 .
note that @xmath312 is not a @xmath313-stopping time and @xmath310 is not @xmath313-adapted since they depend on @xmath314 which is @xmath315-measurable .
implies that there exists a random time @xmath316 such that @xmath317 , otherwise we would have a contradiction of on a subset of @xmath216 : @xmath318 this implies that @xmath319 for almost every @xmath320 , and @xmath321 being a continuous process @xmath322 this is impossible for @xmath254 small enough since @xmath298 is strictly decreasing and thus @xmath216 is a null set . holds .
we show that starting from @xmath297 , @xmath275 reaches @xmath323 in finite time almost surely .
[ prop : r2_to_r4 ] if @xmath324 then @xmath325 .
_ we consider @xmath326 with @xmath327 , and aim to prove that the process @xmath328 with @xmath329 is a supermartingale on @xmath330 , for @xmath331 .
notice that it is bounded in @xmath330 .
applying it to @xmath332 first gives @xmath333dt\ ; .\ ] ] then , noticing that @xmath334 , we obtain @xmath335 it is clear that @xmath336 for all @xmath11 .
now notice that for @xmath337 , we have @xmath338 so that @xmath339\\ & = ( x-\we x)^2 x \we x\l[\frac{y}{x}-\frac{\we y}{\we x}\r ] < 0 \ ; .\end{aligned}\ ] ] now on @xmath340 , @xmath341 implies that @xmath342 , so that @xmath343\\ & = ( x-\we x)^2x\l(y-\we y\r)<0 \;.\end{aligned}\ ] ] denoting @xmath344 , we conclude that @xmath345 is a supermartingale for @xmath20 . using optional sampling theorem , assisted by proposition [ prop : leave r_2 and r_3 ]
, @xmath346 almost surely and @xmath347 since @xmath348 then @xmath349 _ step 2 .
_ according to proposition [ prop : leave r_2 and r_3 ] , @xmath346 almost surely for any @xmath324 , and according to proposition [ prop : r_1 to r_3 ] , @xmath350 for all @xmath268 .
taking @xmath324 , we define the sequence @xmath351 with @xmath352 and @xmath353 we then have @xmath354 for any @xmath355 .
the sequence @xmath356 is decreasing in the sense of inclusion , so that @xmath357 using baye s rule , @xmath358 using step 1 of the present proof and the markov property of @xmath275 , @xmath359 plugging this inequality into concludes the proof .
notice that by choosing @xmath360 properly in the above proof , it is possible to be arbitrarily close to @xmath323 in finite time .
the device is used later in proposition [ prop : r7_r8 ] .
[ prop : r4_to_r5 ] if @xmath361 then @xmath362 .
_ we claim that @xmath363 almost surely .
consider the function @xmath364 .
the process @xmath365 is a positive supermartingale on @xmath366 : @xmath367 according to doob s martingale convergence theorem , @xmath368 converges point - wise with @xmath111 .
let @xmath254 and define @xmath369 .
then @xmath370 on @xmath371 , and similarly to proposition [ prop : leave r_2 and r_3 ] , we use theorem [ th : reccurent domain ] to assert that @xmath275 leaves @xmath372 in finite time almost surely .
this being true for any @xmath254 , @xmath373 for almost every @xmath374 . in @xmath323 ,
this is only possible if @xmath375 also , implying that @xmath376 on this set .
this being improbable , @xmath377 almost surely .
_ by denoting @xmath378 , we then define the sequence @xmath379 by @xmath380 if @xmath381 , then , according to proposition [ prop : r2_to_r4 ] , the process reaches back @xmath382 in finite time . using step 1
, we have that @xmath383 . by construction and proposition [ prop :
r2_to_r4 ] , for @xmath138 @xmath384 therefore , @xmath385 and the sequence of sets @xmath386 is decreasing in the sense of inclusion .
altogether we get @xmath387 now using bayes formula and , we finally obtain for every @xmath291 @xmath388 putting and together , @xmath389 implies that @xmath390 _ step 3 .
let @xmath391 .
according to the process @xmath392 is a supermartingale on @xmath393 $ ] .
fixing @xmath200 and applying optional sampling theorem , we obtain @xmath394 since @xmath395 almost surely , we apply fatou s lemma and obtain [ eq : supermartingale_2 ] . since @xmath396 and @xmath397 for all @xmath138 , implies @xmath398\right ) \ind{x_{\tau_{2,5}^n}>\we x } \big| x_{\tau_4^{n}}=\we x } \le 1 \ ; , \ ] ] leading to [ eq : probable ineq ] 1 - .
if @xmath399 then @xmath400 and by continuity @xmath401 , implying [ eq : exp moins un ] ( ^2(y)[_2,5^n()-_4^n()])>1 , \{x__4^n = x}. let s assume that @xmath402 , so that holds . according to , we get @xmath403\right)-1\right ) \ind{x_{\tau_{2,5}^n}>\we x } \big| x_{\tau_4^{n}}=\we x } \rightarrow 0 \text { as } n \rightarrow \infty \ ; .\ ] ] markov inequality then leads to the following convergence for any @xmath254 : @xmath404\right)-1\right ) \ind{x_{\tau_{2,5}^n}>\we x}>\eps
\big| x_{\tau_4^{n}}=\we x } = 0 \ ; .\ ] ] now bayes rules with provides @xmath405\right)-1\right)>\eps \big| x_{\tau_{2,5}^n}>\we x , \ ; x_{\tau_4^{n}}=\we x } \pro{x_{\tau_{2,5}^n}>\we x | x_{\tau_4^{n}}=\we x } \\ & \hspace{3 cm } = \pro{(\tau_{2,5}^n-\tau_4^{n})>8\ln(1+\eps)/\sigma^2(\we y ) \big| x_{\tau_{2,5}^n}>\we x , \ ; x_{\tau_4^{n}}=\we x } \pro{x_{\tau_{2,5}^n}>\we x | x_{\tau_4^{n}}=\we x}\end{aligned}\ ] ] which leads for any @xmath254 to @xmath406 . from step 2 ,
therefore on this set , the continuous mapping theorem asserts that @xmath408 at consecutive stopping times converge in probability . by continuity , this implies @xmath409 and @xmath410 for @xmath411 . by the markov property of @xmath275 , @xmath412 for almost every @xmath413
we conclude that @xmath414 .
[ prop : r5_to_r7 ] if @xmath415 then @xmath416 .
_ step 1 . _
consider @xmath417 for arbitrarily fixed @xmath254 .
assume that @xmath418 .
denoting @xmath419 with @xmath420 and recalling definition [ def : theta ] , @xmath421 for very @xmath422 .
theorem [ th : reccurent domain ] then states that @xmath275 exits @xmath371 in finite time almost surely .
since that @xmath259 is non - decreasing on this set , and recalling theorem [ th : existence ] , it is only possible via @xmath262 and @xmath423 .
this holds for any @xmath254 .
_ assume now that @xmath424 . according to step 1 , @xmath425 and thus @xmath426 .
because @xmath298 is non decreasing and according to doob s martingale convergence theorem , @xmath298 converges to @xmath427)$ ] on @xmath428 .
this implies that @xmath275 converges with @xmath111 to @xmath429 on @xmath428 . since @xmath430 ,
this convergence is improbable .
finally we prove that if @xmath431 , then the process reaches @xmath276 in finite time almost surely .
one can notice that proofs are very similar to those of subsections [ sec : west ] and [ sec : south ] .
[ prop : leave_r6_union_r7 ] if @xmath432 then @xmath433 .
define the sequence of regions @xmath434 through @xmath435 where @xmath436 is sufficiently small to have @xmath437 .
applying it to @xmath438 we find that for all @xmath439 @xmath440+\frac{1}{4}\frac{x^2}{\wt x - x}\sigma^2(y ) \r ] \le -\frac{x^2 \sigma^2(y)}{8(\wt x - x)^{3/2}}\\ & & \le -\frac{k^2 \sigma^2(1-k / n)}{8\sqrt{n}(n\wt x - k)^{3/2 } } < 0 \end{aligned}\ ] ] while @xmath441 in @xmath442 .
doob s supermartingale convergence theorem implies the existence of the pointwise limit @xmath443 almost surely , where we use the notation @xmath444 .
in addition , theorem [ th : reccurent domain ] guarantees that every set @xmath445 is exited in finite time almost surely . consequently if @xmath446 , we have that either @xmath447 or @xmath448 , a contradiction in either way . [ prop : r7_r8 ] if @xmath449 then @xmath450 .
the proof is identical to the one of proposition [ prop : r2_to_r4 ] , with small modifications . here
@xmath326 with @xmath451 and @xmath452 .
the process @xmath453 is a supermartingale on @xmath454 if we chose @xmath455 where @xmath456 are two positive constants given by @xmath457 \times [ \wt y , \we y ] } x \wt x \sigma^2(y ) $ ] and @xmath458 \times [ \wt y , \we y ] } y(x-\wt x ) \l[\kappa(x)-\gamma\r]-x(y-\wt y)\l[\phi(y)-\alpha\r ] \ ; . $ ] the justification is the following .
the domain @xmath459 contains the area of interest @xmath454 . using proposition [ prop : r2_to_r4 ]
, we can prove that @xmath460 is a supermartingale on @xmath461\x [ \we y , \wt y]$ ] .
on @xmath462\x [ \we y , \wt y]$ ] , @xmath463 - x(y - \we y)[\phi(y)-\alpha]\\ & + \sigma^2(y)\left[y(x-\we x ) - x ( y - \we y ) + \we x \left(y - x(1/c - \we x + \we x)\right)\right]\\ \le & \ y ( x - \we x)[\kappa(x)-\gamma ] - x(y - \we y)[\phi(y)-\alpha ] - x\we x \sigma^2(y)(1/c - \we \theta)\\ \le & \ m - m(1/c - \we \theta)\le 0 \;.\end{aligned}\ ] ] we then reproduce step 2 of the proof of proposition [ prop : r2_to_r4 ] , using propositions [ prop : r5_to_r7 ] and [ prop : leave_r6_union_r7 ] instead of propositions [ prop : r1_r2 ] and [ prop : leave r_2 and r_3 ] .
[ prop : r8_r1 ] if @xmath464 then @xmath465 .
we follow proposition [ prop : r4_to_r5 ] with the minor following modifications . * 1 * we consider @xmath466 the exit time of @xmath467 .
the process @xmath365 with @xmath468 verifies @xmath469 for some @xmath470 .
indeed @xmath471 and is null only if @xmath472 , whereas @xmath473 only if @xmath474 . applying theorem [ th : reccurent domain ] to @xmath475 , @xmath476 almost surely .
_ step 2 . _ if @xmath477 , then the process reaches @xmath261 in finite time almost surely according to proposition [ prop : r7_r8 ] .
we define the sequence @xmath478 with @xmath479 and @xmath480 proceeding as in step 2 proposition [ prop : r4_to_r5 ] , we obtain that @xmath481 implies that [ eq : limite proba ] _
_ define @xmath482 , which is strictly positive according to step 1 . consider the new process @xmath483 with @xmath484 .
it is a positive submartingale on @xmath485 $ ] , and similarly to step 3 of proposition [ prop : r4_to_r5 ] , we can obtain @xmath486 assuming , we have @xmath487 we then proceed exactly as in step 3 of proposition [ prop : r4_to_r5 ] to finish the proof .
in this section we assume that investment follows say s law and philips curve is provided by @xcite . [
ass : form of functions ] we let @xmath488 and @xmath489 .
assumption [ ass : non - linear goodwin ] holds under assumption [ ass : form of functions ] . the unique non - hyperbolic equlibrium point in @xmath490
is given by @xmath491 functions of definition [ def : lyapunov ] are given by @xmath492 with @xmath493 although period @xmath72 given by theorem [ th : periods_goodwin ] is not explicit here , numerical computations allow to approximate it with a linear function of @xmath494 , see first part of fig .
[ fig : linear_rel ] .
following remark [ rem : period ] , @xmath72 does not converge to @xmath179 with solutions of concentrating to @xmath125 .
a local phase portrait with values of @xmath72 is provided in second part of fig .
[ fig : linear_rel ] . [ ass : form of vol ] let @xmath495 \mapsto \sigma_0 ( 1-y)$ ] with @xmath41 .
if we implicitly assume that the perturbation of the average growth rate @xmath4 of the productivity is due to the flow of workers coming in and out of the fraction @xmath3 employed at time @xmath111 , assumption [ ass : form of vol ] conveniently expresses that this perturbation decreases with the employment rate since higher employment implies lower perturbation on the constant average rate @xmath496 .
other models can of course be considered .
assumption [ ass : form of vol ] together with assumption [ ass : form of functions ] , and comparing with , satisfy assumption [ ass : growth conditions ] . indeed for all @xmath154 , @xmath497 and along with the sub - linearity of the log function , @xmath498 in line with assumptions [ ass : non - linear goodwin ] and [ ass : growth conditions ] the vertical asymptote at @xmath474 implies that @xmath499 for some @xmath500 . under assumption
[ ass : growth conditions ] and following and , @xmath501 and @xmath502 .
assumption [ ass : form of vol ] also implies that @xmath503 is the root of a quadratic polynomial @xmath504 the latter shall have a unique root in @xmath505 to satisfy assumption [ ass : unique_root ] .
the following example of condition is sufficient .
[ ass : unique root ] we assume @xmath506 and @xmath507 .
we are now able to claim the existence of @xmath508 such that @xmath509 , where @xmath510 is defined in proposition [ prop : estimate ] .
a direct application provides @xmath511 using , this estimate becomes @xmath512 and @xmath513 is an explicitly calculable constant .
following the same procedure with , @xmath514 with the same @xmath515 and @xmath516 .
now choosing @xmath517 for some @xmath518 , so that @xmath519 , proposition [ prop : estimate ] provides @xmath520 theorem [ th : finite period ] is a straightly observable phenomenon with simulations , see fig .
[ fig : panel_goodwin_stochastic_examples ] . under the assumptions of this section ,
the system has been simulated using xppaut with a fourth order runge - kutta scheme for the deterministic part , and an euler scheme for the brownian part .
[ fig : panel_goodwin_stochastic_examples ] illustrates the effect of the volatility level @xmath521 on trajectories of the system , as for the economic quantity @xmath522 . of subsample paths of trajectories for with different values of volatility @xmath521 , starting from the green star and stopping at the red start .
right column : evolution of output @xmath26 over time for the subsample path.,scaledwidth=90.0%,height=415 ] apart from specific subregions of @xmath0 as @xmath276 or @xmath323 where corollary 3.2 in @xcite can provide an estimate for the expectation of the exit time , a bound for the expected period @xmath523 seems out of reach .
numerical simulations have nevertheless always provide reasonnable finite periods of stochastic orbits of .
we thus expect that @xmath523 is finite for a wide range of values of @xmath21 .
let us start with @xmath524 and reformulate @xmath226 of definition [ def : stochastic orbit ] as the time the process crosses the line @xmath525 for the second time .
this is equivalent to take @xmath526 . resorting to numerical methods , we have simulated the system @xmath527 times for @xmath528 different starting points in @xmath529 and recorded the position at the time when this line is crossed the second time , that is the positions after a full loop . fig .
[ fig : goodwin_stochastic_expected_loop ] contains such examination for an array of values of @xmath521 .
the expected time @xmath523 to complete a full - loop is also illustrated . as observed ,
there seems to be a stable attractive fixed point to @xmath530 for sufficiently large values of @xmath521 . if the starting point is picked too close to @xmath125 , the expected crossing value after one loop is further away from it . on the other hand ,
if the one starts extremely far away from @xmath125 , say with @xmath531 , then the expected value after on loop is higher .
this implies that after many loops , the expectation converges , and so does @xmath523 with the number of loops around @xmath183 . assuming that @xmath532 for enough initial points , theorem [ th : reccurent domain ] can be used with @xmath50 at points @xmath48 and @xmath183 to prove the following conjecture .
[ prop : fixed_point ] consider the function @xmath533 such that @xmath275 is a solution to with @xmath534 , and @xmath226 is the finite stopping time defined by theorem [ th : finite period ]
. then @xmath535 has at least one fixed point in @xmath536 .
after one full loop @xmath537 ( left ) , and expected elapsed time @xmath72 ( right ) .
computation performed in matlab , with @xmath527 simulations for every value single one of the @xmath528 initial values taken along the line @xmath525.,height=415 ]
this contribution attempts to draw the attention of dynamical system analysis onto macroeconomic models . before looking into complex models of finance and crises , e.g. @xcite
, we focus here on a brownian perturbation added into a non - linear version of the lotka - volterra system used in economics , the goodwin model . to begin with
, we recall the usual results for the deterministic planar oscillator : we provide the constant entropy function and describe the period of the closed orbits drawned by the system .
we then provide sufficient conditions for the stochastically perturbed system to stay in the meaningful domain @xmath0 which is a a bounded subset of @xmath133 for the @xmath23-component .
the entropy function is actually of great use for the last result , additionally to prior estimates on variations of the system .
we finally prove what seems a fundamental and staightforward property of the system , namely that a solution @xmath45 rotates with perturbations around a unique point @xmath183 .
the definition of stochastic orbits provided here conventienly suits the intuition of how the deterministic concept can be extended .
however it has clearly not the ambition to be a definitive concept and further investigations might confirm its usefulness or its precarity .
the proof exploits the concept of reccurent domains in an intensive manner .
we expect that economists seek interest in , as other perturbed macroeconomic systems ( e.g. @xcite ) , for the possibility to adjust the model to observed past data ( e.g. @xcite and @xcite ) and find a possible synthetic explanation for perturbations of business cycles ( see @xcite ) .
both authors want to thank matheus grasselli for leading ideas and presentation suggestion .
remaining errors are authors responsibility . | this paper examines the cycling behavior of a deterministic and a stochastic version of the economic interpretation of the lotka - volterra model , the goodwin model .
we provide a characterization of orbits in the deterministic highly non - linear model .
we then study the cycling behavior for a stochastic version , where a brownian noise is introduced via an heterogeneous productivity factor .
sufficient conditions for existence of the system are provided .
we prove that the system produces cycles around an equilibrium point in finite time for general volatility levels , using stochastic lyapunov techniques for recurent domains .
numerical insights are provided . * keywords * : lotka - volterra model ; goodwin model ; brownian motion ; random perturbation ; business cycles ; stochastic lyapunov techniques . |
let @xmath1 be the set of numbers normal in base @xmath2 and let @xmath3 be a function from @xmath4 to @xmath4 .
we say that @xmath3 _ preserves @xmath2-normality _ if @xmath5 .
we can make a similar definition for preserving normality with respect to continued fraction expansions , @xmath6-expansions , the lroth series expansion , etc .
several authors have studied @xmath2-normality preserving functions .
some @xmath2-normality preserving functions naturally arise in h. furstenberg s work on disjointness in ergodic theory@xcite . v.
n. agafonov @xcite , t. kamae @xcite , t. kamae and b. weiss @xcite , and w. merkle and j. reimann @xcite studied @xmath2-normality preserving selection rules . for a real number @xmath7 ,
define real functions @xmath8 and @xmath9 by @xmath10 and @xmath11 . in 1949
d. d. wall proved in his ph.d .
thesis @xcite that for non - zero rational @xmath7 the function @xmath8 is @xmath2-normality preserving for all @xmath2 and that the function @xmath9 is @xmath2-normality preserving functions for all @xmath2 whenever @xmath7 is rational .
these results were also independently proven by k. t. chang in 1976 @xcite .
d. d. wall s method relies on the well known characterization that a real number @xmath12 is normal in base @xmath2 if and only if the sequence @xmath13 is uniformly distributed mod @xmath14 that he also proved in his ph.d .
thesis .
d. doty , j. h. lutz , and s. nandakumar took a substantially different approach from d. d. wall and strengthened his result .
they proved in @xcite that for every real number @xmath12 and every non - zero rational number @xmath7 the @xmath2-ary expansions of @xmath15 and @xmath16 all have the same finite - state dimension and the same finite - state strong dimension .
it follows that @xmath8 and @xmath9 preserve @xmath2-normality .
it should be noted that their proof uses different methods from those used by d. d. wall and is unlikely to be proven using similar machinery . c. aistleitner generalized d. d. wall s result on @xmath9 .
suppose that @xmath17 is a rational number and that the digits of the @xmath2-ary expansion of @xmath18 are non - zero on a set of indices of density zero .
in @xcite he proved that the function @xmath19 is @xmath2-normality preserving .
it was shown in @xcite that c. aistleitner s result does not generalize to at least one notion of normality for some of the cantor series expansions .
there are still many open questions relating to the functions @xmath8 and @xmath9 .
for example , m. mends france asked in @xcite if the function @xmath8 preserves simple normality with respect to the regular continued fraction for every non - zero rational @xmath7 .
the authors are unaware of any theorems that state that either @xmath8 or @xmath9 preserve any other form of normality than @xmath2-normality . in this paper
we will be interested in the function @xmath20 for @xmath21 and @xmath22 , and how this function preserves certain notions of normality of @xmath0-cantor series expansions , namely @xmath0-normality and @xmath0-distribution normality .
( we will provide definitions for all these terms in section [ sec : cantor ] . ) in theorem [ thm : main ] , we will show that there exists a basic sequence @xmath0 and a real number @xmath12 such that @xmath23 is always @xmath0-normal and always _ not _
@xmath0-distribution normal ; in fact , we will show that for this @xmath0 , the set of @xmath12 with this property is big in the sense that it has full hausdorff dimension .
it was first shown in @xcite that the set of numbers that are @xmath0-normal but not @xmath0-distribution normal is non - empty for some basic sequences @xmath0 , but no indication was given to the size of this set .
for a specific basic sequence @xmath0 , we show that there exists a subset @xmath24 of the set of @xmath0-normal numbers that is invariant under @xmath25 for every @xmath21 and @xmath22 ( i.e. @xmath26 ) and has full hausdorff dimension .
related questions for the cantor series expansions are studied in @xcite .
it is an interesting question to know how explicit this @xmath12 and @xmath0 are , so we bring in some definitions from recursion theory .
a real number @xmath12 is _ computable _ if there exists @xmath27 with @xmath28 and a total recursive function @xmath29 that calculates the digits of @xmath12 in base @xmath2 .
a sequence of real numbers @xmath30 is _ computable _ if there exists a total recursive function @xmath31 such that for all @xmath32 we have that @xmath33 .
m. w. sierpiski gave an example of an absolutely normal number that is not computable in @xcite .
the authors feel that examples such as m. w. sierpiski s are not fully explicit since they are not computable real numbers , unlike champernowne s number .
a. m. turing gave the first example of a computable absolutely normal number in an unpublished manuscript .
this paper may be found in his collected works @xcite .
see @xcite by v. becher , s. figueira , and r. picchi for further discussion . in theorem [ thm : computability ] we give a basic sequence @xmath0 and real number @xmath12 , with @xmath12 in the set discussed in theorem [ thm : main ] , that are fully explicit in the sense that they are computable as a sequence of integers and a real number , respectively . throughout this paper
we will use a number of standard asymptotic notations .
by @xmath34 we mean that there exists some real number @xmath35 such that @xmath36 .
by @xmath37 , we mean @xmath34 and @xmath38 . by @xmath39
, we mean that @xmath40 as @xmath41 .
the study of normal numbers and other statistical properties of real numbers with respect to large classes of cantor series expansions was first done by p. erds and a. rnyi in @xcite and @xcite and by a. rnyi in @xcite , @xcite , and @xcite and by p. turn in @xcite .
the @xmath0-cantor series expansions , first studied by g. cantor in @xcite , are a natural generalization of the @xmath2-ary expansions . to a larger class of numbers .
results along these lines may be found in the monograph of j. galambos @xcite .
] let @xmath42 .
if @xmath43 , then we say that @xmath0 is a _
basic sequence_. given a basic sequence @xmath44 , the _ @xmath0-cantor series expansion _ of a real number @xmath12 is the ( unique)-ary expansions . ]
expansion of the form @xmath45 where @xmath46 and @xmath47 is in @xmath48 for @xmath49 with @xmath50 infinitely often .
we abbreviate ( [ eq : cseries ] ) with the notation @xmath51 w.r.t . @xmath0 . a _
block _ is an ordered tuple of non - negative integers , a _ block of length @xmath52 _ is an ordered @xmath52-tuple of integers , and _ block of length @xmath52 in base @xmath2 _ is an ordered @xmath52-tuple of integers in @xmath53 .
let @xmath54 a. rnyi @xcite defined a real number @xmath12 to be _ normal _ with respect to @xmath0 if for all blocks @xmath55 of length @xmath14 , @xmath56 where @xmath57 is the number of occurences of the block @xmath55 in the sequence @xmath58 of the first @xmath59 digits in the @xmath0-cantor series expansion of @xmath12 . if @xmath60 for all @xmath59 and we restrict @xmath55 to consist of only digits less than @xmath2 , then ( [ eq : rnormal ] ) is equivalent to _ simple normality in base @xmath2 _ , but not equivalent to _ normality in base @xmath2_. a basic sequence @xmath0 is _ @xmath52-divergent _ if @xmath61 and _ fully divergent _ if @xmath0 is @xmath52-divergent for all @xmath52 . a basic sequence @xmath0 is _ infinite in limit _ if @xmath62 . [ definition:1.7 ]
a real number @xmath12 is _ @xmath0-normal of order @xmath52 _
if for all blocks @xmath55 of length @xmath52 , @xmath63 we let @xmath64 be the set of numbers that are @xmath0-normal of order @xmath52 .
the real number @xmath12 is _ @xmath0-normal _ if @xmath65 a real number @xmath12 is _
@xmath0-distribution normal _ if the sequence @xmath66 is uniformly distributed mod @xmath14 .
let @xmath67 be the set of @xmath0-distribution normal numbers .
it follows from a well known result of h. weyl @xcite that @xmath67 is a set of full lebesgue measure for every basic sequence @xmath0 .
we will need the following results of the second author @xcite later in this paper .
[ thm : measure ] suppose that @xmath0 is infinite in limit .
then @xmath64 ( resp .
@xmath68 ) is of full measure if and only if @xmath0 is @xmath52-divergent ( resp . fully divergent ) .
we note the following simple theorem .
[ thm : distributionnormal ] suppose that @xmath0 is infinite in limit .
then @xmath69 is @xmath0-distribution normal if and
only if the sequence @xmath70 is uniformly distributed modulo 1 . note that in base @xmath2 , where @xmath60 for all @xmath59 , the corresponding notions of @xmath0-normality and @xmath0-distribution normality are equivalent .
this equivalence is fundamental in the study of normality in base @xmath2 .
another definition of normality , @xmath0-ratio normality , has also been studied .
we do not introduce this notion here as this set contains the set of @xmath0-normal numbers and all results in this paper that hold for @xmath0-normal numbers also hold for @xmath0-ratio normal numbers .
the complete containment relation between the sets of these normal numbers and pair - wise intersections thereof is proven in @xcite .
the hausdorff dimensions of difference sets such as @xmath71 are computed in @xcite .
set @xmath72 our main results of this paper will be the following : [ thm : main ] there exists a basic sequence @xmath0 such that the hausdorff dimension of @xmath24 is @xmath14 .
[ thm : computability ] there exists a computable basic sequence @xmath0 and a computable real number @xmath12 in @xmath24 . in order to prove the main results of this paper
, we will want to understand how the digits of @xmath23 differ from the digits of @xmath12 , when @xmath12 takes a specific form .
we begin with some lemmas based on elementary calculations .
if @xmath73 is a rational number with @xmath74 , @xmath75 and @xmath76 for some @xmath77 , then @xmath12 has a finite @xmath0-cantor series expansion of the form @xmath78 alternately if @xmath12 is a real number in the interval @xmath79 , then @xmath12 has a @xmath0-cantor series expansion of the following form , @xmath80 so that @xmath81 for @xmath82 .
this allows us to prove a number of additional lemmas rather trivially .
suppose that @xmath83 .
if @xmath84 is rational with @xmath74 , @xmath75 and @xmath85 , then @xmath86 has a @xmath0-cantor series expansion of the form @xmath87 so that @xmath86 and @xmath12 differ only in their first @xmath88 digits .
[ lem : rationaladd ] suppose that @xmath0 has the property that for any integer @xmath59 there exists an integer @xmath89 such that @xmath90 . then for any rational number @xmath91 , the @xmath0-cantor series expansion of @xmath12 and of @xmath86 differ on at most finitely many places .
[ lem : rationalmult ] suppose that @xmath12 has a finite @xmath0-cantor series expansion of the form @xmath92 we write @xmath93 @xmath94 so that @xmath95 suppose @xmath7 is a nonzero rational number . if @xmath96 is an integer and @xmath97 , then @xmath98 has a finite @xmath0-cantor series expansion of the form @xmath99
given basic sequences @xmath100 and @xmath101 , sequences of non - negative integers @xmath102 and @xmath103 , and a sequence of sets @xmath104 such that @xmath105 , define the set @xmath106 as follows .
let @xmath107 be the following basic sequence : @xmath108^{s_1 } [ \beta_1]^{t_1}\right]^{\upsilon_1 } \left[[\alpha_2]^{s_2 } [ \beta_2]^{t_2}\right]^{\upsilon_2 } \left[[\alpha_3]^{s_3 } [ \beta_3]^{t_3}\right]^{\upsilon_3 } \cdots.\ ] ] define the function @xmath109 set @xmath110 where @xmath111 and @xmath112 and let the functions @xmath113 , @xmath114 , and @xmath115 be such that @xmath116 .
note this is possible since @xmath117 is a bijection from @xmath118 to @xmath119 .
define the function @xmath120 we consider the condition on @xmath59 @xmath121 define the intervals @xmath122 that is , we choose digits from @xmath123 in positions corresponding to the bases obtained from the sequence @xmath6 and choose a specific digit from @xmath124 for the bases obtained from the sequence @xmath125 .
set @xmath126 we will need the following lemma from @xcite .
[ hdt ] suppose that basic sequences @xmath100 and @xmath101 , sequences of non - zero integers @xmath102 and @xmath103 , and a sequence of sets @xmath104 such that @xmath127 are given where @xmath128 and @xmath129 then @xmath130 provided this limit exists .
let @xmath0 be the base vandehey constructed .
then @xmath131 let @xmath132 be ( one of the numbers constructed by vandehey in some section ) .
let @xmath133 , @xmath134 , @xmath135 , @xmath136 , @xmath137 and @xmath138 . by lemma [ lemma : hdt ] , we know that @xmath139 . but by ( vandehey s main theorem ) , we know that @xmath140 so @xmath141
given integers @xmath143 , let @xmath144 denote the number of blocks of length @xmath59 in base @xmath2 containing exactly @xmath52 copies of a given digit .
( by symmetry it does not matter which digit we are interested in . )
[ lem : bugeaud ] let @xmath145 and @xmath146 be integers .
for every real number @xmath147 with @xmath148 , we have @xmath149 [ lem : k1 ] let @xmath145 and @xmath150 be integers .
for every real number @xmath147 with @xmath151 , we have @xmath152 note that @xmath153 is increasing as a function of @xmath59 , therefore @xmath154 now let @xmath155 and note that @xmath156 likewise one can show that @xmath157 as a result , we have that @xmath158 we now can apply lemma [ lem : bugeaud ] to see that @xmath159 as desired . here
we made use of the fact that @xmath160 .
we will say a block @xmath55 of length @xmath59 in base @xmath2 is @xmath142-normal ( with respect to @xmath2 ) , if the total number of occurrences in @xmath55 of any subblock of length @xmath52 in base @xmath2 is between @xmath161 and @xmath162 .
let @xmath163 denote the number of blocks of length @xmath59 that are _ not _ @xmath142-normal with respect to @xmath2 .
note that lemma [ lem : k1 ] gives a bound on @xmath164 .
the following lemma will give a bound on @xmath163 .
[ lem : epsilonk ] suppose @xmath145 , @xmath165 , @xmath166 are integers .
for every real number @xmath147 with @xmath167 we have @xmath168 let us begin by considering an arbitrary block @xmath169 $ ] of @xmath59 digits in base @xmath2 .
suppose that @xmath170 for some @xmath171 .
let @xmath172 for @xmath173 .
note that @xmath174 . for @xmath175 , let @xmath176
$ ] if @xmath177 and @xmath178 $ ] otherwise . by the pigeon - hole principle , if @xmath55 is not @xmath142-normal with respect to @xmath2 ,
then some @xmath179 is not @xmath180-normal with respect to @xmath181 .
thus , the total number of blocks @xmath55 which are not @xmath142-normal with respect to @xmath2 is at most a sum over @xmath182 of the number of blocks @xmath179 which are not @xmath180-normal with respect to @xmath181 , times either @xmath183 or @xmath184 to account for all possibilities of those digits of @xmath55 which are not contained in @xmath179 .
thus , by lemma [ lem : k1 ] , we have @xmath185 where here again we use that @xmath186 .
given @xmath187 , consider the following definitions .
we let @xmath188 , @xmath189 . with these definitions , we have that the number of @xmath190-normal blocks of @xmath191 digits in base @xmath182 is bounded by @xmath192 provided that @xmath182 is sufficiently large compared to @xmath52 .
when @xmath193 , we shall let @xmath194 . given a block @xmath195 $ ] of @xmath191 in base @xmath182 , let @xmath196 be the naturally associated integer .
let @xmath197 denote the set of all such blocks @xmath55 such that @xmath198 and @xmath199 .
note that @xmath197 always contains the block @xmath200 $ ] .
we denote the size of @xmath197 by @xmath201 , and note that @xmath202 for sufficiently large @xmath182 .
we will let @xmath203 in the moran set construction given in section [ sec : hausdorff ] , let @xmath133 , @xmath134 , @xmath204 , and @xmath205 , with @xmath0 given by .
we shall also let @xmath206 with this definition , we have that @xmath207 tends to @xmath14 and that , as @xmath182 grows , all elements of @xmath208 become arbitrarily small compared to @xmath209 and are eventually divisible by any fixed integer . since @xmath210 , the smallest base in @xmath0 constructed this way is @xmath211 , so that @xmath0 really is a basic sequence . with these definitions ( and any appropriate choice of sequence @xmath212 ) , it is easy to check that all such points satisfy the conditions of theorem [ hdt ] , so that @xmath139 .
it therefore suffices to show that for some proper selection of @xmath124 , we have @xmath213 . to make this selection of @xmath124 ,
let @xmath214^{n_i}[(i!)^2]^{n_i}\right]^{\ell_i},\ ] ] so that we could alternately write @xmath0 as @xmath215^{l_2 } [ x_3]^{l_3 } [ x_4]^{l_4}\cdots.\ ] ] we shall then choose the digits of @xmath124 in such a way so that the digits corresponding to the the @xmath216th occurence of the bases @xmath217^{n_i}$ ] in each copy of @xmath218 are the @xmath216th string from @xmath197 ( when ordered lexicographically ) . with this definition of @xmath124 in mind ,
let @xmath12 be any point in @xmath106 , @xmath219 , and @xmath22 .
we will show that @xmath23 is @xmath0-normal but not @xmath0-distribution normal . by the construction of @xmath0 and corollary [ lem : rationaladd ] , we have for any rational number @xmath91 that the @xmath0-cantor series expansions of @xmath23 and @xmath98 differ on at most finitely many digits .
in addition , we have that @xmath220 for @xmath221 is small compared with @xmath222 and is divisible by @xmath223 , and each digit of @xmath208 is small compared with @xmath224 and is divisible by @xmath225 .
therefore , by lemma [ lem : rationalmult ] , we have that for any nonzero rational number @xmath7 , there will be a sufficiently large @xmath182 such that the digits of @xmath23 corresponding to the bases @xmath218 satisfy the following properties : * each block of digits corresponding to an appearance of @xmath226^{n_i}$ ] is unique .
* the digits corresponding to each appearance of @xmath209 are in the interval @xmath227 .
to see that @xmath23 is not in @xmath67 , we make use of theorem [ thm : distributionnormal ] .
we note that asymptotically half of the bases @xmath228 are of the form @xmath209 for some @xmath182 , and by the previous paragraph , we have that the corresponding digits @xmath47 are @xmath229 .
therefore the sequence @xmath70 is clearly not uniformly distributed modulo @xmath14 . to show that @xmath23 is in @xmath68 , we make use of the following lemma , whose proof is elementary .
[ lem : limitreduction ] let @xmath230 and @xmath231 be sequences of positive real numbers such that @xmath232 .
let @xmath233 be an increasing sequence of positive integers with @xmath234 and define @xmath235 and @xmath236 .
suppose that @xmath237 then @xmath238 let us denote the @xmath216th appearance of @xmath218 in the bases of @xmath0 by @xmath239 .
in particular , this will consist of the bases @xmath228 where @xmath59 falls into the following interval @xmath240:=\left [ \sum_{k=1}^{i-1 } 2 l_k \ell_k + 2(j-1)\ell_k+1 , \sum_{k=1}^{i-1}2l_k \ell_k + 2j \ell_k\right].\ ] ] let us write @xmath241 and let @xmath242 denote the number of occurrences of the block of digits @xmath55 in the @xmath0-cantor series expansion of @xmath23 with the first digit of the block occurring at the @xmath59th place , with @xmath243 $ ] . comparing these two definitions with the definition of @xmath0-normality in ( [ eq : rnormal ] ) , and using lemma [ lem : limitreduction ]
, we see that it suffices to show that @xmath244 as @xmath182 increases ( uniformly for any @xmath245 $ ] ) and that @xmath246 as @xmath182 increases .
to estimate the size of @xmath247 , we note that most of the contribution comes from the terms when @xmath248 .
there are precisely @xmath249 such terms .
if any of the @xmath17 s in the denominator of a term equals @xmath224 ( or , possibly @xmath250 ) , then the entire term is at most @xmath251 .
and there are precisely @xmath252 such summands .
therefore , @xmath253 where the @xmath254 is decreasing as @xmath182 increases and is uniform over @xmath245 $ ] . from this , we derive @xmath255 and therefore derives from comparing and and using the definition of @xmath256 . to estimate the size of @xmath242 ,
let us suppose that @xmath182 is sufficiently large so that the digits of @xmath55 are less than @xmath182 and so that all the digits of @xmath23 corresponding to the large bases @xmath224 are at least @xmath182 in size .
therefore @xmath55 will only occur in the digit strings corresponding to the small blocks @xmath217^{n_i}$ ] .
we know that there are @xmath257 such distinct digit strings and at most @xmath258 of them can not be @xmath190-normal .
therefore , we have @xmath259 as before , the @xmath254 here is decreasing as @xmath182 increases . comparing and gives and completes the proof
. we shall , in fact , prove the following , more explicit theorem .
the basic sequence @xmath0 given in is computable .
let @xmath260 @xmath261 be the real number from the set @xmath106 given in section [ sec : mainproof ] such that @xmath262 if ( [ eq : vncond ] ) holds ( that is , the digits corresponding to the bases @xmath224 will be @xmath223 ) .
the sequence @xmath263 is computable , so @xmath264 is a computable sequence .
we can create a turing machine that , given input @xmath182 , lexicographically enumerates all integers in @xmath265 $ ] .
moreover , we use two turing machines that , given input @xmath182 and the list of integers , check if each integer @xmath220 satisfies the conditions @xmath266 and @xmath267 since the order relation on integers and divisibility of integers are computable relations .
we can then create a turing machine that , given input @xmath182 , lexicographically enumerates the elements of @xmath197 .
another turing machine can be used to output the size of @xmath197 .
thus , @xmath268 is a computable sequence .
since @xmath269 and @xmath268 are computable sequences , the sequence @xmath270 is also computable .
furthermore , @xmath271 is also a computable sequence
. thus the sequences @xmath272 , @xmath273 , @xmath274 , @xmath275 , and @xmath276 are all computable sequences .
therefore we can create a turing machine @xmath277 to output the @xmath59th term of @xmath278 as follows .
first make a turing machine @xmath55 that on inputs @xmath182 and @xmath59 will output the @xmath59th base of @xmath218 as follows . determine the residue class of @xmath59 modulo @xmath279 .
if this residue is less than @xmath191 , return @xmath182 , otherwise return @xmath224 .
this computes the @xmath59th digit of @xmath218 .
finally , create the turing machine @xmath280 that on input @xmath59 determines the maximum @xmath182 such that @xmath281 and computes @xmath282 .
then define @xmath277 as the turing machine that on input @xmath59 computes @xmath283 .
thus , we have a turing machine the outputs the @xmath59th base of @xmath0 , so @xmath0 is a computable sequence . by an argument from the previous paragraphs , we have that there is a turing machine that on input @xmath182 lexicographically enumerates @xmath197
. we can construct a turing machine to compute the sequence @xmath284 as follows . use the turing machine
@xmath285 that on input @xmath59 outputs @xmath286 where @xmath287 and @xmath288 .
create a new turing machine @xmath289 that on input @xmath182 and @xmath59 does the following . if the residue class of @xmath59 modulo @xmath279 is greater than or equal to @xmath191 , output @xmath223 . otherwise , compute @xmath290 and return the @xmath291th digit of the @xmath18th element of @xmath197 . then the turing machine that on input @xmath59 , runs the @xmath285 on @xmath59 , and then runs @xmath289 on the output of the @xmath285 , computes the sequence @xmath284 .
since both @xmath284 and @xmath292 are computable sequences , the real number @xmath293 is computable .
the effect of the rational number @xmath91 on the set we constructed to prove theorem [ thm : main ] was negligible .
we specifically constructed @xmath0 so that the denominator of @xmath91 had to divide some @xmath228 , so addition by @xmath91 would never change more than a finite amount of digits by corollary [ lem : rationaladd ] , and thus had no impact on either @xmath0-normality or @xmath0-distribution normality ( or the lack thereof ) .
this suggests the following natural question .
if we were to restrict @xmath0 so that , say @xmath294 for any @xmath59 , then addition by @xmath295 would have to change an infinite number of digits .
are results similar to those given here possible for such @xmath0 ?
there exist some basic sequences @xmath0 where the set @xmath67 does not contain any computable real numbers .
see @xcite .
what assumptions on @xmath0 must we have to guarantee that there are computable real numbers in @xmath24 ? | we investigate how non - zero rational multiplication and rational addition affect normality with respect to @xmath0-cantor series expansions . in particular , we show that there exists a @xmath0 such that the set of real numbers which are @xmath0-normal but not @xmath0-distribution normal , and which still have this property when multiplied and added by rational numbers has full hausdorff dimension .
moreover , we give such a number that is explicit in the sense that it is computable . |
throughout this paper , all maps , map - germs and vector fields are of class @xmath0 unless otherwise stated .
a map - germ @xmath1 is called a _ legendrian map - germ _ if there exists a germ of unit vector field @xmath2 along @xmath3 such that the following 2 conditions hold , where the dot in the center stands for the scalar product of two vectors .
1 . @xmath4 @xmath5 @xmath6 .
the map - germ @xmath7 defined by @xmath8 is non - singular , where @xmath9 is the unit tangent bundle of @xmath10 .
the vector field @xmath2 , the map - germ @xmath11 and the image of a legendrian map - germ are called a _
unit normal vector field of _ @xmath3 , a _
legendrian lift of _ @xmath3 and a _ wave front _ respectively .
singularities of legendrian map - germs have been relatively well - studied ( for instance , see @xcite ) .
let @xmath12 ( @xmath13 ) be a non - singular plane curve without inflection point ( namely , a non - degenerate curve ) , and let @xmath14 be a point of @xmath15 .
then , the _ pedal curve of @xmath16 relative to the pedal point @xmath14 _ is defined as the trajectory of the foot of perpendicular to the tangent line @xmath17 at @xmath18 from @xmath14 , and it is denoted by @xmath19 .
the given point @xmath14 is called the
_ pedal point_. let @xmath20 be the solution curve of @xmath21 where @xmath22 is the arc - length parameter of @xmath16 .
then , by definition of pedal curve , @xmath23 ( which is the unit tangent vector to @xmath16 at @xmath18 ) is a unit normal vector to @xmath24 at @xmath25 .
thus , the legendrian lift @xmath26 given by @xmath27 is well - defined .
since the original curve @xmath16 is without inflection point , by the serret - frenet formula ( for the serret - frenet formula , see for instance @xcite ) , @xmath28 is non - singular .
thus , the image of @xmath24 must be a wave front curve .
next , we move the pedal point @xmath14 .
let @xmath29 be a map , where @xmath30 is an open neighborhood of the origin of @xmath31 , then , we obtain two corank one maps @xmath32 and @xmath33 defined by @xmath34 respectively .
the map @xmath35 ( resp . , @xmath36 ) is called the _ pedal unfolding of _ @xmath37 ( resp . , the _ wave front unfolding of _
@xmath38 ) . in @xcite ,
the evolution of generic wave fronts by time has been studied .
we want to construct a different method from @xcite to study generic wave front evolutions . in order to do so , we pay attention to the relation between @xmath35 and @xmath36 since we have the following proposition [ proposition 1 ] for @xmath35 .
[ proposition 1 ] let @xmath12 @xmath39 be a non - singular plane curve without inflection point such that @xmath40 and let @xmath30 be an open neighborhood of the origin of @xmath31 .
moreover , we let @xmath29 be a map such that @xmath41 . then
, the map - germ @xmath42 is @xmath43-equivalent to the normal form of ( whitney umbrella)@xmath44 if and only if the origin @xmath45 of @xmath46 is a regular point of the map @xmath47 defined by @xmath48 . here
, two map - germs @xmath49 are said to be @xmath43-_equivalent _ if there exist germs of diffeomorphism @xmath50 and @xmath51 such that @xmath52 , and the _ normal form of _ ( _ whitney umbrella_)@xmath44 is the map - germ defined by @xmath53 where @xmath54 . proposition [ proposition 1 ] in the case @xmath55 is a special case of theorem 1 in @xcite , proof of proposition [ proposition 1 ] is given in 2 . by figures
[ figure 1 ] and [ figure 2 ] , it is easily conjectured that the pedal evolution @xmath35 is @xmath43-equivalent to the normal form of ( whitney umbrella)@xmath44 if and only if the wave front evolution @xmath36 is a ( swallowtail)@xmath56 , where a ( _ swallowtail_)@xmath44 is a map - germ @xmath43-equivalent to @xmath57 @xmath58 ; and in the case @xmath55 this conjecture has been actually proved in @xcite ( such a correspondence is called the _ calculus correspondence_. for more details on the known calculus correspondence , see 2 ) . in this paper , we first extend the known results on the calculus correspondence to evolutions with multi - parameters , and then give applications of calculus correspondence .
moreover , we discuss the possibility of generalization of calculus correspondence to degenerate pedal evolutions and degenerate wave front evolutions . in section [ section 2 ] , known calculus correspondences are extended to evolutions with multi - parameters .
the proof of proposition [ proposition 1 ] is also given in section [ section 2 ] . in section [ section 3 ] , applications of calculus correspondence are given . finally , the possibility of generalization of calculus correspondence is discussed in section [ section 5 ] .
[ definition 1 ] a map - germ @xmath59 having the following form is said to be _ of pedal unfolding type_. @xmath60 where @xmath61 is a function - germ satisfying @xmath62 , @xmath63 is a function - germ and @xmath64 . in the case
@xmath55 , definition [ definition 1 ] has been given in @xcite .
[ proposition 2 ] let @xmath12 @xmath39 be a non - singular plane curve without inflection point such that @xmath40 and let @xmath29 be a map such that @xmath65 , where @xmath30 is an open neighborhood of the origin of @xmath31 . then , @xmath35 is @xmath43-equivalent to a map - germ of pedal unfolding type .
_ proof of proposition [ proposition 2 ] . _
since @xmath12 @xmath39 is non - singular , we may assume that @xmath66 , @xmath67 near @xmath68 .
put @xmath69 .
then , @xmath70 is a normal vector to @xmath16 at @xmath71 .
since @xmath16 is without inflection point , we have that @xmath72 .
then , since @xmath41 , @xmath73 has the form : @xmath74 therefore , we have : @xmath75 @xmath76 [ proposition 3 ] let @xmath12 @xmath39 be a non - singular plane curve without inflection point such that @xmath40 and let @xmath29 be a map such that @xmath65 , where @xmath30 is an open neighborhood of the origin of @xmath31 .
then , @xmath35 is @xmath43-equivalent to a map - germ of the form : @xmath77 .
_ proof of proposition [ proposition 3 ] .
_ as same as the proof of proposition [ proposition 2 ] , we may assume that @xmath66 , @xmath67 near @xmath68 .
by definition of pedal curve , the following holds : @xmath78 here , @xmath79 and @xmath80 are functions defined in the proof of proposition [ proposition 2 ] .
thus , we have the following locally : @xmath81 since @xmath16 is without inflection point and @xmath69 , there exists a function @xmath82 such that @xmath83 and @xmath84 by hadamard s lemma ( for hadamard s lemma , see @xcite ) .
thus , @xmath85 is a morse function - germ . by the morse lemma with parameters ( see @xcite ) , we have that @xmath86 is @xmath43-equivalent to a map - germ of the form : @xmath87 since @xmath88 , by using the malgrange preparation theorem ( for the malgrange preparation theorem , for instance see @xcite ) , @xmath89 is @xmath43-equivalent to @xmath90 @xmath76 note that proposition [ proposition 3 ] may be proved without using the malgrange preparation theorem . alternatively , we may adopt a simple method used to prove the criterion of cuspidal crosscap given in @xcite .
namely , by dividing @xmath91 into the sum of an odd function and an even function with respect to the variable @xmath92 , it is possible to show that the map - germ ( [ equation 1 ] ) is @xmath43-equivalent to the map - germ ( [ equation 2 ] ) .
[ definition s_k ] let @xmath93 be a non - negative integer . then
, a map - germ @xmath94 is said to be _ of @xmath95 type _
if @xmath96 is @xmath43-equivalent to the map - germ @xmath97 . by proposition [ proposition 3 ] , singularities of one - parameter pedal unfoldings
@xmath35 must be of @xmath95 type ( @xmath93 is a certain non - zero integer ) provided that @xmath16 is non - degenerate , @xmath98 and @xmath99 is not flat .
this fact has been already proved in @xcite by using a characterization of spherical pedal given in @xcite .
thus , the proof given here is an alternative proof .
even if we moved the given non - degenerate curve @xmath16 depending on the parameter @xmath100 , the proof of proposition [ proposition 3 ] shows that new singularities never occur for the map - germ of the form @xmath101 provided that @xmath102 is non - degenerate , @xmath103 and @xmath99 is not flat .
_ _ proof of proposition [ proposition 1 ] . _
_ used mather s infinitesimal characterization of stable map - germs ( @xcite ) ) and mather s classification theorem ( @xcite ) . the proof given here , which is self - contained , was suggested by the referee . ] by proposition [ proposition 3 ] , @xmath104 is @xmath43-equivalent to a map - germ @xmath105 under the assumption of proposition [ proposition 1 ] .
it is easily seen that the origin @xmath45 of @xmath46 is a regular point of the map @xmath47 if and only if there exists an integer @xmath106 @xmath107 such that @xmath108 for @xmath99 .
thus , it is sufficient to show that @xmath105 is a ( whitney umbrella)@xmath44 if and only if there exists an integer @xmath106 @xmath107 such that @xmath108 for @xmath99 .
suppose that @xmath109 is a ( whitney umbrella)@xmath44 . let @xmath110 be the set of corank one @xmath111-jets .
note that @xmath112 is transverse to @xmath113 where @xmath96 denotes the normal form of ( whitney umbrella)@xmath44 . since @xmath109 is @xmath43-equivalent to @xmath96 , @xmath114 , too , is transverse to @xmath113 .
since @xmath115 , @xmath114 is transverse to @xmath113 if and only if @xmath116 therefore , there exists an integer @xmath106 @xmath107 such that @xmath108 .
conversely , suppose that there exists an integer @xmath106 @xmath107 such that @xmath108 .
set @xmath117 then , @xmath118 ( resp . , @xmath119 )
is a germ of diffeomorphism of @xmath120 ( resp . , @xmath121 ) .
set also @xmath122 .
then , we have : @xmath123 since @xmath108 , the map - germ @xmath124 is a germ of diffeomorphism .
thus , @xmath125 is @xmath43-equivalent to @xmath109 .
since the map - germ @xmath126 is clearly a ( whitney umbrella)@xmath44 , @xmath109 must be a ( whitney umbrella)@xmath44 .
@xmath76 [ definition 2 ] for a map - germ of pedal unfolding type @xmath127 set @xmath128 the map - germ @xmath129 is called the _ integration of _ @xmath130 . in the case
@xmath55 , definition [ definition 2 ] has been given in @xcite .
[ normalized ] a legendrian map - germ @xmath131 is said to be _ normalized _ if @xmath3 satisfies the following three conditions : 1 .
the map - germ @xmath3 has the form @xmath132 where @xmath64 .
the condition @xmath133 holds .
3 . the vector @xmath134 is @xmath135 or @xmath136 , where @xmath137 denotes the standard coordinate system of @xmath121 . in the case
@xmath55 , definition [ normalized ] has been given in @xcite .
[ definition 5 ] for a normalized legendrian map - germ @xmath138 set @xmath139 the map - germ @xmath140 is called the _ differential of _ @xmath3 . in the case
@xmath55 , definition [ definition 5 ] has been given in @xcite .
[ proposition 4 ] 1 . for a map - germ of pedal unfolding type @xmath141
, @xmath142 is a normalized legendrian map - germ .
2 . for a normalized legendrian map - germ @xmath143 ,
@xmath144 is a map - germ of pedal unfolding type . in the case
@xmath55 , proposition [ proposition 4 ] with its proof can be found in @xcite .
the proof given in @xcite works well even to the case @xmath145 .
the following set is denoted by @xmath146 . @xmath147 and set also @xmath148 where a map - germ @xmath149 is called a ( _ cusp_)@xmath150 if it is @xmath43-equivalent to @xmath151 @xmath152 .
the following theorems [ theorem 1 ] and [ theorem 2 ] are extensions of known calculus correspondences to multi - parameters .
[ theorem 1 ] 1 .
the map @xmath153 defined by @xmath154 is well - defined and bijective .
the map @xmath155 defined by @xmath156 is well - defined and bijective .
[ theorem 2 ] 1 . the map @xmath157 defined by @xmath158 is well - defined and bijective .
the map @xmath159 defined by @xmath160 is well - defined and bijective . in the case
@xmath55 , the proofs of theorems [ theorem 1 ] and [ theorem 2 ] can be found in @xcite . for the proof of theorem [ theorem 1 ] in the case @xmath55 ,
two criteria ( theorems [ theorem 3 ] and [ theorem 4 ] ) have been used in @xcite .
theorem [ theorem 4 ] works well even in the case @xmath145 . although it is uncertain that theorem [ theorem 3 ] works well even in the case @xmath145 , since a ( whitney umbrella)@xmath44 is stable , by mather s infinitesimal characterization of stable map - germs , theorem [ theorem 1 ] in general case can be proved . on the other hand , for the proof of theorem [ theorem 2 ] in the case @xmath55 , theorem
[ theorem 3 ] has not been used in @xcite though theorem [ theorem 4 ] has been used .
hence the proof of theorem [ theorem 2 ] works well even in the case @xmath145 . besides theorems [ theorem 1 ] and [ theorem 2 ] ,
there is one more example of calculus correspondence ( proposition [ proposition 5 ] ) .
since proposition [ proposition 5 ] is almost trivial , its proof is omitted .
put @xmath161 [ proposition 5 ] 1 . the map @xmath162 defined by @xmath163 is well - defined and bijective .
the map @xmath164 defined by @xmath165 is well - defined and bijective .
let @xmath166 be the linear transformation of the form @xmath167 .
two function germs @xmath168 are said to be @xmath169-_equivalent _ if there exists a germ of diffeomorphism @xmath170 having the form @xmath171 and a function - germ @xmath172 having the form @xmath173 , @xmath174 such that @xmath175 . [ theorem 3 ] two map - germs @xmath176 @xmath177 of the following form @xmath178 where @xmath179 and @xmath180 is not flat for each @xmath181 , are @xmath43-equivalent if and only if the function - germs @xmath182 are @xmath169-equivalent .
[ definition 3 ] let @xmath183 be a legendrian map - germ and let @xmath2 be a unit normal vector field of @xmath3 given in the definition of legendrian map - germs .
the function - germ @xmath184 defined by the following is called the _ legendrian - jacobian _ of @xmath3 where @xmath185 .
@xmath186 in the case @xmath55 , definition [ definition 3 ] can be found in @xcite .
note that if @xmath2 satisfies the conditions of unit normal vector field of @xmath3 , then @xmath187 also satisfies them .
thus , the sign of @xmath188 depends on the particular choice of unit normal vector field @xmath2 .
the legendrian jacobian of @xmath3 is called also the _ signed area density function _ ( for instance , see @xcite ) .
although it seems reasonable to call @xmath189 the area density function from the viewpoint of investigating the singular surface @xmath190 ( @xmath30 is a sufficiently small neighborhood of the origin of @xmath15 ) , it seems reasonable to call it the legendrian jacobian from the viewpoint of investigating the singular map - germ @xmath3 .
[ theorem 4 ] let @xmath149 be a normalized legendrian map - germ , 1 .
@xmath3 is a ( swallowtail)@xmath44 if and only if the following two hold where @xmath191 : @xmath192 2 .
@xmath3 is a ( cusp)@xmath150 if and only if the following two hold : @xmath193 here , @xmath194 stands for mather s local algebra for function - germs @xmath195 . for mather s local algebra ,
see @xcite .
theorems [ theorem 3 ] ( resp . , theorem [ theorem 4 ] ) is used as a criterion of whitney umbrella ( resp .
, ( swallowtail)@xmath44 ) .
theorems [ theorem 3 ] and [ theorem 4 ] are connected by the following simple lemma .
[ lemma 1 ] for a normalized legendrian map - germ @xmath149 , @xmath196 here @xmath197 . in the case
@xmath55 , lemma [ lemma 1 ] with its proof can be found in @xcite .
the proof given in @xcite works well in general case .
in order to show that the calculus correspondence is significant and useful , we give two applications of theorem [ theorem 1 ] . [ application ] let @xmath149 be given by @xmath198 where @xmath64 .
then , the following two are equivalent . 1 .
the given @xmath3 is a ( swallowtail)@xmath44 , that is , it is @xmath43-equivalent to the normal form of ( swallowtail)@xmath44 which is the following @xmath199 2 .
the equality @xmath200 holds for any @xmath106 @xmath107 and there exists an @xmath106 @xmath107 such that @xmath201 is satisfied . _
proof of proposition [ application ] .
_ suppose that @xmath3 is a ( swallowtail)@xmath44 .
then , since @xmath3 is legendrian , there exists a unit normal vector field @xmath202 such that the following two hold : @xmath203 @xmath204 since @xmath3 is a ( swallowtail)@xmath44 , @xmath205 ( resp . , @xmath206 ) must be zero by the equality ( [ x ] ) ( resp . , ( [ y ] ) ) .
since @xmath207 is a unit vector , @xmath208 must be @xmath209 .
it is clear that the given @xmath3 satisfies the first and the second conditions of definition [ normalized ] .
thus , @xmath3 is a normalized ( swallowtail)@xmath44 . by theorem [ theorem 1 ]
, @xmath210 is a ( whitney umbrella)@xmath44 of pedal unfolding type . since @xmath144 is of pedal unfolding type , we have that @xmath200 for any @xmath106 @xmath107 .
since @xmath144 is a ( whitney umbrella)@xmath44 , there must exist an @xmath106 @xmath107 such that @xmath211 .
two conditions @xmath200 and @xmath211 imply that @xmath201 .
conversely , suppose that the equality @xmath200 holds for any @xmath106 @xmath107 and there exists an @xmath106 @xmath107 such that @xmath201 is satisfied .
then , @xmath144 is a ( whitney umbrella)@xmath44 of pedal unfolding .
thus , by theorem [ theorem 1 ] , @xmath212 is a normalized ( swallowtail)@xmath44 .
@xmath76 as another application of theorem [ theorem 1 ] , we give an alternative proof of arnold s observation given in @xcite ( for arnold s observation , see also @xcite ) , namely , we show the following : [ observation ] let @xmath213 be the space curve given by @xmath214 . then , the tangent developable of @xmath215 , which is the following , is a swallowtail .
@xmath216 _ proof of observation [ observation ] .
_ put @xmath217 . then , @xmath3 is @xmath218-equivalent to @xmath219 .
it is easily seen that @xmath220 is a whitney umbrella of pedal unfolding type .
thus , by theorem [ theorem 1 ] , @xmath221 is a normalized swallowtail . @xmath76
the following question is a multi - parameter version of the question posed in @xcite . 1 .
let @xmath222 be two map - germs of pedal unfolding type .
suppose that @xmath223 is @xmath43-equivalent to @xmath224 .
is @xmath225 necessarily @xmath43-equivalent to @xmath226 ?
2 . let @xmath227 be two normalized legendrian map - germs .
suppose that @xmath228 is @xmath43-equivalent to @xmath229 .
is @xmath230 necessarily @xmath43-equivalent to @xmath231 ? [ [ s_k - type - singularities - and - legendrian - s_k - type - singularities ] ] @xmath95 type singularities and legendrian @xmath95 type singularities ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ recall that a map - germ @xmath94 is of @xmath95 type if @xmath96 is @xmath43-equivalent to the map - germ @xmath232 ( definition [ definition s_k ] ) .
since the map - germ @xmath233 is of pedal unfolding type , the following map - germ ( which is @xmath234 ) is normalized legendrian map - germ by proposition [ proposition 4 ] .
@xmath235 the legendrian map - germ @xmath234 is called the _ normal form of legendrian @xmath95 type _ and any legendrian map - germ @xmath43-equivalent to @xmath234 is said to be _ of legendrian @xmath95 type_. 1 .
let @xmath236 be a map - germ of pedal unfolding type .
suppose that @xmath130 is of @xmath95 type .
is @xmath142 necessarily of legendrian @xmath95 type ? 2 .
let @xmath237 be a normalized legendrian map - germ .
suppose that @xmath3 is of legendrian @xmath95 type .
is @xmath144 necessarily of @xmath95 type ? in the case
@xmath238 , both @xmath239 are @xmath43-equivalent to the normal form of whitney umbrella , and both @xmath240 are @xmath43-equivalent to the normal form of swallowtail ( namely , the map - germ @xmath241 ) . in this case , we have the calculus correspondence by theorem [ theorem 1 ] . in the case
@xmath242 , @xmath243 ( resp . ,
@xmath244 ) is not @xmath43-equivalent to @xmath245 ( resp . , @xmath246 ) .
it is known that only the map - germs of @xmath113 type are @xmath247-codimension one singularities of mono - germs from the plane to the 3-space ( for @xmath247-codimension , see @xcite and for the classification of @xmath247-codimension one singularities @xmath248 , see @xcite ) .
theorem [ theorem 3 ] can be applied as a criterion of @xmath113 singularities . on the other hand ,
criteria of legendrian @xmath113 singularities have been obtained by izumiya - saji - takahashi ( @xcite ) .
thus , by replacing saji - umehara - yamada criterion ( theorem [ theorem 4 ] ) with izumiya - saji - takahashi criteria given in @xcite , the proof of theorem [ theorem 1 ] is expected to work well to show calculus correspondence between @xmath113 singularities of pedal unfolding type and normalized legendrian @xmath113 singularities .
next , we discuss the case @xmath249 . even in this case ,
theorem [ theorem 3 ] can be applied as a criterion of @xmath95 singularities .
however , there seems to be no criteria for legendrian @xmath95 singularities in the case @xmath249 .
hence , it seems that we can not expect an analogy of the proof of theorem [ theorem 1 ] . 1 .
the map - germ @xmath253 given by @xmath254 is called the _ normal form of legendrian @xmath255 type _ , where @xmath185 .
a map - germ @xmath256 is said to be _ of _ _ legendrian @xmath255 type _ if @xmath3 is @xmath43-equivalent to @xmath257 .
note that the image of @xmath257 is the envelope of the following one parameter family of hyperplanes . by this reason , @xmath257
is called the normal form of @xmath255 type .
@xmath258 for the normal form of legendrian @xmath255 type , we have @xmath259 where @xmath260 and @xmath261 . since , @xmath262 and @xmath62
, @xmath263 is of pedal unfolding type .
therefore , @xmath264 is normalized legendrian by proposition [ proposition 4 ] . 1 .
let @xmath265 be a map - germ of pedal unfolding type .
suppose that @xmath130 is @xmath43-equivalent to @xmath263 .
is @xmath142 necessarily of legendrian @xmath255 type ? 2 .
let @xmath266 be a normalized legendrian map - germ .
suppose that @xmath3 is of legendrian @xmath255 type .
is @xmath144 necessarily @xmath43-equivalent to @xmath263 ?
question [ question gaffney ] was asked by g. ishikawa ( @xcite ) , and independently by t. gaffney during ams spring western section meeting at the university of hawaii ( 2012 ) .
it is easily seen that @xmath267 is non - singular , @xmath268 is the normal form of ( cusp)@xmath150 and @xmath269 is the normal form of ( swallowtail)@xmath44 .
thus , in the case @xmath270 , proposition [ proposition 5 ] , theorem [ theorem 2 ] and theorem [ theorem 1 ] are the affirmative answers to question [ question gaffney ] respectively .
therefore , question [ question gaffney ] asks essentially the case @xmath271 . even in this case , there is a criterion of legendrian @xmath255 singularities ( theorem [ a_k criterion ] ) .
however , there seems to be no criteria for the @xmath43-equivalence class of @xmath263 @xmath272 .
hence , it seems that we can not expect an analogy of the proof of theorem [ theorem 1 ] .
99 [ bibliography ] arnold , v. i. , wave front evolution and equivariant morse lemma , _ comm .
pure appl math .
_ , * 29 * ( 1976 ) , 557582 .
arnold , v. i. , lagrangian manifolds with singularities , asymptotic rays , and the open swallowtail , _ funct . anal .
_ , 15 - 4 ( 1981 ) , 235246 .
arnold , v. i. , gusein - zade , s. m. , varchenko , a. n. , _ singularities of differentiable maps i. _ monographs in mathematics , * 82 * , birkhuser , boston basel stuttgart , 1985 .
bruce , j. w. and giblin , p. j. , _ curves and singularities ( second edition ) _ , cambridge university press , cambridge , 1992 .
chen , x. y. and matumoto , t. , on generic @xmath276-parameter families of @xmath0-maps of an @xmath277-manifold into a @xmath278-manifold , _ hiroshima math . j. _ ,
* 14 * ( 1984 ) , 547550 .
cooper , t. , mond , d. and wik atique , r. , vanishing topology of codimension 1 multi - germs over @xmath279 and @xmath280 , _ compositio math .
_ , * 131 * ( 2002 ) , 121160 .
fujimori , s. , saji , k. , umehara , m. and yamada , k. , singularities of maximal surfaces , _ math . z. _ ,
* 259 * ( 2008 ) , 827848 .
ishikawa , g. , singularities of developable surfaces , _ london mathematical society lecture notes series _ , * 263 * , pp . 403418 , 1999 , ishikawa , g. , private communications , 2011 .
izumiya , s. , on legendrian singularity theory , _ proc .
_ , * 101 * ( 1987 ) , 748752 .
izumiya , s. , saji , k and takahashi , m , horospherical flat surfaces in hyperbolic 3-space , _ j. math .
* 62 * ( 2010 ) , 789849 .
kokubu , m. , rossman , w. , saji , k. , umehara , m. and yamada , k. , singularities of flat fronts in hyperbolic 3-space , _ pacific j. math .
_ , * 221 * ( 2005 ) , 303351 .
mather , j. , stability of @xmath0 mappings , ii , infinitesimal stability implies stability , _ ann .
of math._. * 89*(1969 ) , 2593291 .
mather , j. , stability of @xmath0 mappings , iv , classification of stable map - germs by @xmath279-algebras , _ publ .
hautes tudes sci_. * 37*(1970 ) , 223248 .
mond , d. , on the classification of germs of maps from @xmath15 to @xmath281 , _ proc .
london math .
_ , * 50 * ( 1985 ) , 333369 .
nishimura , t. , normal forms for singularities of pedal curves produced by non - singular dual curve germs in @xmath282 , _ geom dedicata _ * 133 * ( 2008 ) , 5966 .
nishimura , t. , singularities of one - parameter pedal unfoldings of spherical pedal curves , _
j. singularities _ , * 2 * ( 2010 ) , 160169 .
nishimura , t. , whitney umbrellas and swallowtails , _ pacific j. math .
_ , * 252 * ( 2011 ) , 459471 .
saji , k. , umehara , m. and yamada , k. , @xmath250 singularities of wave fronts , _ math .
camb . philos .
_ , * 146 * ( 2009 ) , 731746 .
saji , k. , umehara , m. and yamada , k. , the geometry of fronts , _ ann . of math .
_ , * 169 * ( 2009 ) , 491529 . wall , c. t. c. , finite determinacy of smooth map - germs , _ bull .
london math .
_ , * 13 * ( 1981 ) , 481539 .
zakalyukin , v. m. , lagrangian and legendrian singularities , _ funct .
_ , * 10 * ( 1976 ) , 3745 .
zakalyukin , v. m. , reconstruction of legendrian maps and caustics depending on a parameter and versality of mappings , _ j. soviet math .
_ , * 27 * ( 1984 ) , 27132735 . | the calculus correspondence has been known to exist between generic pedal evolutions and generic wave front evolutions . in this paper , we first extend the known results on the calculus correspondence to evolutions with multi - parameters , and then give applications of calculus correspondence .
moreover , we discuss the possibility of generalization of the calculus correspondence to degenerate pedal evolutions and degenerate wave front evolutions . |
the primary scientific goal of the diffuse infrared background experiment ( dirbe ) aboard the _ cosmic background explorer ( cobe ) _ spacecraft is the measurement of the cosmic infrared background ( cib ) at wavelengths from 1.25 to 240 @xmath0 .
this radiation is the cumulative emission of pregalactic sources , protogalaxies , and evolving galaxies , as well as emission from more exotic processes not common in the local universe ( e.g. bond , carr , & hogan 1986 ) .
these sources of the cib may be found anywhere from the earliest epoch after radiation and matter were decoupled to the present day .
the contribution to the cib from galaxies will be composed of stellar emission from distant galaxies that is redshifted by the cosmological expansion from intrinsically shorter wavelengths , as well as the direct ir emission from stars and dust within galaxies at all distances .
the constraints that dirbe places on the cib are discussed by hauser et al .
( 1998 , hereafter paper i ) and dwek et al .
( 1998 , hereafter paper iv ) .
however , in order to detect the cib , we first need to remove the strong contributions of foreground emission arising within our own solar system and galaxy .
the ir foreground from within our solar system originates from the interplanetary dust ( ipd ) cloud . the modeling and removal of the scattering and emission from the ipd for the entire cold - mission dirbe data set
are reported by kelsall et al .
( 1998 , hereafter paper ii ) . following the removal of this foreground , we attack the next layer of foreground emission by modeling and subtracting the galactic ir emission . the near - ir ( 1.25 4.9 @xmath0 ) emission of the galaxy is dominated by starlight . some bright stars and other compact sources
are resolved as point sources in the dirbe data .
most stars blend into a diffuse background , showing a disk and bulge very similar in appearance to many edge - on spiral galaxies .
extinction effects are clearly present at the shortest near - ir wavelengths as a visible dark lane in the inner galaxy .
papers which have previously examined dirbe observations of the stellar disk and bulge of the galaxy are : weiland et al .
( 1994 ) , arendt et al .
( 1994 ) , freudenreich et al .
( 1994 ) , dwek et al .
( 1995 ) , calbet et al .
( 1996 ) , freudenreich ( 1996 ) , binney , gerhard , & spergel ( 1997 ) , bissantz et al .
( 1997 ) , porcel , c. , battaner , e. , & jimnez - vicente , j. ( 1997 ) , and fux ( 1997 ) .
the mid- and far - ir ( 12 100 and 140 240 @xmath0 ) emission is dominated by thermal emission from dust in the diffuse ism and in more compact star - forming regions .
previously published studies of the galactic ism based on dirbe data include : arendt et al .
( 1994 ) , freudenreich et al .
( 1994 ) , sodroski et al .
( 1994 , 1995 ) , bernard et al . (
1994 ) , boulanger et al .
( 1996 ) , dwek et al .
( 1997 ) , sodroski et al .
( 1997 ) , davies et al .
( 1997 ) , and lagache et al .
( 1998 ) .
this paper details the development of models of the galactic ir emission .
the primary intended use for the models is to permit an accurate measurement of an extragalactic ir background .
the characterization of the sources that give rise to the galactic emission is an important secondary result of the process .
an overview of the procedures used for modeling the galactic foreground is given in section 2 .
section 3 describes the dirbe data set and its preparation , particularly the removal of the ipd scattering and emission ( paper ii ) .
section 4 describes the modeling and removal of starlight from the 1.25 to 25 @xmath0 measurements .
the following section ( 5 ) describes in detail the modeling and removal of the ism emission from dirbe data .
additional investigation into the far - ir emissivity of the diffuse ionized ism is contained in the appendix .
the accuracy of the removal of the galactic emission is a major limitation to the detection of the cib , and consequently section 6 discusses estimates of the uncertainties of the data and procedures .
a brief discussion of the implications of this modeling for galactic properties is contained in section 7 .
a more detailed analysis of the galactic ism as revealed by this work has been reported by dwek et al .
finally , a brief summary of results is given in section 8 .
companion papers ( papers i & iv ) contain analyses of the residual emission , and its implications for the detection of the cib .
we will assume that at any time , @xmath1 , the total intensity observed at a given wavelength and a specified galactic longitude and latitude , @xmath2 , can be represented by : @xmath3 where @xmath4 is the contribution from the interplanetary dust cloud , @xmath5 is the contribution from stars ( resolved and unresolved ) and other compact sources within the galaxy , @xmath6 is the contribution from the interstellar medium , and @xmath7 is residual intensity obtained by subtracting the former three components from the total observed intensity .
the procedures we apply are designed to ensure that the cib is fully contained within this residual emission , i.e. , that no cib component is inadvertently included in the models of the ipd cloud or galactic emission components .
the contribution of the ipd to the infrared sky , @xmath4 , has been calculated in paper ii .
the ipd cloud is modeled using a geometric kernel similar to that developed to represent ipd emission in the _ iras _ data ( wheelock et al .
the model ensures the preservation of the cib intensity by fitting the time variation of the ipd emission rather than the intensity directly .
the model does not contain any explicit isotropic component which would produce no time variation .
details of this model are presented in paper ii . in the dirbe data ,
the stellar emission component of the galaxy , @xmath5 , consists of two components : the first is a discrete component , which includes all bright sources that exceed the local background level by a chosen amount , depending on wavelength ; and the second is a smooth component , consisting of unresolved stellar sources which can be represented by using a statistical model for calculating the brightness of the galaxy . rather than modeling and removing each of the bright sources at all wavelengths ,
the locations of these sources are simply blanked and excluded from any further processing or analysis .
sources excluded by this procedure consist of bright stars at short wavelengths , and diffuse sources such as the magellanic clouds and bright star - forming regions in the galactic plane ( e.g. the @xmath8 oph and orion regions ) , which appear at all wavelengths .
the procedure for constructing the bright source removed maps is described in detail in section 4.1 .
the unresolved stellar emission component is represented by a `` faint source model , '' which closely follows the `` sky '' model described by wainscoat et al .
( 1992 ) , and further developed by cohen ( 1993a , 1994a , 1995 ) .
the sky model of galactic structure was originally developed to fit ground - based @xmath9-band and @xmath10-band and _ iras _ 12 and 25 @xmath0 galactic source counts .
the fitting of source counts of known types of objects , instead of total intensity measurements , ensures that the model intensity represents that of the galactic sources , thereby excluding any possible unresolved extragalactic contribution . to guarantee that the modeled ism contribution to the sky brightness includes only galactic emission ,
we will write this emission component , @xmath6 , as a product of a spatially invariant spectral component @xmath11 times a wavelength - independent spatial template .
the spatial template , such as the galactic h i or co column density , is associated with the galactic gas phase component in which the dust resides .
the advantage of such a procedure is that the line emission from most extragalactic sources is redshifted to velocities higher than those at which galactic line emission is observed .
thus , the h i and co data can be used as templates of the local ( galactic ) ir emission of the ism , without any contributions from extragalactic sources . in principle
, @xmath11 can be obtained directly as the slope of a linear least squares fit to the correlation of @xmath12 [ @xmath13 with the h i or co spatial template for all wavelengths @xmath14 .
the intercept will then yield the value of @xmath7 .
however , the main limitations to such a direct approach are : that there can be relatively large spatial variations in @xmath11 , the ir emission per h atom ; and that high - quality h i and co data sets have restricted sky coverage or different spatial resolution compared to the dirbe data . these limitations can be circumvented by using the 100 @xmath0 ism emission @xmath15 @xmath16 as the spatial template for ism emission at the other wavelengths .
the subtraction of the interstellar component was therefore performed in the following manner .
first , @xmath11 and @xmath17 @xmath16 were obtained as the slope and intercept of the correlation between @xmath18 @xmath16 and the h i spatial template .
this step allows us to separate the ism emission , @xmath15 @xmath19 @xmath20 @xmath21 @xmath16 , from the potential cib , @xmath17 @xmath16 .
then , we chose @xmath15 @xmath16 as the spatial template , and calculated @xmath11 by correlating this template with @xmath22 @xmath16 , so that @xmath23 @xmath16 . for the near - ir wavelengths ( @xmath24 4.9 @xmath0 ) , the subtraction of the starlight is not sufficiently accurate to yield values of @xmath11 and @xmath7 from direct correlations of the near - ir data with @xmath15 @xmath16 .
however , the dust emission in the near - ir can be more easily distinguished from the stellar emission in color - color plots ( e.g. , arendt et al .
1994 ) , as points that are displaced from the reddening line for stellar emission .
this aspect of the subtraction of the ism - related near - ir emission is described in detail in section 5.5 .
the dirbe instrument provides absolutely calibrated intensity sky maps in 10 broad bands at 1.25 , 2.2 , 3.5 , and 4.9 @xmath0 ( near - ir ) , 12 , 25 , 60 , and 100 @xmath0 ( mid - ir ) , and 140 and 240 @xmath0 ( far - ir ) . the instrumental zero point is established by chopping between the sky and a cold zero - flux reference source at 32 hz .
frequent observations of internal reference sources and stable celestial sources allow corrections for small instrumental gain instabilities on both short and long time scales .
measurements of a few well - calibrated celestial sources ( sirius for 1.25 12 @xmath0 , ngc 7027 for 25 @xmath0 , uranus for 60 and 100 @xmath0 , and jupiter for 140 and 240 @xmath0 ) provide the absolute calibration of the dirbe photometric system .
further details on the calibration of the data can be found in the _ cobe _ dirbe explanatory supplement ( 1997 ) .
the dirbe time - ordered data are pixelized and mapped in the _ cobe _ sky - cube format ( _ cobe _ dirbe explanatory supplement 1997 ) .
all analysis is performed on maps in this coordinate system . for illustrational purposes , maps shown in this paper
are reprojected into a galactic mollweide projection .
the mollweide projection is an equal - area projection with the convenient properties that longitudes are equally spaced ( for a fixed latitude ) and all lines of constant latitude are straight horizontal lines ( though unequally spaced ) .
the dirbe surface brightness maps are presented in units of mjy / sr , which can be converted to @xmath25 intensities in units of nw m@xmath26 sr@xmath27 through multiplication by 3000/@xmath14 , where @xmath14 is the wavelength of the band in @xmath0 .
the sky was scanned by dirbe in a manner that provided highly redundant coverage over a @xmath28 wide swath in a single week .
the viewing swath samples solar elongation angles from roughly @xmath29 to @xmath30 , and precesses with an annual period .
the dirbe beam is @xmath31 at all wavelengths .
a more complete description of the dirbe instrument has been given by silverberg et al .
( 1993 ) and the _ cobe _ dirbe explanatory supplement ( 1997 ) .
the _ cobe _
mission is described by boggess et al .
( 1992 ) .
the highly redundant coverage per line of sight and wide range of solar elongation angles sampled provide an excellent database for purposes of modeling the ipd foreground in order to remove it from dirbe observations .
we use the parametric geometrical model described by paper ii to compute the signal from the interplanetary dust cloud , denoted here as @xmath32 , for each line - of - sight ( pixel ) in the dirbe maps .
components of the model are a smooth dust cloud , three pairs of dust bands , and a circumsolar ring in earth - orbit resonance .
details on the construction of this model and on the accuracy of its results are presented in paper ii .
maps of the intensity after subtraction of the ipd emission and scattered sunlight are shown in galactic mollweide projection in figure [ zlsub_maps ] .
these data available as the ` zodi - subtracted mission average ( zsma ) ' maps , through the nssdc _ cobe _ homepage website at @xmath33 .
in order to model and remove the galactic foreground emission on the largest angular scales ,
we excluded small localized regions that are much brighter ( and often had distinctly different properties ) than the average galactic foreground . these regions mainly consist of bright stars at the near - ir wavelengths , star forming regions at mid- and far - ir wavelengths , and a few nearby external galaxies [ see odenwald , newmark , & smoot ( 1997 ) for an analysis of the external galaxies visible to dirbe ] . the locations of these bright sources have been individually blanked from the maps , and were excluded from further analysis . even though the intensities in the unblanked pixels are left unchanged ,
the mean intensity over large areas is reduced by the bright source blanking because the brightest pixels are now excluded .
a number of bright source identification algorithms were tested , both for speed of execution and for source detection accuracy .
we adopted an algorithm based upon pixel brightness relative to a locally determined smooth background .
the smooth background level for each pixel in the map was computed using a morphological filter ( the `` opening '' operation ; see haralick , sternberg & zhuang 1987 ) , followed by a linear smoothing .
pixels in the original map where the brightness exceeded the background level by more than a fixed threshold were blanked .
these thresholds are : 15 , 15 , 15 , 15 , 85 , 110 jy for point sources at 1.25 , 2.2 , 3.5 , 4.9 , 12 , and 25 @xmath0 respectively and 4.5 mjy sr@xmath27 = 135 nw m@xmath26 sr@xmath27 for the more extended sources found at 100 @xmath0 .
the 60 , 140 , and 240 @xmath0 maps were blanked at the same locations as the 100 @xmath0 map . because the dirbe beam is larger than a standard pixel ( @xmath34 ) ,
the four nearest neighbor pixels were blanked as well .
circular regions around the large and small magellanic clouds were blanked at all wavelengths .
the fraction of the sky above @xmath35 that is excluded by the bright source removal ranges from @xmath3635% at the shortest wavelengths to @xmath371% at @xmath14 @xmath38 60 @xmath0 .
figure [ bs_map ] shows the map of 2.2 @xmath0 intensity after the contribution of the ipd has been removed and the bright source blanking has been applied ( cf .
[ zlsub_maps ] ) .
all subsequent intensity maps will contain black regions indicating the locations where bright sources have been blanked .
this source removal technique does not work well at low galactic latitudes , where both source confusion and the curvature of the background are high .
in tests on simulated maps , however , the method was completely successful in blanking sources above @xmath39 .
further tests showed that varying the threshold level by @xmath40 20% ( a generous estimate of the map noise ) changed the median brightness of the sky above @xmath41 by well under one percent at 1.25 and 2.2 @xmath0 , and even less at the other wavelengths .
after the bright sources are removed by blanking , most of the residual diffuse near - ir emission is starlight from unresolved sources .
these sources are sufficiently numerous and evenly distributed that a statistical model well represents their collective emission . to remove the galactic diffuse stellar emission component we constructed a `` faint source model '' ( fsm ) of the integrated brightnesses in the dirbe 1.25 , 2.2 , 3.5 , 4.9 , 12 and 25 @xmath0 bands , based on the statistical model developed by wainscoat et al .
subsequent improvements to the wainscoat et al
. `` sky '' model by cohen ( 1993a , 1994a , 1995 ) are generally not included in the fsm because they do not pertain to emission at near - ir wavelengths or because they are not specified in detail such that they can be accurately reproduced within the framework of the fsm .
the fsm can be used to calculate the surface brightness in any direction of the sky by integrating the flux emitted by the various stellar components along that line of sight .
if the galaxy consisted of a single type of star , with a corresponding absolute magnitude @xmath42 , then the sky brightness along a line of sight would be given by : @xmath43 where @xmath44 is the number density of sources at a given position in the galaxy which is integrated along the line of sight @xmath45 , @xmath46 is the optical depth along the line of sight , and @xmath47 is the luminosity of a zero magnitude star .
a more realistic representation of the sky brightness must include a summation over many different stellar types and their spatial distributions . the simple expression ( eq .
[ eq_fsm ] ) is then replaced by the more complicated one : @xmath48 where the summations over @xmath49 and @xmath50 represent , respectively , those over stellar types , and over galactic structural components , and the integral over @xmath42 represents the fact that stars of a given type have an intrinsic dispersion @xmath51 in absolute magnitudes centered on @xmath52 with a probability @xmath53 .
the fsm follows the basic form of the wainscoat et al . and sky models , including 5 structural components for the galaxy ( disk , spiral arms , molecular ring , bulge , and halo ) , 87 source types , each with a dispersion of absolute magnitudes , and interstellar extinction from dust in an exponential ( in radius and scale height , @xmath54 ) disk .
the spiral arm component includes the local spur as defined by cohen ( 1994a ) .
we have made several modifications to our fsm , deviating from the sky model , either out of necessity or to produce results more suited for comparison with the dirbe data : \a ) since source count models can not accurately represent the brightest point sources , which are unevenly distributed across the sky , the fsm was only integrated over stars fainter than those that were previously blanked from the dirbe maps .
the value of the brightness limit at which the bright source blanking stops and the faint source model begins is not a significant source of uncertainty .
changes in the limits by 20% produce @xmath375% changes in the faint source model intensities , which are largely offset by complementary changes in the flux excluded by the bright source blanking ( section 4.1 . ) .
\b ) we have increased the spatial resolution of the model and calculated the sky brightness at the resolution of the dirbe maps .
\c ) the halo is described by wainscoat et al .
( 1992 ) only as an @xmath55 law in projected surface brightness .
therefore we have adapted the formulation for the volume density presented by young ( 1976 ) for use in our faint source model . at high galactic latitudes ,
the halo is the third most important component of the faint source model after the disk and the spiral arms .
\d ) we omitted the extragalactic component of the sky model , which contributes mainly at faint magnitudes at 25 @xmath0 ( cohen 1994a ) .
\e ) the position of the sun was set at 18 pc above the midplane of the galactic disk .
this was determined during preliminary trials , by requiring equal brightnesses at the north and south galactic poles after subtraction of the fsm in the near ir bands .
this value is independently supported by analysis done with the sky model ( cohen 1995 ) .
\f ) for 3.5 and 4.9 @xmath0 absolute magnitudes , which are not represented in the wainscoat et al . ( 1992 ) source table , we obtained approximate magnitudes for the various source types by extrapolating from the @xmath56 magnitudes using @xmath57 and @xmath58 stellar colors from johnson ( 1966 ) .
\g ) the conversion from magnitudes to fluxes was done in a manner consistent with the absolute calibration of the dirbe data ( see _ cobe _ dirbe explanatory supplement , 1997 ) .
the intensities of the fsm at 2.2 @xmath0 were multiplied by a factor of 0.963 to account for differences between the color corrections of sirius ( which was used as the dirbe absolute calibrator ) and those of k and m giants which dominate the emission of the fsm and have a co absorption band partially within the 2.2 @xmath0 dirbe passband . there
exist uncertainties of 10 15% in the model arising from details of the absolute calibration of the model and the differences and treatment of the various broadband filter responses ( cohen 1993b ) . to check our fsm , we compared the results of our calculations to the average surface brightnesses of 238 large zones covering the entire sky as calculated by the sky model ( cohen 1994b
; see table 7 of wainscoat et al .
( 1992 ) for the zone boundaries ) .
our faint source model reproduces the mean intensities calculated with the sky model to within @xmath365% for most of the zones .
the zones that are not well matched are all low - latitude zones below @xmath39 ( except for the zone containing the lmc and another near the taurus region ) which have been adjusted in the sky model , but not in the fsm , for various localized features within the galaxy ( cohen 1994a ) .
we have also compared the star counts expected from our fsm with the star counts from the prototype 2mass survey in 7 fields at @xmath56 and @xmath59 ( skrutskie 1996 ) .
the fields are all in the first galactic quadrant and span the range @xmath60 .
the fsm star counts resemble the 2mass star counts , although about half the fields show statistically significant differences , mostly at magnitudes @xmath61 and @xmath62 . however , these fainter stars contribute a small enough fraction of total emission from these fields that the integrated brightnesses of the stars in the 2mass fields are consistent with the predictions of the faint source model .
this comparison is limited by the small size of the fields ( 0.12 7 deg@xmath63 ) and thus the large statistical uncertainties on the star counts .
figure [ fsm_map ] shows the 2.2 @xmath0 sky brightness calculated with the fsm .
the model predicts the mean intensity in each pixel , and thus varies smoothly over the sky .
the actual sky is not expected to be as smooth as the model because random poisson deviations from the mean number of stars / magnitude in each pixel create intensity fluctuations .
faint discontinuities in the model are caused by the artificially sharp edges of the spiral arms , as defined in the model ( wainscoat et al .
1992 ; cohen 1994a ) .
maps of the fsm intensities at 1.25 25 @xmath0 can be obtained from the nssdc _ cobe _ homepage website at @xmath33 .
figure [ gfs_map ] shows the 1.25 4.9 @xmath0 residual emission ( @xmath64 ) after the removal of stars and other compact objects with the bright source blanking and subtraction of the fsm . at these wavelengths ,
the subtraction of the stellar emission is very effective at removing the brightness gradients at high galactic latitudes .
the maps are mottled by the emission of the remaining faint , marginally resolved stars , which lie below the bright source threshold .
the average emission of these remaining stars is included in the smooth faint source model .
therefore , over sufficiently large regions this stellar confusion acts as an additional noise term in the residual maps . at wavelengths @xmath65 3.5 @xmath0 and low galactic latitudes ( @xmath66 )
, the fsm tends to overestimate the actual sky brightness , leading to negative residuals .
a longitudinally anti - symmetric residual is present at the location of the galactic bulge ( most clearly seen in the 4.9 @xmath0 map of fig .
[ gfs_map ] ) , which suggests that improvements could result if a bar - like model were used for the bulge ( e.g. , dwek et al .
1995 ; blitz & spergel 1991 ) .
some relatively large differences between the model and the data are found at locations where our implementation of the faint source model has omitted particular adjustments for specific disk and spiral arm features ( cohen 1994a ) .
residual defects from the subtraction of the ipd emission are seen clearly at wavelengths @xmath38 3.5 @xmath0 as s - shaped bands along the ecliptic . in the mid - ir ( after removal of ipd emission )
, the faint source emission contributes @xmath3740% of the observed brightness toward the inner galaxy at low latitudes , and less than 5% and 1% for @xmath67 at 12 and 25 @xmath0 , respectively .
the subtraction of the faint source model in these bands is only apparent in the inner galaxy , and residual intensity maps look nearly the same as those in figure [ zlsub_maps ] . the residual emission at wavelengths @xmath38 12 @xmath0 is clearly dominated by the presence of ism emission , @xmath6 .
the fsm was not optimized to fit the dirbe data , but is calculated from a fixed prescription based on a model designed to fit observed source counts in selected directions in the sky .
it is therefore important to examine to what extent the removal of galactic starlight with the fsm has under- or over - subtracted stellar emission from the maps .
one test designed to look for such effects is to search for structure in the residual map that may be correlated with that of the fsm , or galactic latitude .
figures [ gfs_b_grad ] and [ gfs_beta_grad ] illustrate the gradients of @xmath12 as a function of galactic and ecliptic latitude , respectively , for the 1.25 , 2.2 , 3.5 , and 4.9 @xmath0 data .
table [ tab_fsm_grad ] lists the gradient and correlation coefficient for @xmath12 as a function of @xmath68 .
the figures and table both show that the faint source model removes much of the galactic gradient in the near - ir emission .
subtraction of the model also reduces the longitudinal variation of the near - ir sky brightness , as evidenced by the reduced dispersions at lower latitudes in figures [ gfs_b_grad ] and [ gfs_beta_grad ] . at 4.9 @xmath0 , the residual ipd gradient visible in figure [ gfs_beta_grad ] is responsible for the structure in the residual emission as a function of @xmath68 that is not observed at shorter wavelengths ( fig .
[ gfs_b_grad ] ) .
the last foreground component we need to remove is the emission from the galactic ism .
as outlined in section 2 , we use the 100 @xmath0 dirbe data ( after removal of the ipd emission ) to derive a spatial template of the ism .
we correlate the 100 @xmath0 data with tracers of the gas phase of the ism , and subtract emission that is uncorrelated with the ism to form the 100 @xmath0 ism template .
the external data used to trace the gas phase of the ism are introduced in section 5.1 .
then we derive the 100 @xmath0 emission per h i column density , and assess the possibilities that some of the emission comes from dust within the ionized and molecular phases of the ism ( 5.2 ) . after subtracting the 100 @xmath0 emission that is not associated with the ism
, we correlate the 100 @xmath0 ism template against the emission at other infrared wavelengths to derive the ism intensity at those wavelengths ( section 5.3 ) .
the remaining subsections describe modified procedures that we apply to identify the ism emission at the longest ( 240 @xmath0 ) and shortest ( 3.5 and 4.9 @xmath0 ) wavelengths .
the 100 @xmath0 map is chosen as the template of the high latitude ism because : ( a ) it provides full sky coverage ; ( b ) it has better sensitivity than the maps at longer wavelengths where noisier bolometric detectors were used ; ( c ) it is less affected by errors in the removal of ipd emission than the shorter wavelength maps ; and ( d ) it represents a compromise between potential differences in spatial structure of the shorter and longer wavelength ism emission .
one drawback to this choice is that larger and a greater number of corrections for instrument temperature , charged particle radiation , and photon induced responsivity changes were required at 100 @xmath0 than at any other wavelength ( see _ cobe _ dirbe explanatory supplement 1997 ) . in order to identify that portion of the 100 @xmath0 intensity ( remaining after subtraction of ipd emission ) that arises from the ism
, we need to compare the dirbe data with other data sets that can be safely assumed to trace only galactic ism emission .
three h i 21 cm line surveys have been used for comparison with the long wavelength dirbe data .
the first is the at&t bell laboratories survey of the sky north of @xmath69 at an angular resolution of about @xmath70 ( stark et al .
the velocity - integrated ( @xmath71 km s@xmath27 ) line intensities were converted to h i column densities assuming the line emission is optically thin , and a map of column density was produced in _ cobe _ sky - cube format ( _ cobe _ dirbe explanatory supplement 1997 ) .
the bell laboratories 20 foot horn reflector has very low far - sidelobe response .
lockman , jahoda , and mccammon ( 1986 ) estimated that the maximum far - sidelobe contribution to observed column density is @xmath72 cm@xmath26 , and the typical contribution is @xmath74 cm@xmath26 .
the second data set consists of observations of an @xmath75 region centered on the north ecliptic pole ( @xmath76 ) at @xmath77 resolution ( elvis , lockman , and fassnacht 1994 ) .
elvis et al .
used the bell laboratories survey to correct their spectra for stray radiation , and estimated that the 1@xmath78 random error in column density is @xmath79 cm@xmath26 .
the third data set covers a 300 deg@xmath63 region in ursa major surrounding the direction of lowest h i column density ( @xmath80 ) at a resolution of @xmath77 or better ( snowden et al .
snowden et al . also corrected their spectra for stray radiation , and estimated 1@xmath78 errors ranging from @xmath81 cm@xmath26 to @xmath82 cm@xmath26 .
velocity - integrated maps of the north ecliptic pole and ursa major regions were obtained from f. j. lockman ( private communication ) .
the velocity ranges of @xmath83 to 150 km s@xmath27 for the north ecliptic pole and @xmath83 to 100 km s@xmath27 for ursa major include all significant galactic emission .
the maps were degraded to dirbe resolution and reprojected to _ cobe _ sky - cube format .
the co data used in this analysis comprise the @xmath84co survey of the ophiuchus region by de geus et al .
( 1990 ) , and the @xmath84co surveys of the polaris flare , ursa major , and camelopardalis regions by heithausen et al .
these surveys , which cover a total area of 900 deg@xmath63 at high ( @xmath85 ) galactic latitude , were taken with the 1.2 m millimeter - wave telescope initially located at columbia university in new york city , and later at the harvard - smithsonian center for astrophysics in cambridge , ma , or the nearly - identical columbia southern millimeter - wave telescope in cerro tololo , chile .
both telescopes have an angular resolution of 87 at the @xmath84co j=@xmath86 line frequency observed .
the surveys were used to produce a velocity - integrated @xmath84co map degraded to the resolution of the bell labs h i survey .
the analysis presented in the appendix makes use of pulsar dispersion measures from taylor , manchester , and lyne ( 1993 ) and camilo and nice ( 1995 ) , and h@xmath87 intensities observed by reynolds ( 1980 , 1984 , 1985 , 1991b ) using a fabry - perot spectrometer with a beam @xmath88 in diameter .
our primary identification of the 100 @xmath0 ism emission is made through correlation with h i column densities toward the north ecliptic pole and the region of minimum h i column density ( the lockman hole ) .
the choice of these locations is determined by the low column densities , the availability of high quality h i data sets , and the ability to estimate or limit contributions from the molecular and ionized components of the ism .
the slope of the correlation is the average 100 @xmath0 emissivity per h i column density .
the intercept of the correlation is the mean intensity that is _ not _ associated with the h i component of the ism .
we also estimate the intensity of 100 @xmath0 emission that may be associated with the ionized and molecular phases of the ism .
however , direct correlations between the 100 @xmath0 emission and combinations of @xmath89(h i ) and @xmath89(h ii ) or @xmath89(h@xmath90 ) are found to be relatively ineffective because of low column density and/or emissivity of the molecular and ionized components . to determine the association of the 100 @xmath0 emission with the neutral atomic ism
, we correlate the h i column density , @xmath89(h i ) , against the 100 @xmath0 intensity after subtraction of ipd emission ( and , in principle , any stellar emission ) , @xmath91 , @xmath92 to solve for the emissivity of the neutral ism , @xmath93 , and the mean intensity of emission that is uncorrelated with h i , @xmath94 .
figure [ b10_hi]a shows the pixel - by - pixel comparison for the @xmath95 region centered on the north ecliptic pole ( nep ) . figure [ b10_hi]b shows the comparison for the 300 deg@xmath63 region surrounding the region of lowest h i column density ( the lockman hole ) . figure [ b10_hi]c shows the comparison for a region at @xmath96 @xmath97 25 , @xmath98 @xmath97 25 , and declination @xmath99 , using 100 @xmath0 data smoothed to the resolution of the bell laboratories survey .
this very large region is included to provide an indication of the extent to which the lockman hole and nep regions are typical of other high latitude regions .
data in a @xmath100 dirbe pixel patch centered on the source rafgl 5429 were excluded from figure [ b10_hi]a , and data in @xmath100 pixel patches around five galaxies ( ngc 3079 , 3310 , 3556 , 3690 , and 4102 ) were excluded from figure [ b10_hi]b .
the solid lines show linear fits to the data that minimize @xmath101 calculated using measurement errors in both variables ( press et al .
each plot shows excess 100 @xmath0 emission at the highest h i column densities .
these data were excluded from the fitting by rejecting all data above a cut line that is perpendicular to the fit line on a plot where each variable is divided by its mean measurement error .
the adopted cut lines are shown as short diagonal lines in figures [ b10_hi]a c .
the cut line intersects the fit line at @xmath89(h i)=@xmath102 cm@xmath26 for the nep region , at @xmath103 cm@xmath26 for the lockman hole region , and at @xmath104 cm@xmath26 for the region of figure [ b10_hi]c .
each fit was done iteratively until the fit parameters and the data excluded by the cut stopped changing .
the derived intercept values are not sensitive to the exact location of the cut lines . for the lockman hole region , for example , the derived intercept varied by less than 0.7 nw m@xmath26 sr@xmath27 for a series of test fits in which the location of the cut line intersection ranged from @xmath105 to @xmath106 cm@xmath26 .
figure [ b10_hi_binned ] shows a comparison of the fit line with data averaged within uniformly spaced bins along the fit line for each of the three regions . in each region , the bin - averaged data are consistent with a linear relation over the range of column density used for the fitting .
these correlations begin to deviate from linearity at higher column densities because of the contribution of emission from dust associated with molecular h@xmath107 ( e.g. reach , koo , & heiles 1994 ; reach , wall , & odegard 1998 ) . the fit parameters ( slopes , @xmath93 , and intercepts , @xmath108 @xmath16 ) are listed in table [ tab_b8_hi ] .
the tabulated uncertainties are formal errors determined from the 68% joint confidence region in parameter space .
the average of the intercepts for the nep and lockman hole regions , 19.8 nw m@xmath26 sr@xmath27 , was subtracted from the 100 @xmath0 ipd - subtracted skymap to form the spatial template of 100 @xmath0 ism emission ( see section 5.3 ) .
the difference between the intercepts for the nep and lockman hole regions is 5.0 nw m@xmath26 sr@xmath27 .
this value was adopted as the systematic uncertainty for the zero level of the template .
sources of systematic error that may affect the zero level of the 100 @xmath0 ism template include errors in subtraction of the ipd emission and errors due to variations in 100 @xmath0 emissivity per neutral h atom ( the @xmath109(100 @xmath0)/@xmath89(h i ) ratio for emission originating from the galaxy ) within the nep or lockman hole regions .
the latter could be caused by variations in dust temperature , grain composition , grain size distribution , or dust - to - gas mass ratio within the neutral atomic gas phase , or by emission from dust associated with ionized or molecular gas .
if the 100 @xmath0 emissivity per neutral h atom increases with increasing h i column density within a region , the intercept of the linear @xmath109(100 @xmath0 ) @xmath110 @xmath89(h i ) fit would underestimate the zero level of the emission from the ism , and if it decreases with h i column density the intercept would overestimate the zero level . however , such systematic variations would tend to produce nonlinear @xmath109(100 @xmath0 ) @xmath110 @xmath89(h i ) relations .
a quadratic dependence might be expected for a cloud containing h@xmath90 ( e.g. , dalloglio , et al .
1985 ; reach , koo , & heiles 1994 ) , but specific functional forms are not predicted for other causes of variation in emissivity per neutral h atom .
a simple interpretation of the linear relations found for the nep and lockman hole is that systematic variations of emissivity per neutral h atom with h i column density are not important at low h i column densities in these regions .
the difference between the intercepts found for the nep and lockman hole regions can probably be attributed to error in subtraction of ipd emission .
evidence for this can be seen in figure [ lat_trends ] , which shows that a weak ecliptic latitude dependence remains in the 100 @xmath0 data after subtraction of the ipd emission model .
the figure shows a similar ecliptic latitude dependence for data within four narrow ranges of h i column density .
the intensity difference between @xmath111 and @xmath112 in figure [ lat_trends ] is comparable to the difference in intercepts derived for the lockman hole ( @xmath113 ) and nep ( @xmath114 ) .
because the preceding correlations only identify the 100 @xmath0 emission that is correlated with the neutral h i phase of the ism , we need to estimate any error in the inferred zero level of the 100 @xmath0 ism template that may be caused by dust emission associated with the ionized phase of the ism .
the exact distributions of ionized hydrogen column density within the nep and lockman hole regions are not known . however , for the lockman hole , available data can be used with some assumptions to estimate the possible error of the zero level .
we initially assume that the 100 @xmath0 emissivity per h nucleus within the h ii gas phase equals the slope of the @xmath109(100 @xmath0 ) @xmath110 @xmath89(h i ) relation , and that the h ii column density is constant within the region . under these assumptions , the @xmath109(100 @xmath0 ) @xmath110 @xmath89(h i ) relation would still be linear but its intercept ( @xmath115 ) would overestimate the zero level of emission from the ism by @xmath89(h ii ) times the emissivity per h nucleus .
table [ tab_hi_h2 ] lists information on ionized h column density for the lockman hole based on a pulsar dispersion measure and two h@xmath87 observations .
we are not aware of any useful data on ionized gas in the nep region .
the dispersion measure @xmath116 of psr j1012 + 5307 provides a lower limit to the total @xmath89(h ii ) along the line of sight .
estimates of the distance of the pulsar from the midplane are @xmath54=460 pc from photometry of its companion , assuming it is a @xmath117 white dwarf ( lorimer et al .
1995 ; halpern 1996 ) , and @xmath54=400 pc from the dispersion measure and the galactic electron density model of taylor & cordes ( 1993 ) . estimates of the exponential scale height of the ionized medium in the vicinity of the sun range from 670 to 910 pc ( reynolds 1991a ; nordgren , cordes , & terzian 1992 ; taylor & cordes 1993 ) .
if the ionized gas in the lockman hole region has an exponential @xmath54 distribution with a scale height in this range , and the pulsar is at @xmath54=460 pc , the total @xmath89(h ii ) in the direction of the pulsar would be 2 to 2.5 times the dispersion measure , or about @xmath118 cm@xmath26 .
the h@xmath87 observations provide a measure of @xmath119 ( e.g. , reynolds 1992 ) , since extinction is negligible for lines of sight with such low h i column densities .
h ii column densities have been estimated from the h@xmath87 data in table [ tab_hi_h2 ] using a conversion factor of @xmath120(h@xmath87)/@xmath89(h ii)=0.75 rayleighs/@xmath121 cm@xmath26 ( 1 rayleigh = 10@xmath122/4@xmath123 photons s@xmath27 cm@xmath26 sr@xmath27 = 0.24 nw m@xmath26 sr@xmath27 at h@xmath87 ) .
this value was adopted based on the comparison of h@xmath87 and dispersion measure data shown in figure [ ha_dm ] for high @xmath124 pulsars .
the figure shows that @xmath120(h@xmath125 may increase with increasing h i column density .
the @xmath120(h@xmath125 ratio at the lowest h i column densities in the figure was adopted as the conversion factor for the lockman hole data .
the adopted value corresponds to a characteristic electron density , @xmath126 , of 0.05 @xmath127 for an electron temperature of 8000 k. for comparison , reynolds ( 1991a ) found a mean @xmath128 of 0.08 @xmath127 for lines of sight toward pulsars in 4 globular clusters at @xmath129 4 kpc .
the @xmath89(h ii ) values estimated from the h@xmath87 observations are comparable to or smaller than the pulsar s dispersion measure . even though we consider the @xmath89(h ii ) estimates from the h@xmath87 data to be more reliable than that derived from the dispersion measure , pulsar @xmath54 distance , and scale height of ionized gas , to be conservative we adopt an intermediate value of @xmath89(h ii)=0.4@xmath130 cm@xmath26 as representative of the lockman hole region .
dust associated with ionized gas at this column density would have a 100 @xmath0 intensity of 7 nw m@xmath26 sr@xmath27 under the assumptions described above , and the @xmath109(100 @xmath0 ) @xmath110 @xmath89(h i ) intercept @xmath115 would overestimate the zero level of the ism emission by this amount , about 1/3 of the total .
however , this error estimate is probably too large , for two reasons .
( 1 ) the analysis presented in the appendix suggests that the diffuse ionized gas at high galactic latitudes has a 100 @xmath0 emissivity per h nucleus that is smaller than that of the h i gas .
a 3@xmath78 upper limit of 3/4 times the emissivity per h nucleus of the h i gas was found for a @xmath131 region at @xmath132 , @xmath133 .
analysis by fixsen et al .
( 1998 ) using @xmath134/firas data also indicates that the emissivity of the ionized gas is small .
they correlate far - ir emission against a combination of [ c ii ] line emission , as a tracer of the ionized ism , and linear and quadratic h i column density , as a tracer of the neutral ism . at high latitudes ,
they find little or no far - ir continuum emission correlated with the [ c ii ] line emission .
( 2 ) it is likely that h ii column density is not constant in the lockman hole region , but is correlated to some degree with h i column density .
such correlations are found in the appendix for data along lines of sight toward two samples of high @xmath124 pulsars .
if such a correlation exists in the lockman hole , the error estimate for the 100 @xmath0 zero point should be calculated using only the component of @xmath89(h ii ) that is not correlated with @xmath89(h i ) .
further evidence for correlation between ionized and atomic gas has been shown for the @xmath131 region at @xmath132 , @xmath133 by reynolds et al .
they found that at least 30 percent of the total h@xmath87 emission and 10 to 30 percent of the total 21 cm line emission are kinematically and spatially associated in clouds containing both neutral and ionized gas .
for these reasons , we adopt 4 nw m@xmath26 sr@xmath27 , about 1/5 of the total , as the maximum possible systematic error in the zero level of the 100 @xmath0 ism template that may result from h ii associated dust .
we additionally need to estimate the error in the zero level of the 100 @xmath0 ism template that may be caused by emission from dust associated with the molecular ism .
table [ tab_hi_h2 ] lists 3-sigma upper limits on h@xmath90 column density for both the lockman hole and nep regions , based on upper limits to @xmath84co j=10 line emission within a @xmath135 portion of the lockman hole region and toward some h i emission peaks in the nep region .
[ co emission has been detected toward the 100 @xmath0 brightness peaks of a few cirrus clouds elsewhere in the lockman hole region by heiles , reach , & koo ( 1988 ) , stacy et al .
( 1991 ) , and reach , koo , & heiles ( 1994 ) .
the h i column densities at these positions are @xmath136 cm@xmath26 or greater , so they were excluded from the @xmath109(100 @xmath0 ) vs. @xmath89(h i ) fitting . ] following elvis et al .
( 1994 ) , limits on observed line emission were converted to limits on h@xmath90 column density assuming a 1 km s@xmath27 line width , typical of diffuse clouds , and a co intensity to @xmath89(h@xmath90 ) conversion ratio of @xmath137 mol cm@xmath26 ( k km s@xmath27)@xmath27 .
most determinations of @xmath138 for translucent molecular clouds ( clouds with visual extinction in the range @xmath139 mag ) are consistent with this upper limit ( magnani & onello 1995 , and references therein ) .
dust associated with molecular gas at or below the @xmath89(h@xmath90 ) limits in table [ tab_hi_h2 ] would produce 100 @xmath0 intensities of @xmath3742 and @xmath3723 nw m@xmath26 sr@xmath27 for the lockman hole and nep , respectively , assuming constant @xmath89(h@xmath90 ) across these regions and emissivity per h nucleus equal to the slope of the @xmath109(100 @xmath0)@xmath89(h i ) relation .
these conclusions for the lockman hole and nep can be extended to cover the entire north galactic hemisphere , since a recent co survey for @xmath140 by hartmann , magnani , & thaddeus ( 1998 ) has comparable sensitivity to the measurements in table 3 and finds that detectable co has a filling factor of 0.004 0.006 . for the lockman hole region ,
the h@xmath90 column density and the possible 100 @xmath0 emission from dust associated with h@xmath90 are probably much smaller than allowed by the limit on co line intensity .
ultraviolet absorption line observations toward early type stars have shown that a transition in the fraction of h in molecular form occurs at color excess @xmath1410.08 mag , or at @xmath89(h i ) + 2@xmath89(h@xmath90)@xmath142 atoms cm@xmath26 , above which more than 1% of h atoms are in the form of h@xmath90 ( savage et al .
studies of the h i
h@xmath90 phase transition based on h i , co , and 100 @xmath0 observations of high latitude molecular clouds are consistent with this ( reach , koo , & heiles 1994 ; gir , blitz , & magnani 1994 ) . the @xmath89(h@xmath90 ) limit in table [ tab_hi_h2 ] , together with the maximum h i column density used in the @xmath109(100 @xmath0)@xmath89(h i ) fitting , yields @xmath89(h i ) + 2@xmath89(h@xmath90)@xmath143 atoms cm@xmath26 , suggesting that the gas in the lockman hole is below the threshold for significant fractional h@xmath90 abundance .
however , the @xmath89(h@xmath90 ) limit may not be valid if h@xmath90 is present in clouds that are too diffuse for co to exist .
estimates of visual extinction toward the lockman hole give another indication that the h@xmath90 column density is very low .
the galactic reddening map of burstein & heiles ( 1982 ) , which is based on maps of galaxy counts and h i column density and is calibrated using reddening measurements for globular cluster stars and rr lyrae stars , gives a mean @xmath144 of @xmath145 for the lockman hole region .
the reddening map of holmberg ( 1974 ) , based on counts of distant zwicky galaxy clusters and calibrated assuming @xmath146=0.25 at the galactic poles , gives @xmath147 .
the difference between these results is primarily due to differences in calibration ; if the holmberg map were recalibrated to the mean polar reddening of burstein and heiles , the results would agree within the quoted relative errors .
recent determinations of polar absorption by different methods are in the range @xmath148 ( de vaucouleurs 1995 ; knude 1996 , and references therein ) so the mean @xmath144 for the lockman hole region is probably less than 0.035 and the reddening for most or all of the positions used for the @xmath109(100 @xmath0)@xmath89(h i ) fitting is probably below @xmath149 . adopting the largest fractional h@xmath90 abundance measured by savage et al .
( 1977 ) for @xmath150 would give an average h@xmath90 column density of @xmath151 mol cm@xmath26 for the lockman hole .
assuming constant @xmath89(h@xmath90 ) within the region and emissivity per h atom in the h@xmath90 phase equal to the slope of the @xmath109(100 @xmath0 )
@xmath89(h i ) relation , dust associated with molecular gas would give rise to a 100 @xmath0 intensity of only 0.002 nw m@xmath26 sr@xmath27 .
as in the case for the ionized ism , we have also attempted to determine the 100 @xmath0 emissivity of the molecular ism . for each of the four regions covered by the above @xmath84co surveys ( 5.1.2 ) , we performed a least - squares fit of the equation @xmath152 to the data by varying the parameters @xmath153 , and @xmath154 , where @xmath93 is the 100 @xmath0 emissivity per h atom for the h i gas phase , @xmath155 is the product of the 100 @xmath0 emissivity per molecule for the h@xmath107 gas phase and the ratio of h@xmath107 column density to @xmath84co intensity , and @xmath154 represents any background emission . in each case
the resulting fit to the 100 @xmath0 fluxes is significantly poorer than those obtained in figure [ b10_hi ] , where the contribution of dust associated with molecular gas to the 100 @xmath0 emission is believed to be negligible .
the breakdown of the infrared - gas column density correlation within high - latitude regions having significant co emission may result from one or more of the following : ( 1 ) there may be a greater variability in the dust - to - gas mass ratio or dust temperature within h@xmath107 regions than within h i regions ; ( 2 ) regions with a higher co content may have more dust associated with ionized gas which is not accounted for in the correlations .
the photographic h@xmath87 survey of sivan ( 1974 ) shows that a significant fraction of the ophiuchus region contains extended h ii regions ; ( 3 ) the dust column density within these regions may be too low to effectively shield the co molecule from dissociating uv radiation ( see , for example , heithausen et al .
1993 ) , and therefore the velocity - integrated @xmath84co intensity may not accurately trace the h@xmath107 column density along the line of sight . having estimated the 100 @xmath0 emission that is not associated with the ism ( @xmath156 nw m@xmath26 sr@xmath27 ) , we construct the @xmath15 @xmath16 map from @xmath18 @xmath157 @xmath16 , which is then used as the template for the ism emission for derivation of maps of the background emission at the other dirbe wavelengths : @xmath158 we fit the correlations of @xmath159 to @xmath160 to determine @xmath11 , the @xmath161 @xmath0 color , from the slope of the correlation and a mean @xmath7 from the intercept .
these correlations and associated least - squares fits are shown in figure [ bx_b8 ] .
the slopes and their statistical uncertainties are given in table [ tab_bx_b8 ] .
for the far - ir bands , the correlations are determined only for the regions where @xmath162 . using a cut at lower latitudes resulted in a steeper slope , influenced by bright clouds found only at the lower latitudes .
a higher latitude cut resulted in a shallower slope , possibly an artifact of the poorer correlation provided by the fainter , very high latitude cirrus .
we settled on the @xmath163 cut as a compromise between the weaker correlations at higher latitudes , and the intrinsic color temperature variations between the clouds visible at the lower latitudes . at mid - ir wavelengths ,
residuals for the ipd removal ( paper ii ) were too large to allow use of this entire region for correlation .
smaller regions where the ipd and fsm removal have left no obvious structure were used : @xmath67 and @xmath164 at 60 @xmath0 , and @xmath165 at 12 and 25 @xmath0 . in the near - ir bands ( @xmath166 ) , residuals from removal of starlight and ipd scattering and emission were too large to allow a direct correlation of these bands with the 100 @xmath0 ism template ( see section 5.5 ) . in using linear least - squares fits , it is assumed that the @xmath11 colors of the ism are constant at high latitudes .
the validity of this assumption is indicated by relatively large correlation coefficients , and by dispersions about the least - squares fit relation which are no more than a factor of 2 greater than those expected from the uncertainties associated with the ipd - subtracted data .
these quantities are listed in table [ tab_bx_b8 ] .
another indication of the expected uniformity of the @xmath11 colors is the observed galactic temperature gradient reported by sodroski et al .
even though this radial temperature gradient leads to large color variations at low latitudes , at @xmath162 the gradient implies large scale variation in the @xmath167 color of only @xmath404% between the inner and outer galaxy .
the color variations at other wavelengths should be even smaller , as the radial temperature gradients are smaller ( sodroski et al .
1987 , 1989 ) . in our present results
any such large scale trends are obscured by smaller scale cloud - to - cloud variations and therefore have no direct impact on the derived background level . scaling the 100 @xmath0 ism template by the appropriate colors ( @xmath11 from table [ tab_bx_b8 ] ) and subtracting it from each of the mid- and far - ir wavelength maps yields the maps shown in figure [ gcr_maps ] .
for the 100 @xmath0 residual map we depict the subtraction of @xmath168 nw m@xmath26 sr@xmath27 @xmath169(h i ) from the data ( derived in section 5.2.1 ) .
these maps show the end result of our removal of both ipd and galactic emission .
they should contain the full cib signal as well as any residual emission left by imperfect foreground modeling .
figures [ gcr_b_grad ] and [ gcr_beta_grad ] illustrate the gradients depicted in the residual sky brightness of these maps as functions of galactic and ecliptic latitude .
table [ tab_gcr_grad ] lists these gradients and the correlation coefficients with respect to @xmath68 . at 12 and 25
microns , the subtraction of the ism emission allows the residual gradients of ipd emission to be seen and characterized more clearly , as indicated by figures [ gcr_maps ] , [ gcr_b_grad ] , and [ gcr_beta_grad ] and in table [ tab_gcr_grad ] . at longer wavelengths ,
the ipd residuals are weak compared to the residual ism emission .
the single component model of the ism based on the 100 @xmath0 data removes much of the emission of the ism at high latitudes .
however the residual maps ( fig .
[ gcr_maps ] ) show that at the far - ir wavelengths a strong second component to the ism emission is still present , especially at lower latitudes ( @xmath170 ) .
the similarity of the 140 and 240 @xmath0 residual maps suggests that a two - component model with templates formed from the 100 @xmath0 ism emission and the 140 @xmath0 residual could be effective at removing the 240 @xmath0 emission of the ism . for a linear combination of these two templates
, we can substitute an equivalent linear combination of the 100 @xmath0 ism emission and the 140 @xmath0 ism emission .
we use a linear least squares fit to solve for the coefficients that scale the templates to fit the 240 @xmath0 data . performing this correlation over the entire unblanked sky leads to a 240 @xmath0 ism model of @xmath171 where the ism components at 100 and 140 @xmath0 have had constant terms ( estimated from the one - component ism subtraction residuals ) of 19.8 and 18.5 nw m@xmath26 sr@xmath27 removed respectively .
this two - component model of the ism emission will be referred to as the `` ism2 '' model to distinguish it from the one - component ( ism1 ) model described in 5.3 .
a map of the residuals after subtraction of the ism2 model from the data is shown in figure [ gcr2_10 ] .
the residuals are plotted as functions of galactic and ecliptic latitude , @xmath172 and @xmath173 , in figs .
[ gcr_b_grad ] and [ gcr_beta_grad ]
. this model appears to do an excellent job at removing the ism emission variation all the way down to @xmath174 .
the increase in noise in the residual map occurs as a result of the introduction of the noisy 140 @xmath0 data as part of the model .
the noise partially obscures some deficiencies in the residual map , such as some remaining emission from the ipd which can be seen in a wide band along the ecliptic .
this residual emission arises primarily from the incomplete subtraction of ipd emission at 100 @xmath0 .
physically , the ism2 model subtraction works well because it allows us to account for both the large - scale and local temperature variations of the interstellar dust .
this is demonstrated by noting that the coefficients in equation [ eq2color ] define a line in the 100 140 240 @xmath0 color - color plot ( fig .
[ ccplot ] ) .
this line is a good representation of the data , which are distributed along a nearly linear trend because of temperature variations in the ism .
the bright source blanking ( section 4.1 ) has removed the hottest sources ( mostly h ii regions ) which would extend the range of the data in the color - color plot , but with a distinctly nonlinear trend .
only high signal - to - noise data ( @xmath175 nw m@xmath26 sr@xmath27 ) were used in constructing figure [ ccplot ] , therefore small errors in the zero levels of each band will have little effect on the colors . however , the gain uncertainties ( 6.1 and table [ tab_errors ] ) are large enough to cause systematic changes in the colors .
gain errors of @xmath176 in each color would cause the displacement between the observed colors and the expected color trend for blackbodies with emissivities @xmath177 . for the near - ir wavelengths ,
the subtraction of the starlight is not sufficiently accurate to yield values of @xmath11 and @xmath7 from direct correlations of the near - ir data with @xmath15 @xmath16 .
the 1.25 4.9 @xmath0 emission maps in figure [ gfs_map ] show little similarity to the 100 @xmath0 emission maps in figure [ zlsub_maps ] . therefore , to determine the intensity of the ism emission at these wavelengths we make use of the color differences between the starlight and the ism emission , as revealed in the ipd - subtracted maps ( @xmath178 ) .
colors formed from the simple ratio of two near - ir intensities , e.g. @xmath179 @xmath180 @xmath16 , can reveal extinction effects as well as intrinsic source variations .
we remove the influence of extinction by using the reddening - free parameter : @xmath181^{\zeta } / [ i_z(2.2)/i_z(3.5)]\ ] ] where @xmath182 and @xmath183 is the ratio of extinction at wavelength @xmath14 to that at 1.25 @xmath0 .
the ratio of absorption coefficients is calculated from the extinction law of rieke & lebofsky ( 1985 ) , which has been shown to match the inner galaxy extinction seen by dirbe ( arendt et al .
1994 ) .
the reddening - free parameter is constructed to be independent of the amount of extinction ; it only varies if the sources of emission vary .
in fact , @xmath184 is fairly insensitive to stellar spectral type , so the bulk of the stellar emission of the galaxy can be characterized by a single value , @xmath185 .
thus , variations in @xmath184 will be observed where there is significant _ emission _ from hot interstellar dust .
the map of @xmath184 ( in figure [ q123 ] ) does show more structure of the ism at low latitudes than do the near - ir intensity maps of figure [ gfs_map ] , though residual ipd emission still obscures high latitude structure .
based on the observation that the @xmath179 @xmath180 @xmath16 map shows little or no evidence of diffuse ism features at @xmath186 , while the @xmath187 @xmath188 @xmath16 map shows such features clearly , we assume the ism emission is only significant at 3.5 @xmath0 , i.e. @xmath189 @xmath190 and @xmath191 then we can derive the ism emission as @xmath192.\ ] ] rather than imposing a value of @xmath185 , we can again use a linear least squares fit to derive @xmath185 and the near - ir cirrus color , @xmath193 .
we do this by considering @xmath194 / g_i(100)\\ & = & \frac{q(1.25,2.2,3.5 ) - q_0(1.25,2.2,3.5)}{q(1.25,2.2,3.5 ) \times g_i(100 ) / i_z(3.5)}.\end{aligned}\ ] ] by rewriting this last equation as @xmath195 + q_0(1.25,2.2,3.5)\ ] ] we see that the color @xmath193 can be found as the slope of a least - squares fit of @xmath184 vs. @xmath196 $ ] .
the intercept of the fit is @xmath185 , the nominal stellar reddening - free parameter . in a nearly identical manner
we can also derive the @xmath197 color of the ism emission .
finally , having determined these near - ir colors , we can use them to scale and subtract the 100 @xmath0 template of the ism emission from the 3.5 and 4.9 @xmath0 data . because of the very low intensity of the ism in the near - ir relative to the residual errors of the ipd model , the correlations need to be done at low galactic latitudes ( @xmath170 ) .
the correlations and corresponding least - squares fits are shown in figure [ b3_b8 ] .
the 3.5 and 4.9 @xmath0 colors derived by this method are listed in table [ tab_bx_b8 ] .
the statistical uncertainties in the determination of the 3.5 and 4.9 @xmath0 colors are as good those for the far - ir wavelengths .
the residual maps after subtraction of the derived near - ir emission are shown in figure [ gcr_maps ] .
( at 4.9 @xmath0 , the residual ipd emission is also reduced because the 100 @xmath0 ism template contains some residual ipd emission as well . ) in this derivation of the near - ir emission there are two important assumptions .
the first is that the near - ir emission of the ism within the high - latitude regions of interest is similar to that which is measured at low latitudes by this procedure .
the bright source blanking at low latitudes removes most of the compact emission sources ( mainly h ii regions ) in the area where the ism colors are derived . the remaining emission may still arise from ism with different properties than those of the local high - latitude ism , but the residual maps ( figure [ gcr_maps ] ) give no indication of differences between the unblanked low and high latitude ism emission .
the second assumption is that any near - ir emission of the cib can be neglected in this analysis . at the low galactic latitudes where the ism colors are derived
, stellar emission is likely to be much stronger than the cib emission .
additionally , the slopes of the correlations , which indicate the ism colors @xmath193 and @xmath197 , should not be seriously affected by any isotropic emission , such as that of the cib .
the value of the foreground subtracted maps ( figure [ gcr_maps ] ) as estimators of the cib intensity depends upon of the uncertainties associated with the data and foreground models .
we distinguish two classes of uncertainties .
random uncertainties , such as those arising from the detector noise , are associated with errors exhibiting gaussian or poisson statistics . in general ,
random uncertainties can be reduced by averaging large amounts of data ( i.e. over large regions of the sky , and over the entire mission ) and do not contribute significantly to the total uncertainty of the residual maps .
systematic uncertainties , such as those associated with the detector offsets , affect many or all data in a similar fashion and thus do not average out as the size of the data set increases .
the following sections describe our estimates of the most significant systematic uncertainties associated with each step of the foreground removal process .
the calibration of the dirbe data requires the measurement of instrument gain and offset terms .
errors in the measurement of these quantities will affect all data in a similar fashion .
the evaluation of the gain and offset uncertainties is described in the _ cobe _ dirbe explanatory supplement ( 1997 ) .
the uncertainties in the detector gains and offsets are listed in table [ tab_errors ] .
gain errors will have little impact on the significance of a detected background signal , because the ipd and ism models are scaled directly to the dirbe data .
the fsm is not scaled to the dirbe data directly , but does use the dirbe absolute calibration for conversion from magnitudes to flux densities .
a gain error will cause us to overstate or understate the levels of the residual signal _ and _ its total uncertainty by a multiplicative constant .
the detector offset uncertainties are much more important .
they represent minimum uncertainties that can not be reduced . a true background signal
must at least exceed the offset uncertainties to be detected .
the detector offset uncertainties are worst for the 140 and 240 @xmath0 bands .
after averaging over the entire mission , random instrumental noise is only apparent in the 140 and 240 @xmath0 maps as a graininess visible at the fainter high latitudes .
however , averaging over regions larger than @xmath36 30 pixels ( @xmath198 sr @xmath199 deg@xmath200 ) , reduces the uncertainty from the instrumental noise to levels below that of the total systematic uncertainties . at shorter wavelengths ,
the instrumental noise is dwarfed by the systematic uncertainties even on the scale of a single pixel .
the uncertainties and errors in the subtraction of the emission and scattered light from the ipd are reported in paper ii . of these uncertainties ,
the one likely to be most significant is that arising from uncertainty in the geometry of the interplanetary dust cloud .
the adopted geometric kernel does not uniquely or perfectly describe the data , and consequently we are required to associate systematic errors to the residual intensities that are roughly proportional to the mean intensity of the ipd at each wavelength .
these uncertainties are listed in table [ tab_errors ] .
the systematic uncertainty associated with the removal of galactic stellar emission has two sources .
first , there is uncertainty associated with the bright source blanking . at some locations the blanking may remove confused sources which are individually below the bright source threshold , but
are combined by the dirbe beam into a single source above the threshold .
the result of this effect is to reduce the mean level of the background when measured over large regions .
we have used the faint source model to estimate the frequency with which we expect to find pairs of sources in the same dirbe pixel , where both sources are within a factor of 2 of the bright source threshold .
the number of such double sources at latitudes @xmath67 ranges from @xmath36 140 at 1.25 @xmath0 to @xmath36 6 at 4.9 @xmath0 .
the uncertainties caused by these sources are listed in table [ tab_errors ] .
a more important source of systematic uncertainties in the removal of stellar foreground emission is the accuracy of the faint source model .
systematic errors in the faint source model are not strictly isotropic , but occur on large angular scales such that averages over large patches usually do not reduce the errors .
the clearest indication of systematic errors in the faint source model is the presence of gradients with respect to galactic latitude in the residual emission at 1.25 to 4.9 @xmath0 .
the correlations between these gradients and the fsm at @xmath201 are used to derive the uncertainties reported in table [ tab_errors ] .
the gradients indicate that low latitude emission is oversubtracted relative to the high latitude emission , but they can not indicate at which latitude ( if any ) the subtraction is correct . we have estimated two contributions to the uncertainty associated with the subtraction of the ism .
the first is the uncertainty of the background level that is removed from the 100 @xmath0 map to make the ism template ( see sections 5.2 , 5.3 ) .
any error in this value is propagated into the residual intensities measured at other wavelengths .
the uncertainty in the 100 @xmath0 background intensity is taken to be the difference in the intercepts of the correlations between the h i and 100 @xmath0 emission at the nep and at the lockman hole ( section 5.2.1 ) , which were used to derive the 100 @xmath0 background intensity .
this uncertainty would be larger if we had determined the background using a larger region of the sky , where there would be stronger and less well known contibutions from molecular and ionized components of the ism .
the 100 @xmath0 instrumental offset and ipd uncertainties are added in quadrature to this ism uncertainty before the value is propagated to other wavelengths .
the other ism uncertainty is that associated with the scaling factors , @xmath11 , used to convert the 100 @xmath0 ism emission into ism emission maps at other wavelengths .
these uncertainties were estimated by subdividing the region of sky used to determine @xmath11 into @xmath36 18 regions , and then calculating @xmath11 for each subregion .
the rms variation of @xmath11 for the set of subregions was used as the uncertainty for the global @xmath11 applied in the ism subtraction .
this uncertainty characterizes the variations in the ism spectrum on angular scales of @xmath202 .
both of the systematic uncertainties associated with the ism subtraction are listed in table [ tab_errors ] .
after removal of the ipd and galactic ir foregrounds , the total uncertainty attributed to the residual background intensity at each wavelength is calculated as the quadrature sum of the systematic uncertainties ( @xmath203 ) discussed above and listed in table [ tab_errors ] .
in constructing the fsm , we attempted to reproduce the model described by wainscoat et al .
( 1992 ) as closely as possible .
this preserves the model s basis in source counts and independence from the actual intensity of the cib .
the only modification made to the model in order to improve its fit to the dirbe data was the addition of an 18 pc offset of the sun above the galactic midplane .
this value is supported by the independent results derived at near - ir wavelengths by cohen ( 1995 ) , and other analyses of the dirbe data ( weiland et al .
1994 ; freudenreich 1996 ) bracket the value used here .
however , the final residual maps in the near - ir bands ( fig . [ gcr_maps ] ) , and the gradients with respect to galactic latitude ( table [ tab_gcr_grad ] ) clearly indicate that the fsm has overaccounted for the stellar emission .
this suggests the possibility of adjusting the fsm to minimize these gradients . the simplest adjustment would be to apply a global scaling factor to the model intensities
such a factor could arise if there were an error in the flux attributed to 0magnitude stars in the fsm .
the scaling factors required to minimize the galactic latitude gradients of the residual maps are 0.859 , 0.792 , and 0.674 at 1.25 , 2.2 , and 3.5 @xmath0 , respectively . however , these adjustments are significantly larger than the @xmath204 gain uncertainties expected at these wavelengths ( _ cobe _ dirbe explanatory supplement , 1997 ) .
these scaling factors could also be attributed to errors in the absolute magnitudes of 0.165 , 0.253 , and 0.428 magnitudes for all source types at 1.25 , 2.2 , and 3.5 @xmath0 , respectively . if these scaling factors are applied to the fsm , then the residual intensities at high galactic latitude increase by factors of @xmath362 in all three bands .
alternatively , we can use the star count data from the seven 2mass fields to optimize the fit of the fsm to this independent data set . to minimize the @xmath101 statistic for the comparison of the @xmath56 and @xmath9 band data simultaneously
, we find that the fsm star counts ( number of stars / mag / deg@xmath63 ) need to be reduced by a factor of 0.952 to match the 2mass data .
such a change in the number density of stars in the fsm would eliminate about 1/2 of the gradient that is observed in the residual maps .
changes that optimize the fit to the 2mass star counts do not simultaneously minimize the residual gradients in the dirbe data .
none of these scaling changes were applied to either the intensities or the number densities of the stars in the fsm in the present analysis because we sought to keep the fsm tied closely to the star counts , and because we are unable to determine the physical cause of the apparent discrepancy between the fsm and the dirbe data .
the ir emissivity and colors of the ism derived in this paper differ somewhat from those found in other analyses . using the dirbe data , but a different method for removing the ipd emission , boulanger et al .
( 1996 ) find a 100 @xmath0 emissivity of 15.9 nw m@xmath26 sr@xmath27 / 10@xmath205 cm@xmath26 , which is about 15% lower than the average for the nep and lockman hole regions ( table [ tab_b8_hi ] ) .
dwek et al .
( 1997 ) presented an analysis of an ism spectrum that was derived in a similar manner to that presented in table [ tab_bx_b8 ] , but using an earlier version of the ipd emission removal than that used in this work ( paper ii ) .
the high latitude ism spectrum presented here is considered an improvement on that presented by dwek et al .
( 1997 ) , though differences are generally @xmath206 10% and within the adopted uncertainties .
dwek et al .
( 1997 ) find that the ism spectrum is well fit by emission from silicate grains , graphite grains , and pahs , all heated by the mean interstellar radiation field .
the near - ir emission ( @xmath207 ) provides strong evidence that the pah abundance is higher than expected from most previous studies .
the amount of carbon in graphite grains , pahs , and as c@xmath208 ions is found to be consistent with cosmic abundances . at 240 @xmath0
the results of the two - component ism model ( ism2 ) indicate that it is important for a model of galactic ir emission to account for spatial variation in the dust temperature .
the contrast between figure [ gcr2_10 ] and figure [ gcr_maps ] shows this is especially relevant at low latitude ( @xmath209 ) .
however , because of the increased noise in the ism2 model , it remains unclear how important temperature variation is at high latitudes .
a two - component model , similar in form to the ism2 model used here , is also required to model the galactic emission observed by the _ cobe_/firas experiment ( fixsen et al .
another recent study using both dirbe and firas data ( lagache et al . 1998 ) decomposes the ism emission into warm ( @xmath3617.5 k ) and colder ( @xmath3615 k ) components .
the distribution of this cold component ( figure 5 of lagache et al .
1998 ) is very similar to that of the 140 and 240 @xmath0 residual emission after removal of the one - component ism model ( figure 12 ) . in this two temperature model ,
the apparent temperature variations revealed in the long wavelength colors are ( at least in part ) a result of varying proportions of the warm and colder ism components on different lines of sight .
there are several numbers presented in this paper that should not be used as estimates of the cib , despite their appearances .
these include extrapolation of the @xmath68 or @xmath210 gradients discussed in sections 4.3 and 5.3 , intercepts of the correlations between @xmath109(100 @xmath0 ) and @xmath89(h i ) ( and other ism tracers ) discussed in section 5.2 , and the intercepts of the correlations of other dirbe intensity maps with the 100 @xmath0 ism template discussed in section 5.3 .
the main reason these numbers are not reliable estimates of the cib is that they are derived from regions chosen for characterization of the galactic foreground , not for accuracy or completeness of removal of the galactic and other foregrounds .
frequently these regions include entirely or in part areas where the galactic emission is relatively strong , and small fractional errors in the foreground removal could still have large influences on the derived intensity of the cib .
a more useful investigation of the cib should focus on regions of the sky where errors in the foreground removal have minimal impact on the derived cib .
a proper analysis of the cib also needs a demonstration that the residual emission is isotropic , rather than just providing a single intensity estimate that may average over anisotropic defects in the foreground removal .
such a study of the cib is presented in paper i. any conclusions concerning the cib should be drawn from that paper .
we have modeled and removed the galactic ir emission in the dirbe data in preparation for analysis of the cib .
the procedures used were designed to preserve the emission of the cib in the residual maps , and not remove it inadvertently with the galactic emission .
the procedures concentrated on producing accurate results at high latitudes where galactic emission is weakest . at low latitudes , deficiencies in the models are clearly visible in the residual maps .
we find that the stellar emission of the galaxy is reasonably reproduced by our faint source model , which is based on the sky model ( wainscoat et al .
1992 ; cohen 1993a , 1994a , 1995 ) .
an offset of the sun by @xmath36 18 pc from the galactic plane is required to produce equal residual near - ir intensities at north and south galactic latitudes . however , there is clearly room for improvement in the geometry or calibration of the fsm .
we find that the ism can be fairly well modeled by a single spatial and spectral component if we constrain our study to high galactic latitudes . at 240 @xmath0 , a model of the ism with two spatial components
can produce a much more complete subtraction of the ism , extending to low galactic latitudes .
the two spatial components can combine to produce a range of color temperatures across the sky .
this shows that a complete model of the ism needs to be able to account for a continuous range of dust temperatures .
we are unable to detect any ir emission associated with low density ionized gas at high galactic latitudes .
the authors gratefully acknowledge the contributions over many years of the many engineers , managers , scientists , analysts , and programmers engaged in the dirbe investigation .
the national aeronautics and space administration / goddard space flight center ( nasa / gsfc ) is responsible for the design , development , and operation of _
cobe_. scientific guidance is provided by the _
cobe _ science working group .
gsfc is also responsible for the development of the analysis software and for the production of the mission data sets .
h. t. freudenreich is thanked for his work on developing and implementing the procedures used for blanking the bright sources ( section 4.1 ) .
we thank m. cohen , who provided results from the sky model and technical assistance , which were very useful for the development of our faint source model .
we also thank m. skrutskie , s. price , r. cutri , m. egan , and others in the 2mass project for making some of their prototype camera data available for this research .
the data described in this paper are available to the public through the nssdc _ cobe _ homepage website at @xmath33 .
the mission - averaged zodi - subtracted residual skymaps are contained in the ` zodi - subtracted mission average ( zsma ) ' maps ; the results generated by the faint source model are contained in the ` faint source model ( fsm ) ' maps .
in section 5.2.2 , tracers of ionized gas toward the lockman hole region were used to estimate the possible error in the zero level of the 100 @xmath211 m ism template , as determined from intercepts of @xmath212(100 @xmath211 m )
@xmath89(h i ) correlations , that result from our neglect of emission from h ii associated dust .
the 100 @xmath211 m emissivity per h nucleus was assumed to be the same in the ionized gas and the neutral atomic gas . here
, an attempt is made to determine the mean emissivity per h nucleus for the ionized gas at high galactic latitudes .
previous efforts to measure far - infrared emission from the diffuse ionized medium have been inconclusive .
boulanger et al .
( 1995 ) found that the extended , low density h ii region around the high latitude early b star @xmath87 vir has associated far - ir emission at a brightness that is consistent with a normal dust abundance , but this region may not be representative of the high latitude h ii in general .
boulanger et al .
( 1996 ) searched for far - ir emission from the ionized medium by looking for a @xmath213 dependence in residuals of their high latitude far - ir@xmath89(h i ) fits , which used dirbe and firas data from 100 to 1100 @xmath211 m .
from the slopes of their @xmath213 fits , they placed an upper limit on the far - ir brightness of a component of the h ii gas that is not spatially correlated with @xmath89(h i ) .
they found this limit to be consistent with a normal dust abundance in the ionized medium if about half of the far - ir emission from the ionized medium is uncorrelated with @xmath89(h i ) .
determination of the emissivity per h nucleus in the h ii gas by this approach would require knowledge of the latitude dependence of h@xmath107 associated emission , the latitude dependence of emissivity per h nucleus in the h i gas , and the degree of correlation of @xmath89(h ii ) with @xmath89(h i ) , which may also vary with latitude . also using the firas data , fixsen et al .
( 1998 ) modeled the high latitude ism emission with a combination of 158 @xmath0 [ c ii ] line emission , a tracer of ionized gas , and linear and quadratic terms in @xmath89(h i ) , a tracer of the neutral gas .
their analysis found that most of the high latitude far - ir emission correlated with the quadratic term of the h i column density , and essentially no continuum emission correlated with the [ c ii ] line emission .
there have been a number of studies of correlations between far - ir and h@xmath87 emission or between far - ir and microwave ( largely free - free ) emission at high latitudes ( e.g. , kogut et al .
1996a , 1996b ; de oliveira - costa et al .
1997 ; mccullough 1997 ; kogut 1997 ) .
interpretation of these correlations is not straightforward because they do not separately account for the far - ir emission from the dominant h i phase of the interstellar gas .
sodroski et al . (
1997 ) derived mean far - ir emissivities for the extended low density ionized gas at low galactic latitudes by decomposing dirbe maps into components that correlate with h i , co , and radio continuum emission .
this gas is probably associated with star forming regions , and is probably much denser and subject to a stronger radiation field than the diffuse warm ionized medium observed at high latitudes ( e.g. , heiles , reach , and koo 1996 ; lockman , pisano , and howard 1996 ) . here
, we use observations of h@xmath87 intensity at high latitudes and observations of dispersion measure of pulsars at high @xmath124 as tracers of the ionized gas . we compare results of regression fits of the forms @xmath214 @xmath16 = @xmath215(h i ) + @xmath155 and @xmath214 @xmath16 = @xmath216(h i ) + @xmath217(h ii ) + @xmath218 for data along high latitude lines of sight , where @xmath219 @xmath16 is dirbe intensity after subtraction of ipd emission and @xmath220(h ii ) is either extinction - corrected h@xmath87 intensity or pulsar dispersion measure .
the derived value of the parameter @xmath221 can provide an estimate of the 100 @xmath211 m emissivity per h nucleus of the ionized medium , and the difference between parameters @xmath155 and @xmath218 provides an estimate of error in the zero level of 100 @xmath211 m ism emission that is inferred if the ionized component is not accounted for .
each tracer of ionized gas has advantages and disadvantages .
pulsar dispersion measure is a measure of ionized gas column density , but it pertains to a single line of sight rather than the solid angle sampled by the dirbe beam , and it can significantly underestimate the total ionized column density along the line of sight if the pulsar is not above most of the ionized gas layer .
the h@xmath87 data used here were measured with a beam comparable to the dirbe beam , and sample the entire path through the galaxy .
however , h@xmath87 emissivity is proportional to the square of the ionized gas density while 100 @xmath211 m emissivity is proportional to dust density .
another disadvantage of using h@xmath87 is that it can be affected by extinction .
an extinction correction has been applied to the h@xmath87 intensities used here assuming the ratio of extinction coefficient to h@xmath87 emissivity is constant along each line of sight ( uniformly mixed extinction and emission ) , and the optical depth at h@xmath87 is 0.04 @xmath89(h i ) where @xmath89(h i ) is in @xmath121 cm@xmath26 . with these assumptions ,
the extinction correction factors range from 1.02 to 1.24 for the lines of sight used here .
the regression fitting has been done independently for five different regions or samples of positions .
sample 1 consists of 95 positions in a @xmath131 region centered at @xmath132 , @xmath133 , which was mapped in h@xmath87 with a @xmath222 beam by reynolds ( 1980 ) .
the region contains a number of elongated h@xmath87 enhancements above a smooth background , and positions toward the brightest feature were excluded from our analysis because its estimated electron density is about 5 times higher than that of the other features ( reynolds et al .
sample 2 is a sample of 27 positions within 20 degrees of @xmath87 vir ( at @xmath223 , @xmath224 ) that were observed in h@xmath87 by reynolds ( 1985 ) .
two positions within 1 degree of @xmath87 vir were not included in this sample because an enhanced 60 @xmath211 m to 100 @xmath211 m intensity ratio is observed at these positions .
sample 3 is a sample of 22 pulsar lines of sight at @xmath225 that were observed in h@xmath87 by reynolds ( 1984 , 1991b )
. samples 4 and 5 are different samples of pulsar lines of sight at @xmath226 , for which pulsar dispersion measure is used instead of h@xmath87 as the tracer of ionized gas .
two different criteria were used to select pulsars likely to be at high @xmath124 distances , so the dispersion measure traces most of the ionized gas layer .
the first criterion is based on estimates of the @xmath54 component of the characteristic pathlength @xmath227 occupied by ionized gas between the sun and the pulsar , @xmath228 , where @xmath229 is the distance of the pulsar . following an approach similar to that of reynolds ( 1977 , 1991b ) , lower limits to @xmath227
were calculated for lines of sight to pulsars observed in h@xmath87 by reynolds ( 1984 , 1991b ) , using @xmath230 . here
@xmath231 is the pulsar dispersion measure and @xmath232 is the emission measure along the entire line of sight through the galactic disk , determined from the extinction - corrected h@xmath87 intensity assuming an electron temperature of 8000 k. the calculated lower limit on @xmath233 ranges from 200 to 390 pc for lines of sight toward pulsars in four globular clusters at @xmath234 4 kpc , and it is assumed that other pulsars with @xmath235 200 pc are also above most of the ionized gas .
sample 4 consists of the 9 pulsar lines of sight selected by this criterion .
data for this sample are listed in table [ tab_pulsars ] .
sample 5 consists of 22 pulsar lines of sight at @xmath226 for which @xmath231 @xmath236 is greater than @xmath118 cm@xmath26 , the minimum value of @xmath231 @xmath236 for the globular cluster pulsars in table [ tab_pulsars ] .
some of the pulsars selected by this criterion may not be above most of the ionized gas layer if density enhancements in the ionized medium are common at high latitudes .
the enhancements observed by reynolds et al .
( 1995 ) in the @xmath237 region at @xmath132 , @xmath133 have estimated @xmath89(h ii ) values ranging from 0.2 to 0.6 @xmath238 cm@xmath26 .
results of the regression fits for samples 13 are shown in table [ tab_hi_ha ] and results for samples 4 and 5 are shown in table [ tab_hi_dm ] .
the errors listed were determined from the size of the 68% joint confidence region in parameter space .
the fits for samples 1 and 2 used h i column densities from the leiden / dwingeloo survey ( hartmann and burton 1997 ) , smoothed to the resolution of the dirbe data , and the fits for the other samples used h i column densities from the bell laboratories survey ( stark et al .
except for the region around @xmath87 vir , the derived coefficient of the ionized gas tracer is consistent with zero for each sample .
( for comparison , the mean 100 @xmath211 m emissivities derived by sodroski et al .
( 1997 ) for the extended low density ionized gas at low galactic latitudes correspond to @xmath239 = @xmath240 nw m@xmath26 sr@xmath27/rayleigh inside the solar circle and @xmath241 nw m@xmath26 sr@xmath27/rayleigh outside the solar circle , assuming an ionized gas temperature of 8000 k. ) the derived values of the isotropic terms @xmath155 and @xmath218 agree with each other within the uncertainties , and they are also consistent with the value of 19.8 nw m@xmath26 sr@xmath27 used for making the 100 @xmath211 m ism template ( section 5.3 ) .
figure [ sample_corrs ] illustrates correlations among the variables used for the regression fit for samples 1 , 2 , and 4 . figure [ sample_corrs]a shows linear fits to @xmath212(100 @xmath211m)@xmath89(h i ) correlations , figure [ sample_corrs]b shows linear fits to h@xmath87@xmath89(h i ) or @xmath89(h ii)@xmath89(h i ) correlations , and figure [ sample_corrs]c shows the correlation between residuals of the fit in figure [ sample_corrs]a and residuals of the fit in figure [ sample_corrs]b .
significant trends are not seen in figure [ sample_corrs]c , except for the region around @xmath87 vir .
the results of the regression fits and of figure [ sample_corrs ] show that 100 @xmath211 m emission has been detected from the ionized region around @xmath87 vir , but has not been detected from the general ionized medium at high latitudes . for sample 1 , the three sigma upper limit of @xmath242 nw
m@xmath26 sr@xmath27/rayleigh corresponds to an upper limit on the 100 @xmath211 m emissivity per h nucleus in the ionized medium of @xmath243 nw m@xmath26 sr@xmath27/10@xmath205 cm@xmath26 , assuming electron density @xmath244 @xmath127 and electron temperature of 8000 k ( reynolds et al .
the results for sample 3 give @xmath245 nw m@xmath26 sr@xmath27/10@xmath205 cm@xmath26 under the same assumptions .
comparison with the values of c found for these samples suggests that the 100 @xmath211 m emissivity per h nucleus is smaller in the ionized medium than in the neutral atomic medium .
the results of the fits for samples 4 and 5 are consistent with this conclusion , but the errors in the derived values of @xmath246 are large .
differences in the electron density distribution along different lines of sight may cause errors in the results of the regression fits that use h@xmath87 intensity as the ionized gas tracer . for the region of sample 1 ,
a crude model of the electron density distribution has been constructed .
for each of the elongated features of enhanced h@xmath87 emission , we use values derived by reynolds et al .
( 1995 ) for the mean electron density and the mean linear size along the line of sight .
the densities range from 0.08 to 0.23 @xmath127 . for the gas emitting the smooth h@xmath87 background
, we assume an electron density of 0.08 @xmath127 ( reynolds 1991b ) and determine its extent along each line of sight from the observed h@xmath87 intensity , allowing for the contribution of each enhancement and assuming a gas temperature of 8000 k. the regression fit for sample 1 was then repeated using ionized column density from the model instead of h@xmath87 intensity .
results are @xmath247 nw m@xmath26 sr@xmath27/10@xmath205 cm@xmath26 , @xmath248 nw m@xmath26 sr@xmath27/10@xmath205 cm@xmath26 , and @xmath249 nw m@xmath26 sr@xmath27 .
the electron density model is not unique , but the results are consistent with the conclusions that the 100 @xmath211 m emissivity per h nucleus is smaller in the ionized medium than in the neutral atomic medium , and that 100 @xmath211 m emission from the ionized medium does not cause significant error in the zero level of 100 @xmath211 m ism emission inferred from the @xmath212(100 @xmath211 m )
@xmath89(h i ) correlation .
errors in the fit parameters can also occur if there is significant 100 @xmath211 m emission from h@xmath107-associated dust along one or more lines of sight in a sample , but for the samples used here it appears that such errors are small .
the regression fits were repeated excluding positions in each sample where the 100 @xmath211 m emission exceeds the h i correlated component of 100 @xmath211 m emission by more than about 20 nw m@xmath26 sr@xmath27 . reach et al .
( 1994 , 1997 ) have presented evidence that such excess far - ir emission generally traces molecular gas .
a small fraction of the positions was excluded for each sample , and the results were not significantly different from those of table [ tab_hi_ha ] and table [ tab_hi_dm ] .
other possible sources of error include errors in subtraction of ipd emission , and possible differences in the mean 100 @xmath211 m emissivity per h nucleus , within either the h i or h ii gas , for different lines of sight .
such differences could be caused by differences in mean dust temperature or mean dust - to - gas mass ratio .
use of a small patch of the sky with a large number of lines of sight may reduce these sources of error , so the results for sample 1 are considered to be the most reliable . to summarize
, a multiple regression analysis has been used to estimate an upper limit on the 100 @xmath211 m emissivity per h nucleus in the diffuse ionized gas at high latitudes .
an upper limit of 12 nw m@xmath26 sr@xmath27/10@xmath205 cm@xmath26 is obtained for a @xmath237 region at @xmath132 , @xmath133 .
this value is 3/4 of the emissivity per h nucleus in the neutral atomic gas in the same region .
results for other samples of high latitude positions are consistent with this result , except a greater 100 @xmath0 emissivity is found for ionized gas in the vicinity of the early b star @xmath87 vir . if our derived upper limit on the h ii / h i emissivity ratio is valid for the general high latitude sky , the mean column density ratio @xmath89(h ii)/@xmath89(h i ) of about 1/3 ( reynolds 1991a ) implies that on average less than @xmath3620% of the total 100 @xmath0 emission observed at high latitudes comes from the ionized gas phase .
a low far - ir emissivity per h nucleus might be expected for the diffuse ionized medium since dust destruction by shocks is thought to be more efficient in the warm diffuse interstellar gas than in dense clouds ( e.g. , seab 1987 ; mckee 1989 ; jones et al .
1994 , 1996 ) .
future results from the wisconsin h - alpha mapper survey ( tufte et al .
1996 ) will be useful for extending the type of analysis presented here to large areas of the high latitude sky .
boulanger , f. , abergel , a. , bernard , j .-
p . , desert , f .- x .
, lagache , g. , puget , j .- l . , burton , w. b. , & hartmann , d. 1995 , in unveiling the cosmic infrared background , ed .
e. dwek , ( woodbury , ny : aip ) , p. 87 diffuse infrared background experiment ( dirbe ) explanatory supplement , 1997 , ed .
m. g. hauser , t. kelsall , d. leisawitz , and j. weiland , cobe ref . pub .
97-a , ( greenbelt , md : nasa / gsfc ) , available in electronic form from the nssdc at @xmath33 cccc 1.25 & before & @xmath250 & 0.611.25 & after & @xmath251 & -0.092.2 & before & @xmath252 & 0.562.2 & after & @xmath253 & -0.153.5 & before & @xmath254 & 0.453.5 & after & @xmath255 & -0.074.9 & before & @xmath256 & 0.244.9 & after & @xmath257 & -0.005 cccccc north & @xmath258 & @xmath259 & @xmath260 & 3.3 & 0.95ecliptic pole & @xmath261 & & & lockman & @xmath262 & @xmath263 & @xmath264 & 4.4 & 0.75hole & @xmath265 & & & @xmath266 & & @xmath267 & @xmath268 & 12.9 & 0.81@xmath269 & & & & ccccc 1.25 & & & & 2.2 & & & & 3.5 & @xmath270 & 17.6 & 0.79 & 4.9 & @xmath271 & 22.0 & 0.66 & 12 & @xmath272 & 15.2 & 0.94 & @xmath273 25 & @xmath274 & 8.5 & 0.93 & @xmath273 60 & @xmath275 & 4.0 & 0.91 & @xmath276100 & @xmath277 & & & 140 & @xmath278 & 57.2 & 0.52 & @xmath279240 & @xmath280 & 19.4 & 0.62 & @xmath279 cccc 3.5 & before & @xmath281 & -0.07 3.5 & after & @xmath282 & -0.15 4.9 & before & @xmath283 & -0.005 4.9 & after & @xmath284 & -0.13 12 & before & @xmath285 & -0.17 12 & after & @xmath286 & -0.37 25 & before & @xmath287 & -0.33 25 & after & @xmath288 & -0.45 60 & before & @xmath289 & 0.50 60 & after & @xmath290 & 0.03100 & before & @xmath291 & 0.51100 & after & @xmath292 & -0.04140 & before & @xmath293 & 0.43140 & after & @xmath294 & 0.16240 & before & @xmath295 & 0.46240 & after & @xmath296 & 0.22 ccccccc 1 & @xmath297 & @xmath298 & @xmath299 & @xmath300 & @xmath301 & 95 2 & @xmath302 & @xmath303 & @xmath304 & @xmath305 & @xmath306 & 27 3 & @xmath307 & @xmath308 & @xmath309 & @xmath310 & @xmath311 & 22 | the cosmic infrared background ( cib ) is hidden behind veils of foreground emission from our own solar system and galaxy .
this paper describes procedures for removing the galactic ir emission from the 1.25 240 @xmath0 _ cobe _ dirbe maps as steps toward the ultimate goal of detecting the cib .
the galactic emission models are carefully chosen and constructed so that the isotropic cib is completely retained in the residual sky maps .
we start with dirbe data from which the scattered light and thermal emission of the interplanetary dust ( ipd ) cloud have already been removed .
locations affected by the emission from bright compact and stellar sources are excluded from the analysis .
the unresolved emission of faint stars at near- and mid - ir wavelengths is represented by a model based on galactic source counts .
the 100 @xmath0 dirbe observations are used as the spatial template for the interstellar medium ( ism ) emission at high latitudes .
correlation of the 100 @xmath0 data with h i column density allows us to isolate the component of the observed emission that is associated with the ism .
limits are established on the far - ir emissivity of the diffuse ionized medium , which indicate a lower emissivity per h nucleus than in the neutral medium . at 240 @xmath0
, we find that adding a second spatial template to the ism model can greatly improve the accuracy of the model at low latitudes .
the crucial product of this analysis is a set of all - sky ir maps from which the galactic ( and ipd ) emission has been removed .
we discuss systematic uncertainties and potential errors in the foreground subtraction process that may have an impact on studies seeking to detect the cib in the residual maps .
submitted to the astrophysical journal + received : 8 january 1998 , accepted : 12 may 1998 |
calculations of radiative energy loss via medium - induced gluon radiation ( see e.g. @xcite and refs . therein ) off massive quarks were performed in @xcite .
two effects compete against each other .
the gluon formation time , @xmath1^{-1}\sim \left[t_g^{\rm form}(m= 0)\right]^{-1}+m_q^2/e_q$ ] , of a massive quark ( with mass @xmath2 and energy @xmath3 ) is shorter than that of a massless quark , so gluon emission in the former is less suppressed by the lpm effect than that in the latter . on the other hand , gluon radiation off a massive quark is suppressed at angles smaller than the so - called dead cone angle @xmath4 @xmath5 @xmath6 , similarly to the case in vacuum @xcite .
the generic result of this competition is that a massive quark loses less energy in a medium than a massless one .
but the mentioned calculations do not consider the effects of the interference between emitters and therefore the extension to multi - gluon radiation relies on ad hoc conjectures . in order to investigate these interference effects in vacuum ,
a quark - antiquark ( @xmath0 ) antenna was considered , see @xcite and references therein .
the soft gluon radiation spectrum off a massless @xmath0 antenna in vacuum exhibits angular ordering ( ao ) : radiation is suppressed at @xmath7 @xmath8 @xmath9 ( after averaging over the gluon azimuthal angle ) , where @xmath7 is the angle between the emitted gluon and the parent quark ( antiquark ) and @xmath9 is the antenna opening angle .
the massless antenna spectrum diverges when @xmath7 @xmath10 @xmath11 and @xmath12 @xmath10 @xmath11 .
when the quarks have a non - zero mass , ao is modified and the collinear divergence disappears due to the dead cone effect , but the soft divergence remains .
the result was shown to hold for arbitrary color representations of the @xmath13 pair .
medium - induced soft gluon radiation off a massless @xmath0 antenna has been studied recently @xcite .
the spectrum exhibits antiangular ordering ( aao ) - there is no collinear divergence ( @xmath7 @xmath8 @xmath9 @xmath8 @xmath11 ) - but the soft divergence persists , at variance with the results in @xcite . due to space limitations , here we show some selected results for a massive quark - antiquark pair at first order in opacity .
in fig . [ aspec ] we show the three types of diagrams that we consider .
they correspond to the independent emission off the quark ( antiquark ) , denoted as independent and the interference between the quark and antiquark , denoted as interference i and interference ii , respectively .
the total number of diagrams computed is 96 .
examples of feynman diagrams representing the three contributions to the antenna spectrum .
the cross denotes the position of scattering , the l.h.s . of the dashed line
is the amplitude and the r.h.s . of the dashed line
is the complex conjugate amplitude , with the dashed line being the cut.,scaledwidth=80.0% ] the average radiative energy loss for gluons with energies @xmath14 is @xmath15 dependence of the medium - induced radiative relative energy loss on the antenna opening angle .
the parameters are : debye mass @xmath16 gev and medium length @xmath17 fm for the plots on the left ( right ) .
the solid curves correspond to the massless antenna , the dotted curves to the bottom antenna , the dashed curves to the massless independent spectra and the dash - dotted curves to the bottom independent spectra . from top to bottom , the values used in eq .
( [ deoe ] ) for @xmath18 are 0 , 2 and 6 gev , while those for @xmath19 are 2 , 6 and @xmath3 . ]
the ratio @xmath20 as a function of @xmath9 is shown in fig .
[ hoa ] , where @xmath21 gev for the independent emitter case and for the antenna the spectrum is divided by 2 as it gets contributions from both the quark and antiquark . in the soft and moderate gluon energy regions ( @xmath22 gev and @xmath23 gev @xmath24 gev ) , both the massless and the massive antenna spectra grow monotonously with an increasing opening angle @xmath9 and the former is larger than the latter . in the hard gluon radiation sector ( @xmath25 gev @xmath26 ) ,
the situation is similar for large medium parameters while , for small medium parameters ( lower left plot ) , there is a crossing between the massless and the bottom antennas due to the lpm effect which results in a larger energy loss for larger masses as discussed in the introduction and found previously e.g. in @xcite .
therefore , for large medium parameters the dead cone effect dominates over the lpm in all cases . both massless and massive antenna results approach the ones from independent emitters when @xmath9 is large , showing that the interference between the quark and the antiquark of the antenna reduces with an increasing opening angle .
besides , in the soft gluon emission region and for large medium parameters , there is apparently more energy loss in the antenna than for independent emitters in both the massless and the massive cases .
this reflects the fact that the antenna spectrum exhibits a soft divergence while the independent spectrum is infrared finite . in the moderate and the hard gluon emission sectors , the antenna average energy loss increases and
gradually approaches the independent average energy loss with an increasing antenna opening angle @xmath9 , which indicates that more collimated projectiles lose less energy .
there is no radiation for the antenna when @xmath9 @xmath10 @xmath11 .
the size of the mass effect in the antenna is similar to the one for independent emitters .
in this contribution , we show some results of the medium - induced gluon radiation spectrum off a @xmath0 antenna at first order in opacity for the massive case . in this computation ,
performed in the high - energy limit , both the non - abelian lpm effect and the dead cone for massive quarks ( both contained in the medium - induced gluon spectrum off individual emitting partons - the bdmps - z - w / glv formalism ) , and the interference between emissions off the quark and antiquark , are included .
the antenna radiation is found to be dominated by that off independent emitters for large opening angles of the antenna and for large energies of the emitted gluon .
more collimated antennas lose less energy .
the phase space restriction for gluon emission implied by the dead cone effect is similar in the antenna and in the case of independent emitters .
the effect of the interference between different emitters is to generate predominantly soft radiation at large angles .
10 j. casalderrey - solana and c. a. salgado , acta phys .
b * 38 * , 3731 ( 2007 ) .
n. armesto , c. a. salgado , and u. a. wiedemann , phys .
d * 69 * , 114003 ( 2004 ) .
m. djordjevic and m. gyulassy , nucl .
a * 733 * , 265 ( 2004 ) .
zhang , e. wang , and x .-
wang , phys .
lett . * 93 * , 072301 ( 2004 ) .
y. l. dokshitzer , v. a. khoze , and s. i. troian , j. phys .
g * 17 * , 1481 ( 1991 ) .
l. dokshitzer and d. e. kharzeev , phys .
b * 519 * , 199 ( 2001 ) .
l. dokshitzer , v. a. khoze , a. h. mueller , and s. i. troyan , _
`` basics of perturbative qcd '' , gif - sur - yvette , france , ed .
frontires 1991_. y. mehtar - tani , c. a. salgado , and k. tywoniuk , phys .
lett . * 106 * , 122002 ( 2011 ) .
y. mehtar - tani , c. a. salgado , and k. tywoniuk , arxiv:1102.4317v1 [ hep - ph ] .
y. mehtar - tani and k. tywoniuk , arxiv:1105.1346v1 [ hep - ph ] .
j. casalderrey - solana and e. iancu , arxiv:1105.1760v2 [ hep - ph ] . | the medium - induced one - gluon radiation spectrum off a massive quark - antiquark ( @xmath0 ) antenna traversing a colored qcd medium is calculated in this contribution . the gluon spectrum off the antenna computed at first order in the opacity expansion is collinear finite but infrared divergent , which is different from the result obtained from an independent emitter which is both infrared and collinear finite .
the interference between emitters dominates the soft gluon radiation when the antenna opening angle is small and the emitted gluon is soft , whereas the antenna behaves like a superposition of independent emitters when the opening angle is large and the radiated gluon is hard . as a phenomenological consequence
, we investigate the energy lost by the projectiles due to the radiation . in general , the size of the mass effects is similar in both cases . |
techniques for accelerating ions employing high power lasers are currently attracting considerable interest , thanks to the rapid progress in laser technologies over the past couple of decades . with currently accessible intensities ( in excess of @xmath2 ) of short pulse lasers
, ions can be accelerated to high energies ( up to several tens of mev / nucleon ) with promising beam parameters for several potential applications in science , industry and healthcare @xcite .
a number of acceleration mechanisms , such as target normal sheath acceleration ( tnsa ) @xcite , radiation pressure acceleration ( rpa ) @xcite , break - out afterburner ( boa ) @xcite , are objects of intense investigation both experimentally and theoretically in order to improve the beam parameters .
laser accelerated ion beams from solid targets are typically multi - species , either due to the chemical composition of the target material or due to the layer of contamination over the target ( typically composed of water vapour and hydrocarbons)@xcite .
one of the diagnostics commonly used to characterize these beams is the thomson parabola spectrometer ( tps)@xcite , which has the unique ability to energy resolve the ion spectra while discriminating ions with different charge - to - mass ratio .
ratios , respectively , as shown in * ( b)*. a set of filters were used over the detector in order to discriminate lighter ions from the overlapping heavier species with the same @xmath3 ratio.,scaledwidth=48.0% ] in a typical tps ( fig .
[ fig : tpis_setup ] ) , a pencil beam of ions , selected by the pinhole located at its entrance , travels through regions of parallel magnetic and electric fields , applied transversely to the beam axis , before reaching the detector plane . with reference to fig . [
fig : tpis_setup ] , the magnetic field determines the position of the ions along the @xmath4-axis , depending on their energy per nucleon ( @xmath5 ) ( or longitudinal velocity ( @xmath6 ) ) , while the electric field deflects the ions along the the @xmath7 axis according to their charge(@xmath8 ) to mass(@xmath9 ) ratio . assuming that the inhomogeneities of the fields are negligible and that the ion energies are not relativistic
, particles with given @xmath3 will lay on a parabolic trace on the detector plane described by the expression @xmath10 where @xmath11 and @xmath12 are the electric and magnetic fields , @xmath13 , @xmath14 , @xmath15 and @xmath16 are the dimensions of relevant sections of the tps as labelled in fig .
[ fig : tpis_setup ] . since the locus of the parabolic ion traces produced by the tps is a function of the charge - to - mass ratio ( @xmath3 ) of the ion species , the traces of the species with the same @xmath3 will overlap at the detector plane , preventing their spectra to be characterised .
this is a fundamental limitation of a tps when employed in high intensity laser plasma experiments , where several ion species around the interaction region are simultaneously accelerated to high energies @xcite .
examples of fully ionized ions with the same @xmath3 ratios which are typically observed in laser - plasma experiments are d@xmath17 , c@xmath0 and o@xmath1 ( @xmath3=1/2 ) and c@xmath18 , o@xmath19 ( @xmath3=1/4 ) .
characterization of the overlapping ion species individually is in many cases important not only in order to understand the underlying acceleration mechanism , but also to facilitate the analysis of any data obtained from secondary processes involving these ion species .
this is the case for laser driven deuterium ions ( d@xmath17)@xcite , which are being used to explore the potential of compact , laser - based sources of neutrons@xcite with possible applications in science @xcite , industry @xcite , security @xcite or healthcare @xcite .
the most commonly used targets to produce energetic beams of d@xmath17 ions are deuterated plastic foils , and a direct measurement of the d@xmath17 ion spectra with a tps is non - trivial due to the overlapping c@xmath0 and o@xmath1 ions from the target bulk and contaminant layers .
discrimination of d@xmath17 ions from the overlapping heavier ions can be achieved by filtering out the heavier ions before reaching the detector , for instance , by using a metal foil of appropriate thickness which can preferentially stop heavier ions .
this is possible as , along a ion trace corresponding to a given value of @xmath3 , all the ions will have the same energy per nucleon at any given point of the trace . in principle
, heavier species can therefore be prevented from reaching the detector by exploiting differences in stopping power of the different ions , if a suitable filter can be selected .
however , using a metal foil with a single , fixed thickness in front of the detector may not be sufficient to recover the full d@xmath17 spectrum , depending on the energy range covered by the tps .
therefore , a possible solution is to filter differentially different sections of the ion trace on the detector . in this paper
, we report on the characterisation of the complete spectrum of d@xmath17 ions generated from a laser solid interaction by employing a differential filter ( df ) over fuji image plate detectors @xcite , which are commonly used in such experiments .
the paper is organised as follows .
after a brief description of the experimental setup in section [ sec : experimentalsetup ] , section [ sec : filtering ] describes the concept of differential filtering and the design used in the experiment .
section [ sec : calibration ] reports the absolute calibration of the image plate detector for d@xmath17 ions over a wide range of energies , which was obtained in a single shot by using slotted cr-39 nuclear track detectors placed over the image plates .
finally , a typical d@xmath17 spectrum is shown in section [ sec : application ] , which was obtained by the analysis of a typical tps data from the experiment .
the experiment was carried out at the rutherford appleton laboratory ( ral ) , stfc , uk by employing the petawatt arm of the vulcan laser system .
the data presented in this paper were taken by irradiating the laser onto @xmath20 thick deuterated plastic ( cd ) foil targets . using a @xmath21 off - axis parabolic mirror ,
the laser was focussed down to @xmath22 full width at half maximum spots on the target , delivering peak intensity in excess of @xmath2 .
the ions accelerated by the interaction of the intense pulse with the cd target were diagnosed by employing several tpss , at different angles with respect to the laser axis . as discussed before ,
such interaction produces a multi - species ion source of broad energy spectrum .
therefore , the accelerated deuterium ions were discriminated from the heavier ones with the same @xmath3 value ( such as c@xmath0 , o@xmath1 , etc .
present in both the target bulk and contaminant layers ) by using differential filtering in front of the detector of each tps .
the df ( as discussed in section [ sec : filtering ] ) was attached to a frame which was directly screwed onto the detector drum of the tps , and served as a clamp to hold the detector plate ( see fig .
[ fig : tpis_setup ] ) . in this configuration
the position of the df always remains fixed with respect to the tps , which is crucial for filtering ions in specific energy bands .
bas - tr imaging plates ( ip ) @xcite were used as detectors .
the ips were wrapped with @xmath23 thick al foils to avoid their exposure to ambient light after irradiation . in the shots taken for image plate calibration , slotted cr-39 track detectors
@xcite were placed between the df and the ip .
the cr-39s were slotted uniformly along the energy - dispersion axis in order to select small parts of the ion tracks for absolute particle counting .
more details about the technique will be discussed in section [ sec : calibration ] .
as discussed before , tracks of ions with the same @xmath3 value overlap on the detector , where the particles at a given point of the track share the same energy per nucleon .
this allows using a metal foil of suitable thickness to stop the heavier ions while allowing the lighter ions to pass through due to the higher stopping power of the heavier ions .
however , due to the broad energy spectrum of the ions produced in such experiments , a single filter can not be used across the full spectrum , as the given filter foil would either be too thin to filter the heavier ions at higher energies or too thick to let lower energy deuteron ions to pass through .
the design of a differential filter is based on using several filters of suitable materials and/or thicknesses in different regions of the detector , set along the energy dispersion axis of the detector .
the requirement in this case is that for each zone , with boundaries defined by a pair of y coordinates @xmath24 , such a filter should be capable of stopping the heavier ions reaching the detector plane within this region .
it is to be noted that a df has to be designed specifically for a given tps and for the ion species under consideration , as those are the parameters that defines the energy of the ions to filter at each point of the detector . in this paper
we discuss filtering of c@xmath0 , o@xmath1ions in order to obtain a clean track of d@xmath25 on the detector . in the experiment ,
the tps was setup ( with @xmath12 = 0.988 t , @xmath11 = 11 kv / cm , @xmath13 = 50 mm , @xmath14 = 407 mm , @xmath15 = 150 mm , @xmath16 = 307 mm ) to diagnose d@xmath26 ions from @xmath271 mev / nucleon up to several tens of mev / nucleon , spread over 55 mm along the energy dispersion axis of the detector . considering the energy dispersion of the d@xmath28 and c@xmath29 ions , the 55 mm detection window
was divided into 4 regions , where a suitable filter could be found for each region .
the filter foils for each region were selected by considering the stopping power of both deuterium and carbon ions , obtained via srim simulations @xcite , as shown in fig .
[ fig : filtering ] . ( blue ) and 70 @xmath30 ( red ) aluminium foils , respectively .
the grey blocks represent the regions selected for filtering carbon ions from the low energy part of the deuterium track using the two filters.,scaledwidth=48.0% ] the set of four filters was selected to cover the entire energy range of d@xmath26 ions detectable by the tps as listed in table [ table : filters ] .
it is to be noted that the choice of filter materials and thicknesses were based on their availability during the experiment .
other materials of suitable thicknesses can be found for similar performances , however , higher z and thinner foils would be preferred in order to decrease lateral straggling of ions due to multiple small - angle scattering inside the filters , which may reduce the energy resolution of the diagnostic .
the filters can be assigned to energy bins far from their filtering limits , which was the case in our experiment .
overall , slight variations around the thickness and size of the chosen filter is acceptable , making the filter selection and df deployment easier .
.the four foils selected for the df used in the experiment for different energy bins of d@xmath26 ions in order to filter carbon and heavier ions [ cols="^,^,^,^ " , ] finally , in order to construct the df , the energy bounds for each of the energy bins were converted into corresponding y coordinates on the detector by taking into account the energy dispersion of the tps .
a schematic of the df used in the experiment is shown in the fig .
[ fig : filterexample ] , along with typical data obtained in the campaign . track '' and appearance of other ion species in the gap between two filter foils.,scaledwidth=48.0% ] as can be seen in table [ table : filters ] , a small gap was kept between the selected energy bins , corresponding to @xmath271 mm gap between the filters over the ip .
the gaps between the filters were made not only to check the filter performance , but also to identify the location of different filters from their shadows in the scanned image of the ip .
one can notice the abrupt increase of the signal in the ion trace corresponding to @xmath3=1/2 and the appearance of other ion species tracks ( such as c@xmath31 , o@xmath32 , etc . ) in the gap between consecutive filter sheets ( fig .
[ fig : filterexample]c ) .
this is evidence of the effectiveness of the df design used in the tps for isolating deuterium ion spectra from the other ion species with the same @xmath3 value .
once the carbon and heavier ion species have been filtered from the d@xmath26 track by using the df , the number of deuterium ions incident on the detector need to be calculated using the detector response ( in our case ips ) .
the ip detectors are based on the principle of photo - stimulated luminescence ( psl ) @xcite .
when ionising radiation is incident on the phosphor layer of the ip ( usually europium - doped barium fluorohalide phosphor ) , electrons are excited to higher energy levels in the crystal lattice where they remain in a metastable level until stimulated by a secondary illumination of appropriate wavelength .
an accurate theoretical response function of the detector is difficult to obtain due to the complexity of the mechanism of electron excitation and degree of trapping as a function of the type and energy of the incident ionising radiation .
for this reason , a direct calibration method was performed to calibrate the ip response for deuterium ions .
the calibration curve refers to a relationship between the psl signal produced in the ip per incident particle versus the energy of the particle at the detector .
the calibration method , as used previously by several groups @xcite , involves using cr-39 nuclear track detectors to compare the psl signal obtained from the ip with the number of tracks produced in the cr-39 .
this method provides an absolute ion calibration , as each track produced in the cr-39 refers to a single ion . to obtain the ip response for a range of ion energies , a piece of cr-39 having closely spaced slots carved along the ip energy dispersion axis ( as shown in the fig .
[ fig : calibration_setup ] ) was placed in front of the ip @xcite .
the slotted cr-39 was used after the df , ensuring that only the d@xmath17 ions are exposed to the cr-39 .
several data sets were collected by employing this setup in different tps used in the experiment , where each set of cr-39 and ip data provided 4 - 5 data points towards the calibration curve .
the methods employed for obtaining the calibration curve from the experimental data are discussed below . in order to develop the ion tracks for counting ,
the cr-39 pieces were etched in 6 m sodium hydroxide ( naoh ) solution at a temperature of @xmath33 .
etching was done for short periods of time ( typically 5 - 10 mins ) , depending on the incident flux of particles on the cr-39 , to avoid overlapping of ion tracks . in order to facilitate track counting , sets of pictures of the front surface of cr-39
were taken using an optical microscope with a @xmath34 objective .
the number of particle tracks at any given location was then counted by using the particle analysis subroutine of imagej@xcite software .
the accuracy of this method was cross - checked by counting the tracks manually for several cases , which showed differences smaller than @xmath35 in every case .
scanning resolution .
the solid line represents the fading curve , which is the best fit to the data points.,scaledwidth=49.0% ] the ips were scanned using a commercial ip scanner ( fujifilm fla-5000 @xcite ) with 16-bit dynamic range and @xmath36 pixel size , which was converted from the scan value ( or quantum level ( ql ) ) to psl using the formula given by the manufacturer @xcite @xmath37 where @xmath38 is the scanning resolution , @xmath39 is the sensitivity of the scanner , @xmath40 is the latitude or level and @xmath41 is the bit value of the dynamic range of the image . due to
the spontaneous decay of excited electrons from their metastable state , energy stored in the ip decays over time after the irradiation .
as shown by several authors @xcite the fading of the ip signal can be describe by different rates of decay , form a very fast decay over several tens of minutes after the irradiation to a significantly slower rate of decay lasting for days
. the fading has usually been investigated by exposing the ip to radioactive sources ( e.g. alpha , gamma ) for a period of time ranging from few minutes to tens of minutes and scanning it after the exposure .
although this method is rigorous in describing the slow rate of decay , it does not allow sufficient accuracy for the measuremnt of the decay rate promptly after the irradiation .
moreover , it is likely that the fading rate close to the time of irradiation can vary significantly depending on several factors such as the psl signal strength at the time of exposure , duration of ionising radiation source , type of image plate , etc . in order to overcome these factors , the decay characteristic of the bas - tr image plate
was studied experimentally by exposing the ip to a laser driven pulsed x - ray source of ns burst duration , employing taranis laser facility at the queen s university belfast , uk .
small pieces ( few mm @xmath42 few mm ) of bas - tr ip were placed at a given distance from a laser target in order to illuminate the ips to a spatially homogeneous x - ray beam , while producing equivalent level of psl as obtained by the ions in our experiment .
after the exposure , the image plates were scanned at different times ranging from few minutes to few hours after the exposure .
as shown in fig .
[ fig : fading ] , psl signal suffers a rapid decay in the first half hour after the exposure , followed by a significantly slower decay over several hours .
although the trend of the fading curve broadly matches the measurements taken by other groups ( see the references @xcite ) , the agreement is best for long fading times of the order of hours . in our case
the tps were placed inside the main interaction chamber , which was kept under vacuum during the interaction .
this caused a significant delay in scanning the ips after the irradiation , which ranged between 30 minutes up to 2 hours for different shots .
therefore , instead of obtaining a calibration curve in terms of psl per ion at the time of exposure ( i.e. t = 0 ) , it was decided to set the reference time at @xmath43 minutes after irradiation , which is a typical time required for retrieving the ips from the vacuum interaction chamber for scanning .
based on the data shown in the fig .
[ fig : fading ] , the psl signal at the said reference point of time , i.e. 30 min . after exposure , was estimated by the empirical formula @xmath44 where psl(t ) represents the measured psl signal obtained from the ip scanned ` @xmath45 ' minutes after the irradiation .
as mentioned earlier , a calibration curve refers to the signal produced per ion as a function of ion energy per nucleon reaching the ip ( w@xmath46 ) . due to the use of the df before the detectors in our case ,
the energy of deuterium ions incident on the ip / cr-39 at any given point of the ion track was different from that defined by the tps energy dispersion .
therefore , for each data point obtained for the calibration curve , the energy of the d@xmath17 ion on the image plate , after passing through the df , was calculated by using srim simulation @xcite , as shown in fig . [ fig : filtering ] for the first two filter layers . employing the methods discussed in the subsections a - c ,
the psl@xmath47 per incident particle was obtained for several different d@xmath17 energies at the ip , as shown in fig .
[ fig : calibration ] .
the calibration data points were obtained over a wide range of psl values on the ip , which were divided into two groups depending on the signal strength , given by 0.02 - 0.08 psl/@xmath48 and 0.08 - 0.2 psl/@xmath48 in fig .
[ fig : calibration ] , corresponding to low and high particle fluxes , and they are consistent with previous works@xcite . it can be seen how both groups follow the same calibration function , which implies linear response of the image plate with respect to the flux for the signal levels in our experiment . in terms of the ion energy ,
the psl signal per particle for lower energy ions increases linearly with the incident ion energy , whereas it decreases slowly , following a power law , for higher energy ions .
this behaviour of ip response can be crudely explained on the basis of the finite thickness of the ip active layer and the bragg peak energy deposition process for ions .
the empirical calibration function thus obtained by the best fit to the data points is given by eq .
[ eq : fitting1 ] .
@xmath49 ion for different d@xmath17 energies on the ip .
the data points are divided into two groups depending on the psl signal on the ip , which shows the validity of the calibration over a range of ion flux from @xmath50 to a few ions/@xmath48 on the detector.,scaledwidth=49.0% ] where , @xmath51 represents the d@xmath17 ion energy reaching the ip .
although fig .
[ fig : calibration ] shows the calibration for d@xmath17 ions with energies up to 4 mev / nucleon ,
[ eq : fitting1 ] can be used to obtain d@xmath17 spectra up to significantly higher energies while using df , as the df tends to slow down the ions before reaching the ip .
once the calibration function had been obtained , it was used to analyse the d@xmath17 spectrum obtained in the experiment . as an example
, a typical spectrum obtained along the target normal axis has been analysed ( fig .
[ fig : applicationlin ] ) .
this spectrum corresponds to the interaction of a petawatt laser pulse with a thin ( 100 nm ) deuterated plastic ( cd ) target , using the same laser parameters as mentioned in section [ sec : experimentalsetup ] . in the case of the data shown in fig . [
fig : applicationlin ] , the ips were scanned @xmath27140 minutes after the irradiation and the signal was then re - normalised to the psl value at the 30 min reference time ( psl@xmath47 ) using eq .
[ eq : fading ] .
the ion tracks were measured from the parabola on the ip , and background was subtracted by using the psl values at the regions near the track . for each point on the ion track ,
the energies of the ions at the image plate were calculated taking into account the energy dispersion of the tps and the transmission prperties for each filter obtained by srim .
finally , the number of d@xmath17 ions for each energy was recovered using the calibration function given by the eq .
[ eq : fitting1 ] .
ions obtained in the experiment , together with the detection threshold of the diagnostics due to noise level of the detector .
dotted lines in the ion spectrum correspond to missing spectral information in the regions between adjacent filters.,scaledwidth=48.0% ]
in conclusion , a novel technique to obtain the spectrum of ion species over a broad energy range by removing overlapping traces in a thomson parabola ion spectrometer is demonstrated .
this method is based on differential filtering across the full range of energies produced , by which the lightest of the overlapping ions can be discriminated .
the technique can also be applied for filtering proton track in a magnetic spectrometers , or , to filter high energy part of an ion spectrum in the region where several ion tracks are merged due to insufficient track separation .
the method was employed in a recent experimental campaign to diagnose the deuterium spectrum obtained from the interaction of a high power laser with deuterated plastic targets , which produce a multi - species ion source , with a number of ion species ( such as d@xmath17 , c@xmath0 , o@xmath1 , etc . ) sharing the same @xmath3 value .
in addition to the d@xmath25 spectrum obtained using df , the bas - tr image plate was absolutely calibrated for d@xmath17 response against cr-39 detectors over a broad spectrum .
the authors acknowledge funding from epsrc [ ep / j002550/1-career acceleration fellowship held by s. k. , ep / l002221/1 , ep / e035728/1 , ep / k022415/1 , ep / j500094/1 and ep / i029206/1 ] , laserlab europe ( ec - ga 284464 ) , projects eli ( grant no . cz.1.05/1.1.00/483/02.0061 ) and opvk 3 ( grant no .
cz.1.07/2.3.00/20.0279 ) .
authors also acknowledge the support of the mechanical engineering staff of the central laser facility , stfc , uk .
37ifxundefined [ 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 * * , ( ) http://www.fujifilm.com @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) http://rsbweb.nih.gov/ij/ http://beamline.harima.riken.jp/bl45xu/web_old/info/ bas2500imgspec.pdf @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) | a novel method for characterising the full spectrum of deuteron ions emitted by laser driven multi - species ion sources is discussed .
the procedure is based on using differential filtering over the detector of a thompson parabola ion spectrometer , which enables discrimination of deuterium ions from heavier ion species with the same charge - to - mass ratio ( such as c@xmath0 , o@xmath1 , etc . ) .
commonly used fuji image plates were used as detectors in the spectrometer , whose absolute response to deuterium ions over a wide range of energies was calibrated by using slotted cr-39 nuclear track detectors . a typical deuterium ion spectrum diagnosed in a recent experimental campaign
is presented , which was produced from a thin deuterated plastic foil target irradiated by a high power laser . |
the atlas collaboration has recently observed a localised excess in the invariant mass distribution of pairs of fat jets , hereafter denoted by @xmath1 , around @xmath2 tev @xcite .
fat jets can be produced in the hadronic decay of boosted bosons @xmath3 , where the two quarks from the boson decay merge into a single jet . using jet substructure analyses ,
the fat jets are tagged as resulting from a boson decay .
in addition , in ref .
@xcite the jets are identified as @xmath4 or @xmath5 bosons if the jet mass @xmath6 satisfies @xmath7 gev or @xmath8 gev , respectively .
the excess in the @xmath9 spectrum appears for @xmath10 , @xmath11 and @xmath12 selections , with statistical significances of @xmath13 , @xmath14 and @xmath15 .
can be simultaneously tagged as @xmath4 and @xmath5 with these criteria , as the mass windows for @xmath4 and @xmath5 tagging partially overlap .
this indicates , in particular , that a @xmath10 signal can yield significant excesses in the @xmath12 and @xmath11 selections too . ]
these three channels are not independent and some events fall into two or even the three above categories .
a statistical combination of the three channels must take this fact into account , and has not been yet performed .
@xmath4 and @xmath5 bosons are known to decay leptonically , therefore a potential diboson resonance should also show up in ( semi-)leptonic channels .
but it does not .
for example , for spin-1 @xmath10 resonances with a mass @xmath16 tev the atlas collaboration has the following 95% confidence level ( cl ) upper limits on the cross section @xmath17 : * @xmath18 channel @xcite : @xmath19 fb , with an expected limit of 12 fb .
this corresponds to a @xmath13 excess . as mentioned above , this search is also sensitive to @xmath12 and @xmath11 resonances . * @xmath20 channel , @xmath21 @xcite : @xmath22 fb ( 11 fb expected ) .
this search is also sensitive to @xmath12 resonances .
* @xmath23 channel @xcite : @xmath24 fb ( 16 fb expected ) .
this search is also sensitive to @xmath11 resonances .
* @xmath25 channel @xcite : @xmath26 fb ( 24 fb expected ) .
therefore , the absence of any signal in the semi - leptonic channels excludes a significant excess in the fully hadronic one .
in particular , the @xmath27 channel has better sensitivity than the @xmath28 one the expected limit is smaller but deviations from the standard model ( sm ) predictions are not found .
( we note in passing that the `` definition '' of the fat jet @xmath1 is the same in the three atlas analyses @xcite . )
the tension between the atlas @xmath28 and @xmath27 searches can be quantified with a simple event counting . with the @xmath10 selection , there are 15 observed events in the hadronic channel with @xmath29 $ ] tev , for an expected background of approximately 7 events . with the colected luminosity of 20.3 fb@xmath30 , these eight extra events require a signal cross section @xmath31}$ ] times branching ratio and efficiency factors of @xmath32 }
\times \text{br}(w \to q \bar q ' ) \times \text{br}(z \to q \bar q ) \times \text{eff } = 0.39 ~\text{fb}\,.\ ] ] the superscript in @xmath31}$ ] emphasises that this is the cross section for @xmath29 $ ] tev alone .
( because the reconstructed invariant mass distribution of a 2 tev resonance is wider than the invariant mass interval considered , the actual @xmath10 cross section @xmath33 required to reproduce the excess is around twice larger . ) for the selection efficiency of 0.14 given in ref .
@xcite , which does not include the hadronic branching ratios , we find @xmath31 } = 6 $ ] fb .
now let us reverse the procedure to estimate the number of extra events that should show up in the @xmath27 channel . with the efficiency of 0.25 for @xmath34 given in ref .
@xcite , where @xmath35 includes electrons , muons and taus , @xmath31 } = 6 $ ] fb would yield 7 extra events in the mass interval @xmath36 $ ] tev , which practically coincides with the interval considered for the hadronic channel .
but there are six observed events in this mass interval , for an expected background around 6.5 events
. this would amount to a @xmath37 underfluctuation of the signal observed in the @xmath28 channel .
the cms collaboration has also looked for spin-1 @xmath10 resonances in the @xmath28 channel @xcite , using a slightly different strategy . for a resonance mass @xmath16 tev ,
the limit is @xmath38 fb , with an expected limit of 8 fb .
although there is some excess around @xmath39 tev , the cms upper limit on the cross section is also in tension with the interpretation of the atlas excess as a narrow @xmath10 resonance .
the cms analysis of the semi - leptonic channel @xcite does not observe any excess at @xmath39 tev .
mode a @xmath40 deviation is found at invariant masses around 1.8 tev , though limits are not reported for spin-1 particles .
for a 1.8 tev spin-2 graviton the limits are @xmath41 fb , with 7 fb expected , but these numbers can not be directly compared to the atlas limits @xcite because of the different signal efficiencies . ] the cms search in the fully leptonic channel @xcite gives an upper limit @xmath42 fb for @xmath39 tev .
more generally , a mixture of spin - one @xmath10 , @xmath11 and @xmath12 resonances with nearly the same mass ( they can appear for example in models with a heavy @xmath43 vector boson triplet @xcite ) can not explain the @xmath28 excess .
this can be shown with a simple exercise . since the jet mass cuts in the atlas semi - leptonic searches , @xmath44 @xcite and @xmath45 @xcite , are wide enough to accept @xmath4 and @xmath5 bosons
, one can approximately rewrite the corresponding 95% cl cross section limits in ( ii ) and ( iii ) as @xmath46 by rescaling with the @xmath4 and @xmath5 hadronic branching ratios and ignoring small efficiency differences that may arise from the different @xmath4 and @xmath5 boson masses .
for the fully hadronic final state , a relative acceptance factor @xmath47 for @xmath12 and @xmath11 diboson signals with the @xmath10 selection can be roughly estimated from the overlap of the jet mass distribution for signal samples in ref .
@xcite . the total @xmath28 signal @xmath48 with the @xmath10 selection
can then be estimated as @xmath49 and reaches a maximum @xmath50 fb given the constraints in eqs .
( [ ec : reint ] ) .
besides , the cms search in the @xmath28 channel with a wide mass window @xmath51 gev does not observe such a large excess , as mentioned above .
one can also wonder if a different type of resonance ( _ e.g. _ with spin 2 ) yielding different diboson helicity configurations may have an efficiency in the semi - leptonic channels much smaller than for a vector resonance which mainly produces longitudinal vector bosons therefore turning the non - observation of a large signal compatible with the @xmath28 excess . as it will be shown , this is not the case because the event selection criteria used by the atlas and cms collaborations mainly focus on the kinematics of the reconstructed bosons rather than on its leptonic decay products .
the possibility that the atlas @xmath28 anomaly is due to higgs production incorrectly tagged as @xmath52 is strongly disfavoured by other measurements : * an atlas search for @xmath53 resonances @xcite gives @xmath54 fb for @xmath39 tev , with an expected limit around 45 fb .
* a search for @xmath55 and @xmath0 resonances in the fully hadronic @xmath28 channel by the cms collaboration @xcite yields cross section limits @xmath56 fb and @xmath57 fb , very close to the expected ones . a preliminary @xmath0 resonance search in the @xmath27 final state @xcite
yields a @xmath58 excess at @xmath59 tev , with @xmath60 fb for an expected limit of 20 fb .
but the hypothesis of a @xmath0 resonance behind the latter excess is disfavoured by the former search , which gives @xmath61 fb for @xmath62 tev . a similar analysis by the atlas collaboration @xcite is less sensitive , giving @xmath63 fb , @xmath64 fb , for a resonance mass @xmath65 tev .
since neither @xmath53 nor @xmath66 resonance signals show up in these dedicated searches , it is very unlikely that they could contribute significantly to the @xmath28 excess with a fat dijet selection optimised for @xmath67 production .
additionally , one expects relations between @xmath67 and @xmath66 decay fractions of heavy resonances in definite models @xcite .
all this overwhelming set of related sm - like measurements has motivated the caution by the atlas collaboration regarding this excess , but it has not discouraged early interpretations as new diboson resonances of technicolour models @xcite . while ref .
@xcite only takes into account the limit on the production of @xmath10 resonances from the fully leptonic channel ( the weakest one ) , ref .
@xcite attributes the tension among the searches in different @xmath68 decay channels to statistics .
other @xmath69 interpretations @xcite only focus on the @xmath28 excess overlooking the null results obtained in the other decay modes of the gauge boson pair .
statistical fluctuations aside , experimental data seem to disfavour the possibility that the atlas @xmath28 excess results from a diboson resonance .
we are then led to consider that , it this excess is real , it might be due to something different that looks as a diboson peak due to the kinematical selection applied to reduce sm backgrounds . as we will show in this paper , a requirement on transverse momenta applied in ref .
@xcite shapes certain resonant @xmath70 signals , with @xmath71 an extra particle , making them look like a @xmath67 resonance .
such a requirement is not used by the corresponding analysis of the @xmath28 final state by the cms collaboration @xcite , nor in the atlas analyses of semi - leptonic final states . in section [ sec:2 ]
we explore several final state topologies to find in which cases a diboson peak is kinematically produced without an actual diboson resonance . in section [ sec:3 ]
we show in two benchmark examples how a possible signal would look like with the atlas and cms fat dijet selections , as well as in the atlas analyses of semi - leptonic final states .
we present our conclusions in section [ sec:4 ] . in an appendix
we discuss to what extent the event selection efficiencies in the semi - leptonic channels depend on the different diboson helicities .
in this section we explore different topologies in which a diboson pair plus an extra particle @xmath71 are produced , focusing for definiteness on @xmath10 production .
( results are the same for @xmath12 and @xmath11 , obviously . ) in principle , the @xmath71 particle could either be invisible ( thus a potential dark matter candidate ) , or a new relatively light scalar with dominant hadronic decay , or even just a sm gauge or higgs boson .
we work at the partonic level , calculating matrix elements for processes with new generic scalars , fermions or vector bosons , integrating over phase space and parton distribution functions , and examining how the @xmath67 invariant mass distribution is shaped by the following cuts applied in the atlas dijet analysis @xcite : 1 .
boson pseudo - rapidity @xmath72 and rapidity difference @xmath73 , where the indices 1 and 2 denote the two bosons .
at least one of the two bosons must have transverse momentum @xmath74 greater than 360 gev .
transverse momentum asymmetry @xmath75 . together with the rapidity difference cut
, this requirement selects approximately back - to - back bosons even if they are not produced from the decay of an @xmath76-channel resonance . in this section
we do not impose the requirement on missing energy @xmath77 gev of the analysis in ref .
@xcite to remain as general as possible , since in principle @xmath71 could be invisible or decay hadronically . at any rate , the application of such a cut does not significantly modify the obtained distributions .
matrix elements are evaluated using helas @xcite including the decay of the @xmath4 and @xmath5 bosons .
phase space integration is done by implementing these processes into the generator protos @xcite .
we give our results for specific choices of the new particles , _
e.g. _ assuming that @xmath71 is a neutral scalar .
but our results are more general , as they are mainly based on the kinematics of cascade decays , and we have explicitly checked this fact by using `` flat '' matrix elements , constant except for the resonant propagators , which make no assumption on the spin or charge of the new particles .
results are also independent of the incoming partons . for definiteness
we assume @xmath78 , @xmath79 initial states for the production of @xmath80 , @xmath81 , respectively .
however , the presence of the extra particle @xmath71 opens the possibility of gluon - initiated processes with larger partonic luminosities .
it is clear from the beginning that non - resonant diboson production , with or without an extra particle , can not give a peak at an invariant mass as high as 2 tev merely with the application of the kinematical cuts in ( 13 ) .
however , it is interesting to consider this academic case to investigate how an unaccounted sm contribution could be affected by this event selection .
we consider @xmath82 , @xmath83 ( with @xmath71 a neutral scalar ) plus the charge conjugate processes , mediated by a @xmath84-channel heavy quark @xmath85 , as shown in figure [ fig : topvv ] .
the normalised @xmath10 invariant mass distributions before and after cuts are presented in figure [ fig : distvv ] ( top ) . for @xmath86
we take a scalar mass @xmath87 gev , but the results are rather independent of this value . the cuts
reduce the cross section by a factor of 5 for @xmath10 and 11 for @xmath86 and in both cases they maintain the shape of the distribution , with a shift towards larger invariant masses and a long tail .
a similar analysis can be done for @xmath88 and the atlas @xmath89 selection criteria . in this case
the efficiency variations are rather small , as it is expected from the distributions in figure [ fig : ang ] .
the largest difference is found for @xmath90 with an efficiency a factor of @xmath91 larger than for the @xmath92 helicities .
the same conclusions apply for @xmath10 resonances in the @xmath89 channel .
g. aad _ et al . _ [ atlas collaboration ] , arxiv:1506.00962 [ hep - ex ] .
g. aad _ et al . _
[ atlas collaboration ] , eur .
j. c * 75 * ( 2015 ) 5 , 209 [ arxiv:1503.04677 [ hep - ex ] ] .
g. aad _ et al . _ [ atlas collaboration ] , eur .
j. c * 75 * ( 2015 ) 2 , 69 [ arxiv:1409.6190 [ hep - ex ] ] . g. aad _ et al . _
[ atlas collaboration ] , phys .
b * 737 * ( 2014 ) 223 [ arxiv:1406.4456 [ hep - ex ] ] . v. khachatryan _ et al . _
[ cms collaboration ] , jhep * 1408 * ( 2014 ) 173 [ arxiv:1405.1994 [ hep - ex ] ] .
v. khachatryan _ et al . _
[ cms collaboration ] , jhep * 1408 * ( 2014 ) 174 [ arxiv:1405.3447 [ hep - ex ] ] .
v. khachatryan _ et al . _
[ cms collaboration ] , phys .
b * 740 * ( 2015 ) 83 [ arxiv:1407.3476 [ hep - ex ] ] .
j. de blas , j. m. lizana and m. perez - victoria , jhep * 1301 * ( 2013 ) 166 [ arxiv:1211.2229 [ hep - ph ] ] .
g. aad _ et al .
_ [ atlas collaboration ] , arxiv:1506.00285 [ hep - ex ] .
v. khachatryan _ et al . _
[ cms collaboration ] , arxiv:1506.01443 [ hep - ex ] .
cms collaboration , report cms - pas - exo-14 - 010 .
g. aad _ et al . _ [ atlas collaboration ] , eur .
j. c * 75 * ( 2015 ) 6 , 263 [ arxiv:1503.08089 [ hep - ex ] ] .
j. hisano , n. nagata and y. omura , arxiv:1506.03931 [ hep - ph ] .
h. s. fukano , m. kurachi , s. matsuzaki , k. terashi and k. yamawaki , arxiv:1506.03751 [ hep - ph ] .
d. b. franzosi , m. t. frandsen and f. sannino , arxiv:1506.04392 [ hep - ph ] . k. cheung , w. y. keung , p. y. tseng and t. c. yuan , arxiv:1506.06064 [ hep - ph ] .
h. murayama , i. watanabe and k. hagiwara , report kek-91 - 11 . j. a. aguilar - saavedra .
protos , a program for top simulations .
j. m. butterworth , a. r. davison , m. rubin and g. p. salam , phys .
* 100 * ( 2008 ) 242001 [ arxiv:0802.2470 [ hep - ph ] ] .
s. d. ellis , c. k. vermilion and j. r. walsh , phys .
d * 80 * ( 2009 ) 051501 [ arxiv:0903.5081 [ hep - ph ] ] .
v. khachatryan _ et al .
_ [ cms collaboration ] , eur .
j. c * 74 * ( 2014 ) 11 , 3149 [ arxiv:1407.3683 [ hep - ex ] ] .
f. f. deppisch , t. e. gonzalo , s. patra , n. sahu and u. sarkar , phys .
d * 90 * ( 2014 ) 5 , 053014 [ arxiv:1407.5384 [ hep - ph ] ] .
m. heikinheimo , m. raidal and c. spethmann , eur .
j. c * 74 * ( 2014 ) 10 , 3107 [ arxiv:1407.6908 [ hep - ph ] ] .
j. a. aguilar - saavedra and f. r. joaquim , phys . rev .
d * 90 * ( 2014 ) 11 , 115010 [ arxiv:1408.2456 [ hep - ph ] ] .
j. a. aguilar - saavedra and j. bernabu , arxiv:1508.04592 [ hep - ph ]
. g. l. kane , g. a. ladinsky and c. p. yuan , phys .
d * 45 * ( 1992 ) 124 . | the atlas excess in fat jet pair production is kinematically compatible with the decay of a heavy resonance into two gauge bosons plus an extra particle .
this hypothesis would explain the absence of such a large excess in the analogous cms analysis of fat dijet final states , as well as the negative results of diboson resonance searches in the semi - leptonic decay modes .
if the extra particle is the higgs boson , this hypothesis might also explain statistical fluctuations aside why the cms search for @xmath0 resonances in the semi - leptonic channel finds some excess while in the fully hadronic one it does not have a significant deviation . * triboson interpretations of the atlas + diboson excess * j. a. aguilar
saavedra + departamento de fsica terica y del cosmos , universidad de granada , 18071 granada , spain |
consider solutions @xmath11 to the defocusing nonlinear wave equation @xmath12 where @xmath13 is a parameter . from standard energy methods
( see e.g. @xcite ) , relying in particular on the conserved energy @xmath14 and on the sobolev embedding @xmath15 , we know that given any initial data @xmath16 , @xmath17 , there exists a unique global energy class solution @xmath18 to with initial data @xmath19 , @xmath20 .
one has a similar theory for data that is only locally of finite energy , thanks to finite speed of propagation . in this paper
we investigate the asymptotic behaviour of this solution @xmath0 in the high exponent limit @xmath4 , while keeping the initial data fixed .
to avoid technicalities , let us suppose that @xmath5 are smooth and compactly supported , and that @xmath21 for all @xmath22 .
formally , we expect @xmath23 to converge in some sense to some solution @xmath24 of the `` infinitely nonlinear defocusing wave equation '' @xmath25 with initial data @xmath19 , @xmath20
. of course , does not make rigorous sense .
but , motivated by analogy with infinite barrier potentials to the linear wave equation @xmath26}(x ) \phi$ ] with initial data smooth and supported on @xmath10 $ ] , a simple compactness argument ( or explicit computation ) shows that @xmath8 converges ( in , say , the uniform topology ) to the solution to the free wave equation @xmath27 on @xmath28 $ ] with the reflective ( dirichlet ) boundary conditions @xmath29 .
] , one might wish to interpret the infinite nonlinearity @xmath30 as some sort of `` barrier nonlinearity '' which is constraining @xmath8 to have magnitude at most @xmath31 , but otherwise has no effect .
intuitively , we thus expect the limiting wave @xmath9 to evolve like the linear wave equation until it reaches the threshold @xmath32 or @xmath33 , at which point it should `` reflect '' off the nonlinear barrier in does not directly make sense ) , and so energy should be reflected rather than absorbed by the barrier . ] . the purpose of this paper is to make the above intuition rigorous , and to give a precise interpretation for the equation .
to get some further intuition as to this reflection phenomenon , let us first study ( non - rigorously ) the simpler ode problem , in which we look at solutions @xmath34 to the ode @xmath35 with fixed initial data @xmath36 with @xmath37 , and with @xmath4 . from the conserved energy @xmath38 ( and recalling that @xmath39 ) we quickly obtain the uniform bounds @xmath40 and all times @xmath41 , where the implied constants in the @xmath42 notation depend on @xmath5 .
thus we already see a barrier effect preventing @xmath8 from going too far outside of the interval @xmath10 $ ] .
to investigate what happens near a time @xmath43 in which @xmath44 is close to ( say ) @xmath45 , let us make the ansatz @xmath46 observe from that @xmath47 is positive for @xmath48 and some constant @xmath49 depending only on @xmath50 .
write @xmath51 .
some brief computation then shows that @xmath52 solves the equation @xmath53 for all @xmath54 $ ] ; also , by we obtain an upper bound @xmath55 ( but no comparable lower bound ) , as well as the lipschitz bound @xmath56 . in the asymptotic limit @xmath4 , we thus expect the rescaled solution @xmath57 to converge to a solution @xmath58 of the ode @xmath59 it turns out that this ode can be solved explicitly in phase space . ] , and it is easy to verify that the general solution is @xmath60 for any @xmath61 and @xmath62 .
these solutions asymptotically approach @xmath63 as @xmath64 .
thus we see that if @xmath52 is large and negative but with positive velocity , then the solution to this ode will be approximately linear until @xmath52 approaches the origin , where it will dwell for a bounded amount of time before reflecting back into the negative axis with the opposite velocity to its initial velocity . undoing the rescaling , we thus expect the limit @xmath24 of the original ode solutions @xmath23 to also behave linearly until reaching @xmath65 or @xmath66 , at which point they should reflect with equal and opposite velocity , so that @xmath9 will eventually be a sawtooth function with range @xmath10 $ ] ( except of course in the degenerate case @xmath67 , @xmath68 , in which case @xmath9 should be constant ) . because the ode can be solved more or less explicitly using the conserved hamiltonian , it is not difficult to formalise these heuristics rigorously ; we leave this as an exercise to the interested reader .
note that the above analysis also suggests a more precise asymptotic for how reflections of @xmath23 should behave for large @xmath69 , namely ( assuming @xmath70 for simplicity ) @xmath71 or ( after taylor expansion ) @xmath72 where @xmath73 measures the speed of the reflection , and @xmath43 the time at which reflection occurs , and we are deliberately being vague as to what the symbol @xmath74 means . adapting the above ode analysis to the pde setting , we can now study the reflection behaviour of @xmath23 near the nonlinear barrier @xmath75 at some point @xmath76 in spacetime by introducing the ansatz @xmath77 where @xmath52 can be computed to solve the equation @xmath78 in the region where @xmath8 is near @xmath31 ( and is in particular non - negative ) . in the limit @xmath4
, this formally converges to _
liouville s equation _
remarkably , this nonlinear wave equation can also be solved explicitly@xcite , with explicit solution @xmath80 for arbitrary smooth functions @xmath81 for which the right - hand side is well - defined and @xmath82 . ] .
somewhat less `` magically '' , one can approach the explicit solvability of this equation by introducing the null coordinates @xmath83 and their associated derivatives @xmath84 and rewriting as @xmath85 and then noting the _ pointwise _ conservation laws @xmath86 which can ultimately ( with a certain amount of algebraic computation ) be used to arrive at the solution ; see @xcite for details . using this explicit solution , one can eventually be led to the ( heuristic ) conclusion that the reflection profile @xmath87 should resemble a lorentz - transformed version of , i.e. @xmath88/\sqrt{1-v^2})}\ ] ] for some @xmath89 , @xmath62 , and @xmath90 .
thus we expect @xmath8 to reflect along spacelike curves such as @xmath91 in order to stay confined to the interval @xmath10 $ ] .
we now state the main result of our paper , which aims to make the above intuition precise .
[ main - thm ] let @xmath92 be functions obeying the following properties : * ( regularity ) @xmath5 are smooth . *
( strict barrier condition ) for all @xmath93 , @xmath21 . *
( non - degeneracy ) the sets @xmath94 and @xmath95 have only finitely many connected components in any compact interval are real analytic , since the zeroes of non - trivial real analytic functions can not accumulate . ] . for each @xmath96 ,
let @xmath97 be the unique global solution to with initial data @xmath98 , @xmath99 . then
, as @xmath4 , @xmath23 converges uniformly on compact subsets of @xmath100 to the unique function @xmath101 that obeys the following properties : * ( regularity , i ) @xmath8 is locally lipschitz continuous ( and in particular is differentiable almost everywhere , by radamacher s theorem ) . * ( regularity , ii ) for each @xmath102 and @xmath103 , the functions @xmath104 and @xmath105 are piecewise smooth ( with finitely many pieces on each compact interval ) . *
( initial data ) on a neighbourhood of the initial surface @xmath106 , @xmath8 agrees with the linear solution @xmath107 the free wave equation with initial data @xmath5 . * ( barrier condition ) @xmath108 for all @xmath109 . *
( defect measure ) we have @xmath110 in the sense of distributions , where @xmath111 are locally finite non - negative measures supported on the sets @xmath112 , @xmath113 respectively .
* ( null energy reflection ) for almost every @xmath114 , we have explicitly from in terms of the initial data as @xmath115 and @xmath116 . ]
@xmath117 is almost everywhere equal to a function of @xmath118 only , and similarly for @xmath119 .
the existence and uniqueness of @xmath8 obeying the above properties is not obvious , but is part of the theorem .
the conditions ( i)-(vi ) are thus the rigorous substitute for the non - rigorous equation ; they superficially resemble a `` viscosity solution '' or `` kinetic formulation '' of ( see e.g. @xcite ) , and it would be interesting to see if there is any rigorous connection here to the kinetic theory of conservation laws . the hypotheses ( a ) , ( b ) , ( c ) on the initial data @xmath5 are somewhat stronger than what is likely to be needed for the theorem to hold ; in particular , one should be able to relax the strict barrier condition ( b ) to @xmath120 , and also omit the non - degeneracy condition ( c ) , although the conclusions ( ii ) , ( iii ) the limit @xmath8 would have to be modified in this case ; one also expects to be able to relax the smoothness assumption ( a ) , perhaps all the way to the energy class or possibly even the bounded variation class .
we will not pursue these matters here . to illustrate the reflection in action ,
let us restrict attention to the triangular region @xmath121 and consider the initial data @xmath5 associated to the linear solution @xmath122 where @xmath123 is a small constant ( e.g. @xmath124 is safe ) .
observe that @xmath125 lies between @xmath126 and @xmath31 for most of @xmath127 , but exceeds @xmath31 in the disk @xmath128 .
thus we expect @xmath8 to follow @xmath125 until it encounters this disk , at which point it should reflect .
the actual solution @xmath8 can be described using figure [ fig1 ] . in region
i , @xmath8 is equal to the linear solution @xmath125 . but
@xmath125 exceeds @xmath31 once one passes the circular arc joining @xmath129 and @xmath130 , and so a reflection must occur in region ii ; indeed , one has @xmath131 in this region .
once the solution passes @xmath129 and @xmath130 , though , it turns out that the downward velocity of the reflected wave is now sufficient to drag @xmath8 off of the singular set ( which , in this case , are the left and right arcs of the circle @xmath132 ) are not used as a reflective set for @xmath8 . ]
@xmath133 ( which , in this example , is the circular arc connecting @xmath129 and @xmath130 ) . indeed , in region iii
, we have @xmath134 ( note that this is the unique solution to the free wave equation that matches up with @xmath8 on regions i , ii ) and similarly on region iv we have @xmath135 finally , on region v we have another solution to the free wave equation , which is now on a downward trajectory away from @xmath65 : @xmath136 if one were to continue the evolution of @xmath8 forward in time beyond @xmath127 ( extending the initial data @xmath5 suitably ) , the solution would eventually hit the @xmath66 barrier and reflect again , picking up further singularities propagating in null directions similar to those pictured here .
thus , while the solution remains piecewise smooth for all time , we expect the number of singularities to increase as time progresses , due to the increasing number of reflections taking place .
it is a routine matter to verify that the solution presented here verifies the properties ( i)-(vi ) on @xmath127 ( if @xmath137 is sufficiently small ) , and so is necessarily the limiting solution @xmath8 , thanks to theorem [ main - thm ] ( and the uniqueness theory in section [ unique - sec ] below ) . we omit the details .
the circular arc between @xmath129 and @xmath130 supports a component of the defect measure @xmath138 , which can be computed explicitly from the above formulae .
the defect measure can also be computed by integrating and observing that @xmath139 whenever @xmath140 , @xmath141 , @xmath142 , @xmath143 are the corners of a small parallelogram intersecting this arc . sending @xmath144 ( say ) , we observe that the left - hand side is asymptotic to @xmath145 . since @xmath146 reflects in sign across the arc , we can simplify this as @xmath147 .
this allows us to describe @xmath138 explicitly in terms of the conserved quantity @xmath148 and the slope of the arc ; we omit the details .
the above example shows that the barrier set @xmath149 has some overlap initially with the set @xmath150 , but the situation becomes more complicated after multiple `` reflections '' off of the two barriers @xmath65 and @xmath66 , and the author does not know of a clean way to describe this set for large times @xmath41 , although as the above example suggests , these sets should be computable for any given choice of @xmath41 and any given initial data .
we shall shortly discuss the proof of theorem [ main - thm ] , but let us first pause to discuss two techniques that initially look promising for solving this problem , but end up being problematic for a number of reasons .
each of the nonlinear wave equations enjoy a conserved stress - energy tensor @xmath151 , and it is tempting to try to show that this stress - energy tensor converges to a limit @xmath152 .
however , the author found it difficult to relate this limit tensor to the limit solution @xmath9 .
the key technical difficulty was that while it was not difficult to ensure that derivatives @xmath153 of @xmath23 converged in a weak sense to the derivatives @xmath154 of a limit @xmath9 , this did not imply that the _ magnitudes _ @xmath155 , @xmath156 of the derivatives converged ( weakly ) to the expected limit of @xmath157 , @xmath158 , due to the possibility of increasing oscillations in the sign of @xmath159 or @xmath160 in the limit @xmath4 which could cause some loss of mass in the limit . because of this , much of the argument is instead focused on controlling this oscillation , and the stress - energy tensor conservation appears to be of limited use for such an objective .
instead , the argument relies much more heavily on pointwise conservation ( or almost - conservation ) laws such as , and on the method of characteristics .
another possible approach would be to try to construct an approximate solution ( or parametrix ) to @xmath23 , along the lines of , and show that @xmath23 is close enough to the approximate solution that the convergence can be read off directly ( much as it can be from ) .
while it does seem possible to construct the approximate solution more or less explicitly , the author was unable to find a sufficiently strong stability theory to then close the argument by comparing the exact solution to the approximate solution .
the difficulty is that the standard stability theory for ( e.g. by applying energy estimates to the difference equation ) exhibits losses which grow exponentially in time with rate proportional to @xmath69 , thus requiring the accuracy of the approximate solution to be exponentially small in @xmath69 before there is hope of connecting the approximate solution to the exact one . because of this , the proof below avoids all use of perturbation theory . ] , and instead estimates the nonlinear solutions @xmath23 directly .
it may however be of interest to develop a stability theory for which is more uniform in @xmath69 ( perhaps using bounded variation type norms rather than energy space norms ? ) .
one starting point may be the perturbation theory for , explored recently in @xcite .
our arguments are instead based on a compactness method .
it is not difficult to use energy conservation to demonstrate equicontinuity and uniform boundedness in the @xmath23 , so we know ( from the arzel - ascoli theorem ) that the @xmath23 have at least one limit point .
it thus suffices to show that all such limit points obey the properties ( i)-(vi ) , and that the properties ( i)-(vi ) uniquely determine @xmath8 .
the uniqueness is established in section [ unique - sec ] , and is based on many applications of the method of characteristics . to establish
that all limit points obey ( i)-(vi ) , we first establish in section [ apriori ] a number of _ a priori _ estimates on the solutions @xmath23 , in particular obtaining some crucial boundedness and oscillation control on @xmath8 and its first derivatives , uniformly in @xmath69 . in section [ compact - sec ]
we then take limits along some subsequence of @xmath69 going to infinity to recover the desired properties ( i)-(vi ) .
it seems of interest to obtain more robust methods for proving results for infinite nonlinear barriers ; the arguments here rely heavily on the method of characteristics and so do not seem to easily extend to , say , the @xmath4 limit of the one - dimensional nonlinear schrdinger equation @xmath161 , or to higher - dimensional non - linear wave equations . in higher dimensions
there is also a serious additional problem , namely that the nonlinearity becomes energy - critical in the limit @xmath4 in two dimensions , and ( even worse ) becomes energy - supercritical for large @xmath69 in three and higher dimensions .
however , while global existence for defocusing supercritical non - linear wave equations from large data is a notoriously difficult open problem , there is the remote possibility that the asymptotic case @xmath4 is actually better behaved than that of a fixed @xmath69 . at the very least
, one should be able to _ conjecture _ what the correct limit of the solution should be .
we are indebted to tristan roy for posing this question , and rowan killip for many useful discussions , ut v. le for pointing out a misprint , to patrick dorey to pointing out the link to liouville s equation , and to the anonymous referee for many useful comments and corrections .
the author is supported by a grant from the macarthur foundation , by nsf grant dms-0649473 , and by the nsf waterman award .
we use the asymptotic notation @xmath162 to denote the bound @xmath163 for some constant @xmath130 depending on fixed quantities ( e.g. the initial data ) ; note that @xmath164 may be negative , so that @xmath162 only provides an upper bound .
we also use @xmath165 to denote any quantity bounded in magnitude by @xmath166 ( thus we have both an upper and a lower bound in this case ) , and @xmath167 for @xmath168 .
if the constant @xmath130 needs to depend on additional parameters , we will denote this by subscripts , e.g. @xmath169 or @xmath170 .
it is convenient to use both cartesian coordinates @xmath114 and null coordinates @xmath171 to parameterise spacetime . to reduce confusion we shall use angled brackets to denote the latter , thus @xmath172 and @xmath173 thus for instance we might write @xmath174 for @xmath175
. we will frequently rely on three reflection symmetries of to normalise various signs : the _ time reversal symmetry _ @xmath176
( which also swaps @xmath118 with @xmath177 ) , the _ space reflection symmetry _ @xmath178 ( which also swaps @xmath118 with @xmath179 ) , and the _ sign reversal symmetry _
@xmath180 we will frequently be dealing with ( closed ) _ diamonds _ in spacetime , which we define to be regions of the form @xmath181 for some @xmath182 and @xmath183 .
one can of course define open diamonds similarly .
we will also be dealing with
_ triangles _
@xmath184 which are the upper half of diamonds .
in this section we show that there is at most one function @xmath185 obeying the properties ( i)-(vi ) listed in theorem [ main - thm ] .
it suffices to prove uniqueness on a diamond region @xmath186 for any fixed @xmath187 ; by the time reversal symmetry it in fact suffices to prove uniqueness on a triangular region @xmath188 suppose for contradiction that uniqueness failed on @xmath127 , then there would exist two functions @xmath189 obeying the properties ( i)-(vi ) in theorem [ main - thm ] which did not agree identically on @xmath127 . by property ( iii ) ,
@xmath8 and @xmath190 already agree on some neighbourhood of the time axis .
since @xmath191 are continuous by ( i ) , and @xmath127 is compact , we may therefore find some @xmath192 with @xmath193 such that @xmath194 for all @xmath195 with @xmath196 , but such that @xmath8 and @xmath190 do not agree in any neighbourhood of @xmath76 in @xmath127 .
we will show that @xmath8 and @xmath190 must in fact agree in some neighbourhood of @xmath197 , achieving the desired contradiction .
first suppose that @xmath198 , then of course @xmath199 .
as @xmath191 are both continuous , we thus have @xmath191 bounded away from @xmath126 and @xmath45 on a neighbourhood of @xmath76 in @xmath127 . by ( v ) , @xmath191 both solve the free wave equation in the sense of distributions on this region , and since they agree below @xmath76 , they must therefore agree on a neighbourhood of @xmath76 by uniqueness of the free wave equation , obtaining the desired contradiction .
thus we may assume that @xmath200 has magnitude @xmath31 ; by the sign reversal symmetry we may take @xmath201 by continuity , we thus see that @xmath8 , @xmath190 is positive in a neighbourhood of @xmath76 ; from ( v ) we conclude that @xmath202 is a non - negative measure in this neighbourhood . integrating this
, we conclude that @xmath203 whenever @xmath204 and @xmath205 lie sufficiently close to @xmath76 ; this implies in particular that @xmath206 is non - increasing in @xmath179 ( and @xmath146 non - increasing in @xmath118 ) in this region whenever the derivatives are defined .
similarly for @xmath190 .
write @xmath207 .
we already know that @xmath208 when @xmath171 is sufficiently close to @xmath209 and @xmath210 .
we now make extend this equivalence to the right of @xmath211 : [ phiright ] @xmath8 and @xmath190 agree for all @xmath212 sufficiently close to @xmath213 in the region @xmath214 , @xmath215 .
we may of course assume @xmath216 otherwise this extension is vacuous .
suppose first that @xmath217 for a sequence of positive @xmath218 approaching zero .
since @xmath219 is piecewise smooth by ( ii ) , we see from the mean value theorem that @xmath220 for a sequence of positive @xmath221 approaching zero ; by ( vi ) the same is true for @xmath125 , which by the non - degeneracy assumption ( c ) implies that @xmath222 must in fact vanish on some left - neighbourhood of @xmath223 , which by ( vi ) implies that @xmath224 and @xmath225 vanish almost everywhere whenever @xmath179 is less than @xmath223 and sufficiently close to @xmath223 .
since @xmath191 already agree for @xmath210 , and are lipschitz continuous , an integration in the @xmath179 direction then gives the lemma .
now consider the opposite case , where @xmath226 for all sufficiently small positive @xmath218 .
as @xmath227 is piecewise smooth , and its stationary points have finitely many connected components in @xmath127 by ( vi ) , ( c ) , we conclude that @xmath228 for all @xmath229 for some sufficiently small @xmath230 . by continuity
, we can find a small @xmath231 such that @xmath232 for all @xmath233 . since @xmath234 is piecewise smooth and the stationary points have finitely many connected components in @xmath127 , we see ( after shrinking @xmath235 if necessary ) that @xmath236 is either always positive , always negative , or always zero for @xmath233 .
if @xmath236 is always zero , then by ( vi ) we see that @xmath237 almost everywhere for @xmath118 sufficiently close to and larger than @xmath238 , and any @xmath179 , and the lemma then follows from the fundamental theorem of calculus .
if @xmath236 is always negative , then as @xmath206 is non - increasing in @xmath179 , we see that @xmath239 is negative for almost every @xmath233 and @xmath240 ; combining this with ( vi ) we conclude that @xmath241 almost everywhere in this region , and similarly for @xmath190 .
in particular , @xmath206 and @xmath242 agree in this region , and the lemma again follows from the fundamental theorem of calculus . finally , we handle the most difficult will reflect itself on some spacelike curve containing @xmath76 , which is essentially the only interesting nonlinear phenomenon that @xmath8 can exhibit . ]
case , when @xmath236 is always positive .
let @xmath243 denote the unique lipschitz - continuous solution to the free wave equation on the parallelogram @xmath244 that agrees with @xmath8 ( and hence @xmath190 ) on the lower two edges of this parallelogram ( i.e. when @xmath245 or @xmath246 ) .
we will show that @xmath8 and @xmath190 agree on @xmath247 . call a point @xmath248
good _ if we have @xmath249 for all @xmath250 with @xmath251 ; the set of good points is then a closed subset of @xmath247 .
we first observe that if @xmath171 is good , then @xmath252 . indeed , since @xmath146 is positive on the lower left edge of @xmath247 , and @xmath206 is positive on the lower right edge , we see that @xmath253 for all @xmath250 with @xmath251 , and @xmath254 .
also , from ( v ) , @xmath8 and @xmath190 solve the free wave equation in the neighbourhood of any region near @xmath255 where they are strictly less than @xmath31 .
a continuity argument based on the uniqueness of the free wave equation then shows that @xmath256 for all @xmath236 , and the claim follows .
now suppose that @xmath140 is not good , thus @xmath257 .
excluding a set of measure zero , we may assume that @xmath258 and that @xmath259 exists and is equal to @xmath260 , which is non - zero .
we claim that @xmath261 can not be positive . for
if it is , then we can find @xmath262 less than @xmath179 and arbitrarily close to @xmath179 such that @xmath263 ( see figure [ fig2 ] . )
applying , we conclude that @xmath264 for all @xmath265 . in particular , @xmath266 must be bounded away from @xmath31 .
also , as @xmath243 is continuous , we may also assume that @xmath267 in particular , @xmath8 and @xmath243 disagree at @xmath268 .
this implies that @xmath269 for at least one @xmath270 , @xmath271 , since otherwise by ( v ) @xmath8 would solve the free wave equation in this region and thus be necessarily equal to @xmath243 by uniqueness of that equation . among all such @xmath272 ,
we can ( by continuity and compactness ) pick a pair that maximises @xmath273 .
since @xmath266 must be bounded away from @xmath31 , we have @xmath274 . from ( v ) , @xmath8 solves the linear wave equation in the parallelogram @xmath275 , which implies that @xmath276 but by the fundamental theorem of calculus , @xmath277 , and @xmath269 , hence @xmath278 , contradicting ( iv ) .
hence @xmath279 can not be positive , and hence by hypothesis must be equal to @xmath280 .
the same considerations apply to @xmath190 , and so @xmath281 at almost every point in @xmath247 that is not good . since @xmath282 at good points in @xmath247 , and @xmath281 on all other points of @xmath247
, we obtain the lemma from the fundamental theorem of calculus . using space reflection symmetry
, we can reflect the previous lemma and conclude @xmath8 and @xmath190 agree for all @xmath212 sufficiently close to @xmath213 in the region @xmath283 , @xmath284 . by the previous lemmas , we know that there exist @xmath285 such that @xmath286 for all @xmath287 and @xmath288 for all @xmath287 .
to finish the uniqueness claim , we need to show that @xmath191 also agree in the future parallelogram @xmath289 . as before , by shrinking @xmath235 if necessary , we know that @xmath290 is either always positive , always zero , or always negative for @xmath233 .
the former option is not possible from the barrier condition ( iv ) since @xmath291 , so @xmath290 is always non - positive . using the monotonicity of @xmath206 in @xmath179 and ( vi )
we thus conclude that @xmath292 for almost every @xmath293 , and similarly for @xmath190 , so that @xmath206 and @xmath242 agree almost everywhere on @xmath294 ; similarly @xmath146 and @xmath295 agree .
the claim now follows from the fundamental theorem of calculus , and the uniqueness claim is complete .
one could convert the above uniqueness results , with additional effort , into an existence result , but existence of a solution to ( i)-(vi ) will be automatic for us from the arzel - ascoli theorem , as we will show that any uniformly convergent sequence of @xmath23 will converge to a solution to ( i)-(vi ) .
the next step in the proof of theorem [ main - thm ] is to establish various _ a priori _ estimates on the solution to on the diamond @xmath296 for some large parameter @xmath297 .
accordingly , let us fix @xmath5 obeying the hypotheses ( a),(b),(c ) of theorem [ main - thm ] , and let @xmath298 ; we allow all implied constants to depend on the initial data @xmath5 and @xmath297 , but will carefully track the dependence of constants on @xmath69 .
we will assume that @xmath69 is sufficiently large depending on the initial data and on @xmath297 ; in particular , we may take @xmath299 ( say ) .
standard energy methods ( see e.g. @xcite ) then show that @xmath8 exists globally and is @xmath300 in @xmath301 .
this is sufficient to justify all the formal computations below .
we begin with a preliminary ( and rather crude ) hlder continuity estimate , which we need to establish some spatial separation ( uniformly in @xmath69 ) between the region where @xmath8 approaches @xmath45 , and the region where @xmath8 approaches @xmath126 .
[ hold ] for any @xmath302 we have @xmath303 we use the monotonicity of local energy @xmath304 the standard energy flux identity shows that @xmath305 for all @xmath41 . from the hypotheses ( a ) ,
( b ) we have @xmath306 ( note in particular the uniformity in @xmath69 ) and thus by energy monotonicity @xmath307 for all @xmath41 . by cauchy - schwarz , we see in particular that we have some hlder continuity in space , or more precisely that @xmath308 .
we can also get some hlder continuity in time by a variety of methods .
for instance , from cauchy - schwarz again we have @xmath309 for any @xmath310 and any @xmath22 ; integrating this in @xmath22 on some interval @xmath311 $ ] and using we obtain @xmath312 when @xmath313 . on the other hand , from we have @xmath314 and thus @xmath315 if we optimise @xmath316 , we obtain @xmath317 combining this with we obtain the claim ( possibly after replacing @xmath297 with a slightly larger quantity in the above argument ) . next , we express the equation in terms of the null derivatives as @xmath318 we can use this to give some important pointwise bounds on @xmath8 and its derivatives . for any time @xmath319 , let @xmath320 be the best constant such that @xmath321 and @xmath322 and @xmath323 for all @xmath22 with @xmath324 ( compare with ) .
thus for instance @xmath325 . we now show that the component of @xmath320 , at least , is stable on short time intervals : [ pond ] there exists a time increment @xmath326 ( depending only on the initial data and @xmath297 ) such that @xmath327 for any @xmath319 and @xmath328 with @xmath329 , and either @xmath330 or @xmath331 .
we shall use the method of characteristics .
we take @xmath326 to be a small quantity depending on the initial data to be chosen later .
fix @xmath43 and write @xmath332 .
let @xmath333 be a small quantity depending on @xmath334 and the initial data to be chosen later , and then let @xmath335 be a large quantity depending on @xmath336 to be chosen later . by time reversal symmetry we may take @xmath337 ; by sign reversal symmetry it suffices to establish the upper bound @xmath338 assume this bound fails , then ( by continuity and compact support in space ) there exists @xmath339 and @xmath340 such that @xmath341 and @xmath342 for all @xmath343 and @xmath93 . from , we have @xmath344 if @xmath335 is large enough . from lemma [ hold ]
we see ( if @xmath345 is small enough ) that @xmath346 is positive in the triangular region @xmath347 ( which is contained in @xmath348 ) . from
we conclude that @xmath206 is decreasing in the @xmath179 direction in @xmath127 , and thus has an upper bound @xmath349 on this region thanks to .
similarly we have @xmath350 on @xmath127 . applying the fundamental theorem of calculus and we conclude that @xmath351 for all @xmath195 , which when compared with shows ( if @xmath352 is small enough and @xmath335 is large enough ) that @xmath353 where @xmath354 .
now we consider the diamond region @xmath355 since @xmath356 , this diamond is contained in the triangle @xmath127 ( indeed , it is nestled in the upper tip of that triangle ) . as before , we have the upper bounds @xmath357 on this diamond . from this , , and the fundamental theorem of calculus
, we have @xmath358 on this diamond ( if @xmath352 is small enough ) .
applying we conclude that @xmath359 for some absolute constant @xmath49 . integrating this on the diamond
we conclude that @xmath360 but from , , the left - hand side is bounded below by @xmath361 . we conclude that @xmath362 but from the definition of @xmath363 , we obtain a contradiction if @xmath335 is large enough depending on @xmath352 , and the claim follows .
now we establish a similar stability for the component of @xmath320 .
[ ponder ] there exists a time increment @xmath326 ( depending only on the initial data and @xmath297 ) such that @xmath364 and @xmath365 for any @xmath319 and @xmath328 with @xmath329 , and either @xmath330 or @xmath331 .
this will be a more advanced application of the method of characteristics .
we again let @xmath326 be a sufficiently small quantity ( depending on the initial data ) to be chosen later .
fix @xmath43 and write @xmath366 , and let @xmath367 be a large quantity depending on @xmath334 and the initial data to be chosen later .
it will suffice to show that @xmath368 whenever @xmath329 and @xmath340 .
suppose for contradiction that this claim failed .
as before ( using the symmetries , ) we may assume that @xmath369 and @xmath370 are such that @xmath371 for all @xmath372 and @xmath22 with @xmath373 .
we first dispose of an easy case when @xmath374 is small , say @xmath375 . then by lemma [ hold ] we conclude ( if @xmath345 is small enough ) that @xmath376 on the triangular region @xmath377 , and the claim then easily follows from , the fundamental theorem of calculus , and .
thus we may assume @xmath378 ; replacing @xmath8 with @xmath379 if necessary we may assume @xmath380 . by lemma
[ hold ] , we see that @xmath8 is positive whenever @xmath381 ( say ) , if @xmath345 is small enough
. it will be convenient to make the change of variables @xmath382 then from , @xmath52 solves the equation @xmath383 on the region @xmath384 . from , we have @xmath385 and @xmath386 for @xmath387 and @xmath388 .
meanwhile , while from lemma [ pond ] ( and shrinking @xmath345 as necessary ) we have the upper bound @xmath389 for @xmath384 .
note though that we do not expect @xmath52 to enjoy a comparable lower bound , but since @xmath8 is positive in the region of interest , we have @xmath390 for @xmath384 .
finally , from , we have @xmath391 whenever @xmath388 . motivated by the pointwise conservation laws of the equation , which formally converges to , we consider the quantity @xmath392 using , we compute @xmath393 where @xmath394 . from we see that @xmath395 , and thus by we have @xmath396 from this , , and the fundamental theorem of calculus we see that latexmath:[\[\label{psia } @xmath387 and @xmath398 , and some @xmath399 . from we know that @xmath400 is decreasing in the @xmath179 direction , so from we also have the upper bound @xmath401 for all @xmath387 and @xmath398 . to get a lower
bound , suppose that @xmath402 for some @xmath387 and @xmath403 .
then from we have @xmath404 .
if we move backwards in the @xmath118 direction , we thus see that @xmath400 decreases ; continuing this ( by the usual continuity argument ) until we hit the initial surface @xmath405 and applying we conclude that @xmath406 , a contradiction if @xmath335 is large enough .
we thus conclude that @xmath407 for @xmath387 and @xmath403 . combining this with the upper bound and with we contradict
if @xmath335 is large enough , and the claim follows . combining lemmas [ pond ] , [ ponder ] and the definition of @xmath408 we conclude that @xmath409 whenever @xmath410 and @xmath411 or @xmath412 .
since @xmath325 , we thus conclude on iteration that @xmath413 for all @xmath414 $ ] .
thus we have @xmath415 for all @xmath373 .
now we revisit the conservation laws which were implicitly touched upon in the proof of lemma [ ponder ] .
[ apc ] there exists @xmath326 ( depending on @xmath297 and the initial data ) such that the following claim holds : whenever @xmath416 is such that @xmath417 , then @xmath418 and @xmath419 for all @xmath420 . if instead @xmath421 , then we have @xmath422 and @xmath423 for all @xmath420
. let @xmath326 be sufficiently small to be chosen later .
by sign reversal symmetry we may assume that @xmath424 . by spatial reflection symmetry it suffices to prove .
suppose first that @xmath425 , then by lemma [ hold ] we have @xmath426 ( say ) whenever @xmath427 .
applying and the fundamental theorem of calculus , we see that @xmath428 for all @xmath420 ; similarly , if one differentiates in the @xmath118 direction and applies the bound @xmath426 as well as , we obtain @xmath429 and the claim follows .
henceforth we assume @xmath430 . by lemma [ hold ] ( or )
we see that @xmath174 is positive when @xmath427 , so by making the ansatz @xmath431 as before , we see that @xmath52 obeys for @xmath432 .
also , from , , ( and the positivity of @xmath8 ) we see that @xmath433 for all @xmath432 .
our objective is to show that @xmath434 for all @xmath435 . by and
the fundamental theorem of calculus it suffices to show that @xmath436 applying we can discard the @xmath437 factor .
meanwhile , from , , and the fundamental theorem of calculus we have @xmath438 observe that @xmath439 whenever @xmath440 , and that @xmath441 when @xmath442 , and the claim follows .
we now use this law to show a more precise bound on @xmath206 and @xmath146 than is provided by .
we first handle the case when @xmath8 has large derivative .
[ piece ] let @xmath443 , and let @xmath444 $ ] be an interval such that @xmath445 for all @xmath446 ( note that @xmath179 is irrelevant here ) .
then for each @xmath447 $ ] , we have @xmath448 for all @xmath118 in @xmath449 , excluding at most @xmath450 intervals in @xmath449 of length @xmath451 . similarly with the roles of @xmath118 and @xmath179 reversed .
let @xmath326 be a small number ( depending on the initial data and @xmath297 ) to be chosen later .
we may assume that @xmath69 is sufficiently large depending on @xmath452 , since the claim is trivial otherwise .
by space reflection symmetry it will suffice to prove , ; by time reversal symmetry we may assume that @xmath453 is non - negative .
we introduce an auxiliary parameter @xmath454 , and only prove the claim for @xmath453 between @xmath455 and @xmath456 ; setting @xmath457 will then yield the claim .
we establish this claim by induction on @xmath456 , incrementing @xmath456 by steps of @xmath345 at a time .
let us first handle the base case when @xmath458 .
fix @xmath179 . by sign reversal symmetry and lemma [ hold ]
we may assume that @xmath459 whenever @xmath453 has magnitude at most @xmath460 .
applying lemma [ apc ] , we conclude that @xmath461 for all @xmath462 . from the hypotheses ( a ) , ( b )
and we see that @xmath463 , and so we have @xmath464 if we write @xmath465 , @xmath466 we thus have @xmath467 for all @xmath468 .
let @xmath469 , thus @xmath470 is a ( possibly empty ) interval such that @xmath471 for all @xmath472 . also observe from the smoothness of the initial data and that @xmath473 for all @xmath472 . also from we have @xmath474 for all @xmath472 .
suppose first that @xmath475 and @xmath476 for some @xmath477 and some sufficiently large @xmath130 .
then from the bounds on @xmath478 and , we have @xmath479 a continuity argument ( using gronwall s inequality ) then shows that @xmath480 increases exponentially fast ( with rate @xmath481 ) as @xmath41 increases . since @xmath480 is @xmath450 and was @xmath482 at @xmath43 , we arrive at a contradiction unless @xmath43 lies within @xmath483 of the boundary of @xmath470 . similarly if @xmath484 and @xmath476 for some @xmath477 .
we conclude that @xmath485 for all @xmath472 except for those @xmath41 which are within @xmath451 of the boundary of @xmath470 .
now suppose that @xmath486 , then we see from and the bounds on @xmath81 that @xmath487 ; thus the set of @xmath472 for which this occurs must be contained in a single interval of length @xmath488 .
next , if @xmath489 , then from we obtain a bound of the form @xmath490 where @xmath491 is the sign of @xmath492 . applying the continuity and gronwall argument again , either forwards or backwards in time as appropriate , we see that this event can only occur either within @xmath493 of the boundary of @xmath470 , or on an interval of length @xmath493 adjacent to the interval where @xmath486 . putting all of this together
, we see that @xmath494 for all @xmath472 outside of at most @xmath450 intervals of length @xmath493 .
this gives the desired bound .
the bound then follows from .
now suppose inductively that @xmath495 , and that the claim has already been shown for @xmath496 . by inductive hypothesis we only need to establish the claim for @xmath497 $ ] .
fix @xmath179 . by sign reversal symmetry and lemma [ hold ]
we may assume that @xmath459 whenever @xmath453 lies within @xmath460 of @xmath456 .
we can now repeat the previous arguments , except that the interval @xmath470 must first be subdivided by removing the @xmath450 subintervals of @xmath470 of length @xmath493 for which , ( with @xmath179 replaced by @xmath498 ) already failed ( and which are provided by the inductive hypothesis ) , and then working on each remaining subinterval of @xmath470 separately .
note on each such interval we still have the ode ( using the inductive hypothesis , as a substitute for control of the initial data ) .
we omit the details .
now we handle the opposing case when @xmath8 has small derivative .
[ piece-2 ] let @xmath443 , and let @xmath444 $ ] be an interval such that @xmath499 for all @xmath446 ( again , @xmath179 is irrelevant ) .
then one has @xmath500 and @xmath446 . similarly with the roles of @xmath118 and @xmath179 reversed .
this is very similar to lemma [ piece ] , in that we first establish the base case @xmath501 and then induct by steps of @xmath345 , where @xmath326 is a fixed timestep independent of @xmath456 and @xmath69 .
whereas in the proof of lemma [ piece ] we did the base case in detail and left the inductive step to the reader , here we shall leave the base case to the reader and do the inductive step in detail .
thus , assume @xmath502 and that the claim has already been proven for @xmath496 and @xmath503 .
we may assume @xmath504 since the claim follows from , otherwise . by inductive hypothesis and time and space reversal symmetry , we only need to establish the claims , for @xmath505 $ ] . by the sign reversal symmetry and lemma [ hold ]
we may assume that @xmath506 whenever @xmath507 $ ] .
we can assume that @xmath69 is large compared to @xmath297 , @xmath452 , and the initial data since the claim is vacuous otherwise .
let @xmath508 \}$ ] .
observe ( from the smoothness of the initial data ) that @xmath509 whenever @xmath118 lies within @xmath452 of @xmath470 . by inductive hypothesis ( replacing @xmath452 by @xmath510 )
, we conclude that @xmath511 and @xmath512 for all in the @xmath452-neighbourhood of @xmath470 . applying lemma [ apc ]
, we conclude that @xmath513 for all @xmath41 in the @xmath452-neighbourhood of @xmath470 .
thus , if @xmath514 , then we have @xmath515 for all @xmath118 in the @xmath452-neighbourhood of @xmath470 .
the ode @xmath516 blows up either forward or backward in time within a duration of @xmath517 as soon as @xmath518 exceeds @xmath452 . from this and a continuity and comparison argument
, we see that @xmath518 can not exceed @xmath519 for @xmath520 for some constant @xmath130 depending only on @xmath118 , thus @xmath521 for all @xmath520 . applying we
close the induction as required ; the base case is similar .
the _ a priori _ estimates here did not use the full force of the hypotheses ( a)-(c ) ; the condition ( c ) was not used at all , and the strict barrier condition ( b ) could be replaced by the non - strict condition @xmath120 .
also , a careful examination of the dependence of the implied constants on the initial data , combined with a standard limiting argument using the usual local - wellposedness theory reveals that ( a ) can be replaced with a @xmath522 condition on the initial data @xmath523 .
however , we use the hypotheses ( a)-(c ) more fully in the uniqueness theory of the previous section , and the compactness arguments in the next section .
now we can prove theorem [ main - thm ] . fix @xmath298 .
from lemma [ hold ] , the solutions @xmath0 are equicontinuous and uniformly bounded on the region @xmath524 , and hence ( by the arzel - ascoli theorem ) precompact in the uniform topology on this region . in view of the uniqueness theory in section [ unique - sec ]
, we see that to show theorem [ main - thm ] , it suffices to show that any limit point of this sequence obeys the properties ( i)-(vi ) on @xmath301 . accordingly , let @xmath525 be a sequence such that @xmath526 converges uniformly to a limit @xmath8 .
we can now quickly verify several of the required properties ( i)-(vi ) . from
we obtain the property barrier condition ( iv ) , while from lemma [ hold ] we have the lipschitz condition ( i ) . from the strict barrier hypothesis ( b ) and
lemma [ hold ] , we know that the @xmath527 stay bounded away from @xmath31 in a neighbourhood of the initial interval @xmath528 , and so the nonlinearity @xmath529 converges uniformly to zero in this neighbourhood . because of this and , @xmath526 converges uniformly to @xmath125 in this neighbourhood , yielding the initial condition ( iii ) .
now we verify ( v ) .
suppose @xmath76 is a point in @xmath530 such that @xmath198 .
then by lemma [ hold ] and uniform convergence , we can find a neighbourhood @xmath531 of @xmath76 in @xmath530 and a constant @xmath532 such that @xmath533 for all @xmath534 and all sufficiently large @xmath69 .
in particular , the nonlinearity in converges uniformly to zero on @xmath531 as @xmath4 . taking limits , we see that @xmath535 on @xmath531 in the sense of distributions . taking unions over all such @xmath531 , and using null coordinates
we conclude that the distribution @xmath536 is supported on the set @xmath537 .
next , we consider the neighbourhood of a point @xmath76 where @xmath538 ( say ) .
then by lemma [ hold ] and uniform convergence , we can find a diamond @xmath539 centred at @xmath76 ( with length bounded below uniformly in @xmath76 ) such that @xmath540 is non - negative for all @xmath541 and all sufficiently large @xmath69 .
in particular , the nonlinearity in is non - negative on this diamond , which implies by the fundamental theorem of calculus that @xmath542 whenever @xmath543 lie in @xmath539 . taking uniform limits , we conclude that the same statement is true for @xmath8 . by the usual lebesgue - stieltjes measure construction
( adapted to two dimensions ) we thus see that @xmath544 for some positive finite measure @xmath138 on @xmath539 , which implies that @xmath545 in the sense of distributions on @xmath539 .
similarly when @xmath546 ( now replacing @xmath138 by @xmath547 ) .
piecing together these diamonds @xmath539 and neighbourhoods @xmath531 we obtain the claim .
let us first consider the region where @xmath548 vanishes .
applying lemma [ piece-2 ] and taking weak limits , we see that @xmath549 vanishes almost everywhere when @xmath548 vanishes , which of course gives in this region . as the countable union of null sets
is still null , it thus suffice to verify for almost every @xmath114 in the parallelogram @xmath550 , whenever @xmath449 is an interval such that @xmath551 for all @xmath446 ( @xmath179 is irrelevant ) and some @xmath443 . applying lemma [ piece ] and taking square roots , we know that for all @xmath69 , we have @xmath552 with @xmath446 , where we exclude for each fixed choice of @xmath179 , a union @xmath553 of @xmath450 intervals of length @xmath451 from @xmath449 .
we would like to take limits as @xmath554 , but we encounter a technical difficulty : while we know that @xmath159 converges weakly to @xmath206 , this does not imply that @xmath155 converges weakly to @xmath555 , due to the possibility of increasing oscillation of sign in @xmath159 .
the fact that only fails on a bounded number of short intervals for each @xmath179 rules out oscillation in the @xmath118 direction , but one must also address the issue of oscillation in the @xmath179 direction .
fortunately , from we have some monotonicity of @xmath159 in @xmath179 that allows us to control this possibility .
we turn to the details .
as @xmath8 is lipschitz , we can cover the parallelogram @xmath247 by a bounded number of open diamonds @xmath539 in @xmath247 , on which each @xmath8 varies by at most @xmath556 ( say ) .
if @xmath8 takes any value between @xmath557 and @xmath558 on a diamond @xmath539 , then by @xmath8 solves the free wave equation on @xmath539 , so in particular @xmath206 is constant in @xmath179 ( and agrees with @xmath559 whenever the diamond intersects the initial surface @xmath560 ) .
thus it suffices to establish the claim on those diamonds @xmath539 on which @xmath8 avoids the interval @xmath561 $ ] ; by the symmetry @xmath562 we may assume that @xmath563 on @xmath539 , and hence ( for @xmath564 large enough ) @xmath526 is also positive . by , we conclude that @xmath565 is decreasing in the @xmath179 direction .
let @xmath123 be a small number .
we can partition the diamond @xmath539 into @xmath566 subdiamonds of length @xmath137 in a regular grid pattern .
fix @xmath564 sufficiently large depending on @xmath567 , and call a subdiamond _ totally positive _ ( with respect to @xmath564 ) if @xmath568 at every point on this subdiamond ; similarly define the notion of a subdiamond being _ totally negative_. call a subdiamond _ degenerate _ if it is neither totally positive nor totally negative ( i.e. it attains a zero somewhere in the diamond ) .
we claim that at most @xmath569 degenerate subdiamonds . to see this ,
let @xmath570 be a degenerate sub - diamond .
since @xmath565 is decreasing in the @xmath179 direction , we know that @xmath565 must be negative in at least one point on the northwest edge of @xmath570 , and positive in at least one point on the southeast edge .
suppose that @xmath565 is negative at every point on the northwest edge .
then from the monotonicity of @xmath565 in the @xmath179 direction , we see that there can be at most one degenerate subdiamond of this type on each northwest - southeast column of subdiamonds ; thus there are only @xmath571 subdiamonds of this type .
thus we may assume that @xmath565 changes sign on the northwest edge .
but on the line @xmath572 that that edge lies on , we have holding in @xmath539 outside of @xmath450 intervals of length @xmath573 ; also , by hypothesis , we have @xmath574 for @xmath171 in @xmath539 . we conclude ( for @xmath564 large enough ) that there are at most @xmath450 subdiamonds with northwest edge lying on this line @xmath572 for which @xmath565 changes sign on this edge . summing over all @xmath571 possible edges
, we obtain the claim .
fix @xmath137 , and let @xmath575 .
the set of subdiamonds on which @xmath526 is totally positive or totally negative can change with @xmath564 ; however there are only a finite number of possible values for this set for fixed @xmath137 .
hence , by the infinite pigeonhole principle , we may refine the sequence @xmath576 and assume that these sets are in fact independent of @xmath564 .
for any totally positive or totally negative diamond , @xmath565 has a definite sign ; since @xmath565 converges weakly to @xmath206 , we conclude that @xmath577 converges weakly to @xmath555
. since there are no sign changes on this diamond , must hold throughout the subdiamond ( by the intermediate value theorem ) ; we thus conclude that holds on any such subdiamond .
since the measure of all the degenerate sub - diamonds is @xmath578 , we conclude that holds on @xmath539 outside of a set of measure @xmath578 .
letting @xmath579 we obtain the claim .
the only remaining property we need to verify is ( ii ) . by spatial reflection symmetry
, it suffices to show that for each @xmath447 $ ] , the map @xmath580 is piecewise smooth on @xmath581 $ ] , with only finitely many pieces . from ( c ) , we know that @xmath582 vanishes for @xmath118 in a finite union of intervals and points in @xmath581 $ ] .
on any one of these intervals , we know from ( vi ) that @xmath583 also vanishes almost everywhere , which by the lipschitz nature of @xmath8 and the fundamental theorem of calculus ensures that @xmath584 is constant in @xmath118 on each of these intervals , for any fixed @xmath179 .
so it will suffice to verify the piecewise smoothness of @xmath585 for any @xmath179 and on any compact interval @xmath449 of @xmath118 for which @xmath582 is bounded away from zero , so long as the number of pieces is bounded uniformly in @xmath449 and @xmath179 .
fix @xmath449 and @xmath179 . by hypothesis
, we can find @xmath443 such that @xmath586 for all @xmath446 ; in particular , @xmath587 does not change sign on this interval . by lemma [ piece ] , we conclude for each @xmath564 that @xmath588 for @xmath446 outside of @xmath450 intervals of length @xmath589 intersecting @xmath449 . by pigeonholing , we may assume that the number @xmath590 of such intervals is constant ; denoting the midpoints of these intervals by @xmath591 ; without loss of generality we may take @xmath592 and @xmath593 to be the endpoints of @xmath449 .
we may assume from the bolzano - weierstrass theorem and passing to a further subsequence that each of the @xmath594 converge to some limit @xmath595 . between @xmath594 and @xmath596 , excluding those @xmath118 lying within @xmath589 of either endpoint , we may write @xmath597 , where @xmath598 . by a further pigeonholing we may take @xmath599 independent of @xmath564 . using the fundamental theorem of calculus and then taking limits ,
we conclude that @xmath174 is piecewise smooth for @xmath446 , with possible discontinuities at @xmath600 , and with @xmath174 equal to @xmath601 on the interval @xmath602 for any @xmath603 .
the claim follows , and the proof of theorem [ main - thm ] is complete . | we investigate the behaviour of solutions @xmath0 to the one - dimensional nonlinear wave equation @xmath1 with initial data @xmath2 , @xmath3 , in the high exponent limit @xmath4 ( holding @xmath5 fixed ) .
we show that if the initial data @xmath5 are smooth with @xmath6 taking values in @xmath7 and obey a mild non - degeneracy condition , then @xmath8 converges locally uniformly to a piecewise limit @xmath9 taking values in the interval @xmath10 $ ] , which can in principle be computed explicitly . |
the star formation activity and other fundamental galaxy properties such as morphology and gas content vary systematically with redshift , galaxy mass and environment .
while redshifts and galaxy masses are at least conceptually clearly defined , the definition of a galaxy environment is arbitrary and its optimal choice would require an a priori knowledge of the very thing we are trying to identify , that is the physical environmental driver or drivers of galaxy formation and evolution .
most studies nowadays define the `` environment '' either in terms of the local galaxy number density ( the number of galaxies per unit volume or projected area around the galaxy of interest ) , or the virial mass of the cluster or group to which the galaxy belongs , when this applies , because these are the two most easily measurable quantities from spectroscopic or even imaging surveys . in spite of environment being an elusive and arbitrary concept ,
the fact that galaxy properties depend on environment was recognized earlier than the dependence on galaxy mass and redshift ( hubble & humason 1931 ) .
the first quantitative measurement of systematic differences with local environment was the so - called morphology - density relation ( mdr ) in nearby clusters .
the mdr is the observed variation of the proportion of different hubble types with local density , with ellipticals being more common in high - density regions , spirals being more common in low - density regions , and s0s making up a constant fraction of the total population within the cluster virial radius regardless of density ( dressler 1980 , as revisited in dressler et al .
it was subsequently found that a similar mdr also exists in nearby groups ( postman & geller 1984 also revisited in postman et al .
2005 ) and that a qualitatively similar , but quantitatively different , mdr is present in galaxy clusters at redshifts up to 1 ( dressler et al .
1997 , treu et al .
2003 , smith et al .
2005 , postman et al .
2005 ) .
galaxy stellar populations have also long been known to vary systematically with environment ( spitzer & baade 1951 ) : denser environments have on average older stellar populations . at least at some level
, this must be related to the higher incidence of early - type galaxies in high density regions , i.e. to the morphology - density relation . at low redshift the best characterization of the `` star formation - local density '' ( sfd )
relation has come from large redshift surveys .
these studies have conclusively demonstrated that the average galaxy properties related to star formation depend on local density even at large clustercentric radii ; at low densities in clusters ; and outside of clusters , in groups and the general field ( hashimoto et al .
1998 , lewis et al .
2002 , gomez et al .
2003 , kauffmann et al .
2004 , balogh et al .
2004 , see also pimbblet et al .
moreover , they have highlighted the dependence on both galaxy mass and local environment , showing strong environmental trends at a given galaxy mass ( kauffmann et al .
2004 , baldry et al .
2006 ) . to understand the origin of the sfd relation , it is essential to answer two separate questions : at any given galaxy mass , 1 ) how does the proportion of star - forming galaxies vary with density , and 2 ) how does the star formation activity in star - forming galaxies vary with density ? at low - z ,
the evidence for a change in the relative numbers of red / passively - evolving and blue / star - forming galaxies with local environment is overwhelming , but it remains an open question whether star - forming galaxies of similar mass have star formation histories that depend on local density ( balogh et al . 2004a , b , hogg et al .
2004 , gomez et al .
2003 , baldry et al . 2006 ) .
deep redshift surveys have recently extended the study of the sfd relation in the general field to high redshift .
they have revealed that the number ratio of red to blue galaxies increases with local density out to @xmath4 and that we might be witnessing the establishment of the color - density relation at @xmath5 approaching 1.5 ( cucciati et al .
2006 , cooper et al .
2007 , cassata et al .
this suggests that the transition from a star - forming phase to a passive one occurs for a large number of massive galaxies in groups at @xmath6 ( poggianti et al .
in apparent but not substantial contradiction , the average sfr per galaxy at @xmath7 increases instead of decreasing with local density . therefore the sfd relation is inverted with respect to the local universe ( elbaz et al .
2007 , cooper et al . 2008 ) .
thus , the mdr up to @xmath7 is well studied in clusters and has started to be explored in the field ( capak et al .
2007 ) , and the sfd relation is now being investigated in the general field over a similar redshift baseline . in clusters , moran et al .
( 2005 ) have presented the ew(oii)s of ellipticals and s0 galaxies as a function of local density for a cluster at @xmath8 . however
, a detailed study of the relation between star - formation and local environment in distant clusters has not yet been carried out . as a consequence , a comparison of the mdr and the sfd relation has not been possible to date in clusters at high redshift .
this is due to the limited number of well - studied distant clusters with homogeneous data .
deep galaxy redshift surveys have recently made it possible to characterize the global as well as the local environment of galaxies and to study significant samples of groups at high - z ( wilman et al .
2005 , gerke et al .
2005 , balogh et al .
2007 , gerke et al .
2007 , finoguenov et al . 2007 ) .
groups are typically identified as galaxy associations with masses @xmath9 corresponding to velocity dispersions of @xmath10 .
even the largest field surveys , however , include only very few distant systems above this mass ( finoguenov et al .
2007 , gerke et al . 2007 ) . on the other hand , until recently , distant cluster surveys have studied primarily massive clusters . only the latest surveys of optically - selected samples target structures of a wide range of masses , down to the group level ( hicks et al .
2008 , gilbank et al .
2008 , milvang - jensen et al .
2008 , halliday et al .
2004 ) . nowadays
groups are therefore the meeting point of field and cluster studies at high - z , and it has become possible to study the dependence of the sfd relation on global environment ( clusters , groups and the field ) , approaching the question from both perspectives . in this paper , we investigate the relation between star formation activity , morphology and local galaxy density in @xmath0 clusters and groups observed by the eso distant cluster survey ( ediscs ) . the ediscs data set permits an internal comparison with galaxies in poor groups and the field at the same redshifts in a homogeneous way .
to compare our results with clusters in the nearby universe , we use a cluster sample drawn from the sloan digital sky survey ( sdss ) . after presenting the data set ( 2 ) and the definition of our cluster , group and field samples ( 3 ) ,
we outline our method for measuring the local galaxy density at high - z ( 4 ) and describe the low - z cluster sample we use as local comparison ( 5 ) .
the average trends of the fraction of star - forming galaxies and of the [ oii ] equivalent width with local density in clusters ( 6.1 ) is compared with those found in lower density environments in 6.2 and to those in low - z clusters in 6.4 .
the dependence of the star - forming fraction on global cluster properties is presented in 6.3 .
we then analyze the behavior of the average and specific star formation rates in 7 , summarizing the similarities and differences of the ew - density and sfr - density relations in 7.1 .
cluster - integrated star formation rates are derived in 7.2 where we show their relationship with cluster mass and other global cluster properties .
galaxy morphologies are discussed in 8 , where we present the morphology - density relation , the star - forming properties of each hubble type as a function of local density , and the link between the md and the sfd relations .
the latter is compared with results at low redshift in 8.1 .
finally , we summarize our conclusions in 9 .
all equivalent widths and cluster velocity dispersions are given in the rest frame .
all quantities related to star formation are given uncorrected for dust .
we use proper ( not comoving ) radii , areas and volumes .
we assume a @xmath11cdm cosmology with ( @xmath12 , @xmath13 , @xmath14 ) = ( 70,0.3,0.7 ) .
the eso distant cluster survey ( hereafter , ediscs ) is a multiwavelength survey of galaxies in 20 fields containing galaxy clusters at @xmath15 .
candidate clusters were selected as surface brightness peaks in smoothed images taken with a very wide optical filter ( @xmath16 4500 - 7500 ) as part of the las campanas distant cluster survey ( lcdcs ; gonzales et al .
the 20 ediscs fields were chosen among the 30 highest surface brightness candidates in the lcdcs , after confirmation of the presence of an apparent cluster and of a possible red sequence with vlt 20 min exposures in two filters ( white et al .
2005 ) . for all 20 fields ediscs has obtained deep optical photometry with fors2/vlt , near - ir photometry with sofi / ntt , multislit spectroscopy with fors2/vlt , and mpg / eso 2.2/wfi wide field imaging in @xmath17 .
acs / hst mosaic imaging in @xmath18 of 10 of the highest redshift clusters has also been acquired ( desai et al .
other follow - up programmes include xmm - newton x - ray observations ( johnson et al .
2006 ) , spitzer irac and mips imaging ( finn et al . in prep . ) , @xmath19 narrow - band imaging ( finn et al . 2005 ) , and additional optical imaging and spectroscopy in 10 of the ediscs fields targeting galaxies at @xmath20 ( douglas et al . 2007 ) .
an overview of the survey goals and strategy is given by white et al .
( 2005 ) , who also present the optical ground based photometry .
this consists of @xmath21 , @xmath22 and @xmath23 imaging for the 10 highest redshift cluster candidates , aimed at providing a sample at @xmath24 ( hereafter the high - z sample ) and @xmath25 , @xmath21 and @xmath23 imaging for 10 intermediate
redshift candidates , aimed to provide a sample at @xmath26 ( hereafter the mid - z sample ) . in practice , the redshift distributions of the high - z and the mid - z samples partly overlap ( milvang - jensen et al .
2008 ) .
spectra of @xmath27 galaxies per cluster field were obtained , with typical exposure times of 4 hours for the high - z sample and 2 hrs for the mid - z sample .
spectroscopic targets were selected from _
i_-band catalogs ( halliday et al . 2004 ) . at the redshifts of our clusters
, this corresponds to @xmath28 @xmath3 rest frame .
conservative rejection criteria based on photometric redshifts ( pell et al . in prep . ) were used in the selection of spectroscopic targets to reject a significant fraction of non
members while retaining a spectroscopic sample of cluster galaxies equivalent to a purely _ i_-band selected one .
_ a posteriori _ , we verified that these criteria have excluded at most 1 - 3% of cluster galaxies ( halliday et al .
2004 and milvang - jensen et al .
the spectroscopic selection , observations , and catalogs are presented in halliday et al . ( 2004 ) and milvang - jensen et al . (
2008 ) . in this paper
we make use of the spectroscopic completeness weights derived by poggianti et al .
here we only give a brief summary of the completeness of our spectroscopic sample , referring the reader to the previous paper for details .
given the long exposure times , the success rate of our spectroscopy ( number of redshift / number of spectra taken ) is 97% above the magnitude limit used in this study .
a visual inspection of the remaining 3% of the galaxies reveals that most of these are bright , featureless low - z galaxies .
moreover , in our previous paper we computed the spectroscopic completeness as a function of galaxy magnitude and position within the cluster ( appendix a ) , verified the absence of biases in the completeness - corrected sample ( appendix b ) and found that incompleteness has a negligible effect on the [ oii ] properties of our clusters . in this paper , we analyze 16 of the 20 fields that comprise the original ediscs sample .
we exclude two fields that lack several masks of deep spectroscopy ( cl1122.9 - 1136 and cl1238.5.114 ; see halliday et al .
2004 and white et al . 2005 ) .
we also exclude two additional systems ( cl1037.9 - 1243 and cl1103.7 - 1245 ) , each of which has a neighboring rich structure at a slightly different redshift ( milvang - jensen et al .
2008 ) that is indistinguishable on the basis of photometric properties alone .
the names , redshifts , velocity dispersions , and numbers of spectroscopic members for the remaining 16 clusters are listed in table 1 .
the ediscs spectra have a dispersion of 1.32 / pixel or 1.66 / pixel , depending on the observing run .
they have a fwhm resolution of @xmath29 , corresponding to rest - frame 3.3 @xmath3 at z=0.8 and 4.3 @xmath3 at z=0.4 .
the equivalent widths of [ oii ] were measured from the spectra using a line - fitting technique , as outlined in poggianti et al .
this method includes visual inspection of each 1d spectrum .
each line detected in a given 1d spectrum was confirmed by visual inspection of the corresponding 2d spectrum ; this is especially useful to assess the reality of weak [ oii ] lines .
we do not attempt to separate a possible agn contribution to the [ oii ] line , or to exclude galaxies hosting an agn .
we are unable to identify agns in our data , as the traditional optical diagnostics are based on emission lines that are not included in the spectral range covered by most of our spectra
. agn contamination will be most relevant for our study if there are non - starforming , red galaxies in which the [ oii ] emission originates exclusively from processes other than star formation .
we note that only 13% of the spectroscopic sample used here is composed of red galaxies , see 8 . ] with a detected [ oii ] line in emission
. about 40% of these are spirals of types sa or later . among local field galaxies , about half of the red [ oii]-emitting galaxies are liners
whose source of ionization is still debated , while the rest are either star - forming or seyferts / transition objects in which star formation dominates the line emission ( yan et al .
2006 ) . adopting the distribution of agn types observed locally in the field and conservatively assuming that all liners are devoid of star formation , we estimate that the contamination from pure agns in our sample is at most 7% .
moreover , we have verified that the fraction of red emission - line galaxies is not a function of local density .
it remains true , however , that all the trends we observe , and their evolution , may reflect a combination of the variations in the level of both star formation activity _ and _ agn activity .
this should be kept in mind throughout the paper and when comparing our results with any other work . with these caveats in mind
, we conveniently refer to galaxies interchangeably as `` star - forming '' or `` [ oii ] galaxies '' whenever their ew([oii ] ) @xmath30 @xmath3 , adopting the convention that ews are positive in emission .
the detection of the [ oii ] line above this ew limit is essentially complete in our spectroscopic sample ( see poggianti et al .
2006 for details ) .
llclr & & & & + cluster name & short name & @xmath5 & @xmath31 @xmath32 & @xmath33 + & & & @xmath34 & + cl1232.5 - 1250 & cl1232 & 0.5414 & 1080 @xmath35 & 54 + cl1216.8 - 1201 & cl1216 & 0.7943 & 1018 @xmath36 & 67 + cl1138.2 - 1133 & cl1138 & 0.4798 & 732 @xmath37 & 49 + cl1411.1 - 1148 & cl1411 & 0.5201 & 710 @xmath38&22 + cl1301.7 - 1139 & cl1301 & 0.4828 & 687 @xmath39 & 35 + cl1353.0 - 1137 & cl1353 & 0.5883 & 666 @xmath40 & 20 + cl1354.2 - 1230 & cl1354 & 0.7627 & 648 @xmath41 & 21 + cl1054.4 - 1146 & cl1054 - 11 & 0.6972 & 589 @xmath42 & 49 + cl1227.9 - 1138 & cl1227 & 0.6355 & 574 @xmath43 & 22 + cl1202.7 - 1224 & cl1202 & 0.4244 & 518 @xmath44 & 19 + cl1059.2 - 1253 & cl1059 & 0.4561 & 510 @xmath45 & 41 + cl1054.7 - 1245 & cl1054 - 12 & 0.7498 & 504 @xmath46 & 36 + cl1018.8 - 1211 & cl1018 & 0.4732 & 486 @xmath47 & 33 + cl1040.7 - 1155 & cl1040 & 0.7043 & 418 @xmath48 & 30 + cl1420.3 - 1236 & cl1420 & 0.4959 & 218 @xmath49 & 24 + cl1119.3 - 1129 & cl1119 & 0.5500 & 166 @xmath50 & 17 +
as can be seen in table 1 , ediscs structures cover a wide range of velocity dispersions , from massive clusters to groups . for brevity , we will refer collectively to these structures as `` ediscs clusters '' .
in addition , within the ediscs data set it is possible to investigate the spectroscopic properties of galaxies in even less densely populated environments , at the same redshift as our main structures . in a redshift slice within @xmath51 in @xmath5 from the cluster / group targeted in each field , where we are sure the spectroscopic catalog can be treated as a purely _ i_-band selected sample with no selection bias , we have identified other structures as associations in redshift space as described in poggianti et al .
these associations have between 3 and 6 galaxies and will hereafter be referred to as `` poor groups '' .
we did not attempt to derive velocity dispersions for these systems , given the small number of redshifts per group .
in total , our poor group sample comprises 84 galaxies brighter than the magnitude limits adopted for our analysis ( absolute v magnitude brighter than -20 , see below ) .
finally , within the same redshift slices , any galaxy in the spectroscopic catalogs that is not a member of our clusters , groups or poor group associations is treated as a `` field '' galaxy .
this field galaxy sample is composed of 162 galaxies brighter than our limit and should be dominated by galaxies in regions less populated and less dense than the clusters and groups , although will also contain galaxies belonging to poor structures that went undetected in our spectroscopic catalog .
our field sample is therefore far from being similar to the galaxy sample in general `` field '' studies , which will be dominated by a combination of group , poor group , and field galaxies according to our environment definition .
the median redshift is 0.58 for the field sample and 0.66 for the poor group sample .
redshift and ew([oii ] ) distributions of our poor group and field samples are given in poggianti et al .
computing galaxy masses as outlined in 7 , we find that the mass distribution of galaxies varies significantly with environment , progressively shifting towards higher masses from the field to the poor groups to the clusters .
this corresponds to a difference in the galaxy luminosity distributions , which was shown in poggianti et al .
we build field and poor group samples that are matched in mass to the cluster sample ( hereafter the `` mass - matched '' field and poor group samples ) , drawing for each cluster galaxy a field or poor group galaxy with a similar mass . in the following , we will present the results for both mass - matched and unmatched samples , but show only mass - matched values in all figures .
the projected local galaxy density is computed for each spectroscopically confirmed member of an ediscs cluster .
it is derived from the circular area @xmath52 that in projection on the sky encloses the n closest galaxies brighter than an absolute v magnitude @xmath53 .
the projected density is then @xmath54 = n/@xmath52 in number of galaxies per square megaparsec . in the following we use n=10 , as have most previous studies at these redshifts . for about 7% of the galaxies in our sample , the circular region containing the 10 nearest neighbors extends off the chip . since the local densities of these sources suffer from edge effects , they were excluded from our analysis .
densities are computed both by adopting a fixed magnitude limit @xmath55 and by letting @xmath53 vary with redshift between -20.5 at @xmath56 and -20.1 at @xmath8 to account for passive evolution .
absolute galaxy magnitudes are derived as described in poggianti et al .
we have used two different radial limits to derive the mean properties of galaxies in each density bin .
first , we tried using only galaxies within @xmath57 ( the radius delimiting a sphere with an interior mean density of 200 times the critical density , approximately equal to the cluster virial radius ) .
we also tried including all galaxies in our spectroscopic sample , regardless of distance from the cluster center .
the values of @xmath57 computed for our clusters , as well as sky maps showing @xmath57 relative to the extent of the spectroscopic sample , are given in poggianti et al .
( 2006 ) . for most clusters
, our spectroscopy extends to @xmath57 , while severe incomplete radial sampling occurs for one cluster , cl1232 , which will be treated separately when relevant , e.g. in 7.2 .
the results do not change whether we confine our analysis to @xmath57 or use our full spectroscopic sample , nor whether we use a fixed or a varying magnitude limit .
therefore , to maximise the number of galaxies we can use and to minimize the statistical errors , we show the results for @xmath55 and with no radial limit , unless otherwise stated .
we apply three different methods to identify the 10 cluster members that are closest to each galaxy .
these yield three different estimates of the projected local density , which we compare in order to assess the robustness of our results . in the first method
, the density is calculated using all galaxies in our photometric catalogs and is then corrected using a statistical background subtraction .
the number of field galaxies within the circular area @xmath52 and down to the magnitude limit adopted for the cluster is estimated from the @xmath23-band number counts derived for a 4 deg x 4 deg area by postman et al .
( 1998 ) . in the other two methods we include only those galaxies that are considered cluster members according to photometric redshift estimates .
as described in detail in pell et al .
( in prep . )
, photometric redshifts were computed for ediscs galaxies using two independent codes : a modified version of the publicly available hyperz code ( bolzonella , miralles & pell 2000 ) , and the code of rudnick et al .
( 2001 ) with the modifications presented in rudnick et al .
we use two different criteria to retain cluster members and reject probable non - members . in the first case ,
a galaxy is accepted as a cluster member if the integrated probability that the galaxy lies within @xmath51 in @xmath5 from the cluster redshift is greater than a specific threshold for _ both _ photometric redshift codes .
the probability threshold is chosen to retain about 90% of the spectroscopically confirmed members in each cluster . in the other method
a galaxy is retained as a cluster member if the best photometric estimate of its redshift from the hyperz code is within @xmath51 in @xmath5 from the cluster redshift .
the projected local density distributions obtained with the three methods are shown in fig .
[ vdisden ] .
we note that throughout the paper we use only proper ( not comoving ) lengths , areas , and volumes .
for example , our local densities are given as the number of galaxies per @xmath58 , as measured by the rest - frame observer .
this choice is dictated by the fact that with local densities we are investigating vicinity effects , and gravitation depends on proper distances .
using the sdss , we have compiled a sample of clusters and groups at @xmath59 .
this sample serves as a low - redshift baseline with which we can compare our high - z results .
the sdss cluster sample is described in poggianti et al .
( 2006 ) and comprises 23 abell clusters with velocity dispersions between 1150 and 200 @xmath34 , with an average of 35 spectroscopically confirmed members per cluster . to approximate the ediscs spectroscopic target selection , which was carried out at rest - frame @xmath60 @xmath3 , we used a @xmath61-selected sample extracted from the sdss spectroscopic catalogs .
local densities were computed for spectroscopic cluster members ( within 3@xmath31 from the cluster redshift ) that lie within @xmath57 from the cluster center . for sloan ,
the radial cut is necessary to approximate the ediscs areal coverage , which reaches out to about @xmath57 .
we choose a galaxy magnitude limit of @xmath62 , which maximizes the number of galaxies we can use in our analysis and would correspond to @xmath63 at @xmath8 and @xmath64 at @xmath56 under the assumption of passive evolution . as for ediscs
, local densities were derived using the circular area encompassing the 10 nearest neighbors .
two methods were employed to obtain two independent estimates of local densities . in the first method
, we find the distance to the tenth - nearest projected neighbor considering only spectroscopically confirmed members brighter than @xmath62 . in the second method
, we include the 10 nearest projected neighbors that are within the sloan photometric catalog and that have an estimated absolute magnitude satisfying @xmath62 .
absolute magnitudes were derived from observed magnitudes assuming that all galaxies lie at the cluster redshift and using the transformations of blanton et al .
both the spectroscopic and the photometric densities were computed from catalogs of galaxies located over an area much larger than @xmath57 to avoid edge effects . since the spectroscopic completeness within @xmath57 of the sdss clusters is on average about 84% , the first method is likely to underestimate the `` true '' density . using only the photometric catalog , we ignore the fact that some of the 10 galaxies might be in the background or foreground of the cluster .
the second method therefore overestimates the value of the density .
thus , the spectroscopically - based and photometrically - based local densities represent lower and upper limits to the local density .
density values that have been corrected for spectroscopic incompleteness and for background contamination will lie between these two values .
the local density distributions derived with the two methods are shown in the right panel of fig .
[ vdisden ] .
as expected , the distribution of spectroscopically - based densities is shifted to slightly lower densities than the distribution based on photometry , but in the following we will see that the conclusions reached by the two methods are fully consistent .
compared to the density distribution at high - z shown in the left panel of the same figure , the low - z distribution is shifted to lower densities . in poggianti
( in prep . )
we show this is due to the fact that high - z clusters are on average denser by a factor of @xmath65 compared to nearby clusters , with possible strong consequences on galaxy evolution .
the values of ew([oii ] ) measured with the method we used for ediscs spectra are in very good agreement with those measured by brinchmann et al .
( 2004a , b ) for star - forming galaxies in the sdss ( see fig . 3 in poggianti et al . ( 2006 ) ) . however , for ediscs galaxies , the ew([oii ] ) was measured only when the line was present in emission , and a value ew([oii ] ) = 0 was assigned when no line was present .
in addition , each 1d and 2d spectrum was visually inspected .
in contrast , the sdss [ oii ] measurements of brinchmann et al .
( 2004a , b ) are fully automated and can even yield a ( small ) value in absorption that is compatible with 0 within the error and that can not be ascribed to the [ oii ] 3727 line . to take this into account , for sdss clusters
we have used the ews provided by brinchmann et al .
( 2004a , b ) , but have forced the ew([oii ] ) to be equal to 0 when the value provided by brinchmann et al . is @xmath66 @xmath3 in emission .
moreover , to be compatible with ediscs , we do not exclude agns from our sdss analysis .
finally , the redshift range of our sloan clusters was chosen as a compromise to minimize aperture effects while still sampling sufficiently deep into the galaxy luminosity function .
the @xmath67 sdss fiber diameter covers the central 2.4 - 5.4 kpc of galaxies depending on redshift , compared to the @xmath68 ediscs slit covering 5.4 - 7.5 kpc at high redshift . in the following
, we assume that [ oii ] equivalent widths do not change significantly over these different areas .
we first investigate how the strength of the [ oii ] line varies in ediscs clusters as a function of projected local density .
figure [ main ] shows the mean equivalent width of [ oii ] measured over all galaxies that are spectroscopically confirmed members of ediscs clusters ( black symbols ) , in bins of local density .
the three different estimates of density described in 3 ( empty and filled circles and crosses ) yield similar results within the errors . in the following , errors on mean
equivalent widths are computed as bootstrap standard deviations , and errors on fractions are computed from poissonian statistics .
the mean ew computed over all galaxies is consistent with being flat up to a density @xmath69 ( the kendall s probability for an anticorrelation in the three lowest density bins is only 40% ) , then decreases at higher densities .
these trends arise from a combination of the incidence of star - forming galaxies and the relation between density and ew in star - forming galaxies . as shown in fig .
[ fractions ] , the proportion of star - forming galaxies tends to decline at higher density .
the fraction of star - forming galaxies decreases from about 60% to @xmath7030% . using the average of the values given by the 3 membership methods ,
the kendall test gives a 95% probability of an anticorrelation .
in contrast , fig .
[ oiionly ] shows that the mean ew([oii ] ) computed _ only for galaxies with emission lines _ does not correlate with local density ( the kendall s correlation probability is 38% ) .
it is consistent with being flat over most of the density range , except for the highest density bin centered on @xmath71 galaxies per @xmath58 , where it drops by a factor 2 to 3 .
as will be shown in 8 , the highest density bin is populated only by elliptical galaxies , whose weak [ oii ] may be related to the presence of an agn .
it is thus not surprising that this bin stands out from the other bins , where star - forming spirals dominate the mean behavior .
the constancy of the average [ oii ] equivalent width in star - forming galaxies at most densities suggests that as long as star formation is active , it is on average unaffected by local environment , at least in clusters .
the mean ew([oii ] ) among all of our field galaxies is [email protected] @xmath3 for the mass - matched sample ( dashed lines in fig .
[ main ] ) , and [email protected] @xmath3 for the unmatched sample .
similar values are found for poor group galaxies : [email protected] @xmath3 for the mass - matched sample ( dotted lines ) and [email protected] @xmath3 for the unmatched sample .
these values are comparable within the uncertainties to the values measured in the low density regions of the clusters
. similarly , when considering only galaxies with ongoing star formation , the mass - matched field and poor group values are comparable to those at most densities in clusters ( fig . [ oiionly ] ) .
the mean ew of [ oii ] field and poor group galaxies is [email protected] @xmath3 and [email protected] @xmath3 , respectively ( [email protected]@xmath3 and [email protected]@xmath3 in the unmatched samples ) .
finally , the fraction of [ oii ] galaxies in the field is 62@xmath73% ( matched , fig .
[ fractions ] ) and 72@xmath73% ( unmatched ) . in the poor groups ,
the mass - matched [ oii ] fraction is 80.0 @xmath74% ( matched ) and 77@xmath74% ( unmatched ) . the star - forming fraction in the field is compatible with that observed at most densities in clusters , while the poor group fraction is slightly higher .
therefore , relative to clusters , the unmatched poor groups and field have higher average ews and star - forming fractions .
our results indicate that this is primarily due to differences in the galaxy mass distribution with environment . using galaxy samples with similar mass distributions
, we find that the ew properties of star - forming galaxies do not differ significantly between clusters , poor groups , and the field , neither with local density within clusters as shown in the previous section .
to assess whether the relation between star formation and density is the same in structures of different mass , we divide our cluster sample into different velocity dispersion bins and show the correlation found above , between the star - forming fraction and density , in fig.[fraeach ] .
the analysis is now done using only galaxies within @xmath57 .
as above , errors on fractions are computed from poissonian statistics .
the two most massive clusters ( @xmath75 ) exhibit a flatter relation .
i.e. have lower [ oii ] fractions in the three lowest density bins , than clusters with @xmath76 .
corresponds to the threshold between less(more ) massive clusters with an average total cluster [ oii ] fraction above(below ) 50% ( poggianti et al .
( 2006 ) ) . ] in contrast , we find that at low - z this relation is indistinguishable in clusters with @xmath31 above and below @xmath77 , in agreement with previous works that found no dependence of the correlation between the star - forming ( @xmath19-emitting ) fraction and density from the cluster velocity dispersion in the local universe ( lewis et al .
2002 , balogh et al .
2004 ) .
we note that although the relation between [ oii ] fraction and local density varies with cluster mass at high - z , the relation between star formation and _ physical _ three - dimensional space density may be constant , since the distribution of physical densities in each projected 2d density bin varies with cluster mass ( poggianti et al . in prep . ) .
dividing the high - z sample into finer velocity dispersion bins , we do not find a continuous trend with velocity dispersion ( fig.[fraeach ] ) .
systems with @xmath78 may lie at larger or comparable [ oii ] fractions than systems with @xmath79 , but our errorbars are too large to draw any conclusion . in poggianti
et al . ( 2006 ) we studied how the fraction of [ oii ] galaxies depends on the cluster velocity dispersion @xmath31 . in an [ oii ] fraction-@xmath31 diagram , one can identify three groups of structures : a ) high - mass structures , all with low [ oii ] fractions ; b ) low - mass structures with _
high _ [ oii ] fractions and c ) low - mass structures with _ low _ [ oii ] fractions ( the so - called `` outliers '' in poggianti et al .
( 2006 ) ) .
in addition to having a low [ oii ] fraction , the outliers have a low fraction of blue galaxies ( de lucia et al .
2007 ) , a high fraction of early - type galaxies given the measured velocity dispersions ( simard et al .
2008 ) , and peculiar [ oii ] equivalent width distributions ( poggianti et al .
( 2006 ) ) .
hence , galaxies in the outliers resemble those in the cores of much more massive clusters .
the presence of low - mass structures with low [ oii ] fractions in our sample could be responsible for a non - monotonic trend of the [ oii ] fraction - density relation with cluster mass .
we study the dependence of the [ oii ] fraction on local projected density for the three groups separately in the right panel of fig.[fraeach ] . except for the lowest density bin where the results of all three groups are compatible within the errors ,
the trend with local density is different in the three groups . at any given density ,
the star - forming fraction in low - mass , low-[oii ] groups is significantly lower than those in low - mass high-[oii ] systems . in poggianti
et al . ( 2006 ) we found that the global [ oii ] fraction in distant clusters relates to the system mass , but not _ solely _ to the system mass , at least as estimated from the observed velocity dispersion . here
we find that the [ oii ] fraction does not depend solely on projected local density but also on global environment , and that variations in the star - forming fraction - density relation do not depend uniquely on cluster mass . in principle , the correlations with system mass and local density could have a single common origin , from i.e. a correlation between galaxy properties and physical density in three - dimensional space . in a separate paper ( poggianti et al . in prep .
) , we use numerical simulations to investigate the relations between projected local density , physical 3d density , and cluster mass .
the aim of that paper is a simultaneous interpretation of the observed trends with local density and cluster mass presented in this paper and in poggianti et al .
( 2006 ) .
the sdss results are shown as red symbols ( triangles ) in fig .
[ main ] , fig .
[ fractions ] and fig .
[ oiionly ] . at low - z , the mean ew([oii ] )
computed for all galaxies continuously and smoothly decreases with local density ( fig .
[ main ] ) .
this trend is driven by the decrease in the star - forming fraction with local density ( fig . [ fractions ] ) .
using the average of the values obtained with the two density estimates , the kendall s test yields a 98.5% probability of an anti - correlation .
as at high - z , the mean [ oii ] strength of [ oii ] -galaxies does not vary significantly over most of the density range ( kendall s probability 82.6% ) , except for a decrease in the highest density bin ( fig .
[ oiionly ] ) .
these findings are in agreement with previous low - z results based on sdss and 2dfgrs ( lewis et al .
2002 , balogh et al . 2004 ) . from a quantitative point of view , blindly comparing the high - z and the low - z results , at any projected density in common we observe a lower average ew([oii ] ) at low - z .
this is due to both a lower average [ oii ] fraction and a lower average ew([oii ] ) in star - forming galaxies at low - z .
taken at face value , this indicates that both the proportion of star - forming galaxies and the star formation activity in them decrease with time at a given density .
however , it is worth stressing that observing similar projected densities at different redshifts does not imply similar physical densities , since the correlation between projected and 3d density varies with redshift ( poggianti et al . in prep . ) .
hence , a quantitative comparison of results at different epochs at a given projected density can not be interpreted as a direct measure of the decline of the star formation activity with time for similar `` environmental '' physical conditions .
moreover , we stress again that to the low - z density distribution is shifted to lower densities compared to clusters at high - z .
this is due to the fact that high - z clusters are on average denser by a factor of @xmath65 compared to nearby clusters , as discussed in poggianti et al .
( in prep . ) . from a qualitative point of view , the only difference between high- and low - z observations is the fact that the high - z ew(oii ) averaged over all galaxies is consistent with being flat up to a density @xmath80 ( 40% probability for an anticorrelation ) , while the low - z trend smoothly declines towards higher densities with no discontinuity ( 98.5% probability for an anticorrelation ) . for the rest , the trends of [ oii ] equivalent widths and star - forming fraction with local density are qualitatively very similar at @xmath81 and @xmath56 , showing that an [ oii ] fraction - density relation similar to that observed locally is already established in clusters at these redshifts , and that the activity in star - forming cluster galaxies , when assessed from the ew of the [ oii ] line , does not appear to depend strongly on local density at any redshift .
the star formation rate ( sfr ) of a galaxy can be roughly estimated from the [ oii ] line flux in its integrated spectrum .
the equivalent width is the ratio between the line flux and the value of the underlying continuum .
thus , it is not directly proportional to the star formation rate .
for example , a faint late - type galaxy in the local universe usually has a higher equivalent width , but a comparable or lower star formation rate than a more luminous spiral .
for this reason , the analysis presented above does not yield information on absolute star formation rates , but only on the strength of star formation relative to the galaxy @xmath82 rest - frame luminosity , which itself depends on the current star formation activity .
we have derived star formation rates of galaxies with ediscs spectra by multiplying the value of the observed equivalent width by the value of the continuum flux estimated from our broad - band photometry . for the latter we have used the total galaxy magnitude as estimated from spectral energy distribution ( sed ) fitting ( rudnick et al . 2008 ) , assuming that stellar population differences between the galactic regions falling in and out of the slit are negligible ( see 5 ) .
we use the conversion @xmath83 ) \ , erg \ , s^{-1}/(1.26 \times 10^{41})$ ] ( kewley et al . 2004 ) , adopting an intrinsic ( with no dust attenuation ) flux ratio of [ oii ] and @xmath84 equal to unity with no strong dependence on metallicity , as found by moustakas et al .
( 2006 ) . at this stage
we do not attempt to correct our star formation estimates for dust extinction .
locally , the typical extinction of [ oii ] relative to @xmath19 is a factor 2.5 , and the typical extinction at @xmath19 is an additional factor 2.5 - 3 , so our sfr estimates would be corrected by a factor @xmath85 for extinction ,- based sfrs with those derived from @xmath19 narrow - band photometry from finn et al .
( 2005 ) shows that the relation between the two does not strongly deviate from that derived using the local typical factor @xmath85 for extinction . ]
but there are large galaxy - to - galaxy and redshift variations and they are hard to derive using only optical spectra .
dust - free sfrs based on spitzer data of the ediscs clusters will be presented in finn et al .
( in prep ) . adopting the same criteria and galaxy sample as in sec . 4
, we derive the mean star formation rate for galaxies in bins of local projected density and compute errors as bootstrap standard deviations .
the lower panel of fig .
[ mainsfr ] shows the results including all cluster members as black symbols .
the mean sfr is @xmath86 @xmath87 in low - density regions and declines towards denser regions .
the mean sfr might present a maximum at a density between 15 and 40 galaxies/@xmath88 , though within the errors the values of the two lowest density bins may be consistent .
the corresponding mean sfr for all galaxies in the field and poor groups is 0.8 - 1.2 ( 0.9 - 1.3 in the unmatched samples ) , comparable to the average in cluster low - density regions .
considering only star - forming galaxies , the trend with local density in clusters remains similar and is shifted to higher sfr values with a maximum of @xmath89 @xmath90 between 15 and 40 galaxies/@xmath88 ( blue points in fig.[mainsfr ] ) .
errorbars are larger here due to the reduced number of galaxies .
nevertheless , the presence of a peak is hinted at by the data at the 1 - 2@xmath31 level . to further assess the significance of the peak , we have computed mean and median sfr values for star - forming galaxies in 5 , 4 , and 3 equally - populated density bins .
the latter are shown in the top panel of fig.[mainsfr ] , together with the sfr values for individual galaxies .
the equally - populated bins confirm the presence of the peak at the 2 to 4@xmath31 level .
a ks test rejects the null hypothesis of similar sfr distributions in star - forming galaxies in the peak density bin and in each one of the other bins with a 98.4% and 98.2% probability .
the galaxy mass distribution varies only slightly from one density bin to another . in any case
, we have verified that the significance of the peak in the mean and median sfr remains the same when matching the mass distribution of galaxies in the lowest and highest density bin to that in the bin with the peak .
the distribution of individual points in the top panel of fig .
[ mainsfr ] is also visually consistent with higher sfrs for a significant number of galaxies at densities between 15 and 50 galaxies/@xmath88 .
the peak sfr is higher than the mean values of @xmath91 @xmath87 for mass - matched field star - forming galaxies , but is compatible within the errors with the mass - matched poor group value of @xmath92 ( blue lines in fig.[mainsfr ] ) .
the unmatched samples yield similar results ( @xmath93 ) .
figure [ mainssfr ] shows the average _ specific _ star formation rate ( ssfr ) , defined as the sfr per unit of galaxy stellar mass , as a function of local galaxy density .
galaxy stellar masses were computed from rest - frame absolute photometry derived from sed fitting ( rudnick et al .
2008 ) , adopting the calibrations of bell & de jong ( 2001 ) , which are based on a diet salpeter imf .
cluster trends are similar to the sfr - density diagram , reinforcing the picture of a peak and a declining trend on both sides of the peak .
the average specific star formation rates in mass - matched samples of star - forming field galaxies ( @xmath94 yr@xmath95 ) and of poor groups ( @xmath96 yr@xmath95 ) are comparable to those found in the low - density regions of clusters .
interestingly , using the unmatched samples , the field would be markedly distinct from the other environments , having higher specific star formation rates by a factor of two or more .
this shows that on average our star - forming field galaxies are forming stars at more than twice the rate per unit of galaxy mass than star - forming galaxies in any other environment we have observed , and that this is due to their average lower galaxy mass .
our results show that in distant clusters the average star formation rate and the specific star formation rate per galaxy , computed both over all galaxies and only among star - forming galaxies , may not follow a continuously declining trend with density .
the most striking result is the significance of the peak in the sfr of star - forming galaxies discussed above .
the average star formation rate over all galaxies decreases with density in the general field at @xmath81 ( gomez et al .
2003 ) , but distant field studies have found that the relation between average star formation rate over all galaxies and local density was reversed at @xmath7 , when the sfr _ increases _ with density , at least up to a critical density above which it may decrease again ( cooper et al .
2007 , elbaz et al . 2007 ) .
these high - z surveys sample different regions of the universe ( the general `` field '' ) and slightly higher redshifts than our survey ( @xmath97 ) .
the range of projected densities in these studies is likely to overlap with our range only in their highest density bins , but a direct comparison is hampered by the different measurement methods of local density .
it is compelling , however , that both we and these studies find a possible peak plus a possible decline on either side of the peak .
unfortunately , none of these studies sample a sufficiently broad density range to be sure of the overall trend .
it is possible that the sfr per galaxy at redshifts approaching 1 presents a maximum at intermediate densities ( corresponding to the groups / filaments that are common to all of these studies ) , and declines both towards higher and lower density regions .
large surveys sampling homogeneously a wide range of environments and local densities at @xmath98 should be able to address this question . to summarize the results presented in the previous sections , there are some notable differences between the `` star formation - density '' relation as depicted by the observed equivalent widths ( ew([oii])-density relation ) , and that portrayed by the measured star formation rates ( sfr - density relation ) .
the main differences are best seen by comparing fig .
[ oiionly ] with fig .
[ mainsfr ] , and can be described as follows .
the `` strength '' of star formation in star - forming galaxies , when assessed from the ew([oii ] ) , is consistent with being flat with density in clusters ( except for the strong depression in ellipticals in the densest regions ) , and to be rather similar in equally massive field and poor group galaxies .
the `` strength '' of star formation in star - forming galaxies , when represented by the sfr , possibly peaks in clusters at @xmath99 galaxies/@xmath58 , exceeding the field value .
this finding appears robust to any statistical test we have applied .
however , data for larger galaxy samples will be needed to confirm this result .
from the ew([oii])-density relation one would conclude that on average the star formation activity in currently star - forming galaxies is invariant with both local and global environment , while from the sfr - density relation one may conclude that the star formation rate is possibly boosted by the impact with the cluster outskirts , as several studies have suggested ( see e.g. milvang - jensen et al .
2003 , bamford et al .
2005 , moran et al . 2005 ) .
variations in star formation histories and dust extinction with density must play a role in causing the differences between the ew and sfr trends , and may conspire to keep the ew([oii ] ) relation flat .
we have instead verified that variations in the galaxy mass distributions are not responsible for the sfr peak ( see above ) .
the relation between line ew and local density is often considered equivalent to the sfr - density relation , but we have shown here that they provide different views of the dependence of the star formation activity on environment .
we derive cluster - integrated sfrs by summing up the sfrs of individual galaxies within the projected @xmath57 .
we derive the individual sfrs from the [ oii ] line flux as described in the previous section , and weight each galaxy for spectroscopic incompleteness as outlined in 2 .
we do not attempt to extrapolate to galaxy magnitude limits fainter than the spectroscopic limit adopted for this paper , thus sfrs in galaxies fainter than @xmath100 are not included in our estimate .
the cluster - integrated sfr , normalized by the cluster mass ( sfr / m ) is shown as a function of cluster mass in fig .
[ sfrmass ] .
the cluster mass has been obtained from the cluster velocity dispersion using eqn .
4 in poggianti et al .
error bars are computed by propagating the errors on the observed velocity dispersion and the typical 10% error on the [ oii ] flux .
we reiterate that these sfr estimates are not corrected for extinction . from the millennium simulation , we find that mass and radius estimates based on observed velocity dispersions critically fail for systems below @xmath101 , yielding masses that are up to a factor of 10 lower than the true virial mass of the system ( poggianti et al . in prep . ) . as a consequence ,
the masses and the mass - normalized sfrs for the two lowest velocity dispersion systems in our sample ( cl1119 and cl1420 ) are likely to be blatantly incorrect , and will not be used in the analysis .
nevertheless , for completeness we do show the cl1119 point in the diagrams .
cl1420 has no galaxies showing [ oii ] emission .
it therefore has sfr=0 and is not visible in the plots .
all of our other structures have sfr / m between 5 and 50 m@xmath102 yr@xmath95 per @xmath103 @xmath104 m@xmath102 . having excluded cl1119 and cl1420 , the kendall test gives a 95.7% probability for an anti - correlation between sfr / m and cluster mass ( fig .
[ sfrmass ] ) . again without cl1119 and cl1420 , the average sfr / m is 30.4 and 12.4 @xmath105 yr@xmath95 per @xmath103 @xmath104 m@xmath102 for systems below and above @xmath106 , respectively . at redshift @xmath107
there are very few other clusters in the literature with which we can compare .
cluster - integrated sfrs corrected for incompleteness , within a clustercentric distance @xmath108 , are presented by finn et al .
( 2005 ) based on @xmath19 studies for two additional clusters , cl0024 at @xmath8 ( kodama et al . 2004 ) and cl j0023 at @xmath109 .
a similar analysis was carried out by homeier et al .
( 2005 ) for a cluster at @xmath110 , except that it was based on [ oii ] fluxes . both of these works , when including lower redshift clusters , find a possible anti - correlation between the mass - normalized cluster sfr and the cluster mass similar to ours , although it is impossible to separate the redshift dependence from the mass dependence in such small samples .
an overall evolution of the mass - normalized sfr and a large cluster - to - cluster scatter are also found by geach et al .
( 2006 ) using mid- to far - infrared data .
an upper limit in the mass - normalized sfr versus mass plane has been found to exist for clusters , groups and individual galaxies by feulner , hopp & botzler ( 2006 ) .
the right - hand panel of fig .
[ sfrmass ] shows sfr / m versus m for the three clusters from the literature that were the subject of emission - line studies , plotted alongside the ediscs points restricted to the same radius ( = r@xmath111 ) .
the sfrs for the non - ediscs clusters were corrected to account either for slightly different sfr-[oii ] calibrations or for the extinction of [ oii ] relative to @xmath84 ( a factor of 2.5 ) .
including the three clusters from the literature and excluding cl1119 and cl1420 as above , we find that the average sfr / m is 33.4 and 9.8 @xmath105 yr@xmath95 per @xmath112 for systems below and above @xmath113 , respectively .
the kendall test yields an anti - correlation probability of 99.2% .
in contrast , as shown in the left panel of fig .
[ sfroii ] , the cluster - integrated sfr does not correlate with cluster mass ( 60% probability ) , and there is a large scatter in the mass range occupied by the majority of our clusters ( @xmath114 @xmath115 ) . moreover , the right panel of fig .
[ sfroii ] shows that the sfr per unit mass follows the star - forming fraction ( 98% ) .
we caution that the anticorrelation between sfr / m and m presented in fig .
[ sfrmass ] could be entirely due to the correlation of errors .
we tested this possibility by generating 100 realizations of the dataset used in fig .
[ sfroii ] ( i.e. 100 mass - sfr pairs ) , drawn from gaussians with the same means and intrinsic rms , and by adding gaussian errors as observed .
in 41 out of 100 cases the kendall test gave a probability larger than 95.7% that an anticorrelation between mass and sfr / m exists .
therefore the observed anticorrelation could be mainly driven by correlated errors , although this test can not rule out the existence of an intrinsic anticorrelation .
although our sample increases the number of available cluster - integrated sfrs by a factor of four , larger cluster samples , in particular clusters at the highest and lowest masses , are clearly needed to verify our three findings : the weak anticorrelation of sfr / m with m , the lack of a correlation between the integrated sfr and m , and the presence of a correlation between sfr / m and star - forming fraction . to further investigate the robustness and the possible origin of these three results , in the left panel of fig .
[ sfrnum ] we show that the integrated star formation is linearly proportional to the number of star - forming galaxies @xmath116 .
in fact , the integrated sfr is _ equal _ to the number of star - forming galaxies , because the average sfr per star - forming galaxy is roughly constant in all clusters at about @xmath117 .
the correlation between the integrated sfr and the number of star - forming galaxies is much tighter than the relation between the sfr and the total number of cluster members , also shown in fig .
[ sfrnum ] as empty circles . in poggianti
et al . ( 2006 ) we discovered that the star - forming fraction in distant clusters generally follows an anticorrelation with cluster mass , with some noticeable outliers , while in nearby clusters the average star - forming fraction is constant for @xmath118 , and increases towards lower masses with a large cluster - to - cluster scatter .
the star - forming fraction is given by @xmath119}}= n_{\rm sf}/n_{\rm tot}$ ] . in fig .
[ sfrnum ] we examine the mass dependence of both the numerator and denominator of this expression .
we show that in distant clusters the number of star - forming galaxies @xmath116 does not depend on cluster mass ( central panel ) , while the total number of cluster members @xmath120 grows with cluster mass ( right panel ) according to a least squares fit as : @xmath121 at @xmath81 , the star - forming fraction in systems more massive than @xmath122 is constant .
if the average star formation activity _ in star - forming galaxies _ in these clusters is independent of cluster mass , as it is at high redshift , then the cluster - integrated star formation rate at @xmath81 should be not only linearly proportional to the number of star - forming galaxies , but also to the total number of cluster members , as indeed found by finn et al .
( 2008 ) .
moreover , in low - z clusters the relation between the total number of cluster members and cluster mass is ( triangles in the right panel of fig .
[ sfrnum ] ) @xmath123 as a consequence of eqn .
( 2 ) and of the constancy of the star - forming fraction presented in poggianti et al .
( 2006 ) , the number of star - forming galaxies in clusters with @xmath124 @xmath34 at low - z , as well as the total integrated star formation rate , must increase with cluster mass .
this is at odds with what we find at high - z , where both the number of star - forming galaxies and the total sfr are independent of cluster mass ( fig .
[ sfrnum ] and fig .
[ sfroii ] ) .
the different behavior at @xmath125 and @xmath81 is simply due to the different trends of the star - forming fraction with cluster mass at the two redshifts ( poggianti et al .
( 2006 ) ) . in systems with masses below @xmath126 at @xmath81
, @xmath127rm sf}/n_{\rm tot}$ ] is no longer independent of cluster mass , being on average ( poggianti et al .
( 2006 ) ) @xmath128 based on eqns . 2 and 3 and the relation between cluster mass and @xmath31
, we predict that the average number of star - forming galaxies for low - mass systems at @xmath81 should be equal to between 4 and 6 galaxies regardless of group mass for masses between @xmath129 and @xmath130 .
_ if _ the average star formation rate per star - forming galaxy is independent of group mass at low - z , as it is at high - z , then the _ average _ total group star formation rate in the mass range @xmath131 should also be constant , with a very large scatter from group to group at a given mass reflecting the large scatter in the star - forming fraction .
low - z group samples should be able to verify these predictions , which are based purely on the observed correlations presented in this paper and in poggianti et al .
( 2006 ) . because the integrated sfr is equal to the number of star - forming galaxies in distant clusters ,
the former is by definition proportional ( with a proportionality factor that happens to be equal to 1 ) to the star - forming fraction multiplied by the total number of cluster members @xmath132 . in distant clusters ,
the best - fit relation between @xmath133 and cluster mass was given by poggianti et al .
( 2006 ) : @xmath134 from this and from the fact that the total number of cluster galaxies correlate with cluster mass ( eqns . 4 and 1 ) , and
given that the integrated sfr is equal to the number of star - forming galaxies ( fig .
[ sfrnum ] ) , one can analytically conclude that the sfr / m should correlate with the star - forming fraction , as indeed we observe in fig .
[ sfroii ] . to summarize , in distant clusters
we have observed a weak anticorrelation between sfr / m and cluster mass , the lack of any correlation between cluster - integrated sfr and mass , and the presence of a correlation between sfr / m and star - forming fraction .
these findings can be explained , and actually predicted , on the basis of three observed quantities : a ) the constancy of the average sfr per star - forming galaxy in all clusters , found in this paper ( left panel of fig .
[ sfrnum ] ) ; b ) the correlation between cluster mass and number of member galaxies shown in this paper ( eqn . 1 and right panel of fig . [ sfrnum ] ) ; and c ) the previously observed dependence of star - forming fraction on cluster mass ( poggianti et al . ( 2006 ) ) .
observation ( a ) , that the average sfr per star - forming galaxy is constant for clusters of all masses , suggests that either clusters of all masses affect the star formation activity in infalling star - forming galaxies in the same way , or that , if / when they cause a truncation of the star formation , they do so on a very short timescale . in the latter case ,
star - forming galaxies of a given mass have similar properties inside and outside of clusters . observation ( b ) , the correlation between cluster mass and number of cluster members , stems from the mass and galaxy accretion history of clusters .
these results can be used to test the predictions of simulations .
more importantly , comparisons with simulations can allow to explore how our results are linked with the growth history of clusters , which should play an important role in establishing the star - forming fraction .
the relative numbers of star - forming versus non - star - forming galaxies ( observation ( c ) above ) and , above all , its evolution , remain the key observations that display a strong dependence on cluster mass . ultimately , understanding the observed trends comes down to finding out why the relative proportion of passive and star - forming galaxies varies with
environment , the latter being either cluster mass or local density . in poggianti
et al . ( 2006 ) we proposed a schematic scenario in which there are two channels that cause a galaxy to be passive in clusters today : one due to the mass of the galaxy host halo at @xmath135 ( a `` primordial '' effect ) , and one due to the effects related to the infall into a massive structure ( a `` quenching '' mechanism ) .
the results of this paper are consistent with that simple picture .
for 10 ediscs fields we can study galaxy morphologies from visual classifications of hst / acs images ( desai et al .
2007 ) and thus compare our star formation estimates with galaxy hubble types . in particular , we are interested in knowing whether the trend of sf with local density can be partially or fully ascribed to the existence of a morphology - density relation ( mdr ) . do the sf trends simply reflect a different morphological mix at different densities , with the sf properties of each hubble type being invariant with local density ? or do the sf properties of a given morphological type depend on density ? can the lower average sf activity in denser regions be fully explained by the higher proportion of early - type galaxies in denser regions ?
desai et al .
( 2007 ) have published visual classifications in the form of hubble types ( e , s0 , sa , sb , sc , sd , sm , irr ) . however , for this paper we consider only four broad morphological classes : e ( ellipticals ) , s0s ( lenticulars ) , early - spirals ( sa s and sb s ) and late - spirals ( sc s and later types ) .
we note that irregular galaxies ( irr ) represent only 10% of our late - spiral class and therefore do not dominate any of the late - spiral results we present below .
the morphology - density relation for ediscs spectroscopically - confirmed cluster members brighter than @xmath100 is shown in fig .
we find clear trends similar to what has been observed before in clusters both at high- and low - z ( dressler et al .
1997 , postman et al .
2005 , smith et al .
the fraction of spirals decreases , the fraction of ellipticals increases , and the fraction of lenticulars is flat with local density .
previous high - z studies have not considered early - spirals and late - spirals separately .
we find that the spiral trend is due to the fraction of late - spirals strongly decreasing with density , while the distribution of early - spirals is rather flat with density ( top panel of fig .
the early - spiral density distribution is thus very similar to that of s0 galaxies , suggesting that these objects are the best candidates for the immediate progenitors of the s0 population , which has been observed to grow between @xmath136 and @xmath81 ( dressler et al .
1997 , fasano et al .
2000 , postman et al .
2005 , desai et al .
2007 ) .
we consider three observables related to the star formation activity : the [ oii ] ew and the sfr derived from the [ oii ] flux described in the previous sections , and the break at 4000 . the last is defined as the difference in the level of the continuum just bluer and just redder than 4000 .
it can be thought of as a `` color '' and in fact it usually correlates well with broad band optical colors , though it spans a smaller wavelength range than broad bands and is thus less sensitive to the dust obscuring those stars that dominate the spectrum at these wavelengths .
we use the narrow version of this index , sometimes known as @xmath137 , as defined by balogh et al .
( 1999 ) , and refer to it as d4000 in the following .
the observed values of sfr , ew(oii ) , and d4000 are plotted as a function of local density for each of our four morphological classes in fig .
[ ageden ] ( es , s0s , early - spirals , late - spirals from bottom to upper row ) .
the same figure presents the local density distribution of each morphological class .
the highest density bin is only populated by ellipticals , as previously discussed .
two main conclusions can be drawn from this figure : \a ) neither the sfr , nor the ew(oii ) , nor the d4000 distributions of each given morphological class vary systematically with local density . dividing each morphological class into two equally - populated density bins , we find statistically consistent mean and median values of sfr , ew([oii ] ) , and d4000 .
the only exception may be a possible deficiency of galaxies with high sfr among early spirals at the highest densities .
however , the mean and median sfr in the two density bins differ only at the 1@xmath31 level .
thus , as far as it can be measured in our relatively small sample of galaxies , the sf properties of a given morphological class do not depend on density .
\b ) while the great majority of es and s0s are `` red '' ( = = have high values of d4000 , and null values of ew(oii ) and sfr ) and the great majority of late - type spirals are `` blue '' ( = = have low d4000 values , oii in emission and ongoing sf ) , early - spirals are a clearly bimodal population composed of a red subgroup ( @xmath138 ) and a blue subgroup ( @xmath139 ) .
approximately 40% of the early - spirals are red with absorption - line spectra and 40% are blue with emission - line spectra , and the rest have intermediate colors .
most of the intermediate - color galaxies have some emission , but at least two out of 10 have recently stopped forming stars ( have k+a post - starburst spectra ; poggianti et al .
2008 in prep . ) and therefore are observed in the transition phase while moving from the blue to the red group . for their star - formation properties ,
the red early - spirals can be assimilated to the `` passive spirals '' observed in several previous surveys ( poggianti et al .
1999 , moran et al .
2006 , goto et al . 2003 , moran et al . 2007 ) .
the early - spiral bimodality is not due to the red subgroup being composed mainly of sa s and the blue subgroup consisting mainly of sb s , as the proportion of sa s and sb s is similar in the two subgroups .
interestingly , the relative fractions of `` red '' and `` blue '' early - spirals does not strongly depend on density , as might have been expected , but there is a tendency for the intermediate color galaxies to be in regions of high projected local density .
we now want to calculate whether the observed star formation - density relations can be accounted for by the observed morphology - density relation , combined with the average sf properties of each morphological class . to obtain the trend of star - forming fraction with density expected from the mdr , we compute the fraction of star - forming galaxies in each morphological class and combine this with the fraction of each morphological class in each density bin ( i.e. the mdr ) .
the result is compared with the observed star - forming fractions in fig .
[ mainmorph ] . similarly , to compute the expected sfr - density relation given the md - relation ,
we combine the mean sfr in solar masses per year for each morphological class ( [email protected] for es , [email protected] for s0s , [email protected] for early - spirals and [email protected] for late spirals ) with the fraction of each morphological class in each density bin ( i.e. the mdr ) , and compare it with the observed sfr - density relation in fig .
[ mainmorph ] .
this figure shows that the mdr is able to fully account for the observed trends of star - forming fraction and sfr with density ( and viceversa ) .
hence , we find that the md relation and the `` star formation - density relation '' ( in the different ways it can be observed ) are equivalent .
these observations indicate that at least in clusters , for the densities , redshifts , galaxy magnitudes , sf and morphology indicators probed in this study , these are simply two independent ways of observing the same phenomenon , and that neither of the two relations is more `` fundamental '' than the other .
the equivalence between the sfd and the md relations that we find in ediscs clusters is at odds with a number of studies at low redshift . in local clusters ,
christlein & zabludoff ( 2005 ) have found a residual correlation of current star formation with environment ( clustercentric distance in their case ) for galaxies with comparable morphologies and stellar masses . using the las campanas redshift survey , hashimoto et al .
( 1998 ) demonstrated that the star formation rates of galaxies of a given structure depend on local density , and that `` the correlation between ... star formation and the bulge - to - disk ratio varies with environment '' .
more recently , a series of works on other field low - z redshift surveys have concluded that the star formation - density relation is the strongest correlation of galaxy properties with local density , suggesting that the most fundamental relation with environment is the one with star formation histories , not with galaxy structure ( kauffmann et al .
2004 , blanton et al .
2005 , wolf et al .
2007 , ball , loveday & brunner 2008 ) .
most of these studies are based on structural parameters such as concentration or bulge - to - disk decomposition that are used as a proxy of galaxy `` morphology '' .
visual morphologies such as those we use in this paper are known to be related both to structural parameters and star formation . in fact , from a field sdss galaxy sample at low - z , van der wel ( 2008 ) concludes that structure mainly depends on galaxy mass and morphology depends primarily on environment , and that the md relation at low - z is `` intrinsic and not just due to a combination of more fundamental , underlying relations '' .
similarly , park et al . ( 2007 ) argue that the strongest dependence on local density is that of morphology , when morphology is defined by a combination of concentration index , color and color gradients .
interestingly , having fixed morphology and luminosity , these authors find that both concentration _ and _ star formation related observables are nearly independent of local density .
the difference between a classification based on structural parameters and one obtained from visual morphology may be responsible for the differences between our high - z results and most , but not all , low - z results . using visual morphologies of galaxies in the supercluster a901/2 ,
wolf et al .
( 2007 ) find that the mean projected density of galaxies of a given age does not depend on morphological class , and conclude there is no evidence for a morphology - density relation at fixed age . in their sample , except for the latest spirals , which are all young , galaxies of every other morphological type span the whole range of ages , i.e. there are old , intermediate - age , and young es , s0s , and early spirals .
in contrast , as discussed previously , our morphological classes correspond to a strong segregation in age : practically all ellipticals and s0s are old , all late spirals are young , and only early spirals are a bimodal population in age . to facilitate the comparison with wolf et al .
( 2007 ) , in particular with their fig . 5c , in fig . [ wolf ]
we present our results as mean projected density for galaxies of different stellar `` ages '' as a function of morphological class .
the notation `` old '' and `` young '' separates galaxies with red and blue d4000 ( @xmath140 ) .
figure [ wolf ] shows that we find an `` md - relation '' at fixed age , i.e. a difference in mean density for galaxies of the same age but different morphological type , for example between old es and old s0s . since age trends with density are observed only at faint magnitudes by wolf et al . , the fact that their galaxy magnitude limit is 2 mag deeper than ours may partly or fully explain the discordant conclusions .
the sfd and the md relations may be equivalent at bright magnitudes , and decoupled at faint magnitudes . additionally , it is possible that we are observing an evolutionary effect , with star formation and morphology equally depending on density at high - z , but not at low - z
. this might be the case if at high - z most galaxies still retain the morphological class they had `` imprinted '' in the very early stages of their formation , and if at lower redshifts progressively larger number of galaxies are transformed , having their star formation activity and morphology changed .
such transformations are known to have occurred in a significant fraction of local cluster galaxies .
in fact , approximately 60% of today s galaxies have evolved from star - forming at @xmath141 to passive at @xmath81 according to the results of poggianti et al .
( 2006 ) .
if the changes in star formation are more closely linked with the local environment than the related change in morphology , while the latter retains some memory of the initial structure at very high - z ( mostly dependent on galaxy mass ) , a progressive decoupling between the sfd and the md relations would take place at lower redshifts ( see also capak et al .
2007 ) . since the changes in star formation activity and morphology involve progressively fainter galaxies at lower redshifts in a downsizing fashion ( smail et al .
1998 , poggianti et al .
, 2001a , b , 2004 , de lucia et al . 2004 , 2007 ) , the decoupling at low - z should be prominent at faint magnitudes . in this scenario
, the differences between our analysis and wolf s results would be both an evolutionary and a galaxy magnitude limit effect , the two being closely linked . at low - z it should now be possible to fully address these questions and investigate the galaxy magnitude and global environment dependence of the sfd - md decoupling .
ours is so far the only study comparing the sfd and the md relations at high redshift , so other future works may help clarify the redshift evolution of the link between the two relations in clusters , groups , and the field .
we have measured the dependence of star formation activity and morphology on projected local galaxy number density for cluster , group , poor group , and field galaxies at @xmath0 , comparing with clusters at low redshift . at high - z ,
our 16 main structures have measured velocity dispersions between 160 and 1100 @xmath34 , while for our poor groups we did not attempt a velocity dispersion measurement .
the field sample comprises galaxies that do not belong to any of our clusters , groups , or poor groups .
our analysis is based on the [ oii ] line equivalent widths and fluxes and does not include any correction for dust extinction . all galaxies with an ew([oii ] ) greater than 3 @xmath3 are considered to be currently star - forming . although the contamination from pure agns is estimated to be modest ( 7% at most ) , all the trends shown might reflect a combination of both star formation and agn activity .
\1 ) in distant as in nearby clusters , regions of higher projected density contain proportionally fewer galaxies with ongoing star formation .
both at high and low redshift , the average star formation activity in star - forming galaxies , when measured as mean [ oii ] equivalent width , is consistent with being independent of local density .
\2 ) at odds with low - z results , we find that the correlation between star - forming fraction and projected local density varies for massive and less massive clusters , though it is not uniquely a function of cluster mass .
some low - mass groups can have lower star - forming fractions at any given density than similarly or more massive clusters .
\3 ) in our clusters , the average current star formation rate per galaxy and per star - forming galaxy , as well as the average star formation rate per unit of galaxy mass , do not follow a continuously decreasing trend with density , and may display a peak at densities @xmath1 galaxies mpc@xmath2 .
the significance of this peak ranges between 1 and 4 @xmath31 depending on the method of analysis .
this result could be related to the recent findings of an inverted , possibly peaked sfr - density relation in the field at @xmath7 . the ew - density and the sfr - density relations thus provide different views of the correlation between star formation activity and environment .
the former suggests that the star formation activity in star - forming galaxies does not vary with local density , while the latter suggests the existence of a density range in which the star formation activity in star - forming galaxies is boosted by a factor of @xmath161.5 on average .
\4 ) when using galaxy samples with similar mass distributions , we find variations not larger than 1 @xmath31 in the average ew and sfr properties of star - forming galaxies in the field , poor groups , and clusters .
higher average ews , sfrs , and star - forming fractions in the unmatched field and poor group samples compared to clusters are primarily due to differences in the galaxy mass distribution with global environment . as an example , star - forming field galaxies form stars at more than twice the rate per unit of galaxy mass compared to star - forming galaxies in any other environment . together with point 1 ) above
, this suggests that the current star formation activity in star - forming galaxies of a given galaxy mass does not strongly depend on global or local environment .
\5 ) by summing the ongoing sfr of individual galaxies within each cluster we obtain cluster - integrated star formation rates .
we find no evidence for a correlation with cluster mass .
in contrast , the cluster sfr per unit of cluster mass anticorrelates with mass and correlates with the star - forming fraction , although we caution that the anticorrelation with mass could be mainly driven by correlated errors .
the average star - forming galaxy happens to form about one solar mass per year ( uncorrected for dust ) in all of our clusters , making the integrated star formation rate in distant clusters just equal to the number of star - forming galaxies .
these findings can be understood in the light of three additional results that we show : a ) the cluster integrated sfr is linearly proportional ( equal ) to the number of star - forming galaxies ; b ) the total number of cluster members scales with cluster mass as @xmath142 and c ) the star - forming fraction depends on cluster mass in distant clusters as presented in poggianti et al .
given the invariance of the average star formation with cluster mass , as well as with global and local environment ( see points 1 and 4 above ) , the most important thing that remains to be explained is the cause of the cluster - mass - dependent evolution of the relative number of star - forming versus non - star - forming galaxies .
\6 ) defining galaxy morphologies as visually classified hubble types from hst / acs images , we find a morphology - density relation similar to that observed in previous distant cluster studies .
in addition , we find that the trend of declining spiral fraction with density is entirely driven by late - type spirals of types sc and later , while early spirals ( sa s and sb s ) have a flat distribution with local density as s0s do .
\7 ) the star formation properties ( ongoing sfr , ew(oii ) , and d4000 ) of each morphological class do not depend on local density .
galaxies of a given hubble type in distant clusters have similar star formation properties regardless of the local environment .
\8 ) essentially all es and s0s have old stellar populations and all late spirals have significant young stellar populations , while early spirals are a clearly bimodal population , with 40% of them being red and passively evolving and 40% being blue and having ongoing star formation .
the bimodality of the early spirals , together with their resemblance to s0s as far as the morphology - density distribution is concerned , once more suggests that early spirals are the most promising candidates for the progenitors of a significant fraction of the s0 population in clusters today ( see also moran et al .
2007 ) .
\9 ) from the combination of the morphology - density relation and the average properties of each morphological class , we are able to recover the star formation - density relations we have observed .
the morphology - density and the star formation - density relation are therefore equivalent in our distant clusters , and neither of the two relations is more fundamental than the other .
this is at odds with recent results at low - z . among the possible reasons for the discordant conclusions
are differences between visual morphologies and structural parameters , the fainter galaxy magnitude limit reached in low - z studies , and possibly evolutionary effects that can produce a progressive decoupling of the sfd and the md relations at lower redshifts .
we would like to thank the referee , arjen van der wel , for the constructive and careful report that helped us improving the paper .
bmp thanks the alexander von humboldt foundation and the max planck instituut fur extraterrestrische physik in garching for a very pleasant and productive stay during which the work presented in this paper was carried out . the millennium simulation databases used in this paper and the web application providing online access to them were constructed as part of the activities of the german astrophysical virtual observatory .
the dark cosmology centre is funded by the danish national research foundation .
bmp acknowledges financial support from the firb scheme of the italian ministry of education , university and research ( rbau018y7e ) and from the inaf - national institute for astrophysics through its prin - inaf2006 scheme .
baldry , i.k . ,
balogh , m.l .
, bower , r.g . ,
glazebrook , k. , nichol , r.c . , bamford , s.p .
, budavari , t. , 2006 , mnras , 373 , 469 ball , n.m . ,
loveday , j. , brunner , r.j , 2008 , mnras in press ( astro - ph 0610171 ) balogh , m.l . ,
morris , s.l . ,
yee , h.k.c .
, carlberg , r.g . , ellingson , e. , 1997 , apj , 488 , l75 balogh , m.l .
, morris , s.l .
, yee , h.k.c .
, carlberg , r.g . ,
ellingson , e. , 1999 , apj , 527 , 54 balogh , m.l . , eke , v. , miller , c. , et al .
, 2004a , mnras , 348 , 1355 balogh , m.l . , baldry , i.k . ,
nichol , r. , miller , c. , bower , r. , glazebrook , k. , 2004b , apj , 615 , l101 balogh , m.l .
, wilman , d. , henderson , r.d.e . , et al . , 2007 , mnras , 374 , 1169 bamford , s.p . , milvang - jensen , b. , aragn - salamanca , a. , simard , l. , 2005 , mnras , 361 , 109 bell .
e. , de jong , r. , 2001 , apj , 550 , 212 blanton , m.r . , brinchmann , j. , csabai , i. , doi , m. , eisenstein , d. , fukugita , m. , gunn , j.e . ,
hogg , d.w . ,
schlegel , d.j . , 2003 ,
aj , 125 , 2348 blanton , m.r . , eisenstein , d. , hogg , d.w . ,
schlegel , d.j . , brinchmann , j. , 2005 , apj , 629 , 143 bolzonella , m. miralles , j
.- m . , pell , r. , 2000 , a&a , 363 , 476 brinchmann , j. , charlot , s. , white , s.d.m . ,
tremonti , c. , kauffmann , g. , heckman , t. , brinkmann , j. , 2004a , mnras , 351 , 1151 brinchmann et al .
2004b , astro - ph 0406220 , announcement of data - release capak , p. et al .
2007 , apjs , 172 , 284 cassata , p. et al . , 2007 , apjs , 172 , 270 clowe , d. , et al . , 2006 ,
a&a , 451 , 395 cooper , m.c . , newman , j.a . ,
croton , j.a . , et al
. , 2006 , mnras , 370 , 198 cooper , m.c . , newman , j.a . , weiner , b.j .
, et al . , 2007 , mnras , 376 , 1445 cooper , m.c . , newman , j.a . , weiner , b.j .
, et al . , 2008 , mnras , in press ( astro - ph 07064089 ) cucciati , o. , iovino , a. , marinoni , c. , et al . , 2006 , a&a , 458 , 39 christlein , d. , zabludoff , a.i . , 2005 , apj , 621 , 201 de lucia , g. , poggianti , b.m . , aragn - salamanca , a. , et al . , 2004 ,
apj , 610 , l77 de lucia , g. , blaizot , j. , 2007 , mnras , 375 , 2 de lucia , g. et al . , 2007 , mnras , 374 , 809 desai , v. , et al .
, 2007 , apj , 660 , 1151 douglas , l.s . , bremer , m.n . , stanway , e. , lehnert , m. , 2007 , mnras , 376 , 1393 dressler , a. 1980 , apj , 236 , 351 dressler , a. , oemler , a. , couch , w.j . , smail , i. , ellis , r.s
. , barger , a. , butcher , h. , poggianti , b.m . ,
sharples , r.m . , 1997 , apj , 490 , 577 elbaz , d. , daddi , e. , le borgne , d. , et al . , 2007 ,
a&a , 468 , 33 fasano , g. , poggianti , b. m. , couch , w. j. , bettoni , d. , et al .
2000 , apj , 542 , 673 feulner , g. , hopp , u. , botzler , c.s .
, 2006 , a&a , 451 , l13 finn , r.a .
, zaritsky , d. , mccarthy , d.w .
, poggianti , b.m . ,
rudnick , g. , halliday , c. , milvang - jensen , b. , pell , r. , simard , l. , 2005 , apj , 630 , 206 finn , r.a . ,
balogh , m.l .
, zaritsky , d. , miller , c.j . ,
nichol , r.c .
, 2008 , apj in press ( astro - ph 0802.2282 ) finoguenov , a. , et al . , 2007 , apjs , 172 , 182 gao , l , navarro , j.f . ,
cole , s. , frenk , c.s .
, white , s.d.m . ,
springel , v. , jenkins , a. , neto , a.f . , 2008 , mnras in press ( astro - ph 0711.0746 ) geach , j.e . ,
smail , i. , ellis , r.s . ,
moran , s.m . ,
smith , g.p . , treu , t. , kneib , j .-
edge , a.c . ,
kodama , t. , 2006 , apj , 649 , 661 gerke , b.f .
et al . , 2005 ,
apj , 625 , 6 gerke , b.f . ,
newman , j.a . ,
faber , s.m .
, et al . 2007 , mnras , 376 , 1425 gilbank , d.g . , yee , h.k.c . , ellingson , e. , gladders , m.d . ,
loh , y .- s .
, barrientos , l.f . ,
barkhouse , w.a .
, 2008 , apj , in press ( astro - ph 07102351 ) gomez , p.l . ,
nichol , r.c . ,
miller , c.j .
, et al . , 2003 , apj , 584 , 210 gonzalez , a.h . ,
zaritsky , d. , dalcanton , j.j . , nelson , a. , 2001 , apjs , 137 , 117 goto , t. , et al .
, 2003 , pasj , 55 , 757 halliday , c. et al . , 2004 , a&a , 427 , 397 hashimoto , y. , oemler , a. , lin , h. , tucker , d.l .
, 1998 , apj , 499 , 589 hicks , a.k . ,
ellingson , e. , bautz , m. , cain , b. , gilbank , d.g . , gladders , m.g . , hoekstra , h. , yee , h.k.c . ,
garmire , g. , 2008 , apj in press ( astro - ph 07105513 ) hogg , d.w . ,
et al . , 2004 , apj , 601 , l29 homeier , n.l . ,
et al . , 2005 , apj , 621 , 651 hubble , e. , humason , m.l . , 1931 , apj , 74 , 43 johnson , o. et al . , 2006 , mnras , 371 , 1777 kauffmann , g. , white , s.d . , heckman , t.m . , menard , b. , brinchmann , j. , charlot , s. , tremonti , c. , brinkmann , j , 2004 , mnras , 353 , 713 kewley , l.j . , geller , m.j . , jansen , r.a . , 2004 , aj , 127 , 2002 kodama , t. , balogh , m.l . ,
smail , i. , bower , r.g . ,
nakata , f. , 2004 , mnras , 354 , 1103 lewis , i. , balogh , m. , de propris , r. et al . , 2002 , mnras , 334 , 67 milvang - jensen , b. , aragn - salamanca , a. , hau , g.k.t . , jorgensen , i. , hjorth , j. , 2003 , mnras , 339 , 1 milvang - jensen , b. , et al . , 2008 , a&a in press ( astro - ph 0802.0149 ) moran , s.m . ,
ellis , r.s . ,
treu , t. , smail , i. , dressler , a. , coil , a.l . ,
smith , g.p . , 2005 ,
apj , 634 , 977 moran , s.m . ,
ellis , r.s . ,
treu , t. , salim , s. , rich , r.m . ,
smith , g.p .
, kneib , j .- p . , 2006 ,
apjl , 641 , l97 moran , s.m . ,
ellis , r.s . , treu , t. , smith , g.p . , rich , r.m . , smail , i. , 2007 , apj , 671 , 1503 moustakas , j. , kennicutt , r. , tremonti , c.a . , 2006 , apj , 642 , 775 park , c. , choi , y .- y . , vogeley , m.s . , gott , r. , blanton , m.r .
, 2007 , apj , 658 , 898 pimbblet , k.a . ,
smail , i. , kodama , t. , couch , w.j . ,
edge , a.c . ,
zabludoff , a.i . , ohely , e. , 2002 , mnras , 331 333 poggianti , b. m. , smail , i. , dressler , a. , couch , w. j. , barger , a. j. , butcher , h. , et al . 1999 ,
apj , 518 , 576 poggianti , b.m . , bridges , t.j . ,
mobasher , b. et al . , 2001a , apj , 562 , 689 poggianti , b.m . ,
bridges , t.j . , carter , d. et al . , 2001b , apj , 563 , 118 poggianti , b.m . ,
bridges , t.j . , komiyama , y. et al . , 2004 , apj , 601 , 197 poggianti , b.m . , et al . , 2006 , apj , 642 , 188 postman , m. , geller , m.j . , 1984 , apj , 281 , 95 postman , m. , lauer , t.r . ,
szapudi , i. , oegerle , w. , 1998 , apj , 506 , 33 postman , m. et al . , 2005 , apj , 623 , 721 rudnick , g. , franx , m. , rix , h .- w . , moorwood , a. , kuijken , k. , van starkenburg , l. , van der werf , p. , rottgering , h. , labbe i. , 2001 , aj , 122 , 2205 rudnick , g. , et al .
, 2003 , apj , 599 , 847 simard , l. , et al . , 2008 ,
a&a , in press smail , i. , edge , a.c . ,
ellis , r. s. , blandford , r.d .
, 1998 , mnras , 293 , 124 smith , g.p . , treu .
t. , ellis , r.s . , moran , s.m . ,
dressler , a. , 2005 , apj , 620 , 78 spitzer , l. , baade , w. , a951 , apj , 113 , 413 springel , v. , white , s.d.m . , jenkins , a. , et al .
2005 , nature , 435 , 629 treu , t. , ellis , r.s . , kneib , j .-
p . , dressler , a. , smail , i. , czoske , o. , oemler , a. , natarajan , p. , 2003 ,
apj , 591 , 53 van der wel , a. , 2008 , apjl , in press ( astro - ph 08011995 ) whiley , i. , et al . , 2008 , mnras , in press ( astro - ph ? ? ? ) white , s.d.m . , clowe , d. , i. , simard , l. , et al .
2005 , a&a , 444 , 365 wilman , d.j . ,
balogh , m.l .
, bower , r.g . ,
mulchaey , j.s . ,
oemler , a. , carlberg , r.g . ,
morris , s.l . ,
whitaker , r.j .
, 2005 , mnras , 358 , 71 wolf , c. , gray , m.e .
, aragn - salamanca , a. , lane , k.p .
, meisenheimer , k , 2007 , mnras , l1 yan , r. , newman , j.a . ,
faber , s.m . , .
konidaris , n. , koo , d. , davis , m. , 2006 , apj , 648 , 281 versus projected local density .
the mean ew([oii ] ) is computed over all galaxies .
the high redshift points ( ediscs ) , described in 6.1 , are shown in black as empty circles ( statistical subtraction ) , filled circles ( photo - z probable members ) and crosses ( photo - z within @xmath51 from @xmath143 ) .
the low redshift points ( sdss ) , described in 6.4 , are shown in red as empty triangles ( density computed using only spectroscopic members ) and filled triangles ( full photometric catalog ) .
errors are computed as bootstrap standard deviations from the mean using 100 realizations .
the horizontal dashed and dotted lines delimit the values found in the field and in poor groups at high redshift using mass - matched samples ( see text ) .
[ main],width=604 ] in emission versus local density .
symbols as in fig .
black symbols ( circles ) are ediscs points at @xmath0 ( 6.1 ) .
red symbols ( triangles ) are sdss points at low redshift ( 6.4 )
. errors on data points are computed from poissonian statistics .
the horizontal dashed and dotted lines delimit the value for field and poor group galaxies at @xmath0 .
[ fractions],width=604 ] ) only among galaxies with [ oii ] in emission ( ew@xmath30 ) .
symbols as in fig .
black symbols ( circles ) are ediscs points at @xmath0 ( 6.1 ) .
red symbols ( triangles ) are sdss points at low redshift ( 6.4 ) .
errors are computed as bootstrap standard deviations .
the horizontal dashed and dotted lines delimit the value for mass - matched field and poor group galaxy samples at @xmath0 .
[ oiionly],width=604 ] .
large squares represent the average of the values for the 3 membership methods .
a small shift around the center of each density bin has been applied to the different points to avoid confusion .
the dashed and dotted lines delimit the 1@xmath31 error around the value for field and group galaxies , respectively . for clarity ,
only the star - forming field and poor group values for the mass - matched samples are shown . *
top * sfr in solar masses per year for all individual galaxies .
the mean ( filled circles ) and median ( empty circles ) sfr in star - forming galaxies are shown for three equally populated density bins ( see text ) .
[ mainsfr],width=604 ] for galaxies in different density bins .
all galaxies : black symbols . only star - forming galaxies : blue symbols .
different symbols indicate the values obtained for the 3 membership criteria , as in fig .
large squares represent the average of the values for the 3 membership methods .
a small shift around the center of each density bin has been applied to the different points to avoid confusion .
the dashed and dotted lines delimit the 1@xmath31 error around the value for field and group galaxies , respectively . for clarity ,
only the star - forming field and poor group values for the mass - matched samples are shown .
[ mainssfr],width=604 ] and fig .
[ mainsfr ] .
the large , solid squares represent the values expected given the morphology - density relation and the mean star - forming fraction and sfr of galaxies of each morphological class .
[ mainmorph ] ] and fig .
[ mainsfr ] .
the large , solid squares represent the values expected given the morphology - density relation and the mean star - forming fraction and sfr of galaxies of each morphological class .
[ mainmorph ] ] ) .
practically all ellipticals and s0s are old , and all late spirals are young , while early spirals are cleanly divided into two age populations with similar mean densities .
errors are bootstrap standard deviations .
[ wolf],width=453 ] | we investigate how the [ oii ] properties and the morphologies of galaxies in clusters and groups at @xmath0 depend on projected local galaxy density .
we also compare our results with those derived from the field at similar redshifts and from clusters at low - z .
in both nearby and distant clusters , higher - density regions contain proportionally fewer star - forming galaxies . at odds with low - z results ,
the relation between star - forming fraction and local density at high - z varies from high- to low - mass clusters , although is not solely a function of cluster mass .
both at high- and low - z , the average [ oii ] equivalent width ( ew ) of star - forming galaxies is independent of local density .
however , in distant clusters the average current star formation rate in star - forming galaxies seems to peak at densities @xmath1 galaxies mpc@xmath2 .
when using galaxy samples with similar mass distributions , we do not find large variations in the average ew(oii ) and sfr properties of star - forming galaxies in the distant field , poor groups , and clusters .
overall , these results suggest that at high - z the current star formation activity in star - forming galaxies of a given galaxy mass does not depend strongly on global or local environment , though the possible sfr peak at intermediate densities seems at odds with this conclusion .
we derive cluster - integrated star formation rates and find that the cluster sfr normalized by cluster mass anticorrelates with mass and correlates with the star - forming fraction , although we caution that the anticorrelation with mass could be mainly driven by correlated errors .
we demonstrate that these trends can be understood given a ) that the average star - forming galaxy forms about one solar mass per year ( uncorrected for dust ) in all our clusters ; b ) how the total number of galaxies scales with cluster mass ; and c ) the dependence of star - forming fraction on cluster mass .
these findings show that the most fundamental open question currently remains why the evolution of the star - forming fraction depends on cluster mass .
we present the morphology - density relation for our @xmath0 clusters , and uncover that the decline of the spiral fraction with density is entirely driven by galaxies of types sc or later , while early spirals ( sa s and sb s ) have a flat distribution with density , similarly to that of s0s . for galaxies of a given hubble type , we see no evidence that star formation properties ( ongoing sfr , [ oii ] equivalent width , 4000 @xmath3 break ) depend on local environment .
in contrast with recent findings at low - z , in our distant clusters the star formation - density relation and the morphology - density relation are equivalent , suggesting that neither of the two relations is more fundamental than the other .
we suggest that the contrasting results at high- and low - z are partly due to different classification methods and/or to evolutionary effects giving rise to a progressive decoupling of the two relations at recent epochs . |
unification by orientation is the simplest picture historically invoked to explain the difference between different classes of active galactic nuclei ( agn ; e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and in particular , between luminous radio galaxies and radio - loud quasars ( e.g. * ? ? ?
* ) . in this scheme , radio galaxies ( rgs ) of the frii type @xcite and radio - loud quasars ( rlqsos )
are drawn from the same parent population , and each is only seen from a particular range of viewing angles with respect to the radio jet axis . from along this axis ( `` pole - on '' , viewing angle @xmath0 ) to some limiting angle @xmath1 , one can see directly to the active nucleus and the optical properties will be consistent with radio - loud quasars . beyond @xmath1 , to a completely `` edge - on '' ( @xmath2 ) perspective , one will see optical properties of a radio galaxy .
the limiting angle is assumed to be determined by the opening angle of an obscuring structure , such as a dusty torus , that blocks our view into the nuclear regions of rgs ( e.g. * ? ? ?
* ; * ? ? ?
barthel ( 1989 ) published the first study constraining @xmath1 ( @xmath3 ) and seemingly confirming the simplest prediction of orientation unification ; that the projected sizes of rlqsos in radio images should be smaller than those of rgs , due purely to geometric effects .
although the idea that orientation could explain all properties of agn remained somewhat controversial ( e.g. * ? ? ? * ) , unification through orientation has become quite popular for other types of objects beyond rgs and rlqsos ( e.g. * ? ? ?
the nature of this controversy remains rich enough to have recently expanded to other tests over a wider parameter space , such as environmental measures over broad redshift ranges .
if orientation sufficiently explains agn type , then the environments of different agn should be identical throughout cosmic history .
@xcite demonstrated that rgs and rlqsos occupy similar environments , although obscured agn , which are often discussed within the `` dusty torus '' orientation framework , may have significantly different environments @xcite .
barthel s ( 1989 ) original test remains fertile ground , and newer , larger samples covering both a wider range in redshift and radio luminosity are now being used to revisit the simple projected size tests conducted over 20 years ago . both singal et al . (
2013 ) and boroson et al .
( 2013 , in prep ) find results that contradict those of barthel ( 1989 ) , with independent samples that show no difference in projected radio - source size distributions ( see 2 for more discussion of these results ) .
therefore , they argue , factors other than orientation are necessary to differentiate between rgs and rlqsos .
there is another possibility that can reconcile the seemingly contradictory size distribution results and keep the orientation by unification picture intact . in the simplest case ,
all objects have a single intrinsic size .
thus , if a viewing angle difference is present it will be manifested as a projected size difference in 100% of experiments .
but , this single - size picture is highly idealized
so , assuming , instead , that samples of rgs and rlqsos are drawn randomly from some underlying _ distribution _ of intrinsic sizes ( which is the same for both types of object ) , is it possible to produce both sets of results ? while the effect of viewing angle on the projected sizes of objects is the main focus of all previous work on this subject , _ what is the effect of the underlying intrinsic size distribution _ ?
the purpose of this paper is to explore these effects via monte - carlo modeling of randomly oriented objects generated from a reasonable intrinsic size distribution , using the properties of the samples in the three works mentioned above as constraints .
our main goal is to explore statistically , given an intrinsic size distribution and assuming that unification by orientation is correct , how likely one is to conclude that orientation drives apparent radio - source sizes .
we also explore the effects of varying parameters such as the underlying intrinsic size distribution , the size of the sample , the relative number of rgs to rlqsos in the sample , and the role of different values of @xmath1 .
finally , we attempt to determine the ideal sample that is needed to reliably test the paradigm of unification through orientation for rgs and rlqsos .
we use a cosmology where @xmath4 km s@xmath5 mpc@xmath5 , @xmath6 , and @xmath7 for all calculated parameters ( komatsu et al .
2011 ) , but we note the use of even more recent cosmologies will make no difference in our results
. the notation @xmath8rg@xmath9 and @xmath8rlq@xmath9 will be used throughout to indicate the median projected sizes of rgs and rlqsos , respectively , in a sample or simulation .
we utilize three real samples that have been used for this kind of projected size test ; those of barthel ( 1989 ; b89 ) , singal et al .
( 2013 ; s13 ) , and boroson et al . ( 2013 , in prep ; b13 ; see also boroson 2011 ) .
we briefly summarize these below , but refer the reader to the original references for full details .
some properties of the samples are summarized in table [ samples ] .
see also the appendix for an analysis of the significance of the results in these works when median statistics are taken into account .
we acknowledge that there are other complications that could arise in projected size tests , beyond the effects of viewing angle and the distribution of intrinsic sizes
. there may be issues with how source sizes are measured , bending and rotation of radio jets , and sample selection biases .
additionally , it is well known that there are two subclasses of radio galaxy the high - excitation and low - excitation radio - galaxies ( hergs / lergs , respectively ; hine & longair 1979 , laing et al . 1983 ) .
it is unlikely that lergs participate in orientation unification schemes , as they probably do nt harbor an agn .
singal et al . (
2013 ) discuss this point explicitly .
however , none of the samples discussed here use hergs explicitly . as our goal here is to test whether a distribution of intrinsic sizes can reconcile already published results , we will not make any additional cuts , nor any re - measurements of sizes of the original sources .
this sample is drawn from the 3crr catalog @xcite , which is complete to @xmath1010.9 jy at 178 mhz .
the high flux limit automatically restricts the sample to luminous radio sources at these redshifts ( though not strictly frii sources ) .
it includes all of the quasars ( 17 ) and radio galaxies ( 33 ) in the catalog with redshifts between @xmath11 .
however , we note that there are two rgs in the sample with updated redshifts that place them outside of this range : 3c318 moves from @xmath12 to @xmath13 @xcite , and 3c325 moves from @xmath14 to @xmath15 @xcite .
we include these here with the updated redshifts in order to keep the relative numbers the same as the original test , but excluding them ( or including them with their original @xmath16 values ) does not significantly affect our results .
the results from b89 are consistent with an orientation - based scheme .
the projected size distribution of rgs is shifted toward larger values when compared to the qsos , with the median rg size 2.2 times larger than the median rlqso size ( @xmath17 ) .
this is roughly consistent with the @xmath1045 viewing angle separating the two classes derived from the relative number of objects in each class .
this sample is selected by matching radio data from 3 radio surveys with the sloan digital sky survey ( sdss ) .
the radio surveys utilized are the westerbork northern sky survey ( wenss ; rengelink et al .
1997 ) , faint images of the radio sky at twenty - centimeters ( first ; becker et al .
1995 ) , and the nrao vla sky survey ( nvss ; condon et al .
the low frequency observations of wenss are used to select a sample unbiased in orientation , as these frequencies are not sensitive to beamed radio core emission . a match between the optical position and a radio core
was not required , and a substantial fraction shows lobes but no core .
an explicit cut in radio luminosity ( @xmath18(325 mhz ) @xmath19 w hz@xmath5 ) excludes low - luminosity radio sources .
this threshold is well above the flux limit of wenss within the redshift range ( @xmath20 ) of the sample . only objects targeted with sdss spectra as part of complete samples are included , and comparisons with other radio samples indicate that the only objects that are missing are a small number of optically faint radio galaxies .
essentially all of the objects have steep radio spectra and fr ii radio morphology .
the sample includes 51 rgs and 35 rlqsos , which can be distinguished by the sdss spectroscopy .
the angular sizes were measured from the first images , which have sufficient resolution to easily resolve them at these redshifts .
the findings using this sample contradict the original test by b89 .
rlqsos are found to be larger than rgs by a median factor of 1.6 ( @xmath21 ) , which is used to argue that rlqsos can not be the same objects as rgs seen from smaller viewing angles .
this sample is drawn from the molonglo reference catalog ( mrc ; kapahi et al .
the catalog is also flux limited at low - frequency , but deeper than 3crr ( complete to @xmath100.95 jy at 408 mhz ) .
however , the depth is shallow enough that only luminous radio sources of the type considered for unification are included . the redshift range is larger than that of the b89 sample , with @xmath22 .
the full sample includes 379 rgs and 87 rlqsos , though various subsamples are also studied ( different redshift ranges , radio luminosities , excluding compact steep spectrum ( css ) sources , etc . ) .
we will only consider the full sample here as results are consistent amongst the various subsets considered by s13 .
the s13 results contradict b89 as well , as no difference is seen in the projected size distributions , even when limited to the same redshift range .
again , any difference suggested by the data indicates that the rlqsos are _ larger _ by a median factor of 1.3 ( @xmath23 ) , roughly in agreement with b13 .
lcccccccc b89 & 50 & 33 & 17 & 0.66 & 0.34 & @xmath24-@xmath25 & 2.2 & barthel et al .
( 1989 ) + b13 & 86 & 51 & 35 & 0.59 & 0.41 & @xmath26-@xmath24 & 0.6 & boroson et al .
( in prep ) + s13 & 466 & 379 & 87 & 0.81 & 0.19 & @xmath26-@xmath27 & 0.8 & singal et al . ( 2013 )
in order to build a distribution of intrinsic sizes to draw from , we start with the data from b89 , s13 and b13 . we use the apparent angular sizes and redshifts of each object , and assign each a random viewing angle consistent with a purely geometrically defined distribution
( i.e. an edge - on orientation is much more likely than a pole - on one ; @xmath28 ) .
an intrinsic size is calculated for each object , and the process is repeated @xmath29 times . the resulting distribution of sizes is shown for each sample in the first three panels of figure [ intrinsicmodels ] , in 50 kpc bins .
this method essentially provides the `` widest '' possible distribution of sizes , as the large number of iterations allows each object to be assigned a wide range of orientations . imposing a viewing angle restriction on the rg and rlqso subsamples ( i.e. only allowing rgs to be seen from 90 to @xmath1 degrees and rlqsos from @xmath1 to 0 degrees ) does not change the resulting shape substantially .
it simply changes which objects populate which part of the distribution .
this is due to the fact that the apparent angular size distributions of the two subsets are similar . we next fit functions to these generated distributions .
the forms of the fits are determined purely empirically , and we assign them no physical significance .
we find that two functions are needed in different regions to adequately fit the distributions . in natural logarithmic space , below 500 kpc a power law is used , and above 500 kpc a linear fit is used
. the actual size at which the break between functions occurs is determined by where the fits are closest in @xmath30 ( not necessarily at 500 kpc ) .
a cut at a maximum size of 3 mpc is also imposed , as sources larger than this are extremely rare ; however , we find that where this cut is placed has very little effect on our results . the final exponentials are then normalized by the largest value of @xmath30 .
the fit parameters to this `` wide '' model are given in the middle portion of table [ isizemodels ] , and the fits are shown as dashed lines in figure [ intrinsicmodels ] . as mentioned above , these fits give an upper limit to the size distributions , and may give an unrealistic number of large sources .
if we take 40 kpc as a split between large and small sources , then indeed these fits result in distributions where 8090% of the sources are considered large .
a more conservative large / small ratio for luminous radio sources is 4050% ( bridle & perley 1984 ) .
therefore , we also determine , purely empirically and using the same functional forms , fit parameters for each sample that provide , on average , this ratio when the real sample numbers are randomly drawn from the distributions .
these `` narrow '' fit parameters are given in the top portion of table [ isizemodels ] , and shown as dotted lines in figure [ intrinsicmodels ] .
finally , the fit parameters for the three samples are averaged together ( separately for the wide and narrow distributions ) .
these mean fits are used in the simulations to define the underlying intrinsic size distribution , and are shown in bold in table [ isizemodels ] . in order to completely separate the process of developing an intrinsic size distribution from the samples we are trying to simulate ,
we develop a third distribution from an independent sample .
mullin et al .
( 2008 ; m08 ) present the properties of a sample of strictly frii type sources ( including rgs and qsos ) .
as these are the types of objects for which this kind of projected size difference test is intended , it is an ideal sample to utilize in building an intrinsic size distribution .
while the other samples above will contain many powerful frii type sources , they may not be restricted to _ only _ these objects .
we apply the same method as in the last subsection to turn the apparent linear sizes presented in m08 into intrinsic sizes , by assigning each object a random viewing angle and repeating the process @xmath29 times to build a distribution .
the result is shown in the bottom panel of figure [ intrinsicmodels ] .
no longer does the distribution have the same functional form as the previous three samples , as there are quite a few more intermediate ( 100500 kpc ) sized sources , as well as quite a few more large ( @xmath31 kpc ) sources .
this is reflective of the restriction to only frii objects , which are expected to be large .
we find that two truncated gaussians accurately describe this distribution , one below @xmath10800 kpc , and another above .
the parameters of these gaussians are given at the bottom of table [ isizemodels ] and shown by dot - dashed lines in figure [ intrinsicmodels ] .
times for each sample , and results in the shaded distributions .
the dashed lines are fits consisting of a power law ( at smaller sizes ) and a straight line ( at larger sizes ) in natural - logarithmic space , whose parameters are given in table [ isizemodels ] .
this is taken as the widest possible version of the intrinsic size distribution .
the dotted line is a narrower distribution , defined such that each sample has @xmath1050% of objects above and below 40 kpc .
bottom panel : the same method applied to a pure frii source sample this is best fit by two gaussian components , shown as dot - dashed lines.[intrinsicmodels],width=321 ] times for each sample , and results in the shaded distributions .
the dashed lines are fits consisting of a power law ( at smaller sizes ) and a straight line ( at larger sizes ) in natural - logarithmic space , whose parameters are given in table [ isizemodels ] .
this is taken as the widest possible version of the intrinsic size distribution .
the dotted line is a narrower distribution , defined such that each sample has @xmath1050% of objects above and below 40 kpc .
bottom panel : the same method applied to a pure frii source sample this is best fit by two gaussian components , shown as dot - dashed lines.[intrinsicmodels],width=321 ] times for each sample , and results in the shaded distributions .
the dashed lines are fits consisting of a power law ( at smaller sizes ) and a straight line ( at larger sizes ) in natural - logarithmic space , whose parameters are given in table [ isizemodels ] .
this is taken as the widest possible version of the intrinsic size distribution .
the dotted line is a narrower distribution , defined such that each sample has @xmath1050% of objects above and below 40 kpc .
bottom panel : the same method applied to a pure frii source sample this is best fit by two gaussian components , shown as dot - dashed lines.[intrinsicmodels],width=321 ] times for each sample , and results in the shaded distributions .
the dashed lines are fits consisting of a power law ( at smaller sizes ) and a straight line ( at larger sizes ) in natural - logarithmic space , whose parameters are given in table [ isizemodels ] .
this is taken as the widest possible version of the intrinsic size distribution .
the dotted line is a narrower distribution , defined such that each sample has @xmath1050% of objects above and below 40 kpc .
bottom panel : the same method applied to a pure frii source sample this is best fit by two gaussian components , shown as dot - dashed lines.[intrinsicmodels],width=321 ] lcccccccc & & + + b89 & 220 & @xmath32 & 17.00 & @xmath330.18 & & @xmath32 & @xmath330.003 & 7.10 + b13 & 305 & @xmath34 & 17.00 & @xmath330.18 & & @xmath34 & @xmath330.003 & 7.00 + s13 & 270 & @xmath35 & 19.00 & @xmath330.15 & & @xmath35 & @xmath330.003 & 9.00 + * mean * & * 270 * & @xmath36 & * 17.67 * & @xmath37 & & @xmath36 & @xmath38 & * 7.70 * + & & + + b89 & 220 & @xmath32 & 12.36 & @xmath330.08 & & @xmath32 & @xmath330.003 & 8.54 + b13 & 305 & @xmath34 & 11.49 & @xmath330.05 & & @xmath34 & @xmath330.003 & 9.31 + s13 & 270 & @xmath35 & 15.06 & @xmath330.08 & & @xmath35 & @xmath330.003 & 10.27 + * mean * & * 273 * & @xmath36 & * 12.97 * & @xmath39 & & @xmath36 & @xmath38 & * 9.38 * + & & + + & & & & + & & a & @xmath40 & @xmath41 & & a & @xmath40 & @xmath41 + + * m08 * & * 820 * & * 1 * & * 0 * & * 397.8 * & & * 0.18 * & * 0 * & * 896.8 *
our general approach is to perform simulated versions of the original b89 , s13 , and b13 tests while varying several parameters . in all cases we use the real redshifts of objects in the samples , which are converted into angular size distances .
we assign each object an intrinsic size by drawing from one of the three ( wide , narrow , or gaussian ) distributions discussed in the previous section , and a random viewing angle drawn from a distribution consistent with geometric projection effects .
these parameters are combined to calculate both the projected size on the sky in kpc and the apparent angular size in arcseconds for each object .
we calculate the ratio of the median rg projected size to median rlqso projected size .
the distributions of projected sizes and angular sizes for the rgs and rlqsos are also compared via a kolmogorov - smirnov ( ks ) test .
this process is repeated @xmath42 times .
the final use of these values will be described in detail in the following subsections .
every set of @xmath42 tests is run drawing from the narrow , wide , and gaussian distributions of intrinsic size separately .
we also run each set using various values , and rlqsos to have @xmath43 for @xmath1 ; 30@xmath44 , 45@xmath44 , and 60@xmath44 .
additionally , we run the simulations assuming no viewing angle restriction on rgs or qsos ( i.e. unification by orientation is false ) .
these results will be labeled throughout as `` no restriction '' or
no @xmath1 .
finally , we run each test using the original sample size parameters ( total number of objects @xmath45 , number of rgs @xmath46 , number of rlqsos @xmath47 ) , as well as resampling them to have an equal number of rgs and rlqsos but keeping the same @xmath45 .
this resampling is conducted by simply selecting objects randomly from each subset .
the @xmath16 distributions in this random sampling are not forced to be matched , however we find that in most cases they are .
a new random sample is drawn for each of the @xmath42 iterations .
we note that , to estimate the value of @xmath1 in the orientation scenario , as b89 did , one must carefully select every rg and rlqso available in a sample without bias but to look at projected size distributions this is less critical , so long as selection biases do not affect the rg and rlqso distributions differently .
we are interested in the latter question here .
so while it may not necessarily be possible ( or reasonable ) to force a sample to have an equal number of objects , we are merely interested in seeing how this ratio can affect the outcome .
all of our results are presented using the linear projected sizes , and not the apparent angular sizes . in most cases ,
the results using these values are quite similar , except using the s13 sample where differences in the apparent sizes are more pronounced when using the angular sizes .
this is due to the redshift matching of the rgs and rlqsos .
while statistically the s13 sample has well - matched distributions of redshifts for both rgs and rlqsos , there is a notable overabundance of low-@xmath16 rgs , which they discuss in their paper .
if we neglect some of these objects , and force the @xmath16 distributions to match better by eye , this difference between using projected and angular sizes is diminished .
most of our results are drawn from the distributions of @xmath49 , which are shown in figure [ meddists ] with the peaks normalized to one in order to help illustrate the subtle change in shape when varying the intrinsic size distribution .
these distributions are built by taking the ratio of the median rg projected size to the median rlqso projected size in each of the @xmath42 iterations .
the histograms are converted into probability density functions ( pdfs ) by normalizing their area to 1 , allowing us to simply integrate within various limits to determine the probability of a particular outcome .
the pdfs normalized by area are not shown explicitly here , as their shape is the same as the histograms in figure [ meddists ] and the reader can get a sense of the probabilities from there .
we also show in figure [ meddists ] the @xmath49 distributions when a single intrinsic value is used as opposed to a range .
note that they have some width due to variation in the viewing angles , but they are quite narrow in comparison . , and intrinsic size distribution .
also shown are the results for a single input intrinsic size , as opposed to a distribution with significant width .
we only show these distributions for the original sample sizes and @xmath50 values , as they are nearly identical for the resampled 1:1 ratio simulations.[meddists],width=321 ] , and intrinsic size distribution . also shown
are the results for a single input intrinsic size , as opposed to a distribution with significant width .
we only show these distributions for the original sample sizes and @xmath50 values , as they are nearly identical for the resampled 1:1 ratio simulations.[meddists],width=321 ] , and intrinsic size distribution . also shown
are the results for a single input intrinsic size , as opposed to a distribution with significant width .
we only show these distributions for the original sample sizes and @xmath50 values , as they are nearly identical for the resampled 1:1 ratio simulations.[meddists],width=321 ] the medians of these @xmath49 distributions provide some insight into the effects of the intrinsic size distributions , as well as some checks that our simulations are performing as expected , and are shown in figure [ medresults ] . in the simplest scenario where all rgs and rlqsos have a single intrinsic size , not a distribution
, one can predict what the most probable median projected size ratio will be . for a given @xmath1
, one can directly find the median viewing angle ( @xmath51 ) to rgs or rlqsos , using the fact that the distribution of possible viewing angles obeys the relationship @xmath52 .
the expected ratio of median sizes is then @xmath53 .
this predicts median projected size ratios of 2.5 , 1.8 , and 1.5 , for @xmath54 , and @xmath55 , respectively .
these `` expected '' sizes are shown in figure [ medresults ] for reference . allowing a wider and wider distribution of possible intrinsic sizes will inflate these ratios , which our simulation predicts .
figure [ medresults ] shows that the median projected size ratios start high and tend toward the predicted value in moving from the gaussian to the wide to the narrow intrinsic size distribution .
indeed , if we input a single intrinsic size for all objects , the expected median projected size ratios are always recovered . in the case where there is no viewing angle dependence ( no @xmath1 ) , we find a median projected size ratio of 1 , as expected on average , rgs and rlqsos appear the same size because they have the same average viewing angle .
we note that sample size has virtually no effect on these median values , and that while there is a dependence on the shape of the intrinsic size distribution , the effects are quite small .
the largest difference is due simply to the presence of a distribution of sizes as opposed to a single value .
ratios for each sample , and the bottom shows the results keeping the same @xmath45 but randomly resampling to an equal number of rgs and rlqsos .
the `` expected '' values indicate what the mean projected size ratio would be in the simplest case where there is a single intrinsic size for every source.[medresults],width=321 ] the minimal effect of the shape of the intrinsic size distribution is also seen in figure [ meddists ] . in all cases , while there are differences between the results for the narrow ( shaded ) , wide ( thick outline ) , and gaussian ( dotted ) histograms , they are small and subtle .
the effects of @xmath1 and the sample size are much more prominent .
increasing @xmath1 tends to narrow the projected size ratio distributions . for a given value of @xmath1
, a larger sample size will increasingly narrow the distributions ( recall that the sample size increases from b89 to b13 to s13 ) .
this has important implications for the use of the median projected size ratio to infer things like a specific value of @xmath1 , which we will explore further in
[ sec : discuss ] .
table [ simresults ] gives the results of simulations for each sample , underlying intrinsic size distribution , and value of @xmath1 ; figures [ prob1.5results]-[ksresults ] show them graphically .
table [ simresults ] is divided into four main sections .
each of those four sections is divided into two the top half shows results keeping the original @xmath45 and @xmath50 values , while the bottom gives results with the original @xmath45 but resampling to a 1:1 @xmath50 ratio .
the top three main sections give the probability of various outcomes based on integrating the pdfs .
we discuss each of the four sections in the table in the following subsections .
the first section of table [ simresults ] shows the probability of finding a value of @xmath49 of 1.5 or greater , and the results are shown graphically in figure [ prob1.5results ] .
the value 1.5 was chosen as a limit because it is the smallest ratio we would expect to find ( see figure [ medresults ] ) , unless the value of @xmath1 is significantly larger than 60 . in the case of no orientation unification ( no @xmath1 ) and a single intrinsic size ,
the probability of seeing a projected size difference is essentially zero .
however , one striking result is that when considering the intrinsic size distribution as well as the orientation , depending on the sample size , there is a significant probability that rgs will appear larger than rlqsos , even if there is no difference in viewing angle to rgs and rlqsos .
this effect has little dependence on the shape of the intrinsic size distribution .
however , for a sample size in the several hundreds , such as s13 , the large number of sources overrides the intrinsic size distribution and the probability of finding a projected size difference in the no @xmath1 case again approaches zero .
this means that for small sample sizes , as in b89 , there is a reasonable probability of finding a projected size difference even if there is no viewing angle difference for rgs and rlqsos larger sample sizes are needed to rule out this possibility .
if the hypothesis of unification by orientation is adopted for general values of @xmath1 , an increase in sample size makes it more likely that a projected size ratio greater than 1.5 will be seen .
however , for an increase in sample size of a factor of nearly 10 ( from b89 to s13 ) , the probability only increases by @xmath1020% .
the effect of the intrinsic size distribution in these cases is not large , but is more significant wider distributions are more likely to show a projected size difference .
this is reasonable , as very different intrinsic sizes are more likely to be randomly selected if the intrinsic size distribution is broad .
applying different projections due to the differing allowed viewing angles to a survey that samples a broad intrinsic size distribution then makes the projected size ratio even larger . a ratio of rgs to rlqsos of @xmath101
also slightly increases the probability of finding a large projected size ratio , but the effect is not large .
the particular value of @xmath1 also affects the probability of finding a projected size difference .
of course , in reality we do not control @xmath1 , it just has some intrinsic value ( if it exists at all ) .
however , understanding this dependence is useful as there are ways of estimating @xmath1 ( e.g. b89 estimate it at @xmath1045 ) .
other estimates may be made in the future , and understanding the dependence shown here can be useful in order to determine the robustness of future results .
if we assume the @xmath57 value from b89 is correct , the results in this analysis suggest that the probability of finding a size ratio that can be interpreted as a significant projected size difference between rgs and rlqsos in the b89 , b13 , and s13 experiments ranges from @xmath1065 to 95% . )
greater than 1.5 for each sample ( indicated by color ) , as a function of several parameters ; the value of @xmath1 ( on the x - axis ) , sample size ( increases from b89 to b13 to s13 ) , and the underlying intrinsic size distribution ( narrow ; filled circles , wide ; open triangles , gaussian ; open squares ) .
the top panel shows results keeping the original @xmath50 ratios for each sample , and the bottom shows the results keeping the same @xmath45 but randomly resampling to an equal number of rgs and rlqsos.[prob1.5results],width=321 ] the second part of table [ simresults ] gives the probabilities of finding a median size ratio larger than the `` expected '' value for each @xmath1 ; 2.5 , 1.8 , and 1.5 for @xmath58 , respectively .
essentially , these are the values we would need to obtain , at a minimum , to argue that the assumptions we put into the model are recovered , i.e. for example , that we could use the median projected size ratio to find the input @xmath1 .
the probability of reaching these values is shown in figure [ probcritresults ] .
the probabilities are roughly constant with @xmath1 ( within @xmath105% ) for a given sample and intrinsic size distribution .
figure [ meddists ] implies why this is the case as @xmath1 decreases , the expected value of the median projected size ratio decreases , and the size ratio distributions shift toward lower values in turn , keeping the probabilities roughly constant .
sample size has an even smaller effect on these probabilities than in the previous case , with a factor of @xmath1010 increase in sample size leading to at most a 10% increase in the probability of detecting at least the expected value .
a ratio of @xmath50 closer to @xmath101 can add an additional 10% to the probability that the expected values , or larger , will be recovered .
the results in this analysis indicate that the probability of finding the median projected size ratios ( or larger ) predicted under unification by orientation in the b89 , b13 , and s13 experiments ranges from @xmath1055 - 70% , again assuming @xmath57 . , except showing the probability of finding @xmath48 above the `` expected '' size ratios for a given @xmath1 in the idealized , single intrinsic size case .
these median size ratios are 2.5 , 1.8 , and 1.5 for @xmath59 , respectively , and 1 for the no @xmath1 case .
the top panel shows results keeping the original @xmath50 ratios for each sample , and the bottom shows the results keeping the same @xmath45 but randomly resampling to an equal number of rgs and rlqsos.[probcritresults],width=321 ] we can also determine the probability that the s13 and b13 result ( that rlqsos appear larger than rgs ) could be recovered if the hypothesis of unification by orientation was correct .
this analysis is shown in the third section of table [ simresults ] and in figure [ fracresults ] , as the probability that a simulation will result in a median projected rg to rlqso size ratio of less than one .
we see again that there is a dependence on the value of @xmath1 , though this becomes less critical as sample sizes increase .
the effect of a wide vs. narrow vs. gaussian intrinsic size distribution is also small here , with a change of only a few percent in probability between them .
resampling the sources to have an equal number ratio also has a minimal effect . in this analysis
we see that it is possible in the standard orientation picture to randomly find that rlqsos appear larger than rgs , _ but it is extremely unlikely , especially for the large sample size of s13 . _ .
panels are the same as figure [ medresults ] .
the top panel shows results keeping the original @xmath50 ratios for each sample , and the bottom shows the results keeping the same @xmath45 but randomly resampling to an equal number of rgs and rlqsos.[fracresults],width=321 ] b89 , b13 and s13 generally present their results in the context of @xmath49 .
however , when illustrating the projected sizes , they are presented as cumulative distributions . whether or not these cumulative distributions are different is quantified by a kolmogorov - smirnov ( ks ) test . while the authors do not explicitly use the ks test in their analysis , presenting the cumulative distribution functions essentially allows the reader to perform a ks - test - by - eye .
therefore , we also perform a ks test on the distributions of projected rg and rlqso sizes in each iteration of our simulations , using a p - value of less than 0.05 to indicate that the projected size distributions appear significantly different .
there are some caveats to the use of this test in assessing significant differences between heterogeneous populations , but , for our highly controlled simulated data , the ks test adequately discriminates different populations .
the results are shown in the bottom section of table [ simresults ] and in figure [ ksresults ] .
the major difference between using a ks test or the median size ratio to identify a projected size difference is that the ks test is much more conservative , in particular for smaller samples . for the b89 sample ,
the probability of finding a significant size difference using the ks test is lower by @xmath1040% for a given @xmath1 compared to the probability of finding a median projected size ratio greater than 1.5 .
for the larger s13 sample , this difference in probability is much less significant .
as expected , larger samples are more likely to be able to constrain differences in projected size , regardless of the shape of the intrinsic size distribution .
sample size plays a much larger role in the resulting probability using the ks test than when using the median projected size ratio distributions , with differences in probability of @xmath1080% from the smallest ( b89 ) to largest ( s13 ) samples .
the ratio of rgs to rlqsos does not play a very large role in this test , though for the s13 sample , which has the largest difference in @xmath46 and @xmath47 , we clearly see that having a ratio closer to @xmath101 increases the chance of seeing a difference in sizes .
it is much less apparent in the other samples , which have a more nearly equal number of rgs and rlqsos .
again , the probability of a ks test resulting in @xmath60 depends on the value of @xmath1 .
this probability is more strongly affected by the underlying intrinsic size distributions than in the previous cases , but still behaves as would be expected .
widening the distribution of intrinsic size is more likely to result in a difference in projected size , as it increases the likelihood that the sizes are different before any projection effects are applied .
assuming the b89 value of @xmath57 , b89 should have observed a significant projected size difference in 2530% of experiments , b13 in 4050% of experiments , and s13 in 9095% of experiments . using the gaussian distribution of intrinsic sizes , these results can increase by up to a further 20% . )
difference in projected size distributions of rgs and rlqsos .
the top panel shows results keeping the original @xmath50 ratios for each sample , and the bottom shows the results keeping the same @xmath45 but randomly resampling to an equal number of rgs and rlqsos.[ksresults],width=321 ] lccccccccccccccccc & & & & + + b89 & 50 & 33 & 17 & 0.24 & 0.83 & 0.68 & 0.56 & & 0.24 & 0.87 & 0.72 & 0.59 & & 0.16 & 0.93 & 0.77 & 0.62 + b13 & 86 & 51 & 35 & 0.17 & 0.90 & 0.73 & 0.57 & & 0.16 & 0.94 & 0.79 & 0.63 & & 0.09 & 0.98 & 0.85 & 0.65 + s13 & 466 & 379 & 87 & 0.04 & 0.98 & 0.84 & 0.60 & & 0.04 & 1.00 & 0.90 & 0.68 & & 0.01 & 1.00 & 0.95 & 0.72 + + + b89 & 50 & 25 & 25 & 0.23 & 0.85 & 0.69 & 0.57 & & 0.22 & 0.88 & 0.72 & 0.60 & & 0.15 & 0.95 & 0.79 & 0.62 + b13 & 86 & 43 & 43 & 0.17 & 0.91 & 0.74 & 0.58 & & 0.16 & 0.94 & 0.78 & 0.63 & & 0.08 & 0.98 & 0.85 & 0.66 + s13 & 466 & 233 & 233 & 0.01 & 1.00 & 0.92 & 0.66 & & 0.01 & 1.00 & 0.97 & 0.79 & & 0.00 & 1.00 & 0.99 & 0.82 + + + & & & & + + b89 & 50 & 33 & 17 & 0.50 & 0.55 & 0.56 & 0.56 & & 0.49 & 0.57 & 0.59 & 0.59 & & 0.50 & 0.58 & 0.62 & 0.62 + b13 & 86 & 51 & 35 & 0.51 & 0.55 & 0.57 & 0.57 & & 0.50 & 0.60 & 0.63 & 0.63 & & 0.50 & 0.61 & 0.65 & 0.65 + s13 & 466 & 379 & 87 & 0.51 & 0.56 & 0.60 & 0.60 & & 0.50 & 0.62 & 0.68 & 0.68 & & 0.50 & 0.64 & 0.71 & 0.72 + + + b89 & 50 & 25 & 25 & 0.51 & 0.57 & 0.60 & 0.60 & & 0.50 & 0.55 & 0.57 & 0.57 & & 0.52 & 0.58 & 0.62 & 0.62 + b13 & 86 & 43 & 43 & 0.50 & 0.60 & 0.63 & 0.63 & & 0.50 & 0.56 & 0.58 & 0.58 & & 0.50 & 0.61 & 0.66 & 0.66 + s13 & 466 & 233 & 233 & 0.51 & 0.71 & 0.78 & 0.79 & & 0.50 & 0.61 & 0.67 & 0.66 & & 0.49 & 0.73 & 0.82 & 0.82 + + + & & & & + + b89 & 50 & 33 & 17 & 0.50 & 0.05 & 0.13 & 0.21 & & 0.50 & 0.02 & 0.09 & 0.15 & & 0.49 & 0.00 & 0.04 & 0.09 + b13 & 86 & 51 & 35 & 0.50 & 0.01 & 0.07 & 0.13 & & 0.50 & 0.01 & 0.04 & 0.09 & & 0.50 & 0.00 & 0.01 & 0.03 + s13 & 466 & 379 & 87 & 0.52 & 0.00 & 0.01 & 0.04 & & 0.52 & 0.00 & 0.00 & 0.01 & & 0.52 & 0.00 & 0.00 & 0.00 + + + b89 & 50 & 25 & 25 & 0.51 & 0.04 & 0.11 & 0.18 & & 0.50 & 0.02 & 0.08 & 0.15 & & 0.50 & 0.01 & 0.03 & 0.08 + b13 & 86 & 43 & 43 & 0.49 & 0.01 & 0.05 & 0.12 & & 0.50 & 0.01 & 0.03 & 0.08 & & 0.50 & 0.00 & 0.01 & 0.03 + s13 & 466 & 233 & 233 & 0.50 & 0.00 & 0.00 & 0.00 & & 0.50 & 0.00 & 0.00 & 0.00 & & 0.50 & 0.00 & 0.00 & 0.00 + + + & & & & + + b89 & 50 & 33 & 17 & 0.05 & 0.43 & 0.24 & 0.16 & & 0.04 & 0.54 & 0.30 & 0.18 & & 0.05 & 0.75 & 0.46 & 0.28 + b13 & 86 & 51 & 35 & 0.05 & 0.71 & 0.43 & 0.26 & & 0.04 & 0.81 & 0.51 & 0.30 & & 0.05 & 0.95 & 0.72 & 0.47 + s13 & 466 & 379 & 87 & 0.05 & 0.99 & 0.88 & 0.64 & & 0.05 & 0.99 & 0.94 & 0.72 & & 0.05 & 1.00 & 1.00 & 0.91 + + + b89 & 50 & 25 & 25 & 0.03 & 0.44 & 0.23 & 0.14 & & 0.04 & 0.54 & 0.28 & 0.17 & & 0.04 & 0.76 & 0.45 & 0.26 + b13 & 86 & 43 & 43 & 0.04 & 0.70 & 0.41 & 0.24 & & 0.04 & 0.78 & 0.49 & 0.28 & & 0.04 & 0.95 & 0.71 & 0.44 + s13 & 466 & 233 & 233 & 0.05 & 1.00 & 0.99 & 0.91 & & 0.06 & 1.00 & 1.00 & 0.95 & & 0.05 & 1.00 & 1.00 & 0.99 there is another way to approach this problem : under the assumption that the orientation model is correct , what sample size do we really need to reliably see a difference in the projected sizes of rgs and rlqsos ? our simulations are equipped to answer this question by randomly sampling the original objects to obtain an arbitrary sample size .
we adjust the number of objects needed in each case ( i.e. for each @xmath1 and for each intrinsic size distribution ) until the probability of a ks test with @xmath61 is @xmath1095% , as well as until the probability of finding a median projected size ratio larger than 1.5 is @xmath1095% .
the results are summarized in table [ idealsample ] and figure [ ideal ] .
the required sample size depends heavily on the value of @xmath1 , which again we can not control in reality .
however , we reiterate that this information can be useful if predictions are made about the true value of @xmath1 . if we again assume that the canonical b89 value is correct , the b89 and b13 samples are too small to reliably find a projected size difference , particularly in a median size ratio analysis .
the s13 sample is large enough assuming the wider or gaussian intrinsic size distributions , and nearly large enough for the narrower distribution if tested with a ks test . for the median projected size comparison , the s13 sample is only large enough with the gaussian intrinsic size distribution .
sample sizes must be much larger to detect a projected size difference at a high probability with the median projected size ratio analysis , compared to with a ks test . in the case of
@xmath62 the required sample sizes are in the several thousands .
this is because the median projected size ratio distribution is centered near 1.5 , and a large sample size is required to sufficiently narrow the distribution such that the high median size tail dominates . for the original parameters of the samples ,
s13 requires significantly more objects than b89 or b13 .
this is due to the large discrepancy in @xmath46 and @xmath47 in the s13 sample .
in fact , if the three samples are simulated to have a 1:1 rg to rlqso ratio , then the total number of objects needed is the same regardless of the other details of the sample .
lcccccccccccc + + b89 & 30 & 180 & 119 & 61 & & 135 & 92 & 48 & & 90 & 59 & 31 + & 45 & 350 & 231 & 119 & & 275 & 182 & 93 & & 180 & 119 & 61 + & 60 & 625 & 413 & 212 & & 510 & 337 & 173 & & 330 & 218 & 112 + + b13 & 30 & 170 & 100 & 70 & & 135 & 80 & 55 & & 85 & 50 & 35 + & 45 & 335 & 198 & 137 & & 275 & 162 & 113 & & 170 & 100 & 70 + & 60 & 600 & 354 & 246 & & 490 & 289 & 201 & & 310 & 183 & 127 + + s13 & 30 & 275 & 223 & 52 & & 200 & 110 & 25 & & 135 & 109 & 26 + & 45 & 525 & 425 & 100 & & 425 & 334 & 81 & & 275 & 223 & 52 + & 60 & 950 & 770 & 180 & & 775 & 628 & 147 & & 475 & 385 & 90 + + + + b89 & 30 & 150 & 99 & 51 & & 110 & 73 & 34 & & 66 & 44 & 22 + & 45 & 676 & 446 & 230 & & 400 & 264 & 136 & & 270 & 158 & 82 + & 60 & @xmath63 & 2640 & 1360 & & 2500 & 1650 & 850 & & 1900 & 1254 & 646 + + b13 & 30 & 126 & 74 & 52 & & 96 & 57 & 39 & & 60 & 35 & 45 + & 45 & 690 & 407 & 283 & & 360 & 212 & 148 & & 240 & 142 & 98 + & 60 & @xmath63 & 2360 & 1640 & & 2000 & 1180 & 820 & & 1800 & 1062 & 738 + + s13 & 30 & 250 & 203 & 47 & & 160 & 130 & 30 & & 90 & 73 & 17 + & 45 & 1200 & 972 & 228 & & 640 & 518 & 122 & & 350 & 284 & 66 + & 60 & @xmath645000 & 3240 & 760 & & 3500 & 2835 & 665 & & 2300 & 1863 & 437 95% of experiments , assuming the simple orientation picture is correct .
the top panel shows the number needed if comparisons are made with a ks test , the bottom panel shows the number needed if looking for a median projected size ratio of 1.5 or greater . for @xmath65 and the narrow intrinsic size distribution ,
the required number of sources is @xmath644000 .
these upper limits are cut off from the figure to better illustrate the other values.[ideal],width=321 ]
in order to test unification by orientation via these kinds of projected size tests , sample selection can clearly play an important role .
it is critical to choose a sample that is unbiased , in particular to orientation . in a radio flux - limited sample ,
if selected at too high a frequency , an obvious bias would be toward more beamed ( pole - on ) sources , which could significantly skew the results , particularly with regard to the rlqsos if they are really seen more face - on ( e.g. urry & padovani 1995 ) .
samples selected over different redshift ranges may also produce conflicting results .
however , we can see in these simulations that the three real samples analyzed behave in exactly the same way with regard to all parameters .
the exception to this is when using the apparent angular size as opposed to the projected linear size , as mentioned in 4.1.1 . using the former
, one must be careful to have extremely well - matched ( nearly one - to - one ) redshifts or the size comparisons can be misleading , even if the distributions of @xmath16 are statistically similar .
s13 do , in fact , use the apparent angular sizes only , and as we find that the median size ratios can be inflated even by slight mis - matching in @xmath16 , their finding that qsos are larger than rgs is questionable . at best , the sizes of the two classes are the same .
other than this , the biggest differences seen are due simply to the size of the samples , and in some smaller ways the ratio @xmath50 .
this seems to indicate that sample selection effects are not a primary reason that the results of b89 and s13/b13 are contradictory , and that the problem really is with the underlying orientation model .
discussions in b89 , s13 , and b13 focus on viewing angle , while somewhat ignoring that there is an underlying intrinsic size distribution that might influence projected sizes . if the median size ratio is used as a proxy for how different the projected sizes are , it is interesting to ask whether the underlying intrinsic size distribution can dilute the role of viewing angle for a given sample size .
it appears that the intrinsic size distribution plays a role , but its actual form is relatively unimportant . assuming a gaussian , wide or narrow distribution of intrinsic size
, the effect on @xmath49 is minimal .
once a distribution is present , the different possible values of @xmath1 affect the results more than the shape of the distribution of intrinsic sizes .
while the shape of the intrinsic size distribution is relatively unimportant to the median projected size tests , the presence of a distribution instead of a single size has important consequences in the interpretation of these kinds of projected size results .
first , it makes it possible to identify a significantly large projected size ratio , even when a viewing angle dependence is not present . in the small sample of b89 , where a difference was indeed found ,
our simulations show that there is a 10 - 20% chance of obtaining this result even if no viewing angle difference between rgs and rlqsos exists .
our results also show that the ratio @xmath49 does not simply scale to a value of @xmath1 , as the intrinsic size distribution makes this ratio vary quite a bit compared to the single intrinsic size case . for example , a median projected size ratio of @xmath102 is not that unlikely to find for almost any sample size and any value of @xmath1 . so using this value to obtain ( or confirm ) @xmath57 ,
as b89 did , is unconvincing .
a complication in these simulations and in performing this kind of test in reality can arise due to the effects of the so - called `` receding torus '' @xcite , if such an effect really exists .
this model suggests that as the quasar accretion disk increases in luminosity , it will sublimate away more of the dust in the torus and cause the amount of re - processed ir radiation to decrease as the covering fraction of the dusty material decreases .
this inverse relationship has been observed in many studies ( e.g. * ? ? ? * ; * ? ? ?
* ; * ? ? ?
* ) , but its explanation assumes that the dust is in a torus - like structure and thus assumes that unification by orientation is ( at least partially ) true . if the `` receding torus '' interpretation is correct , then assuming a single @xmath1 for all sources is not strictly correct , as each object has its own value of @xmath1 that determines whether it is seen as a rg or rlqso .
we could in principle estimate this value for each object in each iteration of our simulations by using an optical luminosity as a proxy for the covering fraction and hence @xmath1 , which would further restrict the possible @xmath66 randomly assigned . as a simple test of how this could affect our results , we re - ran the simulations allowing each object in each iteration to be assigned a value of @xmath66 restricted by a randomly generated value of @xmath1 , uniformly distributed between 20 and 60 @xcite . in all cases ,
the results are consistent with what is expected for the average @xmath1 , in this case 40 , and any conclusions would remain the same .
this however assumes that both the rgs and rlqsos have the same range of @xmath1 values , or equivalently ( in the `` receding torus '' model ) accretion disk luminosities .
therefore we also simulate situations where some selection effect in each sample causes the rlqsos to have more luminous accretion disks than the rgs .
we do this by assigning each rg a random @xmath1 between 20 and 40 , and each rlqso a random @xmath1 from 40 to 60 .
in this situation , one is less likely to observe a projected size difference ; in terms of the ks test , even with the large sample of s13 a difference is observed in only @xmath1075% of experiments , which would weaken our conclusions regarding the orientation model .
thus , we stress that if the `` receding torus '' interpretation is valid , samples built for this projected size test must also be somewhat well - matched in accretion disk luminosity , which may be difficult to constrain for rgs in particular , as by definition we can not see through to the nuclear emission .
our modeling also provides the ideal sample size with which to detect a difference in projected size distributions , if one exists due to orientation effects .
this number can be dauntingly large if @xmath1 is large ( @xmath67 ) and especially for samples with few rlqsos compared to rgs .
the situation is worse for the narrower intrinsic size distribution .
it seems that sample sizes thus far have been too small to identify rgs as much larger than rlqsos at a high probability .
however , if the assumptions we have made here are valid ( and we argue that they are at least reasonable ) , it seems quite clear that the conflicting results between b89 and s13/b13 _ can not easily be reconciled with a paradigm in which different projected sizes of radio sources result solely from orientation .
_ b89 , b13 , and s13 should have seen a median projected size ratio of over 1.5 in @xmath1075 , 80 , and 90% of experiments , respectively .
only b89 observed such a ratio , and as mentioned above , there is a small but significant probability that b89 could find this ratio by chance even if there is no unification by orientation .
if , instead of a ratio of 1.5 , we are looking for a size ratio consistent with a particular value of @xmath1 , the probability that b89 , b13 , or s13 would have observed such a ratio is lower . because the projected size ratio distributions roughly accumulate around the `` expected '' ratios , huge numbers of sources are needed to recover these values at high probability . this could make it easier to reconcile conflicting results . in terms of the ks tests , the b13 sample contains a number of objects for which there is a 50/50 chance of seeing a projected size difference ( or closer to 75% if the gaussian distribution is correct ) .
but it seems that b89 would have to have been somewhat fortunate to see this difference given their sample size it would be apparent in @xmath1025 - 40% of experiments , while s13 would simultaneously have to have been very unlucky to miss it
a size difference should be apparent in @xmath6490% of samples that comprise the number of objects studied in s13 .
the advantage of using the median projected size ratio comparison over a ks test is that it is much less sensitive to the shape of the real intrinsic size distribution ( which is not well constrained ) than the ks test .
the ks test however has an advantage in that it requires a smaller overall sample to identify a projected size difference in a large fraction of experiments .
it is possible that b13 or s13 could have failed to measure projected size differences due to random sampling but finding that rlqsos appear larger than rgs in the popular paradigm , while also still possible , is very unlikely . for the s13 sample this result should only occur in @xmath68% of experiments .
so , it appears that the underlying intrinsic size distribution does not matter enough to question the newest results regarding this problem the s13 and b13 tests have therefore convincingly demonstrated that the simplest schemes that unify rgs and rlqsos by orientation are insufficient .
cast solely in terms of viewing angle , recent results in the literature strongly contradict unification by orientation for radio galaxies and radio - loud quasars , as the prediction that rgs should have larger projected radio - source sizes due to a more edge - on perspective is not confirmed .
however , it is highly unlikely that all sources have the same intrinsic size and that results can be interpreted solely in terms of viewing angle .
we explored the role that the underlying intrinsic size distribution of radio sources plays in this type of analysis , paying particular attention to the question of whether the conflicting results can be reconciled by random sampling of this intrinsic size distribution .
we summarize the results as follows : * whether or not a difference in projected radio - source sizes will be seen in a given sample depends heavily on the critical viewing angle that separates rgs and rlqsos ( if one really exists ) .
while we can not control this value in reality , understanding this dependence can be useful for analyzing the robustness of past and future results . *
sample size is critical for discriminating differences in projected sizes for rgs and rlqsos and must number at the very least in the several hundreds for a robust detection . * the chance of detecting a difference in projected size does depend on , but is not particularly sensitive to , the shape of the underlying intrinsic size distribution , at least for the distributions tested here .
the existence of a distribution as opposed to a single intrinsic size is more important , in general , than the form it takes . *
it appears that sample selection biases are unlikely to be the reason conflicting results have been found , as all three real samples tested here behave in an identical manner for all parameters . * in smaller samples ( @xmath10100 or less ) ,
there is a significant probability that rgs will appear larger than rlqsos , _ even if there is no difference in viewing angle to the two subsets _ , due to random sampling of the intrinsic size distribution . *
it is highly unlikely to find rlqsos appearing larger than rgs in the standard orientation model , even with a wide distribution of intrinsic sizes . *
it is possible , _ but very unlikely _
, under the simplest unification by orientation model , for conflicting results in the literature to be reconciled .
our simulations demonstrate that there is , indeed , a major problem with using the projected size distribution of luminous radio sources to argue that radio galaxies and radio - loud quasars can be unified by orientation .
our analyses suggest that if orientation does play a role in our view of radio galaxies and radio - loud quasars , it can not be the dominant factor based solely on comparison of the apparent sizes of these sources on the sky .
we would like to thank ritaban chatterjee for helping develop the idea for this project , as well as for useful discussions .
mad , jcr , and adm were partially supported by nasa through adap award nnx12ae38 g and epscor award nnx11am18a and by the national science foundation through grant number 1211112 .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
b89 report that rgs appear larger than rlqsos , in contradiction with the results of b13/s13 . as both sets of results rely on measurements of the median of a distribution , and as scientists predominantly work with significances based on _ the mean _ , it is worth reviewing these results in the context of inferring significances from _ the median_. to do this , we follow the method outlined in @xcite . to locate the 2@xmath41 limits in the smallest - to - largest ordered list of projected sizes , we first calculate @xmath69 , where @xmath70 is either @xmath46 or @xmath47 .
the variable @xmath71 ranges from 0 to 1 and has an expected value of 0.5 .
@xmath72 represents the locations ( indices ) of the 2@xmath41 size limits in the ordered list of size measurements , normalized by @xmath70 .
multiplying these normalized indices by @xmath70 and rounding up to the nearest integer allows us to approximate the location in the list of values of the 2@xmath41 upper and lower limits . reading off these values from the list provides the range of median size values for either rgs or rlqsos within the 2@xmath41 errors in each of the above samples .
we see that in all cases there is significant overlap between the median sizes of the subsamples .
thus , when the errors on the median apparent sizes are taken into account , it is clear that _ none of these samples are of sufficient size to determine reliably if rgs really appear larger than rlqsos _ , and claims based simply on the median sizes in the samples are highly suspect .
helmboldt , j.f . ,
taylor , g.b . ,
tremblay , s. , fassnacht , c.d . , walker , r.c . ,
myers , s.t . , sjouwerman , l.o . ,
pearson , t.j . ,
readhead , a.c.s . ,
weintraub , l. , gehrels , n. , romani , r.w . , healey , s. , michelson , p.f .
, blandford , r.d . , & cotter , g. 2007 , , 658 , 203 | unification by orientation is a ubiquitous concept in the study of active galactic nuclei .
a gold standard of the orientation paradigm is the hypothesis that radio galaxies and radio - loud quasars are intrinsically the same , but are observed over different ranges of viewing angles .
historically , strong support for this model was provided by the projected sizes of radio structure in luminous radio galaxies , which were found to be significantly larger than those of quasars , as predicted due to simple geometric projection .
recently , this test of the simplest prediction of orientation - based models has been revisited with larger samples that cover wider ranges of fundamental properties and no clear difference in projected sizes of radio structure is found .
cast solely in terms of viewing angle effects , these results provide convincing evidence that unification of these objects solely through orientation fails . however , it is possible that conflicting results regarding the role orientation plays in our view of radio sources simply result from _ insufficient sampling of their intrinsic size distribution_. we test this possibility using monte - carlo simulations constrained by real sample sizes and properties .
we develop models for the real intrinsic size distribution of radio sources , simulate observations by randomly sampling intrinsic sizes and viewing angles , and analyze how likely each sample is to support or dispute unification by orientation .
_ we find that , while it is possible to reconcile conflicting results purely within a simple , orientation - based framework , it is very unlikely .
_ we analyze the effects that sample size , relative numbers of radio galaxies and quasars , the critical angle that separates the two subclasses , and the shape of the intrinsic size distribution have on this type of test . 0.5 in |
our aim in this paper is to study the _ boundary of hyperbolicity _ of certain families of two dimensional maps .
we say that a family of @xmath2 plane diffeomorphisms is called a _
hnon - like family _ if it can be written in the form @xmath3 where @xmath4 , @xmath5 and @xmath6 is a @xmath0 `` perturbation '' of the standard _
hnon family _
@xcite satisfying @xmath8 in this paper we consider @xmath9 fixed sufficiently small and investigate the dynamics as the parameter @xmath10 is varied . for simplicity we shall therefore omit @xmath11 and @xmath12 from the notation and denote a hnon - like family by @xmath13 . for future reference
we remark that the inverse of @xmath14 is given by an equation of a similar form : @xmath15 where @xmath16 as @xmath17 .
we shall suppose without loss of generality that @xmath18 hnon and hnon - like families have been extensively studied over the last almost 30 years .
one of the earliest rigorous results on the subject is @xcite in which it was shown that the non - wandering set @xmath19 is uniformly hyperbolic for all @xmath20 and all sufficiently large @xmath10 ( depending on @xmath21 ) . on the other hand , for small @xmath22 and @xmath23
there exists positive probability of `` strange attractors '' which contain tangencies between stable and unstable leaves .
this was first proved in @xcite for the hnon family and later generalized in @xcite to hnon - like families , see also @xcite .
these attractors can not be uniformly hyperbolic due to the presence of tangencies but turn out to satisfy weaker _ nonuniform _ hyperbolicity conditions @xcite .
more recently bedford and smillie have described the transition between these two regimes for hnon families by identifying and describing some of the properties of the _ boundary of uniform hyperbolicity _ @xcite . in particular
they show that for small @xmath24 , the nonwandering set is uniformly hyperbolic up until the first parameter @xmath10 at which a tangency occurs between certain stable and unstable manifolds . combining this with the statements contained in @xcite
their results also imply uniform bounds on the lyapunov exponents of all invariant probability measures at the bifurcation parameter @xcite .
their methods rely crucially on previous work @xcite which in turn is based on the polynomial nature of the hnon family , a feature which allows them to consider its complexification and to apply original and highly sophisticated arguments of holomorphic dynamics . in this paper
we develop a new and completely different strategy to the problem , based purely on geometric `` real '' arguments , which have the advantage of applying to general @xmath0 _ hnon - like _ families .
we also obtain the analogous _ uniformity _ results by showing that the hyperbolicity expansion and contraction rates are uniform right up to the point of tangency and that even _ at _ the point of tangency some strong version of nonuniform hyperbolicity continues to hold : all lyapunov exponents of all invariant measures are uniformly bounded away from 0 .
for all @xmath25 and @xmath26 sufficiently small we have the following property .
for every hnon - like family @xmath27 of plane diffeomorphisms there exists a unique @xmath28 such that 1 . for all @xmath29 the nonwandering set
@xmath30 is uniformly hyperbolic ; 2 . for @xmath31
the nonwandering set @xmath32 contains an orbit of tangency but is `` almost uniformly hyperbolic '' in the sense that all lyapunov exponents of all invariant probability measures supported on @xmath33 are uniformly bounded away from 0
. moreover , the bounds on the expansion and contraction rates for all @xmath34 are independent of @xmath10 and of the family .
we remark that this is not the only existing definition of hnon - like in the literature .
one standard approach is to consider `` singular '' perturbations of the limiting one - dimensional map corresponding to the case @xmath35 : @xmath36 this formulation however has some slight technical issues .
for example , one can not assume that @xmath37 is small on all of @xmath38 since that would violate the requirement that @xmath39 be a global diffeomorphism of @xmath1 .
this can be dealt with by restricting our attention to some compact region , say @xmath40 \times [ -2,2 ] $ ] , and supposing only that @xmath41 in this region .
our arguments apply in this case also and yield a more local result on the hyperbolicity of the nonwandering set restricted to @xmath40 \times [ -2,2 ] $ ] .
we recall that a point @xmath42 belongs to the _ nonwandering _ set @xmath43 of @xmath44 if it has the property that for every neighbourhood @xmath45 of @xmath42 there exists some @xmath46 such that @xmath47 .
the nonwandering set is always invariant and closed ( and thus if bounded , also compact ) .
we say that a compact invariant set @xmath33 is _ uniformly hyperbolic _
( with respect to @xmath44 ) if there exists constants @xmath48 and a _ continuous _
decomposition @xmath49 of the tangent bundle such that for every @xmath50 , every non - zero vector @xmath51 and @xmath52 and every @xmath46 we have @xmath53 by standard hyperbolic theory , the stable and unstable subspaces @xmath54 are tangent everywhere to the stable and unstable manifolds .
in particular uniform hyperbolicity is incompatible with the presence of any tangencies in @xmath33 between any stable and any unstable invariant manifolds associated to points of @xmath55 .
a weaker notion of hyperbolicity can be formulated in terms of invariant measures .
for simplicity we restrict our discussion to the two - dimensional setting , as relevant to the situation we consider in this paper .
let @xmath56 be an @xmath44-invariant ergodic probability measure with support in some compact invariant set @xmath33 . by oseledec
s ergodic theorem @xcite there exist constants @xmath57 and a measurable decomposition @xmath49 such that for @xmath56-almost every @xmath58 and every non - zero vector @xmath51 and @xmath52 we have @xmath59 the constants @xmath60 and @xmath61 are called the _
lyapunov exponents _ associated to the measure @xmath56 .
we say that @xmath56 is _ hyperbolic _
@xcite if @xmath62 clearly implies for any @xmath56 .
the converse however is false in general : the measurable decomposition may not extend to a continuous one on all of @xmath33 and the exponential expansion and contraction in implies only a limited version of in which the constants @xmath63 are measurable functions of @xmath64 and not bounded away from 0 .
this definition of hyperbolicity in terms of lyapunov exponents is sometimes called _ nonuniform hyperbolicity _ and is consistent in principle with the existence of tangencies between stable and unstable manifolds . in general
there may be many ergodic invariant probability measures supported in @xmath33 of which some may be hyperbolic and some not . even if they are all hyperbolic the corresponding lyapunov exponents may not be uniformly bounded way from @xmath65 .
the situation in which all lyapunov exponents of all ergodic invariant measures are uniformly bounded away from zero is , in some sense , as `` uniformly hyperbolic '' as one can get while admitting the existence of tangencies .
this situation can indeed occur , for example in the present context of hnon - like maps .
a first example of a set satisfying this property was given in @xcite .
after completing the proof of the theorem 1 we realized that _
much simpler _ versions of our arguments yield an analogous , new and non - trivial , result in the context of one - dimensional maps . we explain and give a precise formulation of this result .
we consider first the quadratic family @xmath66 we choose this particular parametrization for convenience and consistency with our two dimensional results , but any choice of smooth family of unimodal or even multimodal maps with negative schwarzian derivative would work in exactly the same way .
it is well known that for @xmath67 the nonwandering set @xmath30 is uniformly expanding although we emphasize here that this depends crucially on the negative schwarzian derivative property .
the negative schwarzian property is not robust with respect to @xmath0 perturbations and standard methods do not therefore yield this statement for such perturbations .
there exists a constant @xmath68 such that if a family @xmath69 of @xmath70 one - dimensional maps satisfies @xmath71 then there exists a unique parameter value @xmath28 such that 1 . for all @xmath72 the non - wandering set
@xmath33 is uniformly hyperbolic ; 2 . for @xmath31
the lyapunov exponents of all ergodic invariant probability measures are all positive and uniformly bounded away from 0
. moreover the rates of expansion and the bound on the lyapunov exponents are uniform , independent of the family and of the parameter .
the proof of this result is exactly the same as that of theorem 1 but hugely simpler as all more geometrical arguments concerning curvature etc become essentially trivial .
we emphasize that the uniform expansivity of @xmath30 for a particular parameter value @xmath67 is of course robust under sufficiently small perturbations of @xmath39 , by standard hyperbolic theory .
however this approach _ requires the size of the perturbation to depend on the parameter _
@xmath10 and in particular to shrink to zero as @xmath10 tends to @xmath73 . the crucial point of our approach , both in this one - dimensional setting , as in the two - dimensional setting is that the size of the perturbation _ does not _
depend on the parameter .
we have divided our argument into three main sections . in section [ proofnonwan ]
we analyze the geometric structure of stable and unstable manifolds of the two fixed points and define the parameter @xmath28 as the first parameter for which a tangency occurs between some compact parts of these manifolds .
we also identify a region @xmath74 which we show contains the non - wandering set .
in section [ sectionhyp ] we define a `` critical neighbourhood '' @xmath75 outside of which our maps are uniformly hyperbolic by simple perturbation arguments
. however @xmath76 does contain points of @xmath33 and thus we can not ignore this region . to control the hyperbolicity in @xmath75 we introduce the notions of hyperbolic coordinates and critical points which form the key technical core of our approach .
finally , in section [ sectionhypest ] we apply these techniques to prove the required hyperbolicity properties .
in this section we define the parameter @xmath77 as in the statement of our main theorem , and show that for @xmath78 the nonwandering set is contained in the closure if the unstable manifold of a hyperbolic fixed point restricted to a certain compact region of @xmath79 .
we define the bifurcation parameter @xmath28 below as the first parameter for which there is a tangency between certain compact parts of the stable and unstable manifolds of the fixed points .
this does not immediately imply that it is a first parameter of tangency though this will follow from our proof of the fact that the nonwandering set is uniformly hyperbolic for all @xmath29 . for the endomorphisms
@xmath81 with @xmath82 , there are two fixed points , @xmath83 both hyperbolic .
for the special parameter value @xmath84 , to simplify the notation below , we write @xmath85 since @xmath86 and @xmath87 are repelling in the horizontal direction , their stable sets are simply their preimages : @xmath88 in particular these sets contain the following curves @xmath89 and @xmath90 and @xmath91.,scaledwidth=30.0% ] the first preimage of @xmath86 is a parabola with a minimum at @xmath92 and intersecting the @xmath64-axis at @xmath93 , and having slope equal to @xmath94 at the point @xmath95 and the second is a parabola with a minimum at @xmath96 .
similarly we can compute @xmath97 which is a parabola with a minimum at @xmath98 , intersecting the @xmath64-axis at @xmath99 and having slope equal to @xmath73 at the point @xmath100 , and @xmath101 which is a parabola with a minimum at @xmath102 .
the unstable manifolds going @xmath103 and @xmath104 can be defined and computed in a similar way and are easily seen to be horizontal .
consider first the _
hnon family _
@xmath105 for @xmath106 , @xmath107 is a diffeomorphism .
the hyperbolicity of the fixed points implies that there exists a neighbourhood of the set @xmath108 corresponding to pairs of parameters for which there is an _ analytic continuation _ @xmath109 as hyperbolic fixed points of @xmath107 . considering @xmath110 small , we also have that the analytic continuations @xmath111 and @xmath112 are also well defined and hyperbolic . for simplicity we shall often just refer to these two points as @xmath113 leaving implicit their dependence on @xmath44 .
explicit formulas for @xmath109 can be easily derived from the equation @xmath114 but these would not be particular useful . instead we just observe that the fixed points must lie on the line @xmath115 and so in particular this means that for @xmath116 and @xmath117 the vertical coordinates of @xmath118 and @xmath119 are negative and positive respectively , and the converse for @xmath120 .
clearly the same holds for @xmath121 and @xmath122 if @xmath12 is sufficiently small .
moreover , the determinant of @xmath123 is given by @xmath124 in particular the determinant is constant and negative if @xmath21 is positive and positive if @xmath21 is negative .
we thus refer to the case @xmath125 as the _ orientation - reversing _ case , and the case @xmath126 as the _ orientation - preserving _ case . ) and the orientation - preserving ( @xmath127 ) case ( dotted curves indicate negative eigenvalues),title="fig:",scaledwidth=45.0% ] ) and the orientation - preserving ( @xmath127 ) case ( dotted curves indicate negative eigenvalues),title="fig:",scaledwidth=45.0% ] recall that the determinant of a matrix is the product of the eigenvalues , and thus in particular , the sign of the determinant has implications for the sign of the eigenvalues which , as we shall see , in turn has implications for the geometry of the stable and unstable manifolds of the fixed points . for @xmath35
the fixed points @xmath128 and @xmath129 have derivatives @xmath130 and @xmath131 respectively , and thus , for @xmath22 and @xmath12 small , the expanding eigenvalues of @xmath132 and @xmath133 are @xmath134 and @xmath135 respectively .
this implies that for @xmath136 , the orientation - reversing case , the contracting eigenvalues of @xmath137 and @xmath132 must be @xmath138 and @xmath139 respectively , while for @xmath127 , the orientation preserving case , they must be @xmath139 and @xmath140 respectively .
the two situations are illustrated in figure [ fixedpoints ] with dashed lines showing the invariant manifolds corresponding to negative eigenvalues . by classical hyperbolic theory ,
compact parts of the stable manifolds depend continuously on the map ( see e.g. @xcite ) .
therefore , for small @xmath21 and small @xmath12 the analytic continuations @xmath113 of the fixed points @xmath129 and @xmath128 have stable and unstable manifolds which are close to those computed above for the limiting case .
elementary calculations show that the actual geometrical relations between these continuations depend on whether we consider the orientation reversing ( @xmath125 ) or the orientation preserving ( @xmath127 ) case , and are as illustrated in figure [ intersection ] .
, width=340 ] we let @xmath141 denote the compact parts of the stable and unstable manifold as shown in figure [ intersection ] and notice that , in particular , since for @xmath35 and @xmath142 the unstable manifold of @xmath143 and @xmath144 extend to the whole of the line , for each @xmath67 and @xmath125 sufficiently small we have that @xmath145 crosses @xmath146 four times , and for each @xmath67 and @xmath127 sufficiently small we have that @xmath147 crosses @xmath148 four times , and also we can ensure that the compact parts defined above and in the figure intersect transversally .
again this continues to hold also for a hnon - like family for sufficiently small @xmath149 .
we are now ready to define the parameter @xmath28 .
we fix @xmath22 .
+ for an _ orientation - reversing _ ( @xmath125 ) hnon - like family @xmath39 , we let @xmath150 for an _ orientation - preserving _
@xmath127 hnon - like family @xmath39 , we let @xmath151 we also define a parameter @xmath152 as the @xmath153 of parameters @xmath10 for which @xmath145 crosses @xmath146 four times ( @xmath125 ) or @xmath154 crosses @xmath155 four times ( @xmath127 ) .
clearly this is a weaker condition and thus @xmath156 .
notice that @xmath28 and @xmath157 converge to @xmath158 as @xmath21 and @xmath159 tend to 0 . in this section
we carry out a detailed geometrical study aimed at showing that the nonwandering set is contained in a relatively restricted region . to prove hyperbolicity
we will then just have to focus our efforts in this region .
for the moment we restrict ourselves to the orientation - reversing case . at the end we shall remark how the orientation - preserving case follows by identical arguments with a few minor changes of notation .
first of all we let @xmath160 denote the closed topological disc bounded by compact pieces of the @xmath161 and @xmath162 as shown in figure [ regiond ] . , width=151 ] the main result of this section is the following [ nonwan ] for all @xmath163 we have @xmath164 \times ( -4b , 4b)\}.\ ] ] in this paper we are interested in parameters @xmath165 , but it is worth observing that if follows from proposition [ nonwan ] that for all @xmath166 , and so in particular for a certain range of parameter values which may contain multiple tangencies the recurrent dynamics is captured to some extent by the dynamics on @xmath161 .
this includes in particular all complex dynamical phenomena associated to the unfolding of the tangency at the parameter @xmath28 ( indeed , this includes the range of parameters considered by benedicks - carleson in @xcite .
we split the proof of proposition [ nonwan ] in several lemmas .
once again we deal first with the case @xmath125 . at the end of the proof
we indicate the necessary modifications in order to deal with the case @xmath167 . we first define a relatively `` large '' region @xmath168 and show that @xmath169 and then show in separate arguments that @xmath170 and @xmath171 , and finally refine our estimate to obtain the statement in the proposition .
let @xmath172 we also define the following 6 ( overlapping ) regions ( see figure [ regionv ] ) : @xmath173 to @xmath174,width=340 ] then @xmath175 we prove the following two statements . [ 1.1 ] @xmath176 .
we show that the orbit of every point @xmath177 , @xmath178 is unbounded in either backward or forward time .
this implies in particular that no such point is nonwandering . for @xmath179 ,
let @xmath180 .
we shall use repeatedly the fact that @xmath116 and @xmath181 . for @xmath182
we have @xmath183 and therefore @xmath184 and @xmath185 , and @xmath186 . repeating the calculation we have @xmath187 and so @xmath188 .
for @xmath189 we have @xmath190 and @xmath191 .
thus @xmath192 and so @xmath188 .
similarly , for @xmath193 we have @xmath194 and @xmath191 .
thus @xmath192 and we argue as above . for @xmath195 we consider backward iterations of @xmath196 .
note that @xmath197
. then @xmath198 and @xmath199 .
thus @xmath200 and @xmath201 .
therefore @xmath202 and so @xmath203 we have @xmath204 .
thus @xmath205 and @xmath206 .
so we have that @xmath200 , and we argue as above . for @xmath207 , we have @xmath208 and @xmath209 . therefore , @xmath210 and again , we argue as above .
@xmath211 .
the arguments used above have implications for the locations of the stable and unstable manifolds of the fixed points .
indeed the stable manifolds of the fixed points can not intersect @xmath212 since all points in this region tend to infinity in forward time , whereas , by definition , points in the stable manifolds tend to the fixed points under forward iteration .
similarly the unstable manifolds of the fixed points can not intersect @xmath213 since all points in this region tend to infinity in backwards time . by definition @xmath74
is bounded by arcs of stable and unstable manifolds of the fixed point as in the figure and therefore @xmath214 .
, width=340 ] @xmath215 to show that @xmath216 we refine the strategy used in the proof of the previous lemma , and show that the orbits of all points outside @xmath217 are unbounded in either backward or forward time . since we have already shown that @xmath218 , we need to consider only points in the region @xmath219 .
we write @xmath220 where the regions @xmath221 are defined as follows .
consider the points @xmath222 and @xmath223 of intersection of @xmath224 and @xmath225 and @xmath226 , @xmath227 of intersection of @xmath225 and @xmath228 as in the figure [ regionsr ] .
we let @xmath229 denote the closed region bounded by the arcs of manifold @xmath230 , @xmath231 and @xmath232 , and the segment @xmath233 .
we let @xmath234 denote the region bounded by the arc of manifolds @xmath235 , @xmath236 and @xmath230 , and the segment @xmath237 .
we let @xmath238 denote the region bounded by the arcs of @xmath239 and @xmath225 between the points @xmath240 and @xmath241 , as in figure [ regionsr ] .
we let @xmath242 .
we also define @xmath243 as the region satisfying @xmath244 at the left side of the arc of @xmath225 between the points @xmath245 and @xmath246 , of intersection of that manifold with the lines @xmath247 and @xmath248 , as in the figure [ regionsrnew ] ( b ) .
since @xmath250 is small , we have that all the points @xmath251 satisfy @xmath252 .
notice that , for the unperturbed hnon map @xmath253 , we have that any piece of curve @xmath254 with slope less than @xmath255 contained in the region where @xmath256 is mapped to another curve with slope less than @xmath255 . indeed , letting @xmath257 denote a tangent vector to @xmath254 with @xmath258 , we have @xmath259 whose slope is @xmath260 , provided @xmath250 is small and @xmath261 is close to 2 . for future reference ,
notice that , if @xmath262 , we also have that the norm of @xmath263 is uniformly expanded .
so , since @xmath264 is close to @xmath107 in the @xmath265 topology , we can assume that @xmath264 also has this property in @xmath229 .
now call @xmath266 the successive images of the segment @xmath233 intersected to @xmath229
. since they can not intersect each other , and @xmath233 has a point of the stable manifold of @xmath267 , the curves @xmath266 determine a system of `` fundamental domains '' in @xmath229 : they cross @xmath229 from one stable boundary to the other , and they converge to the arc of unstable manifold @xmath231 . call @xmath268 the region of @xmath229 between @xmath269 and @xmath270 , @xmath271 , and notice that @xmath272 . we also have that @xmath273 falls outside @xmath274 .
that implies that @xmath275 does not intersect @xmath276 , and any point which has an iterate in @xmath277 is not in @xmath278 .
we show that @xmath281 .
indeed , the unstable eigenvalue of @xmath132 is positive and therefore @xmath282 must remain on the same side of @xmath283 as @xmath284 .
also , since @xmath285 $ ] we have that @xmath282 does not intersect any of @xmath286 . to @xmath287,width=340 ] we now show that @xmath289 .
again , we use the fact that @xmath285 $ ] . then
, since one of the components of the boundary of @xmath290 is an arc of stable manifold of @xmath291 containing the fixed point @xmath291 , and the unstable eigenvalue of @xmath291 is positive , we conclude that the image of @xmath292 is contained @xmath293 . we can assume , if @xmath250 is small , that all the points @xmath294 in @xmath295 satisfy @xmath296 ( notice that , for @xmath297 , we have @xmath298 . take @xmath299 a point in @xmath300 , and connect @xmath299 to the boundary of @xmath295 through a horizontal line inside @xmath301 , determining a point @xmath302 . again , by the proximity of @xmath303 and @xmath107 , and the fact that vectors with slope smaller than @xmath304 in @xmath305 are sent by @xmath306 in vectors with slope smaller than @xmath304 , and uniformly expanded , we have that the horizontal distance between @xmath307 and @xmath308 is uniformly expanded .
applying @xmath303 repeatedly , as long as the image is inside @xmath309 , we have that the horizontal distance between the successive images of @xmath299 and @xmath225 increases exponentially
. then the forward images of @xmath299 leave @xmath274 for some positive time .
notice that @xmath311 since all the other regions in @xmath274 outside @xmath312 are mapped forward to the region @xmath295 , and so do not contain points of the backward image of @xmath313
. moreover , the unstable boundary of @xmath313 belongs to @xmath239 and approaches @xmath267 as we apply @xmath314 , and the stable boundary can not cross @xmath225 , then @xmath315 does not intersect @xmath312 . since @xmath316 , the points in there that are not in @xmath239
leave @xmath274 for backward iterations . for @xmath167,width=340 ]
@xmath317 .
notice first of all that by the @xmath318-lemma we have @xmath319 .
now suppose by contradiction that there exists @xmath320 with @xmath321
. then there exists @xmath322 and an @xmath322 neighbourhood @xmath323 of @xmath42 with @xmath324 . since @xmath33 is @xmath44-invariant we have @xmath325 for all @xmath326 and therefore @xmath327 for all
notice that the boundary @xmath329 ( where @xmath330 denotes the piece of @xmath331 between @xmath332 and @xmath333 , as in figure [ regionsr ] .
it is enough therefore to show that , for large @xmath334 , the @xmath335 is @xmath322-dense in @xmath336 as this will imply that @xmath337 , contradicting the assumptions .
this follows easily from the fact that @xmath44 is ( strongly ) area contracting , and thus the area of @xmath338 tends to zero as @xmath339 .
in particular we must have that @xmath340 for all @xmath341 sufficiently large .
moreover , the length of the part of the boundary which belongs to @xmath283 also tends to zero .
thus most of the boundary belongs to @xmath161 and thus we must have @xmath342 for all @xmath343 sufficiently large . combining the results of the lemmas stated above we have that @xmath344 .
the statement in the proposition now follows immediately by observing that @xmath345 implies @xmath346 and that @xmath347\times [ -4b , 4b ] $ ] directly from the definition of @xmath44 if @xmath12 is sufficiently small .
finally , in the case @xmath348 , we consider the stable and unstable manifolds of @xmath291 crossing as in figure [ negativeb ] ( the rectangle @xmath274 is exactly the same ) , determining the region @xmath349 in this case .
the proof is entirely analogous considering the points @xmath350 , etc .
, corresponding to the points @xmath351 etc .
the key idea of our whole strategy is the notion of _ dynamically defined critical point _ which relies on the fundamental notion of _ hyperbolic coordinates_. in this section we introduce these notions and develop the main technical ideas which we will use . in section [ prelim ]
we clarify the relations between various constants used in the argument and introduce some preliminary geometric constructions . in section [ hypcoord ]
we discuss the definition and basic theory of hyperbolic coordinates . in section [ curvature ]
we introduce the idea of _ admissible curves _ and prove certain estimates concerning the images of admissible curves .
finally , in section [ hypcrit ] we introduce the notion of dynamically defined critical point and prove that such critical points always exist in images of certain admissible curves .
we now explain the required relations between the different constants used in the proof , and the order in which these constants are chosen .
all constants are positive .
first of all we fix two constants @xmath352 these will be introduced in sections [ fixptnhbd ] and [ curvature ] below .
even though we specify the actual numerical value of these constants we shall continue to use the constants in the argument because they have some specific geometric meaning and it is useful to keep track of their occurrence throughout the paper .
we then fix a constant @xmath353 large enough so that @xmath354 in section [ critnhbd ] we fix a constant @xmath322 which will then remain unchanged .
finally , at some finite number of places in the argument , we will require @xmath10 to be sufficiently close to 2 and @xmath24 and @xmath12 to be sufficiently small .
we remark that we can suppose that @xmath10 is close to 2 without compromising the fact that hyperbolicity holds for all larger values of @xmath10 . indeed ,
once we fix a neighbourhood of @xmath73 in the @xmath10 parameter space , we can always guarantee uniform hyperbolicity for values of @xmath355 outside this neighbourhood by taking @xmath24 and @xmath12 sufficiently small , ( depending on this neighbourhood ) .
recall first of all that the map @xmath356 has two fixed points @xmath128 and @xmath357 with @xmath358 .
for @xmath359 we let @xmath360 be the open ball of radius @xmath361 centred at @xmath129 and @xmath362 be the component of @xmath363 not intersecting @xmath364 , see figure [ q ] .
the expanding eigenvalue at the point @xmath129 is equal to 4 and so we can suppose that @xmath365 are all small enough so that @xmath366 for all @xmath367
. then , for @xmath368 , let @xmath369 notice that @xmath370 is just the component of @xmath371 containing @xmath372 . and
@xmath373,width=340 ] since @xmath374 is a neighbourhood of @xmath133 for every @xmath343 , @xmath375 , where @xmath376 denotes the connected component containing @xmath133 of @xmath377 , has two components : we let @xmath378 notice that a piece of @xmath283 forms the boundary between @xmath379 and @xmath380 .
we mention for future reference a simple estimate which we shall use below .
[ distance1 ] @xmath381 for all @xmath382 @xmath383 implies , by definition , @xmath384 . for points @xmath42 close to @xmath385
this also means @xmath386 since such points come very close to the fixed point @xmath137 and escape the @xmath361 neighbourhood of @xmath133 along the direction of @xmath387 .
thus , using the fact that the norm of the derivative @xmath388 in @xmath74 is uniformly bounded above by 5 we obtain the result . for @xmath389
we define a _ critical neighbourhood _ @xmath390 notice that we can take @xmath322 sufficiently small so that @xmath391 and @xmath392 from now on we consider @xmath322 fixed .
we also let @xmath393 for @xmath10 sufficiently close to @xmath73 and @xmath24 and @xmath12 sufficiently small we have uniform hyperbolicity outside @xmath394 .
we state this fact more formally in the following [ lemmaunifhyp ] for every @xmath395 and @xmath396 sufficiently small , there exists a constant @xmath397 such that for all @xmath398 and points @xmath42 with @xmath399 , and vector @xmath400 with slope @xmath401 we have @xmath402 @xmath403 if ,
_ moreover _ , @xmath404 then we have @xmath405 this is a standard result ( see for example @xcite or @xcite ) and so we omit the details .
we just mention that it follows from the fact that the limiting one - dimensional map @xmath406 satisfies uniform expansivity estimates outside an arbitrary critical neighbourhood @xmath75 ( with constant @xmath407 arbitrarily close to @xmath408 but constant @xmath409 depending on @xmath322 and arbitrarily small for @xmath410 small ) , see e.g. @xcite . considering this one - dimensional map as embedded in the space of two - dimensional maps and using the fact that uniform hyperbolicity is an open condition we obtain the statement in the lemma for @xmath411 sufficiently small .
the notion of hyperbolic coordinates is inspired by some constructions in @xcite , developed in @xcite and formalized in @xcite as an alternative framework with which to approach the classical theory of invariant manifolds .
here we review the basic definitions and theory to the extent to which they will be required for our purposes .
we recall the notion of hyperbolic coordinates and give the basic definitions and properties in the general context of @xmath0 diffeomorphisms of a riemannian surface @xmath412 .
for @xmath413 we let @xmath414 denote the maximum expansion and the maximum contraction respectively of @xmath415 .
then we think of the quantity @xmath416 as the _ hyperbolicity _ of @xmath415 .
notice that @xmath417 always .
the condition @xmath418 implies that the linear map @xmath419 maps the unit circle @xmath420 to an ellipse @xmath421 with well defined major and minor axes .
the unit vectors @xmath422 which are mapped to the minor and major axis respectively of the ellipse , and are thus the _ most contracted _ and _ most expanded _
vectors respectively , are given analytically as solutions to the differential equation @xmath423 which can be solved to give the explicit formula @xmath424 } { ( { \ensuremath{\partial_{x}f_{1}}}^{k})^2+({\ensuremath{\partial_{x}f_{2}}}^{k})^2 - ( { \ensuremath{\partial_{y}f_{1}}}^{k})^2 -({\ensuremath{\partial_{y}f_{2}}}^{k})^2}.\ ] ] here @xmath425 are the two coordinate functions of @xmath44 . notice that @xmath426 and @xmath427 are always _ orthogonal _ and _ do not _ in general correspond to the stable and unstable eigenspaces of @xmath419 .
we define the _ hyperbolic coordinates of order _
@xmath428 at the point @xmath42 as the orthogonal coordinates @xmath429 given at @xmath42 by the most contracted and most expanded directions for @xmath430 .
if @xmath44 is @xmath0 and @xmath431 then hyperbolic coordinates are defined in some neighbourhood of @xmath42 and define two orthogonal @xmath432 vector fields .
in particular they are locally integrable and thus give rise to two orthogonal foliations .
we let @xmath433 denote the _
stable foliation of order _
@xmath428 formed by the integral curves of the vector field @xmath434 and @xmath435 denote the _ unstable foliation of order _
@xmath428 formed by the integral curves of the vector field @xmath436 .
a crucial property of hyperbolic coordinates and finite order stable and unstable foliations is that , under very mild assumptions , they _
converge _ in quite a strong sense as @xmath437 .
we formulate a version of this property here in our specific context .
[ c2close ] for every @xmath438 , hyperbolic coordinates @xmath439 and stable and unstable foliations @xmath433 and @xmath435 are defined in @xmath440 moreover 1 . the angle between each stable direction @xmath426 and the slope of @xmath441 .
the curvature of each stable leaf , are both @xmath442 . also , the @xmath0 distance between leafs of @xmath433 and leaves @xmath443 is @xmath444 .
analogous convergence results are formulated and proved in great generality in @xcite under weak ( subexponential ) growth of the derivative . here
we shall need only some very particular cases of these estimates and therefore we first describe the specific setting in which they will be applied here .
the main ingredient for the proof is that fact that by our choice of @xmath361 and assuming that @xmath445 , @xmath446 and @xmath12 are small enough we have that @xmath447 is small for all @xmath448 and thus in particular @xmath449 it follows immediately that hyperbolic coordinates , and their associated foliations , of order @xmath428 are well defined in @xmath450 .
points in @xmath451 are then re - injected into @xmath452 and these hyperbolicity estimates can no longer be guaranteed , a priori , for all time .
points in @xmath453 however are outside @xmath74 and therefore , by the arguments of section [ proofnonwan ] , eventually escape towards infinity .
in particular the required hyperbolicity conditions can be guaranteed to hold for all positive iterates .
this implies that hyperbolic coordinates of order @xmath428 are well defined in @xmath454 as in the statement of the proposition .
the statements about the direction of the stable directions , the curvature of the leaves and the @xmath0 distance between stable leaves of different orders , all follow directly from @xcite*main theorem .
these calculations are purely technical and do not add to our geometrical understanding of this situation , we therefore omit the details and refer the reader to that paper . recall that the curvature @xmath455 of a parametrized curve @xmath456 is given by @xmath457 the equivalence between the two formulas
is given by the formula @xmath458 .
for @xmath459 , we say that a @xmath0 curve @xmath460 is _ admissible _ if @xmath461 and @xmath462 for all @xmath463 .
we remark that both the curvature and the slope of tangent vectors of a curve are independent of the parametrization , and thus so is the definition of admissibility .
we shall want to compare the curvature at a point of a curve and at the corresponding point of its image .
so , we suppose @xmath464 is a parametrized @xmath0 curve and @xmath465 . for simplicity we shall often omit the parameter @xmath463 and simply write @xmath388 to denote the derivative at the point @xmath464 .
[ smallcurv ] let @xmath466 be a sequence of @xmath0 curves with @xmath467 .
suppose that for some @xmath463 , @xmath343 is a `` hyperbolic time '' in the sense that @xmath468 for all @xmath469 .
then for @xmath470 sufficiently small , @xmath471 implies @xmath472 [ admisstoadmiss ] if @xmath473 is admissible , then @xmath474 is also admissible this follows from proposition [ smallcurv ] and the hyperbolicity outside @xmath75 .
condition implies that the slope of each tangent vector to @xmath475 is @xmath476 and condition together with lemma [ smallcurv ] gives the curvature @xmath476 . to prove proposition [ smallcurv ] we first prove a general curvature estimate .
we fix some bounded neighbourhood @xmath477 of @xmath168 and , as above , we suppose @xmath466 is a sequence of @xmath0 ( not necessarily admissible ) curves with @xmath467 , all contained in @xmath478 .
[ curv ] there exists @xmath479 independent of @xmath480 such that for all @xmath481 we have @xmath482 we use the formula @xmath483 for the curvature .
we have @xmath484 and @xmath485 therefore @xmath486 is given by @xmath487 where @xmath488 and @xmath489 we shall estimate the two terms of separately .
these will yield the two terms in the statement of the proposition .
for the second term we have @xmath490 is the area of the parallelogram defined by the two vectors @xmath491 and @xmath492 , and @xmath493 is the are of the parallelogram defined by the two vectors @xmath494 and @xmath495 which of course just the image of the first parallelogram under @xmath388 .
the second equality just follows immediately from the definition of @xmath496 .
so it just remains to show that the value of @xmath497 is bounded above by some multiple of @xmath11 and @xmath12 .
indeed , writing @xmath498 we have , by the `` row - linearity '' of the determinant , @xmath499 using @xmath500 , @xmath501 , @xmath502 , @xmath503 and @xmath504 this gives @xmath505 where in the last step we have used the fact that @xmath10 is close to @xmath506 . substituting this above gives the required bound for the second term of .
to bound the first term we write @xmath507 then the norm of this expression is bounded above by @xmath508 all the terms contain a factor @xmath494 ; each of the terms @xmath509 , see , contains a bounded constant multiplied by the factor @xmath12 ; the term @xmath510 , see , is of the order of @xmath511 but here it is multiplied by either @xmath512 or @xmath513 each one of which contains a term which is bounded by @xmath12 .
therefore , there exists a constant @xmath514 such that @xmath515 applying lemma [ curv ] recursively we get @xmath516 using the expansivity assumption and @xmath517 small , this gives @xmath518 the next proposition makes precise the notion of a _
critical point of order _ @xmath519 .
we recall that @xmath254 is a @xmath0 _ admissible curve _ if all its tangent vectors have slope @xmath476 and it has curvature @xmath520 .
we say that @xmath254 is a _ long admissible curve _ if it is an admissible curve which crosses the entire length of @xmath75 .
[ tang ] let @xmath521 be a long admissible curve .
then there exists a unique point @xmath522 such that @xmath523 has a ( quadratic ) tangency at @xmath524 with the stable foliation @xmath525 , for any @xmath526 .
moreover there exists a constant @xmath527 , independent of @xmath517 , such that @xmath528 .
in particular , the sequence @xmath529 is cauchy .
we call @xmath530 and @xmath531 respectively a _ critical point _ and _ critical value _ of order @xmath428 , associated to the long admissible curve @xmath254 .
we remark that critical values @xmath531 of finite order are not guaranteed to be outside @xmath74 , however we shall show below that their limit points as @xmath532 , i.e. the `` real '' critical points always fall strictly outside @xmath74 for @xmath533 . given a parametrized curve @xmath534 and its image @xmath535 we denote by @xmath536 the curvature of @xmath537 at the point @xmath538 and by @xmath539 the curvature of @xmath540 at the point @xmath541 .
[ poscurv ] let @xmath538 be an admissible curve and let @xmath542 .
suppose that for some @xmath543 we have @xmath544 and @xmath545 .
then @xmath546 .
lemma [ poscurv ] essentially says that if the tangent direction of the image of an admissible curve at a certain point is roughly vertical ( or at least contained in the `` vertical '' cone between the positive and the negative diagonals ) then the curvature at this point is strictly bounded away from 0 .
this does not apply to admissible curves outside @xmath75 since we have shown above ( corollary [ admisstoadmiss ] ) that images of such curves are still admissible and therefore their tangent directions are roughly horizontal .
we will instead apply it below to the images of admissible curves inside @xmath75 as a way of pinpointing the location of _
folds_. first recall that the curvature @xmath547 is independent of the choice of parametrization and also the condition @xmath545 is independent of the parametrization since @xmath548 is just the slope of the tangent vector .
therefore we choose the parametrization @xmath549 for simplicity we also omit the subscript @xmath550 from the coordinate functions of @xmath540 and just write @xmath551 . from the definition of @xmath44 we have @xmath552^{2 } , d^{2}\varphi_{2}(\gamma_{0}(t))[\dot\gamma_{0}(t)]^{2 } ) \end{aligned}\ ] ] choosing @xmath12 sufficiently small , for example so that @xmath553 this implies @xmath554 we can now compute the curvature @xmath547 .
first notice that the condition @xmath545 implies in particular @xmath555 dividing numerator and denominator by @xmath556 , using the condition @xmath545 and we get @xmath557 finally , from the formulas for @xmath558 and @xmath559 and the fact that @xmath560 by the admissibility of @xmath537 , we get @xmath561 as long as @xmath12 is sufficiently small .
the existence of a tangency between @xmath475 and the stable foliation @xmath433 follows by the simple geometric observation that the image of a long admissible curve necessarily `` changes direction '' between one endpoint and the other .
thus , by a simple intermediate value argument it follows that there is some point of tangency . now , proposition [ c2close ] says that the leaves of the stable foliations @xmath562 are close to the piece of stable manifold @xmath563 and thus have slope close to @xmath73 , and that their curvature is small .
in particular the point of tangency must occur at some point at which the tangent direction to @xmath475 is close to 2 and therefore proposition [ tang ] shows that at this point of tangency @xmath475 has strictly positive curvature .
this implies that this tangency is quadratic as well as unique .
this is the final and main section of the paper .
we apply the notion of hyperbolic coordinates and dynamical defined critical points to prove theorem 1 . in section
[ shadowing ] we combine the hyperbolic coordinates and the curvature estimates to show that all components of the unstable manifold @xmath161 in @xmath75 are almost horizontal curves with small curvature . in particular they all have well - defined critical points . in section [ hypret ]
we take advantage of the structure of critical points on such components to show that points in the critical region @xmath564 recover hyperbolicity after some bounded number of iterations depending only on the parameter @xmath10 . in section [ unifhyp ]
we then extend these estimates to uniform expansion estimates on all of @xmath161 with a hyperbolicity constant @xmath565 depending only on the parameter . in section [ unifhypom ]
we then show how to extend this hyperbolicity to the closure of @xmath161 and thus to the whole nonwandering set @xmath33 .
finally , in section [ lyap ] , we consider the bifurcation parameter value @xmath31 and show that all lyapunov exponents are uniformly bounded away from 0 . [ recovering ] for all @xmath78 all components of @xmath567 are long admissible curves .
moreover , for all @xmath568 and any vector @xmath569 tangent to @xmath161 at @xmath42 and @xmath438 such that @xmath570 we have @xmath571 if @xmath572 is a long admissible curve we consider the tangent vectors @xmath573 and their images @xmath574 . by proposition [ tang ]
, @xmath575 is tangent to the stable direction @xmath426 at the point @xmath531 . for this and other nearby points on @xmath254 we can write the tangent vector @xmath576 where @xmath577 is the orthogonal basis given by the most expanded and most contracted direction for @xmath419 and @xmath578 and @xmath579 are the components of @xmath580 in this basis .
notice that the basis itself depends on the point .
proposition [ c2close ] implies that the basis varies _ very slowly _ with the base point , and proposition [ tang ] implies that the tangent vector @xmath575 is varying at _ positive speed _ with respect to this basis .
we omit the calculations which are relatively standard , see for example @xcite .
specifically this implies that the component @xmath581 of the tangent vector @xmath575 at some point @xmath582 is proportional to the distance between @xmath42 and the critical point of order @xmath428 , @xmath530 . in our setting , the constants actually give @xmath583 we can now prove the following [ binding ] suppose @xmath584 is an admissible curve , @xmath585 , @xmath586 and @xmath530 is the critical point of order @xmath428 in @xmath254 .
then for a vector @xmath587 tangent to @xmath254 at @xmath42 and all @xmath588 we have @xmath589 in particular @xmath590 the first equality follows immediately from and . to prove the second we need to find a bound for @xmath591 in terms of @xmath428 . using the quadratic nature of @xmath537 and the proximity to the one - dimensional map @xmath592 with @xmath116
, we obtain @xmath593 to estimate @xmath594 we use the observation that the `` real '' critical value @xmath595 on @xmath537 , i.e. the point of tangency between @xmath537 and the limiting stable foliation @xmath596 lies necessarily either on @xmath597 ( this is only a possibility if @xmath31 ) or to the right of @xmath283 in @xmath598 .
we write this as @xmath599 . combining this with lemma [ distance1 ] and the rate of convergence of critical points of finite order @xmath600 as mentioned in proposition [ tang ] and taking @xmath21 sufficiently small
, we get @xmath601 substituting this into and using the fact that we can assume @xmath602 as well as the definition of @xmath353 in and of @xmath318 in , we have @xmath603 returning to the proof of the proposition , to obtain the statement about admissibility , notice first of all that combining lemma [ binding ] with lemma [ smallcurv ] we immediately obtain the statement that if @xmath604 is admissible and @xmath428 is the first time that @xmath605 , then @xmath606 is admissible .
now , by choosing @xmath24 and @xmath12 small we can guarantee that @xmath607 is a long admissible curve .
moreover , every piece of @xmath608 is the image of some curve in @xmath609 and is therefore admissible .
proposition [ recovering ] gives a pointwise recovery time for the hyperbolicity of points in the critical region , based on their position .
the following proposition gives a key uniformity estimate in the phase space for each parameter @xmath29 .
[ n ] for all @xmath29 there exists a constant @xmath611 such that for @xmath612 , and @xmath569 a tangent vector to @xmath161 at @xmath613 , there exists @xmath614 such that @xmath615 is almost horizontal and @xmath616 we remark that the constant @xmath617 is _ not _ uniformly bounded in @xmath618 and in particular _ does not _ apply to @xmath31 .
however it gives us a _
uniformity statement _ in @xmath42 which will implies , as we shall see below , uniform hyperbolicity for each given parameter value @xmath29 .
for the proof we need to extend the definition of admissibility naturally to curves which are only differentiable of class @xmath619 ( lipschitz continuous derivative ) .
[ unique critical ] for every @xmath72 , every @xmath628 lies on a @xmath619 admissible curve @xmath254 which is the limit of @xmath0 admissible curves in @xmath161 and @xmath254 contains a unique critical point @xmath629 with @xmath630 .
we show first of all that every point @xmath631 lies on a @xmath619 admissible curve which is the limit of @xmath0 admissible curves in @xmath161 .
let @xmath631 and let @xmath632 be a sequence in with @xmath633 . by proposition [ recovering ] each @xmath634 belongs to a long admissible curve @xmath635 . we can write these as functions @xmath636 with @xmath637 $ ] and suppose that converge pointwise to @xmath638 . since @xmath639 is compact and @xmath640 are bounded and equicontinuous sequences we have that @xmath254 is @xmath432 and @xmath641 in the @xmath432 topology . to see that @xmath642 is lipschitz , let @xmath643 and observe that each @xmath644 is a lipschitz function with uniformly bounded lipschitz constant @xmath645 .
then we have @xmath646 and hence @xmath646 .
we now show that any such curve @xmath254 contains a unique critical point .
we show first that it must contain at most one , and then argue that it must contain at least one .
let @xmath647 be the angle between the vectors @xmath648 and @xmath649 .
since the image of each admissible curve is quadratic with respect to @xmath596 we have that @xmath647 has a strictly non - zero derivative having at most one root corresponding to a point of tangency between @xmath650 and @xmath596 .
since @xmath651 in the @xmath652 topology , we have that @xmath653 also has strictly non - zero derivative having at most one root also corresponding to a point of tangency between @xmath475 and @xmath596 . to see that such a point exists , observe that if @xmath29 then the graph of @xmath654 crosses the boundary of @xmath655 twice and @xmath656 is outside @xmath74 where the foliation @xmath596 is well defined and the extreme points of @xmath656 both lie on a piece of @xmath283 which is a leaf of the foliation @xmath657 .
this implies that there exists a point outside the interior of @xmath74 where @xmath475 is tangent to @xmath657 .
lemma [ unique critical ] allows us to define a canonical set @xmath658 of _ critical points _ as the union of all critical points @xmath629 for every @xmath619 which are @xmath432 limits of long admissible curves of @xmath659 . in the next lemma
we show that this set is bounded away from the set of nonwandering points .
notice first of all that @xmath663 and thus in particular is bounded .
let @xmath664 be a sequence converging to some point @xmath665 .
we need to show that @xmath666 . since each @xmath667 is the limit of long admissible curves , we can consider sequences @xmath668 for each @xmath428 . using lemma [ unique critical ] and the fact that @xmath669 converges pointwise to @xmath254 ,
we conclude that this convergence is in fact @xmath432 .
since @xmath670 we have that @xmath671 and this implies that @xmath665 is a critical point as required .
we have therefore shown that the critical set @xmath672 is compact . since @xmath33 is also compact , it is sufficient to show that @xmath673 to imply that they are at some positive distance apart .
disjointness follows from the observation that the image of a critical point is always outside @xmath74 , while @xmath33 is an invariant set contained in @xmath674 . by lemma [ crit ] and the uniform approximation of the critical set @xmath675 by the finite order critical sets @xmath676
, there exists @xmath662 sufficiently large so that the following two conditions hold ( using also @xmath677 : @xmath678 now consider @xmath679 and let @xmath46 be such that @xmath680 .
recall @xmath681 and therefore such an @xmath343 is always well defined except for those points which map exactly to the the curve @xmath682 which forms the boundary between @xmath683 and @xmath684 . for these points
we let @xmath685 .
then we let @xmath686 if @xmath687 the statement follows from proposition [ recovering ] .
otherwise our choice of @xmath662 in gives @xmath688 the first inequality follows from lemma [ binding ] , the second one follows from @xmath689 , the third one follows from the first part of , and the last one follows from the second part of .
the following proposition is is essentially a corollary of proposition [ n ] .
however we state it separately as it gives an explicit construction of the constant @xmath565 of hyperbolicity for each @xmath29 . before stating the result we define this constant .
let @xmath694 and @xmath695 denote the maximum possible contraction and the maximum possible expansion exhibited by any vector @xmath696 for any point @xmath697 in at most @xmath662 iterations .
letting @xmath409 denote the constant of hyperbolicity as in on page , we then let @xmath698 let @xmath699 and let @xmath587 tangent to @xmath161 at @xmath42 . since we do not assume anything about the location of @xmath42 the vector @xmath587 may or may not be almost horizontal .
we distinguish these two possibilities .
let @xmath702 be the sequence of returns of the iterates of @xmath42 to @xmath75 ( with @xmath703 if @xmath704 and @xmath705 otherwise ) .
then for each @xmath706 we have an integer @xmath707 given by proposition [ recovering ]
. then we can write @xmath708 where @xmath709 is the number of iterates during which the point remains outside @xmath75 . from proposition [ recovering ] properties and ,
the images of the vector at these iterates remains horizontal and we have @xmath710 for all @xmath711 , in particular for @xmath712 . if @xmath713 , applying to the remaining iterates gives @xmath714 as required . if @xmath715 we have expansion for the first @xmath716 iterates which gives @xmath717 .
there follow @xmath718 iterates ( since @xmath719 ) during which we can bound the contraction coarsely by the @xmath662th power of the maximum contraction in the region @xmath74 which gives @xmath720 we show first of all that some preimage of @xmath42 lies in @xmath724 .
indeed , @xmath725 implies that @xmath726 as @xmath727 and therefore that @xmath728 is almost horizontal for sufficiently large @xmath343 since the local unstable manifold of @xmath132 is admissible . by the invariance of the unstable conefield
outside @xmath75 images of @xmath729 remain almost horizontal unless some return to @xmath724 takes place .
now let @xmath730 be the smallest integer such that @xmath731 .
then @xmath732 is almost horizontal since every component of @xmath733 in @xmath75 is almost horizontal . from proposition [ n ]
it follows that @xmath734 is almost horizontal and therefore it follows necessarily that @xmath735 .
otherwise @xmath736 will be almost horizontal .
returning to the proof of the proposition , we can now argue as in the previous case to obtain exponential growth starting from time @xmath737 : @xmath738 where @xmath739 moreover @xmath740 substituting this back into completes the proof .
we have obtained uniform expansion estimates for vectors tangent to @xmath161 . in this section
we show that these estimates can be extended to @xmath33 .
this part of the argument uses very little of the specific hnon - like form of the map and therefore we state it in a more abstract and general context .
[ hypprop2 ] let @xmath742 be a @xmath432 diffeomorphism and @xmath33 a compact invariant set with @xmath743 on @xmath33 .
suppose that there exists some invariant submanifold @xmath744 dense in @xmath33 and such that there exist constants @xmath745 such that @xmath746 for all @xmath747 and @xmath569 tangent to @xmath744 .
then @xmath33 is uniformly hyperbolic with hyperbolic constants @xmath748 and @xmath318 .
we shall show that @xmath33 is uniformly hyperbolic by constructing an invariant hyperbolic splitting @xmath750 at every point of @xmath33 and then showing that this splitting is continuous .
we carry out this construction in several steps .
the starting point is the observation that @xmath751 is already given by the tangent direction to @xmath744 for all points @xmath752 .
[ limit ] for any @xmath753 and any sequence @xmath754 with @xmath755 , the sequence @xmath756 converges to a ( unique ) limit direction @xmath757 .
each vector @xmath758 satisfies @xmath759 for all @xmath46 .
suppose @xmath753 and let @xmath754 be a sequence with @xmath760 . consider the sequence of corresponding directions @xmath756 . by compactness ( of the space @xmath761 of possible directions ) there must exist some subsequence @xmath762 such that the corresponding directions @xmath763 converge to some direction which we call @xmath757 .
notice that a priori this direction is not unique since it depends on the choice of subsequence .
we shall show first that the forward expansion and backward contraction estimates hold and then show that this actually implies uniqueness .
let @xmath764 and @xmath765 be a sequence with @xmath766 .
then , for each @xmath767 we have , by the continuity of @xmath768 , @xmath769 by assumption we know that @xmath770 and therefore @xmath771 for any @xmath389 .
therefore @xmath772 and , since this holds for every @xmath343 , we have the required statement as far as the expansion in forward time is concerned . to prove contraction in backward time it is sufficient to prove it for points on @xmath744 and then apply exactly the same approximation argument . for @xmath773
this follows immediately from the uniform expansivity assumption in forward time .
indeed , letting @xmath774 , the expansivity assumption gives @xmath775 which immediately implies @xmath776 .
it remains to show uniqueness of @xmath757 for each @xmath777 .
suppose by contradiction that we could find two sequences @xmath778 and @xmath779 with corresponding directions @xmath780 and @xmath781 converging to two different directions @xmath751 and @xmath782 .
let @xmath783 and @xmath784
. then @xmath785 must be linearly independent and thus every other vector @xmath786 can be written as a linear combination @xmath787 for some @xmath788 . by linearity and
the backward contraction estimates obtained above this implies that @xmath789 as @xmath727 .
since @xmath587 was arbitrary this implies that all vectors are shrinking to zero in backward time .
but this is impossible since we have assumed that @xmath790 and thus @xmath791 at every point @xmath792 there exists a unique tangent space splitting @xmath793 which is invariant by the derivative @xmath388 and which satisfies the standard uniform hyperbolicity expansion and contraction estimates .
lemma [ limit ] gives the expanding direction @xmath751 of the splitting with the required hyperbolic expansion estimates in forward time .
the invariance for points in @xmath744 is automatic ( since tangent directions to @xmath744 are mapped to tangent directions to @xmath744 ) , and the invariance for general points follows immediately from the definition of @xmath794 , the invariance of @xmath780 for @xmath754 , and the continuity of @xmath388 . the stable direction @xmath795 is given immediately by as the limit of the sequence @xmath796 of vectors most contracted by @xmath797 , as discussed in section [ shadowing ] .
this also automatically gives the uniqueness and invariance . to complete the proof of the proposition , we just need to show that the given tangent space splitting is continuous
this follows by standard arguments from the uniqueness proved in corollary [ unique ] . indeed ,
for any @xmath753 and any sequence @xmath798 with @xmath799 , every limit point of the corresponding sequence of splittings @xmath800 must also be a splitting @xmath801 . by approximation arguments identical to those used above
it follows that this splitting must also satisfy the uniform hyperbolic contraction and expansion estimates . therefore . by uniqueness
, it must coincide with the existing splitting @xmath793 .
this completes the proof that @xmath33 is uniformly hyperbolic .
finally it remains to consider the dynamics of @xmath803 .
recall that @xmath28 is defined on page as the first parameter for which a tangency occurs between the compact parts of @xmath283 and @xmath161 , see figure [ tangency ] for the pictures in the two cases @xmath125 and @xmath167 .
we need to show that , for @xmath31 , all lyapunov exponents are uniformly bounded away from 0 .
we show that for each point @xmath804 _ not in the orbit of tangency _
@xmath805 ( it is not necessary to consider the orbit of tangency since this is a countable set without recurrence and can therefore not support any invariant probability measure ) there exists a constant @xmath806 , a vector @xmath807 , and a sequence @xmath808 with @xmath809 such that , for all @xmath810 , @xmath811 this is obviously true if the orbit of @xmath42 never enters @xmath75 in forward time or it enters @xmath75 only a finite number of times .
indeed suppose that there exists some @xmath428 such that @xmath812 for all @xmath813 .
then let @xmath587 be a vector which is mapped to the horizontal vector @xmath814 after @xmath428 iterations .
then by we have @xmath815 for all @xmath46 .
this implies that there exists a constant @xmath816 such that @xmath817 for all @xmath46 .
otherwise there exists an infinite sequence @xmath818 such that @xmath819 and @xmath820 . by lemma [ unique critical ] , @xmath821 lies on either a @xmath0 long admissible curve or a @xmath619 long admissible curve which is the @xmath432 limit of @xmath0 long admissible curves in @xmath161 . since @xmath42 has an infinite number of returns to @xmath394 , this implies in particular that @xmath822 and so @xmath823 and so there exists @xmath824 such that @xmath825 . therefore _
exactly _ the same arguments as in lemmas [ binding ] and [ limit ] show that for a vector @xmath826 tangent to such an admissible curve @xmath254 at @xmath827 we have @xmath828 notice that since the @xmath432 limits of @xmath0 admissible curves are unique , as proved above , we have @xmath829 .
then , by and we have @xmath830 then we can define @xmath831 and we have @xmath832 . where the constant @xmath806 is required simply to compensate for the possible lack of expansion for the first @xmath833 iterates . in particular it can be chosen by considering the maximum possible contraction along the orbit of @xmath42 for the first @xmath833 iterations @xmath834 we have shown therefore that for each @xmath753 @xmath835 this clearly implies the same bound for the limit wherever it exists . in particular any point which is typical for some ergodic invariant probability measure and for which therefore such a limit does exist , will have a positive lyapunov exponent @xmath836 . by dissipativity
this immediately implies also that the other lyapunov exponent is negative and uniformly bounded away from 0 both in the dynamical and in the parameter space . | we consider @xmath0 hnon - like families of diffeomorphisms of @xmath1 and study the boundary of the region of parameter values for which the nonwandering set is uniformly hyperbolic . assuming sufficient dissipativity
, we show that the loss of hyperbolicity is caused by a first homoclinic or heteroclinic tangency and that uniform hyperbolicity estimates hold _ uniformly in the parameter _ up to this bifurcation parameter and even , to some extent , at the bifurcation parameter . |
entanglement is a fundamental physical resource in quantum information and computation @xcite .
once two remote parties , say alice and bob , share a suitable amount of entanglement , they can implement a variety of powerful protocols , including teleportation of quantum states @xcite and quantum gates @xcite , and the distribution of unconditionally secure keys @xcite .
the problem of entanglement distribution is therefore a central topic of investigation in the quantum information community .
unfortunately , this distribution is also fragile : quantum systems inevitably interact with the external environment whose decoherent action typically degrades their entanglement . in realistic implementations where the effect of decoherence is non - negligible , entanglement distribution may become challenging and may need distillation protocols @xcite , where a large number of weakly - entangled states are converted into fewer but more entangled states .
the worst case scenario is when decoherence is so strong as to destroy any input entanglement .
mathematically , this is represented by the concept of entanglement - breaking channel @xcite . in general , a quantum channel @xmath0 is entanglement - breaking when its local action on one part of a bipartite state always results into a separable output state , no matter what the initial state were . in other words ,
given two systems , @xmath1 and @xmath2 , in an arbitrary bipartite state @xmath3 , the output state @xmath4 is always separable , where @xmath5 is the identity channel applied to system @xmath1 and @xmath6 is the entanglement - breaking channel applied to system @xmath2 . despite entanglement - breaking channels having been the subject of an intensive study by the community
, they have only been analyzed under markovian conditions of no memory .
in other words , when the distribution involves two or more systems , these systems are typically assumed to be perturbed in an independent fashion , each of them subject to the same memoryless channel .
for instance , consider the symmetric scheme in fig .
[ scenario ] , where a middle station ( charlie ) has a bipartite system @xmath7 in some entangled state , but its communication lines with two remote parties ( alice and bob ) are affected by entanglement - breaking channels @xmath8 and @xmath6 . and @xmath2 .
the communication lines with alice and bob are affected by two independent entanglement - breaking channels , @xmath8 and @xmath6 . in this environment , no entanglement can be distributed , neither via single transmission ( charlie @xmath9 alice or bob ) , nor via double ( charlie @xmath9 alice and bob).,scaledwidth=65.0% ] under the assumption of memoryless channels , there is clearly no way to distribute entanglement among any of the parties .
suppose that charlie tries to share entanglement with one of the remote parties by sending one of the two systems while keeping the other ( a scenario that we call 1-system transmission or just single transmission ) . for instance , charlie may keep system @xmath1 while transmitting system @xmath2 to bob .
the action of @xmath10 destroys the initial entanglement , so that systems @xmath1 ( kept ) and @xmath11 ( transmitted ) are separable .
symmetrically , the action of @xmath12 destroys the entanglement between system @xmath13 ( transmitted ) and system @xmath2 ( kept ) .
now suppose that charlie sends system @xmath1 to alice and system @xmath2 to bob ( a scenario that we call 2-system transmission or double transmission ) .
since the joint action of the two channels is given by the tensor product @xmath14 quantum entanglement must necessarily be destroyed . in other words , since we have 1-system entanglement - breaking , then we must have 2-system entanglement - breaking . in this paper
we discuss how the previous implication is false when we introduce correlations , i.e. , a memory , between the two entanglement - breaking channels . in the presence of a correlated - noise environment
, the double transmission can successfully distribute entanglement despite the single transmission being subject to entanglement - breaking : charlie can transmit entanglement to alice and bob , despite not being able to share any entanglement with them .
surprisingly , this effect can be induced by the presence of separable correlations in the joint environment , so that the broken entanglement is restored in a subtle way and it is not just replaced by other entanglement coming from the environment . in other words , to achieve this effect we do not need to consider arbitrary joint dilations of the two channels , but just two independent unitary dilations which are coupled by a separable environmental state . and
@xmath6 , are dilated in two independent unitaries , @xmath15 and @xmath16 , combining the input systems with two environmental systems , @xmath17 and @xmath18 , described by a product - state @xmath19 ( see text for more details ) .
_ right panel_. symmetric distribution in a correlated - noise environment . here we create correlations between the two entanglement - breaking channels .
the cheapest way is to replace @xmath19 with a separable state @xmath20 . in this scenario , despite entanglement can not be distributed by single transmissions ( charlie can not be entangled with alice or bob ) , still it can be distributed via the double transmission ( so that alice and bob can become entangled ) .
this reactivation of entanglement distribution is induced by the separable correlations injected from the environment . for twirling environments ( in any dimension ) , the separable state @xmath21 is also classical ( zero - discord ) , so that entanglement reactivation is mediated by purely - classical correlations.,scaledwidth=62.0% ] to better clarify these points , consider fig .
[ scenario2 ] . in the left panel , the two memoryless entanglement - breaking channels , @xmath8 and @xmath6 ,
are dilated in two unitaries , @xmath15 and @xmath16 , coupling the input systems , @xmath1 and @xmath2 , with environmental systems , @xmath17 and @xmath18 , described by a product - state @xmath22 .
in other words , the reduced state @xmath23 is transformed by the map@xmath24,\ ] ] and , similarly , @xmath25 is transformed as@xmath26.\ ] ] now , we note that there are many ways in which we can correlate the two channels .
one could consider a completely general joint unitary @xmath27 involving both the input systems , together with ancillary systems @xmath28 however this general approach is not interesting and somehow trivial , since it always includes cases where the input entanglement is broken ( @xmath29 and @xmath30 both separable ) and replaced by fresh entanglement coming from the environment ( so that @xmath31 is entangled ) .
more interestingly , our analysis regards more realistic scenarios where the environment has minimal resources , i.e. , it is weakly correlated , therefore far from being entangled .
this is interesting for its potential applications to realistic non - markovian systems , and theoretically non - trivial , since no swapping of entanglement can occur between systems and environment . in other words
, we are interested in studying the weakest models of correlated - noise environment , close to the memoryless paradigm , which are able to reactivate the entanglement distribution .
driven by such a goal , the simplest and cheapest way to combine two entanglement breaking channels and create a weakly - correlated joint environment , is to keep the previous independent unitaries , @xmath15 and @xmath16 , and replace the environmental product - state @xmath32 with a separable state @xmath20 having the same marginals @xmath33 and @xmath34 , but correlated @xmath35 ( see the right panel of fig . [ scenario2 ] ) . assuming this minimal dilation , we create a joint channel @xmath36 , transforming the input state @xmath3 as@xmath37~,\label{eqdil}%\ ] ] where @xmath38 .
clearly , the joint channel is not memoryless @xmath39 .
also note that the two local channels , @xmath8 and @xmath6 , are still well defined .
for instance , if only system @xmath1 is transmitted , we have@xmath40.\nonumber\end{aligned}\ ] ] in our work we show that , despite @xmath8 and @xmath6 being entanglement - breaking ( @xmath29 and @xmath30 separable for any input @xmath3 ) , the joint channel @xmath0 is able to transmit entanglement , in the sense that @xmath31 can be entangled for a suitable choice of the input entangled state @xmath3 .
this entanglement reactivation is clearly an effect of the injected correlations , coming from the separable state of the environment @xmath21 . as mentioned earlier , the interesting point is thar the distribution is restored as an effect of weak , local - type correlations .
the reactivation of entanglement occurs in different kinds of separable environments , with rather different physical properties .
the simplest environments to construct are the twirling environments , which are based on the twirling operators @xmath41 ( or @xmath42 ) . in this case , a random unitary @xmath43 is directly applied to system @xmath1 and the same unitary ( or its conjugate ) is correspondingly applied to system @xmath2 . while the local action of a random unitary ( @xmath44 or @xmath45 ) is entanglement - breaking , the correlated action ( @xmath41 ) perfectly preserves specific classes of entangled states , belonging to an invariant decoherent - free subspace of the joint correlated environment @xcite .
since these environments are based on local operations and classical communication ( locc ) , they are expected to introduce correlations which are separable and , more precisely , purely - classical .
this feature can be explicitly checked by performing the unitary dilation of these environments according to eq .
( [ eqdil ] ) , and checking that the environmental state @xmath20 is a classical state , therefore , with zero quantum discord . twirling environments can easily be constructed for quantum systems with hilbert spaces of any dimension , both finite ( qudits ) and infinite ( continuous - variable systems @xcite ) . in the case of two qudits
, the randomization of the twirling operators @xmath41 ( or @xmath46 ) is performed over the entire unitary group . in this case , it is easy to identify states which are invariant under @xmath41-twirling ( werner states @xcite ) and @xmath42-twirling ( isotropic states @xcite ) . in the specific case of qubits ( @xmath47 ) , we can restrict the randomization to the basis of the pauli operators , with the qubit werner state being invariant under pauli twirling .
things are less trivial for infinite dimension , in particular , for bosonic modes @xcite . in this case , we restrict the twirling operators to the compact group of orthogonal symplectic transformations , i.e. , phase - space rotations .
the bosonic twirling environment so defined is non - gaussian . here
, it is interesting to see that , only for the @xmath42-twirling , we can identify gaussian states which are invariant and entangled : these are the two mode squeezed vacuum states , also known as einstein - podolsky - rosen ( epr ) states @xcite . more generally , this class can be extended to continuous variable werner states , which are non - gaussian states given by mixing an epr state with a tensor product of thermal states @xcite .
thanks to the existence of an invariant subspace of entangled states , twirling channels allow for the perfect distribution of entanglement in the two - system transmission , despite being one - system entanglement - breaking . however , the presence of such a subspace is not necessary if we relax the condition of perfect transmission and we accept that only a fraction of the input entanglement survives the two - system transmission .
remarkably , this is possible for bosonic systems evolving in a realistic model of correlated - noise gaussian environment , which is the direct generalization of the standard thermal - lossy environment . our model of gaussian environment can be represented by two beam - splitters ( @xmath15 and @xmath16 ) whose environmental ports are subject to a correlated ( separable ) two - mode thermal state @xmath20
. such an environment does not have decoherence - free subspaces , apart from the trivial invariant space given by its own thermal state @xmath20 .
then , for a sufficiently large amount of entanglement at the input , charlie can distribute a distillable amount to alice and bob , despite the single quantum channels charlie - alice ( @xmath8 ) and charlie - bob ( @xmath6 ) being entanglement - breaking . in particular , we find a threshold behavior according to which remote entanglement is restored only after a critical amount of correlations has been injected by the environment .
once this reactivation has taken place , we may ask if the amount of remote entanglement is then increasing in the amount of injected correlations .
unfortunately such a simple monotonicity does not seem to hold in the general case .
in fact , for these gaussian environments , we adopt gaussian discord as a measure of quantum correlations ( upper bound to the actual discord ) .
then , we can easily evaluate the number of entanglement bits ( ebits ) which are remotely distributed in terms of the number of discordant bits ( dbits ) and classical bits ( cbits ) , which are injected by the environment .
for some classes of gaussian environments , remote entanglement is proven to be increasing in the injected correlations ( both classical and quantum ) . however
, this monotonic behavior is not a general property , since we can exhibit two examples of environments , with identical classical correlations and increasing gaussian discord , for which the number of ebits remotely restored is strictly decreasing . because of this non - monotonic behavior , the phenomenon of entanglement restoration in gaussian environments does not seem to be easily or directly related to a specific quantification of the correlations ( classical , quantum or total ) . to clarify relations with previous literature
, we remark that our study is specifically devoted to show the potentialities of separable correlations in breaking the mechanism of entanglement - breaking , so that an entangled state is able to keep some of the initial entanglement during the transmission towards remote parties .
it is clear that this is different from what considered in previous studies @xcite , where separable probes were used to distribute fresh entanglement between two quantum systems that have never interacted before .
this basic difference leads to completely different roles for quantum discord . while in refs .
@xcite discord is identified as the fundamental physical resource needed to distribute fresh entanglement , in our study discord is not necessary in general , since purely - classical correlations can indeed be sufficient to restore broken entanglement ( e.g. , see twirling environments ) . in terms of potential impact
, our work opens new possibilities for entanglement distribution in environments with extreme decoherence , where the presence of noise correlations and memory effects can be exploited to recover from entanglement - breaking .
since entanglement restoration can be achieved by separable correlations , our work also poses fundamental questions on the intimate relations between local and nonlocal correlations and , more generally , between classical and quantum correlations .
it is important to note that memory channels and correlated ( in particular , non - markovian ) environments are present in a wide series of practical scenarios @xcite , including spin chains @xcite , atoms in optical cavities @xcite , quantum dots in photonic crystals @xcite , photonic propagation through linear optical systems @xcite and atmospheric turbulence @xcite .
the general structure of the paper is the following . in sec .
[ sectwirl ] , we study twirling environments showing how entanglement is perfectly restored by classical correlations .
this is proven for quantum systems of any dimension ( qubits , qudits and bosonic systems ) . in sec .
[ secgaussmain ] , we consider bosonic gaussian environments . here
distillable entanglement can be restored by separable correlations , whose composition in terms of classical correlations and gaussian discord is also analyzed .
finally , sec .
[ secconclusion ] is for conclusion and discussion . for the sake of completeness
, we also include an appendix [ appmisce ] , containing simple technical proofs , and an appendix [ appbosonic ] , which is a brief review on bosonic systems and gaussian states .
in this first analysis , we combine two entanglement - breaking channels to form a twirling environment , where entanglement distribution is reactivated by the presence of classical correlations .
in particular , two - system transmission from charlie to alice and bob allows for a perfect transfer of entanglement , thanks to the existence of an invariant subspace of entangled states .
as already said , this is proven for hilbert spaces of any dimension . in sec .
[ qubitsec ] , we start by considering the transmission of qubits in the simplest examples of twirling environments ( correlated pauli environments ) . then , we consider the general case of qudits evolving in multidimensional twirling environments ( sec . [ twirlsection ] )
finally , we consider bosonic systems subject to non - gaussian twirling environments which are based on random phase - space rotations ( sec .
[ bossections ] ) .
the simplest example can be constructed for qubits considering the basis of the four pauli operators @xmath48 , @xmath49 , @xmath50 and @xmath51 @xcite .
given an arbitrary two - qubit state @xmath3 , we consider the correlated pauli channel @xmath52 where @xmath53 and @xmath54 are probabilities .
this channel is clearly simulated by random loccs .
in fact , it is equivalent to extract a random variable @xmath55 , apply the pauli unitary @xmath56 to qubit @xmath57 , communicate @xmath58 and then apply the same unitary @xmath56 to qubit @xmath2 .
it is easy to check that the two - qubit channel of eq .
( [ qubitenv ] ) does not change if we replace the pauli twirling operator @xmath59 with the alternative operator @xmath60 .
it is easy to write the unitary dilation of the correlated pauli channel .
it is sufficient to introduce an environment composed by two systems @xmath17 and @xmath18 , each being a qudit with dimension @xmath61 and orthonormal basis @xmath62 .
then , we can write the dilation of eq .
( [ eqdil ] ) , where the environment is prepared in the correlated state@xmath63 and the unitary interactions are control - pauli unitaries , @xmath64 as evident from eq .
( [ stateenv1 ] ) , the state of the environment is separable , which means that only separable ( i.e. , local - type ) correlations are injected into the travelling qubits . more precisely , since @xmath20 is expressed as a convex combination of orthogonal projectors , it is a purely - classical state , i.e. , a state with zero quantum discord @xcite . as a result , this environment contains correlations which are not only local but also purely classical .
we now show the conditions under which the correlated pauli environment is simultaneously one - qubit entanglement - breaking and two - qubit entanglement - preserving .
we start by considering the transmission of one qubit only , e.g. , qubit @xmath1 .
this is subject to a depolarizing channel @xmath65 it is easy to show that @xmath8 is entanglement - breaking when @xmath66 for every @xmath58 ( see appendix [ depapp ] for a simple proof ) .
a particular choice can be @xmath67 and @xmath68 as for instance used in ref .
@xcite . assuming the condition of one - qubit entanglement - breaking ( @xmath66 ) , charlie is clearly not able to share any entanglement with alice or bob . despite this
, charlie can still distribute entanglement to alice and bob .
this is possible because we can identify a class of entangled states @xmath3 which are invariant under the action of the correlated map ( [ qubitenv ] ) , i.e. , @xmath69 .
this class is simply given by the werner states@xmath70 with parameter @xmath71 , where @xmath72 is the maximally - mixed state and @xmath73 is the maximally - entangled ( singlet ) state .
for @xmath74 , the state @xmath75 is known to be entangled and distillable .
the remarkable property of the werner states is that they are invariant under twirling operators @xmath41 , i.e.,@xmath76 for any unitary @xmath43 . as a result , they are clearly invariant under the action of the correlated pauli environment of eq .
( [ qubitenv ] ) .
thus , if charlie sends werner states with @xmath74 , these entangled states are perfectly distributed to alice and bob ( two - qubit entanglement - preserving ) . in this section
, we consider the general case of qudits , i.e. , quantum systems with hilbert space of arbitrary dimension @xmath77 .
for these systems , we can easily construct classically - correlated environments which are simultaneously 1-qudit entanglement - breaking and 2-qudit entanglement - preserving .
consider two qudits , @xmath1 and @xmath2 , with hilbert spaces of the same dimension @xmath78 and prepared in a bipartite state @xmath3 .
a multidimensional @xmath41 twirling channel is described by the following completely - positive trace - preserving map@xmath79 where the integral is over the entire unitary group @xmath80 acting on the @xmath78-dimensional hilbert space , and @xmath81 is the haar measure .
this channel is clearly realizable by random loccs : a random unitary @xmath43 is drawn and applied to qudit @xmath1 , the choice of @xmath43 is classically communicated to the other qudit @xmath2 , which is then subject to the same unitary .
similarly , we can define a @xmath42 twirling channel , by replacing the twirling operator @xmath41 with the alternative twirling @xmath42 in the definition of eq .
( [ gentwmap ] ) .
compactly , we refer to the @xmath82 twirling channel@xmath83 where @xmath84 or @xmath85 .
this channel can be dilated to introduce an environmental system . since we have an integral in eq .
( [ gentwirl ] ) , this dilation seems to involve the introduction of continuous variable systems .
in fact , the unitary group @xmath80 is labelled by @xmath86 real parameters @xcite , which means that @xmath87 continuous variable systems are needed to embed these parameters and describe the environment .
actually , such a continuous dilation is not necessary , since we can always replace the previous haar integral with a discrete sum over a finite number of suitably - chosen unitaries .
in fact , any twirling channel ( [ gentwirl ] ) can be written as@xmath88 where @xmath89 belongs to the set of unitary 2-design @xmath90 @xcite and @xmath91 or @xmath92 .
the set @xmath93 has a finite number of elements which depends of the dimension of the hilbert space @xmath94 ( to see how the cardinality @xmath95 scales with the dimension @xmath78 , see ref .
the proof of the equivalence between eqs .
( [ gentwirl ] ) and ( [ design1 ] ) can be found in ref .
@xcite for the @xmath41 twirling channel .
see appendix [ twirlaveapp ] for a simple extension of the proof to the other @xmath42 twirling channel .
note that , in the case of qubits ( @xmath47 ) , an example of unitary 2-design is provided by the clifford group @xcite , which is the normalizer of the pauli group and typically employed in quantum error correction @xcite . now using the unitary 2-design
, we can dilate the twirling channel into an environment made by finite - dimensional systems , i.e. , two larger qudits @xmath17 and @xmath18 , each with dimension @xmath95 and orthonormal basis @xmath96 .
in other words , for @xmath97 we can write the dilation of eq .
( [ eqdil ] ) , where the environment is prepared in the uniformly - correlated state@xmath98 and the interactions are the following control - unitaries @xmath99 where @xmath100 and @xmath91 or @xmath92 . as evident from eq .
( [ stateenv2 ] ) , the state of the environment is separable and purely - classical ( zero discord ) . as expected , twirling environments only contain purely - classical correlations .
once we have fully characterized the properties of these environments , we show the conditions under which they are simultaneously one - qudit entanglement - breaking and two - qudit entanglement - preserving .
first of all , we explicitly show that one - qudit transmission is always subject to entanglement - breaking .
in fact , suppose that charlie transmits qudit @xmath1 to alice while keeping qudit @xmath2 .
for any input state @xmath3 , the output state is given by@xmath101 it is clear that the random map @xmath44 implements an entanglement - breaking channel ( a symmetric result holds for the other random map @xmath102 which describes the transmission of qudit @xmath2 ) .
as shown in appendix [ haaravesec ] , the proof of eq .
( [ depola2 ] ) is an application of the identity @xmath103 which is the haar average of a linear operator @xmath104 . in turn , eq .
( [ twirlingid ] ) is a consequence of schur s lemma and the invariance of the haar measure @xcite .
the next step is the analysis of the two - qudit transmission , from charlie to alice and bob . in this case , we look for entangled states which are preserved by the correlated action of the twirling environment .
luckily , we can easily find states which are invariant under the action of the twirling operator @xmath82 , i.e. , @xmath105 thanks to this invariance , such states are fixed points of the @xmath82 twirling channel of eq .
( [ gentwirl ] ) .
in the specific case of @xmath84 , it is well known that the unique solution of eq .
( [ twirlinva ] ) is provided by the multidimensional werner states @xcite . for two qudits @xmath1 and @xmath2 ( with the same dimension @xmath78 ) ,
the werner states are defined by the one - parameter class @xcite@xmath106 where @xmath107 , and @xmath108 is the unitary flip operator @xmath109 .
these states are known to be entangled and distillable for any physical value of @xmath110 .
thus , if charlie has a werner state @xmath111 with suitable @xmath112 ( in the entanglement regime ) , he is able to perfectly transmit this state to alice and bob , who can therefore share and distill entanglement .
such a distribution is possible , despite charlie not being able to share any entanglement with alice or bob due to the entanglement - breaking condition of eq .
( [ depola2 ] ) . coming back to eq .
( [ twirlinva ] ) , one can find a similar solution for @xmath85 .
in fact , as shown in ref .
@xcite , there exist states which are invariant under @xmath42-twirling .
these are called isotropic states , and they are defined by the one - parameter class @xcite@xmath113 involving a convex combination of the maximally mixed state @xmath114 and the maximally entangled state @xmath115 with parameter @xmath116 . in general
, these states are entangled and distillable for any physical value of @xmath117 . thus ,
if charlie has an isotropic state @xmath75 with suitable @xmath118 ( i.e. , in the entanglement regime ) , he is able to perfectly transmit this state to alice and bob , who therefore can share and distill entanglement .
again , this is possible despite the transmission of a single qudit being subject to an entanglement - breaking channel according to eq .
( [ depola2 ] ) .
it is clear that these results can be specialized to the case of qubits ( @xmath119 ) . for qubits , the classes of multidimensional werner states of eq .
( [ multiwerner ] ) and isotropic states of eq .
( [ isostates ] ) coincide up to a local unitary @xcite .
multidimensional werner states reduce exactly to the qubit werner state of eq .
( [ 2qubitwerner ] ) which is @xmath41-invariant , we have @xmath120 which , replaced in eq .
( [ multiwerner ] ) , gives the state of eq .
( [ 2qubitwerner ] ) by setting @xmath121 . ] . on the other hand , isotropic states reduce to eq .
( [ 2qubitwerner ] ) , proviso that the singlet @xmath122 is replaced by the triplet @xmath123 this state is now @xmath42-invariant and known as werner - like state ( this definition is commonly adopted when the singlet is replaced by another maximally entangled state ) . if we restrict the random unitaries to the basis of the pauli operators @xmath124 as done in sec .
[ qubitsec ] , we see that both these states are fixed points of the correlated pauli environment of eq .
( [ qubitenv ] ) , since this map does not change under the replacement @xmath125 . here
we extend the analysis to the case of continuous variable systems , i.e. , quantum systems with infinite dimensional hilbert spaces ( @xmath126 ) .
in particular , we consider the case of two bosonic modes of the electromagnetic field ( see appendix [ appbosonic ] for a brief review of the main concepts on bosonic systems ) . the simplest generalization of the notion of twirling environment involves the use of rotations in the phase space . given a single bosonic mode with number operator @xmath127 ,
the rotation operator is defined as @xmath128 . in the phase space
, the symplectic action of this operator is described by the rotation matrix @xmath129{cc}% \cos\theta & \sin\theta\\ -\sin\theta & \cos\theta \end{array } \right ) ~.\ ] ] in terms of the second - order statistical moments , we have that the covariance matrix ( cm ) @xmath130 of the input mode is transformed by the congruence @xmath131 now , given an input state @xmath3 of two modes , @xmath1 and @xmath2 , we can synchronize two random rotations and define the bosonic twirling channel as@xmath132 where @xmath133 or @xmath134 .
this is a non - gaussian channel , since the twirling operator @xmath135 , despite gaussian , is averaged using a uniform ( i.e. , non - gaussian ) distribution .
it is clearly based on random loccs , since random rotations are locally applied to each bosonic mode and can be correlated by means of classical communication .
it is interesting that the unitary dilation of this channel can be restricted to a finite - dimensional environment .
this is because the single - mode rotation operator @xmath136 belongs to the compact subgroup of the orthogonal symplectic transformations @xmath137 ( these correspond to passive gaussian unitaries , i.e. , unitary transformations preserving both the gaussian statistics and the energy of the state ) .
then , @xmath138 is isomorphic to the unitary group @xmath139 , which is the multiplicative group composed by all complex numbers with unit module , also known as the circle group . because of this isomorphism
, a unitary 2-design @xmath140 can be mapped into a unitary 2-design for @xmath141 @xcite . as a result , we can write @xmath142 for a suitable set of angles @xmath143 , and @xmath144 or @xmath145 . in this form ,
the channel is manifestly non - gaussian .
in particular , it can be represented by introducing two environmental qudits , @xmath17 and @xmath18 , prepared in a correlated state @xmath20 as in eq .
( [ stateenv2 ] ) , and interacting with the two bosonic modes by two control - unitaries as in eq .
( [ cunitaries2 ] ) , where now the target unitaries , @xmath89 and @xmath146 , are rotations in the phase space .
the environmental state is not only separable but also purely - classical ( zero discord ) , which means that only classical correlations are injected by the bosonic twirling environment .
it is easy to show that one - mode transmission is always subject to entanglement - breaking in this kind of environment .
for instance , if mode @xmath1 is transmitted from charlie to alice , then the output state @xmath147 is separable , no matter what the input state @xmath3 is .
in fact , it is easy to prove that a uniformly dephasing channel as in eq .
( [ unidepha ] ) is entanglement - breaking . for completeness
, we give this simple proof in appendix [ dephaseapp ] .
the next step is to find two - mode states which are invariant under bosonic twirling@xmath148 so that they are perfectly transmitted by the bosonic twirling environment@xmath149 let us start this search within the set of zero - mean gaussian states , therefore completely characterized by their cms @xmath150 . finding a solution of eq .
( [ invrot ] ) is therefore equivalent to solving @xmath151 depending on the type of environment , i.e. , @xmath152 ( correlated rotations ) or @xmath153 ( anti - correlated rotations ) , we have two different classes of invariant gaussian states . unfortunately , in the case of correlated rotations , the invariant gaussian states are separable .
in fact , it is easy to check that , for @xmath154 and arbitrary @xmath155 , the unique solution of eq .
( [ invrot2 ] ) is given by the quasi - normal form@xmath156{cc}% \mathbf{a } & \mathbf{c}\\ \mathbf{c}^{t } & \mathbf{b}% \end{array } \right ) = \left ( \begin{array } [ c]{cccc}% \alpha & & \omega & \varphi\\ & \alpha & -\varphi & \omega\\ \omega & -\varphi & \beta & \\ \varphi & \omega & & \beta \end{array } \right ) ~ , \label{cm_appfinite}%\ ] ] with @xmath157 and @xmath158 are real numbers ( which must satisfy a set of bona - fide conditions in order to make the previous matrix a quantum cm @xcite ) .
then , it is easy to check that the previous cm can describe separable gaussian states only .
see appendix [ invgausapp ] to see how to derive eq .
( [ cm_appfinite ] ) and check its separability properties .
thus , despite there are two - mode gaussian states @xmath3 which are invariant under correlated rotations ( @xmath152 ) , these states must be separable .
this negative result can be generalized : no entangled gaussian state is invariant under twirlings of the form @xmath159 , with @xmath43 gaussian unitary ( apart from the trivial case @xmath160 ) . for a simple proof see appendix [ appgtwirl ] . luckily , the scenario is completely different when we consider the other type of environment
. we can easily find entangled gaussian states which are invariant under anti - correlated rotations .
in fact , we can easily check that , for @xmath153 and arbitrary @xmath155 , a possible solution of eq .
( [ invrot2 ] ) is given by the cm@xmath161{cc}% \mu\mathbf{i } & \mu^{\prime}\mathbf{z}\\ \mu^{\prime}\mathbf{z } & \mu\mathbf{i}% \end{array } \right ) ~ , \label{eprcmmu}%\ ] ] where@xmath162 and@xmath163{cc}% 1 & \\ & 1 \end{array } \right ) , ~\mathbf{z}=\left ( \begin{array } [ c]{cc}% 1 & \\ & -1 \end{array } \right ) .\ ] ] this is the cm of a two - mode squeezed vacuum state @xmath111 , also known as epr state @xcite .
thus , in the presence of the anti - correlated environment @xmath164 , despite charlie can not share any entanglement with alice or bob ( one - mode entanglement - breaking ) , he is still able to distribute them entanglement by transmitting epr states perfectly ( two - mode entanglement - preserving ) .
more generally , we can extend the previous analysis and include non - gaussian input states in charlie s hands .
one possible choice is the continuous - variable version of the werner state @xcite@xmath165 \label{cvwerner}%\ ] ] which corresponds to a mix , with probability @xmath166 , of an epr state @xmath167 and a tensor - product of two identical single - mode thermal states , @xmath168 and @xmath169 , with cms @xmath170 .
this state is known to be entangled for @xcite@xmath171 where @xmath172 is the mean number of thermal photons in each mode [ for large @xmath173 , we have @xmath174 , which reminds the threshold valid for qubit werner states ] . since thermal states
are invariant under phase - space rotations , the convex combination of eq .
( [ cvwerner ] ) is invariant under the action of anti - correlated rotations @xmath153 , therefore representing a fixed point of @xmath164 . as a result , any continuous - variable werner state @xmath175 with @xmath176
can be used to perfectly transmit entanglement to alice and bob , despite the environment @xmath164 being one - mode entanglement - breaking .
in the previous sec . [ sectwirl ] , we have studied the distribution of entanglement in twirling channels , considering quantum systems with hilbert spaces of any dimension , both finite and infinite .
we have shown the conditions under which these environments are simultaneously one - system entanglement - breaking and two - system entanglement - preserving .
this peculiar situation is possible due to the existence of invariant subspaces of entangled states ( werner states and isotropic states ) . in this case ,
entanglement preservation is induced by the injection of purely - classical correlations in the travelling systems . in this section ,
we consider a rather different scenario .
we construct a correlated - noise gaussian environment which generalizes the standard model of memoryless thermal - lossy environment to include cross - correlations between two travelling modes .
this is done by using two beam - splitters which mix the travelling modes , @xmath1 and @xmath2 , with two environmental modes , @xmath17 and @xmath18 , prepared in a correlated gaussian state @xmath20 ( in particular , it can be chosen to be separable ) .
each single beam - splitter introduces losses and thermal noise such to realize an entanglement - breaking channel .
for this reason , charlie is not able to share any entanglement with alice or bob . despite this ,
the noise - correlations enable charlie to distribute entanglement to alice and bob .
as we will show , an input epr state @xmath3 with sufficiently - high epr correlations is able to generate an output state which is entangled and even distillable by alice and bob .
this is remarkable considering that : ( i ) for these environments , there is no decoherence - free subspace of entangled states ( which means that entanglement can not be preserved ) ; and ( ii ) entanglement distribution can be reactivated by the injection of separable correlations ( so that lost entanglement is not replaced by other entanglement , i.e. , coming from the environment ) . in detail , this section is structured as follows . in sec .
[ secgauss ] , we characterize our basic model of correlated - noise gaussian environment , identifying the physical conditions under which it is separable or entangled ( sec .
[ sepcmsubsec ] ) , and discussing specific kinds of quadrature correlations , that we call symmetric or asymmetric ( sec . [ subsecspecific ] ) . in sec .
[ secdirect ] , we study the distribution of entanglement . assuming the condition of one - mode entanglement - breaking , we show how entanglement distribution can be restored by the environmental correlations ( in particular , separable correlations ) identifying the regime where remote entanglement is also distillable .
we show that entanglement restoration is a threshold process starting only after a critical amount of correlations is present in the environment . in particular , in secs .
[ subsym ] and [ subasym ] , we specify this study to environments with symmetric and asymmetric correlations , analyzing the process of entanglement restoration in terms of the injected classical and quantum correlations . here
we compute the critical number of correlation bits ( cbits and dbits ) which are needed for entanglement restoration .
once reactivated , the number of ebits remotely restored is increasing in the number of correlation bits which are injected .
this monotonic behavior holds true when we compare environments with the same types of correlations ( symmetric or asymmetric ) but fails to be true in the general case as explicitly discussed in sec .
[ nonmono ] . finally , in sec .
[ subeprevolution ] , we also analyze the evolution of the epr correlations in this peculiar process .
general notions and technical details on bosonic systems and gaussian states can be found in appendix [ appbosonic ] . in this section ,
we introduce and fully characterize our model of correlated - noise gaussian environment , which represents the simplest and most direct generalization of the standard memoryless gaussian environment with losses and thermal noise .
we consider a gaussian decoherence process which affects two bosonic modes , @xmath1 and @xmath2 , in a symmetric way . in order to describe the introduction of losses and thermal noise , we consider two beam splitters ( with the same transmissivity @xmath177 ) which combine modes @xmath1 and @xmath2 with two environmental modes , @xmath17 and @xmath18 , respectively .
these ancillary modes are prepared in a zero - mean gaussian state @xmath20 which is symmetric under @xmath17-@xmath18 permutation . in the standard memoryless model depicted in the left panel of fig .
[ envschemes ] , the environment is in a tensor - product state @xmath178 , meaning that @xmath17 and @xmath18 are fully independent . in particular , @xmath179 is a thermal state with covariance matrix ( cm ) @xmath180 , where the noise variance @xmath181 quantifies the mean number of thermal photons @xmath173 entering the beam splitter .
each interaction is then equivalent to a lossy channel with transmissivity @xmath177 and thermal noise @xmath182 . and thermal noise @xmath182 .
_ right_. correlated - noise gaussian environment , with losses @xmath177 , thermal noise @xmath182 and cross - correlations @xmath183 .
the state of the environment @xmath20 can be separable or entangled.,scaledwidth=65.0% ] in our work , we generalize this gaussian process to include the presence of correlations between the environmental modes , as depicted in the right panel of fig .
[ envschemes ] .
the simplest extension of the model consists of taking the ancillary modes , @xmath17 and @xmath18 , in a zero - mean gaussian state @xmath20 with cm in the symmetric normal form @xmath184{cc}% \omega\mathbf{i } & \mathbf{g}\\ \mathbf{g } & \omega\mathbf{i}% \end{array } \right ) ~ , \label{eve_cmapp}%\ ] ] where @xmath185 is the thermal noise variance associated with each ancilla , and the off - diagonal block@xmath186{cc}% g & \\ & g^{\prime}% \end{array } \right ) \label{gblock}%\ ] ] accounts for the correlations between the ancillas . such a correlated thermal state can be separable or entangled ( explicit conditions will be given below ) , and its quantum discord is always non - zero ( apart when @xmath187 ) .
it is clear that , when we consider the two interactions @xmath188 and @xmath189 separately , the environmental correlations are washed away . in fact , by tracing out @xmath18 , we are left with mode @xmath17 in a thermal state ( @xmath190 ) which is combined with mode @xmath1 via the beam - splitter . in other words , we have again a lossy channel with transmissivity @xmath177 and thermal noise @xmath182 .
the scenario is identical for the other mode @xmath2 when we trace out @xmath17 .
however , when we consider the joint action of the two environmental modes , the correlation block @xmath183 comes into play and the global dynamics of the two travelling modes becomes completely different from the standard memoryless scenario . before studying the system dynamics and the corresponding evolution of entanglement ,
we need to characterize the correlation block @xmath183 more precisely .
in fact , the two correlation parameters , @xmath191 and @xmath192 , can not be completely arbitrary but must satisfy specific physical constraints . these parameters must vary within ranges which make the cm of eq .
( [ eve_cmapp ] ) a bona - fide quantum cm @xcite . given an arbitrary value of the thermal noise @xmath185
, the correlation parameters must satisfy the following three bona - fide conditions @xmath193 ( details of the proof can be found in appendix [ app_sub2modes ] ) .
once the bona - fide conditions for the environment are fully clarified , the next step is the characterization of its separability properties . for this aim , we compute the smallest partially - transposed symplectic ( pts ) eigenvalue @xmath194 associated with the cm of the environment @xmath195 ( for more details on this formalism see appendix [ app_sub2modes ] ) . for gaussian states
, this eigenvalue represents an entanglement monotone , being fully equivalent to the log - negativity @xcite .
after simple algebra , we get@xmath196 provided that the conditions of eq .
( [ cmconstraints ] ) are satisfied , the separability condition @xmath197 is equivalent to@xmath198 the various conditions of bona - fide and separability can be combined together .
an environment of the form ( [ eve_cmapp ] ) with thermal noise @xmath185 is bona - fide and separable when the correlation parameters @xmath191 and @xmath199 satisfy@xmath200 where@xmath201 by contrast , it is bona - fide and entangled when@xmath202 for the gaussian environment , corresponding to thermal noise @xmath203 .
the black area identifies forbidden environments ( correlations are too strong to be compatible with quantum mechanics ) .
white area identifies physical environments , i.e. , the subset of points which satisfy the bona - fide conditions of eq .
( [ cmconstraints ] ) . within this area ,
the inner region labelled by s identifies separable environments [ eq . ( [ sepenv ] ) is satisfied ] while the two outer regions labelled by e identify entangled environments [ eq . ( [ entenv ] ) is satisfied ] . _ right .
_ correlation plan where we display the memoryless environment ( origin @xmath187 ) , sc environments ( bisector @xmath204 ) and ac environments ( bisector @xmath205 ) .
the msc environments are the sc environments with maximal correlations ( extremal points on the bisector @xmath204 ) .
the epr environments are the ac environments with maximal correlations ( extremal points on the bisector @xmath205 ) .
finally , the mac environments are those ac environments which are simultaneously separable and maximally correlated.,scaledwidth=47.0% ] to better clarify the structure of the environment , we provide a numerical example in fig .
[ envcha ] . in the left panel of this figure
, we consider the _ correlation plan _ which is spanned by the two parameters @xmath191 and @xmath192 . for a given value of the thermal noise @xmath182
, we identify the subset of points which satisfy the bona - fide conditions of eq .
( [ cmconstraints ] ) .
this subset corresponds to the white area in the figure . within this area
, we then characterize the regions which correspond to separable environments ( area labelled by s ) and entangled environments ( areas labelled by e ) .
here we specify our model of gaussian environment to particular cases with specific correlation properties . before discussing these cases , note that the memoryless gaussian environment corresponds to the origin of the correlation plan ( @xmath187 ) , as also depicted in the right panel of fig .
[ envcha ] .
in particular , the memoryless environment represents the unique physical solution for @xmath206 ( vacuum noise ) , in correspondence to which all the correlation plan collapses into its origin . in the remainder of this section
we implicitly take @xmath207 , therefore excluding the trivial case of a singular correlation plan .
the first type of specifically - correlated environment is called symmetrically correlated ( sc ) .
this corresponds to taking @xmath204 , which means that positions and momenta of the two modes , @xmath17 and @xmath18 , are correlated exactly in the same way . on the correlation plan , these environments are points on the bisectors of the first and third quadrants ( see the right panel of fig .
[ envcha ] ) .
it is easy to check that the bona - fide conditions of eq .
( [ cmconstraints ] ) simplify to the single inequality@xmath208 with maximal symmetrical correlations ( msc ) achieved at @xmath209 and @xmath210 , which are the two extremal points shown in the figure .
it is important to note that sc environments are always separable , since eq .
( [ sepcon ] ) becomes @xmath211 , which is always satisfied by eq .
( [ qsymmcon ] ) .
then , they are also mixed , with von neumann entropy equal to @xmath212 where is intended to be base @xmath213 , so that the corresponding quantities are measured in bits .
we write @xmath214 for the natural base @xmath215.]@xmath216 given an arbitrary sc environment , we can investigate the nature of its separable correlations in terms of classical correlations ( measured in classical bits or cbits ) and gaussian discord ( measured in discordant bits or dbits ) . using the formulas of ref .
@xcite ( see also appendix [ app_subdiscord ] ) we compute the amount of classical correlations which is here equal to@xmath217 and the amount of gaussian discord @xmath218 which is given by@xmath219 the bits of total correlations are quantified by the quantum mutual information @xmath220 , here equal to@xmath221 it is important to note that gaussian discord @xmath218 is an upper bound to the actual discord , even if they are conjectured to coincide @xcite .
correspondingly , the measure for classical correlations which is here considered @xmath222 represents a lower bound to the actual amount of classical correlations .
as expected , both types of correlations , @xmath223 and @xmath218 , are increasing in @xmath224 at any fixed value of thermal noise @xmath182 . despite increasing ,
gaussian discord must be bounded by @xmath225 , which is compatible with the fact that sc environments are always separable ( @xmath226 is a sufficient condition for entanglement @xcite ) .
finally , one can check @xcite that msc environments have maximal gaussian discord not only between the sc environments but , more generally , among all the separable gaussian environments , represented by the s region in fig .
[ envcha ] . the second type of environment , that we call _ _ _ _ asymmetrically correlated ( ac ) , corresponds to the condition @xmath205 .
this means that positions and momenta have opposite correlations , i.e. , if positions are correlated ( anticorrelated ) , then momenta are anticorrelated ( correlated ) . on the correlation plan ,
these environments are represented by the points lying on the bisectors of the second and fourth quadrants ( see right panel of fig . [ envcha ] ) . in this case , the bona - fide conditions of eq .
( [ cmconstraints ] ) simplify to the inequality@xmath227 for maximal correlations @xmath228 we have an epr state , which is pure and maximally entangled .
depending on the sign of @xmath191 , we have two different epr environments : the positive epr environment ( @xmath229 ) with positions correlated , and the negative epr environment ( @xmath230 ) with positions anticorrelated .
apart from these extremal cases , ac environments are mixed with entropy@xmath231 in general , we note that ac environments can be separable or entangled . by using eq .
( [ sepcon ] ) , we see that they are separable for @xmath232 while they are entangled for stronger correlations @xmath233 we call maximally - separable ac environments ( macs ) those ac environments with maximal correlations but still separable
. they are characterized by the border condition @xmath234 , and they correspond to the two intersection points shown in the right panel of fig .
[ envcha ] . as before
, we can easily quantify the amounts of classical correlations @xmath223 and gaussian discord @xmath218 , both increasing functions in @xmath235 . for ac environments , these quantities
are respectively given by eq .
( [ csym ] ) and@xmath236 the quantum mutual information is now equal to@xmath237 the correlation properties of the ac environments can be easily compared with those of the previous sc environments . for fixed values of @xmath182 and @xmath191 with @xmath238 ( separable environments ) ,
ac and sc environments have identical classical correlations but different gaussian discord , i.e.,@xmath239 this means that they also have different amounts of total correlations @xmath240 as clear from the previous formulas , this is simply a consequence of the fact that ac environments are more entropic than sc environments , i.e. , @xmath241 the difference in entropy directly quantifies the difference in the correlations between these environments@xmath242 however , contrary to the sc environments , the ac environments can also become entangled . when the correlation parameter @xmath191 is chosen in the entanglement regime of eq .
( [ qcorrelatedenv ] ) , classical correlations and gaussian discord become larger ( in particular , @xmath218 may exceed @xmath243 ) .
the optimum is reached at the two epr states whose classical and quantum correlations are identical and maximal in the entire correlation plan .
in particular , we have @xmath244 which is just the entropy of entanglement of the epr state .
let us study the propagation of entanglement in a correlated - noise gaussian environment .
suppose that charlie has an entanglement source described by an epr state @xmath3 with cm @xmath245 given in eq .
( [ eprcmmu ] ) . * * * * we consider the different scenarios depicted in the two panels of fig .
[ twomodeeb ] .
charlie may attempt to distribute entanglement to alice and bob as shown in the left panel of fig .
[ twomodeeb ] , or he may try to share entanglement with one of the remote parties , e.g. , bob , as shown in the right panel of fig .
[ twomodeeb ] ( by symmetry , our derivation is the same if we consider alice in the place of bob ) .
and @xmath2 prepared in an epr state @xmath3 . in order to distribute entanglement to the remote parties ,
charlie transmits the two modes through the correlated gaussian environment , characterized by transmissivity @xmath177 , thermal noise @xmath182 and correlations @xmath183 .
_ right_. charlie aims to share entanglement with a remote party ( e.g. , bob ) .
he then keeps mode @xmath1 while sending mode @xmath2 through the lossy channel @xmath6.,scaledwidth=60.0% ] let us start from the scenario where charlie aims to share entanglement with bob , which means that he keeps mode @xmath1 while sending mode @xmath2 .
the action of the environment is therefore reduced to @xmath246 , where @xmath6 is a lossy channel with transmissivity @xmath177 and thermal noise @xmath182 .
it is easy to check @xcite that the output state @xmath30 , shared by charlie and bob , is gaussian with zero mean and cm@xmath247{cc}% \mu\mathbf{i } & \mu^{\prime}\sqrt{\tau}\mathbf{z}\\ \mu^{\prime}\sqrt{\tau}\mathbf{z } & x\mathbf{i}% \end{array } \right ) , \label{abpcm}%\ ] ] where @xmath248 we can compute closed analytical formulas in the limit of large @xmath112 , i.e. , large input entanglement . in this case , the entanglement of the output state @xmath30 is quantified by the pts eigenvalue@xmath249 the entanglement - breaking condition corresponds to the separability condition @xmath250 , which provides@xmath251 or , equivalently , @xmath252 for the mean number of thermal photons . despite the entanglement - breaking condition of eq .
( [ ebcond ] ) being derived for an epr input , it is indeed valid for any input state . in other words ,
a lossy channel @xmath6 with transmissivity @xmath177 and thermal noise @xmath253 destroys the entanglement of any input state @xmath3 .
as one can check , eq .
( [ ebcond ] ) corresponds exactly to the entanglement - breaking condition for lossy channels derived in ref .
in particular , the threshold condition @xmath254 is sufficient to guarantee one - mode entanglement - breaking , i.e. , the impossibility for charlie to share any entanglement with a remote party .
now remember our main question : suppose that charlie can not share any entanglement with alice or bob ( one - mode entanglement - breaking ) , can he still distribute entanglement to them ?
in other words , suppose that the correlated - noise gaussian environment has transmissivity @xmath177 and thermal noise @xmath255 , so that the individual lossy channels @xmath8 ( charlie@xmath9alice ) and @xmath6 ( charlie@xmath9bob ) are entanglement - breaking .
is it still possible to use the joint channel @xmath256 to distribute entanglement to the remote parties ? in the following , we explicitly reply to this question . in particular
, we will show that entanglement can be distributed by separable environments and can be large enough to be distilled by one - way distillation protocols .
let us derive the general evolution of the two modes @xmath1 and @xmath2 under the action of the joint environment , as depicted in the left panel of fig .
[ twomodeeb ] .
since the input epr state @xmath3 is gaussian and the environmental state @xmath20 is gaussian , the output state @xmath31 is also gaussian .
this state has zero mean and cm given by@xmath257{cc}% x\mathbf{i } & \mathbf{h}\\ \mathbf{h } & x\mathbf{i}% \end{array } \right ) ~,\ ] ] where@xmath258 for large input entanglement @xmath259 , we can easily derive the symplectic spectrum of the output state@xmath260 and its smallest pts eigenvalue@xmath261 the latter eigenvalue determines the remote entanglement distributed to alice and bob , equivalently quantified by the log - negativity@xmath262 which is measured in entanglement bits ( ebits ) .
in particular , @xmath263 represents an upper bound to the mean number of ebits which are distillable per output copy @xmath31 . in the same limit , we can also compute the coherent information @xmath264 between the two remote parties , which provides a lower bound to the mean number of ebits per copy that can be distilled using one - way distillation protocols , i.e. , protocols based on local operations and one - way classical communication ( see appendix [ app_sub2modes ] for more details ) .
this is clearly a lower bound since one - way distillability implies two - way distillability , where both forward and backward communication are generally exploited . in the remainder of the paper , by distillability we implicitly mean the sufficient condition of one - way distillability . for large input entanglement
@xmath259 , we can write @xmath265 which is derived by expanding the coherent information for diverging spectra ( see appendix [ app_sub2modes ] ) . thus , remote entanglement is distributed for @xmath266 , according to eq .
( [ lognegmain ] ) , and it is furthermore distillable for @xmath267 , according to eq .
( [ cohedir ] ) .
now suppose that the gaussian environment has thermal noise @xmath254 , so that it is one - mode entanglement - breaking . replacing in eq .
( [ epsmain ] ) , we can write @xmath268[1+\tau+(1-\tau)g^{\prime}% ] } \nonumber\\ & : = \varepsilon(\tau , g , g^{\prime } ) \label{ebentexp}%\end{aligned}\ ] ] answering our previous question corresponds to finding environmental parameters @xmath177 , @xmath191 and @xmath192 , for which @xmath269 is sufficiently low : for a given value of the transmissivity @xmath177 , we look for regions in the correlation plan @xmath270 where @xmath266 ( remote entanglement is distributed ) and , more strongly , @xmath267 ( remote entanglement is distillable ) .
this is done in fig .
[ dirtot ] for several numerical values of the transmissivity . on the correlation plan @xmath270 for different values of the transmissivity @xmath271 ,
@xmath272 , @xmath273 , and @xmath274 ( from top left to bottom right ) .
corresponding values of the thermal noise are determined by the condition of one - mode entanglement - breaking @xmath255 . in each inset
, the non - black area identifies the set of physical environments , which are divided into separable ( s ) and entangled ( e ) environments by the solid lines .
the gray region is the reactivation area and identifies those environments for which charlie is able to distribute entanglement to alice and bob ( @xmath266 ) . within the reactivation area ,
the points below the dashed curve are those environments for which the distributed entanglement is also distillable ( @xmath267).,scaledwidth=48.0% ] in fig .
[ dirtot ] , the environments identified by the gray reactivation area allow charlie to distribute entanglement to alice and bob ( @xmath266 ) , despite being impossible for him to share any entanglement with the remote parties .
in other words , these environments allow two - mode entanglement - distribution , despite being one - mode entanglement - breaking . furthermore , we can identify sufficiently - correlated environments for which the entanglement remotely distributed is also distillable ( @xmath267 ) . from fig .
[ dirtot ] , it is evident that entanglement restoration is a threshold process occurring only after a critical amount of correlations is injected by the environment . in particular , the critical values of the correlation parameters , @xmath191 and @xmath192 , correspond to the border of the reactivation area . the most remarkable feature in fig . [ dirtot ]
is represented by the presence of separable environments in the reactivation area . in other words , there are separable gaussian environments which contain enough correlations to restore the distribution of entanglement to alice and bob .
furthermore , for sufficiently - high transmissivities and correlations , these environments enable charlie to distribute a distillable amount of entanglement .
for instance , see the bottom right panel of fig .
[ dirtot ] , where separable environments are predominant .
thus , we have just proven that a distillable fraction of the large input entanglement can be remotely restored by the presence of separable correlations in the gaussian environment , which are therefore sufficiently strong to break the mechanism of entanglement - breaking induced by the thermal noise . as we have already said , achieving the simultaneous conditions of one - mode entanglement - breaking and two - mode entanglement - distribution is here surprising , since there is no decoherence - free subspace which can preserve the input entanglement and there is no entanglement coming from the environment which can replace the amount lost in the transmission .
by contrast , the conditions of one - mode entanglement - breaking and two - mode entanglement - distribution can be trivially achieved in entangled environments .
for instance , we may consider two beam - splitters with zero transmissivity , so that charlie s state is completely reflected into the environment and the entangled state of the environment is reflected to alice and bob .
this scenario is certainly one - mode entanglement - breaking , since charlie has no chance of sharing part of his state with alice or bob . at the same time ,
two - mode entanglement - distribution is realized , since the loss of charlie s initial entanglement is just replaced by the injection of entanglement from the environment .
here we specify the previous analysis to sc environments ( @xmath204 ) .
we consider these separable environments at the thermal threshold @xmath254 , so that they are one - mode entanglement - breaking . in the limit of large @xmath112
, we derive the parameter regimes for which remote entanglement can be distributed by the two - mode transmission and , more strongly , result into distillable entanglement . in particular , we provide simple threshold conditions for the correlation parameter @xmath191 .
finally , we study the relations between remote entanglement and the various types of separable correlations ( quantum and classical ) present in these environments , quantifying the critical number of correlation bits which are needed for entanglement restoration . from eq .
( [ ebentexp ] ) , we see that alice and bob s remote entanglement can be quantified by@xmath275 since @xmath276 does not depend on the sign of @xmath191 , its study can be reduced to sc environments with positive @xmath191 .
it is easy to check that @xmath276 takes its optimal ( minimum ) value@xmath277 when the environment has maximal correlations ( msc environment ) , with correlation parameter@xmath278 from eq .
( [ epssc ] ) , we can see that remote entanglement is restored ( @xmath279 ) for transmissivities @xmath280 and values of the correlation parameter @xmath281 , where @xmath282 is the entanglement - restoration threshold@xmath283 more strongly , remote entanglement is distillable ( @xmath284 ) for @xmath285 and @xmath286 , where @xmath287 is the entanglement - distillation threshold@xmath288 see fig .
[ linea0 ] for a pictorial representation . of the sc environment ( which is always separable ) . here
we consider the regime of high transmissivity @xmath285 , where remote entanglement can be distilled for sufficiently high values of the correlation parameter ( @xmath289).,scaledwidth=56.0% ] as discussed in sec .
[ subsecspecific ] , we can easily quantify the separable correlations of the sc environments in terms of classical correlations @xmath223 and gaussian discord @xmath218 .
these environmental quantities can be connected with the amount of entanglement which is restored at the remote stations under the usual conditions one - mode entanglement - breaking ( @xmath255 ) and large input entanglement ( @xmath259 ) . as we can see from the top panel of fig .
[ discordsc1 ] , remote entanglement can be restored ( @xmath279 ) only after a certain amount of gaussian discord @xmath218 and classical correlations @xmath223 .
plot refers to the numerical case of @xmath290 , but the same behavior appears for any other value of transmissivity @xmath280 . in the bottom panel of fig .
[ discordsc1 ] , we have plotted the number of ebits remotely restored ( as quantified by the log - negativity @xmath263 ) versus the environmental gaussian discord @xmath218 ( solid line ) and classical correlations @xmath223 ( dashed line ) .
we can clearly see the threshold behavior of the entanglement , which starts to be restored only after a certain numbers of correlation bits ( dbits and cbits ) are injected by the environment . after these critical values , the number of remote ebits is monotonically increasing .
( dimensionless units ) , gaussian discord @xmath218 ( dbits ) and classical correlations @xmath223 ( cbits ) are plotted versus the correlation parameter @xmath191 of the sc environment ( note that we can restrict our analysis to @xmath291 by symmetry ) .
remote entanglement is generated ( @xmath292 ) for sufficiently high values of @xmath218 and @xmath223 , and it is optimal at the border ( msc environments ) , where @xmath218 and @xmath223 are maximal . in this numerical example , transmission is @xmath290 and thermal noise is @xmath293 @xmath294 ( one - mode entanglement - breaking ) . _
bottom panel_. remote entanglement , as quantified by the log - negativity ( ebits ) , is plotted versus the gaussian discord ( dbits , solid line ) and the classical correlations ( cbits , dashed line ) .
remote entanglement starts to be restored only after a critical number of correlation bits are injected .
then , it is monotonically increasing .
numerical parameters are identical as before ( @xmath290 and @xmath295 @xmath294).,scaledwidth=36.0% ] entanglement restoration is a threshold process at any transmissivity @xmath280 .
this is particularly evident from fig .
[ criticalscpic ] , where we plot the critical numbers of dbits and cbits , after which remote entanglement is established . for completeness
, we also plot the critical number of bits of total correlations ( quantified by the quantum mutual information ) which are needed for the restoration . as we can see , we need less than 2 bits of total correlations to recover entanglement at any @xmath280 . also note that , for extremely low loss ( @xmath296 ) and high thermal noise ( @xmath297 ) , restoration is achieved in correspondence of a negligible amount of gaussian discord and @xmath213 bits of classical correlations . , we plot the critical number of correlation bits after which remote entanglement starts to be restored .
we show the critical number of dbits ( solid ) , cbits ( dashed ) and their total ( dotted ) , corresponding to bits of quantum mutual information.,scaledwidth=43.0% ] here we repeat the previous analysis for the case of ac environments ( @xmath205 ) , under the same conditions of one - mode entanglement - breaking ( @xmath255 ) and large input entanglement ( @xmath259 ) . using eq .
( [ ebentexp ] ) , we see that alice and bob s remote entanglement is quantified by@xmath298 we can have @xmath299 only for positive values of the correlation parameter @xmath191 .
such an asymmetry , which is also evident from fig .
[ dirtot ] , depends on the fact that charlie s input state has epr correlations of the type @xmath300 and @xmath301 .
these input correlations tend to be preserved by ac environments with positive @xmath191 ( whose correlations are of the same type ) while they tend to be destroyed by ac environments with negative @xmath191 ( opposite type ) .
for this reason , in our analysis we only consider positive ac environments ( @xmath291 ) which are separable for @xmath302 , and entangled for@xmath303 the optimal restoration of entanglement is achieved by the positive epr environment @xmath304 , for which we have @xmath305 .
one can easily check that this is the global optimum in the entire correlation plan .
optimal restoration in separable ac environments is achieved by the positive mac environment ( @xmath306 ) for which @xmath307 in general , from eq .
( [ epsepr ] ) , we see that remote entanglement is restored at any transmissivity @xmath177 for values of the correlation parameter @xmath308 , where the entanglement - restoration threshold is here equal to@xmath309 this threshold satisfies @xmath310 , which means that remote entanglement can always be recovered by separable ac environments with sufficiently high correlations , as pictorially shown in fig .
[ linea1 ] . ) . for @xmath311 , there are separable ac environments able to restore remote entanglement.,scaledwidth=53.0% ] thus , at any transmissivity @xmath177 , there always exist separable ac environments able to reactivate the entanglement distribution . in particular , this is true for @xmath312 ( and therefore @xmath313 ) . in fact , by expanding at the leading order in @xmath177 , we get@xmath314 which means that we can always find a correlation parameter @xmath315 such that @xmath316 .
remarkably , in conditions of extreme losses , where entanglement is broken by a negligible amount of thermal noise , entanglement distribution can be reactivated by a vanishing amount of separable correlations ( @xmath317 ) . proceeding with our analysis of the ac environments
, we see that entanglement distillation is possible for transmissivities @xmath318 and values of the correlation parameter @xmath319 , where the entanglement - distillation threshold is now @xmath320 at high transmissivities @xmath321 , we have @xmath322 , which means that distillable entanglement can be restored in the presence of separable ac environments . for a pictorial representation see fig .
[ linea2 ] . ) . here
we consider the regime of high transmissivity @xmath323 , where remote entanglement can be distilled even in the presence of separable environments ( @xmath324).,scaledwidth=55.0% ] finally , we analyze the behavior of the remote entanglement in terms of the different types of environmental correlations . as evident from the top panel of fig .
[ discordac1 ] , remote entanglement starts to be restored ( @xmath299 ) only after certain values of gaussian discord and classical correlations are present in the positive ac environment .
after these critical values , the number of remote ebits of log - negativity are increasing in both correlations , as shown in the bottom panel of fig .
[ discordac1 ] .
curves refer to the same numerical example as before ( @xmath290 ) and similar behavior holds at the other transmissivities .
in general , the positive ac environments outperform the sc environments in restoring entanglement , as it is clear by comparing the curves of fig .
[ discordac1 ] to those of fig .
[ discordsc1 ] . such a better performance can also be recognized from the asymmetric shape of the reactivation area in fig .
[ dirtot ] .
( dimensionless units ) , gaussian quantum discord @xmath218 ( dbits ) and purely - classical correlations @xmath223 ( cbits ) are plotted versus the correlation parameter @xmath191 of the ac environment .
remote entanglement is restored ( @xmath299 ) only in positive ac environments ( @xmath291 ) and after critical values of @xmath218 and @xmath223 .
optimal restoration is achieved for the positive epr environment , where @xmath325 are maximal .
numerical example for @xmath290 and @xmath326 @xmath294 . _ bottom panel_. remote entanglement quantified by the log - negativity ( ebits ) is plotted versus gaussian discord ( dbits , solid line ) and classical correlations ( cbits , dashed line ) present in the environment .
once entanglement is reactivated , the number of remote ebits is increasing in the number of environmental dbits and cbits ( here @xmath177 and @xmath182 are chosen as before).,scaledwidth=36.0% ] the better performance of these environments is also evident from fig .
[ criticalacpic ] , where we plot the critical number of correlation bits which are needed for the restoration of remote entanglement ( dbits , cbits and bits of total correlations ) .
we need less than @xmath327bits of total correlations to trigger the reactivation at any transmissivity @xmath328 .
in particular , the critical number of dbits is always less than @xmath329 . for @xmath177 close to zero ,
a negligible amount of correlations is needed .
this is compatible with the fact that , for these environments , the threshold for entanglement - restoration @xmath282 goes rapidly to zero for @xmath330 , so that a small @xmath331 is sufficient to reactivate the entanglement distribution in the high loss regime . in the opposite regime of extremely low loss ( @xmath296 ) and high thermal noise ( @xmath297 )
, the critical number of dbits falls to zero while the numbers of cbits tend to its maximum . , we plot the critical number of correlation bits after which remote entanglement starts to be restored .
we show the critical numbers of dbits ( solid ) , cbits ( dashed ) and their total ( dotted ) , corresponding to bits of quantum mutual information.,scaledwidth=45.0% ] according to the previous sections , the restoration of entanglement is a threshold process which starts only after a critical amount of environmental correlations is injected into the travelling modes . after this threshold , the number of ebits remotely recovered is monotonically increasing in the number of correlation bits ( dbits , cbits or total bits ) .
however , this monotonic behavior holds true as long as we compare environments within the same class , i.e. , we analyze sc environments separately from ac environments . in general , the monotonicity is lost when we compare two arbitrary points in the correlation plan .
for instance , it is sufficient to compare the msc environment with the ( positive ) mac environment to find that entanglement restoration is not monotonic in any type of correlations , i.e. , classical , quantum or total . as we have already discussed in sec .
[ subsecspecific ] , at any fixed thermal noise @xmath207 , msc and mac environments have exactly the same amount of classical correlations @xmath332 , but different gaussian discord @xmath333 and , therefore , different total correlations @xmath334 .
this is because mac environments are more entropic than msc environments .
their difference in entropy@xmath335 directly quantifies the difference in the correlations , i.e. , @xmath336 in particular , all these relations hold true at the threshold for entanglement - breaking @xmath255 , where the number of ebits remotely restored is bigger for the positive mac environment , i.e. , the log - negativities satisfy @xmath337 which is equivalent to @xmath338 , where the two eigenvalues @xmath339 and @xmath340 are given in eqs .
( [ epsmsc ] ) and ( [ epsmac ] ) , respectively .
thus , in this specific example , an increase in the environmental correlations , quantified by the gaussian discord ( @xmath341 ) or the total correlations ( @xmath342 ) , corresponds to a decrease in the number of ebits remotely recovered ( @xmath343 ) . at the same time
, we also see that this decrease happens for constant classical correlations ( @xmath344 ) .
such a phenomenon could have different potential explanations .
there could be a problem with gaussian discord , which could not be equal to the actual discord for these environments .
if this were true , we would have a different quantification for both quantum correlations @xmath218 and classical correlations @xmath223 .
then , monotonicity could be re - established in terms of these actual quantities .
more drastically , it could be that entanglement restoration might be monotonically connected with a different and more subtle quantification of the environmental correlations .
it is also important to note that this analysis does not change if we consider a different quantification for the remote entanglement .
for instance , we could consider the entanglement of formation @xcite in the place of the log - negativity to quantify the entanglement of alice and bob s two - mode state .
however , such a state turns out to be a symmetric gaussian state , so that its entanglement of formation is equal to its gaussian entanglement of formation @xcite , which is monotonically related to its log - negativity @xcite . in order to give another point of view to the dynamics of the process
, we describe the evolution of the epr correlations of charlie s input state when the two modes are transmitted through the correlated - noise gaussian environment , with transmission @xmath177 , thermal noise @xmath182 and correlations @xmath183 , as depicted in the left panel of fig .
[ twomodeeb ] ( see appendix [ app_subepr ] for more details on epr correlations ) . as we will see below , the entanglement restored at alice s and bob s stations is not necessarily in the form epr correlations .
we can easily write the input - output bogoliubov transformations for the bosonic modes involved in the process@xmath345 where @xmath346 is a vector of quadratures . from these equations
, we can extract the output epr operators @xmath347 now , using the cm of the input epr state ( [ eprcmmu ] ) and that of the environment ( [ eve_cmapp ] ) , we can compute their variances@xmath348{cc}% v(\hat{q}_{-}^{\prime } ) & \\ & v(\hat{p}_{+}^{\prime } ) \end{array } \right ) \nonumber\\ & = \tau(\mu-\mu^{\prime})\mathbf{i}+(1-\tau)(\omega\mathbf{i}-\mathbf{zg})~.\end{aligned}\ ] ] in the limit of @xmath259 , we have@xmath349 and assuming entanglement - breaking ( @xmath255 ) we get@xmath350 analyzing eq .
( [ eprheis2 ] ) we see that the epr condition @xmath351 can be realized by suitable choices of the correlation block @xmath183 ( here the epr condition corresponds to having quadrature correlations between the output modes , quantified by the variances @xmath352 and @xmath353 , which are below the unit - variance of the vacuum noise ) .
let us explicitly compare the different types of environments .
for the memoryless environment ( @xmath354 ) we have @xmath355 which means that the epr variances are always greater than or equal to one , i.e. , epr correlations do not survive .
for the ac environment ( @xmath356 ) we have @xmath357\mathbf{i~}.\ ] ] it is easy to check that the epr condition @xmath358 is achieved by physical values of @xmath359 , where @xmath282 is given in eq .
( [ gegasymm ] ) .
this means that the remote entanglement generated by this environment is always in the form of epr correlations ( of the same type of the original epr state at charlie s station ) . for the sc environment ( @xmath360 ) we have @xmath361 and we can check that the condition @xmath362 is not realizable by any choice of @xmath191 .
thus , the initial epr correlations do not survive in this case .
nevertheless we have explicitly proven that remote entanglement can be distributed in the presence of this environment .
in conclusion , we have shown how the combination of two entanglement - breaking channels into a joint correlated - noise environment can restore the distribution of entanglement . interestingly
, this reactivation is successfully induced by the presence of separable correlations and occurs for quantum systems with hilbert spaces of any dimension ( qubits , qudits and bosonic modes ) . in our first analysis ,
we have considered twirling environments which are classically - correlated , being realizable by random loccs and described by zero - discord environmental states .
we have considered correlated pauli environments for qubits , multidimensional twirling environments for qudits and , finally , non - gaussian bosonic environments based on phase - space rotations .
all these scenarios are characterized by one - system entanglement - breaking ( so that charlie can not share any entanglement with alice or bob ) , and two - system entanglement - preserving ( so that charlie is able to transmit entanglement to alice and bob ) . achieving these simultaneous conditions is somehow easy for these environments , since there are well - known classes of entangled states which are invariant under their action ( thus forming decoherence - free subspaces ) . depending on the type of twirling ,
i.e. , @xmath41 or @xmath42 , these invariant states are werner or isotropic states , respectively . in the case of bosonic systems , we can find entangled states which are invariant under the action of anti - correlated phase - space rotations @xmath363 , and they coincide with the continuous - variable werner states .
more interestingly , in our second analysis on bosonic systems in gaussian environments , we have shown how entanglement distribution is still possible even in the absence of decoherence - free subspaces ( a similar relaxation of conditions can also be found in the context of quantum error correction @xcite ) . in this case ,
entanglement is no longer preserved but still can survive the double transmission and be distilled by the remote parties .
more precisely , we have considered a realistic model of gaussian environment with correlated noise , which generalizes the standard memoryless environment with losses and thermal noise .
then , we have imposed the condition of one - mode entanglement - breaking by choosing a sufficiently high value of thermal noise @xmath364 for any value of the transmission ( so that charlie can never share entanglement with the remote parties ) . under these conditions of extreme decoherence
, we have shown that a distillable amount of entanglement is able to survive the two - mode transmission and reach alice s and bob s stations .
this reactivation of entanglement distribution is a direct effect of the correlations which are injected by the gaussian environment .
remarkably , the injection of separable correlations is sufficient to break the mechanism of entanglement - breaking and trigger the reactivation . in a further analysis
, we have shown that remote entanglement starts to be restored only after a critical amount of correlations is present in the environment , where these correlations can be quantified in terms of gaussian discord , classical correlations and total correlations ( quantum mutual information ) .
after this critical amount has been reached , the number of remote ebits is not always increasing in the number of correlation bits injected by the environment , despite such a monotonicity being true when we restrict the environments to classes with specific types of noise correlations ( sc and ac environments ) .
it is possible that the monotonicity could be extended to the entire correlation plan if gaussian discord were proven to be strictly larger than the actual discord .
more drastically , the monotonicity could hold in terms of a different and unknown quantification of the environmental correlations . from a theoretical point of view
, the fact that separability can be exploited to recover from entanglement - breaking is a paradoxical behavior which poses fundamental questions on the intimate relations between local and nonlocal correlations .
while the reactivation process is easily understandable in the presence of entangled environments , where the original system entanglement is partly or fully replaced by the environmental entanglement , its interpretation is puzzling in the case of separable environments , where no injection of entanglement may take place . in conclusion , in terms of potential practical impact ,
our analysis shows new perspectives for entanglement distribution and distillation in conditions of extreme decoherence as long as the presence of noise correlations and memory effects can be identified in the environment , such as those arising from non - markovian dynamics of open quantum systems @xcite . according to our findings
, such correlations and effects do not need to be strong , since separable correlations and classical memories may be sufficient to restore broken entanglement .
note that memory channels and non - markovian environments are ubiquitous .
for instance , they naturally arise in the context of spin chains @xcite and micromasers @xcite .
other important examples can be found in condensed matter , in particular when we consider the dynamics of quantum dots in photonic crystals @xcite . in the bosonic
setting , memory gaussian channels come into play when electromagnetic modes propagate through dispersive media , such as linear optical systems or free - space . in this case ,
correlations and memory effects are naturally introduced by diffraction @xcite .
finally , other examples of bosonic memory channels can also be found in the propagation of electromagnetic radiation through atmospheric turbulence @xcite .
this work has been supported by epsrc under the research grant hipercom ( ep / j00796x/1 ) and by the leverhulme trust .
special thanks of the author are for s. l. braunstein , m. paternostro and c. ottaviani .
the author would also like to thank ( in random order ) p. horodecki , o. oreshkov , a. furusawa , g. spedalieri , g. adesso , m. gu , s. mancini , g. chiribella , o. hirota , b. munro , n. metwally , r. namiki , p. tombesi , s. guha , m. j. w. hall , s. danilishin , m. barbieri , m. bellini , r. filip , and j. eschner for discussions and comments .
here we report some of the proofs related to our analysis of entanglement preservation in twirling environments .
this appendix contains new results but also known facts which are given to the reader for the sake of completeness ( e.g. , sec .
[ dephaseapp ] ) . here
we show the conditions under which the depolarizing channel of eq .
( [ qubitdepo ] ) becomes an entanglement - breaking channel .
this means to find a specific regime for the probabilities @xmath365 characterizing the channel .
first of all note that , for hilbert spaces of finite dimension @xmath78 , a simple way to check if a quantum channel @xmath0 is entanglement - breaking is to test it on the maximally - entangled state @xmath366 of eq .
( [ maxentstate ] ) . in other words , if @xmath367 is separable , then @xmath368 is separable for any input state @xmath3 @xcite . in the case of qubits , we can test the channel on the triplet state @xmath369 of eq .
( [ tripletstate ] ) .
we then compute the output state@xmath370 adopting the computational basis @xmath371 and using @xmath372 , @xmath373 and @xmath50 , we get@xmath374 where the coefficients @xmath375 are the elements of the following density matrix@xmath376{cccc}% p_{0}+p_{3 } & 0 & 0 & p_{0}-p_{3}\\ 0 & p_{1}+p_{2 } & p_{1}-p_{2 } & 0\\ 0 & p_{1}-p_{2 } & p_{1}+p_{2 } & 0\\ p_{0}-p_{3 } & 0 & 0 & p_{0}+p_{3}% \end{array } \right ) ~.\ ] ] to check the separability properties we adopt the peres - horodecki criterion @xcite .
this corresponds to compute the partial transpose ( pt ) of the state which is given by the following linear map @xmath377 at the level of the density matrix we then have @xmath378 i.e. , @xmath379 swapping .
it is easy to check that the partially - transposed matrix @xmath380{cccc}% p_{0}+p_{3 } & 0 & 0 & p_{1}-p_{2}\\ 0 & p_{1}+p_{2 } & p_{0}-p_{3 } & 0\\ 0 & p_{0}-p_{3 } & p_{1}+p_{2 } & 0\\ p_{1}-p_{2 } & 0 & 0 & p_{0}+p_{3}% \end{array } \right)\ ] ] has eigenvalues @xmath381 thus the partially - transposed state has the following spectral decomposition@xmath382 with @xmath383 orthogonal eigenstates .
this operator is positive ( @xmath384 ) if and only if@xmath385 as a result , the output state @xmath386 is separable ( and the channel is entanglement - breaking ) if and only if @xmath66 for every @xmath58 , as reported in the main text .
let us consider two qudits with the same dimension @xmath78 , so that the composite system is described by an hilbert space @xmath387 with finite dimension @xmath86 . from the literature @xcite , we know that we can write the following equality for any input state @xmath3 subject to a @xmath41 twirling channel@xmath388 which is valid for @xmath100 , where @xmath93 is a unitary 2-design with @xmath95 elements .
here we can easily show that @xmath389 where @xmath89 belongs to the same design @xmath93 as before . in a few words , the two twirling channels , @xmath390 and @xmath391 ,
can be decomposed using the same unitary 2-design . in the first step of the proof
we show that @xmath390 and @xmath392 are connected by a partial transposition .
consider an arbitrary input state @xmath3 decomposed in the orthonormal basis of @xmath393 @xcite@xmath394 its partial transpose corresponds to transposing system @xmath2 only , i.e.,@xmath395 we first prove that @xmath396\}~. \label{ptconnection}%\ ] ] in fact , by linearity we have @xmath397=\int_{\mathcal{u}(d)}du~(u\otimes u)\mathrm{pt}(\rho_{ab})(u\otimes u)^{\dagger}\nonumber\\ = \sum_{ijkl}\rho_{ij}^{kl}\int_{\mathcal{u}(d)}du~u\left\vert i\right\rangle _ { a}\left\langle j\right\vert u^{\dagger}\otimes u\left\vert l\right\rangle _ { b}\left\langle k\right\vert u^{\dagger}.\end{gathered}\ ] ] since @xmath398^{t}=u^{\ast}\left\vert k\right\rangle _ { b}\left\langle l\right\vert ( u^{\ast})^{\dagger}$ ] we get@xmath399\}=\nonumber\\ = \sum_{ijkl}\rho_{ij}^{kl}\int_{\mathcal{u}(d)}du~u\left\vert i\right\rangle _ { a}\left\langle j\right\vert u^{\dagger}\otimes u^{\ast}\left\vert k\right\rangle _ { b}\left\langle l\right\vert ( u^{\ast})^{\dagger}\nonumber\\ = \int_{\mathcal{u}(d)}du(u\otimes u^{\ast})\rho_{ab}(u\otimes u^{\ast } ) ^{\dagger}=\mathcal{e}_{uu^{\ast}}(\rho_{ab})~.\end{gathered}\ ] ] now the second step is to combine eq .
( [ ptconnection ] ) with the unitary @xmath213-design for @xmath390 .
first it is important to note that the equivalence between eqs .
( [ euu1 ] ) and ( [ euu2 ] ) is valid not only when @xmath3 is a density operator but , more generally , when it is an hermitian linear operator .
this extension is straightforward to prove .
suppose that the linear operator @xmath400 is hermitian .
then its spectral decomposition involves real eigenvalues and orthonormal eigenvectors , i.e. , we can write@xmath401 where @xmath402 and @xmath403 .
now we can write @xmath404 where ( [ s1])@xmath9([s2 ] ) by linearity , ( [ s2])@xmath405([s3 ] ) by the fact that @xmath406 are projectors ( and therefore states ) and ( [ s3])@xmath9([s4 ] ) by linearity again . as a result
, we can apply the equivalence between eqs .
( [ s1 ] ) and ( [ s4 ] ) to the linear operator @xmath407 which fails to be a density operator when @xmath3 is entangled but still it is hermitian ( and unit trace ) in the general case .
thus , we can write@xmath397=\frac{1}{k}\sum_{t=0}^{k-1}% ( u_{t}\otimes u_{t})\mathrm{pt}(\rho_{ab})(u_{t}\otimes u_{t})^{\dagger } \nonumber\\ = \frac{1}{k}\sum_{t=0}^{k-1}\sum_{ijkl}\rho_{ij}^{kl}u_{t}\left\vert i\right\rangle _ { a}\left\langle j\right\vert u_{t}^{\dagger}\otimes u_{t}\left\vert l\right\rangle _ { b}\left\langle k\right\vert u_{t}^{\dagger}~.\end{gathered}\ ] ] now , using the connection in eq .
( [ ptconnection ] ) and the fact that @xmath408^{t}=u_{t}^{\ast}\left\vert k\right\rangle _ { b}\left\langle l\right\vert ( u_{t}^{\ast})^{\dagger}$ ] , we can write@xmath409\}\nonumber\\ = \frac{1}{k}\sum_{t=0}^{k-1}\sum_{ijkl}\rho_{ij}^{kl}u_{t}\left\vert i\right\rangle _ { a}\left\langle j\right\vert u_{t}^{\dagger}\otimes u_{t}^{\ast}\left\vert k\right\rangle _ { b}\left\langle l\right\vert ( u_{t}^{\ast})^{\dagger}\nonumber\\ = \frac{1}{k}\sum_{t=0}^{k-1}(u_{t}\otimes u_{t}^{\ast})\rho_{ab}(u_{t}\otimes u_{t}^{\ast})^{\dagger}~ , \label{finapt}%\end{gathered}\ ] ] which gives the equivalence between eqs . ( [ euustar1 ] ) and ( [ euustar2 ] ) . in this short appendix
we give a simple proof of eq .
( [ depola2 ] ) . consider an hilbert space @xmath410 with finite dimension @xmath411 ( where @xmath412 and @xmath413 are generally different ) .
given a linear operator @xmath414 , we can always decompose it in an orthonormal basis @xmath415 then , we can write the following partial haar average , where only system @xmath1 is averaged on the unitary group@xmath416 now we use eq .
( [ twirlingid ] ) with linear operator @xmath417 , which gives@xmath418 then , by replacing this expression in eq .
( [ tui ] ) , we get@xmath419 this is a simple extension of eq .
( [ twirlingid ] ) to considering the presence of a second ( unaveraged ) system @xmath2 .
in particular , for @xmath420 density operator , we have the result of eq .
( [ depola2 ] ) . here
we prove that the uniformly dephasing channel of eq .
( [ unidepha ] ) is entanglement - breaking .
this is a simple proof of a very intuitive fact , which is given to the reader for completeness .
first of all , let us consider a pure input state @xmath421 , expressed in the fock basis of the two modes as @xmath422 since @xmath423 , we get@xmath424 then , by re - distributing the sum , we get@xmath425 where we have introduced @xmath426 in the last step ( clearly @xmath427 ) . now introducing the pure state@xmath428 we can write the following spectral decomposition for the output state@xmath429 in particular , note that we can always write the tensor product @xmath430 so that the output state is manifestly in separable form @xmath431 proof
can trivially be extended to mixed states via their spectral decomposition into pure states . to derive the cm of eq .
( [ cm_appfinite ] ) just check that a @xmath433 real matrix @xmath434 is invariant under rotations @xmath435 ( or , equivalently , it commutes with rotations @xmath436=0 $ ] ) if and only if it takes the asymmetric form@xmath437{cc}% m & t\\ -t & m \end{array } \right ) ~,\ ] ] with @xmath438 real numbers .
thus , the @xmath439 blocks of the cm ( [ cm_appfinite ] ) must have this general form , with the blocks @xmath440 and @xmath441 diagonal by imposing the additional condition of symmetry . then , it is easy to check that cm of eq .
( [ cm_appfinite ] ) describes a separable gaussian state .
in fact , using suitable local rotations @xmath442 ( therefore not changing the separability properties of the state ) , we can transform @xmath150 into the simpler form@xmath443{cc}% \alpha\mathbf{i } & \gamma\mathbf{i}\\ \gamma\mathbf{i } & \beta\mathbf{i}% \end{array } \right ) ~ , \label{vabprimeapp}%\ ] ] where @xmath444 . without loss of generality ,
suppose that @xmath445 and set @xmath446 .
we can always generate @xmath447 with cm @xmath448 by applying local gaussian channels @xmath449 to the symmetric gaussian state @xmath450 with cm @xmath451 .
it is sufficient to choose the identity channel @xmath5 and a gaussian channel @xmath452 with additive noise @xmath453 ( also known as canonical b2 form @xcite ) .
it is now trivial to check that the state @xmath454 is separable .
in fact , @xmath455 is a bona - fide quantum cm when its parameters @xmath456 and @xmath118 satisfy the conditions @xmath457 and @xmath458 .
then , we can check that the partially - transposed symplectic eigenvalues of @xmath455 are greater than @xmath243 , i.e. , the state is separable , when @xmath459 , which is a condition always satisfied .
finally , since @xmath460 is separable , then also @xmath447 and @xmath3 must be separable ( local operations can not create entanglement ) .
the proof is easy .
suppose that we have an arbitrary gaussian state @xmath461 , with mean value @xmath462 and covariance cm @xmath150 .
the action of two gaussian unitaries @xmath41 on @xmath3 corresponds to apply two identical displacements to its mean value @xmath463 and two identical symplectic matrices to its cm @xmath464 in general , there is no chance to find an invariant gaussian state , since any nonzero displacement @xmath465 maps the input state into a different output state .
we then restrict the search to considering canonical gaussian unitaries ( @xmath466 ) which are one - to - one with the symplectic transformations . according to euler s decomposition @xcite
, any single - mode symplectic transformation @xmath467 can be decomposed into an orthogonal rotation @xmath468 , a single - mode squeezing @xmath469 , and another orthogonal rotation @xmath470 , where @xmath471 are angles and @xmath472 is a squeezing parameter .
as long as some squeezing is present ( @xmath473 ) , the trace of the cm changes ( physically this corresponds to increase the mean number of photons ) . as a result ,
the cm and , therefore , the gaussian state must change .
thus , we are left to find gaussian states which are invariant under rotations only ( which are passive transformations , i.e. , preserving the trace of the cm ) .
however , we have already seen that , despite they exist , these gaussian states must be separable .
in this appendix , we provide notions on bosonic systems , gaussian states and their properties , which are useful for understanding the mathematical details of the main part of the paper . in detail
, we discuss gaussian states and their statistical moments ( sec .
[ app_subgauss ] ) , the symplectic manipulation of covariance matrices ( sec . [ app_subspectrum ] ) , the bona - fide conditions for bipartite gaussian states and their separability properties ( sec . [ app_sub2modes ] ) , epr correlations ( sec . [ app_subepr ] ) and , finally , the notions of discord and gaussian discord ( sec
. [ app_subdiscord ] ) .
[ [ gaussian - states - and - operations - representation - in - terms - of - the - statistical - momentsapp_subgauss ] ] gaussian states and operations : representation in terms of the statistical moments[app_subgauss ] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ the most important systems in continuous variable quantum information are the bosonic modes of the electromagnetic field .
a bosonic mode is a quantum system with an infinite - dimensional hilbert space and described by a pair of quadrature operators : position @xmath474 and momentum @xmath475 , satisfying the commutation relation @xmath476=2i$ ] ( here we set @xmath477 ) .
more generally , a bosonic system of @xmath478 modes is described by a vector of @xmath479 quadrature operators@xmath480 satisfying @xmath481=2i\omega_{ij}$ ] , where @xmath482 and @xmath483 is the generic element of the symplectic form@xmath484{cc}% 0 & 1\\ -1 & 0 \end{array } \right ) ~. \label{symplectic_form}%\ ] ] in experimental quantum optics , bosonic modes are typically prepared in gaussian states . by definition ,
a quantum state @xmath485 is gaussian when its wigner phase - space representation is gaussian @xcite .
a gaussian state is very easy to describe , being completely characterized by its first and second - order statistical moments .
the first - order moment is known as the mean value , and defined by @xmath486 , where @xmath487 denotes the average of the arbitrary operator @xmath488 on the state @xmath485 .
the second - order moment is known as the covariance matrix ( cm ) @xmath130 , with generic element@xmath489 where @xmath490 is the deviation and @xmath491 is the anticommutator .
note that the diagonal elements @xmath492 are just the variances of the quadratures @xmath493 the cm is a @xmath494 real symmetric matrix , which must satisfy the uncertainty principle @xcite@xmath495 implying the positive - definiteness @xmath496 @xcite .
the simplest gaussian states are thermal states .
a single - mode thermal state has zero mean and cm@xmath497 , where @xmath498 is the @xmath433 identity matrix and @xmath499 is the mean number of thermal photons ( vacuum state for @xmath500 ) .
multimode thermal states are constructed by tensor product .
tensor product of states @xmath501 corresponds to direct sum of cms @xmath502 .
conversely , the partial trace @xmath503 corresponds to collapsing the total cm @xmath504 into the block @xmath505 spanned by @xmath506 . in the experimental practice ,
gaussian states are typically processed by gaussian operations .
the simplest ones are gaussian unitaries , defined as those unitary operators which transform gaussian states into gaussian states . at the level of the statistical moments , the action of a gaussian unitary @xmath507 corresponds to@xmath508 where @xmath465 is a real displacement vector and @xmath467 is a symplectic matrix , i.e. , a real matrix preserving the symplectic form @xmath509 . in the heisenberg picture , a gaussian unitary corresponds to the affine map @xmath510 in particular , a gaussian unitary is called canonical when @xmath511 .
the most important example of canonical unitary is the beam - splitter transformation , which involves two bosonic modes .
this unitary is characterized by the symplectic matrix@xmath512{cc}% \sqrt{\tau}\mathbf{i } & \sqrt{1-\tau}\mathbf{i}\\ -\sqrt{1-\tau}\mathbf{i } & \sqrt{\tau}\mathbf{i}% \end{array } \right ) ~ , \label{bsmatrix}%\ ] ] with transmissivity parameter @xmath513 .
other important examples of gaussian operations are the gaussian channels @xcite .
these channels describe all those cases where bosonic systems and environment interact by means of linear and/or bilinear hamiltonians .
the action of a gaussian channel @xmath514 corresponds to the following transformation for the cm of the state@xmath515 where @xmath516 and @xmath517 are @xmath494 real matrices , satisfying the condition
@xmath518 @xcite .
the most important example of gaussian channel is the lossy channel , which is typically used to model the optical propagation through dissipative linear media .
when a single bosonic mode propagates through a lossy channel , its cm is transformed by eq .
( [ transformationch ] ) with@xmath519 where @xmath513 is the transmissivity of the channel and @xmath520 its thermal number .
a single - mode lossy channel can always be dilated into a beam - splitter with transmissivity @xmath177 , which mixes the incoming mode with an environmental thermal mode with @xmath173 mean photons .
a central result in the theory of gaussian states is williamson s theorem @xcite . given a cm @xmath130
, there is a symplectic matrix @xmath467 realizing the diagonalization @xmath521 where the diagonal matrix @xmath522 is known as the williamson normal form , and @xmath523 are the @xmath478 symplectic eigenvalues of the cm . since @xmath524 , we have that@xmath525 using the symplectic spectrum , we can write the uncertainty principle of eq .
( [ unc_princ ] ) in the equivalent form@xmath526 which implies @xmath527 @xcite .
the symplectic spectrum fully determines the entropic and purity properties of the gaussian states .
in fact , the von neumann entropy @xmath528tr@xmath529 of an arbitrary @xmath478-mode gaussian state is expressed by @xcite@xmath530 where the function @xmath531 is given in eq .
( [ hvonneumann ] ) , with the logarithmic base equal to @xmath213 for bits or @xmath215 for nats .
the purity of a gaussian state is given by@xmath532 so that the state is pure ( mixed ) iff @xmath533 ( @xmath534 ) . in our paper , we consider situations where the symplectic eigenvalues are large . in this case , it is useful to use the asymptotic expansion@xmath535 which is valid for large @xmath536 . in particular , if the whole symplectic spectrum is diverging ( @xmath537 large for any @xmath58 ) , then we can use eq .
( [ expa ] ) to write the asymptotic formula@xmath538 where the entropy is directly related to the purity of the gaussian state .
[ [ two - mode - gaussian - states - and - bipartite - entanglementapp_sub2modes ] ] two - mode gaussian states and bipartite entanglement[app_sub2modes ] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ since we are interested in the distribution of bipartite entanglement , we devote a specific section to gaussian states of two bosonic modes and their separability properties .
let us consider two modes , @xmath1 and @xmath2 , with quadrature vector @xmath539 .
we assume that these modes are described by a zero - mean gaussian state @xmath3 .
since @xmath540 , this state is fully characterized by its cm , that we can always express in the blockform @xmath541{cc}% \mathbf{a } & \mathbf{c}\\ \mathbf{c}^{t } & \mathbf{b}% \end{array } \right ) ~ , \label{blockform}%\ ] ] where @xmath440 , @xmath441 and @xmath542 are @xmath433 matrices . finding the symplectic spectrum is straightforward , since @xcite@xmath543 where @xmath544 .
the uncertainty principle is then equivalent to the bona - fide condition@xmath545 where @xmath546 is the smallest symplectic eigenvalue . as an example , consider the two - mode cm of the form @xmath541{cccc}% \omega & & g & \\ & \omega & & g^{\prime}\\ g & & \omega & \\ & g^{\prime } & & \omega \end{array } \right ) \label{cm_app2}%\ ] ] where @xmath185 quantifies the thermal noise present in each mode , while @xmath191 and @xmath192 are cross - correlation parameters .
this is the cm of the correlated - noise gaussian environment which is defined in the main text [ see eq .
( [ eve_cmapp ] ) ] .
it is easy to derive simple bona - fide conditions in terms of the parameters @xmath182 , @xmath191 and @xmath192 .
in particular , given @xmath185 , we can easily derive conditions for the two correlation parameters , by imposing eq .
( [ bonafide ] ) .
the positivity @xmath496 is equivalent to the positivity of the principal minors of the matrix .
the positivity of the first two minors is trivially implied by @xmath185 .
the third minor gives @xmath547{ccc}% \omega & 0 & g\\ 0 & \omega & 0\\ g & 0 & \omega \end{array } \right ) > 0\leftrightarrow\omega(\omega^{2}-g^{2})>0,\ ] ] which is equivalent to@xmath548 the fourth minor corresponds to the determinant@xmath549 and its positivity @xmath550 leads to the condition@xmath551 we then derive the symplectic spectrum of the cm ( [ cm_app2 ] ) , which is given by @xmath552 now , we find that @xmath553 is equivalent to @xcite@xmath554 thus , an environmental cm of the form ( [ cm_app2 ] ) with thermal noise @xmath185 is a bona - fide cm if the correlation parameters @xmath191 and @xmath192 satisfy the three bona - fide conditions of eqs .
( [ gcon1 ] ) , ( [ gcon2 ] ) and ( [ gcon3 ] ) which are also given in eq . ( [ cmconstraints ] ) of the main text .
using eqs .
( [ purity ] ) and ( [ detvapp ] ) , we also see that the purity of the gaussian state with cm ( [ cm_app2 ] ) is given by @xmath555 ^{-1/2}. \label{purcon}%\ ] ] its von neumann entropy is computed by using eq .
( [ vn_entropy ] ) and the spectrum in eq .
( [ spectrum123 ] ) .
it is easy to study the separability properties of an arbitrary two - mode gaussian state with cm in the generic blockform of eq .
( [ blockform ] ) .
in fact , it is sufficient to compute the smallest partially - transposed symplectic ( pts ) eigenvalue @xmath269 .
this eigenvalue can be computed using the formula of @xmath546 , given in eq .
( [ symform ] ) , proviso that we replace @xmath556 with @xmath557 .
then , a gaussian state is separable ( entangled ) if and only if @xmath558 ( @xmath266 ) .
more strongly , this eigenvalue provides a quantification of entanglement , being an entanglement monotone for two - mode gaussian states .
in fact , it is monotonically related to the log - negativity @xmath559 , which is another entanglement monotone and can be expressed in entanglement bits ( ebits ) .
the log - negativity @xmath263 provides an upper - bound to the distillable entanglement , corresponding to the average number of ebits per copy which can be extracted from infinitely - many copies of the state @xmath560 we can also consider a lower - bound to distillable entanglement . according to the hashing inequality @xcite , this lower bound is given by the coherent information @xcite @xmath561 where @xmath25 is the reduced state of bob , here corresponding to a zero - mean gaussian state with cm @xmath441 .
using eq .
( [ vn_entropy ] ) , the coherent information becomes@xmath562 where @xmath563 is the symplectic eigenvalue of @xmath441 , and @xmath564 is the symplectic spectrum of @xmath130 . in the case where these eigenvalues are all large , we can use the expansion of eq .
( [ expa ] ) to get the formula@xmath565 more precisely , the coherent information is a lower bound to the distillable entanglement which is achievable by means of one - way protocols between alice and bob ( one - way distillability ) . in these protocols
, alice applies a quantum instrument to her copies , i.e. , a collection of completely - positive trace - preserving maps labelled by a classical index @xmath58 .
then , she classically communicates @xmath58 to bob , who applies a conditional quantum operation on his copies ( see ref .
@xcite for more details ) .
the most important example of two - mode gaussian state is the two - mode squeezed vacuum state , also known as the epr state .
this state represents the most common source of continuous - variable entanglement .
an epr state has zero mean and cm given by eq .
( [ eprcmmu ] ) in the main text .
the quadratures @xmath566 have equal variances @xmath567 and maximal correlations @xmath568 the parameter @xmath112 can be used to quantify the entanglement between the two modes @xmath1 and @xmath2 .
in fact , the smallest pts eigenvalue takes the form@xmath569 which is monotonically decreasing in @xmath112 .
we can easily check that we have entanglement ( @xmath266 ) for any @xmath570 .
the epr correlations can be characterized in terms of the variance of the epr quadratures @xmath571 the epr condition corresponds to @xmath572 , which means that the quadrature correlations between the two modes are below the vacuum noise . for the epr state described by eq .
( [ eprcmmu ] ) indeed we have @xmath573 which is less than @xmath243 for any @xmath570 . for this type of state , epr correlations and entanglement are equivalent conditions . in general , for an arbitrary two - mode gaussian state , the presence of epr correlations is only a sufficient condition for entanglement . in the limit of large entanglement ( @xmath574 )
, the epr state becomes ideal .
in fact , we have @xmath575 , which corresponds to realizing the ideal epr correlations @xmath576 and @xmath577 , where positions ( momenta ) are perfectly correlated ( anticorrelated ) .
note that an alternative epr state can be defined using the other reflection matrix @xmath578 in the place of @xmath579 in eq .
( [ eprcmmu ] ) .
this alternative state has epr correlations in the quadratures @xmath580 and @xmath581 , therefore tending to realize the ideal conditions @xmath582 and @xmath583 . in our work
, we do not consider this state , but similar results can easily be derived .
quantum entanglement is synonymous of quantum correlations for pure states . for mixed states
the scenario is more subtle .
in fact , separable mixed states may still contain features of quantumness , one of which is known as quantum discord @xcite .
quantum discord derives from different quantum extensions of the notion of classical mutual information .
classically , the mutual information of two random classical variables , @xmath1 and @xmath2 , is defined as @xmath584 where @xmath585 is the shannon entropy . using the chain rule for the entropy , we can also write the equivalent formulation@xmath586 where @xmath587 is the conditional shannon entropy . in quantum mechanics , we can have two inequivalent generalizations , the first being provided by the quantum mutual information which accounts for the total correlations present in a bipartite state @xmath3 .
it is expressed in terms of the von neumann entropy as @xmath588 another generalization is given by the entropic quantity@xmath589 where @xmath590 is a povm acting on system @xmath2 and @xmath591 where @xmath54 is the probability of the outcome @xmath58 and @xmath592 is the conditional state of @xmath1 .
the quantity @xmath593 , which is expressed in bits or classical bits ( cbits ) , quantifies the classical correlations within the state , corresponding to the maximal amount of common randomness which can be extracted using local operations and one - way classical communication @xcite .
quantum discord aims to quantify the quantum correlations in the state .
thus , it is defined as the difference between total and classical correlations@xmath594 and measured in bits or discordant bits ( dbits ) .
it is important to note that eq .
( [ cdeve ] ) is generally non - symmetric under system permutation . as a result , we have that the @xmath2-type quantities @xmath595 and @xmath596 are generally different from the @xmath1-type quantities of eqs .
( [ cdeve ] ) and ( [ discordeq ] ) .
such ambiguity clearly disappears for symmetric states ( @xmath597 ) , as the states considered in our paper . for bosonic states
, one can define the gaussian discord by restricting the previous minimization from arbitrary to gaussian povms @xcite .
it is clear that gaussian discord represents an upper bound to the actual discord , and the classical correlations of eq .
( [ cdeve ] ) restricted to gaussian povms represent a lower bound to the actual classical correlations of the state .
gaussian discord is conjectured to be the actual discord in the case of bosonic gaussian states , for which we know closed formulas @xcite .
in fact , consider a two - mode gaussian state with cm in the blockform of eq .
( [ blockform ] ) and symplectic spectrum @xmath564 .
then , the ( restricted ) classical correlations and gaussian discord are given by@xmath598 where the expression of the @xmath599-term is given in ref .
apart from the specific case of tensor - product states @xmath600 , all two - mode gaussian states have non - zero gaussian discord @xcite and non - zero ( actual ) discord @xcite . in particular , gaussian discord @xmath226 implies the presence of entanglement @xcite .
in fact , the global input state @xmath601 has cm @xmath602 .
this cm is subject to the symplectic transformation @xmath603 , where @xmath604 is the beam splitter matrix of eq .
( [ bsmatrix ] ) acting on modes @xmath2 and @xmath18 .
tracing out the environment from the output cm , we get eq .
( [ abpcm ] ) .
note that the uncertainty principle is known to be equivalent to the bona - fide condition @xmath496 and @xmath527 @xcite .
however , since @xmath496 and @xmath522 is congruent with @xmath130 , then it is automatically implied that @xmath522 has strictly - positive eigenvalues ( this is a simple consequence of sylvester s law of inertia @xcite ) . as a result
, we have that @xmath527 is equivalent to @xmath605 and , therefore , the previous bona - fide condition can also be stated as in eq .
( [ bfnew ] ) .
note that eq .
( [ gcon3 ] ) does not imply eqs .
( [ gcon1 ] ) and ( [ gcon2 ] ) .
for instance , if we set @xmath606 with @xmath607 , then eq . ( [ gcon3 ] )
becomes @xmath608 which is always satisfied .
in other words , eq .
( [ gcon3 ] ) alone is not sufficient to guarantee the bona - fide of the cm . minor remark on matrix notation .
we use @xmath609 when we refer to kets @xmath610 and bras @xmath611 .
we prefer to use @xmath612 when we distinguish between system @xmath1 ( @xmath613 ) and system @xmath2 ( @xmath614 ) .
clearly , we have @xmath615 . | combining two entanglement - breaking channels into a correlated - noise environment restores the distribution of entanglement .
surprisingly , this reactivation can be induced by the injection of separable correlations from the composite environment . in any dimension ( finite or infinite )
, we can construct classically - correlated twirling environments which are entanglement - breaking in the transmission of single systems but entanglement - preserving when two systems are transmitted . here
entanglement is simply preserved by the existence of decoherence - free subspaces .
remarkably , even when such subspaces do not exist , a fraction of the input entanglement can still be distributed .
this is found in separable gaussian environments , where distillable entanglement is able to survive the two - mode transmission , despite being broken in any single - mode transmission by the strong thermal noise . in the gaussian setting , entanglement restoration is a threshold process , occurring only after a critical amount of correlations has been injected .
such findings suggest new perspectives for distributing entanglement in realistic environments with extreme decoherence , identifying separable correlations and classical memory effects as physical resources for breaking entanglement - breaking . |
we consider the gross - pitaevskii equation for @xmath0 @xmath1 with the boundary condition @xmath2 this equation is a dynamical model for superfluids and bose - einstein condensates , and has been extensively studied , especially concerning traveling wave solutions of the form @xmath3 , and dynamics of vortices ( zeros of @xmath4 ) .
see [ 210 , 1216 , 2025 , 27 , 28 ] and references therein .
however we know very little about long - time dynamics of general solutions , for example about stability of vortices and traveling waves , or even of the constant solution @xmath5 .
heuristically , the main difficulty is that small perturbations can continue to interact with the non - zero background @xmath6 , and so do not easily disperse and decay .
thus we started in @xcite an investigation of large - time behavior of solutions @xmath7small " as a first step toward understanding dispersive processes in this equation .
the perturbation @xmath8 from the equilibrium satisfies the equation @xmath9 the conserved energy and charge are written respectively @xmath10 for @xmath11 and @xmath12 , unique global existence for the cauchy problem has been proved first in @xcite for @xmath13 , and later in @xcite for any finite energy solution . in order to investigate dispersive properties of small solutions @xmath14 ,
it is natural to linearize the equation around @xmath15 .
the left hand side of can be made complex linear by the following change of variable : @xmath16 then the new function @xmath17 satisfies the equation @xmath18 the linear evolution @xmath19 is expected to approximate small perturbations from the equilibrium .
we have proved in the previous paper @xcite that this is indeed the case if @xmath20 for small solutions @xmath21 . in this paper
we turn to the physical dimensions @xmath22 , where the dispersion becomes much weaker .
actually there exist traveling wave solutions with arbitrarily small energy for @xmath11 @xcite , so it seems unlikely that the same result holds as in @xmath20 .
however , it is still possible that small solutions @xmath14 disperse in general if they are well localized in space at some time , since those traveling waves are spatially spread - out and do not belong to @xmath23 @xcite .
our theorems [ thm:3d ] , [ thm:2d ] show that at least there exist plenty of global dispersive solutions even for @xmath22 .
since is rather complicated , it seems natural to compare it with a simpler nonlinear schrdinger equation ( nls ) with the same nonlinearity : @xmath24 the scaling argument suggests that the quadratic nonlinear terms can be treated as a perturbation in @xmath25 only if @xmath26 .
hence we are led to work in weighted spaces , which provides more decay in time .
even with the optimal decay rate of the free evolution , the quadratic terms exhibit in general the critical decay order @xmath27 in @xmath28 if @xmath11 . for @xmath29
, we can generally expect asymptotically free behavior for dispersive solutions .
coming back to our equation , the operator @xmath30 has a singularity at 0 frequency @xmath31 similar to the wave equation , which is worse for time decay than the schrdinger equation .
in addition , apparently contains a singularity due to @xmath32 in the nonlinearity .
nevertheless , it turns out that our equation is better than the nls , and in fact our argument for @xmath11 does not apply to the latter , which appears simpler at first glance .
this is because has a special structure and thereby we can transform it to another equation with a derivative nonlinearity , in effect .
we give the details of the transform in section [ ss : normal ] . before stating our results , we review the known results on the nls for comparison .
dispersive global solutions have been constructed for the quadratic nls in @xmath11 only in the following two cases : 1 .
@xmath33 , ( @xmath34 ) , @xcite [ case1 ] 2 .
@xmath35 ^ 2 $ ] , ( @xmath36 ) , @xcite in the first case , the asymptotic profile is modified from the free evolution by a phase factor which depends only on @xmath37 , while in the second case it is modified by the addition of a term with @xmath23 scaling concentration at @xmath31 .
each argument depends essentially on the form of the modification , and so it seems difficult to combine these results to cover our @xmath38 .
for example when @xmath39 , it is only known that there are no asymptotically free solutions in the @xmath23 sense with the natural decay property @xcite . for @xmath29 , it is known @xcite that small initial data in certain weighted spaces lead to asymptotically free solutions in the above case [ case1 ] .
the final data problem ( or construction of the wave operators ) is easier and arbitrary quadratic terms can be treated in the same way as in section [ s:3d ] .
now we state the main results in this paper . @xmath40 and @xmath41 denote the inhomogeneous and homogeneous sobolev spaces respectively ( cf .
@xcite ) , and we omit the subscript when @xmath42 .
denote by @xmath43 the homogeneous besov spaces . in three dimensions
, we have wave operators without size restriction . [ thm:3d ]
let @xmath29 and @xmath44 be small ( @xmath45 is sufficient ) .
for any @xmath46 , we define a banach space @xmath47 by the following norm @xmath48 where @xmath49 and @xmath50 . for any @xmath51 satisfying @xmath52 , there exists a unique global solution @xmath53 of satisfying @xmath54 the above condition on @xmath55 is satisfied if @xmath56 .
we have the same result in the critical case @xmath57 if @xmath58 is small enough .
the threshold @xmath57 is related to the scaling property of the nls with quadratic nonlinearity in @xmath29 . in two dimensions
, we can construct asymptotically free solutions for small final data .
[ thm:2d ] let @xmath11 and @xmath51 .
assume that @xmath59 for all multi - indices @xmath60 with @xmath61 , and that @xmath62 is sufficiently small .
then there exists a unique global solution @xmath53 for satisfying @xmath63 for any @xmath64 and @xmath65 , where the constants depend on @xmath66 .
the correction term @xmath67 is coming from the normal form ( see sect .
[ ss : normal ] ) .
it has a singularity at @xmath31 , which can be worse than @xmath68 , because we do not know whether our solution @xmath14 belongs to @xmath28 .
that is also the reason we describe it in terms of @xmath14 , not @xmath55 .
the correction term @xmath69 is essentially the same as in @xcite for the nls with @xmath70 , although we do not know whether it can be simplified as there , because of the singularity of our @xmath71 at @xmath31 .
it is probably not in @xmath28 in general .
however , these correction terms have no essential effect in the nonlinearity , and so they can be regarded as error terms if one does not require @xmath23 asymptotics for @xmath17 or @xmath72 . in the next subsections
, we explain our basic tools , namely the normal form and the @xmath73 decay estimate .
to eliminate the singularity at zero frequency @xmath31 , we introduced in @xcite the following transformation of normal form type : @xmath74 where @xmath75 was a fourier multiplier cutting - off the higher frequency @xmath76 . the new function @xmath77 satisfies the following equation @xmath78 , \end{split}\ ] ] where @xmath79 and @xmath80 .
it was crucial in @xcite for @xmath20 that @xmath81 is essentially of derivative form .
we also exploited the fact that the quadratic part does not contain @xmath82 in the low frequency . here
we make a new observation that a special choice of @xmath75 related to the equation leads to even better and much simpler nonlinearity .
let @xmath83 and @xmath84 .
then we have @xmath85 and @xmath86 hence the equation for @xmath87 is given by @xmath88 ds,\\ & u = vz - p\frac{|u|^2}{2},\quad z^0=e^{-iht}{\varphi } , \end{split}\ ] ] where we denote @xmath89 the new nonlinearity is roughly of the form @xmath90 .
it is vital for our analysis in @xmath11 that the quadratic terms consist only of derivatives .
we will solve the above equation for @xmath91 and for @xmath92 by the fixed point argument . then solving the equation for @xmath14 from @xmath93 by the result in @xcite ( or by @xcite for @xmath29 ) , and using local uniqueness of @xmath91 satisfying @xmath94 ds,\quad u = vz - p\frac{|u|^2}{2 } , \end{split}\ ] ]
we can deduce that our solution @xmath95 satisfies the gross - pitaevskii equation and extends globally in time .
we recall the linear decay estimate proved in @xcite .
we call the pair of exponents @xmath96 admissible if @xmath97 , @xmath98 and @xmath99 .
we denote by @xmath100 the hlder conjugate .
[ th:4 - 2 ] let @xmath101 .
( i ) let @xmath102 and @xmath103 .
then we have @xmath104 ( ii ) let @xmath96 , @xmath105 and @xmath106 be admissible .
then we have @xmath107 where @xmath108 is some positive continuous function of @xmath109 , but diverges as @xmath110 when @xmath11 .
the above estimates are exactly the same as for the schrdinger evolution @xmath111 .
we had in @xcite some gain at @xmath31 for @xmath29 , but we ignore it in this paper .
the second last statement in theorem [ thm:3d ] follows from the above estimate ( i ) . for any @xmath112 and @xmath113
, we denote the full set of strichartz norms of @xmath114 solutions for @xmath115 by @xmath116 when @xmath29 , this is just @xmath117 . when @xmath11 , it is slightly bigger than @xmath118 .
we define the weighted lebesgue space @xmath119 by the following norm for any @xmath120 , @xmath112 and @xmath121 : @xmath122 we denote the mixed norm by ( where @xmath123 is a banach space ) @xmath124 the hlder inequality implies that @xmath125 we have also @xmath126 iff @xmath127 moreover we have @xmath128 the rest of this paper is organized as follows . in the section 2 , we deal with the three dimensional case , and the other sections are devoted to two dimensions . after explaining the main ideas in section 3 ,
we give the main bilinear estimate in section 4 , and then prove theorem [ thm:2d ] in section 5 .
in this section , we construct the wave operators in @xmath29 .
the nonlinear terms are estimated simply by the hlder and sobolev inequalities , and the wave operators are constructed for the equation in our normal form by the standard fixed point theorem using the linear decay estimate .
@xmath129 regularity would be sufficient for the final state problem , but we do not pursue it in this paper .
once the solution @xmath14 is constructed in @xmath130 for some large @xmath131 , it is uniquely extended to a global one by the result in @xcite .
we will construct the asymptotically free solution by the fixed point theorem in the space @xmath132 for large @xmath133 .
we start with the simpler critical case @xmath57 . by using the @xmath134 decay estimate
, we have for the quadratic term for @xmath115 , @xmath135 and for the cubic term @xmath136 the decay in @xmath137 is derived for the quadratic terms by @xmath138 and for the cubic terms by @xmath139 as for the normal form , we have @xmath140 for any @xmath141 .
thus we get the unique solution @xmath91 for by the standard fixed point argument in the space , provided that @xmath142 is sufficiently small and @xmath131 .
then the solution is extended globally by the result in @xcite , and local uniqueness of the solution @xmath91 in @xmath143 for , which follows easily from the strichartz , sobolev and hlder inequalities .
next we consider the case @xmath144 without size restriction .
we define exponents @xmath145 by @xmath146 the decay estimate implies that for @xmath115 , @xmath147 the strichartz estimate implies that @xmath148 the cubic term has additional @xmath149 decay due to the @xmath150 bound by the same argument as in
. then we use the complex interpolation @xmath151_{\theta}= { { l^{1/2-{\theta}/4-{\varepsilon}(2 - 3{\theta})}_{{\theta}/2,t}}}(h^1_{q } ) , \end{split}\ ] ] where @xmath152 $ ] should be chosen to satisfy @xmath153 i.e. , @xmath154 .
the last inequality is because @xmath155 . to embed the above space into @xmath47
, we need @xmath156 i.e. , @xmath157 , which is satisfied by the above @xmath158 with strict inequalities .
in fact we have @xmath159 therefore we get @xmath160 as a small factor for the nonlinear term in the space .
the rest of proof is the same as in the critical case .
in the rest of the paper , we deal with the case @xmath11 . in this section
we describe the outline , and derive the key estimate in the next section , then finally prove the main theorem in the last section .
let @xmath161 .
the integral equation is decomposed as follows @xmath162 ds\\ & + \int_\infty^t e^{-ih(t - s)}n^2(u^0 ) ds = : { { \it tri}}(u)+{{\it dif}}(u)+{{\it asy}}(u^0 ) \end{split}\ ] ] the first two terms are estimated by simple hlder and sobolev type inequalities , and the main task is to derive enough time decay for the last term @xmath163 , which is explicitly given by the data @xmath55 . in estimating @xmath164
, we use @xmath165 decay of @xmath166 , which forces us to assume smallness of the data @xmath55 ( this is usual in the case of critical decay ) .
we further decompose @xmath163 as follows : @xmath167 @xmath69 is the only part where the oscillation of @xmath166 is completely canceled at @xmath31
. we will derive in the next section @xmath168 notice that we have by the simple @xmath169 decay and the hlder that @xmath170 so in we are gaining roughly @xmath27 decay by losing @xmath171 at @xmath31 , which is acceptable for our nonlinearity .
the main idea of the decay estimate is as follows . for simplicity , consider the schrdinger evolution @xmath172 .
our quadratic terms are roughly of the form @xmath173 in the fourier space , where the phase function @xmath174 is given by one of @xmath175 @xmath176 corresponds to @xmath177 and @xmath178 to @xmath179 and @xmath180 .
we can gain @xmath27 by integration by parts in @xmath181 , picking up the divisor @xmath182 , where @xmath183=\mp(2{\eta}-{\xi } ) .
\end{split}\ ] ] hence the singularity of @xmath184 is canceled by @xmath185 in .
we need to integrate twice , since we want to have @xmath186 after the integration in @xmath187 .
then we get @xmath188 in the case of @xmath176 , but it is almost in @xmath189 and so ok if we allow the loss of @xmath190 . in the case of @xmath178 ,
@xmath191 depends on @xmath181 , which reflects the fact that @xmath179 and @xmath180 are oscillatory .
but now we can integrate in @xmath187 , because at the stationary point @xmath192 , the phases @xmath178 do not vanish : @xmath193 we are getting strong divisor @xmath194 , but it is still ok around the stationary point , where we have @xmath195 and the nonlinearity supplies @xmath196 decay .
since our actual symbol @xmath71 is degenerate at @xmath31 , we get a stronger singularity , where the @xmath196 gain plays a crucial role .
in addition , we should carefully compare the unbalanced radial and angular components .
the detail starts in the next section .
as seen above , we are going to have a non - stationary phase estimate for the bilinear expression with integration in space - time .
now we state the main estimate in a slightly more general setting .
let @xmath11 and @xmath197 .
assume that @xmath198 satisfies @xmath199 for all @xmath200 and some nonnegative functions @xmath201 and @xmath202 , where @xmath203 .
@xmath204 and @xmath205 are not related to @xmath206 and @xmath207 then we have @xmath208 for @xmath209 , @xmath210 , @xmath211 and @xmath212 the main part of proof is to derive precise lower bounds on the first derivative of the phase and compatible upper bounds for the higher derivatives . in doing that , we should carefully distinguish the radial and angular components , otherwise we would get too much singularity at @xmath31 . for any vectors @xmath213 , we denote @xmath214=\sqrt{2+|{\xi}|^2},\quad { \widehat}{{\xi}}=\frac{{\xi}}{|{\xi}|},\quad { \xi}_{\eta}:={\xi}\cdot{\widehat}{{\eta}},\quad { \xi}{^\perp_}{\eta}={\xi}-{\widehat}{{\eta}}{\xi}_{\eta}. \end{split}\ ] ] then the phase function @xmath71 is written as @xmath215 .
\end{split}\ ] ] we will denote @xmath216 , etc . first we need to see that the above lemma applies to @xmath163 .
its fourier transform is a linear combination of the form @xmath217 with @xmath218^{-1}[{\eta}-{\xi}]^{-1 } & ( \text{for } u_1 ^ 2),\\ { \widehat}{{\xi}}\cdot|{\eta}|({\eta}-{\xi})[{\xi}]^{-1}[{\eta}]^{-1 } & ( \text{for } pu^{-1}{\nabla}\cdot(u_1{\nabla}u_2 ) ) , \end{cases}\\ & g({\xi},{\eta}):=\begin{cases } { \mathcal{f}}{\varphi}({\xi}-{\eta}){\overline}{{\mathcal{f}}{\varphi}}(-{\eta } ) & ( \phi=\phi_0=h({\xi})+h({\eta})-h({\eta}-{\xi})),\\ { \mathcal{f}}{\varphi}({\xi}-{\eta}){\mathcal{f}}{\varphi}({\eta } ) & ( \phi=\phi_+=h({\xi})-h({\eta})-h({\eta}-{\xi})),\\ { \overline}{{\mathcal{f}}{\varphi}}(-{\xi}+{\eta}){\overline}{{\mathcal{f}}{\varphi}}(-{\eta } ) & ( \phi=\phi_-=h({\xi})+h({\eta})+h({\eta}-{\xi } ) ) .
\end{cases } \end{split}\ ] ] then in all six cases , our assumption implies the first condition of with @xmath219 and some nonnegative functions @xmath201 determined by @xmath55 , and the second one follows from the assumption @xmath51 .
in addition , we observe that @xmath220 does not contain the terms with @xmath221 . by symmetry
, we will mainly restrict our attention to the region where @xmath222 since we are going to integrate by parts twice , we need up to the third derivatives of the phases
. let @xmath223 .
explicit computations give us @xmath224}},\quad
h''(r ) = \frac{2r(3+r^2)}{{[r]}^3},\quad
h'''(r ) = \frac{12}{{[r]}^5},\\ & h''''(r ) = -\frac{60r}{{[r]}^7},\quad i(r ) = -\frac{4}{r^2[r]^3},\quad i'(r)=\frac{4(4 + 5r^2)}{r^3{[r]}^5 } , \end{split}\ ] ] as for the differences , we have for any @xmath225 , @xmath226 for any vector @xmath17 , we denote the partial derivative with respect to @xmath181 in the direction @xmath17 by @xmath227
. we will omit the estimate with @xmath228 , which is easier than that with @xmath229 .
for the phases @xmath176 and @xmath229 , and for any vectors @xmath230 , we have @xmath231 where the upper and lower signs correspond to @xmath176 and @xmath229 respectively , and @xmath232 denotes the symmetric 3-tensor defined by @xmath233 we will use the following elementary geometry . for any @xmath234 and unit vectors @xmath235 , we have @xmath236 where @xmath237 follows just by the triangle inequality and @xmath238 follows by squaring the both sides .
for any nonzero vectors @xmath239 , we have @xmath240 hence by putting @xmath241 and @xmath242 , we have @xmath243 first we consider the phase @xmath176 , for which the integration in @xmath187 does not play any role . for a fixed @xmath244 and each @xmath245 , we split the integral region of @xmath181 into the following three overlapping domains : @xmath246 and choose a partition of unity @xmath247 satisfying @xmath248 for @xmath249 and @xmath250 such functions can be given in the form @xmath251 with some one - dimensional cut - off function @xmath252 , then its @xmath181 derivatives are given by @xmath253\frac{\chi'}{|{\xi}|}. \end{split}\ ] ] the above bounds follows from these identities , , and @xmath254 in @xmath255 . by symmetry , it suffices to estimate only in @xmath256 and @xmath257 .
we first consider @xmath256 . here
we use the polar coordinates @xmath258 , or in other words , we choose the direction @xmath259 for the partial integration . by the definition of @xmath256 , we have @xmath260 , and by , @xmath261 partial integration in @xmath262 gives @xmath263 where the operator @xmath264 is defined by @xmath265 for the phase derivatives , we have the following estimates @xmath266|{\widehat}{{\eta}}{^\perp_}{{\eta}-{\xi}}|^2\\ |{\partial}_r^2\phi_0| & { { \ \lesssim \ } } \frac{|{\eta}|-|{\eta}-{\xi}|}{{{{\langle}{\eta}{\rangle } } } } + \frac{{{{\langle}{\eta}-{\xi}{\rangle}}}}{|{\eta}-{\xi}|}(1-{\widehat}{{\eta}}_{{\eta}-{\xi } } ) { { \ \lesssim \ } } \frac{{\partial}_r\phi_0}{|{\eta}-{\xi}|},\\ { \partial}_r^3\phi_0 & = h'''({\eta})-h'''({\eta}-{\xi } ) + h'''({\eta}-{\xi})(1-{\widehat}{{\eta}}_{{\eta}-{\xi}}^3)\\ & \qquad + 3i({\eta}-{\xi}){\widehat}{{\eta}}_{{\eta}-{\xi}}|{\widehat}{{\eta}}{^\perp_}{{\eta}-{\xi}}|^2\\ |{\partial}_r^3\phi_0| & { { \ \lesssim \ } } \frac{|{\eta}|-|{\eta}-{\xi}|}{{{{\langle}{\eta}-{\xi}{\rangle}}}^5{{{\langle}{\eta}{\rangle } } } } + \frac{1-{\widehat}{{\eta}}_{{\eta}-{\xi}}}{|{\eta}-{\xi}|^2{{{\langle}{\eta}-{\xi}{\rangle}}}^3 } { { \ \lesssim \ } } \frac{{\partial}_r\phi_0}{|{\eta}-{\xi}|^2 } , \end{split}\ ] ] where we have used @xmath267 .
thus we obtain @xmath268 next we consider @xmath257 . here
the main part of @xmath269 is its angular component , which is not always close to either @xmath270 or @xmath271 , so we simply integrate in the direction of @xmath269 .
is not adequate in @xmath256 , where the angular difference terms such as @xmath272 in can not be controlled . ]
partial integration gives @xmath273 where the operator @xmath274 is defined by @xmath275 which satisfies @xmath276|f| .
\end{split}\ ] ] in the region @xmath257 , we have @xmath277 .
using together with and , we have @xmath278|{\widehat}{{\eta}}\otimes{\widehat}{{\eta}}-{\widehat}{{\eta}-{\xi}}\otimes { \widehat}{{\eta}-{\xi}}|\\ & { { \ \lesssim \ } } \frac{|{\xi}|}{|{\eta}|^2 } + \frac{{{{\langle}{\eta}{\rangle}}}}{|{\eta}|}|{\widehat}{{\eta}}-{\widehat}{{\eta}-{\xi}}|,\\ |{\nabla}^3\phi_0| & { { \ \lesssim \ } } |h'''({\eta})-h'''({\eta}-{\xi})| + |i({\eta})-i({\eta}-{\xi})|\\ & \quad + ( h'''({\eta})+|i({\eta})|)|{\widehat}{{\eta}}-{\widehat}{{\eta}-{\xi}}|\\ & { { \ \lesssim \ } } \frac{|{\xi}|}{|{\eta}|^3{{{\langle}{\eta}{\rangle } } } } + \frac{|{\widehat}{{\eta}}-{\widehat}{{\eta}-{\xi}}|}{|{\eta}|^2{{{\langle}{\eta}{\rangle}}}^3}. \end{split}\ ] ] by the definition of @xmath257 and , we have @xmath279 hence we deduce that @xmath280 and therefore @xmath281 which is slightly better than the bound in .
in conclusion , we obtain @xmath282 , \end{split}\ ] ] where we used the schwarz inequality and the condition @xmath197 for the second integral . for @xmath283 , we have also @xmath284 applying this estimate in the region @xmath285 and in the rest , using the young inequality @xmath286 , @xmath287 , and appending the same estimate in the opposite region @xmath288 , we obtain @xmath289 for @xmath290 . after integration in @xmath291 ,
this estimate corresponds to the case @xmath292 in .
the remaining case @xmath293 is covered by interpolation , see section [ ss : interp ] .
next we consider the phase @xmath229 , for which we need to take account of the time oscillation , and so split the integral into more regions .
let @xmath294 satisfy @xmath295 for @xmath296 and @xmath297 for @xmath298 , and denote @xmath299 .
hence we have @xmath300 $ ] and @xmath301 .
we also denote @xmath302 for a fixed positive @xmath303 , and each @xmath245 and @xmath304 , we introduce partitions of unity for @xmath305 by the following identities : @xmath306 and @xmath307 where @xmath308 and @xmath309 are the same as in . hence denoting by @xmath310 ,
we have ( see figure [ fig:1 ] ) @xmath311 remark that @xmath312 and @xmath313 cover the stationary - phase regions in space and time , respectively .
@xmath313 is an annular region , separating @xmath312 into two connected components ( if @xmath314 ) .
the derivatives of the cut - off functions satisfy @xmath315 for @xmath316 , where @xmath313 is identified with its characteristic function .
we can easily derive these bounds using @xmath317 together with similar estimates for the radial derivative in @xmath318 .
in @xmath319 , @xmath320 and @xmath321 , we use the polar coordinates @xmath258 and integrate twice in the radial direction . because there is no cancellation between the angular terms @xmath322 and @xmath323 in @xmath324 , and also because the cut - off function has larger derivative in the angular direction . ] since @xmath325 , we have @xmath326 in @xmath319 , we have @xmath327 , which implies that @xmath328 and also @xmath329 therefore , defining @xmath264 by with @xmath330 , we get @xmath331 since @xmath332 , we obtain by the schwarz inequality @xmath333 in @xmath320 , we have @xmath334 and by the radial component @xmath335 whereas in @xmath321 we have @xmath336 and by the angular component @xmath337 and we have in both cases .
hence we have for @xmath338 , @xmath339 for @xmath340 , this gives the same bound as in the last term in . for @xmath341 , we use the shape of @xmath313 and the polar coordinates for @xmath318 , @xmath342 where the factor @xmath343 is coming from integration in @xmath271 .
we can treat @xmath344 in the same way by symmetry .
thus we obtain for @xmath345 , @xmath346 next we exploit the time oscillation in @xmath312 , which is split into the following two cases : @xmath347 in the first case , we have @xmath348 and @xmath349+\sqrt{2 } } + \frac{|{\eta}-{\xi}|^3}{[{\eta}-{\xi}]+\sqrt{2 } } - \frac{|{\xi}|^3}{[{\xi}]+\sqrt{2 } } { { \ \gtrsim \ } } \frac{|{\xi}|^3}{{{{\langle}{\xi}{\rangle } } } } , \end{split}\ ] ] while in the second case , we have @xmath350 and @xmath351 + [ h(|{\eta}|+|{\eta}-{\xi}|)-h({\eta})-h({\eta}-{\xi})]\\ & { { \ \gtrsim \ } } \frac{|{\xi}|^3}{{{{\langle}{\xi}{\rangle } } } } , \end{split}\ ] ] where we used and the identity @xmath352}+{[a ] } } + \frac{ab(a+2b)}{{[a+b]}+{[b ] } } \quad(a , b>0 ) .
\end{split}\ ] ] therefore we can integrate in @xmath187 for @xmath245 : @xmath353 for @xmath354 .
in particular , we have @xmath355 hence by using the triangle inequality we obtain @xmath356 in @xmath357 , we use the polar coordinate @xmath358 and partial integration in @xmath359 : @xmath360 where the operator @xmath361 is defined by @xmath362 the first derivative is given by @xmath363 where the radial component can be estimated by @xmath364 and the angular component by @xmath365 to rewrite it in terms of @xmath366 , let @xmath367 and @xmath368 be the angles @xmath369 $ ] such that @xmath370 then by the sine theorem and @xmath371 by @xmath372 , we have @xmath373 and so @xmath374
thus we obtain @xmath375 the second derivative is estimated by @xmath376 then by using @xmath377 and the above estimate , we obtain @xmath378 .
\end{split}\ ] ] thus we obtain @xmath379\\ & = \frac{m^2}{{{{\langle}{\xi}{\rangle}}}^{1+{\sigma}}|{\xi}|^2{{{\langle}m { \rangle}}}^{1+{\sigma}}|{\zeta}|\left({\omega}^2+\frac{m|{\xi}|}{{{{\langle}{\xi}{\rangle}}}^2}\right)}\left[\frac{1}{m}+\frac{1}{|{\zeta}|}+\frac{d_t{{{\langle}{\xi}{\rangle}}}}{m^{1/2}|{\xi}|^{3/2}}\right ] . \end{split}\ ] ] using @xmath380 , we have for @xmath340 , @xmath381 the first term is treated as the last term of , and the second term by using the triangle and the schwarz inequalities @xmath382 for @xmath341 , we have @xmath383 and so in the polar coordinates @xmath384 , @xmath385 thus we obtain @xmath386 we have the same bound for @xmath387 by symmetry . in @xmath388
, we use the operator @xmath274 defined by with @xmath330 . here
we have @xmath195 and @xmath389 , which implies by the same argument as in that @xmath390 hence we have @xmath391 and so @xmath392 { { \ \lesssim \ } } \frac{1}{{{{\langle}{\xi}{\rangle}}}^{2 + 2{\sigma}}|{\zeta}|}\left[\frac{1}{|{\zeta}| } + \frac{{{{\langle}{\xi}{\rangle}}}d_t}{|{\xi}|^2}\right ] . \end{split}\ ] ] this bound is better than , and so in the same way we obtain @xmath393 on the other hand , we have for @xmath394 , @xmath395 putting the pieces together , we obtain @xmath396 here the condition @xmath397 is inevitable for @xmath229 . the difference from the case of @xmath176
is the inbalance between the @xmath185 from @xmath398 and the @xmath399 from the partial integration , which costs one regularity . on the other hand , @xmath229 is better at @xmath31 than @xmath176 , which is non - oscillatory at @xmath31 .
thus we have proved in the case @xmath292 . on the other hand , we have for any @xmath400 , @xmath401 which implies that for any @xmath65 , @xmath402 this corresponds to the case @xmath403 in , although it is divergent for the integral @xmath404 . by applying real interpolation or the hlder inequality to dyadic sequences on @xmath185 and @xmath291
, we obtain the desired estimate for @xmath209 .
we decompose our solution @xmath405 and @xmath406 , where @xmath407 and starting from the above estimates , we will derive @xmath408 with some @xmath235 satisfying @xmath409 by the standard iteration argument .
@xmath410 can be arbitrarily close , but not equal , to @xmath411 .
we fix @xmath412 such that @xmath413 let @xmath414 .
the @xmath73 decay implies the following bounds on the free part @xmath415 : @xmath416 the last quantity is finite for high frequency by interpolation of @xmath417 and @xmath418 .
the low frequency part is also finite , because implies that @xmath419 for all @xmath420 and @xmath421 , which is proved simply by estimating the inverse fourier transform . in the following three subsections ,
we derive estimates on the normal form , the trilinear terms and the quadratic difference terms , where we need not assume that @xmath14 is the solution .
for any function @xmath14 , we denote @xmath422 remark that @xmath423 is not a norm , but it is designed to measure different types of decay of @xmath166 and @xmath424 , namely dispersive and dissipative . since @xmath425 , we have @xmath426 the quadratic part is estimated just by the hlder inequality : @xmath427 for @xmath428 , we have @xmath429 we apply this estimate after expanding @xmath406 , choosing different @xmath109 for each term . for example , @xmath430 is expanded into the following spaces @xmath431 where we used the interpolation inequalities of gagliardo - nirenberg type : @xmath432 the other terms containing @xmath433 are estimated in the same way .
thus we obtain @xmath434 ^ 2\|u - w\|_{{\mathcal{z}}_t'}. \end{split}\ ] ] the quadratic difference term can be expanded by putting @xmath406 @xmath435 and each term is estimated in @xmath436 by using @xmath437 where @xmath438 is the small factor coming from @xmath62 , and we choose @xmath439 and @xmath440 such that @xmath441 thus we obtain @xmath442 \|u\|_{{\mathcal{z}}_t},\\ & \|{{\it dif}}(u)-{{\it dif}}(w)\|_{stz_t^1 } { { \ \lesssim \ } } t^{-{\alpha } } \left[{\delta}+t^{-{\kappa}}(\|u\|_{{\mathcal{z}}_t}+\|w\|_{{\mathcal{z}}_t})\right ] \|u - w\|_{{\mathcal{z}}_t'}. \end{split}\ ] ] we define an iteration sequence @xmath443 for @xmath444 and @xmath92 , by @xmath445 we introduce the following norm for @xmath446 : @xmath447 since @xmath448 , we have @xmath449 . using the sobolev embedding @xmath450 and @xmath448
, we have for any @xmath451 , @xmath452 and then the quadratic part is estimated by using and @xmath453 , @xmath454\|u_{(k)}-u_{(j)}\|_{{\mathcal{z}}_t'}. \end{split}\ ] ] for the first iteration , we have @xmath455 , and so @xmath456 as for @xmath457 , we use and , deriving @xmath458\\ & \times \|u_{(k)}-u_{(j)}\|_{{\mathcal{z}}_t'}. \end{split}\ ] ] for the first iteration , we apply and .
then we get @xmath459 , \end{split}\ ] ] where we denote @xmath460 gathering the above estimates , we deduce that @xmath461,\\ & e_k { { \ \lesssim \ } } \sum_{j=1}^{k-1 } d_j + \|u_{(1)}\|_{{\mathcal{z}}_t},\quad \|u_{(1)}\|_{{\mathcal{z}}_t } { { \ \lesssim \ } } \|u^0\|_{{\mathcal{z}}_t}(1+t^{-{\kappa}}\|u^0\|_{{\mathcal{z}}_t } ) .
\end{split}\ ] ] hence for sufficiently small @xmath438 and large @xmath462 , @xmath463 converges to some function @xmath91 satisfying the equation and @xmath464 the uniqueness for @xmath115 is proved also by the above difference estimates .
in addition , implies that @xmath465 where we used the fact that @xmath466 does not contain @xmath176 . by using this estimate together with and the @xmath73 decay
, we also obtain @xmath467 for any small @xmath144 .
in particular , @xmath468 for @xmath469 .
the local uniqueness for in this class is easily derived from the strichartz , sobolev and hlder inequalities .
the final task is to extend our solution @xmath14 to @xmath470 .
we can not apply the @xmath471 global wellposedness of @xmath14 by @xcite , since @xmath163 barely falls out of @xmath23 .
however the nonlinear energy is still finite , because @xmath472 , and the @xmath23 singular part at low frequency belongs to @xmath473 .
hence we can apply the global existence results in @xcite , and we have only to see persistence of our function space , namely @xmath474 .
the conserved energy can be written as @xmath475 where we denote @xmath476 .
* theorem 1.1 ) gives @xmath477 hence we have @xmath478 for @xmath469 , which implies for @xmath479 that @xmath480 on the other hand , by using the identities @xmath481 we get @xmath482 and the integral equation for @xmath446 can be written as @xmath483 ds , \end{split}\ ] ] where the nonlinearity is in @xmath484 .
hence the strichartz estimate implies that @xmath485 and therefore @xmath486 for @xmath469 .
combined with , this implies that @xmath487 for @xmath469 .
the research of gustafson and tsai is partly supported by nserc grants .
the research of nakanishi was partly supported by the jsps grant no .
15740086 .
f. bethuel and j. c. saut , _ vortices and sound waves for the gross - pitaevskii equation _ nonlinear pde s in condensed matter and reactive flows , 339354 , nato sci . ser .
, bf 569 kluwer acad .
publ . , dordrecht , 2002 .
f. bethuel , g. orlandi , d. smets , _ vortex rings for the gross - pitaevskii equation_. j. euro .
* 6 * ( 2004 ) , no .
1 , 1794 . c. a. jones , s. j. putterman and p. h. roberts , _ motions in a bose condensate : v. stability of solitary wave solutions of non - linear schrdinger equations in two and three dimensions _ , j. phys .
a : math . gen .
* 19 * ( 1986 ) 29913011 . | we study asymptotic behaviour at time infinity of solutions close to the non - zero constant equilibrium for the gross - pitaevskii equation in two and three spatial dimensions .
we construct a class of global solutions with prescribed dispersive asymptotic behavior , which is given in terms of the linearized evolution . |
and @xmath18 surfaces for a model @xmath77-wave superconductor .
the first bz and one extended zone are drawn .
the solid black curves denote the fermi surface and the shaded region is filled in the normal state .
red arrows are the unit vectors in eq .
( [ n ] ) .
the blue dots with @xmath12 signs represent nodes with vorticity @xmath78 respectively .
top left : the slanted thin black line segment is the surface bz of the @xmath20 edge .
the black arrow with letter `` p '' indicates the direction of projection .
thick red line segments mark the surface momenta with zero energy bound states .
top right : the thin black horizontal line segment is the bz of the @xmath18 edge .
the @xmath12 vorticity nodes overlap after projection .
the vorticity of the node enclosed by the parallelogram is equal to the winding number difference along the two side blue segments because the winding along the top and bottom segments cancel due to the periodicity in momentum space .
this can be explicitly seen by following the turning of the red arrows ( the actual winding number is twice of the winding shown by the arrows , due to the spin degeneracy ) .
( b ) the ( 11 ) edge bandstructure . the surface flat bands are marked red . in constructing this figure we have used @xmath66 and @xmath79 in eq .
( [ dwave ] ) . here
@xmath80.,title="fig : " ] and @xmath18 surfaces for a model @xmath77-wave superconductor .
the first bz and one extended zone are drawn .
the solid black curves denote the fermi surface and the shaded region is filled in the normal state .
red arrows are the unit vectors in eq .
( [ n ] ) .
the blue dots with @xmath12 signs represent nodes with vorticity @xmath78 respectively .
top left : the slanted thin black line segment is the surface bz of the @xmath20 edge .
the black arrow with letter `` p '' indicates the direction of projection .
thick red line segments mark the surface momenta with zero energy bound states .
top right : the thin black horizontal line segment is the bz of the @xmath18 edge .
the @xmath12 vorticity nodes overlap after projection .
the vorticity of the node enclosed by the parallelogram is equal to the winding number difference along the two side blue segments because the winding along the top and bottom segments cancel due to the periodicity in momentum space .
this can be explicitly seen by following the turning of the red arrows ( the actual winding number is twice of the winding shown by the arrows , due to the spin degeneracy ) .
( b ) the ( 11 ) edge bandstructure . the surface flat bands are marked red . in constructing this figure we have used @xmath66 and @xmath79 in eq .
( [ dwave ] ) . here
@xmath80.,title="fig : " ] the idea behind the criterion presented in the main text is best illustrated by using the cuprate superconductor as an example .
the bogoliubov - de gennes ( bdg ) hamiltonian of the cuprate superconductor read h_cuprates()=()_0_3+()_0_1,[dwave ] where @xmath81 is the identity matrix acting in the spin space and @xmath82 are @xmath83 pauli matrices in the nambu space , @xmath25 is the normal state dispersion satisfying @xmath84 and @xmath85 is the d - wave gap function .
( since the cuprates are quasi two dimensional materials , we shall use two dimensional notations in the following discussions . ) the fermi surface and the gap nodes are shown in fig .
[ fig : d - wave]a , therefore @xmath86 . in the same figure , the normalized vector ( )=((),())/[n]is plotted as a function of @xmath7 over two bzs ( see red arrows ) . inspecting these arrows
one notices each node is a `` vortex '' in @xmath87 . around each vortex
the arrows exhibit non - zero winding . the total winding number associated with each node is given by w=22dp .
[ w ] ( the extra factor of 2 is due to spin degeneracy ) .
clearly each node is characterized by an even integer winding number .
the bdg hamiltonian defined on all one ( @xmath88 ) dimensional loop enclosing the node are topologically nontrivial .
now consider the bandstructure projected along the @xmath20 direction .
for each transverse momentum @xmath61 along @xmath19 we have a 1d chain running in @xmath20 .
so long as @xmath61 does not coincide with the projection of the nodes the spectrum is fully gapped and characterized by the winding number defined in eq .
( [ w ] ) .
any two chains whose @xmath61 straddle the projection of a node their winding numbers must differ by @xmath78 ( see fig .
[ fig : d - wave]a captions ) , hence at least one of them is topologically non - trivial and possess @xmath63 end states when the boundary condition along @xmath20 changes from closed to open .
this implies e=0 bound states exists for _ intervals _ of @xmath61 .
therefore @xmath89 is indeed @xmath90 .
an example of the ( 11 ) boundary bandstructure is shown in fig . [ fig : d - wave]b .
the @xmath61 intervals showing the flat bands are represented by the thick red line segments in the top left corner of fig .
[ fig : d - wave]a .
the only edges which do not possess zbabs are the @xmath29 ( miller s notation is used ) edges where the projection of positive and negative nodes overlap ( see top right corner of fig .
[ fig : d - wave]a ) . for the real material @xmath91 and
the only modification is @xmath89 changes from 1 to 2 .
y. zhang , z. r. ye , q. q. ge , f. chen , juan jiang , m. xu , b. p. xie , and d. l. feng , nature phys . *
8 * , ( 2012 ) .
j. k. dong , s. y. zhou , t. y. guan , h. zhang , y. f. dai , x. qiu , x. f. wang , y. he , x. h. chen , and s. y. li , phys .
* 104 * , 087005 ( 2010 ) . j .- ph .
reid , m. a. tanatar , a. juneau - fecteau , r. t. gordon , s. rene de cotret , n. doiron - leyraud , t. saito , h. fukazawa , y. kohori , k. kihou , c. h. lee , a. iyo , h. eisaki , r. prozorov , louis taillefer , arxiv:1201.3376 ( 2012 ) . | in the past few years materials with protected gapless surface ( edge ) states have risen to the central stage of condensed matter physics .
almost all discussions centered around topological insulators and superconductors , which possess full quasiparticle gaps in the bulk . in this paper we argue systems with _
topological stable _ bulk nodes offer another class of materials with robust gapless surface states . moreover
the location of the bulk nodes determines the miller index of the surfaces that show ( or not show ) such states . measuring the spectroscopic signature of these zero modes allows a phase - sensitive determination of the nodal structures of unconventional superconductors when other phase - sensitive techniques are not applicable .
we apply this idea to gapless iron - based superconductors and show how to distinguish accidental from symmetry dictated nodes .
we shall argue the same idea leads to a method for detecting a class of the elusive spin liquids .
the interest in topological insulators and superconductors lie mainly in studying their robust gapless surface states@xcite .
systems with bulk gap nodes are usually neglected for it is generally felt bulk nodes render surface states ill defined .
gap nodes in the form of band crossing are singularities in the electronic structure .
examples include the dirac nodes in graphene , the d - wave superconducting gap nodes of cuprates and the weyl nodes@xcite .
these singularities have dramatic effects on the boundary electronic structure .
for example , the zigzag edge of graphene has zero energy ( e=0 ) flat bands@xcite , and the \{110 } surfaces of the cuprate superconductors have zero bias andreev bound states ( zbabs)@xcite .
recently it has been shown that the weyl nodes will give rise to surface `` fermi arcs''@xcite .
we shall argue the above examples are special cases of a more general fact , namely translational invariant systems with _ topologically stable _ bulk nodes also possess protected gapless surface bound states . in the presence of disorder the otherwise degenerate zero energy band broadens into finite width , but does not lose the signature of gapless surface bound state .
an important class of `` nodal material '' is nodal superconductor .
usually , the presence of gap nodes reflects the overall repulsive nature of the ( effective ) cooper pairing hamiltonian .
the location of the gap nodes provide detailed information about the momentum dependence of the pairing interaction .
this information is necessary for the understanding of the pairing mechanism of nodal , hence unconventional , superconductors . among these superconductors
the pairing symmetry of iron based superconductors ( fescs ) is still very much under debate .
experimentally roughly half of the materials in this family show the evidence of gap nodes .
examples include lafepo@xcite , bafe@xmath0(as@xmath1p@xmath2)@xmath0@xcite , kfe@xmath0as@xmath0@xcite , lifep@xcite etc .
the nodal structures of the majority in this group are unknown .
the notable exception is bafe@xmath0(as@xmath1p@xmath2)@xmath0 whose node locations have been determined by angle - resolved photoemission spectroscopy@xcite . because of the complex fermiology and the likelihood of `` accidental nodes '' these materials are not accessible to conventional phase - sensitive probes .
finding experimental means to pin down the location of the nodes for these materials has been a major experimental challenge . in the following
we propose to study the correlation between the orientation of the surfaces with the presence / absence of zero energy peaks in surface sensitive spectroscopy .
we show this method does not suffer the above difficulty .
we will illustrate the idea by focusing on lafepo . near the end
we shall briefly discuss gapless non - centrosymmetric superconductors , and generalize the idea presented here for the detection of a class of spin liquids . * bulk nodes and zero energy surface bound states . *
the gap nodes of a d - dimensional quasiparticle hamiltonian can have non - zero dimension @xmath3 . for examples @xmath4 for point node and @xmath5 for line node etc .
a prerequisite of a stable nodal structure is in @xmath6 dimension the symmetry of the hamiltonian@xcite _ at a generic @xmath7 point _ admits topological insulator / superconductors@xcite .
_ if a nodal structure is stable _ , and if @xmath8 , then suitably oriented boundaries of this material will have e=0 bound state .
generically the dimension of the sub boundary bz possessing e=0 bound states is @xmath9 . for graphene @xmath10 , @xmath4 ; for cuprates @xmath11 , @xmath5 ; and for the weyl semimetal@xcite @xmath11 , @xmath4 . for the purpose of this paper
we focus on the cases where the band topology of the @xmath6 dimensional system is classified by the group of integers@xcite . under this condition a winding number@xcite(or vorticity ) can be assigned to each independent nodal manifold ( e.g. point or line ) .
_ as long as the projection of opposite - winding - number nodal manifolds does not completely overlap in the boundary brillouin zone ( bz ) , there will be gapless surface bound states .
_ special cases of the above argument have already been put forward in ref .
@xcite , especially in volovik et al.@xcite .
a more illustrative example is given in appendix [ app : example ] .
+ signs indicate nodes with @xmath12 sign of winding numbers .
( a , b ) two different @xmath13-wave gap nodes ( red dots ) , ( c ) @xmath14 nodes and ( d ) @xmath15 nodes . the black circle in part ( a ) is a close loop enclosing a gap nodes.,title="fig : " ] signs indicate nodes with @xmath12 sign of winding numbers .
( a , b ) two different @xmath13-wave gap nodes ( red dots ) , ( c ) @xmath14 nodes and ( d ) @xmath15 nodes . the black circle in part ( a ) is a close loop enclosing a gap nodes.,title="fig : " ] + signs indicate nodes with @xmath12 sign of winding numbers .
( a , b ) two different @xmath13-wave gap nodes ( red dots ) , ( c ) @xmath14 nodes and ( d ) @xmath15 nodes . the black circle in part ( a ) is a close loop enclosing a gap nodes.,title="fig : " ] signs indicate nodes with @xmath12 sign of winding numbers .
( a , b ) two different @xmath13-wave gap nodes ( red dots ) , ( c ) @xmath14 nodes and ( d ) @xmath15 nodes . the black circle in part ( a ) is a close loop enclosing a gap nodes.,title="fig : " ] * application to lafepo . * lafepo is a stoichiometric compound , it is clean enough for quantum oscillation to be observed@xcite . according to ref .
@xcite lafepo has nearly two dimensional fermi surfaces : two hole pockets at the center and two electron pockets on the face of the unfolded bz ( for a caricature see fig .
[ fig1]a ) . in the following we shall used 2d notations .
lafepo is the first fesc where strong evidence of line nodes have been reported@xcite . at the present time
there is no direct information about the location of the line nodes .
in addition , early knight shift measurement has revealed a rather puzzling result . unlike conventional singlet - paired superconductors , the knight shift increases rather than decreases below @xmath16@xcite .
this unusual behavior calls for the examination of possible nodal triplet pairing .
theoretically it has been proposed that the _ singlet _ gap function of lafepo possesses the so - called `` accidental nodes''@xcite where the nodal gap function transforms without any sign change under all point group operations . a schematic illustration of the proposed nodal structure is shown in fig . [ fig1]a .
first let s discuss the stability of these accidental nodes .
consider a loop encircling one of nodes ( e.g. the black circle in fig .
[ fig1]a ) . like the cuprates
the bdg hamiltonian defined on the loop is characterized by an even integer winding number , and has a fully gapped spectrum .
thus the loop has a non - trivial band topology .
if the node is removed from the interior of the circle the winding number must change , which requires the spectrum on the circle to become gapless , i.e. the node must pass through the circle .
of course the above statement applies to _ every _ loop enclosing the node .
this leads to the `` vorticity conservation law '' - a _ single _ node can be displaced but not annihilated .
annihilation can occur only when two opposite vorticity nodes hit each other . therefore although `` accidental '' these nodes are stable against small perturbations . because the opposite vorticity nodes distribute symmetrically around the @xmath17 , @xmath18 and @xmath19 , @xmath20 axes in fig .
[ fig1]a , there will be no zbabs on the edges .
this is in sharp contrast to the d@xmath21 nodes ( fig .
[ fig1]c ) which possess zbabs on the @xmath19 and @xmath20 edges , as well as the d@xmath22 nodes which have zbabs on the @xmath17 and @xmath18 edges . the lowest index edge that does show zbabs are the @xmath23 edges .
( the corresponding orientation of the real surfaces can be obtained by adding an extra zero , e.g. , @xmath24 etc . ) using the tight - binding fit of lda bandstructure adapted from ref .
@xcite and the following bdg hamiltonian h_lafepo()=()_0_3+()i_55_0_1,[lfp ] we calculated the surface local density of states for a number of different _ singlet _ pairing symmetries and surface orientations . in eq .
( [ lfp ] ) @xmath25 is the @xmath26 bandstructure matrix , @xmath27 and @xmath28 is the identity matrix in the orbital space .
( the calculation uses the iterative green s function method of ref .
@xcite . )
a few examples of the results for nodal singlet gap function are shown in fig .
[ fig2]a - d .
of course the @xmath13-wave accidental nodes can fall on the central hole pockets too , an example is shown in fig . [ fig1]b .
( not dictated by symmetry , there is no reason for accidental nodes to occur on both hole fermi surfaces . ) due to the reflection symmetry about the @xmath29 and @xmath30 axes , the projection of the opposite vorticity nodes coincide , hence there will be no zbabs on those surfaces .
in contrast there is no reflection symmetry about the @xmath23 directions hence generically these edges will have zbabs .
this argument holds when the accidental nodes exist on both hole pockets and/or on the electron pockets . for some of the pairing symmetries in table [ t1 ] . for each symmetry
two surface orientations are shown : one with zero bias peak and the other without .
the unit of the density of states is arbitrary .
the @xmath31 used in this calculation is @xmath32ev .
thermal broadening with temperature @xmath33mev is used .
, title="fig : " ] for some of the pairing symmetries in table [ t1 ] . for each symmetry
two surface orientations are shown : one with zero bias peak and the other without .
the unit of the density of states is arbitrary .
the @xmath31 used in this calculation is @xmath32ev .
thermal broadening with temperature @xmath33mev is used .
, title="fig : " ] for some of the pairing symmetries in table [ t1 ] . for each symmetry
two surface orientations are shown : one with zero bias peak and the other without .
the unit of the density of states is arbitrary .
the @xmath31 used in this calculation is @xmath32ev .
thermal broadening with temperature @xmath33mev is used .
, title="fig : " ] for some of the pairing symmetries in table [ t1 ] . for each symmetry
two surface orientations are shown : one with zero bias peak and the other without .
the unit of the density of states is arbitrary .
the @xmath31 used in this calculation is @xmath32ev .
thermal broadening with temperature @xmath33mev is used .
, title="fig : " ] for some of the pairing symmetries in table [ t1 ] . for each symmetry
two surface orientations are shown : one with zero bias peak and the other without .
the unit of the density of states is arbitrary .
the @xmath31 used in this calculation is @xmath32ev .
thermal broadening with temperature @xmath33mev is used .
, title="fig : " ] for some of the pairing symmetries in table [ t1 ] . for each symmetry two surface orientations
are shown : one with zero bias peak and the other without .
the unit of the density of states is arbitrary .
the @xmath31 used in this calculation is @xmath32ev .
thermal broadening with temperature @xmath33mev is used .
, title="fig : " ] because the unusual temperature dependence of the knight shift we have also considered triplet pairing .
the bdg hamiltonian for this case is given by h^_lafepo()=()_0_3+i_55()_1 , [ equ : p - wave ] where @xmath34 with @xmath35 .
in general if there is no further restriction on @xmath36 the node should not be topologically stable .
this is because for a loop consists of generic momenta around a node , the mapping from @xmath7 to @xmath37 ( @xmath38 ) has the homotopy group @xmath39 . the situation can change if we impose further constraints on the form of @xmath36 . for example , ignoring the spin - orbit interaction , if we require @xmath40 where @xmath41 is a partner function of the point group irreducible representation , the node becomes stable , and is characterized by the winding number of the aiii class@xcite . under such condition
we expect boundary zero energy states . for simplicity we considered only the following @xmath42-wave pairings with @xmath43 ( @xmath15 pairing ) , @xmath44 ( @xmath45 pairing ) and @xmath46 ( @xmath47 pairing ) .
in table [ t1 ] we summarize the results for eight different time reversal symmetric nodal pairing scenarios .
in addition to the four triplet gap functions discussed above we have also studied nodal @xmath13 , @xmath48 ; @xmath14 , @xmath49 and @xmath50 , @xmath51 ( here @xmath52ev ) .
moreover we have also considered the horizontal ring nodes observed in ref .
@xcite for bafe@xmath0(as@xmath1p@xmath2)@xmath0 .
surface orientation wise we considered @xmath53 , @xmath54 , @xmath55 , @xmath56 and @xmath57 . since there are no identical rows , by studying the dependence of the zero - bias peak on the surface orientation one can differentiate different types of nodal pairing symmetries .
.``@xmath58 '' denotes the presence of zbabs and `` @xmath59 '' the absence of it .
the pairing symmetries are defined in main text . [ cols="^,^,^,^,^,^",options="header " , ] [ t1 ] although our discussion is in the context of lafepo , our results should apply to other iron - based superconductors with the same lattice symmetry and time - reversal symmetry as well . for example
the proposed @xmath14-wave pairing in kfe@xmath0as@xmath0@xcite should have the same surface orientation dependence of zbabs as the cuprates .
* effects of disorder . *
a natural question one might ask in reading this work is how protected these e=0 surface states really are . because @xmath60 , the second pauli matrix in the nambu space , anticommutes with all hamiltonians
we have discussed , the e=0 states can be made eigenstates of it .
it can be shown that for a fixed winding number the e=0 surface states with different @xmath60 eigenvalues localize at the opposite ends of the sample . reversing the winding number exchanges these localized states .
thus local disorder potential can only mix e=0 states at surface momenta where _ opposite _ winding numbers occur .
it turns out time reversal symmetry requires the 1d winding number to change sign from surface momentum @xmath61 to @xmath62 . as the result
there are equal , opposite , intervals in the boundary bz with opposite winding numbers .
smooth surface potentials do not mix them but sharp impurity potentials do . in the presence of the mixing the delta function density of states broadens into a peak centered around @xmath63 .
the width of the peak is proportional to the disorder strength .
therefore although the singular density of states at e=0 is not protected , a broadened version of it is .
interestingly the @xmath64-wave pairing studied in conjunction with lafepo evades disorder broadening hence is strictly protected .
this is because opposite winding number occurs for opposite spin .
as long as the disorder potential does not flip spins the @xmath63 singularity is unaffected .
of course the ability to see the zero bias conductance peak@xcite and its surface orientation dependence@xcite in real experiment on cuprates , where disorder should mix zbabs , attests for the robustness of the phenomenon discussed here
. * nodal non - centrosymmetric superconductors . *
another interesting application of the idea presented here is to nodal non - centrosymmetric superconductors@xcite . in these systems due to the spin - orbit interaction the fermi surfaces
are spin non - degenerate .
this is shown by the black and gray curves in fig .
[ fig3]a .
several known non - centrosymmetric superconductors exist near an antiferromagnetic phase .
these include cept@xmath65si , cerhsi@xmath65 , ceirsi@xmath65 and cecoge@xmath65@xcite . for these systems
it is natural to expect the superconducting gap function to possess nodes ( in particular d - wave nodes ) .
this is indeed substantiated by experimental findings@xcite . therefore according to earlier arguments
there should be surface e=0 bound states .
we consider the following simple model bdg hamiltonian h()= & ( )_0_3-()_3 + & + _ 0 ( k_x - k_y)[_0+()]_1 .
[ nch ] here @xmath66 , @xmath67 and @xmath68 .
the bandstructure on the @xmath30 edges of this model hamiltonian is shown in fig .
[ fig3]b .
there are e=0 flat bands in different portions of the edge bz .
generally , when exist , such flat bands are two fold degenerate , i.e. , on each edge there are two @xmath63 bound states .
the only exception are the flat bands ( marked by the red thick line segments in fig .
[ fig3]b ) in the momentum intervals between the yellow arrows .
the momenta of these two intervals are opposite to each other ( up to reciprocal lattice vector ) .
in addition the zero energy bound states there are _ non - degenerate_. the e=0 quasiparticle operator associated with the two different red intervals are hermitian conjugate of each other . in ref .
@xcite similar surface flat bands are found in the nodal states bordering non - centrosymmetric ( fully gapped ) topological superconductors .
+ ) .
the nodes are designated by the red dots .
( b ) the bandstructure of the @xmath30 edges .
the edge states exist between the momenta marked by the yellow arrows .
the parameters we used in producing fig .
[ fig3]b are @xmath69 , @xmath70 and @xmath71.,title="fig : " ] ) .
the nodes are designated by the red dots .
( b ) the bandstructure of the @xmath30 edges .
the edge states exist between the momenta marked by the yellow arrows .
the parameters we used in producing fig .
[ fig3]b are @xmath69 , @xmath70 and @xmath71.,title="fig : " ] * detecting nodal spin liquids . *
the search for spin liquid is a focused interest in current condensed matter physics .
the difficulty with finding a spin liquid is precisely why they are novel - they are `` featureless '' in conventional sense . among different spin liquids
there is a class that are nodal ( to ensure their stability we focus on the ones having @xmath72 gauge symmetry ) .
for example such a spin liquid was proposed for the pseudogap state of the cuprates@xcite . like nodal superconductors described above
, nodal spin liquid should have surface `` spinon '' flat bands .
if exist these flat bands will give rise to curie behavior of local magnetic susceptibility hence allows for experimental detection , for example the low temperature susceptibility will diverge as @xmath73 .
of course , as discussed earlier , disorder broadens the singular density of state at @xmath63 into a peak hence renders the local magnetic susceptibility finite .
in addition one might need to worry about the effect of residual spinon interaction on these flat bands .
we believe if the residual spinon interaction is repulsive surface ferromagnetism will result . * discussions . *
experimentally one can prepare differently oriented surfaces by thin film growth and perform scanning tunneling microscopy ( stm ) measurements to measure the local density of states associated with zbabs .
the advantage of using stm is it can probe @xmath74 length scale
. therefore even when the surface is not ideal , it can pick up signals from surfaces with different local orientations .
thus by scanning the tip it is possible to acquire the required data from a single sample .
after completion of this work we noticed a very recent experiment observing zero - bias conductance peak for fese@xmath75te@xmath76@xcite . |
following refs . , and consider charge asymmetric case , when excess of @xmath5 provides effective suppression of positively charged species . in the period @xmath9 at @xmath10 , @xmath6
has already been formed in the sbbn and virtually all free @xmath5 are trapped by @xmath6 in o - helium atoms " @xmath11 . here
the o - helium ionization potential is nucleus leads to smaller value @xmath12 @xcite . ] i_o = z_x^2 z_he^2 ^2 m_he/2 1.6 , [ io]where @xmath13 is the fine structure constant,@xmath14 and @xmath15 stands for the absolute value of electric charge of @xmath5 .
the size of these atoms " is @xcite r_o ~1/(z_x z_hem_he ) 2 10 ^ -13 [ rehe ] here and further , if not specified , we use the system of units @xmath16 . due to nuclear interactions with nuclei of cosmic plasma ,
the o - helium gas is in thermal equilibrium with plasma and radiation on the radiation dominance ( rd ) stage , while the energy and momentum transfer from plasma is effective .
the radiation pressure acting on the plasma is then transferred to density fluctuations of the o - helium gas and transforms them in acoustic waves at scales up to the size of the horizon . at temperature
@xmath17 the energy and momentum transfer from baryons to o - helium is not effective @xcite because @xmath18 where @xmath19 is the mass of the @xmath20 atom , @xmath21 , @xmath22 is the mass of proton , @xmath23 and @xmath24 is the baryon thermal velocity
. then o - helium gas decouples from plasma .
it starts to dominate in the universe after @xmath25 at @xmath26 and o - helium atoms " play the main dynamical role in the development of gravitational instability , triggering the large scale structure formation .
the composite nature of o - helium determines the specifics of the corresponding dark matter scenario , which has qualitative feature of a warmer than cold dark matter model @xcite .
being decoupled from baryonic matter , the @xmath20 gas does not follow the formation of baryonic astrophysical objects ( stars , planets , molecular clouds ... ) and forms dark matter halos of galaxies .
it can be easily seen that o - helium gas is collisionless for its number density , saturating galactic dark matter . taking the average density of baryonic
matter one can also find that the galaxy as a whole is transparent for o - helium in spite of its nuclear interaction .
only individual baryonic objects like stars and planets are opaque for it .
o - helium atoms can be destroyed in astrophysical processes , giving rise to acceleration of free @xmath5 in the galaxy .
if the mechanisms of @xmath5 acceleration are effective , the anomalous low @xmath27 component of @xmath28 charged @xmath5 can be present in cosmic rays at the level @xcite @xmath29 and be within the reach for pamela and ams02 cosmic ray experiments . in the framework of walking technicolor model
the excess of both stable @xmath5 and @xmath30 is possible @xcite , the latter being two - three orders of magnitude smaller , than the former .
it leads to the two - component composite dark matter scenario with the dominant ohe accompanied by a subdominant wimp - like component of @xmath31 bound systems .
technibaryons and technileptons can be metastable and decays of @xmath5 and @xmath30 can provide explanation for anomalies , observed in high energy cosmic positron spectrum by pamela and in high energy electron spectrum by fermi and atic .
o - helium collisions in the galactic bulge can lead to excitation of o - helium . if 2s level is excited , pair production dominates over two - photon channel in the de - excitation by @xmath32 transition and positron production with the rate @xmath33
is not accompanied by strong gamma signal . according to ref .
this rate of positron production for @xmath34 is sufficient to explain the excess in positron annihilation line from bulge , measured by integral ( see ref . for review and references ) .
if @xmath20 levels with nonzero orbital momentum are excited , gamma lines should be observed from transitions ( @xmath35 ) @xmath36 ( or from the similar transitions corresponding to the case @xmath37 ) at the level of @xmath38 .
the evident consequence of the o - helium dark matter is its inevitable presence in the terrestrial matter , which appears opaque to o - helium and stores all its in - falling flux .
the nuclear cross section of the o - helium interaction with matter escapes the severe constraints on strongly interacting dark matter particles ( simps ) @xcite imposed by the xqc experiment @xcite .
therefore , a special strategy of direct o - helium search is needed , as it was proposed in ref . .
after they fall down terrestrial surface the in - falling @xmath20 particles are effectively slowed down due to elastic collisions with matter .
then they drift , sinking down towards the center of the earth with velocity v = 80 s_3 a_med^1/2 /.
[ dif]here @xmath39 is the average atomic weight in terrestrial surface matter , @xmath40 is the number of terrestrial atomic nuclei , @xmath41 is the rate of nuclear collisions and @xmath42 . then the o - helium abundance the earth
is determined by the equilibrium between the in - falling and down - drifting fluxes .
the in - falling o - helium flux from dark matter halo is @xmath43 where @xmath44-speed of solar system ( 220 km / s ) , @xmath45-speed of earth ( 29.5 km / s ) and @xmath46 is the local density of o - helium dark matter . here , for simplicity
, we do nt take into account velocity dispersion and distribution of particles in the incoming flux that can lead to significant effect . at a depth @xmath47 below the earth s surface ,
the drift timescale is @xmath48 , where @xmath49 is given by eq .
( [ dif ] ) .
it means that the change of the incoming flux , caused by the motion of the earth along its orbit , should lead at the depth @xmath50 to the corresponding change in the equilibrium underground concentration of @xmath20 on the timescale @xmath51 . in underground detectors ,
@xmath20 `` atoms '' are slowed down to thermal energies and give rise to energy transfer @xmath52 , far below the threshold for direct dark matter detection .
it makes this form of dark matter insensitive to the severe cdms constraints @xcite .
however , in @xmath20 reactions with the matter of underground detectors can lead to observable effects .
the equilibrium concentration , which is established in the matter of underground detectors , is given by @xmath53 where @xmath54 , @xmath55 and @xmath56 is the phase .
the averaged concentration is given by @xmath57 and the annual modulation of concentration is characterized by @xmath58 the rate of nuclear reactions of ohe with nuclei is proportional to the local concentration and the energy release in these reactions should lead to observable signal .
there are two parts of the signal : the one determined by the constant part and annual modulation , which is concerned by the strategy of dark matter search in dama experiment @xcite .
our explanation @xcite is based on the idea that ohe , slowed down in the matter of dama / nai or dama / libra detector , can form a few kev bound state with nucleus , in which ohe is situated * beyond * the nucleus .
therefore the positive result of this experiment is explained by reaction @xmath59+\gamma \label{heeaz}\ ] ] with sodium and/or iodine . in detectors with different chemical content
such level may not exist at all , or has other value of energy .
the rate of reaction ( [ heeaz ] ) is proportional to temperature and suppressed in cryogenic detectors , making the comparison of their results with dama a nontrivial task .
the approach of refs . and assumes the following picture : at the distances larger , than its size , ohe is neutral and it feels only yukawa exponential tail of nuclear attraction , due to scalar - isoscalar nuclear potential . it should be noted that scalar - isoscalar nature of he nucleus excludes its nuclear interaction due to @xmath60 or @xmath61 meson exchange , so that the main role in its nuclear interaction outside the nucleus plays @xmath62 meson exchange , on which nuclear physics data are not very definite . when the distance from the surface of nucleus becomes smaller than the size of ohe , the mutual attraction of nucleus and ohe
is changed by dipole coulomb repulsion .
inside the nucleus strong nuclear attraction takes place . in the result
the spherically symmetric potential appears , given by @xmath63 here @xmath64 , @xmath65 are atomic weight and charge of helium , @xmath66 and @xmath67 are respectively atomic weight and charge of nucleus , @xmath68 and @xmath69 are the mass and coupling of scalar - isoscalar meson - mediator of nuclear attraction , @xmath70 is the size of ohe and @xmath71 is its electromagnetic formfactor , which strongly suppresses the strength of dipole electromagnetic interaction outside the ohe `` atom '' . to simplify the solution of schrodinger equation the potential ( [ epot ] )
was approximated in @xcite by a rectangular potential , presented on fig .
[ pic23 ] .
solutions of schrodinger equation for each of the four regions , indicated on fig .
[ pic23 ] , are given in textbooks ( see e.g.@xcite ) and their sewing determines the condition , under which a low - energy ohe - nucleus bound state appears in the region iii .
the energy of this bound state and its existence strongly depend on the parameters @xmath68 and @xmath69 of nuclear potential ( [ epot ] ) . on the fig .
[ nai ] the region of these parameters , giving 2 - 6 kev energy level in ohe bound states with sodium and iodine are presented . in these calculations
@xcite the mass of ohe was taken equal to @xmath72 .
the rate of radiative capture of ohe by nuclei can be calculated with the use of the analogy with the radiative capture of neutron by proton with the account for : i ) absence of m1 transition that follows from conservation of orbital momentum and ii ) suppression of e1 transition in the case of ohe .
since ohe is isoscalar , isovector e1 transition can take place in ohe - nucleus system only due to effect of isospin nonconservation , which can be estimated by factor @xmath73 , corresponding to relative mass difference of neutron and proton . in the result
the rate of ohe radiative capture by nucleus with atomic number @xmath66 and charge @xmath67 to the energy level @xmath74 in the medium with temperature @xmath75 is given by @xmath76 formation of ohe - nucleus bound system leads to energy release of its binding energy , detected as ionization signal . in the context of our approach the existence of annual modulations of this signal in the range 2 - 6
kev and absence of such effect at energies above 6 kev means that binding energy of na - ohe and i - ohe systems in dama experiment should not exceed 6 kev , being in the range 2 - 4 kev for at least one of these elements .
the amplitude of annual modulation of ionization signal ( measured in counts per day per kg , cpd / kg ) is given by @xmath77 here @xmath78 is avogadro number , @xmath79 denotes na or i , @xmath80 ( corresponding to 1 kg of the matter of detector ) , @xmath81 , @xmath82 is the binding energy of na - ohe ( i - ohe ) system and @xmath46 is the local density of o - helium dark matter . the value of @xmath83 should be compared with the integrated over energy bins signals in dama / nai and dama / libra experiments and the result of these experiments can be reproduced e.g. for @xmath84 and @xmath85 . at the corresponding values of @xmath68 and @xmath69 energy of ohe binding with other nuclei can strongly differ from 2 - 6 kev .
in particular , energy release at the formation of ohe bound state with thallium can be larger than 6 kev .
however , for the cross section of radiative capture of thallium by ohe , given by eq .
( [ radcap ] ) and taking into account that thallium content in dama detector is 3 orders of magnitude smaller , than nai , such signal is below the experimental errors . it should be noted that the results of dama experiment exhibit also absence of annual modulations at the energy of mev - tens mev .
energy release in this range should take place , if ohe - nucleus system comes to the deep level inside the nucleus ( in the region i of fig .
[ pic23 ] ) .
this transition implies tunneling through dipole coulomb barrier and is suppressed below the experimental limits .
the actual rate of these transitions is under our current study .
since ohe capture rate is proportional to the temperature , it is suppressed in cryogenic detectors by a factor of order @xmath86 .
the predicted effects of ohe radiative capture in different cryogenic detectors at @xmath87 , @xmath88 and @xmath89 are given in table [ ta1 ] .
to conclude , the existence of heavy stable charged particles may not only be compatible with the experimental constraints but even lead to composite dark matter scenario of nuclear interacting warmer than cold dark matter .
this new form of dark matter can provide explanation of excess of positron annihilation line radiation , observed by integral in the galactic bulge .
the search for stable -2 charge component of cosmic rays is challenging for pamela and ams02 experiments .
decays of heavy charged constituents of composite dark matter can provide explanation for anomalies in spectra of cosmic high energy positrons and electrons , observed by pamela , fermi and atic . in the context of our approach search for heavy stable charged quarks and leptons at lhc
acquires the significance of experimental probe for components of cosmological composite dark matter .
the results of dark matter search in experiments dama / nai and dama / libra can be explained in the framework of our scenario without contradiction with negative results of other groups .
our approach contains distinct features , by which the present explanation can be distinguished from other recent approaches to this problem @xcite ( see also review and more references in ref . ) .
the proposed explanation is based on the mechanism of low energy binding of ohe with nuclei . within the uncertainty of nuclear physics parameters
there exists a range at which ohe binding energy with sodium and/or iodine is in the interval 2 - 6 kev .
radiative capture of ohe to this bound state leads to the corresponding energy release observed as an ionization signal in dama detector .
ohe concentration in the matter of underground detectors is determined by the equilibrium between the incoming cosmic flux of ohe and diffusion towards the center of earth .
it is rapidly adjusted and follows the change in this flux with the relaxation time of few minutes .
therefore the rate of radiative capture of ohe should experience annual modulations reflected in annual modulations of the ionization signal from these reactions . an inevitable consequence of the proposed explanation is appearance in the matter of dama / nai or dama / libra detector anomalous superheavy isotopes of sodium and/or iodine , having the mass roughly by @xmath19 larger , than ordinary isotopes of these elements .
our results show that the ionization signal , detected by dama , is proportional to the temperature and should be suppressed in cryogenic detectors
. therefore test of results of dama / nai and dama / libra experiments by other experimental groups can become a very nontrivial task , especially , in view of their rejection of electromagnetic part of counting rate in the absence of nuclear recoil .
the presented approach sheds new light on the physical nature of dark matter .
specific properties of composite dark matter and its constituents are challenging for their experimental search .
ohe interaction with matter is an important aspect of these studies . in this context positive result of dama / nai and dama / libra experiments
may be a signature for exciting phenomena of o - helium nuclear physics .
we would like to thank pierluigi belli , rita bernabei , jean pierre gazeau and bernard sadoulet for discussions and important comments .
we express our gratitude to organizers of iwara09 for cooperation .
khlopov in _ cosmion-94 _ ,
m.yu.khlopov et al .
( editions frontieres , 1996 ) p. 67 ; m. y. khlopov , .
m. y. khlopov , ; in arxiv:0711.4681 , p. 114 ; m. y. khlopov and n. s. mankoc - borstnik , ibid , p. 195 .
s. l. glashow , arxiv : hep - ph/0504287 .
d. fargion and m. khlopov , arxiv : hep - ph/0507087 .
et al_. , m.yu .
khlopov , .
k. belotsky _ et al_. , arxiv : astro - ph/0602261 . k. belotsky _ et al_. , ; k. belotsky _ et al_. , arxiv:0806.1067 [ astro - ph ] .
m. y. khlopov , arxiv : astro - ph/0607048 .
m. y. khlopov and c. kouvaris , .
f. sannino and k. tuominen , ; d. k. hong _ et al_. , ; d. d. dietrich _ et al_. , ; s. b. gudnason _
et al_. , ; s. b. gudnason _ et al _ , .
manko bortnik , ; n.s .
manko bortnik , ; g. bregar , m. breskvar , d. lukman , n.s .
manko bortnik , . c. b. dover _
et al_. , ; s. wolfram , ; g. d. starkman _
et al_. , ; r. bernabei _ et al_. , ; d. javorsek _ et al_. , ; s. mitra , ; g. d. mack _ et al_. , ; b. d. wandelt _ et al_. , arxiv : astro - ph/0006344 ; p. c. mcguire and p. j. steinhardt , arxiv : astro - ph/0105567 ; g. zaharijas and g. r. farrar , .
m. y. khlopov , arxiv:0806.3581 [ astro - ph ] ; m. y. khlopov , arxiv:0801.0167 [ astro - ph ] ; arxiv:0801.0169 [ astro - ph ] .
m. yu . khlopov _ et al_. , m. y. khlopov and c. kouvaris , m. yu .
khlopov , a. g. mayorov , e.yu .
soldatov , ; arxiv:0911.5606 [ astro - ph ] ; m. yu . khlopov , the puzzles of dark matter searches , to appear in _ aip proceedings of the invisible universe international conference _ , arxiv:0911.5685 [ astro-ph.co ] . k. belotsky _ et al_. , arxiv : astro - ph/0606350 .
cdms collaboration ( d. s. akerib _ et al_. ) , ; cdms collaboration ( z. ahmed _
et al_. ) , ; the cdms - ii collaboration ( z. ahmed _
et al_. ) , arxiv:0912.3592 [ astro-ph.co ] ; o. kamaev for the cdms collaboration , arxiv:0910.3005 [ hep - ex ] . f. petriello and k. m. zurek , ; r. foot , ; j. l. feng , j. kumar and l. e. strigari , arxiv:0806.3746 [ hep - ph ] ; j. l. feng , j. kumar , j. learned and l. e. strigari , arxiv:0808.4151 [ hep - ph ] ; e. m. drobyshevski , arxiv:0811.0151 [ astro - ph ] . | positive results of dark matter searches in dama / nai and dama / libra experiments , being put together with the results of other groups , can imply nontrivial particle physics solutions for cosmological dark matter .
stable particles with charge -2 , bound with primordial helium in o - helium `` atoms '' ( ohe ) , represent a specific warmer than cold nuclear - interacting form of dark matter . slowed down in the terrestrial matter
, ohe is elusive for direct methods of underground dark matter detection used in cryogenic experiments .
however radiative capture of ohe by na and i nuclei can lead to annual variations of energy release in the interval of energy 2 - 5 kev in dama / nai and dama / libra experiments .
the widely shared belief is that the dark matter , corresponding to @xmath0 of the total cosmological density , is nonbaryonic and consists of new stable particles .
one can formulate the set of conditions under which new particles can be considered as candidates to dark matter ( see e.g. refs .
for review and reference ) : they should be stable , saturate the measured dark matter density and decouple from plasma and radiation at least before the beginning of matter dominated stage .
the easiest way to satisfy these conditions is to involve neutral weakly interacting particles .
however it is not the only particle physics solution for the dark matter problem . in the composite dark matter scenarios new stable particles
can have electric charge , but escape experimental discovery , because they are hidden in atom - like states maintaining dark matter of the modern universe .
it offers new solutions for the physical nature of the cosmological dark matter .
the main problem for these solutions is to suppress the abundance of positively charged species bound with ordinary electrons , which behave as anomalous isotopes of hydrogen or helium .
this problem is unresolvable , if the model predicts stable particles with charge -1 , as it is the case for tera - electrons @xcite . to avoid anomalous isotopes overproduction , stable particles with charge -1
should be absent , so that stable negatively charged particles should have charge -2 only .
elementary particle frames for heavy stable -2 charged species are provided by : ( a ) stable `` antibaryons '' @xmath1 formed by anti-@xmath2 quark of fourth generation @xcite ( b ) ac - leptons @xcite , predicted in the extension @xcite of standard model , based on the approach of almost - commutative geometry @xcite .
( c ) technileptons and anti - technibaryons @xcite in the framework of walking technicolor models ( wtc ) @xcite .
( d ) finally , stable charged clusters @xmath3 of ( anti)quarks @xmath4 of 5th family can follow from the approach , unifying spins and charges @xcite . in the asymmetric case ,
corresponding to excess of -2 charge species , @xmath5 , their positively charged antiparticles effectively annihilate in the early universe . in all the models ,
in which new stable species belong to non - trivial representations of electroweak su(2 ) group sphaleron transitions at high temperatures provide the relationship between baryon asymmetry and excess of -2 charge stable species .
after it is formed in the standard big bang nucleosynthesis ( sbbn ) , @xmath6 screens the @xmath5 charged particles in @xmath7 _ o - helium _ `` atoms '' @xcite . in all the forms of o - helium
, @xmath5 behaves either as lepton or as specific `` heavy quark cluster '' with strongly suppressed hadronic interaction .
therefore o - helium interaction with matter is determined by nuclear interaction of @xmath8 .
these neutral primordial nuclear interacting objects contribute to the modern dark matter density and play the role of a nontrivial form of strongly interacting dark matter @xcite . here after a brief review of main features of ohe universe we concentrate on its effects in underground detectors .
we present a quantitative confirmation of the earlier guess @xcite that the positive results of dark matter searches in dama / nai ( see for review ref . ) and dama / libra @xcite experiments can be explained by effect of o - helium , resolving the controversy between these data and negative results of other experimental groups . |
the purpose of the present paper is to study some aspects of condensation in the ground state of the gross - pitaevskii energy functional with a disordered background potential . as they can be treated very similar , we consider the discrete and the continuous setting simultaneously
. + * the continuous setting : * + in @xmath0 , consider the cube @xmath1^d$ ] of side length @xmath2 and volume @xmath3 . in @xmath4 , on the domain @xmath5 ,
consider @xmath6 the continuous self - adjoint anderson model on @xmath7 with periodic boundary conditions .
we assume * @xmath8 is the continuous laplace operator ; * @xmath9 is an ergodic random potential i.e. an ergodic random field over @xmath0 that satisfies + @xmath10 \end{aligned}\ ] ] where @xmath11 ( resp .
@xmath12 ) denotes the supremum norm in @xmath13 ( resp .
@xmath14 ) .
these assumptions are for example satisfied by a continuous anderson model with a smooth compactly supported single site potential i.e. if @xmath15 where @xmath16 and @xmath17 are bounded , non negative identically distributed random variables . *
the discrete setting : * + on the finite discrete cube @xmath1^d\cap{{\mathbb z}}^d\subset{{\mathbb z}}^d$ ] the cube of side length @xmath18 and volume @xmath19 , let @xmath20 the discrete anderson model on @xmath21 with periodic boundary conditions .
we assume * @xmath22 is the discrete laplacian ; * @xmath9 is a potential i.e. a diagonal matrix entries of which are are given by bounded non negative random variables , say @xmath23 . for the sake of definiteness
, we assume that the infimum of the ( almost sure ) spectrum of @xmath24 be @xmath25 .
we define [ discrete gross - pitaevskii model ] the ( one - particle ) gross - pitaevskii energy functional on the cube @xmath7 ( in the discrete or in the continuous ) is defined by @xmath26= \langle { \rm h^p_{\omega , l } } \varphi,\varphi \rangle + { \rm u } \|\varphi\|_4 ^ 4\ ] ] for @xmath27 and @xmath28 is a positive coupling constant . for applications ,
it is natural that this coupling constant is related to @xmath29 .
we refer to the discussion following theorem [ theorem1 ] for details .
one proves [ pro:1 ] for any @xmath30 and @xmath31 , there exists a ground state @xmath32 i.e. a vector @xmath33 such that @xmath34 minimizing the gross - pitaevskii energy functional , i.e. @xmath35 = { \min_{\substack{\varphi \in \mathcal{d}_l \\ \|\varphi\|_2=1 } } } { \mathcal e}^{gp}_{\omega , l}[\varphi].\ ] ] the ground state @xmath32 can be chosen positive ; it is unique up to a change of phase .
@xmath36 denotes the ground state energy of the discrete gross - pitaevskii functional .
the proof in the continuous case is given in @xcite ; the proof in the discrete case is similar .
+ let @xmath37 and @xmath38 respectively denote the neumann and dirichlet restrictions of @xmath24 to @xmath7 .
our main assumptions on the random model are : * ( h0 ) decorrelation estimate : * the model satisfies a finite range decorrelation estimate i.e. there exits @xmath39 such that , for any @xmath40 and any sets @xmath41 , if @xmath42 then the restrictions of @xmath43 to the domains @xmath44 , i.e. the functions @xmath45 , are independent random fields . + * ( h1 ) wegner estimate : * there exists @xmath46 such that , for any compact interval @xmath47 and @xmath48 , @xmath49\leq c|i|l^d;\ ] ] @xmath50 + * ( h2 ) minami estimate : * there exists @xmath46 such that , for @xmath47 a compact interval and @xmath48 , @xmath51\leq c(|i|l^d)^2;\ ] ] @xmath50 + * ( h3 ) lifshitz type estimate near energy @xmath52 : * there exist constants @xmath53 such that , for @xmath31 and any parallelepiped @xmath54 where the intervals @xmath55 satisfy @xmath56 , one has @xmath57\}],\\ { \mathbb{p}}[\{h^n_{\omega|p_l}\text { has at least one eigenvalue in } [ 0,l^{-2}]\}]\leq c e^{-l^d / c } \end{gathered}\ ] ] where @xmath58 ( resp .
@xmath58 ) is the dirichlet ( resp .
neumann restriction ) of @xmath24 to @xmath59 .
@xmath50 + let us now discuss the validity of these assumptions .
+ the decorrelation assumption ( h0 ) is satisfied for the discrete anderson model described above if the random variables @xmath60 are i.i.d .
( h0 ) clearly allows some correlation between the random variables . for the continuous anderson model , it is satisfied if the single site potential has compact support and the random variables are i.i.d . + under the assumption that the random variables are i.i.d and that their distribution is regular , it is well known that the wegner estimate ( h1 ) holds at all energies for both the discrete and continuous anderson model ( see e.g. @xcite ) .
+ the minami estimate ( h2 ) is known to hold at all energies under similar regularity assumptions for the discrete anderson model ( see e.g. @xcite ) and for the continuous anderson model in the localization regime under more specific assumptions on the single site potential ( see e.g. @xcite ) .
+ finally , the lifshitz tails estimate ( h3 ) is known to hold for both the continuous and discrete anderson model under the sole assumption that the i.i.d .
random variables be non degenerate , non negative and @xmath25 is in their essential range ( see e.g. @xcite ) . though the lifshitz tails estimate is usually not stated for parallelepipeds but for cubes , the proof for cubes applies directly to parallelepipeds satisfying the condition stated in ( h3 ) . .4cmthe
main result of the present paper is [ * condensation in the single particle ground state * ] [ theorem1 ] assume assumptions ( h0)-(h3 ) hold .
denote by @xmath61 the single particle ground state of @xmath62 ( chosen to be positive for the sake of definiteness ) and by @xmath32 the gross - pitaevskii ground state .
+ if for @xmath63 large , one assumes that @xmath64 where @xmath65 and @xmath66 in the discrete setting , resp .
@xmath67 arbitrary in the continuous case , then , there exists @xmath68 when @xmath69 such that @xmath70{\operatornamewithlimits{\to}}_{l\to+\infty}0.\]]@xmath50 + the proof of theorem [ theorem1 ] also yields information on the size of @xmath71 and on the probability estimated in .
note that the assumption ( h1)-(h3 ) can be relaxed at the expense of changing the admissible size for @xmath72 .
+ to appreciate theorem [ theorem1 ] maybe some comments about the physical background of the gross - pitaevskii model , its relationship to bose - einstein condensation and to known results are of interest .
motivated by recent experiments with weakly interacting bose gases in optical lattices ( see for example @xcite ) the fundamental objects of interest are the ground state density and energy , i.e. @xmath73\phi\right\rangle.\ ] ] the optical lattice is modeled by the background potential @xmath74 as shown in figure [ fig:1 ] .
[ fig:1 ] assuming a weak interaction limit of the interaction potential @xmath75 , the continuous @xmath76-particle gross - pitaevskii energy functional @xmath77 is a mean field approximation of the ground state energy ( [ ground state energy ] ) , e.g. in three dimensions one has @xmath78 ( see for example @xcite ) the discrete gross - pitaevskii model is then a tight binding approximation of the continuous one - particle gross - pitaevskii functional @xcite @xmath79=\int_{{\mathbb{r}}^3}\left ( |\nabla\varphi(x)|^2 + v|\varphi(x)|^2 + 4 n \pi \mu a|\varphi(x)|^4\right)dx.\ ] ] another way to derive the discrete gross - pitaevskii model starts with a discretization of ( [ ground state energy ] ) yielding the standard description of optical lattices using the bose - hubbard - hamiltonian @xmath80 where @xmath81 , @xmath82 are bosonic creation and annihilation operators and @xmath83 gives the particle number at site @xmath84 ( see the survey article @xcite and references therein ) .
a mean field approximation then yields the discrete gross - pitaevskii energy functional @xcite .
+ one motivation to study bose gases is bose - einstein condensation , i.e. the phenomena that a single particle level has a macroscopic occupation ( a non - zero density in the thermodynamic limit ) @xcite . introduced in @xcite in the context of an ideal bose gas , it was due to naturally arising interactions a difficult problem to realize bose - einstein condensation experimentally @xcite .
+ as we will see , also the formal description is more elaborated . to motivate the definition of bec for vanishing temperature we follow the continuous approach in @xcite . to formalize the concept of a macroscopic occupation of a single particle state
we remember the definition of the one - particle density matrix @xcite , i.e. the operator on @xmath85 given by the kernel @xmath86 with the normalized ground state wave function @xmath87 of the many boson hamiltonian .
bec in the ground state is then defined that the projection operator @xmath88 has an eigenvalue of order @xmath76 in the thermodynamic limit .
+ remembering that for the ideal bose gas the multi - particle ground state can be represented as a product @xmath89 of the single particle ground state @xmath90 the one - particle density matrix becomes @xmath91 thus the definition of bec above is natural and can also be related to the thermodynamic formalism ( see e.g. @xcite and references ) .
in particular , it is of interest to consider bec for the ideal bose gas with a random background potential . in this case
the lifshitz tail behavior at the bottom of the spectrum makes a generalized form of bose - einstein condensation possible even for @xmath92 , @xmath93 ( see @xcite and references cited there ) .
+ the situation in the gross - pitaevskii - limit is close to the situation for the ideal bose gas @xcite .
the one - particle density matrix is asymptotically given by @xmath94 physically the content of ( [ condensation in the ground state ] ) is that all bose particles will condensate in the gp ground state motivating the definition of complete ( or 100% ) bec in @xcite .
+ the purpose of the present publication is a first step to analyze the fine structure of the gross - pitaevskii ground state . under the assumption of a random background potential
we want to understand how @xmath32 is related to the eigenstates of the single particle hamiltonian .
more familiar is this problem in the following two settings .
+ if the bosons are trapped by a potential tending to @xmath95 , i.e. @xmath96 , the spectral properties of the single particle are invariant in the thermodynamic limit , i.e. the discrete spectrum and the strictly positive distance between the first two eigenvalues . assuming @xmath97 in the continuous setting , respectively @xmath98 in the context of the discrete gross - pitaevskii model , the interaction energy is a small perturbation of the single particle energy functional .
in this situation it is natural , that in the thermodynamic limit @xmath99 and the single particle ground state @xmath90 coincide @xcite . +
a complementary situation is given if the bosons are confined to a cube @xmath7 with @xmath100 but without a background potential .
as described in @xcite assuming @xmath101 and @xmath102 in the limit @xmath103 one can prove @xmath104 i.e. bec in the normalized single particle ground state @xmath105 . as explained in @xcite @xmath106
is in this context the natural interaction parameter since `` in the gp limit the interaction energy per particle is of the same order of magnitude as the energy gap in the box , so that the interaction is still clearly visible '' . + as emphasized in the physics literature ( see e.g. @xcite ) , new phenomena like fragmented bec ( lifshitz glasses ) should occur when bosons are trapped in a random background potential .
our purpose in this publication is more modest .
we want to understand the natural interaction parameter in a random media , s.t .
the gross - pitaevskii ground state is close to the ground state of the single particle hamiltonian as it is suggested by the situation in the ideal bose gas .
as we will see the setting of bosons trapped in a random potential is not really comparable to the two situations described above .
+ under our assumptions , near @xmath25 which is almost sure limit of @xmath107 , we are in the localized regime , i.e. one has pure point spectrum and localized eigenfunctions .
in contrast to the situation with vanishing potential the eigenstates close to the bottom of the spectrum are localized in a small part of @xmath7 , i.e. the interaction energy will be larger than in the case of the homogeneous bose gas . in the random case ,
we determine the almost sure behavior of the ground state from information on the integrated density of states ( see lemma [ lemma1 ] ) . under our weak lifshitz tails assumption ( h3 )
, we obtain that the ground state energy is of size @xmath108 .
when @xmath69 , the difference between the first two eigenvalues will tend to zero ; the speed at which this happens is crucial in our analysis ( see proposition [ le:2 ] ) . in our case , we estimate that , with good probability , it must be at least of order @xmath109 .
this difference is much smaller than the one obtained in the homogeneous bose gas where it typically is of order @xmath110 .
we deem that the estimate @xmath109 for the spacing is not optimal in the present setting .
this estimate is the correct one in the bulk of the spectrum ; at the edges , the spacings should be larger .
it seems that getting an optimal estimate requires a much better knowledge of the integrated density of states or , in other words , much sharper lifshitz tails type estimates ( see ( h3 ) ) and minami type estimates that take into account the fact that we work at the edge of the spectrum ( see ( h2 ) ) . combining these observations
explains the interaction parameter @xmath111 that we do nt believe to be optimal .
+ let us now briefly outline the structure of our paper . to prove our result we need two ingredients .
we need an upper bound of the interaction term , i.e. we have to estimate the @xmath112- norm of the single particle ground state @xmath61 . at the same time , we need a lower bound of the distance of the first two single particle eigenvalues asymptotically almost surely ( a.a.s . )
i.e. with a probability tending to @xmath113 in the thermodynamic limit .
comparing these two estimates we will see that under the assumptions of theorem [ theorem1 ] it is energetically favorable , that the gross - pitaevskii ground state and the single particle ground state coincide .
this will be proven at the end of this publication .
+ to estimate the interaction term we will prove in lemma [ lemma1 ] that almost surely in the thermodynamic limit the single particle ground state is flat , i.e. @xmath114 this then yields an estimate of the interaction term which is the purpose of proposition [ lemma2 ] .
+ the a.a.s .
lower bound of the distance of the first two single particle eigenvalues is a little bit more intricate and uses the wegner and minami estimates ; it is related to the methods developed in @xcite . in lemma
[ lemma3 ] , we first estimate the probability that the first two eigenstates and also their localization center are close together . if the localization centers are relatively far away , one can decouple the eigenstates and treat the first two eigenvalues of each other .
this is used in lemma [ lemma5 ] .
the main result of this section is an upper bound on @xmath115}$ ] .
this quantity is non negative ( see ( [ ground state ] ) ) and we prove [ lemma2 ] there exists @xmath46 , such that , for any @xmath116 , one has @xmath117 } \leq c u f_d(\log l)\right ] \geq 1 -l^{-p}\ ] ] where @xmath118 is defined in .
+ by definition , for @xmath119 the ground state of @xmath120 , one has @xmath121={\rm e_0^p[\omega , l]}+ u \|\varphi_0(\omega , l)\|_4 ^ 4\ ] ] resp .
@xmath122 } \leq u \|\varphi_0(\omega , l)\|_4 ^ 4.\ ] ] to prove proposition [ lemma2 ] , resp .
control the interaction term , we first estimate the ground state energy of the random schrdinger operator and derive in corollary [ le:4 ] an estimate on the `` flatness '' of its ground state .
we start with the dirichlet and neumann boundary cases .
[ lemma1 ] assume ( h3 ) is satisfied .
let @xmath123 be the ground state energy of @xmath120 and denote by @xmath119 the associated positive normalized ground state .
+ then , for any @xmath124 , there is a constant @xmath46 such that , for @xmath63 sufficiently large , @xmath125\geq 1 -l^{-p}.\]]@xmath50 + as @xmath43 is non negative and @xmath119 normalized , one has @xmath126 .
hence , proposition [ lemma2 ] implies the following `` flatness '' estimate of the ground state .
[ le:4 ] under the assumptions of proposition [ lemma2 ] , for any @xmath124 , there is a constant @xmath46 such that , for @xmath63 sufficiently large , @xmath127\geq 1 -l^{-p}.\]]@xmath50 + it is maybe interesting to note that from a lifshitz tail type estimate ( i.e. the annealed estimate ) , we recover the ( approximate ) almost sure behavior of the ground state energy of @xmath37 ( i.e. the quenched estimate ) ( see e.g. @xcite ) .
+ we note that proposition [ lemma2 ] and corollary [ le:4 ] also hold if we replace the periodic ground state and ground state energy by the neumann or dirichlet ones .
fix @xmath128 .
decompose the interval @xmath129 $ ] into intervals of length comprised between @xmath130 and @xmath131 .
this yields a partition of @xmath7 in parallelepipeds i.e. @xmath132 such that * @xmath133 where the intervals @xmath134 satisfy @xmath135 * for @xmath136 , @xmath137 , * @xmath138 , the number of parallelepiped , satisfies @xmath139 . in the continuous model
, one can take the parallel piped to be cubes .
+ denote by @xmath140 the restriction of @xmath14 to @xmath7 .
furthermore , let @xmath141 be the periodic extension of @xmath140 to @xmath142 i.e. for @xmath143 and @xmath144 , @xmath145 where @xmath146 in the discrete case and @xmath2 in the continuous one . as @xmath147 is the periodic restriction of @xmath24 to @xmath7 , we know that @xmath148}=\inf\sigma(h_{\omega^{p , l}})$ ] where this last operator is considered as acting on the full space @xmath0 or @xmath142 ( see e.g. @xcite )
. + we can now decompose @xmath0 or @xmath142 into @xmath149 . by dirichlet - neumann bracketing ( see e.g. @xcite ) @xmath150 satisfies as an operator on @xmath142 or @xmath0 @xmath151 define @xmath152}=\inf \sigma({\rm h}^\bullet_{\omega|p_j})\quad\text{for } \quad\bullet\in\{n , d\};\ ] ] here , the superscripts @xmath153 and @xmath76 refer respectively to the dirichlet and neumann boundary conditions .
as @xmath141 is @xmath154-periodic , @xmath155 and @xmath156 are unitarily equivalent .
the bracketing ( [ eq:14 ] ) then yields @xmath157}\leq { \rm e_0^p[\omega , l ] } \leq \inf_{1\leq j\leq j } { \rm e_0^d[\omega,\ell , j]}.\ ] ] labeling every second interval of the partition of @xmath129 $ ] used to construct the partition of @xmath7 , we can partition the interval @xmath158 into @xmath159 sets , say @xmath160 such that 1 . if @xmath161 , @xmath162 , 2 . for @xmath163 and @xmath164 such that @xmath165 , one has dist@xmath166 , 3 . there exists @xmath46 such that for @xmath167 , @xmath168
assume @xmath169 is given by ( h0 ) . by ( 2 ) of the definition of the partition above , for any @xmath170 , all the @xmath171 , resp .
all the @xmath172)_{j\in \mathcal{j}_l}$ ] ( for @xmath173 ) are independent .
hence , using ( [ eq:19 ] ) , we compute @xmath174 } > e]&\leq { \mathbb{p } } [ \inf_j { \rm e_0^d[\omega,\ell , j ] } > e]\leq \sum_{l=1}^{2^d}\prod_{j\in\mathcal{j}_l}{\mathbb{p } } [ { \rm e_0^d[\omega,\ell , j ] } > e]\\ & = \sum_{l=1}^{2^d}\prod_{j\in\mathcal{j}_l}\left ( 1-{\mathbb{p } } [ { \rm e_0^d[\omega,\ell , j]}\leq e]\right ) .
\end{split}\ ] ] pick @xmath175 where @xmath176 is given by assumption ( h3 ) and @xmath177 where @xmath178 will be chosen below . applying the lifshitz estimate ( h3 ) , we obtain @xmath174 } > e]&\leq \sum_{l=1}^{2^d}\left ( 1- e^{-k \log l / c}\right)^{\#\mathcal{j}_l } \\&\leq \sum_{l=1}^{2^d}\exp\left(-\#\mathcal{j}_l\,e^{-k \log l / c}\right)\leq o(l^{-\infty } ) \end{split}\ ] ] if we choose @xmath179 as @xmath180 for @xmath167 .
+ hence , we have @xmath181}\leq e]\geq 1-o(l^{-\infty}).\ ] ] to estimate from below , we use again ( [ eq:19 ] ) to get @xmath182}\leq e ] \leq \sum_{j\in\mathcal{j } } { \mathbb{p } } [ { \rm e_0^n[\omega,\ell , j]}\leq e]).\ ] ] pick @xmath183 where @xmath184 is given by assumption ( h3 ) and @xmath185 where @xmath178 will be chosen below . as @xmath186 , applying the lifshitz estimate ( h3 ) , we obtain @xmath181}\leq e]\leq c\left(\frac{l}{\ell}\right)^d e^{-k \log l / c}\leq l^{-p}\ ] ] if we choose @xmath187 . hence , we have @xmath181}\geq e]\geq 1-l^{-p}.\ ] ] this completes the proof of lemma [ lemma1 ] . to prove estimate ( [ eq:4 ] ) , we will use the spectral decomposition of @xmath188 .
though the arguments in the discrete and continuous cases are quite similar , it simplifies the discussion to distinguish between the discrete and the continuous case rather than to introduce uniform notations . we start with the discrete case .
[ le:1 ] there exists @xmath46 such that , for @xmath189 and @xmath190 satisfying @xmath191 one has for @xmath192 with @xmath193 and @xmath194 the estimate @xmath195 the spectral decomposition of @xmath188 is given by the discrete fourier transform that we recall now .
identify @xmath7 with the abelian group @xmath196 . for @xmath197 , set @xmath198 then , one checks that ( see e.g. @xcite ) @xmath199 pick @xmath192 with @xmath193 and @xmath194 and write @xmath200 where @xmath201 , @xmath202 and * @xmath203 * for @xmath204 , @xmath205 * @xmath206 where @xmath207 denotes the discrete fourier transform defined defined in .
+ then , for @xmath208 , @xmath209 and , using ( [ eq:7 ] ) , for @xmath210 , @xmath211 i.e. @xmath212 hence , using ( [ eq:9 ] ) and hlder s inequality , we compute @xmath213 so , for @xmath214 , one gets @xmath215 then , using ( [ eq:8 ] ) , we compute @xmath216 where @xmath217 is defined in ( [ eq:6 ] ) .
this completes the proof of lemma [ le:1 ] .
[ rem:1 ] lemma [ le:1 ] is essentially optimal as , for @xmath63 sufficiently large , * the trial function @xmath218 satisfies @xmath219 , @xmath220 and @xmath221 ; * the trial function @xmath222 satisfies @xmath223 , @xmath224 and @xmath225 .
we now turn to the continuous case .
[ le:3 ] fix @xmath226 .
there exists @xmath46 such that , for @xmath189 , @xmath227 and @xmath190 satisfying @xmath191 one has for @xmath228 with @xmath229 the norm estimate @xmath230 we now use the fourier series transform to decompose @xmath188 .
identify @xmath7 with the abelian group @xmath231 . for @xmath197 , set @xmath232 then , @xmath233 and @xmath234 pick @xmath235 as in lemma [ le:3 ] and decompose it as in the proof of lemma [ le:1 ] i.e. write @xmath200 where @xmath201 , @xmath236 and * @xmath203 * for @xmath204 , @xmath205 * @xmath206 where @xmath207 denotes the fourier series transform defined in and .
+ the control on @xmath237 for @xmath238 is obtained in the same way as in the proof of lemma [ le:1 ] namely the estimate holds for @xmath238 .
the additional ingredient that we need is to obtain a control over the large frequency components .
+ recall that @xmath239 fix @xmath240 . for notational convenience ,
write @xmath241 and compute @xmath242 \left|\frac{\pi\gamma}{l}\right|^{-r/2}\\ & \leq \frac1{(2l)^{d/2}}\left(\sum_{e^{k-1}\varepsilon l \leq |\gamma| < e^k\varepsilon l } \left|\frac{\pi\gamma}{l}\right|^{r } |(\hat v)_\gamma|^2\right)^{1/2 } \left(\sum_{e^{k-1}\varepsilon l \leq |\gamma|}\left|\frac{\pi\gamma}{l}\right|^{-r } \right)^{1/2}\\ & \leq c \left(\sum_{e^{k-1}\varepsilon l \leq |\gamma|
< e^k\varepsilon l } \left|\frac{\pi\gamma}{l}\right|^{r-2/q } |(\hat v)_\gamma|^{2/p}\cdot \left|\frac{\pi\gamma}{l}\right|^{2/q } |(\hat v)_\gamma|^{2/q}\right)^{1/2}\\ & \leq c \|v\|^{1/p}_{h^{rp-2p / q}}\cdot\langle -\delta^p_l v , v\rangle^{1/(2q)}\\ & \leq c \|v\|^{\eta}_{h^n}\cdot\epsilon^{1-\eta } \end{split}\ ] ] if @xmath243 , @xmath244 and @xmath245 as @xmath227 .
one then completes the proof of lemma [ le:3 ] in the same way as that of lemma [ le:1 ] . in the discrete case proposition [ lemma2 ]
is a consequence of lemma [ lemma1 ] and lemma [ le:1 ] with @xmath246 .
+ to be able to apply lemma [ le:3 ] to @xmath119 in the continuous case , we need to show that , for any @xmath247 , @xmath248 with a bounded depending only on @xmath84 not on @xmath63 or @xmath14 .
therefore we use the first assumption on the random field @xmath43 i.e. that , for any @xmath249 , @xmath250 .
hence , as @xmath119 is an eigenvector of @xmath251 , using the eigenvalue equation @xmath252-v_\omega)\varphi_0(\omega , l)\ ] ] inductively , we see that @xmath253 proposition [ lemma2 ] in the continuous case is then a consequence of lemma [ lemma1 ] and lemma [ le:3 ] with @xmath246 .
this completes the proof of proposition [ lemma2 ] .
the main result of the present section is [ le:2 ] let the first two eigenvalues of @xmath62 be denoted by @xmath254}<{\rm e_1^p[\omega , l]}$ ] .
then , for @xmath124 , there exists @xmath46 such that , for @xmath63 sufficiently large and @xmath255 , one has @xmath256}-{\rm e_0^p[\omega , l ] } \leq \eta l^{-d}\right]\leq c\eta\left[1+(\log l)^{d-2/d+\epsilon}\right]+l^{-p}\ ] ] with @xmath66 in the discrete setting resp . @xmath67
arbitrary in the continuous case . in the localization regime , both the level - spacing and the localization centers spacing have been studied in e.g. @xcite .
the main difficulty arising in the present setting is that the interval over which we need to control the spacing is of length @xmath257 ; it is large compared to the length scales dealt with in @xcite .
+ our analysis of the spectral gap relies on the description of the ground state resulting from the analysis of the anderson model @xmath24 in the localized regime ( see e.g. @xcite , @xcite ) . under the assumptions made above on @xmath24 ,
there exists @xmath47 a compact interval containing @xmath25 such that , in @xmath47 , the assumptions of the aizenman - molchanov technique ( see e.g. @xcite ) or of the multi - scale analysis ( see e.g.@xcite ) are satisfied .
one proves [ lemma4 ] there exists @xmath258 such that , for any @xmath124 , there exists @xmath259 such that , for any @xmath31 and @xmath260 , there exists @xmath261 such that * @xmath262\geq 1-l^{-p}$ ] , * for @xmath263 , one has that , if @xmath264 is a normalized eigenvector of @xmath265 associated to @xmath266 , and @xmath267 is a maximum of @xmath268 on @xmath7 then , for @xmath269 , one has , @xmath270 note that , for a given eigenfunction , the maximum of its modulus need not be unique but two maxima can not be further apart from each other than a distance of order @xmath271 .
so for each eigenfunction , we can choose a maximum of its modulus that we dub center of localization for this eigenfunction .
+ to prove proposition [ le:2 ] , we will distinguish two cases whether the localization centers associated to @xmath254}$ ] and @xmath272}$ ] , say , respectively @xmath273 and @xmath274 are close to or far away from each other .
+ in lemma [ lemma3 ] , we show that the centers of localization being close is a very rare event as a consequence of the minami estimate .
+ in lemma [ lemma5 ] , we estimate the probability of @xmath275}$ ] and @xmath276}$ ] being close to each other when @xmath273 and @xmath274 are far away from each other . in this case ,
@xmath277}$ ] and @xmath276}$ ] are essentially independent of each other , and the estimate is obtained using wegner s estimate .
[ lemma3 ] for @xmath124 , there exists @xmath278 such that , for @xmath279 , @xmath280 and @xmath255 , one has @xmath281}-{\rm e_0^p[\omega , l ] } \leq \eta\ , l^{-d},\\ & |x_0(\omega)-x_1(\omega)|\leq \lambda(\log l)^{1/\xi } \end{aligned } \right]\ , \leq c\eta(\log l)^{d/\xi-2/d } + l^{-p}\ ] ] with @xmath282 in the discrete setting resp .
@xmath283 arbitrary in the continuous case .
let us start with the discrete setting .
fix @xmath124 and let @xmath284 be given by lemma [ lemma4 ] . the basic observation following from lemma [ lemma4 ]
is that , for @xmath263 , if @xmath285 is the localization center of @xmath286 and @xmath287 , then @xmath288 where * @xmath289_{|x_n(\omega)+\lambda_l}$ ] is @xmath120 restricted to the cube @xmath290 , * @xmath291 is the eigenfunction @xmath286 restricted to the cube @xmath290 . to apply the observation above
we pick a covering @xmath292 of @xmath7 by cubes of side length of order @xmath271 i.e. @xmath293 .
then the number of cubes @xmath138 can be estimated by @xmath294 and there exists @xmath46 ( depending on @xmath295 , @xmath284 and @xmath296 ) such that , if @xmath297 and @xmath298 , there exists a cube @xmath299 ( containing @xmath273 ) such that , for @xmath63 sufficiently large @xmath300 where we have set @xmath301 ( see ) and * @xmath302 is the operator @xmath24 restricted to the cube @xmath303 , * @xmath304 for @xmath305 .
@xmath50 + let @xmath184 be given by lemma [ lemma1 ] and define @xmath306 $ ] . decompose @xmath307 where * @xmath308 are intervals of length @xmath309 , * for @xmath310 , @xmath311 , * for @xmath312 , @xmath313 is of length @xmath314
. one can choose @xmath315 .
this implies that , for @xmath63 sufficiently large , @xmath316}-{\rm e_0^p[\omega , l ] } \leq \eta\ , l^{-d}\\ & |x_0(\omega)-x_1(\omega)|\leq \lambda\log l \end{aligned } \right\}\subset\omega_1\cup\omega_2\ ] ] where @xmath317 and @xmath318 by lemma [ lemma4 ] , we know that @xmath319\leq l^{-p}\ ] ] minami s estimate ( h.2 ) and the estimate on @xmath320 tells us that @xmath321\leq cl^{2d}(\log l)^{-d-2/d}\eta^{-1 } ( \eta l^{-d}(c\log l)^d)^2\leq c\eta(\log l)^{d-2/d}.\ ] ] this completes the proof for the discrete setting .
the proof for the continuous case is very similar .
one has to replace @xmath322 by a smooth version of the characteristic function of the cube @xmath290 ( see for example @xcite ) , resp . change the length scale @xmath271 to @xmath323 in the side length of the boxes where one restricts the eigenfunctions .
this is necessary because of the weaker estimate in lemma [ lemma4 ] .
this completes the proof of lemma [ lemma3 ] .
we now estimate the probability of the spectral gap being small conditioned on the fact that the localization centers are far away from one another .
we prove [ lemma5 ] for any @xmath124 , there exists @xmath279 and @xmath324 such that , for @xmath63 sufficiently large and @xmath255 , one has @xmath325}-{\rm e_0^p[\omega , l ] } \leq \eta\ , l^{-d},\\ & |x_0(\omega)-x_1(\omega)|\geq \lambda(\log l)^{1/\xi } \end{aligned } \right]\ , \leq c\eta+l^{-p}\ ] ] with @xmath282 in the discrete setting , resp .
@xmath283 arbitrary in the continuous case .
using the same line of reasoning as in the proof of lemma [ lemma3 ] we give the proof in the discrete setting . + fix @xmath326 and split the interval @xmath327 $ ] into intervals of length @xmath328 as in the proof of lemma [ lemma3 ] . by minami
s estimate , we know that , for @xmath63 sufficiently large @xmath329}-{\rm e_0^p[\omega , l ] } \leq l^{-\nu } \right]\,\leq c(\log l)^{-2/d}l^\nu l^{2(d-\nu)}\leq l^{-p}.\ ] ] so we may assume that @xmath272}-{\rm e_0^p[\omega , l ] } \geq l^{-\nu}$ ] . + as in the proof of lemma [ lemma3 ] , pick a covering of @xmath7 by cubes , say @xmath330 of side length less than @xmath271 such that @xmath138 , the number of cubes , satisfies @xmath331 .
+ assume that @xmath299 is the cube containing @xmath274 , @xmath276}-{\rm e_0[\omega , l ] } \leq \eta\ , l^{-d}$ ] and @xmath332 .
let @xmath333 .
define the operators @xmath334 , resp .
@xmath335 to be the restriction of @xmath120 to @xmath336 , resp .
@xmath337 , with dirichlet boundary conditions . if @xmath338 and @xmath63 is large enough , we know that * dist@xmath339 , * dist@xmath340 * dist@xmath341 with @xmath39 as in the decorrelation assumption ( h0 ) .
hence , for @xmath295 sufficiently large , using the estimate ( [ eq:10 ] ) for the operators @xmath334 and @xmath335 , we know that : * the operator @xmath335 admits an eigenvalue , say @xmath342 , that satisfies @xmath343 ; * the operator @xmath334 admits an eigenvalue , say @xmath344 , that satisfies @xmath345 .
moreover , as @xmath346 is the dirichlet restriction of @xmath120 , its eigenvalues are larger than those of @xmath120 .
in particular , its second eigenvalue is larger than @xmath347 .
hence , up to a small loss in probability , we may assume it is larger than @xmath348 as we know the estimate ( [ eq:11 ] ) .
this implies that we may assume that @xmath344 is the ground state of @xmath349 .
so we obtain @xmath350}-{\rm e_0^p[\omega , l]}&\leq\eta\,l^{-d},\\ |x_0(\omega)-x_1(\omega)|&\geq\lambda\log l \end{aligned } \right\}\subset\omega_1 \cup\bigcup_{1\leq j\leq j}\omega_j\ ] ] where @xmath351}-{\rm e_0[\omega , l ] } \leq l^{-\nu}\}$ ] and @xmath352 as @xmath335 and @xmath334 are independent of each other , we estimate the probability of @xmath353 using wegner s estimate to obtain @xmath354\leq c(\eta l^{-d}+l^{-\nu})(\log l)^d.\ ] ] hence , one obtains @xmath355}-{\rm e_0[\omega , l ] } \leq \eta\ , l^{-d},\\ & |x_0(\omega)-x_1(\omega)|\geq \lambda\log l \end{aligned } \right ] & \leq c(\eta l^{-d}+l^{-\nu})(\log l)^d\frac{l^d}{(\log l)^d}+2l^{-p } \\&\leq c(\eta+l^{-p } ) \end{split}\ ] ] if @xmath356 .
+ this completes the proof in the discrete setting . to prove lemma [ lemma5 ] for the continuous case
, one does the same modifications as in the proof of lemma [ lemma3 ] in the continuous setting .
setting @xmath357 , proposition [ le:2 ] then follows from lemma [ lemma3 ] and lemma [ lemma5 ] .
defining @xmath358 and applying the definition of the ground state , we can estimate @xmath359}&\|(1-\pi_0)\varphi^{\rm gp}\| + { \rm e_0^p[\omega , l]}\| \pi_0 \varphi^{\rm gp}\|\\ & \leq { \rm e}_{\omega , l}^{\rm gp}\|(1-\pi_0)\varphi^{\rm gp}\| + { \rm e}_{\omega , l}^{\rm gp}\| \pi_0 \varphi^{\rm gp}\|,\end{aligned}\ ] ] respectively @xmath360}-{\rm e}_{\omega , l}^{\rm gp}\right ) \|(1-\pi_0)\varphi^{\rm gp}\| \leq \left ( { \rm e}_{\omega , l}^{\rm gp}-{\rm e_0^p[\omega , l]}\right ) \| \pi_0 \varphi^{\rm gp}\|.\ ] ] as a consequence of proposition [ lemma2 ] and proposition [ le:2 ] , we know with a probability larger than @xmath361 that , for @xmath255 and @xmath118 defined in ( [ eq:12 ] ) the estimates @xmath362}-{\rm e}_{\omega , l}^{\rm gp } \geq { \rm e_1^p[\omega , l]}-{\rm e_0^p[\omega , l ] } \geq \eta l^{-d } [ 1+(\log l)^{d-2/d+\epsilon}]^{-1}\ ] ] and @xmath363}\leq c
u f_d(\log l)\ ] ] are satisfied .
we obtain @xmath364\ ; \|\pi_0\varphi^{\rm gp}\|\ ] ] and @xmath365\right]^2 .
\end{split}\ ] ] applying the assumption concerning the coupling constant @xmath72 i.e. @xmath366^{-1}[f_d(\log l)]^{-1}\right)\ ] ] and setting @xmath367f_d(\log l)|}\ ] ] we get that , @xmath368 when @xmath69 and for some @xmath46 , @xmath369 this completes the proof of theorem [ theorem1 ]
. 99 aizenman , m. , elgart , a. , naboko , s. , schenker , j. , stolz , g. : _ moment analysis for localization in random schrdinger operators _ , inv . math . * 163 * , 343 - 413 ( 2006 ) aizenman m. , schenker j. h. , friedrich r. m. , and hundertmark d. : _ finite - volume fractional - moment criteria for anderson localization _ , comm .
phys . , 224(1):219253 , 2001 .
dedicated to joel l. lebowitz .
bellissard j. , hislop p. , and stolz g. : _ correlation estimates in the anderson model .
_ , _ j. stat .
_ , 129(4):649662 , 2007 .
bloch , i. , dalibard , j. , zwerger , w. : _ many - body physics with ultracold gases _ , rev .
mod . phys .
80 ( 2008 ) , 885 combes jm .
, germinet f. , klein a. : _ generalized eigenvalue - counting estimates for the anderson model
_ j. stat .
phys . , 135(2):201216 , 2009 .
combes , j.m . ,
germinet , f. , klein , a. : _ poisson statistics for continuum random schrdinder operator _ , to appear in j. stat .
combes jm . , hislop p. , frdric klopp f. : _ an optimal wegner estimate and its application to the global continuity of the integrated density of states for random schrdinger operators _ , duke math .
j. , 140(3):469498 , 2007 .
davis , k.b . ,
mewes , m.o . ,
andrews , m.r . ,
van druten , n.j . , durfee , d.s . ,
kurn , d.m . ,
ketterle , w. : _ bose - einstein condensation in a gas of sodium atoms _ , phys .
lett . 75 ( 1995 ) , 3969 .
einstein a. , _ quantentheorie des einatomigen idealen gases _ , sitzber .
akad . wiss . , 261267 ( 1924 ) , and 314 ( 1925 ) .
fernndez , r. , frhlich , j. , ueltschi , d. : _ mott transition in lattice boson models _
266 ( 2006 ) , 777 .
germinet , f. , klein , a. : _ new characterizations of the region of complete localization for random schrdinger operators _ , j. stat
. phys . * 122 * , 73 - 94 ( 2006 ) germinet , f. , klopp , f. : _ level , level spacing and localization center statistics for random schrdinger operators in the regime of anderson localization _ , in preparation graf gm . ,
vaghi a. : _ a remark on the estimate of a determinant by minami _ , lett .
phys . , 79(1):1722 , 2007 .
killip , s. , nakano , f. : _ eigenfunction statistics in the localized anderson model _ ,
henri poincar , 8 , 27 - 36 ( 2007 ) kirsch , w. : _ invitation to random schrdinger operators _ , panoramas et syntheses 25 ( 2008 ) , 1 - 119 .
w. kirsch and f. martinelli . large deviations and lifshitz singularities of the integrated density of states of random hamiltonians .
, 89:2740 , 1983 .
kirsch , w. , metzger , b.:_the integrated density of states for random schrdinger operators _ in spectral theory and mathematical physics : a festschrift in honor of barry simon s 60th birthday , 649 - 696 , proc .
pure math .
, * 76 * , part 2 , amer .
soc . , providence , ri , 2007 .
kirsch w. , simon b. : _ lifshitz tails for periodic plus random potentials _
, j. statist . phys . , 42(5 - 6):799808 , 1986 .
kirsch w. , simon b. : _ comparison theorems for the gap of schrdinger operators _ , j. funct .
, 75(2):396410 , 1987 .
ketterle w. , van druten , n.j . : _ evaporative cooling of trapped atoms _ , in : b. bederson , h. walther , eds .
, advances in atomic , molecular and optical physics , * 37 * ( 1996 ) , 181236 , academic press .
klopp , f. : _ weak disorder localization and lifshitz tails _ ,
( 2002 ) , 125 - 155 .
klopp , f. : _ resonances for `` large '' ergodic systems _ , in preparation lee , d.k.k . , gunn , j.m.f . : _ bosons in a random potential : condensation and screening in a dense limit _ , j. phys . : condens .
matter * 2 * ( 1990 ) , 7753 - 7768 .
lenoble o. , pastur l.a .
, zagrebnov v.a . : _ bose - einstein condensation in random potentials _ , comptes rendus de lacadmie des sciences ( paris ) , physique * 5 * ( 2004 ) , 129 - 142 .
lewin , m. : private communication lieb , e.h .
, seiringer , r. , solovej , j.p . ,
yngvason , j. : _ the mathematics of the bose gas and its condensation _ , oberwolfach seminars , * 34 * , birkhauser verlag , basel , 2005 .
lugan , p. , clement , d . ,
bouyer , p. : aspect , a. , lewenstein , m. , sanchez - palencia l. : _ ultracold bose gases in 1d disorder : from lifshits glass to bose - einstein condensate _ , phys .
* 98 * , 170403 ( 2007 ) .
minami n : _ local fluctuation of the spectrum of a multidimensional anderson tight binding model _ , comm .
, 177(3):709725 , 1996 .
sanchez - palencia l. : _ strongly correlated bosons in optical lattices _ , http://atomoptic.iota.u-psud.fr/members/lsp/www/teach/hubbard.pdf simon b. , the statistical mechanics of lattice gases , princeton univ . press ( 1993 ) smerzi , a. , trombettoni , a. : _ discrete solitons and breathers with dilute bose - einstein condensates _ , phys .
* 86 * , 11 ( 2001 ) , 2353 - 2356 smerzi , a. , trombettoni , a. : _ nonlinear tight - binding approximation for bose - einstein condensates in a lattice _ ,
arxiv : cond - mat/0309285v1 stollmann , p. : _ caught by disorder .
bound states in random media _ , progress in mathematical physics , * 20 * birkhuser boston , 2001 stollmann , p. , stolz , g. : _ singular spectrum for multidimensional schrdinger operators with potential barriers _ , j. operator theory ,
* 32 * ( 1995 ) , 91 - 109 sznitman , a.s .
brownian motion , obstacles and random media .
springer monographs in mathematics .
springer - verlag , berlin , 1998 .
ueltschi , d. : _ geometric and probabilistic aspects of boson lattice models _ , progr .
* 51 * ( 2002 ) , 363 - 391 birkhuser , boston .
ivan veseli . existence and regularity properties of the integrated density of states of random schrdinger operators , volume 1917 of _ lecture notes in mathematics_. springer - verlag , berlin , 2008 . | for discrete and continuous gross - pitaevskii energy functionals with a random background potential , we study the gross - pitaevskii ground state .
we characterize a regime of interaction coupling when the gross - pitaevskii ground state and the ground state of the random background hamiltonian asymptotically coincide . |
stellar population models of absorption line indices are a key tool for the analysis of star cluster and galaxy absorption spectra .
they are used to derive the fundamental stellar population properties such as age , metallicity and element abundance ratios . in particular , optical absorption line diagnostics in the spectra of evolved stellar populations have successfully been adopted in the past in studies on galaxy evolution ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) and globular cluster formation ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the lick / ids system @xcite is the standard set of absorption line indices that has been used extensively during the last two decades for studying absorption features of stellar populations .
this system consists of index definitions for 25 prominent absorption features between 4000 and @xmath2 present in the spectra of evolved stellar populations . for studies of galaxy and star cluster evolution , absorption lines need to be modelled for stellar populations ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
a convenient way goes through the use of empirical calibrations .
this is motivated by the fact that theoretical model atmospheres are known to suffer from incomplete line lists and continuum uncertainties .
( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
empirical calibrations on the other hand have the disadvantage to be hardwired to the chemical abundance pattern of the milky way , which can be overcome in a semi - empirical approach as in the models by @xcite , @xcite and @xcite empirical calibrations can be inserted in the models in two ways . in the first and most widely used approach , absorption line indices enter stellar population modelling through calibrations of the empirical relationship between the indices and the stellar atmospheric parameters t@xmath3 , @xmath4 and [ fe / h ] as provided by stellar libraries . as these calibrations are usually obtained through polynomial fitting procedures
they are commonly referred to as fitting functions. the quality of the final stellar population model critically depends on the accuracy with which these relationships can be inferred from stellar libraries , i.e. the coverage of stellar parameter space and the reliability of the index measurements .
the computational procedure with which the fitting functions are determined is a further crucial step in producing accurate models .
a number of studies in the literature are devoted to such empirical calibrations for various stellar libraries , either for the lick indices , parts of the lick indices or other prominent absorption features @xcite .
alternatively to the use of fitting functions , absorption line indices can be measured directly on the synthetic spectral energy distribution ( sed ) from stellar population models that are based on empirical stellar libraries .
the benefit of this method is that the full sed can be compared pixel - by - pixel to observations ( e.g. * ? ? ?
* ; * ? ? ?
the major strength of fitting functions , instead , lies in the fact that they allow for interpolation between well populated regions of stellar parameter space which increases the accuracy of the model in stellar parameter space that is only sparsely sampled by empirical stellar libraries .
moreover , each absorption index or spectral feature is represented by an individual fitting function , which is optimised to best reproduce its behaviour in stellar parameter space .
fitting functions are also easier to implement in a stellar population synthesis code , and models based on fitting functions are better comparable . the widely used fitting functions of @xcite and @xcite are based on the lick / ids stellar library @xcite .
they are adopted in most stellar population models @xcite in the literature .
other fitting functions based on the same stellar library exist @xcite and lead to overall consistent results in the final stellar population model @xcite .
major progress has been made with the advent of a new generation of stellar libraries @xcite that have led to considerable improvements regarding coverage of stellar parameter space , spectral resolution , signal - to - noise ratio , and flux calibration .
in particular the latter is a critical step forward . as the lick / ids system is not flux calibrated
, observations have to be re - calibrated onto the lick / ids system through comparison with lick standard stars .
this requirement hampers the analysis of data samples for which spectra of such calibration stars are either not available at sufficient quality or do not cover the appropriate rest - frame wavelength range .
this problem is most imminent in high redshift observations and in galaxy redshift surveys such as the sloan digital sky survey @xcite .
the new flux calibrated libraries allow the analysis of flux calibrated spectra at any redshift without spectroscopic standard stars .
flux - calibrated stellar libraries in the literature that are suitable for stellar population modelling include the _ jones _
@xcite , _ elodie _
@xcite , _ stelib _ @xcite and _ miles _ @xcite libraries .
the _ miles _
library is particularly well suited for stellar population modelling of absorption line indices owing to its favourable combination of spectral resolution , wavelength range , stellar parameter coverage , and quality of flux calibration . in this paper
we present new lick index fitting functions based on the _ miles _ stellar library . to take advantage of the full spectral resolution of the _ miles _ library we have produced fitting functions for both the lower lick / ids resolution ( @xmath5 fwhm ) and the higher resolution of the _ miles _ library ( 2.3 fwhm ) .
a new version of the tmb stellar population model of absorption line indices based on these new fitting functions will be presented in a subsequent paper .
this paper is organized as follows . in section [ stlib ]
we present the lick indices measured on the _ miles _ library and a quality evaluation of the index measurements .
we discuss offsets between the flux calibrated miles and the lick / ids systems .
the empirical fitting method is presented in section [ fitf ] along with the resulting fitting functions . in section [ comps ]
we compare the fitting functions of this work with fitting functions from the literature .
we summarise in section [ concs ] . [ cols= " > ,
< , < , < , < , < , < , < , > , < , < , > , < , < , > , < " , ] * h@xmath6 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ warm _ _ giants _ extend well beyond the 2rms level where this work show much weaker indices .
we find both _
warm _ and _ cold _ _ dwarfs _ to show stronger indices for this work , even extending beyond the 2rms level for the metal - poor and metal - rich ends , respectively .
otherwise , this work show slightly weaker indices extending to the 1rms level .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ _ cold _ _ dwarfs _ show weaker indices for this work , beyond the 1rms level .
_ warm _ and _ intermediate _ temperature _ dwarfs _ show stronger indices for this work out to the 2rms level in the metal - poor regime .
_ intermediate _ temperature _ giants _ show stronger indices out to the 2rms level at the ends of the metallicity scale .
otherwise are mainly differences within the 1rms level found .
+ * h@xmath7 * + _ wff comparison ( fig . [ wffcomp ] ) : _ the most obvious difference is found for _ warm _ _ giants _ where this work show much weaker indices , extending well beyond the 2rms level .
otherwise are differences mainly within the 1rms level , except for the metal - rich end of _ cold _ and _ warm _ _ dwarfs _ that show stronger indices for this work beyond the 2rms level .
+ _ sff comparison ( fig . [ sffcomp ] ) : _ this work shows in general stronger indices in the metal - poor regime , beyond the 1 rms level for _ intermediate _ temperature and _ warm _ _ dwarfs _ and beyond the 2rms level for _ intermediate _ temperature _ giants_. in the metal - rich regime we instead find weaker for this work , out to the 2rms level for _ intermediate _ temperature and _ cold _
_ giants_. + + + + + + + + + + + + + * cn@xmath8 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ the _ warm _
end for _ giants _ show significantly stronger indices for this work , extending well beyond the 2rms level . otherwise ,
this work show in general stronger indices at the metal - poor end and weaker indices at the metal - rich end , out to the 2rms level in both cases .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ _ intermediate _ temperature _ giants _ and _ dwarfs _ show weaker and stronger indices for this work , respectively , at the metal - rich end .
otherwise are agreements within the 1rms level mainly found .
+ * cn@xmath9 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ similar to the previous index , but _ warm _ _ dwarfs _ show weaker indices for the entire metallicity scale for this work , out to the 2rms level .
+ _ sff comparison ( fig . [ sffcomp ] ) : _ due to problems with implementing the sffs we can not make a reliable comparison . + * ca4227 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ cold _ _ dwarfs _ show stronger indices for this work , extending out to the 2rms level in the metal - poor regime , while _ cold _ _ giants _ instead show weaker indices for this work out to the 2rms level at the metal - poor end .
_ warm _ _ giants _ show stronger indices for this work beyond the 1rms level at the metal - poor end . _
cold _ _ dwarfs _ show stronger indices for this work , even extending beyond the 2rms level at the metal - poor end .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ the most prominent difference is found for _ cold _ _ giants _ in the metal - poor regime , extending well beyond the 2rms level . _
cold _ _ dwarfs _ show stronger indices for this work at the metal - rich end , beyond the 1rms level .
otherwise are differences within the 1rms level found .
+ * g4300 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ _ warm _ _ giants _ extend well beyond the 2rms level with stronger indices for this work .
metal - poor _ cold _ _ giants _ show stronger indices for this work , extending to the 2rms level .
_ cold _ metal - poor _ dwarfs _ extend beyond the 1rms level , showing weaker indices for this work . _
cold _ and _ warm _ metal - rich _ dwarfs _ show stronger and weaker indices for this work , respectively , beyond the 1rms level .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ differences beyond the 1rms and 2rms levels are found in several regimes , strongest at the ends of the metallicity scale .
+ * h@xmath10 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ warm _ _ giants _ show significantly weaker indices for this work , well beyond the 2rms level . _
warm _ _ dwarfs _ and _ cold _ _ giants _ show stronger indices for this work out to the 2rms level . otherwise are mainly differences within the 1rms level found .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ the most significant difference is found for _ intermediate _ temperature _ dwarfs _ , showing weaker indices for this work in the metal - poor regime well beyond the 2rms level .
otherwise are differences mainly found around the 1rms level .
+ * h@xmath11 * + _ wff comparison ( fig . [ wffcomp ] ) : _ weaker indices are found for this work for _ warm _ _ giants _ and _ cold _ _ dwarfs _ beyond the 2rms and 1rms level , respectively . otherwise are mainly differences within the 1rms level found .
+ _ sff comparison ( fig . [ sffcomp ] ) : _ due to problems with implementing the sffs we can not make a reliable comparison .
+ * fe4383 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ warm _ _ giants _ and _ warm _ _ dwarfs _ show stronger indices out to the 2rms level at the metal - poor end .
_ cold _ _ giants _ show weaker indices , out to the 2rms level at the metal - poor and metal - rich ends . otherwise are mainly differences within the 1rms level found .
+ _ sff comparison ( fig . [ sffcomp ] ) : _ _ cold _ _ dwarfs _ show significantly stronger indices for this work , well beyond the 2rms level .
_ warm _ _ dwarfs _ instead show weaker indices , out to the 2rms level .
otherwise are mainly differences within the 1rms found .
+ + + * ca4455 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ cold _ _ giants _ show stronger indices for this work , extending well beyond the 2rms level in the metal - rich regime .
_ cold _ _ dwarfs _ show weaker indices for this work , extending well beyond the 2rms level .
warm _ regime show stronger indices for this work , extending beyond the 1rms level in the metal - poor regime .
+ * fe4531 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ cold _ _ dwarfs _ show stronger indices for this work at the 1rms level for the metal - poor end and increasing well beyond the 2rms level at the metal - rich end .
_ intermediate _ temperature _ giants _ show stronger indices for this work , out to the 2rms level at the metal - rich end .
+ * c@xmath12 * + _ wff comparison ( fig . [ wffcomp ] ) : _ the metal - rich end show weaker indices for this work , mainly down to the 1rms level .
the metal - poor end show weaker and stronger indices for this work extending beyond the 1rms level for the _ warm _ bins and _ cold _ _ dwarfs _ , respectively .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ _ cold _ _ giants _ show weaker indices for this work beyond the 1rms level in the metal - rich regime . _
intermediate _ temperature _ dwarfs _ show stronger indices for this work at the metal - rich end out to the 2rms level .
+ * h@xmath13 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ we find this work to show weaker indices for _ warm _ _ giants _ well beyond the 2rms level .
_ cold _ _ dwarfs _ show stronger indices for this work beyond the 2rms level . _
cold _ _ giants _ show weaker indices for this work beyond the 1rms level .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ _ cold _ _ dwarfs _ show the biggest differences right beyond the 1 rms level .
otherwise are differences within the 1rms level are found .
+ * fe5015 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ cold _ _ giants _ show weaker indices for this work , beyond the 2rms level in the metal - poor regime . _
intermediate _ temperature _ giants _ show stronger indices for this work beyond the 2rms level in the metal - rich regime .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ _ cold _ _ giants _ show significantly weaker indices for this work , well beyond the 2rms level .
this work shows weaker indices for _ cold _ _ dwarfs _ beyond the 1rms level in the metal - poor regime .
otherwise are differences within the 1rms mainly found .
+ * mg@xmath8 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ the _ cold _ _ dwarfs _ show stronger indices for this work , out to the 2rms level in the metal - poor regime .
the _ cold _ _ giants _ show stronger indices for this work at intermediate metallicities .
no differences found beyond the 2rms level .
+ * mg@xmath9 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ _ cold _ _ dwarfs _ show stronger indices for this work beyond the 1rms level in the metal - poor regime .
_ cold _ _ giants _ show stronger indices for this work beyond the 1rms level for intermediate metallicities .
otherwise are mainly differences within the 1rms level found .
+ _ sff comparison ( fig . [ sffcomp ] ) : _ the _ cold _ end show weaker indices for this work in the metal - poor regime , beyond the 2rms level .
_ cold _ _ dwarfs _ instead show stronger indices for this work beyond the 2rms level at the metal - rich end .
+ * mgb * + _ wff comparison ( fig .
[ wffcomp ] ) : _ _ cold _ and _ intermediate _ temperature _ giants _ show weaker indices for this work , extending beyond the 2rms level . _
intermediate _ temperature and _ cold _ _ dwarfs _ show weaker and stronger indices , respectively , for this work in the metal - poor regime , beyond the 1rms level .
+ _ sff comparison ( fig .
[ sffcomp ] ) : _ due to problems implementing the sffs we can not make a reliable comparison . + * fe5270 * + _ wff comparison ( fig . [ wffcomp ] ) : _ the _ warm _ end show stronger indices beyond the 2rms level in the metal - poor regime .
_ cold _ _ giants _ show weaker indices for this work beyond the 2rms level .
the metal - rich end show stronger indices for this work beyond the 1rms level for _ intermediate _ temperature _ cold _ _
dwarfs_. + _ sff comparison ( fig .
[ sffcomp ] ) : _ we find weaker indices for this work beyond the 1rms level for _ cold _ _ giants_. stronger indices for this work beyond the 2rms level are found for _ cold _ _ dwarfs _ in the intermediate metallicity regime .
otherwise are differences well within the 1rms level found .
+ * fe5335 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ cold _ _ giants _ show weaker indices for this work well beyond the 2rms level .
warm _ end show stronger indices out to the 1rms level at intermediate metallicities .
_ cold _ _ dwarfs _ show weaker and stronger indices out to the 1rms level a the metal - poor and metal - rich ends , respectively .
+ _ sff comparison ( fig . [ sffcomp ] ) : _ _ cold _ _ dwarfs _ show stronger indices beyond the 2rms level at intermediate metallicities and weaker indices beyond the 2rms level at the metal - rich end .
_ cold _ _ giants _ show weaker indices for this work beyond the 1rms level . _
intermediate _ temperature _ dwarfs _ instead show stronger indices for this work , beyond the 1rms level in the metal - rich regime .
+ * fe5406 * + _ wff comparison ( fig . [ wffcomp ] ) : _ the _ cold _ end show stronger indices for this work beyond the 1rms level in the metal - rich regime .
_ intermediate _ temperature _ giants _ show weaker indices for this work beyond the 1rms level .
no differences found beyond the 2rms level .
+ * fe5709 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ _ cold _ _ dwarfs _ show stronger indices for this work , extending beyond the 2rms level at the metal - rich end . otherwise , no significant differences beyond the 1rms level .
+ * fe5782 * + _ wff comparison ( fig .
[ wffcomp ] ) : _ _ warm _ _ giants _ show stronger indices for this work in the metal - poor regime , extending beyond the 2rms level at the metal - poor end .
_ cold _ _ giants _ show stronger indices for this work regime , beyond the 1rms level .
_ warm _ _ dwarfs _ show stronger indices for this work beyond the 1rms level at the metal - poor end .
+ * nad * + _ wff comparison ( fig .
[ wffcomp ] ) : _ the _ cold _ end show stronger indices for this work extending well beyond the 2rms level , especially in the metal - poor regime . _
intermediate _ temperature _ giants _ show stronger indices for this work , extending beyond the 2rms level in the metal - rich regime .
+ * tio@xmath14 * + _ wff comparison ( fig . [ wffcomp ] ) : _ _ cold _ _ dwarfs _ show weaker indices for this work , extending well beyond the 1rms level at the metal - rich end . otherwise , no differences found beyond the 1rms level . + *
tio@xmath15 * + _ wff comparison ( fig . [ wffcomp ] ) : _ we find significantly weaker indices for this work for _ cold _ _ dwarfs _ , extending very far beyond the 2rms level .
otherwise , no significant differences found beyond the 1rms level .
we have derived new empirical fitting functions for the relationship between lick absorption indices and stellar atmospheric parameters ( t@xmath3 , [ fe / h ] and @xmath4 ) described by the _ miles _ library of stellar spectra , both for the resolution of the _ miles _ library and for the resolution of the lick / ids library .
the _ miles _ library consists of 985 stars selected to produce a sample with extensive stellar parameter coverage .
the _ miles _
library was also chosen because it has been carefully flux - calibrated , making standard star derived offsets unnecessary .
this becomes important when comparing stellar population models to high redshift data where no resolved individual stars are available .
we find the index measurements of the _ miles _ spectra to have very high quality in terms of observational index errors .
these errors are also found to be significantly smaller than for the lick / ids library .
this was expected since the _ miles _ library was observed nearly thirty years after the lick / ids library . given the high quality of the index measurements , index errors should not be the major error sources for the final fitting functions .
we instead find indications that the stellar parameter estimates are significant error sources .
lick index offsets between the _ miles _ library and the classic lick / ids library are derived in order to be able to compare stellar population models based on this work with models in the literature .
we find these offsets to be dependent on index strength and have therefore derived least - square fits for the residual between the two libraries .
offset to the lick / ids library are also derived for the flux - calibrated _ elodie _ and _ stelib _ libraries .
we find clear offset deviations between the libraries .
the largest deviations are found for the _ stelib _ library compared to the other two libraries , which is also the library having least stars in common with the lick / ids library .
the deviations in offsets found between the three libraries undermine the derivation of universal offsets between the lick / ids and these flux - calibrated systems .
we compare the fitting functions of this work to fitting functions in the literature , namely the fitting functions of @xcite , @xcite and @xcite .
generally we find good agreement within the rms of the residuals between the data and the fitting functions of this work .
the differences found in the comparisons vary significantly from index to index and especially from one stellar parameter region to another for individual indices .
however , the major differences are found in the outskirts of stellar parameter space , i.e. at the temperature and metallicity ends .
this is probably due to a low number of data points in these regimes for the stellar libraries , inducing uncertainties which result in the major differences found . in a forthcoming paper ( thomas et al . in prep . )
the fitting functions of this work will be implemented in stellar population models following the techniques of @xcite and @xcite .
a user friendly fortran 90 code is available online at www.icg.port.ac.uk/@xmath16johanssj to easy the implementation of our fitting functions in population synthesis codes .
we thank patricia snchez - blzquez for help with the _ miles _ library .
we thank ricardo schiavon and rita tojeiro for useful discussions regarding the empirical fitting method .
we also thank an anonymous referee for a prompt and constructive report .
annibali f. , bressan a. , rampazzo r. , zeilinger w. w. , danese l. , 2007 , a&a , 463 , 455 borges a. c. , idiart t. p. , de freitas pacheco j. a. , thvenin f. , 1995 , aj , 110 , 2408 brodie j. p. , strader j. , denicol g. , beasley m. a. , cenarro a. j. , larsen s. s. , kuntschner h. , forbes d. a. , 2005 , aj , 129 , 2643 bruzual g. , charlot s. , 2003 , mnras , 344 , 1000 burstein d. , faber s. m. , gaskell c. m. , krumm n. , 1984 , apj , 287 , 586 buzzoni a. , gariboldi g. , mantegazza l. , 1992 , aj , 103 , 1814 buzzoni a. , mantegazza l. , gariboldi g. , 1994 , aj , 107 , 513 carney b. w. , 1996 , pasp , 108 , 900 cenarro a. j. , gorgas j. , cardiel n. , vazdekis a. , peletier r. f. , 2002 , mnras , 329 , 863 cenarro a. j. , peletier r. f. , snchez - blzquez p. et al . , 2007 , mnras , 374 , 664 coelho p. , bruzual g. , charlot s. , weiss a. , barbuy b. , ferguson j. w. , 2007 , mnras , 382 , 498 davies r. l. , sadler e. m. , peletier r. f. , 1993 , mnras , 262 , 650 edvardsson b. , andersen j. , gustafsson b. , lambert d. l. , nissen p. e. , tomkin j. , 1993 , a&a , 275 , 101 faber s. m. , friel e. d. , burstein d. , gaskell c. m. , 1985 , apjs , 57 , 711 forbes d. a. , beasley m. a. , brodie j. p. , kissler - patig m.,2001 , apj , 563 , 143 fuhrmann k. , 1998 , a&a , 338 , 161 gorgas , j. , faber s. m. , burstein d. , gonzalez j. j. , courteau s. , prosser c. , 1993 , apjs , 86 , 153 gorgas j. , cardiel n. , pedraz s. , gonzlez j. j. , 1999 a&as , 139 , 29 henry r. b. c. , & worthey g. , 1999 , pasp , 111 , 919 jones l. a. , 1999 , ph.d .
thesis , univ .
north carolina kissler - patig m. , brodie j. p. , schroder l. l. , forbes d. a. , grillmair c. j. , huchra j. p. , 1998
, aj , 115 , 105 korn a. j. , maraston c. , thomas d. , 2005 , a&a , 438 , 685 kuntschner h. , 2000 , mnras , 315 , 184 kuntschner h. , 2001 , ap&ss , 276 , 885 kuntschner h. , davies r. l. , 1998 , mnras , 295 , 29 kuntschner h. , ziegler b.l . , sharples r. m. , worthey g. , fricke k. j. , 2002 , a&a , 395 , 761 le borgne j. f. , bruzual g. , pell r. et al . , 2003 , a&a , 402 , 433 lee h. , worthey g. , dotter a. , 2009 , aj , 138 , 1442 leitherer c. , schaerer d. , goldader j. d. et al . , 1999 ,
apjs , 123 , 3 maraston c. , 1998 , mnras , 300 , 872 maraston c. , 2005 , mnras , 362 , 799 maraston c. , greggio l. , renzini a. , ortolani s. , saglia r. p. , puzia t. h. , kissler - patig m. , 2003 , a&a , 400 , 823 maraston c. , nieves colmenrez l. , bender r. , thomas d. , 2009 , a&a , 493 , 425 milone a. , sansom a. e. , snchez - blzquez p. , submitted , in bruzual g. r. , charlot s. , eds , proc .
262 , stellar populations - planning for the next decade .
panter b. , jimenez r. , heavens a. f. , charlot s , 2007 , mnras , 378 , 1550 prugniel , ph . , soubiran c. , 2001 , a&a , 369 , 1048 puzia t. h. , saglia r. p. , kissler - patig m. , maraston c. , greggio l. , renzini a. , & ortolani s. , 2002 , a&a , 395 , 45 puzia t. h. , kissler - patig m. , thomas d. , maraston c. , saglia r. p. , bender r. , goudfrooij p. , hempel m. , 2005 , a&a , 439 , 997 rodrguez - merino l. h. , chavez m. , bertone e. , buzzoni a. , 2005 , apj , 626 , 411 rose j. a. , bower r. g. , caldwell n. , ellis r. s. , sharples r. m. , teague p. , 1994
, aj , 108 , 2054 snchez - blzquez p. , peletier r. f. , jimnez - vicente j. et al . , 2006 , mnras , 371 , 703 snchez - blzquez p. , jablonka p. , noll s. et al . , 2009 , a&a , 499 , 47 schiavon r. p. , 2007 ,
apjs , 171 , 146 thomas d. , maraston c. , bender r. , 2003 , mnras , 343 , 279 thomas d. , maraston c. , korn a. , 2004 , mnras , 351 , 19 thomas d. , maraston c. , bender r. , mendes de oliveira c. , 2005 , apj , 621 , 673 tojeiro r. , heavens a. f. , jimenez r. , panter b. , 2007 , mnras , 381 , 1252 trager s. c. , worthey g. , faber s. m. , burstein d. , gonzalez j. j. , 1998 , apjs , 116 , 1 trager s. c. , faber s. m. , worthey g. , gonzlez j. j. , 2000 , aj , 119 , 1645 tripicco m. j. , bell r. a. , 1995 , aj , 110 , 3035 vazdekis a. , 1999 , apj , 513 , 224 vazdekis a. , casuso e. , peletier r. f. , beckman j. e. , 1996 , apjs , 106 , 307 vazdekis a. , peletier r. f. , beckman j. e. , casuso e. , 1997 , apjs , 111 , 203 walcher c. j. , coelho p. , gallazzi a. , charlot s. , 2009 , mnras , 398 , 44 worthey g. , 1994 , apjs , 95 , 107 worthey g. , 1998 , pasp , 110 , 888 worthey g. , ottaviani d. l. , 1997 , apjs , 111 , 377 worthey g. , faber s. m. , gonzalez j. j. , 1992 , apj , 398 , 69 worthey g , faber s. m. , gonzalez j. j. , burstein d. , 1994 , apjs , 94 , 687 york d. g. , adelman j. , anderson j. e. jr .
et al . , 2000 , aj , 120 , 1579
short version , full appendix can be found at www.icg.port.ac.uk/@xmath16johanssj + +
short version , full appendix can be found at www.icg.port.ac.uk/@xmath16johanssj & & + term & @xmath17 & @xmath18 & @xmath19 & @xmath20 & @xmath21 & @xmath22 & @xmath23 & @xmath24 + const . & -8.002 & -136.8 & 36.39 & -127.4 & -3.369 & -134.7 & -49.75 & -41.96 + @xmath25 & -12.19 & 487.8 & x & 23.06 & 16.05 & 596.3 & 297.8 & 42.42 + & x & -2.367 & 41.22 & x & x & -13.31 & 31.61 & x + @xmath4 & -4.330 & 21.43 & 2.662 & x & x & x & -0.5889 & x + @xmath26 & x & -550.1 & -88.27 & 103.1 & 13.26 & -922.4 & -423.0 & -9.752 + @xmath27 & x & x & 0.5080 & x & x & -3.267 & x & x + @xmath28 & 5.072 & x & x & x & 0.1957 & x & x & x + @xmath25[fe / h ] & x & x & -78.94 & x & x & 31.04 & -68.15 & x + @xmath29 & x & -44.24 & -6.149 & x & 0.2840 & 26.90 & x & x + @xmath4 & x & 0.2318 & x & x & x & x & x & x + @xmath30 & 273.0 & 195.6 & 48.49 & 49.23 & x & 448.91 & 175.3 & x + @xmath31 & x & 0.5286 & 0.3213 & x & x & -0.1733 & -0.1461 & x + @xmath32 & -0.8813 & x & x & x & x & x & x & x + @xmath26[fe / h ] & x & x & 34.99 & x & x & -20.90 & 33.01 & x + @xmath33 & -16.26 & 22.55 & 3.167 & x & x & -17.20 & -1.042 & x + @xmath25[fe / h]@xmath27 & x & 1.297 & x & x & x & 2.919 & -0.7101 & x + @xmath34 & x & x & x & x & x & 0.02043 & x & x + @xmath35 & 2.150 & x & x & x & x & -1.628 & x & x + @xmath36 & -275.2 & x & x & -108.1 & x & x & x & x + @xmath37 & x & x & x & 31.52 & x & x & x & x + rms & 1.033 & 1.076 & 0.8149 & 0.6510 & 0.8884 & 0.5558 & 0.5954 & 1.477 + n & 96 & 346 & 359 & 41 & 49 & 278 & 276 & 33 + | stellar population models of absorption line indices are an important tool for the analysis of stellar population spectra .
they are most accurately modelled through empirical calibrations of absorption line indices with the stellar parameters effective temperature , metallicity , and surface gravity , the so - called fitting functions . here
we present new empirical fitting functions for the 25 optical lick absorption line indices based on the new stellar library miles .
the major improvements with respect to the lick / ids library are the better sampling of stellar parameter space , a generally higher signal- to - noise , and a careful flux calibration .
in fact we find that errors on individual index measurements in miles are considerably smaller than in lick / ids .
instead we find the rms of the residuals between the final fitting functions and the data to be dominated by errors in the stellar parameters .
we provide fitting functions for both lick / ids and miles spectral resolutions , and compare our results with other fitting functions in the literature .
a fortran 90 code is available online in order to simplify the implementation in stellar population models .
we further calculate the offsets in index measurements between the lick / ids system to a flux calibrated system . for this purpose
we use the three libraries miles , elodie , and stelib .
we find that offsets are negligible in some cases , most notably for the widely used indices h@xmath0 , mg@xmath1 , fe5270 , and fe5335 . in a number of cases , however , the difference between flux calibrated library and lick / ids is significant with the offsets depending on index strengths .
interestingly , there is no general agreement between the three libraries for a large number of indices , which hampers the derivation of a universal offset between the lick / ids and flux calibrated systems .
[ firstpage ] line : profiles stars : abundances stars : fundamental parameters |
calcium imaging , first used to measure the activity of neurons in the early 1990s @xcite , has been successfully applied throughout the nervous system .
it allows us to see the behavior of neurons in awake behaving mice , using either chemical or genetic calcium indicators , with confocal microscopy , two - photon microscopy , or wearable imaging devices @xcite . as a result , it is an increasingly useful tool for identifying the neural substrate of mouse behaviors . however , calcium imaging videos have difficult noise properties , including white noise and motion artifacts which must be corrected in a preprocessing step before proper analysis can be undertaken .
motion correction is the first step in the analysis of calcium images .
after they are motion - corrected , rois are identified , and time - activity graphs are made from each roi . if the motion - correction is low - quality , then the time - activity graphs suffer , and the reconstructed networks may have errors . for real time
closed loop operation , if the motion correction is slow , it can not be done while the mouse is in the microscope , and the experiment fails .
turboreg @xcite is a commonly used algorithm to do motion correction .
it uses a downsampling strategy , which is prerequisite for speed , and it uses a template image , which is necessary for accuracy .
we have developed a similar method , called _ moco _ ( motion corrector ) , which adopted both strategies , since correcting one image against the next in the stack results in unacceptable roundoff errors .
other approaches use hmms @xcite , @xcite or other techniques @xcite , @xcite , @xcite , @xcite , @xcite .
@xcite is the only one similar mathematically to , and may be slightly faster than , moco , but it has accuracy problems ( see figure 2 ) .
moco uses downsampling and a template image , and it can be called from imagej .
however , it is faster than turboreg @xcite at translation - based motion correction because it uses dynamic programming and two - dimensional fft - acceleration of two - dimensional convolutions .
@xcite also uses the fft approach but uses a different objective function that does not require dynamic programming ; we believe that our approach is more robust to corrupted data , see figure 2 .
image stabilizer is as fast for small images , but is very slow for standard - size images .
running on our own datasets , moco appears faster than all approaches compatible with imagej .
moco corrects every image in the video by comparing every possible translation of it with the template image , and chooses the one which minimizes the @xmath0 norm of the difference between the images in the overlapping region , @xmath1 , divided by the area of @xmath1 . the fact that it is
so thorough makes it robust to long translations in the data .
more complicated non - translation image warps are usually unnecessary for fixing calcium images , which suffer from spurious translations , which moco corrects , and spurious motion in the z - direction , which is very hard to correct .
our approach also uses cache - aware upsampling : when an image is aligned with the template in the downsampled space , it must be jittered when it is upsampled to see which jitter best aligns with the upsampled template .
we do this in such a way that data that is used recently is reused immediately , making the implementation faster . hence , moco is an efficient motion correction of calcium images , and is likely to become a useful tool for processing calcium imaging movies .
let @xmath2 , for @xmath3 and @xmath4 be an image in the stack .
we assume @xmath5 is downsampled if it is larger than @xmath6 .
let @xmath7 be the template image against which to align @xmath5 .
we want to pick @xmath8 such that @xmath9 , where @xmath10 is input by the user , and @xmath11 is minimal , where @xmath12 is the set of ordered pairs of integers @xmath13 such that @xmath14 , @xmath15 , @xmath16 , and @xmath17 .
if we do this for every image @xmath5 in the stack , we have then motion corrected the video , and we are done , up to a short upsampling step . to upsample , multiply the optimal @xmath8 by @xmath18 and do a local search to minimize @xmath19 on the finer grid .
now , @xmath20 the first two sums can be computed via dynamic programming .
let s consider @xmath5 when @xmath21 and @xmath22 are negative .
let @xmath23 we have that @xmath24 hence , the first two sums can be computed for all @xmath8 in @xmath25 time , which is unaffected by a constant amount of downsampling .
it suffices to compute for all @xmath8 such that @xmath9 , @xmath26 let @xmath27 be @xmath28 rotated @xmath29 degrees . using matlab notation , let @xmath30;{\rm zeros(w , n+w)}]),\ ] ]
@xmath31;{\rm zeros(w , n+w)}]).\ ] ] commas denote horizontal concatenation , semicolons denote vertical concatenation , and @xmath32 is an @xmath33 matrix of zeros . for equally sized matrices
@xmath34 , @xmath35 , let @xmath36 mean @xmath37 .
then @xmath38 is a rearrangement of @xmath39 .
since @xmath40 s are fast , that means @xmath39 can be computed for all @xmath8 in @xmath41 time .
hence , after upsampling , the entire video can be aligned in @xmath42 time , where @xmath43 is the number of slides in the video .
after @xmath8 are chosen to minimize @xmath19 , they are multiplied by two multiple times to upsample .
every time they are multiplied by @xmath18 , @xmath44 are computed for @xmath45 to see which @xmath46 and @xmath47 are minimal .
these nine evaluations of @xmath48 are done with a cache - aware algorithm for speed .
we compare moco in speed to turboreg @xcite on its translation mode , using both the `` fast '' and `` accurate '' settings .
we also compare it to image stabilizer using its default settings @xcite ( it can be made faster by changing the settings but the accuracy is poor ) .
we use several real calcium imaging videos , which we say are @xmath49 if they contain @xmath43 slides of size @xmath50 .
if the images are larger than @xmath6 , we downsample once , otherwise , we do not downsample .
we have found that dowsampling @xmath51 and @xmath52 times causes severe errors so we avoid those settings .
in addition , we have compared moco to turboreg on synthetic images with severe translational motion artifacts and have found that moco is slightly more accurate .
all times are in seconds .
the template used for every video is the first image in the video for both moco and turboreg .
moco uses a maximum translation width of @xmath53 in both the @xmath54 and @xmath55 directions .
.2 in [ cols="^,^,^,^,^",options="header " , ]
the first author would like to thank julia sable for her help , and inbal ayzenshtat , jesse jackson , jae - eun miller , luis reid , weijian yang , and weiqun fang for their datasets .
the first author would also like to thank the columbia university data science institute for their ongoing support .
this research is supported by the nei ( dp1ey024503 , r01ey011787 ) , nihm ( r01mh101218 , r41mh100895 ) and darpa contract w91nf-14 - 1 - 0269 .
this material is based upon work supported by , or in part by , the u. s. army research laboratory and the u. s. army research office under contract number w911nf-12 - 1 - 0594 ( muri ) .
david s. greenberg , jason n.d .
kerr , `` automated correction of fast motion artifacts for two - photon imaging of awake animals , '' _ journal of neuroscience methods _ , volume 176 , issue 1 , 15 january 2009 , pages 115 .
patrick kaifosh , jeffrey d. zaremba , nathan b. danielson , and attila losonczy , `` sima : python software for analysis of dynamic fluorescence imaging data , '' front neuroinform .
2014 ; 8 : 80 . published online 2014 sep 23 | motion correction is the first in a pipeline of algorithms to analyze calcium imaging videos and extract biologically relevant information , for example the network structure of the neurons therein .
fast motion correction would be especially critical for closed - loop activity triggered stimulation experiments , where accurate detection and targeting of specific cells in necessary .
our algorithm uses a fourier - transform approach , and its efficiency derives from a combination of judicious downsampling and the accelerated computation of many @xmath0 norms using dynamic programming and two - dimensional , fft - accelerated convolutions .
its accuracy is comparable to that of established community - used algorithms , and it is more stable to large translational motions .
it is programmed in java and is compatible with imagej . |
it is thought that the critical phenomena of the anderson transition may be classified according to three universality classes ( orthogonal , unitary and symplectic ) depending on the symmetries of the hamiltonian , i.e. time reversal and spin rotation symmetries .
a different value of the critical exponent @xmath1 describing the divergence of the localization length is expected to characterize each universality class . in the absence of any diagonal disorder the su(2 )
model hamiltonian may have chiral symmetry .
this depends on the boundary conditions and the number of sites in each direction . near the band center chiral symmetry
affects the localization of electrons , e.g. , an even - odd system size dependence of the quasi-1d localization length is observed @xcite . here
we investigate whether or not chiral symmetry also affects the critical exponent @xmath1 .
a system belongs to the symplectic universality class if its hamiltonian commutes with a time reversal operator @xmath2 that satisfies @xmath3 , e.g. , systems with significant spin - orbit coupling . systems in this universality class exhibit an anderson transition even in 2d .
recently we analysed the su(2 ) model with an on - site random potential ( diagonal disorder ) , which breaks chiral symmetry , and estimated the critical exponent @xmath4 @xcite . here
we study the su(2 ) model without an on - site random potential . in spite of the fact that there is no random potential
, we find that there is an anderson transition at a critical energy @xmath5 .
further this energy @xmath5 is far from the the band center and chiral symmetry does not affect the observed critical phenomena .
the hamiltonian of the su(2 ) model describes non - interacting electrons on a simple square lattice with nearest neighbour su(2 ) random hopping @xmath6 where @xmath7(@xmath8 ) denotes the creation ( annihilation ) operator of an electron at the site @xmath9 with spin @xmath10 .
we distribute hopping matrices randomly and independently with uniform probability on the group su(2 ) .
this corresponds to @xmath11 and @xmath12 uniformly distributed in the range @xmath13 , and @xmath14 distributed in the range @xmath15 $ ] according to the probability density , @xmath16 .
we calculate the localization length @xmath17 on a quasi-1d strip whose width is @xmath18 with the transfer matrix method @xcite and analyse the dependence of the re - normalized localization length @xmath19 on the energy and the width of the strip @xcite .
we also calculate the density of states with the green s function iteration method @xcite .
in the course of our calculations we noticed a technical problem with the method described in @xcite . when fitting the numerical data it is important to have not only an estimate of the localization length @xmath17 but also an estimate @xmath20 of how accurately @xmath17 has been estimated .
otherwise a reliable fitting of the data is not possible .
we have found that the error @xmath20 is over estimated by the method of @xcite . to overcome this problem we divide the quasi-1d strip into blocks whose lengths are @xmath21 and calculate the quasi-1d localization length @xmath22(@xmath9 denotes the block ) in each block and accumulate @xmath22 . from the distribution of @xmath23
we estimate @xmath17 and @xmath20 . vs fermi energy @xmath24 for system sizes
the solid lines are the fit . ]
we calculated @xmath26 as a function of fermi energy for systems with widths @xmath27 and @xmath28 .
the accuracy of the localization length data is @xmath29 , with the exceptions of @xmath30 where the accuracy is @xmath31 . to achieve this accuracy systems of length of the order of @xmath32 to @xmath33
are required .
we impose the periodic boundary conditions in the transverse direction . for this choice of boundary condition systems with
even @xmath18 have chiral symmetry , while those with odd @xmath18 do not .
we fit even and odd data separately first and then in combination .
the results are displayed in table [ table : result ] .
the estimates of the critical parameters for all three fits are in close coincidence and we conclude that chiral symmetry does not change the universality class . in fig .
[ fig : su2w0cp ] we show the result for the combined even and odd data .
the lines cross at single point , indicating both that corrections to scaling due to irrelevant variables are negligible and also that an even - odd effect does not exist .
the estimates @xmath1 and @xmath34 are consistent with those in @xcite for the symplectic universality class .
the density of states for a system of the size @xmath35 averaged over @xmath21 ensembles is shown in fig.[fig : su2w0dos ] . in the critical region , the density of states changes slowly .
this behaviour is different from that of the random magnetic flux u(1 ) model in 3d where the density of states changes rapidly in the critical region @xcite .
this might be the reason why we obtain clearer single parameter scaling here than in a study of the critical phenomena of the u(1 ) model @xcite . .
the number of data @xmath36 and goodness of fit @xmath37 .
the best fit estimates of the critical energy @xmath5 , @xmath38 and the critical exponent @xmath1 with @xmath39 confidence intervals .
there are 7 fitting parameters . [ cols="<,<,<,<,<,<",options="header " , ] | we report an analysis of the anderson transition in an su(2 ) model with chiral symmetry .
clear single parameter scaling behaviour is observed .
we estimate the critical exponent for the divergence of the localization length to be @xmath0 indicating that the transition belongs to the symplectic universality class . ,
, anderson transition ; spin - orbit coupling ; chiral symmetry |
extra gauge bosons , both charged ( @xmath2 ) and/or neutral ( @xmath1 ) , arise in many models of physics beyond the standard model ( sm ) @xcite .
examples include extended gauge theories such as grand unified theories @xcite and left - right symmetric models @xcite along with the corresponding supersymmetric models , and other models such as those with finite size extra dimensions @xcite . to elucidate what physics lies beyond the standard model it is necessary to search for manifestations of that new physics with respect to the predicted particle content , both fermions and extra gauge bosons .
such searches are a feature of ongoing collider experiments and the focus of future experiments .
the discovery of new particles would provide definitive evidence for physics beyond the standard model and , in particular , the discovery of new gauge bosons would indicate that the standard model gauge group was in need of extension .
there is a considerable literature on @xmath1 searches . in this paper
we concentrate on @xmath2 searches , for which much less work has been done .
limits have been placed on the existence of new gauge bosons through indirect searches based on the deviations from the sm they would produce in precision electroweak measurements . for instance , indirect limits from @xmath4-decay constrain the lrm @xmath2 to @xmath5 gev @xcite .
a more severe constraint arises from @xmath6 mass - splitting which gives @xmath7 tev @xcite . in obtaining the above limits
, it was assumed that the coupling constants of the two @xmath8 gauge groups are equal .
new gauge boson searches at hadron colliders consider their direct production via the drell - yan process and their subsequent decay to lepton pairs . for @xmath2 bosons ,
decays to hadronic jets are sometimes also considered .
the present bounds on neutral gauge bosons , @xmath1 s , from the cdf and d0 collaborations at the tevatron @xmath9 collider at fermilab are @xmath10 gev with the exact value depending on the specific model @xcite . for @xmath2 s the limits are @xmath11 gev ; again the limits depend on the details of the model @xcite .
the search reach is expected to increase by @xmath12 gev with 1 fb@xmath13 of luminosity @xcite .
the large hadron collider is expected to be able to discover @xmath1 s up to masses of 4 - 5 tev @xcite and @xmath2 s up to masses of @xmath14 tev @xcite .
the @xmath2 limits assume sm strength couplings and decay into a light stable neutrino which is registered in the detector as missing @xmath15 .
they can be seriously degraded by loosening the assumptions in the model .
in addition , one can place limits on new gauge bosons by looking for deviations from sm expectations for observables measured at @xmath16 and @xmath17 colliders
. searches for new gauge bosons at @xmath17 colliders are kinematically limited by the available center - of - mass energy so that one searches for indirect effects of extra gauge bosons in cross sections and asymmetries for @xmath18 .
there is a considerable body of work on @xmath1 searches at @xmath19 colliders and , although the discovery limits are very model dependent , they lie in the general range of 2 - 5 tev for @xmath20 gev with 50 fb@xmath13 luminosity @xcite .
in contrast to the @xmath1 case , there are virtually no studies of indirect searches for @xmath2 bosons at @xmath17 colliders .
recently , hewett suggested that the reaction @xmath21 would be sensitive to @xmath2 s with masses greater than @xmath22 @xcite . in the standard model
, this process proceeds through @xmath23-channel @xmath24 and @xmath25-channel @xmath26 exchange with the photon being radiated from every possible charged particle . in
extended gauge models the process is modified by both @xmath23-channel @xmath1 and @xmath25-channel @xmath2 exchange . in this paper
, we examine this process for various extended electroweak models . the first model we consider is the left - right symmetric model @xcite based on the gauge group @xmath27 which has right - handed charged currents .
the second model we consider is the un - unified model @xcite which is based on the gauge group @xmath28 where the quarks and leptons each transform under their own @xmath8 .
the final type of model , which has received considerable interest lately , contains the kaluza - klein excitations of the sm gauge bosons which are a possible consequence of theories with large extra dimensions @xcite .
the models under consideration are described in more detail in section ii . additionally , we study discovery limits for various combinations of @xmath2 and @xmath1 bosons with sm couplings .
although these are not realistic models , they have been adopted as benchmarks to compare the discovery reach of different processes .
we will find that , while the process @xmath29 can indeed extend the discovery reach for @xmath2 s significantly beyond @xmath22 , with the exact limit depending on the specific model , it is not in general competitive with limits obtainable at the lhc .
however , if extra gauge bosons are discovered which are not overly massive , the process considered here could be used to measure their couplings .
this would be crucial for determining the origins of the @xmath1 or @xmath2 .
as such , it would play an important complementary role to the lhc studies . in the next section we review the relevant details of the various models that we use in our calculations . in section iii
, we describe the details of our calculations .
the resulting @xmath2 discovery limits and projected sensitivities for @xmath2 couplings and @xmath30 couplings are given in section iv .
we conclude with some final comments .
in this section , we describe the models considered in our investigation .
the so - called sequential standard model ( ssm ) includes additional weak gauge bosons of higher mass , with sm couplings .
this is a rather arbitrary scenario which we include only as a benchmark .
since our emphasis here is on extra @xmath26 s , we consider a ssm with a @xmath2 only , which we refer to as ssm(@xmath2 ) , and a ssm with both @xmath2 and @xmath1 , denoted by ssm(@xmath31 ) . in the latter
, we will take @xmath32 for simplicity .
the general left - right symmetric model ( lrm ) @xcite is based on the extended electroweak gauge group @xmath33 .
left - handed fermion fields transform as doublets under @xmath34 and as singlets under @xmath35 .
the reverse is true for right - handed fermions .
a right - handed neutrino is included in the fermion content .
the model is parametrized by the ratio of the coupling constants of the two @xmath8 gauge groups , which we denote as @xmath36 .
this parameter is allowed to vary here in the approximate range @xmath37 @xcite .
the lower bound on @xmath38 arises from the condition @xmath39 ( or , equivalently , @xmath40 ) , which expresses the positivity of a ratio of squared couplings . in principle , @xmath38 is restricted to be less than 1 based on symmetry breaking scenarios and coupling constant evolution arguments .
however , it is conceivable that this bound may be violated in some grand unified theory so we take a phenomenological approach and loosen this upper bound @xcite .
additionally , a parameter , @xmath41 , describes the higgs content of the model .
if only higgs doublets are used to break the gauge symmetry to @xmath42 , @xmath41 is 1 . for higgs triplets ,
@xmath41 is 2 .
a combination of doublets and triplets leads to an intermediate value of @xmath41 between 1 and 2 @xcite .
we will use @xmath43 , corresponding to higgs doublets . in the lrm
, there is a relationship between the @xmath1 and @xmath2 masses , as follows : @xmath44 the couplings of the extra gauge bosons relevant to our calculation can be read from the following parts of the lagrangian .
@xmath45 + h.c.\end{aligned}\ ] ] where @xmath46 denotes a right - handed electron field .
note that we neglect two angles , usually denoted as @xmath47 and @xmath48 , which parametrize the @xmath49 and @xmath50 mixings , respectively .
limits on these angles are rather severe so this is justified @xcite .
neglect of these angles implies sm couplings for the @xmath24 and @xmath26 .
additionally , we assume light dirac - type neutrinos . the un - unified model ( uum ) @xcite employs the alternative electroweak gauge symmetry @xmath51 with left - handed quarks and leptons transforming as doublets under their respective @xmath8 groups .
all the right - handed fields transform as singlets under both @xmath8 groups .
the uum may be parametrized by an angle @xmath52 , which represents the mixing of the charged gauge bosons of the two @xmath8 groups , and by a ratio @xmath53 , where @xmath54 and @xmath55 are the vacuum expectation values of the scalar multiplets which break the symmetry to @xmath42 .
the existing constraint on @xmath52 is @xmath56 , based on the validity of perturbation theory . for @xmath57 ,
the @xmath1 mass is approximately equal to that of the @xmath2 and the parameter @xmath58 may be replaced by @xmath59 .
the lepton couplings of interest to us here arise from the following part of the lagrangian .
@xmath60 + h.c.\end{aligned}\ ] ] as expected , the fermion couplings to the additional gauge bosons are all left - handed in the uum .
additional fermions must also be included in order to cancel anomalies .
this is rather difficult to do without generating flavour changing neutral currents and some considerations of this problem lead to rather high lower bounds on the @xmath1 mass of about 1.4 tev @xcite .
however , lower @xmath1 and @xmath2 masses may be allowed in other scenarios ; hence we take a phenomenological approach in this investigation .
finally , we consider the consequences of models which have been of considerable interest lately , those containing large extra dimensions @xcite .
in particular , we consider an extension of the sm to 5-dimensions ( 5dsm ) @xcite .
the presence of an extra dimension of size @xmath61 tev@xmath13 may imply an infinite tower of kaluza - klein ( kk ) excitations of the sm gauge bosons . the mass of the excitations is associated with the compactification scale of the extra dimension as @xmath62 ( @xmath63 ) , where @xmath64 .
the properties of and relationships among electroweak observables are modified by the presence of these kk towers .
we treat this possibility in a manner similar to the other models described above ; that is , we include in our process the exchange of a @xmath2 and @xmath1 corresponding to the first kk excitations
. the model can be parametrized by an angle @xmath65 which is correlated with the properties of its higgs sector , which includes two doublets ; for @xmath66 , the sm higgs may propagate in all 5-dimensions ( the bulk ) while for @xmath67 , it is confined to the 4-dimensional boundary . in terms of this parameter ,
the physical masses of the lightest electroweak gauge bosons ( corresponding to the experimentally measured masses ) are given , to first order in @xmath68 , as @xmath69 \\ m^{(ph)2}_z & = & m^2_z \left[1 - s^4_{\beta } \frac{\pi^2}{3 } \frac{m^2_z}{m^2_c } \right]\end{aligned}\ ] ] where @xmath70 , as usual.the gauge couplings of the physical @xmath26 and @xmath24 are also modified by a term of order @xmath71 .
global analyses of electroweak parameters put a lower limit on @xmath72 of about 2.5 tev so this is a very small effect .
we will therefore neglect it and thus eliminate @xmath73 as a parameter . on the other hand ,
the fermion coupling of the first kk excitations , @xmath2 and @xmath1 , each of mass @xmath74 , is enhanced by a factor of @xmath75
. hence our consideration of the 5dsm amounts to including a @xmath2 and a @xmath1 , of equal mass , each coupling as in the sm apart from an extra factor of @xmath75 .
the process under consideration is @xmath76 the relevant feynman diagrams are given in fig . 1
. the kinematic observables of interest are the photon s energy , @xmath77 , and its angle relative to the incident electron , @xmath78 , both defined in the @xmath17 center - of - mass frame .
the invariant mass of the @xmath79 pair , @xmath80 , and @xmath77 are related via @xmath81 where @xmath82 .
let @xmath83 denote the sum of the amplitudes shown in fig .
1 , over a given number of @xmath1 s and @xmath2 s . the doubly differential cross section is related to @xmath84 via @xmath85 where @xmath86 and @xmath87 are the polar and azimuthal angles , respectively , of @xmath88 in a frame where @xmath88 and @xmath89 are back - to - back .
the explicit momentum parametrizations are given in the appendix .
two approaches to determining @xmath84 are possible .
one can determine @xmath83 analytically , using spinor techniques @xcite for instance , then square it numerically or one can find @xmath84 analytically .
we have followed both approaches , which provides an independent check . obtaining @xmath84
analytically has been done both via the trace method , using the symbolic manipulation program form @xcite , and by squaring the helicity amplitudes and summing over the final state helicities .
the latter approach leads to a rather compact result which we present below .
in order to present @xmath84 , we define the following kinematic variables .
we follow the notation of @xcite , where the sm contribution for this process was calculated : @xmath90 the decay width of the extra neutral gauge boson , @xmath91 , into fermion- antifermion pairs is calculated in each of the models we consider .
we include the one - loop qed , three - loop qcd and @xmath92 corrections , although their effect on the cross section is negligible . in the following ,
we denote generalized couplings as may be inferred from the vertices @xmath93 thus , in the sm , @xmath94 , @xmath95 , @xmath96 , @xmath97 , @xmath98 , and @xmath99 .
it is only necessary to present the unpolarized squared amplitude as the individual polarized contributions may be inferred from the coupling structure .
the spin - averaged unpolarized @xmath84 is given by : @xmath100 \\ \nonumber & & + \frac{4}{s ' } \sum_{\stackrel{\scriptstyle i=1,nw}{j = i , nw } } w_{ij } [ ( a^2_{w_i}a^2_{w_j}+b^2_{w_i}b^2_{w_j } ) ( u^2+u'^2 ) + 2 a_{w_i}a_{w_j}b_{w_i}b_{w_j}(s^2+s'^2 ) ] \\ \nonumber & & + \frac{4}{c^2_w } \sum_{\stackrel{\scriptstyle i=1,nw}{j=1,nz } } [ \left ( wz \right)_{ij } ( u^2 a^2_{w_i } a^e_{z_j } a^\nu_{z_j } + u'^2 b^2_{w_i } b^e_{z_j } b^\nu_{z_j } ) \\ & & \,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\ , \,\,\,\,\,\,\ , + \left ( wz \right)'_{ij } ( u'^2 a^2_{w_i } a^e_{z_j } a^\nu_{z_j } + u^2 b^2_{w_i } b^e_{z_j } b^\nu_{z_j } ) ] \left .
\mbox{\raisebox{4ex } { } } \right\ } , \end{aligned}\ ] ] where @xmath101 using the notation @xmath102 , where @xmath103 is the completely antisymmetric levi - civita tensor and @xmath104 . in eq .
( [ m2 ] ) we have assumed lepton universality with regards to the @xmath30 couplings . although it may not be immediately apparent , the contribution to the cross section from the state where the @xmath105 and @xmath106 are both left - handed is equal to the contribution from the state where they are both right - handed and the sum is given by the term in eq .
( [ m2 ] ) proportional to @xmath107 .
a relation which was quite useful in simplifying @xmath84 is @xmath108 in the sm limit , eq .
( [ m2 ] ) agrees with the expression given in @xcite after correcting for the known missing factors of @xmath109 in @xcite required on dimensional grounds .
the calculation of @xmath110 may be performed analytically or numerically .
we have followed both approaches and verified numerical agreement .
further checks were performed using the program comphep @xcite .
before discussing the discovery limits obtained in the various models , we present the total cross sections and the differential cross sections @xmath111 and @xmath112 . in doing so ,
all the essential features are illustrated .
we take the sm inputs @xmath113 gev , @xmath114 gev , @xmath115 , @xmath116 , @xmath117 gev @xcite . since we work only to leading order in @xmath84
, there is some arbitrariness in what to use for the above input , in particular @xmath118 .
kinematically , the maximum allowed value for @xmath77 is @xmath119 .
in addition , to take into account detector acceptance , @xmath77 and @xmath78 have been restricted to the ranges @xmath120 the cuts also serve to remove the singularities which arise when the emitted photon is soft or collinear with the beam .
further , we restrict the photon s transverse momentum to @xmath121 where @xmath122 is the minimum angle down to which the veto detectors may observe electrons or positrons .
we take @xmath123 mrad .
this cut has the effect of removing the largest background to our process , namely radiative bhabha - scattering where the scattered @xmath106 and @xmath105 go undetected down the beam pipe .
this study was performed in leading order , but qed corrections to @xmath124 must be taken into account in a precision analysis of real data .
they have been known to @xmath125 for some time @xcite .
see @xcite for a short review _ of _ and further references _ to _ higher order qed corrections , and @xcite for a description of a related mc generator .
since our aim is to determine the statistical power of the process in discovering @xmath2 s , there is no need to include in this study the radiative corrections which will only marginally influence the number of events .
complete consistency at nlo , however , would require determination of the bremsstrahlung corrections to the _ generalized _ expression ( [ m2 ] ) and corresponding loop graphs . as well , we do not explicitly take into account any higher order backgrounds .
a background , which can not be suppressed , comes from the reaction @xmath126 .
the authors of @xcite have provided the following cross sections of relevance here : @xmath127 and @xmath128 .
these results are for the same conditions as in table 1 of @xcite but for @xmath129 .
the cuts used in obtaining the above numbers differ from ours . nonetheless , these cross sections give an idea of the magnitude of the background .
assuming lepton universality , the total cross section is @xmath130 for the process @xmath126 . imposing our @xmath131 cut will suppress it even further .
this background must be included in an `` @xmath132 + nothing '' analysis of real data .
we expect that the cross sections of @xmath133 and of @xmath134 are so small that they need not be taken into account in the analysis .
the errors generated from the subtraction of the above backgrounds form part of the systematic error .
as the backgrounds themselves are much smaller than the signal , though comparable to the new physics effect , we expect that the error in the sm prediction of the backgrounds would be much smaller than the systematic errors arising from detector and beam uncertainties .
we shall return to the issue of systematics in connection with their influence on the discovery limits presented in the next section .
we have calculated three distinct total cross sections : unpolarized : @xmath135 , for left - handed @xmath105 : @xmath136 , and for right - handed @xmath105 : @xmath137 .
2 shows all three plotted versus @xmath22 , with @xmath136 and @xmath137 calculated using 100% beam polarization .
results are shown for the sm , lrm ( @xmath138 ) , uum ( @xmath139 ) , ssm(@xmath2 ) , ssm(@xmath140 ) and kk model , with @xmath141 gev in each case .
these mass and coupling parameter choices are rather arbitrary , made to illustrate general behaviour .
it is worth noting at this point that in the uum and ssm(@xmath140 ) , the correction to the sm cross section changes sign as @xmath22 is varied .
this arises , for certain @xmath22 and @xmath59 , due to a negative interference term between the sm and @xmath142 diagrams in these models .
it is clear from the presence of the peaks in fig . 2 that we are also probing @xmath1 s , in those models which include them .
( there is also a very sharp peak at lower @xmath22 , off the plot , due to the sm @xmath24 . )
the @xmath1 peaks generally occur for @xmath22 slightly above the @xmath1 mass since the photon carries away some of the energy . at very high energies ,
the sm @xmath24 contribution is negligible .
further , by using a right - handed @xmath105 beam , we can reduce the sm @xmath26 contribution ( depending on the degree of polarization ) .
then we directly probe the @xmath2 ( and @xmath1 ) in the lrm , while in the ssm(@xmath140 ) and kk model , we probe only the @xmath1 .
the latter two models as well as the two remaining models all require some component of left - handed polarization to probe the @xmath2 .
the above features are borne out in fig . 2 .
in order to see which regions of @xmath77 are most sensitive to the new physics , we plot for left- and right - handed electron beams respectively , in figs .
3(a ) and 4(a ) @xmath111 versus @xmath77 and in figs .
3(b ) and 4(b ) the deviation from the sm result divided by the square root of the predicted cross section versus @xmath77 .
we show results for @xmath143 gev with 100% @xmath105 beam polarization in these figures .
first , we note the shape of @xmath111 in figs 3(a ) and 4(a ) . for left - handed electrons ,
the bulk of the cross section comes from the low @xmath77 region ; the reduction at very low @xmath77 is due to the @xmath131 cut and the sharp peak at @xmath144 gev is due to the radiative return to the @xmath24 pole . for 100% right polarized electrons ,
the cross section is rather flat in the low to moderate @xmath77 region , then increases as a result of the @xmath24 peak at high @xmath77 . on the other hand ,
since the right - handed cross section is two orders of magnitude smaller than the left - handed cross section away from the @xmath24 peak , any realistic degree of polarization ( i.e. 90% ) will lead to a large contribution from @xmath136 to the low @xmath77 region .
in general , there can also be a peak due to a @xmath1 for @xmath145 which occurs at @xmath146 in analogy with the sm @xmath24 .
most important , however , is the relative statistical significance , shown in figs .
3(b ) and 4(b ) . in both the left- and right - handed cases ,
the low @xmath77 region is the most sensitive to the new physics .
there are two reasons for this .
first , for left - handed electrons , the cross section is largest at low @xmath77 , as mentioned above .
second , the lower @xmath77 , the higher the mass probed in the @xmath1 propagator via eq .
( [ egamma ] ) .
the relative effect is even larger when combining the @xmath147 s from the different bins , since it is the squares of the plotted quantities which will enter .
overall , the kk model leads to the most statistically significant deviations , except for the 100% left polarized case where the ssm(@xmath2 ) exhibits the largest deviation .
we can also see clearly how the sign of the deviation from the sm depends on the beam polarization . for the kk model and ssm(@xmath140 ) , we observe a negative deviation with right - handed polarization , implying a negative @xmath1 contribution , versus a positive overall contribution coming from the left - handed channel .
clearly , interference effects will make probing @xmath2 s nontrivial .
we shall return to this point in the next section . in figs .
5 and 6 we plot the analogous quantities relevant to @xmath112 , versus @xmath148 . we note that both @xmath112 and the relative statistical significance are peaked in the forward and backward directions and both are very nearly symmetric in @xmath148 .
the latter implies that the forward - backward asymmetry will be small and , therefore , the deviation from the sm forward - backward asymmetry will also be small , at least in absolute magnitude .
we therefore do not expect the forward - backward asymmetry to serve as a useful probe of the new physics , which is confirmed by explicit calculation .
an important observation is that our @xmath131 cut , while eliminating a large background , has also eliminated much of our signal ( both from the small angle and soft events ) which was appreciably stronger prior to the cut . a more detailed study , including a detector simulation , would be required to determine whether the background could be accurately subtracted with a looser @xmath131 cut .
the best discovery limits were in general obtained using the observable @xmath111 , combined with beam polarization , while @xmath112 was less sensitive .
comparable or equal limits were obtained using the total cross section , with an additional cut on the energy to eliminate the @xmath24 pole radiative return events : @xmath149 as can be seen from figs .
3(b ) and 4(b ) , the @xmath24 pole region is quite insensitive to new physics . in the cases that @xmath111 provided a better limit than the total cross section ,
the improvement was of order 50 gev .
however , the @xmath147 obtained using the total cross section is a somewhat less stable function of @xmath59 as the sign of the deviation from the sm cross section may change with @xmath59 leading to isolated regions of insensitivity at low @xmath59 . also , when systematic errors are included , the limits obtained using @xmath111 are affected much less than those obtained using the total cross section .
substantially weaker limits were obtained using the left - right asymmetry , @xmath150 even when including systematic errors only one half those used in the @xmath111 calculation ( since one expects some cancellation of errors between the numerator and denominator in @xmath151 ) . as expected from the discussion of the previous section , the forward - backward asymmetry , @xmath152 , was quite insensitive to the new physics . in light of the above , we restrict the remaining discussion to limits obtained using @xmath111 as an observable . in obtaining the @xmath147 for @xmath111 , we used 10 equal sized energy bins in the range @xmath153 , where @xmath154 follows from the @xmath131 cut eq .
( [ ptcut ] ) : @xmath155 which supersedes the acceptance cut of eq .
( [ accept ] ) .
we have @xmath156 where @xmath157 is the error on the measurement and analogous formulae hold for other observables .
one sided 95% confidence level discovery limits are obtained by requiring @xmath158 for discovery .
systematic errors , when included , were added in quadrature with the statistical errors . in determining the limits for the case of polarized electron beams , we show results for the polarization state which in general has the largest sensitivity ( deviation from the sm ) for a given model ; a right - handed @xmath105 beam for the lrm and a left - handed beam for all other models .
we used one half the unpolarized luminosity for the polarized case , assuming equal running time in each polarization state .
the discovery limits for all five models are listed in table i , for @xmath159 , 1.0 and 1.5 tev , using the same input parameters as for the cross sections presented in the previous section .
we show limits for both an unpolarized @xmath105 beam and for a 90% polarized one . for each center - of - mass energy ,
two luminosity scenarios are considered and we present limits obtained with and without systematic errors .
our prescription is to include a 2% systematic error per bin .
this number is quite arbitrary but seems reasonable , if not conservative , considering the clean final state .
in addition to detector systematics , which we expect will dominate , there are uncertainties associated with the beam luminosity and energy , which will be spread over a range .
the systematic errors associated with the background subtraction should be much smaller than 2% as should be the errors in the calculation of the qed corrections .
the 2% number should not be taken too seriously therefore , except to highlight the fact that a precision measurement is required to take full advantage of the large event rate .
certain features are common to all models . with
no systematic error included , we observe quite an improvement in the limits with increased luminosity .
the only exception is the uum at @xmath22 of 1.5 tev , where the improvement is minimal .
the reason is that the @xmath147 decreases very rapidly as @xmath59 is increased in the vicinity of the limit , hence increasing the luminosity by a factor of 2.5 does little .
the unusual @xmath22 dependence can be attributed to the interference effect noted in the previous section , which results in , for example , for the uum with @xmath139 and an integrated luminosity of 500 fb@xmath13 , a lower discovery limit at @xmath160 tev than at 0.5 and 1 tev .
we will return to this peculiar behaviour later in the section .
when 2% systematic errors are included , the high luminosity scenario yields little improvement in the limits in any of the models , since the systematic error now dominates the statistical .
perhaps surprising at first is the observation that 90% beam polarization does not improve the limits very much .
this follows from taking into account the reduced luminosity and the fact that the left - handed component tends to dominate the unpolarized cross section by a considerable amount . on the other hand
, we observed that if the polarization is pushed beyond 90% , then the right - polarized limits can increase significantly in those models in which the beyond - sm bosons have a non - zero right - handed coupling : the lrm , kk model and ssm(@xmath140 ) . in the latter two models
, it is however , the @xmath1 which is being probed . the higher degree of polarization is required to eliminate the contamination from the much larger left - handed component .
thus , the primary advantage of beam polarization is to distinguish between models and measure the new couplings , as will be investigated in the next section .
7 presents the @xmath2 mass discovery limits obtainable in the lrm with an unpolarized beam , plotted versus @xmath38 for @xmath161 and @xmath162 @xmath163 , 1.0 , 1.5 and 2 tev using a luminosity of 50 fb@xmath13 for @xmath159 tev and 200 fb@xmath13 for the higher energies .
only statistical errors are included .
depending on @xmath22 and @xmath38 , the limits range from 0.8 to 2.8 tev .
we expect greater deviations from the sm , and hence larger limits , as @xmath38 is increased since this increases the @xmath2 coupling strength , as can be seen from eq .
( [ lrmlagr ] ) . the predicted dependence on @xmath38
is generally observed , except at low @xmath38 where we notice a moderate _
increase _ in the limits , even though the @xmath2 couplings have weakened .
we attribute this effect to the @xmath1 , whose couplings are enhanced ( but its mass increased ) in the low @xmath38 region .
this was indicated by an appreciable improvement in the limits for low @xmath38 and @xmath161 versus those obtained using @xmath164 and consequently a heavier @xmath1 , via eq .
( [ massrel ] ) .
8 demonstrates the improvement in bounds in the moderate to large @xmath38 region obtained when a 90% or 100% polarized right - handed @xmath105 beam is used .
the beam polarization picks out the lrm @xmath2 and suppresses the sm @xmath26 .
8(a ) shows that for @xmath165 , 90% beam polarization improves the limits . further increasing the polarization leads to substantial improvements , even at lower @xmath38 , as demonstrated in fig .
8(b ) . the dependence of the limits in the uum on @xmath166 is shown in fig .
9 , for @xmath162 @xmath163 , 1.0 , 1.5 and 2.0 tev , under the same running conditions as fig .
7 . only the unpolarized case is considered as beam polarization was not beneficial .
again , only statistical errors are included . at each @xmath22 , we note that the contour defining the exclusion region as a function of @xmath166 is a complicated curve .
the consequence is that for @xmath167 tev , we obtain better limits over a range of @xmath166 than we do for @xmath160 and even @xmath168 tev .
essentially , this is due to the complicated interference with the sm diagrams . in general , as @xmath166 increases , the uum couplings also increase , as can be seen from eq .
( [ uumlagr ] ) , so that higher mass scales are probed .
so , referring to fig .
2 , the peak in the cross section ( due to the @xmath1 ) at the scale being probed shifts to the right .
but the sign of the deviation from the sm changes with @xmath22 for fixed @xmath59 ( or vice - versa ) such that the uum cross section dips below the sm over some region to the left of the peak , then goes back above it for small @xmath22 ( or large @xmath169 for fixed @xmath22 ) .
hence , there is a small step in the limits near @xmath170 , corresponding to passing the rightmost crossing with the sm and another structure in the contour at some higher @xmath59 such that the leftmost crossing is situated near @xmath22 .
one sees this explicitly by plotting @xmath147 versus @xmath59 for fixed @xmath22 and @xmath166 and observing a dip in the @xmath147 at relatively low @xmath59 .
had we used @xmath135 as an observable , the dip would be much more pronounced since @xmath171 passes through zero , but @xmath172 may differ in sign between bins , leading to a nonzero @xmath147 at the crossing points .
once @xmath166 is large enough that we are probing the region to the left of the leftmost crossing , the limits shoot up in an impressive fashion as the dip in @xmath147 never goes back down to 2.69 .
the shape of the plot is luminosity dependent since , as pointed out earlier in this section , the degree to which increased luminosity improves the limits depends on the rate at which the @xmath147 decreases with increasing @xmath59 in the vicinity of the limit .
that , in turn , varies with @xmath22 for fixed @xmath166 and with @xmath166 for fixed @xmath22 . in this section
, we consider constraints which can be put on the couplings of extra gauge bosons by the process @xmath173 .
these constraints are significant only in the case where the mass of the corresponding extra gauge boson is considerably lower than its search limit in this process . in most models ,
the process @xmath174 and/or searches at the lhc are more sensitive to a @xmath1 or @xmath2 ( lhc ) than the process @xmath173 .
we assume here that a signal for an extra gauge boson has been detected by another experiment .
given such a signal , we derive constraints ( at 95% c.l . ) on the couplings of extra gauge bosons .
we present the constraints in terms of couplings normalized as follows relative to eqs .
( [ zcoup ] ) and ( [ wcoup ] ) .
@xmath175 the constraints correspond to @xmath176 where @xmath177 is the prediction for the observable @xmath178 in the sm , @xmath179 is the prediction of the extension of the sm and @xmath180 is the expected experimental error .
the index @xmath181 corresponds to different observables such as @xmath135 and @xmath151 .
our assumptions concerning beam polarization are as follows . for single beam ( @xmath105 ) polarization , we assume , as in the previous section , equal running in left and right polarization states . for double beam polarization , we assume equal running in the @xmath182 and @xmath183 states , but no running in the @xmath184 and @xmath185 states .
thus , @xmath186 where the first subscript of @xmath135 refers to the @xmath105 helicity .
note that for 100% polarized @xmath105 and @xmath106 , @xmath187 in all the models we consider .
this remains approximately valid as the couplings deviate from their model - defined values . in figs .
[ nnng1 ] and [ nnng2 ] , we present @xmath30 coupling constraints assuming there is no signal for a @xmath2 .
this is the case when the sm is extended by @xmath188 factors only .
it can also happen in models where the @xmath2 has purely right - handed couplings and the right - handed neutrino is heavy .
then , the process @xmath173 would be one of the best for constraining the couplings of the @xmath1 to sm neutrinos _ below _ the @xmath1 resonance .
if there is also a signal for a @xmath2 , a similar analysis could be performed including the @xmath2 parameters , as measured in other experiments .
the resulting bounds would be larger than those shown in the two figures .
however , the main points of the discussion would remain unchanged .
[ nnng1 ] illustrates the resulting constraints on a 1.5 tev @xmath1 at a 500 gev collider for different observables and experimental parameters , including luminosity and beam polarization .
we see that we can get some interesting constraints even though the @xmath1 is considerably heavier than the centre - of - mass energy .
the region which can not be resolved by the observables is between the two corresponding lines and contains the couplings of the sm .
hence the star in this figure corresponds to the ssm(@xmath1 ) .
for the cases where only one bounding line is shown , the second line is outside the figure . @xmath189 and
@xmath190 are mainly constrained by the interference of the @xmath1 exchange with the sm .
the strongest constraint is on the @xmath1 coupling to left - handed neutrinos .
this makes the constraints especially simple .
first we consider an integrated luminosity of 500 fb@xmath13 .
the total unpolarized cross section gives the strongest constraint .
the constraints from energy and angular distributions ( with 10 equal size bins ) were also considered but they give no improvement . the constraint from @xmath151 is shown for two polarization cases : 90% electron beam polarization and the case of a collider with a @xmath191 polarized electron beam and a @xmath192 polarized positron beam . even for the latter case , the constraint from @xmath151 is worse than that from the total cross section .
we mention here for completeness that two polarized beams give not only a high effective polarization but also a small effective polarization error @xcite . the constraint obtained with an integrated luminosity of @xmath193 fb@xmath13 is also shown in fig .
[ nnng1 ] , to contrast with the high luminosity case .
we see that for @xmath194 fb@xmath13 a systematic error of 1% relaxes the constraints considerably and dilutes the advantage of high luminosity .
thus , both small systematic errors and a high luminosity collider are highly desired for the proposed measurement . fig
. [ nnng2 ] shows the possible constraints on @xmath189 and @xmath190 from @xmath135 and @xmath151 , including systematic errors , for two representative @xmath1 masses , 0.75 tev and 1 tev .
the constraints become much stronger as the @xmath1 mass is decreased .
so far , we assumed that the @xmath195 couplings , @xmath196 and @xmath197 , are precisely known .
however , they must be measured ( with errors ) by another experiment .
4(b ) of @xcite illustrates such a measurement for a collider with a luminosity of 20 fb@xmath13 .
to estimate their influence on the @xmath189 , @xmath190 constraint , we make use of the errors on @xmath196 and @xmath197 given in @xcite . our input for the errors of the @xmath195 couplings for @xmath198 tev and @xmath199 tev
are obtained from those for @xmath200 tev by the scaling relation ( 2.63 ) in @xcite .
we see that the uncertain knowledge of the @xmath195 couplings leads to only slightly weaker constraints on @xmath189 and @xmath190
. however , fig .
[ nnng2 ] shows that this effect is only important for a relatively heavy @xmath1 and for @xmath189 ( even at lower @xmath1 masses ) for which the constraints are already weak .
finally , we mention that there is no sign ambiguity in the measurement of @xmath189 and @xmath190 if the signs of the @xmath195 couplings are known .
it was noted @xcite that the @xmath195 couplings have a two - fold sign ambiguity if measured in the process @xmath201 alone .
if this ambiguity exists , it induces a related sign ambiguity for @xmath189 and @xmath190 . if the sign ambiguity in the @xmath195 couplings is resolved @xcite ( i.e. by measurements obtained from the process @xmath202 below the @xmath1 resonance or by measurements at the @xmath1 resonance ) it also disappears in our constraints on @xmath189 and @xmath190 . in figs .
[ nnng3 ] to [ nnng6 ] , we shall assume that there is no signal from a @xmath1 but that a signal from a @xmath2 has been observed .
this could happen in models where the @xmath2 is considerably lighter than the @xmath1 .
we recognize that this particular scenario is unlikely in the context of the models we consider .
for instance , in the uum , the @xmath2 and @xmath1 masses are approximately equal and there would most likely be a signal observed for the @xmath1 in addition to the @xmath2 .
the situation is similar in the lrm , where the relationship between the @xmath2 and @xmath1 masses is given in eq .
( [ massrel ] ) .
thus , it should be understood that our results for the case of a @xmath2 only represent an estimate of the reach of this process in constraining @xmath2 couplings , rather than precision limits in the context of a full understanding of the physics realized in nature .
we use this simple scenario in order to indicate sensitivity to various parameters , such as the observables used and the luminosity .
alternatively , a known @xmath1 could be included in the following analysis .
again , the experimental errors on the measured @xmath1 parameters would enlarge the errors of the @xmath2 measurements but not change the main conclusions .
we will see that the process @xmath173 can give model independent constraints on the quantities @xmath203 and @xmath204 for @xmath2 masses considerably larger than the center - of - mass energy .
we only probe @xmath205 directly , but we are assuming lepton universality throughout .
[ nnng3 ] is similar to fig .
[ nnng1 ] , but it shows the constraints on the @xmath2 couplings . in this figure , for illustration , we assume there exists a @xmath2 with sm couplings but with a mass of 1.5 tev and that the right - handed neutrino is light enough to be produced .
we find that the left- and , to some extent , the right - handed @xmath2 coupling can be constrained .
the figure illustrates the use of different combinations of @xmath135 and @xmath151 , and of different beam polarizations .
the unpolarized cross section mainly constrains the left - handed @xmath2 coupling because left - handed electrons give its dominant contribution .
the constraints from energy and angular distributions give almost no improvement for the model considered here .
the constraint from @xmath151 is complementary to that from @xmath135 .
it is shown for the two cases of 90% electron beam polarization and for 90% electron beam polarization with 60% positron polarization .
we see that @xmath135 and @xmath151 together give the best constraints on the couplings .
the constraints on the @xmath2 couplings have a two - fold sign ambiguity ; nothing is changed by a simultaneous change of the sign of @xmath203 and @xmath204 .
the reason for this ambiguity lies in the squared amplitude , eq.([m2 ] ) , where these couplings always enter as squares or as a product of left and right @xmath2 couplings . in the case where we have only a weak @xmath2 signal , the two regions allowed by this ambiguity overlap into one large region . in fig .
[ nnng4 ] , we show constraints on the @xmath2 couplings from @xmath135 and @xmath151 combined . in this figure , we illustrate the use of different luminosities and the inclusion of a systematic error .
we have the same two well separated regions for the case of high luminosity and no systematic error as in fig . [ nnng3 ] .
these two regions become larger for low luminosity and no systematic error .
we are left with one large region after the inclusion of a systematic error of 2% for @xmath135 and 1% for @xmath151 . as in the case of extra neutral gauge bosons , small systematic errors _ and _ high luminosity are necessary for a coupling measurement . in fig .
[ nnng5 ] , we show how the constraints on the @xmath2 couplings vary for different @xmath2 masses .
the constraint for @xmath206 tev is identical to that from fig .
[ nnng4 ] .
we see that the constraint on the @xmath2 couplings improves dramatically for lower @xmath2 masses .
[ nnng6 ] illustrates the possibility of discrimination between different models .
we see that a @xmath2 with sm couplings ( @xmath207 ) can be separated from the sm . a @xmath2 with pure right - handed couplings ( @xmath208 ) with a strength of the left - handed coupling of the sm @xmath26
can not be distinguished from the sm case .
looking at the squared amplitude , eq .
( [ m2 ] ) , we see that the constraints shown in figs .
[ nnng3 ] to [ nnng6 ] are , to a good approximation , valid for the combinations @xmath209 and @xmath210 , and not for the couplings and the mass separately .
we have fixed the @xmath2 mass here for illustrational purposes .
if a @xmath2 is found with a mass different from our assumptions , the constraint on its couplings can be found by the appropriate scaling of our results .
so far , we considered model independent bounds on the couplings of a single extra gauge boson while neglecting the existence of other extra gauge bosons . however ,
typically , extra neutral and charged gauge bosons simultaneously influence the observables .
we consider this situation for the lrm and the uum . in fig .
[ nnng7 ] , we consider the left - right symmetric model . for @xmath211 tev ,
( [ massrel ] ) gives @xmath212 tev for @xmath213 and @xmath214 .
we show the constraints on the couplings of the @xmath2 for @xmath161 obtained by two different fitting strategies .
first , we ignore the @xmath1 completely , and second , we take the @xmath1 into account assuming exact knowledge of its couplings .
we see that the two curves are quite close .
the reason is that our process is not very sensitive to such a @xmath1 .
these two curves are very similar to those for the @xmath215 and the sm in fig .
[ nnng6 ] because we are not very sensitive to a right - handed @xmath2 .
the case of @xmath164 predicts a heavier @xmath1 , which produces constraints differing even less from each other than those for @xmath161 , so we do not show them . to demonstrate how the constraints change for a larger signal , we repeated the same procedure with @xmath216 gev .
this number ( and the mass of the associated @xmath1 ) are at the edge of the present exclusion limit @xcite .
although the constraints improve a bit , they are still not very impressive .
[ nnng8 ] is similar to fig .
[ nnng7 ] but here we consider the un - unified model . we examine the cases
@xmath217 tev and @xmath218 tev .
we show the constraints on the couplings of the @xmath2 obtained using the same two fitting strategies described for fig .
[ nnng7 ] . even for masses of @xmath199 tev ,
the two curves are better separated than in lrm . for masses of 0.55 tev ,
the wrong fitting strategy gives a region which is outside the true @xmath2 coupling .
this shows that such a light @xmath1 can not be ignored in the fitting procedure . the process @xmath174 and searches in hadron collisions are more sensitive to @xmath1 discovery than @xmath173
. a @xmath1 signal will always be detected in the cases where the @xmath1 contribution is relevant for a @xmath2 constraint from @xmath173 .
this information from other experiments will be required for a reliable @xmath2 constraint from @xmath173 .
in this paper , we studied the sensitivity of the process @xmath219 to extra gauge bosons .
we used this process to find discovery limits and to see how well one could measure the couplings of extra gauge bosons that are expected in extensions of the standard model . for the discovery limits we focused on @xmath2 s since one can put better limits on @xmath1 s from other processes , such as @xmath220 , while , on the other hand , no similar limits exist on @xmath2 s .
the highest reach was obtained by binning the @xmath221 distribution although comparable results were obtained using the total cross section after the @xmath24 radiative return was eliminated .
the discovery reach is typically in the 1 - 6 tev range depending on the specific model , the center of mass energy , and the assumed integrated luminosity .
these results are substantially degraded if one includes systematic errors . for the @xmath215 boson , for which lhc discovery limits are available , the discovery limits are , for @xmath222 , @xmath59= 1.2 , 1.6 , and 1.9 tev for @xmath162 500 , 1000 , and 1500 gev respectively assuming @xmath194 fb@xmath13 relative to a reach of @xmath223 tev at the lhc .
although the discovery reach for @xmath2 s of this process is not competitive with the reach of the lhc , precision measurements can give information on extra gauge boson couplings which complements the lhc .
in particular , if the lhc were to discover a @xmath1 or @xmath2 the process @xmath29 could constrain @xmath1 and @xmath2 couplings . for a @xmath1
, this would be the best measurement of the @xmath224 couplings . for @xmath2 couplings
, reliable measurements would require information from , for example , @xmath174 and searches in hadron collisions which would always detect a @xmath1 signal in the cases where its contribution is relevant for a @xmath2 constraint by @xmath173 . finally , we emphasize that to make measurements of the extra gauge boson couplings , high luminosity will be needed and it will be very important to reduce the systematic uncertainties as much as possible . this research was supported in part by the natural sciences and engineering research council of canada .
s.g . , p.k . , and b.k .
thank dean karlen and a.l .
thanks graham wilson for useful discussions .
here we give explicit parametrizations of the momenta defined in the frame where @xmath225 are back - to - back and @xmath226 defines the @xmath227 axis , suitable for use with the phase - space ( [ phasesp ] ) : @xmath228 where @xmath229 it is arbitrary whether @xmath230 is taken as positive or negative as long as one is consistent . for a recent review
see m. cvetic and s. godfrey ; in _ electroweak symmetry breaking and beyond the standard model _ , eds . t. barklow _ et al . _
( world scientific , 1995 ) , p. 383 , hep - ph/9504216 .
see also ref .
@xcite .
robinett , phys .
d * 26 * , 2388 ( 1982 ) ; r.w .
robinett and j.l .
rosner , _ ibid .
_ d * 25 * 3036 ( 1982 ) ; _ ibid .
_ d * 26 * , 2396 ( 1982 ) ; _ ibid . _ d * 27 * , 679(e ) 1983 ; p. langacker , r.w .
robinett , and j.l .
rosner , phys .
d * 30 * , 1470 ( 1984 ) .
i. antoniadis , phys .
b * 246 * , 377 ( 1990 ) ; j. lykken , phys .
d * 54 * , 3693 ( 1996 ) ; e. witten , nucl .
* b471 * , 135 ( 1996 ) ; n. arkani - hamed , s. dimopoulos , and g. dvali , phys .
b * 429 * , 263 ( 1998 ) ; i.antoniadis , n. arkani - hamed , s. dimopoulos , and g. dvali , phys . lett .
b * 436 * , 257 ( 1998 ) ; n. arkani - hamed , s. dimopoulos , and g.dvali , phys . rev .
d * 59 * , 086004 ( 1999 ) .
t. rizzo ; in _ new directions for high - energy physics : proceedings of the 1996 dpf / dpb summer study on high - energy physics _ , snowmass , colorado , 1996 , edited by d.g .
cassel , l. trindle gennari , and r.h .
siemann ( stanford linear accelerator center , 1997 ) , p. 900
, hep - ph/9609248 ; p. 864
, hep - ph/9612440 .
j. hewett ; in _ new directions for high - energy physics : proceedings of the 1996 dpf / dpb summer study on high - energy physics _ , snowmass , colorado , 1996 , edited by d.g .
cassel , l. trindle gennari , and r.h .
siemann ( stanford linear accelerator center , 1997 ) , p. 887
, hep - ph/9704292 .
m. masip and a. pomeral , phys .
d * 60 * , 096005 ( 1999 ) ; t.g .
rizzo and j.d .
wells , report nos .
slac - pub-8119 , cern - th/99 - 139 , hep - ph/9906234 ; g.f .
giudice , r. rattazzi , and j.d .
wells , nucl .
* b544 * , 3 ( 1999 ) ; t. han , j.lykken , and r .- j .
zhang , phys . rev .
d * 59 * , 105006 ( 1999 ) . p.a .
baikov et al . , physical results by means of comphep ; in _ proc . of x workshop on high energy physics and quantum field theory ( qfthep-95 ) _ , moscow , 1996 , edited by b. levtchenko and v. savrin , p. 101
, hep - ph/9701412 ; e.e .
boos , m.n .
dubinin , v.a .
ilyin , a.e .
pukhov , and v.i .
savrin , hep - ph/9503280 .
g. montagna , m. moretti , o. nicrosini , and f. piccinini , nucl .
* b541 * , 31 ( 1999 ) . g. montagna , o. nicrosini , and f. piccinini , comput .
commun . * 98 * , 206 ( 1996 ) .
a. denner , s. dittmaier , m. roth , and d. wackeroth , report no .
bi - tp-99 - 10 , hep - ph/9904472 .
ginzburg _ et al .
_ , nucl .
instrum .
methods , * 205 * , 47 ( 1983 ) ; _ ibid _ * 219 * , 5 ( 1984 ) ; c. akerlof , report no .
um - he-81 - 59 ( 1981 ; unpublished ) .
a. czarnecki and w.j .
marciano , int .
j. mod .
phys . * a13 * , 2235 ( 1998 ) .
a. leike and s. riemann , z. phys .
c * 75 * , 341 ( 1997 ) .
a. leike , phys .
* 317 * , 143 ( 1999 ) .
s. godfrey , p. kalyniak , b. kamal , and a. leike ; in _ proceedings of mrst99 : high energy physics at the millennium _ , ottawa , canada , 1999 , edited by p. kalyniak , s. godfrey , and b. kamal ( aip , 1999 ) , p. 1 , hep - ph/9907571 ; s. godfrey , p. kalyniak , b. kamal , and a. leike ; presented at _ the international workshop on physics and detectors at the linear collider _ , sitges , spain , 1999 , hep - ph/9909422 . to appear in proceedings . .@xmath2
95% c.l .
discovery limits obtained in the ssm ( @xmath2 ) , ssm ( @xmath140 ) , lrm ( @xmath231 ) , uum ( @xmath139 ) , and the kk model using @xmath111 as the observable .
results are presented for @xmath20 , 1000 , and 1500 gev and for various luminosity and polarization scenarios , with and without a 2% systematic error included .
for the lrm , the polarized scenario corresponds to a right - handed @xmath105 beam , while for all other models the beam is left - handed .
[ cols= " < , < , < , < , < , < , < , < , < , < " , ] | we examine the sensitivity of the process @xmath0 to extra gauge bosons , @xmath1 and @xmath2 , which arise in various extensions of the standard model .
the process is found to be sensitive to @xmath2 masses up to several tev , depending on the model , the center of mass energy , and the assumed integrated luminosity .
if extra gauge bosons were discovered first in other experiments , the process could also be used to measure @xmath3 and @xmath2 couplings .
this measurement would provide information that could be used to unravel the underlying theory , complementary to measurements at the large hadron collider .
0.0 cm |
a _ generator _ of a metric space is a set @xmath7 of points in the space with the property that every point of the space is uniquely determined by its distances from the elements of @xmath7 .
given a simple and connected graph @xmath8 , we consider the metric @xmath9 , where @xmath10 is the length of a shortest path between @xmath11 and @xmath12 .
@xmath13 is clearly a metric space .
a vertex @xmath14 is said to distinguish two vertices @xmath11 and @xmath12 if @xmath15 .
a set @xmath16 is said to be a _ metric generator _ for @xmath0 if any pair of vertices of @xmath0 is distinguished by some element of @xmath7 .
a minimum generator is called a _ metric basis _ , and its cardinality the _ metric dimension _ of @xmath0 , denoted by @xmath17 . motivated by the problem of uniquely determining the location of an intruder in a network , the concept of metric dimension of a graph was introduced by slater in @xcite , where the metric generators were called _ locating sets_. the concept of metric dimension of a graph was introduced independently by harary and melter in @xcite , where metric generators were called _ resolving sets_. applications of this invariant to the navigation of robots in networks are discussed in @xcite and applications to chemistry in @xcite .
this invariant was studied further in a number of other papers including for example , @xcite .
several variations of metric generators including resolving dominating sets @xcite , independent resolving sets @xcite , local metric sets @xcite , and strong resolving sets @xcite , etc . have been introduced and studied . in this article
we are interested in the study of strong resolving sets @xcite .
a vertex @xmath18 _ strongly resolves _ two vertices @xmath19 if @xmath20 or @xmath21 , i.e. , there exists some shortest @xmath22 path containing @xmath3 or some shortest @xmath23 path containing @xmath2 .
a set @xmath7 of vertices in a connected graph @xmath0 is a _ strong metric generator _ for @xmath0 if every two vertices of @xmath0 are strongly resolved by some vertex of @xmath7 .
the smallest cardinality of a strong resolving set of @xmath0 is called _ strong metric dimension _ and is denoted by @xmath24 .
so , for example , @xmath25 if and only if @xmath0 is the complete graph of order @xmath26
. for the cycle @xmath27 of order @xmath26 the strong metric dimension is @xmath28 and if @xmath29 is a tree with @xmath30 leaves , its strong metric dimension equals @xmath31 ( see @xcite ) .
a _ strong metric basis _ of @xmath0 is a strong metric generator for @xmath0 of cardinality @xmath24 .
given a simple graph @xmath8 , we denote two adjacent vertices @xmath32 by @xmath33 .
the _ neighborhood of a vertex _ @xmath3 of @xmath0 is @xmath34 and the degree of @xmath3 is @xmath35 .
the _ open neighborhood of a set _ @xmath7 of vertices of @xmath0 is @xmath36 and the _ closed neighborhood of _
@xmath7 is @xmath37=n_g(s)\cup s$ ] .
the _ subgraph induced by a set _
@xmath38 will be denoted by @xmath39 .
a vertex @xmath2 of @xmath0 is _ maximally distant _ from @xmath3 if for every vertex @xmath1 in the open neighborhood of @xmath2 , @xmath40 .
if @xmath2 is maximally distant from @xmath3 and @xmath3 is maximally distant from @xmath2 , then we say that @xmath2 and @xmath3 are _ mutually maximally distant_. the _ boundary _ of @xmath8 is defined as @xmath41 there exists @xmath14 such that @xmath32 are mutually maximally distant@xmath42 . for some basic graph classes , such as complete graphs @xmath43 , complete bipartite graphs
@xmath44 , cycles @xmath27 and hypercube graphs @xmath45 , the boundary is simply the whole vertex set .
it is not difficult to see that this property holds for all @xmath46-antipodal is defined as @xmath47 .
we recall that @xmath8 is @xmath46-antipodal if for each vertex @xmath48 there exists exactly one vertex @xmath49 such that @xmath50 . ]
graphs and also for all distance - regular graphs .
notice that the boundary of a tree consists exactly of the set of its leaves .
a vertex of a graph is a _ simplicial vertex _
if the subgraph induced by its neighbors is a complete graph . given a graph @xmath0 , we denote by @xmath51 the set of simplicial vertices of @xmath0 .
notice that @xmath52 .
we use the notion of strong resolving graph introduced in @xcite .
the _ _ strong resolving graph _ _ of a graph @xmath0 has vertex set @xmath53 and two vertices @xmath32 are adjacent in @xmath54 if and only if @xmath2 and @xmath3 are mutually maximally distant in @xmath0 .
so , the strong resolving graph defined here is a subgraph of the strong resolving graph defined in @xcite and it can be obtained from the latter graph by deleting its isolated vertices .
] of @xmath0 is a graph @xmath55 with vertex set @xmath56 where two vertices @xmath32 are adjacent in @xmath55 if and only if @xmath2 and @xmath3 are mutually maximally distant in @xmath0 .
there are some families of graphs for which its strong resolving graph can be obtained relatively easy .
for instance , we emphasize the following cases . * if @xmath57 , then @xmath58 . in particular , @xmath59 and for any tree @xmath29 with @xmath30 leaves , @xmath60 .
* for any @xmath46-antipodal graph @xmath0 of order @xmath26 , @xmath61
. in particular , @xmath62 . *
@xmath63 .
a set @xmath7 of vertices of @xmath0 is a _ vertex cover _ of @xmath0 if every edge of @xmath0 is incident with at least one vertex of @xmath7 .
the _ vertex cover number _ of @xmath0 , denoted by @xmath64 , is the smallest cardinality of a vertex cover of @xmath0 .
we refer to an @xmath64-set in a graph @xmath0 as a vertex cover of cardinality @xmath64 .
oellermann and peters - fransen @xcite showed that the problem of finding the strong metric dimension of a connected graph @xmath0 can be transformed to the problem of finding the vertex cover number of @xmath55 .
the following result will be an important tool of this article . _
let @xmath0 be connected graph .
a set @xmath65 is a strong metric generator for @xmath0 if and only if @xmath6 is a vertex cover for @xmath55 .
it was shown in @xcite that the problem of computing @xmath24 is np - hard .
this suggests finding the strong metric dimension for special classes of graphs or obtaining good bounds on this invariant .
an efficient procedure for finding the strong metric dimension of distance hereditary graphs was described in @xcite . in this paper
we study the problem of finding exact values or sharp bounds for the strong metric dimension of rooted product of graphs and express these in terms of invariants of the factor graphs .
notice that the metric dimension of rooted product graphs has been recently studied in @xcite . and
@xmath66 , where @xmath3 has degree two.,title="fig:",scaledwidth=30.0% ] and @xmath66 , where @xmath3 has degree two.,title="fig:",scaledwidth=30.0% ] a _ rooted graph _ is a graph in which one vertex is labeled in a special way so as to distinguish it from other vertices .
the special vertex is called the _ root _ of the graph .
let @xmath0 be a labeled graph on @xmath26 vertices .
let @xmath67 be a sequence of @xmath26 rooted graphs @xmath68
, @xmath69, ... ,@xmath70 .
the _ rooted product graph _
@xmath71 is the graph obtained by identifying the root of @xmath72 with the @xmath73 vertex of @xmath0 @xcite . in this paper
we consider the particular case of rooted product graph where @xmath67 consists of @xmath26 isomorphic rooted graphs @xcite .
more formally , assuming that @xmath74 and that the root vertex of @xmath75 is @xmath3 , we define the rooted product graph @xmath76 , where @xmath77 and @xmath78 note that for any @xmath79 the subgraph @xmath80 of @xmath81 is isomorphic to @xmath75 . given @xmath79 , @xmath82 and @xmath83 we will denote by @xmath84 the set of element of @xmath85 whose first component is @xmath11 , i.e. , @xmath86 .
if @xmath75 is a vertex transitive graph , then @xmath81 does not depend on the choice of @xmath3 , up to isomorphism .
in such a case we will denote the rooted product by @xmath87 .
figure [ p-4-c-3 ] shows the case of the rooted product graphs @xmath88 and @xmath66 , where @xmath3 has degree two .
we also recall that the _ corona product _
@xmath89 is defined as the graph obtained from @xmath0 and @xmath75 by taking one copy of @xmath0 and @xmath26 copies of @xmath75 and joining by an edge each vertex from the @xmath73-copy of @xmath75 with the @xmath73-vertex of @xmath0 .
if @xmath0 and @xmath75 are connected graphs of order @xmath90 , @xmath75 is a connected graph of order @xmath91 , then we note that the corona product graph @xmath89 is a particular case of a rooted product graph , _
i.e. _ , @xmath92 , where @xmath3 denotes the vertex of @xmath93 .
metric dimension and strong metric dimension of corona product graphs were studied in @xcite and @xcite , respectively .
we emphasize that given @xmath94 and @xmath95 it follows , @xmath96 and if @xmath97 , then @xmath98 .
this article is composed by two main sections . in section [ formulae ]
we obtain closed formulae for the strong metric dimension of some classes of rooted product graphs while section [ bounds ] is devoted to obtain tight bounds for the strong metric dimension of rooted product graphs .
we start by stating the following easily verified lemmas .
[ maximally_distant with root ] let @xmath0 and @xmath75 be two connected graphs .
let @xmath94 , @xmath97 , @xmath95 and let @xmath99 be the set of vertices of @xmath100 which are maximally distant from @xmath3
. then @xmath101 and @xmath102 are mutually maximally distant vertices in @xmath81 if and only if @xmath103 . ( sufficiency ) suppose that @xmath101 and @xmath102 are not mutually maximally distant vertices in @xmath104 .
so , there exists a vertex @xmath105 such that @xmath106 , or there exists @xmath107 such that @xmath108 .
we consider , without loss of generality , that @xmath105 and @xmath106 .
so we have , @xmath109 thus , @xmath110 . since @xmath111 and @xmath112 , we have a contradiction .
( necessity ) let us suppose that @xmath113 .
so , there exists @xmath114 such that @xmath115 .
thus , @xmath116 .
hence , there exists a vertex @xmath117 such that @xmath118 , which is a contradiction since @xmath101 and @xmath102 are mutually maximally distant .
[ rooted mutually_maximally_distant ] let @xmath0 and @xmath75 be two connected nontrivial graphs .
let @xmath119 be vertices of @xmath75 such that @xmath120 .
for every vertex @xmath121 of @xmath0 we have that @xmath101 and @xmath122 are mutually maximally distant vertices in @xmath81 if and only if the vertices @xmath11 and @xmath12 are mutually maximally distant in @xmath75 .
the result follows directly from the fact that for every vertex @xmath123 of @xmath0 and every vertex @xmath124 of @xmath75 we have that @xmath125 if and only if @xmath126 and also that @xmath127 for every @xmath128 of @xmath75 .
[ maximally distant with v ] let @xmath75 be a connected graph , let @xmath82 and let @xmath99 be the set of vertices of @xmath100 which are maximally distant from @xmath3
. then @xmath129 .
let @xmath130 .
if @xmath3 is not maximally distant from @xmath2 , then there exists a vertex @xmath131 such that @xmath132 .
so @xmath2 is maximally distant from @xmath133 .
by repeating this argument , since @xmath75 is finite , we will find a vertex @xmath134 such that @xmath134 and @xmath2 are mutually maximally distant .
therefore , @xmath135 .
[ boundary - of - rooted ] let @xmath0 be a connected graph of order @xmath90 and let @xmath75 be a connected graph . 1 . if @xmath136 , then @xmath137 2 .
if @xmath138 , then @xmath139 let @xmath140 and @xmath141 be two mutually maximally distant vertices in @xmath142 . since @xmath143 , it follows @xmath144 .
we differentiate two cases .
case 1 : @xmath145 . by lemma
[ rooted mutually_maximally_distant ] we conclude that @xmath140 and @xmath141 are mutually maximally distant in @xmath142 if and only if @xmath12 and @xmath146 are mutually maximally distant in @xmath75 .
case 2 : @xmath147 . by lemma [ maximally_distant with root ] the vertices @xmath140 and @xmath141 are mutually maximally distant in @xmath142 if and only if @xmath148 .
note that , by lemma [ maximally distant with v ] , @xmath149 . according to the above cases we conclude that if @xmath150 , then @xmath151 .
moreover , if @xmath151 , then for every @xmath79 we have @xmath150 .
therefore , if @xmath136 , then @xmath152 and if @xmath153 , then @xmath139 [ simplicial - of - rooted ] let @xmath0 be a connected graph of order @xmath90 and let @xmath75 be a connected graph . 1 .
if @xmath154 , then @xmath155 2 . if @xmath156 , then @xmath157 note that @xmath158 is not simplicial in @xmath142 . since the following assertions are equivalent , the result immediately follows .
* the vertex @xmath159 is simplicial in @xmath142 . * for @xmath79 and @xmath160
the vertex @xmath140 is simplicial in @xmath161 . *
the vertex @xmath162 is simplicial in @xmath75 .
[ thcompletesr ] let @xmath0 be a connected graph of order @xmath90 and let @xmath75 be a connected graph such that @xmath163 . 1 . if @xmath136 , then @xmath164 2 .
if @xmath138 , then @xmath165 since @xmath163 , as a direct consequence of proposition [ boundary - of - rooted ] and proposition [ simplicial - of - rooted ] we obtain that if @xmath138 , then @xmath166 and if @xmath136 , then @xmath167 hence , if @xmath138 , then @xmath168 and , if @xmath136 , then @xmath169 . therefore , the result follows by theorem [ lem_oellerman ] .
we emphasize the following particular cases of theorem [ thcompletesr ] .
[ corollari - simplicial ] let @xmath0 be a connected graph of order @xmath90 . 1 . for any complete graph of order @xmath170 ,
@xmath171 2 .
for any tree @xmath29 with @xmath30 leaves , @xmath172 3 .
let @xmath173 be a connected graph of order @xmath170 and let @xmath174 , where @xmath175 , @xmath176 . then + @xmath177 [ thk_2sr ] let @xmath0 be a connected graph of order @xmath90 and let @xmath75 be a connected graph such that @xmath178 .
let @xmath82 and let @xmath99 be the set of vertices of @xmath75 which are maximally distant from @xmath3 .
let @xmath179 be the set of isolated vertices of the subgraph of @xmath180 induced by @xmath99 . 1 . if @xmath138 , then @xmath181 .
if @xmath136 , then @xmath182 let @xmath183 be the vertex set of @xmath0 and let @xmath85 be a vertex cover for @xmath184 .
first we note that by premiss for every @xmath185 there exists exactly one vertex @xmath186 such that @xmath121 and @xmath187 are adjacent in @xmath180 .
we consider the set @xmath188 defined in the following way : @xmath189 if and only if there exists @xmath190 such that @xmath121 and @xmath187 are mutually maximally distant in @xmath75 .
note that @xmath191 and , if @xmath136 and @xmath192 are mutually maximally distant , then @xmath193 and @xmath194 . also , since there are no edges in @xmath180 connecting vertices belonging to @xmath195 to vertices belonging to @xmath196 , by lemmas [ maximally_distant with root ] and [ rooted mutually_maximally_distant ]
we conclude that there are no edges in @xmath197 connecting vertices belonging to @xmath198 to vertices belonging to @xmath199 . with this idea in mind
, we proceed to prove the results . in order to prove ( i ) we consider that @xmath138 .
note that in this case by proposition [ boundary - of - rooted ] ( ii ) , @xmath200 . by lemma [ rooted mutually_maximally_distant ]
we have that for every mutually maximally distant vertices @xmath201 and every @xmath202 the vertices @xmath203 and @xmath204 are mutually maximally distant in @xmath142 and , as a consequence , @xmath205 if and only if @xmath206 .
thus , the subgraph of @xmath207 induced by @xmath208 is composed by @xmath209 components isomorphic to @xmath210 .
on the other hand , by lemma [ maximally_distant with root ] we have that @xmath211 are mutually maximally distant in @xmath142 , for every @xmath212 and @xmath213 .
thus , if @xmath205 for some @xmath214 , then @xmath215 for every @xmath216 .
moreover , as above , lemma [ rooted mutually_maximally_distant ] allows us to conclude that given two mutually maximally distant vertices @xmath217 it follows that @xmath205 if and only if @xmath206 .
thus , @xmath85 contains exactly @xmath218 vertices belonging to @xmath199 .
therefore , latexmath:[\[\begin{aligned } the proof of ( i ) is complete . from now on
we suppose @xmath136 .
note that in this case by proposition [ boundary - of - rooted ] ( i ) we have @xmath220 . to prove ( ii ) we proceed by analogy to the proof of ( i ) . in this case
we obtain that the subgraph of @xmath207 induced by @xmath198 is composed by @xmath221 components isomorphic to @xmath210 and @xmath85 contains exactly @xmath222 vertices of @xmath81 belonging to @xmath223 .
thus , latexmath:[\[\begin{aligned } the proof of ( ii ) is complete .
we conjecture that if @xmath153 , then @xmath225 . in order to show a particular case of theorem [ thk_2sr ] where @xmath226 we consider the graph @xmath75 shown in the left hand side of figure [ grafos - i(v ) ] where @xmath227 , @xmath228 and @xmath229 . in the case of the graph @xmath75 shown in the right hand side of figure [ grafos - i(v ) ] we have @xmath227 , @xmath230 , @xmath231 and @xmath232 . in both cases
@xmath233 is a strong metric basis of @xmath81 for any graph @xmath0 with vertex set @xmath234 . and @xmath229 .
in right hand side graph @xmath231 and @xmath232.,title="fig:",scaledwidth=40.0% ] and @xmath229 . in right hand side graph
@xmath231 and @xmath232.,title="fig:",scaledwidth=40.0% ] [ corollaryantipodal ] let @xmath0 be a connected graph of order @xmath90 and let @xmath75 be a connected 2-antipodal graph of order @xmath170 . then @xmath235 .
[ h is cycle ] let @xmath236 be a cycle of order @xmath237 . for any connected graph @xmath0 of order @xmath175 , @xmath238 let @xmath239 and @xmath240 be the vertex sets of @xmath0 and @xmath236 , respectively .
we assume @xmath241 in @xmath236 and from now on all the operations with the subscripts of @xmath134 are done modulo @xmath242 .
since @xmath236 is a vertex transitive graph , we can take without loss of generality @xmath243 as the root of @xmath236 .
if @xmath242 be an even number , then @xmath236 is 2-antipodal .
so the result follows by corollary [ corollaryantipodal ] .
now let @xmath242 be an odd number .
note that exactly two vertices @xmath244 and @xmath245 are maximally distant from @xmath3 in @xmath236 .
so , from lemma [ maximally_distant with root ] we have that every vertex @xmath246 is mutually maximally distant from @xmath247 in @xmath248 , with @xmath249 and @xmath250 .
moreover , from lemma [ rooted mutually_maximally_distant ] we have that for every @xmath251 , @xmath252 is mutually maximally distant from @xmath253 and @xmath254 in @xmath248 with @xmath255 . also , the vertex @xmath256 is mutually maximally distant from @xmath257 and the vertex @xmath258 is mutually maximally distant from @xmath259 .
thus , we obtain that the graph @xmath260 is isomorphic to a graph with set of vertices @xmath261 where @xmath262 is isomorphic to a complete @xmath263-partite graph @xmath264 and for every @xmath265 , @xmath266 is isomorphic to a path graph @xmath267 . notice that the leaves of @xmath267 belong to @xmath268 , so for every @xmath265 , @xmath269
thus , we have the following : @xmath270 the proof is complete .
we recall that the _ clique number _ of a graph @xmath75 , denoted by @xmath271 , is the number of vertices in a maximum clique in @xmath75 .
two distinct vertices @xmath11 , @xmath12 are called _ true twins _ if @xmath272 = n_h[y]$ ] .
we say that @xmath273 is a _ twin - free clique _ in @xmath75 if the subgraph induced by @xmath38 is a clique and for every @xmath274 it follows @xmath275\ne n_h[v]$ ] , i.e. , the subgraph induced by @xmath38 is a clique and it contains no true twins .
we say that the _ twin - free clique number _ of @xmath75 , denoted by @xmath276 , is the maximum cardinality among all twin - free cliques in @xmath75 .
so , @xmath277 . _
@xcite_[lem_strdimcoronagen ] let @xmath0 be a connected graph of order @xmath263 .
let @xmath75 be a graph of order @xmath242 and maximum degree @xmath278 .
if @xmath279 or @xmath175 , then @xmath280 given a vertex @xmath3 of a graph @xmath75 , we denote by @xmath281 the graph obtained by removing @xmath3 from @xmath75 .
now , if @xmath3 is a vertex of @xmath75 of degree @xmath282 , then the rooted product graph @xmath81 is isomorphic to the corona product graph @xmath283 .
so , as a direct consequence of theorem [ lem_strdimcoronagen ] we obtain the following result .
let @xmath0 be a connected graph of order @xmath175 .
let @xmath75 be a connected graph of order @xmath91 and let @xmath3 be a vertex of @xmath75 of degree @xmath284
. then @xmath285 .
the next result gives the exact value for the strong metric dimension of @xmath89 when @xmath75 is a triangle free graph . _
@xcite_[lem_strdimcoronatrianglefree ] let @xmath0 be a connected graph of order @xmath263 and let @xmath75 be a triangle free graph of order @xmath237 and maximum degree @xmath278 . if @xmath175 or @xmath286 , then @xmath287 as a direct consequence of theorem [ lem_strdimcoronatrianglefree ] we have the following .
let @xmath0 be a connected graph of order @xmath175 .
let @xmath75 be a connected graph of order @xmath91 and let @xmath3 be a vertex of @xmath75 of degree @xmath284 . if @xmath281 is a triangle free graph .
then @xmath288 as the next theorem shows , the strong metric dimension of @xmath89 depends on the diameter of @xmath75 . _
@xcite_[lem_strdimcoronadiameter ] let @xmath0 be a connected graph of order @xmath263 .
let @xmath75 be a graph of order @xmath242 and maximum degree @xmath278 . 1 .
if @xmath75 has diameter two and either @xmath279 or @xmath175 , then @xmath289 2 . if @xmath75 is not connected or its diameter is greater than two , then @xmath290 therefore , as a consequence of theorem [ lem_strdimcoronadiameter ] we obtain the following result for @xmath142
let @xmath0 be a connected graph of order @xmath175 .
let @xmath75 be a graph of order @xmath91 and let @xmath3 be a vertex of @xmath75 of degree @xmath284 . 1 . if @xmath281 has diameter two , then @xmath291 2 .
if @xmath281 has diameter greater than two , then @xmath292 the strong metric dimension of @xmath89 depends on the existence or not of true twins in @xmath75 . in this sense , the following result was presented in @xcite . _
@xcite_[lem_strdimcoronatwins ] let @xmath0 be a connected graph of order @xmath263 and let @xmath75 be a graph of order @xmath242 .
let @xmath293 be the number of vertices of @xmath75 having degree @xmath284 . 1 .
if @xmath75 has no true twins and @xmath175 , then @xmath294 2 .
if the only true twins of @xmath75 are vertices of degree @xmath284 and @xmath175 , then @xmath295 our next result is an interesting consequence of theorem [ lem_strdimcoronatwins ] .
[ prop strdimroottwins ] let @xmath0 be a connected graph of order @xmath175 .
let @xmath75 be a connected graph of order @xmath91 and let @xmath3 be a vertex of @xmath75 of degree @xmath284 .
let @xmath296 be the number of vertices of @xmath281 having degree @xmath297 . 1 .
if @xmath281 has no true twins , then @xmath298 2 .
if the only true twins of @xmath281 are vertices of degree @xmath297 , then @xmath299
[ lemmmadivide ] let @xmath0 and @xmath75 be two connected graphs .
given @xmath79 , @xmath82 and a strong metric basis @xmath85 of @xmath81 let @xmath86 and let @xmath99 be the set of vertices of @xmath100 which are maximally distant from @xmath3 . then the following assertions hold .
first we consider a pair @xmath303 of adjacent vertices in @xmath304 , where @xmath144 . since @xmath85 is a vertex cover of @xmath184 , either @xmath305 or @xmath306 .
thus , @xmath307 is a vertex cover of @xmath304 .
note that @xmath308 and , as a consequence , @xmath309 .
hence , @xmath310 . therefore , ( i ) follows .
now we suppose @xmath301 .
if @xmath140 and @xmath158 are adjacent in @xmath304 , then @xmath311 . so the edge @xmath312 of @xmath304 is covered by @xmath305 .
thus , @xmath84 is a vertex cover of @xmath304 and , as a result , @xmath302 .
therefore , ( ii ) follows .
finally , suppose that @xmath3 does not belong to any strong metric basis of @xmath75 .
since the function @xmath313 , where @xmath314 , is a graph isomorphism and @xmath307 is a strong metric generator for @xmath315 , the set @xmath316 is a strong metric generator for @xmath75 .
thus , since @xmath3 does not belong to any strong metric basis of @xmath75 , @xmath317 .
taking into account that @xmath309 we obtain @xmath318 .
the proof is complete .
let @xmath6 be a strong metric basis of @xmath75 such that @xmath321 and let @xmath85 be a strong metric basis of @xmath142 . since @xmath3 belongs to a metric basis of @xmath75 , we have @xmath136 .
suppose there exists @xmath79 such that @xmath322 for some @xmath130 .
by lemma [ lemmmadivide ] ( i ) we obtain @xmath300 .
moreover , by lemma [ maximally_distant with root ] we have that for @xmath323 and @xmath324 the vertices @xmath325 and @xmath326 are mutually maximally distant in @xmath81 .
hence , since @xmath322 and @xmath85 is a vertex cover of @xmath207 , for every @xmath323 we have @xmath327 .
so , according to lemma [ lemmmadivide ] ( ii ) we have @xmath328 . therefore , @xmath329 on the other hand , since @xmath136 , proposition [ boundary - of - rooted ] ( ii ) leads to @xmath137 we will show that @xmath330 is a vertex cover for @xmath207 , where @xmath331 and @xmath332 .
let @xmath140 and @xmath141 be two adjacent vertices in @xmath207 . if @xmath333 or @xmath334 , then @xmath335 or @xmath336 .
now let , @xmath337 . since @xmath338 and @xmath6 is a vertex cover for @xmath75
, @xmath339 is a vertex cover for @xmath340 and , as a consequence , @xmath341 or @xmath342 .
hence , @xmath7 is a vertex cover for @xmath207 .
therefore , @xmath343 the proof of ( i ) is complete . from now on
we assume that @xmath3 does not belong to any strong metric basis of @xmath75 .
the lower bound of ( ii ) is a direct consequence of lemma [ lemmmadivide ] ( iii ) .
suppose @xmath138 . in this case , by proposition
[ boundary - of - rooted ] ( i ) we conclude @xmath139 by analogy with the proof of the upper bound of ( i ) we show that @xmath344 is a vertex cover for @xmath207 , where @xmath345 , @xmath332 and @xmath346 is a strong metric basis of @xmath75 .
hence , @xmath347 finally , for the case @xmath136 we have @xmath152 and proceeding by analogy with the proof of the upper bound of ( i ) we show that @xmath348 is a vertex cover for @xmath207 , where @xmath349 , @xmath332 and @xmath350 is a strong metric basis of @xmath75 .
thus , in this case @xmath351 the proof of ( ii ) is complete . as corollary [ corollari - simplicial ] shows , the bounds of theorem [ h is not t v in basis ] ( i ) are tight and the upper bound @xmath352 of theorem [ h is not t v in basis ] ( ii ) is tight . to show the tightness of the upper bound @xmath353 we consider the graph @xmath354 shown in figure [ graph - j ] .
notice that any strong metric basis of @xmath354 is formed by the vertices @xmath355 , @xmath356 and three vertices of the set @xmath357 .
let @xmath234 be the set of vertices of @xmath0 . from figure [ graph - j ]
we have that there exits six vertices @xmath360 and @xmath361 which are maximally distant from @xmath3 .
so , by using lemma [ maximally_distant with root ] , we have that every two vertices @xmath362 , where @xmath363 , are mutually maximally distant .
moreover , by lemma [ rooted mutually_maximally_distant ] for every two mutually maximally distant vertices @xmath364 in @xmath354 we have that @xmath365 are mutually maximally distant in @xmath366 for every vertex @xmath367 of @xmath0 .
thus , @xmath368 is isomorphic to @xmath369 .
therefore , @xmath370 . to see the tightness of the lower bound of theorem
[ h is not t v in basis ] ( ii ) we define the family @xmath371 of graphs @xmath75 containing a vertex of degree one not belonging to any strong metric basis of @xmath75 .
we begin with the cycle @xmath236 , where @xmath242 is an odd number such that @xmath372 , with set of vertices @xmath373 .
to obtain a graph @xmath374 we add the sets of vertices @xmath375 , @xmath376 and @xmath377 , where @xmath378 , and edges @xmath379 , @xmath380 , @xmath381 , for every @xmath382 , and @xmath383 , for every @xmath384 . notice that vertices of @xmath385 have degree one in @xmath386 and they are mutually maximally distant between them . also , for any vertex @xmath387 , @xmath388 , where @xmath384 .
similarly , for any vertex @xmath389 , @xmath390 , where @xmath382 .
moreover , we can observe that @xmath391 and @xmath392 are mutually maximally distant for every @xmath393 .
so , @xmath394 is formed by @xmath395 connected components , that is , @xmath396 connected components isomorphic to @xmath210 and also , a connected component isomorphic to a graph with set of vertices @xmath397 where @xmath398 is isomorphic to @xmath399 , @xmath400 is adjacent to every vertex @xmath401 , @xmath384 , and @xmath402 is adjacent to every vertex @xmath403 , @xmath382 .
notice that every @xmath404-set is formed only by the vertices of @xmath405 and one vertex from each subgraph isomorphic to @xmath210 .
therefore , @xmath406 and @xmath12 is a vertex of degree one not belonging to any strong metric basis of @xmath386 .
the graphs @xmath407 and @xmath408 are shown in figure [ family - f-1 ] . and @xmath408 .
the set @xmath409 is a strong metric basis of @xmath407.,title="fig:",scaledwidth=40.0% ] and @xmath408 .
the set @xmath409 is a strong metric basis of @xmath407.,title="fig:",scaledwidth=40.0% ] let @xmath411 be the vertex set of @xmath0 and let @xmath374 with set of vertices @xmath412 , where @xmath376 , @xmath373 , @xmath375 and @xmath377 . since every vertex @xmath413 is maximally distant from @xmath3 , by lemma [ maximally_distant with root ]
, we have that every two different vertices @xmath414 , @xmath147 , are mutually maximally distant . moreover , by lemma [ rooted mutually_maximally_distant ] for every two mutually maximally distant vertices @xmath415 in @xmath386 we have that @xmath416 are mutually maximally distant in @xmath417 for every vertex @xmath2 of @xmath0 .
thus , @xmath418 is formed by @xmath419 connected components , i. e. , @xmath420 connected components isomorphic to @xmath210 and one connected component isomorphic to a graph @xmath421 with set of vertices @xmath422 where @xmath423 is isomorphic to @xmath424 and for every @xmath425 , @xmath426 is adjacent to every vertex @xmath427 , @xmath384 , and @xmath428 is adjacent to every vertex @xmath429 , @xmath382 .
since in @xmath421 every vertex of @xmath423 has a neighbor not belonging to @xmath430 we have that @xmath431 .
therefore , we obtain that @xmath432 according to the remark [ rem f-1-second ] we have that for every graph @xmath433 and any connected graph @xmath0 of order @xmath26 , @xmath434 where @xmath3 is the vertex of degree one not belonging to any strong metric basis of the graph @xmath75 .
[ vertex degree 1 ] let @xmath0 be a connected graph of order @xmath90 and let @xmath3 be a vertex of a graph @xmath75 . if @xmath3 does not belong to the boundary of @xmath75 and there exists a vertex different from @xmath3 , of degree one in @xmath75 , not belonging to any strong metric basis of @xmath75 , then @xmath436 let @xmath1 be a vertex of degree one in @xmath75 not belonging to any strong metric basis of @xmath75 . notice that the vertices of the set @xmath437 are also vertices of degree one in @xmath81 .
thus , they are simplicial vertices and from lemma [ lem_simplicial_v ] we have that at least all but one vertices of @xmath438 belongs to every strong metric basis of @xmath81 .
thus , @xmath439 [ rem f-1 ] let @xmath0 be a connected graph of order @xmath26 .
let @xmath3 be the vertex of the graph @xmath374 adjacent to the vertex of degree one not belonging to any strong metric basis of @xmath386 .
then @xmath440 let @xmath411 be the vertex set of @xmath0 .
now , according to the construction of the family @xmath371 , let the graph @xmath386 with set of vertices @xmath412 , where @xmath376 , @xmath373 , @xmath375 and @xmath377 . since every vertex @xmath441 is maximally distant from @xmath3 , by lemma [ maximally_distant with root ] , we have that every two different vertices @xmath442 , @xmath147 , are mutually maximally distant .
moreover , by lemma [ rooted mutually_maximally_distant ] for every two mutually maximally distant vertices @xmath415 in @xmath386 we have that @xmath416 are mutually maximally distant in @xmath417 for every vertex @xmath2 of @xmath0 .
thus , @xmath418 is formed by @xmath419 connected components , that is , @xmath420 connected components isomorphic to @xmath210 and one connected component isomorphic to a graph @xmath421 with set of vertices @xmath443 where @xmath444 is isomorphic to @xmath445 and for every @xmath425 , @xmath426 is adjacent to every vertex @xmath427 , @xmath384 , and @xmath428 is adjacent to every vertex @xmath429 , @xmath382 .
notice that @xmath446 .
therefore , we obtain that @xmath447 j. cceres , c. hernando , m. mora , i. m. pelayo , m. l. puertas , c. seara , and d. r. wood , on the metric dimension of cartesian product of graphs , _ siam journal on discrete mathematics _ * 21 * ( 2 ) ( 2007 ) 273302 . | let @xmath0 be a connected graph .
a vertex @xmath1 strongly resolves a pair @xmath2 , @xmath3 of vertices of @xmath0 if there exists some shortest @xmath4 path containing @xmath3 or some shortest @xmath5 path containing @xmath2 .
a set @xmath6 of vertices is a strong resolving set for @xmath0 if every pair of vertices of @xmath0 is strongly resolved by some vertex of @xmath6 .
the smallest cardinality of a strong resolving set for @xmath0 is called the strong metric dimension of @xmath0 .
it is known that the problem of computing this invariant is np - hard .
this suggests finding the strong metric dimension for special classes of graphs or obtaining good bounds on this invariant . in this paper
we study the problem of finding exact values or sharp bounds for the strong metric dimension of rooted product of graphs and express these in terms of invariants of the factor graphs .
_ keywords : _ strong metric dimension ; rooted product graphs ; strong metric basis ; strong resolving set . _
ams subject classification numbers : _ 05c12 ; 05c69 ; 05c76 . |
for the last 6 - 8 months , an intense discussion has been going on in the relativistic heavy - ion collision ( rhic ) community whether one can consider the discovery of the quark - gluon plasma ( qgp ) as settled and if it is time for the rhic program to move on , entering a new phase of detailed and precise measurements to explore the quantitative properties of this new state of hot and dense matter @xcite . whereas theorists seem to generally agree that qgp has been successfully created in au+au collisions at rhic @xcite , the experimental collaborations are more cautious in their assessments @xcite .
general agreement exists , however , that rhic has produced thermalized matter at unprecedented high energy density .
this conclusion is based on the successful hydrodynamic description of the measured hadron momentum spectra , in particular the elliptic flow and its `` fine structure '' , i.e. its dependence on the hadron mass @xcite .
hydrodynamics provides a direct link between the equation of state ( eos ) of the expanding fluid and the flow pattern manifested in the emitted hadron spectra . just like any other phase transition
, the quark - hadron phase transition in qcd generates a `` soft '' region in the eos within which the ability of the pressure to generate flow is greatly reduced .
this is expected to leave visible traces in the _ flow excitation function_. for example , hydrodynamics predicts that the elliptic flow coefficient @xmath0 should show a non - monotonic dependence on the collision energy @xcite , arising from a reduction in the elliptic flow signal if the collision fireball spends the crucial early stage , where such flow would be created , in the phase transition region .
a quantitative determination of the eos requires precision flow data as well as systematic theoretical studies of the influence of the initial conditions , equation of state , non - ideal transport effects , and the final decoupling kinetics on the observed hadron spectra . with the present theory and rhic data ,
we are just at the beginning of such a program . that the rhic data so far provide strong support for the basic validity of the hydrodynamic approach
is a good omen , raising the hope that such quantitative studies will be ultimately successful .
there would be much less optimism if rhic had repeated the history of lower - energy heavy - ion collisions where hydrodynamics never worked well enough to provide more than a rough qualitative understanding of the collision dynamics .
when the fireball dynamics proceeds too far away from local thermal equilibrium , it requires a microscopic transport theoretical description , involving a variety of non - equilibrium mechanisms with typically poorly constrained parameters .
this renders the extraction of the equation of state difficult and uncertain .
of course , even at rhic the hydrodynamic approach has its limitations .
ideal fluid dynamics , which assumes instantaneous local thermal equilibration , always remains an idealization ; however , quantitative control of possibly small non - ideal effects due to shear and bulk viscosity , heat conduction and diffusion requires a numerical implementation of _ viscous relativistic hydrodynamics _ which has its technical and conceptual difficulties and is only now beginning to become practical @xcite .
furthermore , in very peripheral heavy - ion collisions , where the size and lifetime of the collision fireball become small , even a viscous hydrodynamic approach is expected to eventually break down .
also , particles with very large transverse momenta ( jets ) are never expected to suffer sufficiently many interactions with the fireball medium to fully thermalize before escaping ; hence a hydrodynamic approach can never work at very high @xmath1 .
however , we can turn this inescapable failure of hydrodynamics in small collision systems and at high @xmath1 to our favor : since ideal fluid dynamics appears to work well in near - central collisions , near mid - rapidity , and at low @xmath2gev/@xmath3 @xcite , we can study its gradual breakdown at larger impact parameters , rapidities , and transverse momenta in order to learn something about the mechanisms for the _ approach to thermal equilibrium _ at the beginning of the collision and the _ decay of thermal equilibrium _ near the end of the expansion stage , and hence about the transport properties of the early quark - gluon plasma and the late hadron resonance gas created in these collisions . in this talk
i will lay out the beginnings of such a program .
if the dense matter formed in the nuclear collision thermalizes locally on a time scale much shorter than any macroscopic dynamical scale , its evolution can be described by ideal fluid dynamics .
the fluid s state is then completely determined by the space - time profiles for its energy density @xmath4 , baryon density @xmath5 , and flow 4-velocity @xmath6 .
the equation of state @xmath7 closes the set of equations .
small deviations from local thermal equilibrium result in viscous corrections to the ideal fluid decomposition of energy - momentum tensor and baryon current , involving gradients of the above profiles multiplied by transport coefficients . for large deviations from local thermal equilibrium
this expansion in terms of gradients of @xmath8 , @xmath9 and @xmath10 breaks down , and the macroscopic hydrodynamic approach must be replaced by a microscopic kinetic one which evolves the distribution functions in both coordinate and momentum space .
ideal fluid dynamics does not work during the initial collision stage , when the two colliding nuclei penetrate each other , converting some of their initial kinetic energy into creation of new matter , and it is also inapplicable during the last stage of the collision when the expanding matter becomes so dilute that interactions among its constituents are rare and local equilibrium breaks down .
the hydrodynamic model therefore needs _ initial conditions _
( supposedly the result of an early pre - equilibrium quantum kinetic evolution leading to initial thermalization ) and _ final conditions _ ( a freeze - out criterium describing the transition from a locally thermalized fluid to an ensemble of free - streaming , noninteracting hadrons ) .
both initial and final conditions involve free parameters which must be tuned to data .
as described in some detail in @xcite , they can all be fixed by only using data on the collision centrality dependence of the charged particle multiplicity and the _ pion _ and _ proton _ transverse momentum spectra in _ central _ ( @xmath11 ) collisions .
all other hadron spectra from central collisions , and all features of all hadron spectra in non - central collisions ( in particular the elliptic flow for all hadron species ) are then parameter - free predictions of the model @xcite . for 130@xmath12gev and 200@xmath12gev au+au collisions at rhic , these predictions were surprisingly successful ( see review in @xcite ) .
figure [ f1 ] shows the fit to the ( absolutely normalized ) pion and antiproton spectra and , ( color online ) negative pion , kaon , antiproton , and @xmath13 spectra from central au+au collisions at @xmath14gev , as measured by the four rhic experiments @xcite .
the curves show hydrodynamical calculations @xcite as described in the text.,width=234 ] ( color online ) negative pion , kaon , antiproton , and @xmath13 spectra from central au+au collisions at @xmath14gev , as measured by the four rhic experiments @xcite .
the curves show hydrodynamical calculations @xcite as described in the text.,width=234,height=169 ] as examples , the predictions for the kaon and @xmath13 spectra from 200@xmath12gev au+au collisions .
two sets of theoretical curves @xcite are shown : the lower ( blue ) bands correspond to kinetic decoupling directly after hadronization at @xmath15mev , whereas the upper ( red ) bands assume decoupling at @xmath16mev .
the width of the bands indicates the sensitivity of the calculated spectra to an initial transverse flow of the fireball already at the time of thermalization ( for details see @xcite ) .
the hydrodynamic model output shows @xcite that it takes about 9 - 10fm/@xmath3 until the fireball has become sufficiently dilute to completely convert to hadronic matter , and another 7 - 8fm/@xmath3 to completely decouple .
figure [ f1 ] shows clearly that by the time of hadronization hydrodynamics has not yet generated enough radial flow to reproduce the measured @xmath17 and @xmath13 spectra ; these heavy hadrons , which are particularly sensitive to radial flow effects , require the additional collective `` push '' created by resonant ( quasi)elastic interactions during the fairly long - lived hadronic rescattering stage between @xmath18 and @xmath19 .
the late hadronic stage of the collision thus plays an important role for the quantitative understanding of the shape of the single particle spectra at rhic and their radial flow .
this is different for the elliptic flow which is driven by the spatial anisotropy of the reaction zone in non - central collisions and the resulting anisotropic pressure gradients : as shown in @xcite , at rhic energies , for not too peripheral au+au collisions , the momentum anisotropy _ saturates before the completion of hadronization _ , due to the disappearance of the spatial deformation of the fireball .
the measured @xmath0 thus reflects almost exclusively the early qgp dynamics and the phase transition region .
figure [ f2 ] shows that the corresponding hydrodynamic predictions work very well , at least for transverse momenta below about 1.5 - 2gev/@xmath3 where most ( @xmath20 ) of the particles are emitted .
( color online ) differential elliptic flow @xmath21 for several identified hadron species from minimum bias au+au collisions at 130 ( right @xcite ) and 200@xmath12gev ( left @xcite ) , compared with hydrodynamic predictions @xcite . ,
width=264,height=188 ] ( color online ) differential elliptic flow @xmath21 for several identified hadron species from minimum bias au+au collisions at 130 ( right @xcite ) and 200@xmath12gev ( left @xcite ) , compared with hydrodynamic predictions @xcite . ,
width=236,height=178 ] hydrodynamics predicts a characteristic rest mass dependence of @xmath0 @xcite which is nicely reproduced by the data in fig .
this `` fine structure '' constitutes compelling evidence that the hot matter produced at rhic behaves like a thermalized fluid .
the `` mass scaling '' of @xmath0 seen at low @xmath1 is quite distinct from the `` quark number scaling '' of @xmath0 seen at larger transverse momenta @xmath22gev/@xmath3 @xcite ; the latter is characteristic of a quark - coalescence mechanism for hadron production which dominates in that @xmath1 range @xcite .
the mass scaling seen in the hydrodynamic regime is sensitive to the eos of the fireball matter : the right panel in fig .
[ f2 ] shows two calculations , one without a phase transition ( eos h ) , the other including the quark - hadron transition ( eos q ) .
although neither one fits the star proton data particularly well , the latter show a clear preference for the eos including a phase transition .
the elliptic flow coefficient @xmath21 is a measure for the ( relatively small ) differences between the transverse momentum spectra with momenta pointing into and perpendicular to the reaction plane . as such
, it is more sensitive to deviations from ideal hydrodynamic behaviour than the overall shapes ( slope parameters ) of the transverse momentum spectra .
two model studies @xcite showed that @xmath0 reacts particularly strongly to shear viscosity .
as the mean free path of the plasma constituents ( and thus the fluid s viscosity ) goes to zero , @xmath0 approaches the ideal hydrodynamic limit from below @xcite ( see fig.[f3]a ) . at higher transverse momenta
it does so more slowly than at low @xmath1 @xcite , approaching a constant saturation value as seen in the data rather than following the hydrodynamic almost linear rise as a function of @xmath1 shown by the lines in fig.[f2 ] . the increasing deviation from
the ( color online ) left : elliptic flow from a parton cascade @xcite , compared with star data , for different parton - parton scattering cross sections .
larger cross sections lead to smaller mean free paths .
right : perturbative effects of shear viscosity on the elliptic flow @xmath21 @xcite ( see text for discussion ) . , width=264,height=196 ] + ideal fluid limit for growing @xmath1 is qualitatively consistent with the expected influence of shear viscosity : teaney @xcite showed that viscous corrections to the local thermal equilibrium distribution function increase in first order quadratically with @xmath1 , leading to stronger and stronger reduction of @xmath0 below the ideal fluid limit as @xmath1 grows ( see fig.[f3]b ) . from the results in fig.[f3]b teaney
concluded that at rhic the normalized sound attenuation length @xmath23 ( where @xmath24 is the shear viscosity , @xmath25 the temperature and @xmath26 the entropy density ) can not be much larger than about 0.1 .
this puts a stringent limit on the dimensionless ratio @xmath27 , bringing it close to the recently conjectured absolute lower limit for the viscosity of @xmath28 @xcite .
this would make the quark - gluon plasma the most ideal fluid ever observed @xcite .
these arguments show that deviations from ideal fluid dynamics at high @xmath1 must be expected , and that they can be large even for fluids with very low viscosity .
at which @xmath1 non - ideal effects begin to become visible in @xmath21 can be taken as a measure for the fluid s viscosity .
figure [ f2]a , however , shows that mesons seem to deviate from the ideal fluid dynamical prediction earlier than baryons which follow the hydrodynamic curves to considerably larger @xmath1 , thereby reaching significantly larger saturation values for @xmath0
. why should viscous effects be different for baryons and mesons ?
in fact , both the @xmath1-value where @xmath0 begins to drop below the hydrodynamic prediction and the asymptotic saturation values of @xmath0 appear to be independent of the hadron mass and only care about whether the hadron is a meson or a baryon !
this `` valence quark number scaling '' of @xmath0 at intermediate and high @xmath1 can be explained through quark coalescence @xcite .
this model predicts that the elliptic flow divided by the number of valence quarks @xmath9 , plotted against @xmath29 , should be a universal function which reflects the partonic collective elliptic flow just before hadronization : @xmath30 @xcite . for @xmath31mev ,
this scaling works extremely well @xcite , at _ all collision centralities _ ( note that @xmath0 is a strong function of centrality ! ) .
strange and non - strange partons seem to carry the same elliptic flow @xcite , consistent with strong rescattering and thermalization in the partonic phase .
below @xmath32mev the extracted partonic elliptic flow follows the hydrodynamically predicted almost linear rise with @xmath1 , and quark number scaling is violated and replaced by the hydrodynamically predicted mass scaling . in summary ,
the critical value of @xmath1 where qgp viscosity manifests itself in the partonic elliptic flow is @xmath33mev/@xmath3 .
what this implies for the value of the qgp shear viscosity @xmath24 remains to be seen , by comparing the data to simulations with a viscous relativistic hydrodynamic code .
hydrodynamics also fails to describe the elliptic flow @xmath0 in more peripheral au+au collisions at rhic , in central and peripheral collisions at lower energies ( for both see fig .
[ f4]a ) , and in minimum bias collisions at rhic at forward rapidities @xcite .
whereas hydrodynamics predicts a non - monotonic beam energy dependence of @xmath0 @xcite , with largest values at upper ags and lower sps energies ( see fig .
[ f4]a ) , somewhat lower values at rhic and again larger values at the lhc , the data seem to increase monotonically with @xmath34 .
( color online ) left : scaled elliptic flow @xmath35 ( where @xmath36 denotes the initial spatial eccentricity ) vs. the charged multiplicity density per unit initial transverse overlap area @xmath37 @xcite .
right : elliptic flow @xmath0 for minimum bias au+au collisions at various collision energies , parametrized by the final charged multiplicity density at midrapidity @xcite .
see text for discussion . , width=264,height=207 ] + ( color online ) left : scaled elliptic flow @xmath35 ( where @xmath36 denotes the initial spatial eccentricity ) vs. the charged multiplicity density per unit initial transverse overlap area @xmath37 @xcite .
right : elliptic flow @xmath0 for minimum bias au+au collisions at various collision energies , parametrized by the final charged multiplicity density at midrapidity @xcite .
see text for discussion . , title="fig:",width=235,height=30 ] + such a monotonic rise is consistent with `` hybrid '' calculations by teaney @xcite ( fig .
[ f4]b ) where the fireball undergoes ideal fluid dynamic evolution only to a point just below the hadronization phase transition , followed by hadronic _ kinetic _ evolution using the rqmd code until final decoupling .
figure [ f4]b shows several curves corresponding to different equations of state during the hydrodynamic evolution ( see @xcite for details ) , with lh8 being closest to the data .
the difference between the points labelled lh8 and the hydrodynamic values at the top of the figure is due to the different evolution during the late hadronic stage .
obviously , at lower collision energies and for impact parameters @xmath38fm ( which dominate @xmath0 in minimum bias collisions ) , ideal fluid dynamics continues to build additional elliptic flow during the hadronic stage , in contrast to rqmd .
the reason for the additional hadronic @xmath0 in the hydrodynamic simulations is that at lower beam energies the initial energy densities are smaller and the fireball does not spend enough time in the qgp phase for the spatial eccentricity @xmath36 to fully disappear before entering the hadron resonance gas phase .
anisotropic pressure gradients thus still exist in the hadronic phase , and hydrodynamics reacts to them according to the stiffness of the hadron resonance gas eos ( @xmath39 ) .
calculations @xcite show that rqmd reacts to these remaining anisotropies much more weakly , building very little if anny additional elliptic flow during the hadronic phase .
the hadron resonance gas , as modelled in rqmd , is a highly viscous medium , unable to reach or maintain local thermal equilibrium and therefore not behaving as an ideal fluid at all .
the failure of the hydrodynamic model in peripheral collisions at rhic and in central and peripheral collisions at the sps and ags is therefore _ not _ necessarily caused by the absence of an ideal fluid qgp phase during the early expansion stage , but rather by the highly viscous late hadronic stage which is unable to efficiently react to the remaining spatial fireball eccentricity .
similar arguments hold at forward rapidities at rhic @xcite where the initial energy densities are also significantly smaller than at midrapidity while the initial spatial eccentricities are similar . the large hadron gas viscosity spoils one of the clearest experimental signatures for the existence of the quark - hadron phase transition , the predicted non - monotonic beam energy dependence of @xmath0 @xcite
. the ideal fluid dynamic limit for @xmath0 scales with the speed of sound @xmath40 .
as one passes from the hadron resonance gas through the phase transition into the qgp , @xmath41 varies from about 0.15 in the hadron gas to about @xmath42 in the qgp , going through a minimum near zero in the soft transition region .
ideal fluid dynamics maps this behaviour directly onto @xmath0 as one changes the initial energy density by increasing the collision energy . as one comes down from infinite beam energy
, @xmath0 is therefore predicted to first decrease ( due to the softening of the eos in the phase transition region ) and then recover somewhat in the moderately stiff hadron gas phase .
the hadron gas viscosity spoils this recovery , leading to an apparently monotonous decrease of @xmath0 with falling beam energy ( fig .
however , recent phenix data from au+au collisions at @xmath43gev @xcite indicate that this decrease may not be quite as monotonous as suggested by fig .
they show essentially constant elliptic flow over the entire energy range explored at rhic ( from 62 to 200 @xmath12 gev ) , decreasing only when going further down to sps and ags energies .
while this is does not confirm the hydrodynamically predicted _ rise _ of @xmath0 , it may still be a strongly diluted reflection of this predicted non - monotonic structure in the elliptic flow excitation function .
obviously , many and more systematic hybrid calculations of the type pioneered by teaney @xcite are necessary to explore to what extent we can eventually prove the existence of a qcd phase transition using elliptic flow data .
we recently pointed out @xcite a very nice possibility to extend fig .
[ f4]a towards the right , in order to confirm that for higher initial energy densities the measured @xmath0 indeed settles on the ideal hydrodynamic curve . by colliding deformed uranium nuclei at rhic , using the zero degree calorimeters to select only full overlap collisions , and then exploiting the known binary collision component in the produced charged particle multiplicity to select configurations with different initial spatial eccentricities by cutting on multiplicity
, one can extend the explored region in fig .
[ f4]a by about 60% along the horizontal axis .
in fact , the lowest value for @xmath44 obtainable in this way is for u+u collisions in the side - on - side configuration , which reproduce similar initial energy densities as central au+au collisions , but with a large initial eccentricity of 25% .
the opposite limit is reached for u+u collisions in the nose - on - nose configuration where the elliptic flow approaches zero but @xmath44 is about 60% higher than the largest value plotted in fig .
the excellent reproduction of all aspects of single - particle hadron spectra measured in central and semicentral au+au collisions at rhic , including the elliptic flow and its fine structure , by relativistic hydrodynamics provides strong evidence for the creation of a thermalized medium at high energy density @xmath45gev/@xmath46 with thermalization time @xmath47fm/@xmath3 .
the only known thermalized state at such energy densities it the quark - gluon plasma .
the observation of quark - number scaling of the elliptic flow and other observables at intermediate @xmath1 indicates that deconfined valence quarks play a dynamical role in hadron production , providing at least indirect evidence for color deconfinement .
the almost ideal fluid behaviour of the expanding fireball shows that the qgp is a strongly coupled plasma with fluid - like rather than gas - like properties and with the lowest viscosity of any known liquid .
in contrast to the qgp , the hadron gas dominating the late expansion stage is highly viscous and does not behave like an ideal fluid .
hydrodynamics works for the first time quantitatively at rhic since the fireball dynamics is for the first time dominated by the qgp . we have reached an interesting stage where we are beginning to explore the detailed properties of this new qgp state of matter . clearly , many detailed studies , both theoretical and experimental , are required to achieve that goal .
systematic hydrodynamic studies , combined with high statistics data on spectra and elliptic flow , should help to further constrain the thermalization time @xmath48 and equation of state .
hybrid hydro+cascade calculations should be performed to isolate the non - ideal effects from the late hadronic stage .
a 3 + 1-dimensional relativistic viscous hydrodynamic code must be developed to quantitatively constrain the transport properties of the qgp ( such as its viscosity ) from the data , and thereby provide phenomenological constraints for theoretical efforts to calculate the latter from first principles .
99 see articles by gyulassy m and mclerran l , shuryak e v , mller b , wang x n , stcker h , and blaizot j p and gelis f 2004 in : _ new discoveries at rhic the strongly interacting qgp _
( rbrc scientific articles , vol . 9 , brookhaven national laboratory , upton , ny , usa ) , to appear as a special volume in _ nucl . phys .
_ a. adler c _ et al . _ ( star collab . ) .
greco v , ko c m , and lvai p and _ phys .
_ c * 68 * 034904 ; fries r j , mller b , nonaka c , and bass s a and _ phys .
_ c * 68 * 044902 ; molnar d and voloshin s a .
heinz u and wong s m h .
communication .
molnar d and gyulassy m ( erratum _ ibid .
_ * 703 * 893 ) .
kovtun p , son d t , and starinets a o .
alt c _ et al . _
( na49 collab . ) .
teaney d , lauret j , and shuryak e v .
heinz u and kolb p f 2004 _ j. phys .
_ g : nucl . part .
phys . * 30 * s1229 .
adler s s _ et al . _
( phenix collab . ) . | i review the successes and limitations of the ideal fluid dynamic model in describing hadron emission spectra from au+au collisions at the relativistic heavy ion collider ( rhic ) . |
explicit strong stability preserving ( ssp ) runge kutta methods were developed for the time evolution of hyperbolic conservation laws @xmath0 , with discontinuous solutions @xcite .
these works studied total variation diminishing ( tvd ) spatial discretizations that can handle discontinuities .
the spatial discretizations used to approximate @xmath1 were carefully designed so that when the resulting system of odes @xmath2 ( where @xmath3 is a vector of approximations to @xmath4 , @xmath5 ) is evolved in time using the _ forward euler method _ , the solution at time @xmath6 satisfies a strong stability property of the form @xmath7 under a step size restriction @xmath8 the term @xmath9 can represent , as it did in @xcite the total variation semi - norm , or indeed any other semi - norm , norm , or convex functional , as determined by the design of the spatial discretization .
these spatial discretizations satisfy the strong stability property @xmath10 _ when coupled with the forward euler time discretization _ , but in practice a higher order time integrator , that will still satisfy this property , is desired . to accomplish this
, we attempt to re - write a higher order time discretization as a convex combination of forward euler steps , so that any convex functional property that is satisfied by the forward euler method will still be satisfied by the higher order time discretization .
an @xmath11-stage explicit runge kutta method can be written in the form @xcite , @xmath12 if all the coefficients @xmath13 and @xmath14 are non - negative , and a given @xmath13 is zero only if its corresponding @xmath14 is zero , then each stage can be rearranged into a convex combination of forward euler steps @xmath15 where if any of the @xmath16 s are equal to zero , the corresponding ratio is considered infinite .
the last inequality above follows from the strong stability conditions and @xmath17 and the consistency condition @xmath18 . from this
we can conclude that whenever the explicit runge kutta method can be decomposed into convex combinations of forward euler steps , then any convex functional property satisfied by forward euler will be _ preserved _ by the higher - order time discretizations , perhaps under a different time - step restriction @xmath19 @xcite .
thus , this type of decomposition where @xmath20 is clearly a sufficient condition for strong stability preservation .
it has also been shown @xcite that this convex combination condition is necessary for strong stability preservation .
if a method does not have a convex combination decomposition into forward euler steps we can find some ode with some initial condition such that the forward euler condition is satisfied but the method does not satisfy the strong stability condition for any positive time - step @xcite .
methods that can be decomposed like this with with @xmath21 are called strong stability preserving ( ssp ) , and the coefficient @xmath22 is known as the _ ssp coefficient _ of the method .
ssp methods guarantee the strong stability of the numerical solution for any ode and any convex functional provided only that the forward euler condition is satisfied under a time step .
this is a very strong requirement that leads to severe restrictions on the allowable order of ssp methods , and on the size of the allowable time step @xmath23 we seek high order ssp runge kutta methods with the largest allowable time - step .
the forward - euler time step @xmath24 is a property of the spatial discretization method only , and so our aim in searching for time - stepping methods that preserve the strong stability property is to maximize the _ ssp coefficient _
@xmath22 of the method .
a more relevant quantity may be the total cost of the time evolution , which in our case translates into the allowable time step relative to the number of function evaluations at each time - step ( typically the number of stages of a method ) .
for this purpose we define the _ effective ssp coefficient _ @xmath25 where @xmath11 is the number of stages .
this value allows us to compare the efficiency of explicit methods of a given order .
it has been shown @xcite that all explicit @xmath11-stage runge kutta methods have an ssp bound @xmath26 , and therefore @xmath27 , but this upper bound is not always attained .
in @xcite it was shown that explicit ssp runge kutta methods can not exist for order @xmath28 however , in the special case where we consider only linear autonomous problems , explicit ssp runge kutta methods exist for any _ linear _ order @xmath29 @xcite .
the linear and nonlinear order conditions are equivalent up to and including order @xmath30 , so in this work we consider explicit ssp runge kutta methods that have nonlinear order @xmath31 and @xmath32 , and have higher linear orders @xmath33 . in section [ sec : background ]
we review the ssp properties of explicit runge kutta methods and discuss the linear and nonlinear order conditions . using these order conditions and
optimization problem described in @xcite , in section [ sec : optimization ] we describe the optimization code in matlab ( based on @xcite ) used to find explicit runge kutta methods that have @xmath34 with optimal ssp coefficient . in section
[ sec : optimal ] we list some of the new methods and their effective ssp coefficients , and in section [ sec : test ] we demonstrate the performance of these methods on a selection of test problems .
strong stability preserving methods were first developed by shu @xcite for use with total variation diminishing spatial discretizations . in these works , the authors presented second and third order methods that have @xmath35 .
the explicit ssp runge kutta method of order @xmath36 @xmath37 and the @xmath38 method @xmath39 these methods were proven optimal in @xcite .
it was shown in @xcite that no four stage fourth order explicit runge kutta methods exist with positive ssp coefficient . by considering methods with @xmath40 , fourth order methods with order @xmath32
have been found .
notable among these is the @xmath41 method with @xmath42 ( @xmath43 ) in @xcite @xmath44 and the @xmath45 method with @xmath46 ( @xmath47 ) in @xcite @xmath48 it was shown @xcite that no methods of order @xmath49 with positive ssp coefficients can exist .
this restrictive order barriers on explicit ssp runge kutta methods stem in part from the nonlinearity of the odes . for order of accuracy on linear autonomous ode systems , explicit ssp runge
kutta methods need only satisfy a smaller set of order conditions .
if we require only that the method have high linear order ( @xmath29 ) , then the order barrier is broken and explicit runge kutta methods with positive ssp coefficients exist for arbitrarily high linear orders .
optimally contractive explicit runge kutta methods were studied by kraaijevanger in @xcite , where he gives optimal _ linear _ methods for many values of @xmath11 and @xmath50 , including @xmath51 , and @xmath52 for any @xmath11 .
these methods are interesting because their ssp coefficients serve as upper bounds for nonlinear methods , but they may also be useful in their own right .
although ssp methods were first developed for nonlinear problems , the strong stability preserving property can be useful for linear problems such as maxwell s equations and linear elasticity .
first and second methods that have @xmath11 stages have been shown to attain the theoretical bound @xmath53 .
methods of order @xmath29 with @xmath54 and @xmath55 also exist with @xmath53 .
these methods can be found in @xcite , and are given here in their canonical shu - osher form : the family of @xmath11-stage , linear order @xmath56 methods has @xmath57 and @xmath58 : @xmath59 where the coefficients @xmath60 of the final stage of the @xmath11-stage method are given iteratively by @xmath61 starting from the coefficients of the @xmath62-stage , first order method @xmath63 and @xmath64 .
the family of @xmath11-stage , linear order @xmath65 methods has @xmath35 and @xmath66 : @xmath67 here the coefficients @xmath60 of the final stage of the @xmath11-stage method are given iteratively by @xmath68 starting from the coefficient of the forward euler method @xmath69 . however , all these methods with high linear order @xmath29 have low nonlinear order @xmath30 .
the idea that we pursue in this paper is the construction of explicit ssp runge kutta methods that have a high linear order @xmath70 while retaining the highest possible nonlinear order @xmath32 .
we also consider methods with @xmath31 and @xmath71 for comparison .
the idea behind these methods is that they would be the best possible methods ( in terms of ssp coefficient and order ) for linear problems , without compromising order when applied to nonlinear problems .
the shu - osher form of an explicit runge kutta method , given in , is most convenient for observing the ssp coefficient .
however , this form is not unique , and not the most efficient form to use for the optimization procedure @xcite .
the butcher form of the explicit method given by @xmath72 ( where the coefficients @xmath73 are place into the matrix @xmath74 and @xmath75 into the row vector @xmath76 ) is unique , so rather than perform a search for the optimal convex combination of the shu - osher form , we define the optimization problem in terms of the butcher coefficients . the conversion from the shu - osher form to butcher form , and from an optimal butcher form to the canonical shu - osher form is discussed in @xcite . we follow the approach developed by david ketcheson and successfully used in @xcite : we search for coefficients @xmath74 and @xmath76 that maximize the value @xmath77 subject to constraints : @xmath78 where @xmath79 are the order conditions . after this optimization
we have the coefficients @xmath74 and @xmath76 and an optimal value @xmath80 that define the method .
the order conditions @xmath81 appear as the equality constraints on the optimization problem . in this work
, we consider methods that have @xmath31 and @xmath32 but have higher linear order @xmath70 . in this subsection
, we list these order conditions . *
linear order conditions : * given a runge kutta method written in the butcher form with coefficients @xmath74 and @xmath76 ( and @xmath82 where @xmath83 is a vector of ones ) , the order conditions that guarantee @xmath29 order accuracy for a linear problem can be simply expressed as @xmath84 * nonlinear order conditions : * if we want a method to demonstrate the correct order of accuracy for nonlinear problems , the first and second order conditions are the same as above : @xmath85 a method that satisfies these conditions will be second order for both linear and nonlinear problems .
two additional conditions are required for third order accuracy : @xmath86 note that when the first of these conditions is satisfied , the second condition is equivalent to @xmath87 , which is the linear third order condition .
four more conditions are required for the method to be fourth order for a nonlinear problem @xmath88 note that the first three conditions together imply the fourth order linear order condition @xmath89 in this work we consider the nonlinear order conditions only up to @xmath32 because it is known that there are no explicit ssp runge kutta methods greater than fourth order , but we consider higher order linear order conditions @xmath70 .
using david ketcheson s matlab optimization code @xcite with our modified order conditions ( described in section [ orderconditions ] ) we produce the optimal linear / nonlinear ( lnl ) methods in this section .
this r0.625 code formulates the optimization problem in section [ sec : optimization ] in matlab and uses fmincon to find the coefficients @xmath74 and @xmath76 that yield the largest possible @xmath22 .
we set the tolerances on fmincon to @xmath90 .
we used this code to generate methods with @xmath91 and @xmath92 .
we compare these methods with @xmath30 `` linear '' methods that we generated and matched to known optimal methods .
our primary interest is the size of the ssp coefficient for each method .
we denote the ssp coefficient for a method with @xmath11 stages , linear order @xmath29 and nonlinear order @xmath93 method @xmath94 .
the ssp coefficients for the @xmath31 methods with a given number of stages and linear order are the same as for the corresponding linear methods , ( i.e. @xmath95 ) .
this indicates that , for these values of @xmath11 and @xmath29 the additional condition @xmath96 needed for nonlinear third order does not pose additional constraints on the strong stability properties of the method .
table [ tab : p2p3 ] shows the ssp coefficients of the @xmath30 and @xmath31 methods with @xmath97 .
the coefficients for @xmath98 are known to be optimal because they match the linear threshold in kraiijevanger s
paper @xcite .
r0.67 table [ tab : p4 ] shows the ssp coefficients of the @xmath32 methods for @xmath97 .
in bold are the coefficients that match those of the @xmath30 methods .
in general , as we increase the number of stages the ssp coefficients for the @xmath32 methods approach those of the @xmath30 methods , as shown in figure [ fig : p2p4a ] .
the tables clearly show that the size of the ssp coefficient depends on the relationship between @xmath29 and @xmath11 , so it is illuminating to look at the methods along the diagonals of these tables . clearly , for @xmath54 methods we have an optimal value of @xmath35 and @xmath66 .
the @xmath30 and @xmath31 methods all attain this optimal value , but for @xmath32 we have @xmath99 and @xmath100 . however , once we get to a high enough number of stages , all the methods with @xmath32 and @xmath101 that have @xmath35 and @xmath66 .
figure [ fig : p2p4b ] shows that for the linear methods ( @xmath30 ) the ssp coefficient is fixed for @xmath102 ( blue dotted line ) , @xmath103 ( red dotted line ) , @xmath104 ( green dotted line ) , @xmath105 ( black dotted line ) , and @xmath106 ( cyan dotted line ) , and that the ssp coefficient of the corresponding @xmath32 methods ( solid lines ) approach these as the number of stages increases .
( dotted line ) and @xmath32 ( solid line ) for @xmath102 ( blue ) , @xmath103 ( red ) , @xmath104 ( green ) , @xmath105 ( black ) , and @xmath106 ( cyan ) .
the ssp coefficient of the @xmath32 methods ( solid lines ) approach those of the corresponding @xmath30 method as the number of stages increases , scaledwidth=113.0% ] ( dotted line ) and @xmath32 ( solid line ) for @xmath102 ( blue ) , @xmath103 ( red ) , @xmath104 ( green ) , @xmath105 ( black ) , and @xmath106 ( cyan ) .
the ssp coefficient of the @xmath32 methods ( solid lines ) approach those of the corresponding @xmath30 method as the number of stages increases , scaledwidth=113.0% ] it is interesting to note that the linear stability regions of the @xmath30 , @xmath31 methods are generally identical .
the @xmath32 methods have stability regions that are virtually identical to those of the linear methods when the ssp coefficient is identical .
in addition , methods with @xmath54 and @xmath32 all have the same stability regions as the corresponding linear methods , which is not surprising as the stability polynomial of @xmath54 is unique . for the rest of the methods
, we observe that for a given number of stages @xmath11 , as the linear order @xmath29 increases the linear stability regions of the @xmath32 methods look closer to those of the linear methods .
a nice illustration of this is the family of @xmath107 methods , shown in figure [ fig : linstab2 ] .
it is known in the literature that some methods with nonlinear orders @xmath31 and @xmath32 achieve the linear threshold value .
a nice example of this is ketcheson s ssp runge
kutta method of @xmath108 , @xmath109 , which achieves the threshold value @xmath110 .
this suggests that the linear order conditions are very significant to the value of the ssp coefficient .
indeed , we see this relationship in tables [ tab : p2p3 ] and [ tab : p4 ] , as we move right from column to column we see a significant drop in ssp coefficient . for each application
, one must decide if a higher linear order is valuable , as we pay a price for requiring additional @xmath29 .
however , once one has decided that the cost of a higher linear order is useful , there is no penalty in terms of ssp coefficient for requiring a higher nonlinear order @xmath31 and , in most cases , little reason not to use @xmath32 .
our results show that if one wishes to use a method with high linear order @xmath70 , then requiring @xmath31 or even @xmath32 rather than the standard @xmath30 is not usually associated with significant restriction on the ssp coefficient .
this can be beneficial in cases where the solution has linear and nonlinear components that need to be accurately captured simultaneously , or in different regions , or at different time - levels , so that the use of an ssp method that has optimal nonlinear order and higher linear order would be best suited for all components of the solution . for linear ( blue )
, @xmath31 ( red ) and @xmath32 ( green ) methods .
the @xmath32 methods approach the @xmath31 and @xmath30 methods as @xmath29 increases.,title="fig : " ] for linear ( blue ) , @xmath31 ( red ) and @xmath32 ( green ) methods .
the @xmath32 methods approach the @xmath31 and @xmath30 methods as @xmath29 increases.,title="fig : " ] + for linear ( blue ) , @xmath31 ( red ) and @xmath32 ( green ) methods .
the @xmath32 methods approach the @xmath31 and @xmath30 methods as @xmath29 increases.,title="fig : " ] for linear ( blue ) , @xmath31 ( red ) and @xmath32 ( green ) methods .
the @xmath32 methods approach the @xmath31 and @xmath30 methods as @xmath29 increases.,title="fig : " ] +
in this section , the optimized lnl methods described in section [ sec : optimal ] are tested for convergence and ssp properties .
first , we test these methods for convergence on both odes and pdes to confirm that the new methods exhibit the desired linear and nonlinear orders .
next , we study the behavior of these methods in conjunction with a higher order weno spatial discretizations , and show that although the weno method is nonlinear , when applied to a linear smooth problem the higher order linear order @xmath29 is beneficial .
on the other hand , for nonlinear problems , both with shocks and without , the higher order nonlinear order @xmath32 is advantageous . finally , the lnl methods are tested on linear and nonlinear problems with spatial discretizations that are provably total variation diminishing ( tvd ) and positivity preserving .
we show that for the linear case , the observed time - step for the time stepping method to preserve the tvd property is well - predicted by the theoretical ssp coefficient , while for positivity and for the nonlinear problem the theoretical time - step serves as a lower bound .
* example 1 : nonlinear ode convergence study . *
the van der pol problem is a nonlinear system of odes : @xmath111 we use @xmath112 and initial conditions @xmath113 .
this was run using @xmath114 ( 9,6,2 ) , ( 9,6,3 ) , ( 9,6,4 ) , ( 10,9,2 ) , ( 10,9,3 ) , and ( 10,9,4 ) lnl runge kutta methods to final time @xmath115 , with @xmath116 where @xmath117 .
the exact solution ( for error calculation ) was calculated by matlab s ode45 routine with tolerances set to abstol=@xmath90 and reltol=@xmath118 . in figure
[ fig : vdpconv ] we show that the @xmath119 of the errors in the first component vs. the @xmath119 of the number of points .
the slopes of these lines ( i.e the orders ) are calculated by matlab s polyfit function . as expected
, the rate of convergence follows the nonlinear order of the method .
in fact , we observe that a higher linear order is of no benefit at all for this example .
* example 2 : pde convergence study .
* in this study we solve the linear advection equation with sine wave initial conditions and periodic boundaries @xmath120 \\
\nonumber u(0,x ) & = & \sin(4\pi x ) \ ; \ ; \ ; \ ; u(t,0)=u(t,1 ) \nonumber\end{aligned}\ ] ] the fourier spectral method was used to discretize in space using @xmath121 points . the exact solution to this problem
is a sine wave with period @xmath122 that travels in time , so the fourier spectral method gives us an exact solution in space @xcite once we have two points per wavelength , allowing us to isolate the effect of the time discretization on the error .
we run this problem for five methods with orders @xmath123 , @xmath124 .
our final time is @xmath125 , and the time step @xmath126 , where @xmath127 .
errors are computed at the final time by comparison to the exact solution .
figure [ fig : spectral ] shows the @xmath119 of the @xmath128 norm of the errors vs. @xmath119 of the number of points .
the slopes of these lines ( i.e the orders ) are calculated by matlab s polyfit function , and demonstrate that the methods achieved the expected linear convergence rates .
, on the right @xmath32 . in both cases , the linear order dominates.,title="fig : " ] , on the right @xmath32 . in both cases , the linear order dominates.,title="fig : " ] * example 3 : linear advection with weno * we repeat the example above , this time using the 15th order ( @xmath129 ) weno method to discretize in space with @xmath127 for @xmath130 .
the weno method is a nonlinear method , so that even if the pde is linear , the resulting system of odes is not .
however , we can decompose the weno method into a linear part and a nonlinear correction term that suppresses oscillations . in theory ,
when the equation is linear and solution is smooth , the weno method is close to linear .
we test this problem with selected lnl runge
kutta time discretizations of linear order @xmath131 and @xmath132 , and with the shu - osher ssp runge kutta ( 3,3 ) and ketcheson s ssp runge kutta ( 10,4 ) . as above
, our final time is @xmath125 , and the time step @xmath126 , where @xmath127 .
errors are computed at the final time by comparison to the exact solution .
figure [ fig : linweno ] shows the @xmath119 of the @xmath128 norm of the errors vs. @xmath119 of the number of points , and the slopes of these lines ( i.e the orders ) as calculated by matlab s polyfit function .
we observe that the linear order dominates for this problem , which indicates that in regions where the problem is primarily linear and the solution smooth , the new lnl methods with higher linear orders could be of benefit .
* example 4 : burgers equation with weno .
* in the previous example we demonstrated the advantages of using a time - stepping method with higher @xmath29 with weno in the case of a linear , smooth problem . in this example
, we show how a higher nonlinear order @xmath50 is beneficial when dealing with a nonlinear equation with possibly discontinuous solution .
consider burgers equation with symmetric sine wave initial conditions and periodic boundaries .
@xmath133 \\
\nonumber u(0,x ) & = & \sin(2\pi x ) \ ; \ ; \ ; \ ; u(t,0)=u(t,1 ) .
\nonumber\end{aligned}\ ] ] this problem develops a standing shock .
we use a 15th order weno scheme with @xmath134 points in space , and test the lnl time - stepping methods of linear order @xmath135 and nonlinear order @xmath136 .
we use a time - step @xmath137 where @xmath138 . in figure [
fig : burgersweno ] we show the absolute values of the pointwise errors at spatial location @xmath139 for @xmath140 ( top ) and for @xmath141 ( bottom ) .
these errors are shown before the shock forms ( at time @xmath142 , solid line ) and after the shock forms ( at time @xmath143 , dotted line ) .
observe that for smaller number of spatial points the errors decays very fast , however once we reach a spatial refinement that is small enough we see that the methods with higher @xmath50 have significantly smaller errors . if we consider only @xmath144 points , we see the nonlinear order generally dominating : the linear methods feature second order convergence both pre- and post - shock , while the @xmath32 methods are fourth order pre - shock , but jump to twelfth order post - shock ( probably capturing the high order weno behavior )
. taken together with the problem in example 3 , this suggests that using a method with high linear and high nonlinear order may be beneficial in examples that have smooth and linear regions and non - smooth nonlinear regions . , before ( solid lines ) and after ( dotted line ) the shock .
the top figures show the @xmath119 of the errors vs. the number of points @xmath134 .
the bottom plots show the @xmath119 of the errors vs. @xmath141 for @xmath144 .
the time stepping methods use are @xmath145 and @xmath146 with nonlinear orders @xmath30 ( red ) and @xmath32 ( blue ) .
clearly , methods with higher nonlinear order in time give smaller errors.,title="fig : " ] , before ( solid lines ) and after ( dotted line ) the shock .
the top figures show the @xmath119 of the errors vs. the number of points @xmath134 .
the bottom plots show the @xmath119 of the errors vs. @xmath141 for @xmath144 .
the time stepping methods use are @xmath145 and @xmath146 with nonlinear orders @xmath30 ( red ) and @xmath32 ( blue )
. clearly , methods with higher nonlinear order in time give smaller errors.,title="fig : " ] + , before ( solid lines ) and after ( dotted line ) the shock .
the top figures show the @xmath119 of the errors vs. the number of points @xmath134 .
the bottom plots show the @xmath119 of the errors vs. @xmath141 for @xmath144 . the time stepping methods use
are @xmath145 and @xmath146 with nonlinear orders @xmath30 ( red ) and @xmath32 ( blue ) .
clearly , methods with higher nonlinear order in time give smaller errors.,title="fig : " ] , before ( solid lines ) and after ( dotted line ) the shock .
the top figures show the @xmath119 of the errors vs. the number of points @xmath134 .
the bottom plots show the @xmath119 of the errors vs. @xmath141 for @xmath144 .
the time stepping methods use are @xmath145 and @xmath146 with nonlinear orders @xmath30 ( red ) and @xmath32 ( blue ) .
clearly , methods with higher nonlinear order in time give smaller errors.,title="fig : " ] * example 5 : positivity and tvd time - step for a linear advection equation with first order finite difference in space .
* consider the linear advection equation with a step function initial condition : @xmath147 on the domain @xmath148 $ ] with periodic boundary conditions .
we used a first - order forward difference to semi - discretized this problem on a grid with @xmath149 points and evolved it to a final time of @xmath150 . for this problem
it is known that euler s method is tvd and positive for step sizes up to @xmath151 .
we computed the numerical solution using all the @xmath30 and @xmath32 ssp lnl runge kutta methods described in section [ sec : optimal ] and , for each one , found the largest @xmath152 for which tvd and positivity are preserved . in figure
[ fig : linadvfd ] we plot these values ( blue for tvd , green for positivity ) compared to the time - step guaranteed by the theory , @xmath153 ( in red ) .
we observe that the theoretical value is an excellent predictor of the observed tvd time - step ; in fact , the blue line is frequently indistinguishable from the red line .
the positivity preserving time - step is slightly larger , but follows a similar line .
* example 6 : tvd and positivity for buckley - leverett with centered scheme and koren limiter .
* we solve the buckley - leverett equation , a nonlinear pde used to model two - phase flow through porous media : @xmath154 on @xmath155 $ ] , with periodic boundary conditions .
we take @xmath156 and initial condition @xmath157 the problem is semi - discretized using a second order conservative scheme with a koren limiter as in @xcite with @xmath158 , and run to @xmath159 . for this problem
euler s method is tvd for @xmath160 .
we computed the numerical solution using all the @xmath136 ssp lnl runge kutta methods described in section [ sec : optimal ] and , as above found the largest @xmath152 for which tvd and positivity are preserved . in figure
[ fig : bltvd ] we plot these values ( blue for tvd , green for positivity ) compared to the time - step guaranteed by the theory , @xmath153 ( in red )
. the observed tvd and positivity time - step are typically significantly larger than the theoretical value .
as before , the positivity preserving time - step is larger than the tvd time - step .
using the optimization procedure described in @xcite , we find ssp - optimized explicit runge kutta methods that have nonlinear order of @xmath31 and @xmath32 and a higher @xmath161 order of convergence on linear autonomous problems .
the order barrier of @xmath162 for explicit ssp runge kutta methods indicates the critical importance of the nonlinear order on the ssp property .
nevertheless , we find that the _ size _ of the ssp coefficient is typically more constrained by the linear order conditions . as the number of stages increases , the ssp coefficient becomes primarily a function of the relationship between the number of stages and the linear order of the method , and not the nonlinear order .
this means that in many cases , we can obtain methods of nonlinear order @xmath32 and linear order @xmath161 that have the same ssp coefficient as methods with nonlinear order @xmath30 and linear order @xmath161 .
we verified the linear and nonlinear orders of convergence of the new methods on a variety of test cases .
we also showed the behavior of these new lnl time - stepping methods coupled with the weno method for both linear and nonlinear problems , which suggests that these lnl methods may be useful for problems that have regions that are dominated by linear , smooth solutions and other regions where the solution is discontinuous or dominated by nonlinear behavior .
finally , we studied the total variation diminishing and positivity preserving properties of these lnl methods on linear and nonlinear problems , and showed that for the linear problems , the theoretical ssp time - step is a very accurate predictor of the observed behavior , while serving only as a lower bound in the nonlinear case .
we conclude that where methods with high linear order are desirable , it is usually advantageous to pick those methods that also have higher nonlinear order ( @xmath163 ) . * acknowledgment .
* the authors wish to thank prof .
bram van leer for the motivation for studying this problem , and prof .
david ketcheson for many helpful discussions .
this publication is based on work supported by afosr grant fa-9550 - 12 - 1 - 0224 and kaust grant fic/2010/05 . | high order spatial discretizations with monotonicity properties are often desirable for the solution of hyperbolic pdes .
these methods can advantageously be coupled with high order strong stability preserving time discretizations .
the search for high order strong stability time - stepping methods with large allowable strong stability coefficient has been an active area of research over the last two decades .
this research has shown that explicit ssp runge kutta methods exist only up to fourth order . however ,
if we restrict ourselves to solving only linear autonomous problems , the order conditions simplify and this order barrier is lifted : explicit ssp runge kutta methods of any _ linear order _ exist .
these methods reduce to second order when applied to nonlinear problems . in the current work we aim to find explicit ssp runge kutta methods with large allowable time - step , that feature high linear order and simultaneously have the optimal fourth order nonlinear order .
these methods have strong stability coefficients that approach those of the linear methods as the number of stages and the linear order is increased .
this work shows that when a high linear order method is desired , it may be still be worthwhile to use methods with higher nonlinear order . |
the light propagation through a one - dimensional chain , as illustrated in fig .
[ fig1_si ] , can be described by the transfer matrix formalism @xcite . for an atom at position @xmath50
one can introduce the matrix @xmath51 that relates components of the backward- and forward - traveling electric field on the right side @xmath52 and @xmath53 to the ones on the left side @xmath54 and @xmath55 : @xmath56 the boundary conditions for the wave equation allows to obtain the following expression for the matrix ( [ m_a - matrix ] ) : @xmath57 where @xmath58 and @xmath20 are single - atom transmission and reflection coefficients respectively .
the transfer matrix @xmath59 for propagation between neighboring atoms is simply given by @xmath60 where @xmath10 is the distance between the two atoms .
the transfer matrix for the full ensemble is obtained as a product of matrices @xmath51 and @xmath59 as follows : @xmath61 the transmission and reflection coefficients of the atomic chain are finally given by : @xmath62 atoms are trapped in the vicinity of a nanoscale waveguide with radiative decay rates @xmath1 into the right and left - propagating modes and @xmath2 into all the other modes . ] the single - atom transmission and reflection coefficients @xmath58 and @xmath20 depend on the radiative decay rates .
each atom has two channels of decay @xmath63 that can be associated with the guided mode @xmath64 and the radiation modes @xmath65 where @xmath15 is the radiative decay in free - space .
the decay to the guided mode can occur in forward @xmath66 and backward @xmath67 directions . given the polarization of the guided light and positions of the atoms , @xmath66 and @xmath67 can be equal or asymmetric @xcite .
in the case of quasilinear polarization along @xmath68 , i.e. orthogonal to the atomic array , the decay rates in the forward and backward directions are equal @xmath69 . in this symmetric case
, the reflection and transmission coefficients can be expressed as @xcite : @xmath70 where @xmath21 is the detuning from the atomic resonance . in accordance with ( [ m_a - matrix2 ] )
the explicit expression of the transfer matrix @xmath51 can be written as follows : @xmath71 in the case of asymmetric scattering in forward and backward directions , a situation that occurs if the guided mode is quasilinearly polarized along the @xmath27-direction ( pointing towards the atomic chain ) , the full decay rate can be written as @xmath72 .
the explicit expressions for these coefficients can be found in accordance with @xcite as follows : @xmath73 these expressions finally provide the transfer matrix ( [ m_a - matrix2 ] ) for the asymmetric case as : @xmath74
the dipole trapping of the atoms can result in a resonance shift and an inhomogeneous broadening .
this effect is limited in the reported experiment : the shift has been measured equal to 3 mhz and the broadening to @xmath75 3 mhz . to study the effect of these parameters ,
we vary randomly the detuning for each atom of the chain , @xmath76 in accordance with the gaussian distribution : @xmath77 where @xmath78 is the standard deviation .
each atom has an individual transfer matrix @xmath79 and the total transfer matrix for the chain can be read as : @xmath80 in the main manuscript , simulations include the shift but not the broadening .
the measured broadening has indeed small effect , below what can be measured given the precision of our measurements .
figure [ fig2_si ] provides the simulations of the reflexion spectra for different broadenings @xmath78 . .
the trap detuning is @xmath81 nm , the number of atoms @xmath5 and the decay rate @xmath33 .
the spectra are averaged over 15 realizations . ] of each trapping site .
we consider two parallel arrays .
the trap detuning is @xmath81 nm , the number of atoms @xmath5 and the decay rate @xmath33 .
the spectra are averaged over 15 realizations . ]
the filling factor of the trapping sites can induce randomness in the distribution of the atoms across the lattice .
we consider here two parallel lines of atoms , as obtained with a nanofiber , with at most one atom per site . given a filling factor per site @xmath43 , at each trapping axial position
, there are either no atom at all , one atom on the upper chain , one atom on the lower chain or one atom in both upper and lower chains
. this situation can be simulated by a single random array with , for each site , either no atom ( probability @xmath82 ) , one atom ( probability @xmath83 ) or two atoms ( probability @xmath84 ) . as shown on fig .
[ fig3_si ] , a filling factor of 50% , which is the ideal case in the collisional blockade regime , leads here to the same spectrum than a full single array for the same number of atoms . when the filling factor decreases , the spectrum is narrower and the maximum reflection decreases . in the main manuscript , a filling factor @xmath34 was used to fit the experimental data . in the inset of figure 4 of the main manuscript
, we finally provided the maximal reflection as a function of the number of remaining trapped atoms .
for this simulation , we started from two parallel arrays with a filling factor @xmath34 and subsequently added random loss of the atoms .
if one denotes @xmath85 the probability for an atom to be lost , the new probability for having one atom is @xmath86 and for two atoms @xmath87 . for the near resonant trap for cs atoms outside a cylindrical waveguide of 200 nm radius
here , @xmath81 nm and the red - detuned trapping power is @xmath88 .
the power of each blue - detuned fields at 686.1 nm and 686.5 nm is 4 mw .
blue line corresponds to the axial potential for the ground state manifold @xmath89 , while the red lines correspond to the substates of the excited state @xmath90 . ]
following the calculation presented in @xcite , we calculated the trapping potential and trapping frequency in the axial direction .
figure [ fig4_si ] shows the axial dependence of the trapping potential @xmath91 for atoms trapped at a distance @xmath92 234 nm from the fiber surface .
the @xmath93 values for the @xmath94 ground state manifold and @xmath95 excited state manifold are plotted as blue and red lines respectively . for this calculation
, we use @xmath96 and a power of @xmath97 for the red detuned beams , and a power of @xmath98 for the blue detuned beams .
the potential depth at minimum is @xmath99 mk and the trap frequency in the axial direction is @xmath100 . in the case of @xmath101 and
a trapping power of @xmath102 , the potential depth at minimum is @xmath103 and the axial trap frequency is reduced to @xmath104 .
both axial frequencies for this near resonant trap are comparable with previous nanofiber - based traps @xcite . using these calculated frequencies , we compute the root mean square of the spatial spread in the axial direction in the harmonic approximation : @xmath105 where @xmath106 is boltzmann constant , @xmath107 is the temperature
, @xmath108 is the cesium atomic mass and @xmath109 is the axial frequency of the trap . in our case ,
the temperature of the atomic cloud is estimated to @xmath110 @xmath111 from a time - of - flight measurement .
thus , @xmath112 is equal to @xmath113 @xmath114 and @xmath115 @xmath114 for @xmath96 and @xmath101 respectively . for atoms cooled in the ground state ,
the axial spread would be @xmath116 @xmath114 and @xmath117 @xmath114 .
the reduction of bragg reflected intensity due the spread of the atomic position in the potential wells is usually estimated by a so - called debye - waller factor @xcite .
this factor is given by @xmath118 where @xmath119 is the guided mode wavenumber .
in our case , we find values of @xmath120 and @xmath121 for @xmath96 and @xmath101 respectively ( @xmath122 for atoms cooled to the motional ground state ) . in fig .
[ fig5_si ] , we provide the simulated spectra taking into account a gaussian spread @xmath123 for the atom position .
as it can be seen , this close - to - unity debye waller factor has a very limited effect in our configuration where bragg reflection is obtained out of resonance while it would be an important factor close to resonance @xcite .
we note that the reduction at resonance is larger than the debye - waller factor , which is valid in the single - scattering regime .
the debye - waller factor of 0.9 corresponds here to @xmath124 nm , as estimated from the temperature of the atoms and the simulated trap axial frequency .
the number of atoms is @xmath5 and the decay rate is @xmath33 .
the spectra are averaged over 15 realizations . ] | we report experimental observations of large bragg reflection from arrays of cold atoms trapped near a one - dimensional nanoscale waveguide . by using an optical lattice in the evanescent field surrounding a nanofiber with a period nearly commensurate with the resonant wavelength
, we observe a reflectance of up to 75 % for the guided mode .
each atom behaves as a partially - reflecting mirror and an ordered chain of about 2000 atoms is sufficient to realize an efficient bragg mirror
. measurements of the reflection spectra as a function of the lattice period and the probe polarization are reported .
the latter shows the effect of the chiral character of nanoscale waveguides on this reflection .
the ability to control photon transport in 1d waveguides coupled to spin systems would enable novel quantum network capabilities and the study of many - body effects emerging from long - range interactions .
in recent years , the coupling of one - dimensional bosonic waveguides and atoms , either real or artificial , has raised a large interest @xcite . beyond the remarkable ability to couple a single emitter to a guided mode @xcite
, the 1d reservoir would also enable the exploration and eventual engineering of photon - mediated long - range interactions between multiple qubits , a challenging prospect in free - space geometries .
this emerging field of waveguide quantum electrodynamics promises unique applications to quantum networks , quantum nonlinear optics and quantum simulation @xcite . in this context
, progress has been reported on various fronts . in the microwave regime
, the coupling of superconducting qubits to a one - dimensional transmission line provides a versatile platform to study such photon - mediated interactions @xcite . at optical frequencies ,
recent experimental advances include the development of 1d nanoscale dielectric waveguides coupled to cold atoms trapped in the vicinity @xcite . in these experiments , tight transverse confinement of the electric field achieves an effective mode area comparable to the atomic cross section and thereby a strong atom - photon interaction in a single - pass configuration @xcite .
coupling of atom arrays to 1d waveguides could lead to a variety of remarkable cooperative phenomena .
this coupling can strongly modify the photon transport properties @xcite , resulting for instance in sub- and superradiant decays as recently observed for two coupled atoms @xcite .
it can also lead to photonic bandgaps and provide atomic bragg mirrors , with envisioned applications to integrated cavity - qed @xcite .
this setting is as well at the basis of a recently proposed deterministic state engineering protocol @xcite and constitutes the building block of chiral spin networks in which the emission into the left- and right - propagating modes is asymmetric @xcite .
moreover , strong optomechanical couplings resulting from photon - mediated forces can give rise to rich spatial atomic configurations , including self - organization @xcite .
optical nanofibers offer a promising platform for exploring these effects .
their subwavelength diameter results in a large evanescent field that can be used for both trapping and interacting with atoms @xcite .
seminal works achieved the trapping of cold atoms near an optical nanofiber @xcite and recently enabled the demonstrations of all - fibered optical memories @xcite .
these works were realized with disordered atoms or incommensurate arrays , and relied on the optical depth of the medium .
an important capability that has not been demonstrated heretofore is the realization of cooperative effects emerging from the spatial order of the atoms , when the lattice is commensurate or close - to - commensurate with the resonant wavelength . in this letter ,
we report a large bragg reflection from atom arrays near a one - dimensional waveguide in a situation where each trapping site is occupied by at most one atom . due to the tight transverse confinement of the guided light , a few thousand atoms trapped in the evanescent field of a nanofiber are sufficient to strongly reflect the incoming light . by using a near - resonant dipole trap ,
the maximum bragg reflection is obtained for a slightly detuned probe .
we provide detailed characterizations of the reflection spectra and finally show the effect of the waveguide chirality arising from the complex polarization pattern of tightly focused light .
the long - range order of trapped atoms can indeed dramatically change the scattering properties . in incommensurate arrays
, the interference between forward- and backward - scattered light leads to absorption and a vanishing reflection .
in contrast , in commensurate arrays , this interference can result in strong reflection close to resonance @xcite .
this effect , well - known as bragg reflection , has been largely studied in crystals as well as in multilayer dielectric structures .
it has also been observed with ordered cold atoms in free - space , either with three - dimensional @xcite or one - dimensional optical lattices @xcite .
a reflectance as high as 80% was demonstrated in @xcite .
this observation required around @xmath0 atoms distributed over 7700 layers to reach the regime of multiple reflections .
in contrast , in this work we demonstrate that 2000 atoms are sufficient to achieve large reflectance in a waveguide - mediated scenario .
atoms are trapped near of a waveguide and exhibit radiative decay rates @xmath1 into the right and left - propagating modes , and @xmath2 into all the other modes .
( b ) electric field distribution in the transverse plane of a nanofiber for a guided probe with a quasilinear polarization ( indicated by the arrow ) .
( c ) theoretical reflection spectra for a probe quasilinearly polarized along the y - direction ( symmetric decay rates ) and ( d ) along the x - direction ( asymmetric decay rates ) .
the spectra are given for different distances between the atoms , with values close to the commensurate case .
@xmath3 stands for the trap detuning to resonance , with @xmath4 .
( @xmath5 , @xmath6 , @xmath7 , @xmath8 ) .
theoretical values of the couplings are taken from @xcite . ] to introduce the scattering properties and associated photon transport , we first consider a typical configuration as depicted in fig .
[ fig1](a ) .
@xmath9 atoms are trapped in the vicinity of a waveguide , with a lattice constant @xmath10 close to @xmath11 , where @xmath12 corresponds to wavelength of the atomic transition . due to the complex polarization structure of tightly focused light @xcite , which includes a significant longitudinal component ,
the scattering in the guided mode can be asymmetric @xcite .
each atom exhibits a radiative decay rate @xmath13 and @xmath14 into the right- and left - propagating mode respectively , and @xmath2 into all the other modes .
@xmath15 is the radiative decay rate in free space . for a guided probe field
quasilinearly polarized along the y - direction , the two decay rates are equal , @xmath16 . for an orientation along the x - direction , i.e. pointing towards the atoms , the couplings to the waveguide become strongly asymmetric .
it has also been shown that when the ground state zeeman levels are equally populated , the two x- and y - polarization are not coupled to each other by the linear coherent scattering @xcite .
for nanofiber - trapped atoms , typically around 200 nm from the surface , the ratio @xmath17 amounts to a few @xmath18 . in the case of asymmetric coupling , the forward decay rate
is increased by sixfold while the backward decay rate is suppressed by about one order of magnitude .
ground state and the probe addresses the transition @xmath19 .
around 2000 atoms in total are trapped along two parallel chains .
reflection and transmission spectra are measured with avalanche photodiodes after polarization and frequency filterings .
dm stands for dichroic mirror , fbs for fiber beamsplitter and vbg for volume bragg grating . ]
the single - atom amplitude reflection @xmath20 can be calculated from the couplings to the waveguide and the probe detuning @xmath21 to resonance . in the symmetric decay case for instance ,
the reflection is given by @xmath22 @xcite . at resonance , one atom thus reflects @xmath23 in intensity .
the transfer matrix formalism @xcite enables then to calculate the single - photon propagation through the ensemble ( see appendix ) .
figure [ fig1](c ) and [ fig1](d ) provide theoretical reflection spectra for different small detunings @xmath3 of the trap wavelength to atomic resonance and for the two orthogonal polarizations . for atoms separated exactly by @xmath11 ,
the reflection spectrum is a broadened lorentzian in the symmetric coupling case while the reflectance is strongly suppressed in the chiral one .
indeed , the amount of chirality and number of atoms result in a finite bandwidth around resonance where reflection is suppressed , as detailed in @xcite . for close - to - commensurate traps ,
as studied here , the bragg condition is fulfilled out of resonance .
this leads to a maximum reflectance shifted to the blue @xcite but also results in an increased reflectance for the chiral case .
large reflectance values can then be obtained for both polarizations as the single - atom reflection coefficients are similar in our configuration .
our experimental setup is illustrated in fig .
[ fig2 ] .
arrays of trapped atoms are prepared in the evanescent field of a 400-nm diameter nanofiber suspended inside an ultrahigh vacuum chamber .
the nanofiber is produced from a single - mode fiber ( oz optics smf-780 - 5/125 ) by the standard heating - pulling technique .
the polarization of the different guided beams can be aligned by measuring the polarization properties of the light scattered from imperfections at the fiber surface @xcite .
our two - color dipole trap is based on the combination of laser beams all guided by the nanofiber . by using the appropriate combination of attractive red - detuned light and repulsive blue - detuned light , the atoms are trapped in the potential minima located at a sub - wavelength distance from the waveguide @xcite .
the longitudinal periodic structure is given by the standing wave formed by the two counter - propagating red - detuned beams , with @xmath3 the detuning from resonance .
we perform our experiment at two different values of @xmath3 , 0.12 nm and 0.2 nm , with a total power equal to @xmath24w and @xmath25w respectively .
a pair of counter - propagating beams ( @xmath26 mw ) at 686.1 nm and 686.5 nm are chosen as blue - detuned beams and contribute to the trap in a compensated manner @xcite .
the polarization of the beams is oriented along the transverse @xmath27-direction .
the resulting intensity pattern leads to trapping sites aligned along two lines parallel to the fiber .
because of the loading dynamics , each site hosts one atom at most @xcite . simulations of the trapping potential ( see appendix ) provide for @xmath28 nm ( 0.2 nm ) a trap depth at minimum equal to -0.15 mk ( -0.1 mk ) and an axial trap frequency @xmath29 khz ( 215 khz ) . for @xmath30 nm , we measure a shift of the transition equal to 3 mhz relative to free - space and an inhomogeneous broadening limited to @xmath31 . 0.12
nm and @xmath320.2 nm . ( c ) and ( d ) simulated spectra for @xmath5 atoms , @xmath33 and a filling factor of the lattice sites @xmath34 . the coupling value and filling factor
have been adjusted to fit the measured spectra reported in ( a ) . ] to load the dipole trap , which is continuously on , we overlap a cigar - shaped magneto - optical trap ( mot ) along the fiber axis .
the atomic cloud is then further cooled and the lattice is loaded during an optical molasses phase : the mot coils are switched off , and during a 10 ms interval , the mot cooling - beam detuning is increased from @xmath35 to @xmath36 , while the powers of both the cooling and repumping beams are decreased to zero .
the atoms are then optically pumped into the @xmath37 hyperfine ground state for @xmath38 . by a saturation measurement
, we estimate the number of trapped atoms to be @xmath39 .
residual magnetic fields are canceled with bias coils and characterized via zeeman sublevel microwave spectroscopy .
the fields are compensated down to the 20 mg level . to measure the reflection and transmission spectra , a probe pulse first passes through a fibered beamsplitter that we use to separate the reflected pulse from the input probe .
the probe is then combined with the pair of trapping beams , via a volume bragg grating ( vbg ) , and sent into the nanofiber .
the weak probe pulse arriving on the atoms has a mean photon number of [email protected] .
the transmitted pulse is filtered by another vbg and an additional filtering system , and finally detected by an avalanche photodiode .
the filtering system combines a polarization beamsplitter and a commercial lens - based cavity @xcite .
the cavity transmission is around 75% , with a bandwidth of 80 mhz and a rejection around 40 db for the trapping beams .
these cascaded stages are required to filter out efficiently the dipole beams and reach the single - photon level .
the reflected pulse is directed to a similar filtering and detection stage .
the reflectance values are obtained by comparing the reflected pulse when atoms are trapped to the transmitted pulse without atoms , and by correcting for the different losses in the system .
these losses , which are detuning - dependent due to the filtering cavities , are calibrated before and after each measurement .
the transmittance values are obtained by comparing the transmitted pulse with and without trapped atoms .
error bars in the data include the calibration uncertainties . and subsequent random loss ( @xmath320.2 nm , @xmath33 ) . ]
we first characterize the reflection and transmission spectra for a probe quasilinearly polarized along the y - direction . in this configuration ,
the radiative decay rates are expected to be symmetric .
the measurements are given in fig .
[ fig3 ] for the two different detunings of the dipole trap . for @xmath320.12 nm , as shown in fig .
[ fig3](a ) , we observe a broad reflection spectrum , with a maximum reflectance of ( [email protected] ) for a detuning of 25 mhz . for @xmath320.2 nm ,
the reflection peak is narrower and the maximum reflectance ( [email protected] ) is shifted to a lower frequency ( fig . [ fig3](b ) ) .
the spectra resulting from the two trapping distances are well explained by our simple theoretical model , as presented in fig . [
fig3](c ) and fig .
[ fig3](d ) . in this model
, we take into account the disorder in the distribution of the atoms along the two parallel arrays due to a limited filling factor @xmath43 of the individual trapping sites ( see appendix ) . the coupling to the waveguide and the filling factor have been adjusted to provide the best fit to the spectra of fig .
[ fig3](a ) .
a value @xmath34 was obtained .
larger values lead to a pronounced dip on the reflection spectrum while smaller values result in a strongly reduced reflectance and a narrowing of the spectra width .
the simulations take also into account the shift induced by the dipole trapping .
inhomogeneous broadening is not included as the limited broadening @xmath44 leads to a negligible modification of the spectra ( see appendix ) .
apart from a finite filling factor , a second contribution to disorder can arise from the thermal distribution of the atoms in the potential wells .
the good agreement between the achieved reflectances and the predicted ones supports a tight axial localization , as expected with our microscopic traps . with a temperature @xmath45 estimated by a time - of - flight measurement and the predicted axial trap frequency of @xmath29 khz
, the spread is given in the harmonic approximation by @xmath46 nm , i.e. @xmath47 .
this value leads to a debye - waller factor @xmath48 close to 0.9 @xcite .
as shown in the appendix , this disorder has a very limited effect in our configuration where bragg reflection is observed out of resonance @xcite
. 0.2 nm .
( b ) theoretical simulations ( @xmath5 , @xmath33 , @xmath7 , @xmath8 , @xmath34 ) . ]
the measurements described above were performed 1 ms after loading the trap .
we now investigate the reflection for longer trapping time and compare its decay with the trap lifetime .
figure [ fig4 ] shows the maximal reflectance and optical depth of the medium as a function of the hold time . using exponential fits ,
we obtain a decay time of 2.7 ms and 1.9 ms respectively . from the measured optical depth
, we then estimate the number of remaining trapped atoms at each time .
the inset in fig .
[ fig4 ] finally provides the maximal reflectance as a function of this inferred atom number .
the green line takes into account a random loss of the atoms from the initial arrays with a total atom number of 2000 and a filling factor @xmath34 ( see appendix ) .
this model agrees well with the data .
finally , we investigate the effect of the chiral character of the waveguide .
the previous measurements were realized with the y - polarization , which leads to symmetric decay rates for the backward and forward scattering .
we now consider the asymmetric case .
this situation can be obtained with a probe quasilinearly polarized along the x - direction .
the measured reflection spectra for both probe polarizations are compared in fig . [ fig5](a ) .
the trap detuning is fixed at @xmath320.2 nm . as it can be seen , the spectrum is significantly shifted and broadened in the asymmetric case .
these features are compelling signatures of the chiral character of the waveguide on the reflection , as confirmed by the associated simulations in fig .
[ fig5](b ) .
the maximal observed reflectance of ( [email protected] ) is obtained in this asymmetric case , at a probe detuning of 25 mhz . in conclusion
, we have realized an efficient bragg atomic mirror based on a nanoscale one - dimensional waveguide coupled to about 2000 atoms .
the effect of the chiral character of the waveguide on the reflection features has also been observed . beyond their fundamental significance ,
these observations demonstrate key ingredients for the exploration of a variety of emerging and potentially rich protocols based on 1d reservoirs coupled to atoms . to enable an exact commensurate array of trapped atoms , and therefore an enhanced reflection closer to resonance , a bichromatic optical superlattice , whose trap periodicity is given by the tunable beat frequency of the trapping lasers @xcite ,
can be developed .
such a trapping scheme , which has been demonstrated in free space , would need to be adapted to the evanescent field configuration around a nanofiber .
this superlattice could also include a double primitive cell enabling a richer photonics spectrum @xcite .
an additional classical driving field in a three - level atomic configuration would finally provide a dynamical control of the transport properties @xcite .
+ _ note during the preparation of this manuscript , a related experiment by h.l . s@xmath49rensen and
coworkers was presented in @xcite . _
+ this work is supported by the european research council ( starting grant hybridnet ) , the emergence program from ville de paris and the ifraf dim nanok from rgion ile - de - france .
the authors acknowledge interesting discussions within the capes - cofecub project ph 740 - 12 and with h.j .
kimble when j.l .
was in a visiting professor position at caltech in 2015 .
the authors also thank a. nicolas , d. maxein and o. morin for their contributions in the early stage of the experiment , and a.c .
mcclung for constructive comments on the manuscript .
a.s . and d.v.k .
acknowledge rfbr ( grant no .
15 - 02 - 01060 ) .
n.v.c .
is supported by the eu ( marie curie fellowship ) .
j.l . is a member of the institut universitaire de france .
10 d.e .
chang , a.s .
s@xmath49rensen , e.a .
demler , and m.d .
lukin , a single - photon transistor using nanoscale surface plasmons , nat .
physics * 3 * , 807 ( 2007 ) .
a.l .
van loo , a. fedorov , k. lalumire , b.c .
sanders , a. blais , and a. wallraff , photon - mediated interactions between distant artificial atoms , science * 342 * , 1494 ( 2013 ) .
p. lodahl , s. mahmoodian , and s. stobbe , interfacing single photons and single quantum dots with photonic nanostructures , rev .
mod .
phys .
* 87 * , 347 ( 2015 ) .
h.j .
kimble , the quantum internet , nature * 453 * , 1023 ( 2008 ) .
d.e .
chang , v. vuleti , and m.d .
lukin , quantum nonlinear optics - photon by photon , nat .
photonics * 8 * , 685 ( 2014 ) .
a. gonzlez - tuleda , c .-
. hung , d.e .
chang , j.i .
cirac , and h.j .
kimble , subwavelength vacuum lattices and atom - atom interactions in two - dimensional photonic crystals , nat . photonics * 9 * , 320 ( 2015 ) .
e. vetsch , d. reitz , g. sagu , r. schmidt , s.t .
dawkins , and a. rauschenbeutel , optical interface created by laser - cooled atoms trapped in the evanescent field surrounding an optical nanofiber , phys .
rev .
lett . * 104 * , 203603 ( 2010 ) a. goban _
et al . _ ,
demonstration of a state - insensitive , compensated nanofiber trap , phys .
rev .
lett . *
109 * , 033603 ( 2012 ) .
a. goban _
et al . _ , atom - light interactions in photonic crystals , nat
. commun . * 5 * , 3808 ( 2014 ) .
p. domokos , p. horak , and h. ritsch , quantum description of light - pulse scattering on a single atom in waveguides , phys .
rev .
a * 65 * , 033832 ( 2002 ) .
j.t .
shen and s. fan , coherent photon transport from spontaneous emission in one - dimensional waveguides , opt .
lett . * 30 * , 2001 ( 2005 ) .
d. dzsotjan , a.s .
sorensen , and m. fleischhauer , quantum emitters coupled to surface plasmons of a nanowire : a green s function approach , phys .
rev .
b * 82 * , 075427 ( 2010 )
. f. le kien and a. rauschenbeutel , propagation of nanofiber - guided light through an array of atoms , phys .
rev .
a * 90 * , 063816 ( 2014 ) .
a. goban _
et al . _ , superradiance for atoms trapped along a photonic crystal waveguide , phys .
rev .
lett . * 115 * , 063601 ( 2015 ) .
d.e .
chang , l. jiang , a.v .
gorshkov , and h.j .
kimble , cavity qed with atomic mirrors , new j. phys . * 14 * , 063003 ( 2012 ) .
y. chang , z.r .
gong , and c.p .
sun , multiatomic mirror for perfect reflection of single - photons in a wide band of frequency , phys .
rev .
a * 83 * , 013825 ( 2011 ) .
p .- o .
guimond , a. roulet , h. nguyen le , and v. scarani , rabi oscillation in a quantum cavity : markovian and non - markovian dynamics , phys . rev . a * 93 * , 023808 ( 2016 ) .
a. gonzlez - tudela , v. paulisch , d.e .
chang , h.j .
kimble , and j.i .
cirac , deterministic generation of arbitrary photonic states assisted by dissipation , phys .
rev .
lett . * 115 * , 163603 ( 2015 ) h. pichler , t. ramos , a.j .
daley , and p. zoller , quantum optics of chiral spin networks , phys .
rev . a * 91 * , 042116 ( 2015 ) .
j.k .
asbth , h. ritsch , and p. domokos , optomechanical coupling in one - dimensional optical lattice , phys .
rev .
a * 77 * , 063424 ( 2008 ) .
d.e .
chang , j.i .
cirac , and h.j .
kimble , self - organization of atoms along a nanophotonic waveguide , phys .
rev .
lett . * 110 * , 113606 ( 2013 ) .
t. nieddu , v. gokhroo , and s. nic chormaic , optical nanofibres and neutral atoms , j. opt .
* 18 * , 053001 ( 2016 ) .
b. gouraud , d. maxein , a. nicolas , o. morin , and j. laurat , demonstration of a memory for tightly guided light in an optical nanofiber , phys .
rev .
lett . * 114 * , 180503 ( 2015 ) .
c. sayrin , c. clausen , b. albrecht , p. schneeweiss , and a. rauschenbeutel , storage of fiber - guided light in a nanofiber - trapped ensemble of cold atoms , optica * 2 * , 353 ( 2015 ) .
i.h .
deutsch , r.j.c .
spreeuw , s.l .
rolston , and w.d .
phillips , photonic band gaps in optical lattices , phys .
rev .
a * 52 * , 1394 ( 1995 ) .
g. birkl , m. gatzke , i.h .
deutsch , s.l .
roston , and w.d .
phillips , bragg scattering from atoms in optical lattices , phys .
rev .
lett . * 75 * , 2823 ( 1995 ) .
m. weidemller , a. hemmerich , a. grlitz , t. esslinger , and t. w. hnsch , bragg diffraction in an atomic lattice bound by light , phys .
rev .
lett . * 75 * , 4583 ( 1995 ) .
s. slama , c. von cube , b. deh , a. ludewig , c. zimmermann , and p.w .
courteille , phase - sensitive detection of bragg scattering at 1d optical lattices , phys .
rev .
lett . * 94 * , 193901 ( 2005 ) .
a. schilke , c. zimmermann , p.w .
courteille , and w. guerin , photonic band gaps ine one - dimensionally ordered cold atomic vapors , phys .
rev .
lett . * 106 * , 223903 ( 2011 ) .
f. le kien , j.q .
liang , k. hakuta , and v.i .
balykin , field intensity distributions and polarization orientations in a vacuum - clad subwavelength - diameter optical fiber , opt . commun . * 242 * , 445 ( 2004 ) .
f. le kien and a. rauschenbeutel , anisotropy in scattering of light from an atom into the guided modes of a nanofiber , phys .
rev .
a * 90 * , 023805 ( 2014 ) .
f. le kien and a. rauschenbeutel , propagation of nanofiber - guided light through an array of atoms , phys .
rev .
a * 90 * , 063816 ( 2014 ) .
e. vetsch , s.t .
dawkins , r. mitsch , d. reitz , p. schneeweiss , and a. rauschenbeutel , nanofiber - based optical trapping of cold neutral atoms , ieee j. quantum electron . *
18 * , 1763 ( 2012 ) .
f. le kien , v.i .
balykin , and k. hakuta , atom trap and waveguide using a two - color evanescent field around a subwavelength - diameter optical fiber , phys .
rev .
a * 70 * , 063403 ( 2004 ) .
n. schlosser , g. reymond , and p. grangier , collisional blockade in microscopic optical dipole traps , phys .
rev .
lett . *
89 * , 023005 ( 2002 ) .
p. palittapongarnpim , a. macrae , and a.i .
lvosvky , a monolithic filter cavity for experiments in quantum optics , rev .
sci .
instrum . * 83 * , 066101 ( 2012 ) .
n.w . ashcroft and n.d .
mermin , _ solid state physics _
( saunders , philadelphia , 1976 ) .
s. slama , c. von cube , m. kohler , c. zimmermann , and ph . w. courteille , multiple reflections and diffuse scattering in bragg scattering at optical lattices , phys .
rev .
a * 73 * , 023424 ( 2006 ) .
a. grlitz , t. kinoshita , t.w .
hnsch , and a. hemmerich , realization of bichromatic optical superlattices , phys .
rev .
a * 64 * , 011401(r ) ( 2001 ) .
a. camposeo _
et al . _ , one - dimensional bichromatic standing - wave cooling of cesium atoms , j. opt .
b : quantum semiclass . opt . * 5 * , s29 ( 2003 ) .
s. rist , p. vignolo , and g. morigi , photonic spectrum of bichromatic optical lattices , phys .
rev .
a * 79 * , 053822 ( 2009 ) .
d. witthaut and a.s .
sorensen , photon scattering by a three - level emitter in a one - dimensional waveguide , new j. phys .
* 12 * , 043052 ( 2010 ) .
h.l .
s@xmath49rensen _ et al . _ , coherent backscattering of light off one - dimensional atomic strings , arxiv:1601.04869 . |
material bodies moving within a surrounding fluid are typically subject to a force associated with the fluid - body interaction .
for instance , in the case of a photon gas this force acts like a friction contrasting the particle motion , as first considered by poynting @xcite and robertson @xcite .
the force term entering the equations of motion was given there by the 4-momentum density of radiation observed in the particle s rest frame with a multiplicative constant factor expressing the strength of the interaction itself .
in the present paper we follow the same approach to investigate the general features of particle motion in a schwarzschild spacetime region filled with a fluid with equation of state @xmath6 .
the parameter @xmath1 characterizes different kind of fluids , from massive free scalar fields ( @xmath2 ) , to non - relativistic dust ( @xmath7 ) , all the way down deep into the so - called phantom " energy ( @xmath3 ) .
as it is well known from mathematical cosmology @xcite , the parameter @xmath1 has a profound influence on the time evolution of the universe .
for instance , in the friedmann - robertson - walker ( frw ) models it determines the power law exponent of the scale factor as a function of time .
according to our analysis , which relies on an explicit form of the fluid - particle interaction , the parameter @xmath1 plays a major role also in discriminating among different types ( escape versus capture ) of particle motion , for a given black hole spacetime metric . in particular , it is shown that the condition @xmath8 ( the case of a cosmological constant ) identifies with geodesic motion in the schwarzschild metric , and sets the borderline between capture ( positive sign ) and escape ( negative sign ) , respectively .
this sharp borderline could in principle lead to measurable effects , hence new hints on the nature of the cosmological matter , based on the observation of the motion of test particles embedded in the cosmological fluid surrounding a schwarzschild black hole .
we begin by writing the schwarzschild metric in the standard form [ sch ] ds^2=-n^2dt^2+n^-2dr^2+r^2(d^2+^2d^2 ) , with the lapse function `` @xmath9 given by n= , where @xmath10 is the mass of the black hole .
the metric ( [ sch ] ) is static , i.e. , @xmath11 is a hypersurface - forming timelike killing vector .
observers at rest with respect to the coordinates ( or static observers '' ) have their @xmath12-velocity vector aligned along the killing direction itself , namely u = n^-1_t . an orthonormal frame @xmath13 ( @xmath14 ) adapted to them
is given by @xmath15 for later use , let us introduce co - rotating @xmath16 and counter - rotating @xmath17 circular timelike geodesic orbits on the equatorial plane @xmath18 , characterized by the keplerian " @xmath12-velocity u_k=_k ( u_k e_)=_k ( _ t _ k _ ) , where @xmath19 and @xmath20 , with [ kepler ] _
k=,_k=. these orbits only exist for @xmath21 .
next , let us consider a ( test ) perfect fluid , superposed to the schwarzschild spacetime , described by the following stress - energy tensor @xcite [ tensorefotoni2 ] t^(r)= _ 0(r ) , i.e. , with equation of state @xmath22 , where @xmath1 is a constant characterizing the type of cosmic fluid , as detailed below .
the form of @xmath23 is obtained by integrating the equations @xmath24 ; in this case , due to spherical symmetry of the background , the solution of these equations is easily obtained , i.e. , _ 0(r)= _
0^ n^- , where @xmath25 is a constant representing the value of the energy density at infinity . noted values of @xmath1 in the literature ( see , e.g. , refs .
@xcite ) are as follows : * @xmath2 : massive free scalar field ; * @xmath26 : radiation ; * @xmath7 : dust ( non - relativistic matter ) ; * @xmath27 : quintessence ; * @xmath28 : cosmological constant ; * @xmath3 : phantom energy .
as is well known , each of these values corresponds to a different growth of the cosmological scale factor in the simple frw model , that is @xmath29 .
this shows that the capture scenario , i.e. , @xmath4 , corresponds to initial time singularities ( big - bang ) , whereas the escape scenario , i.e. , @xmath30 , associates with finite - time future singularities ( big - rip ) ( see ref . @xcite and references therein ) .
based on the energy momentum tensor ( [ tensorefotoni2 ] ) , we model the force acting on a massive particle with @xmath12-velocity @xmath31 , in the form of a linear drag term , proportional to the particle velocity [ frad_gen ] f_(scat)(u)_=- p(u ) _
t^_u^ , where [ udef ] u=(u+^a e_a),=(1-_ab^a^b)^-1/2 , @xmath32 projects orthogonally to @xmath33 and @xmath34 is a numerical coefficient .
this expression for the scattering force has been already used in the literature to study radiation scattering @xcite . in that case , @xmath35 models the absorption and consequent re - emission of radiation by the particle . here , we extend its form in such a way as to cover more general situations . the entire discussion about deviations from geodesic motion given below relies on this choice of the force .
although by no means unique , the definition ( [ frad_gen ] ) has the merit of mathematical simplicity , as combined with a clear physical basis . the motion of a massive particle is thus governed by the equations [ eq_mot_gen ] m a(u)=f_(scat)(u ) , where @xmath36 denotes the particle @xmath12-acceleration with frame components @xmath37\,,\nonumber \\ a(u)^{\hat \phi}&= & \frac{d ( \gamma \nu^{\hat \phi})}{d\tau}+\frac{\gamma^2}{r\sin\theta}[n\sin \theta
\nu^{\hat r}+\cos \theta \nu^{\hat \theta}]\nu^{\hat \phi}\,.\nonumber\end{aligned}\ ] ] the quantities @xmath38 and @xmath39 have been introduced in eq .
( [ kepler ] ) . from the orthogonality condition @xmath40 , the following relation holds [ formula_acc ] a(u)^t=_r a(u)^r+
a(u)^. in the sequel , we shall focus on the special case of a particle confined to the equatorial plane ( @xmath18 ) , namely with @xmath41 . in this case ,
@xmath42 and the remaining components simplify to @xmath43 the general equatorial motion is thus fully described by the following equations @xmath44\,,\nonumber \\
\frac{d \nu^{\hat \phi}}{d\tau } & = & \frac{1}{\gamma}\left(\frac{a(u)_{\hat \phi}}{\gamma_\phi^2}- a(u)_{\hat r}\nu^{\hat r}\nu^{\hat \phi}\right ) -\frac{\gamma n}{\gamma_k^2 r}\nu^{\hat r}\nu^{\hat \phi}\,,\end{aligned}\ ] ] where @xmath45 and @xmath46 , once the acceleration components are replaced by the force components according to @xmath47 the evolution equations for the radial and azimuthal coordinates follow directly from the definition of @xmath33 ( see eq .
( [ udef ] ) ) , i.e. , @xmath48 we note that when @xmath28 , the force term @xmath49 vanishes identically , implying geodesic motion . therefore ,
if the superposed fluid corresponds to a cosmological constant term in the einstein s equations , the interaction between particles and gas can not be modeled by eq .
( [ frad_gen ] ) .
this is exactly the case of the de sitter spacetime .
in fact , for @xmath50 , @xmath51 being the cosmological constant , we have t_u^=-u _ , which vanishes upon projecting orthogonally to @xmath33 . in this case
, a more elaborate force term should be used .
let us further note that at @xmath28 there is a sign change in the force , yielding an outward push as @xmath3 .
this sign change holds the key to the transition from capture to escape scenarios . moving to a polar representation of the velocity ,
i.e. , @xmath52 and @xmath53 , the above equations take the following form @xmath54 where the interaction parameter a= _ 0^ , has been introduced so that @xmath55 identifies the equatorial geodesic case .
note that the presence of @xmath56 affects only the first equation of ( [ poleqs ] ) .
apart from very special situations , the analysis of these equations can only be performed numerically . figs .
[ fig.1 ] and [ fig.2 ] show typical orbits for the same choice of initial conditions , but different values of the interaction strength parameter @xmath56 , at different values of the parameter @xmath1 .
simple inspection shows that values of @xmath57 ( @xmath3 ) imply capture ( escape ) from the hole , as already stated . in fig .
[ fig.1 ] , escaping orbits are characterized by a monotonically increasing speed , up to the speed of light .
this does not occur in fig .
[ fig.2 ] , due to the very small value of the coupling parameter @xmath56 . in both cases , however , the deviations from geodesic motion are clearly appreciated . .
the solid lines starting at @xmath58 ( with @xmath59 , @xmath60 , @xmath61 ) represent the particle motion associated with the scattering of the fluid , with parameters @xmath62 and @xmath63 $ ] .
the short - dashed curve corresponds to @xmath28 , hence to geodesic motion . with this choice of @xmath56 ,
all particles scattered by a fluid with @xmath64 fall into the hole .
negative values of @xmath65 , instead , allow particles to escape capture , and imply a sudden increase of the speed , which reaches the speed of light , i.e. , @xmath66 , in a finite time ( in order to preserve the causality condition for the particle world line , the numerical integration has been stopped at @xmath66 ) . ] , but with @xmath67 and @xmath68 $ ] . with this choice of @xmath56 ( denoting a weaker interaction with respect to the previous case ) , escaping particles ( @xmath3 ) never reach the speed of light . ] in principle , it is interesting to consider multi - component fluid scenarios , in which the matter around the hole corresponds to regions of different fluids ( with different @xmath1 ) , superposed to the schwarzschild background .
this is only an approximation , since the various ( ideal ) fluids in the different regions would rapidly mix and yield a mixture with intermediate values of @xmath1 .
however , if only as a hypothetical speculation , it is of interest to explore the way the equation for @xmath69 ( and only that ) slightly modifies to account for the presence of multi - component matter , say with @xmath70 components , associated with the parameters @xmath71 and @xmath72 . the new evolution equation for @xmath69 is thus given by @xmath73 explicitly , for @xmath74 we have , for example , @xmath75 this is equivalent to a heterogeneous mixture with a radially - dependent @xmath1 parameter , namely w(r)= , where we have set @xmath76 .
one may then consider the presence of both ordinary or exotic matter , with special choices of @xmath77 and @xmath78 .
numerical integration of the orbits in this case , confirms the previous analysis , namely that the dominating value of @xmath1 determines the final fate , capture or escape , of the particle trajectory .
a more in depth analysis of the multi - component fluid scenario will make the object of future investigations .
the motion of massive test particles interacting with a perfect fluid superposed to a schwarzschild black hole is investigated .
the equation of state describing the fluid is taken as @xmath79 , which corresponds to different physical scenarios for different walues of @xmath1 , from ordinary matter to exotic matter .
the interaction with the fluid distribution is modeled by a force term entering the equations of motion given by the 4-momentum density observed in the particle s rest frame with a multiplicative constant factor expressing the strength of the interaction itself .
deviations from geodesic motion are analyzed for different values of the scattering parameter as well as of the parameter @xmath1 .
the latter turns out to be crucial in distinguishing among capture ( @xmath64 ) or escape ( @xmath80 ) from the black hole .
the measurement of such effects may offer a new means of gleaning information on the nature of the cosmic matter surrounding astrophysical objects .
d. bini , a. geralico , r.t .
jantzen , o. semerk , l. stella , the general relativistic poynting - robertson effect : ii . a photon flux with nonzero angular momentum
quantum gravity * 28 * , 035008 ( 2011 ) | the motion of a massive test particle in a schwarzschild spacetime surrounded by a perfect fluid with equation of state @xmath0 is investigated .
deviations from geodesic motion are analyzed as a function of the parameter @xmath1 , ranging from @xmath2 which corresponds to the case of massive free scalar fields , down into the so - called phantom " energy , with @xmath3 .
it is found that the interaction with the fluid distribution leads to capture ( escape ) of the particle trajectory in the case @xmath4 ( @xmath5 ) , respectively .
based on this result , it is argued that inspection of the trajectories of test particles in the vicinity of a schwarzschild black hole may offer a new means of gaining insights into the nature of cosmic matter . |
as policymakers , academics and the public at large grow increasingly concerned about the cost of an aging society , we believe it is worthwhile to go back in time and examine the capital market instruments used to finance retirement in a period before social insurance , defined benefit ( db ) pensions , and annuity companies .
indeed , in the latter part of the 17th century and for almost two centuries afterwards , one of the most popular `` retirement income '' investments in the developed world was not a stock , bond or a mutual fund ( although they were available ) .
in fact , the method used by many individuals to generate income in the senior years of the lifecycle was a so - called tontine scheme sponsored by government .
part annuity , part lottery and part hedge fund , the tontine which recently celebrated its 360th birthday offered a lifetime of income that increased as other members of the tontine pool died off and their money was distributed to survivors .
the classical tontine investment pool is quite distinct from its public image as a lottery for centenarians in which the longest survivor wins all the money in a pool .
in fact , the tontine is both more subtle and more elegant . for those readers who are not familiar with the historical tontine , here
is a simple example .
imagine a group of 1,000 soon - to - be retirees who band together and pool $ 1,000 each to purchase a million - dollar u.s .
treasury bond ( with a very long , or even perpetual maturity date ) paying 3% coupons .
the bond generates $ 30,000 in interest yearly , which is split amongst the 1,000 participants in the pool , for a 30,000 / 1,000 = guaranteed $ 30 dividend per member .
a custodian holds the big bond taking no risk and requiring no capital and charges a trivial fee to administer the annual dividends . in a tontine scheme
the members agree that if and when they die , their guaranteed $ 30 dividend is split amongst those who still happen to be alive .
for example , if one decade later only 800 original investors are alive , the $ 30,000 coupon is divided into 800 , for a $ 37.50 dividend each . of this , $ 30 is the guaranteed dividend and $ 7.50 is _ other people s money_. then , if two decades later only 100 survive , the annual cash flow to survivors is $ 300 , which is a $ 30 guaranteed dividend plus $ 270 .
when only 30 remain , they each receive $ 1,000 in dividends .
the extra payments above and beyond the guaranteed $ 30 dividend
are the mortality credits .
in fact , under this scheme payments are expected to increase at the rate of mortality , which is a type of super - inflation hedge .
the tontine , of course , differs from a conventional life annuity .
although both offer income for life and pool longevity risk , the mechanics and therefore the cost to the investor are quite different .
the annuity ( or its issuing insurance company ) promises predictable guaranteed lifetime payments , but this comes at a cost and regulator - imposed capital requirements that inevitably makes its way to the annuitant . indeed , the actuaries make very conservative assumptions regarding how long annuitants are likely to live . in future ( especially under the regulations of solvency
ii ) annuity - issuers might have to hold capital and reserves against aggregate longevity risk .
in contrast , the tontine custodian divides the variable @xmath3 ( bond coupons received ) by the variable @xmath4 ( participants alive ) and sends out cheques .
it is cleaner to administer , less capital - intensive and in its traditional form results in an increasing payment stream over time ( assuming , of course , that you are alive ) .
underlying this paper is the argument that properly - designed tontines should be on the menu of products available to individuals as they transition into their retirement years .
lorenzo de tonti was a colorful italian banker who in the 1650s promoted the scheme which now shares his name .
he described the tontine as a mixture of lottery and insurance equally driven by fear of old age , and economic greed .
tontines were first introduced in holland , then very successfully in france and a few decades later , in 1693 , england s king william iii presided over the first government - issued tontine . in this first government - issued tontine , in exchange for a minimum payment of 100 , participants were given the option of ( a ) investing in a tontine scheme paying guaranteed 10 dividends until the year 1700 and then 7 thereafter , or ( b ) a life annuity paying 14 for life , but with no survivorship benefits . in our view , the choice between two possible ways of financing retirement income is something that could conceivably be offered again in the future .
perhaps the choice between a tontine ( in which longevity risk is pooled ) and a life annuity ( in which payments are guaranteed ) is a paradigm for many of the choices retirees , and society , now face .
back to the 1693 tontine of king william , over a thousand englishmen ( but very few women ) decided to invest in the tontine .
the sums involved were nt trivial
. the 100 entry fee would be worth close to 100,000 today .
so , this was no impulse lottery purchase .
rather , some investors picked the tontine because they wanted the skewness that is , the potential for a very large payout if they survived while others wanted the more predictable annuity income option .
the oldest survivor of king william s tontine of 1693 lived to age 100 , earning 1,000 in her final year and no doubt well tended - to by her family .
see milevsky ( 2014 ) for more on king william s tontine .
our main practical points will be that ( i ) tontines and life annuities have co - existed in the past , so perhaps they can in the future ; ( ii ) there is no reason to construct the tontine payout function such that the last survivor receives hundreds of multiples of their initial investment ; and ( iii ) a properly - constructed tontine can result in lifetime utility that is comparable to the utility of a life annuity .
indeed , both the tontine and annuity hedge individuals against their idiosyncratic mortality risk : the difference is that the tontine leaves individuals exposed to aggregate longevity risk , while the annuity transfers this risk to the vendor for a fee .
if this fee is sufficiently high , the tontine will offer greater utility than the annuity . to our knowledge , we are the first to derive the properties of an optimized tontine payout structure , in contrast to the optimal life annuity which is well - known in the literature .
we believe this is part of the `` optimal insurance design '' literature , similar to the recent paper by bernard , et .
an interesting insurance product , that lies between a tontine and annuity , is the _
pooled annuity_. stamos ( 2008 ) analyzes the optimal design of such a pool , while donnelly , guillen and nielsen ( 2014 ) compare its utility to that of alternatives and a recent article by donnelly ( 2015 ) examines the `` fairness '' of pooled annuity funds in general . as with a tontine ,
pooled annuity funds insulate the issuer from longevity risk .
these hybrid designs yield utility intermediate between that of a tontine and an annuity , at the expense of a more complex product design ( from the client point of view ) .
we will discuss this , and review the broader literature , in section [ lit ] .
the remainder of this paper is organized as follows .
section [ 1693 ] very briefly describes the highlights of the first ever english government tontine of 1693 , which interestingly had some aspects of our _ optimal tontine .
_ it also provides some motivation and colour to the paper .
section [ theory ] is the theoretical core of the paper , which derives the properties of the optimal tontine structure and compares this optimal structure with the life annuity .
section [ numerics ] uses and applies the results from section [ theory ] and addresses how 21st - century tontines might be constructed .
section [ conc ] concludes the paper and section [ apen ] is an appendix with most proofs and derivations .
in early 1693 , during the fourth year of the reign of king william iii and queen mary ii , the english parliament passed the so - called _ million act _ , which was an attempt to raise one million pounds towards carrying on the war against france .
the _ million act _ specified that prior to may 1 , 1693 , any british native or foreigner could purchase a tontine share from the exchequer for 100 and thus gain entry into the first british government tontine scheme . for 100 an investor could select any nominee of any age including the investor himself on whose life the tontine would be contingent .
dividend payments would be distributed to the investor as long as the nominee was still alive .
now , to put the magnitude of the minimal 100 investment in perspective , the average annual wage of building laborers in england during the latter part of the 17th century was approximately 16 and a few shillings per year so the entry investment in the tontine pool far exceeded the average industrial wage , and the annual dividends alone might serve as a decent pension for a common labourer . accordingly
, it is quite plausible to argue that the 1693 tontine was an investment for the rich and perhaps even one of the first exclusive hedge funds .
this was a simpler structure compared to the original tontine scheme envisioned by lorenzo de tonti in the year 1653 , which involved multiple classes . in the 1693 english tontine
, each share of 100 would entitle the investor to an annual dividend of 10 for seven years ( until june 1700 ) , after which the dividends would be reduced to 7 per share .
the 10% and 7% tontine dividend rate exceeded prevailing ( market ) interest rates in england at the end of the 17th century , which were officially capped at 6% .
note the declining structure of the interest payments , which is a preview of our soon - to - come discussion ( in section 3 ) about the optimal tontine payout function . moving on to the annuity side of the offering , to further entice investors to participate in the tontine scheme ,
the act included a unique sweetener " or bonus provision .
it stipulated that if the entire 1,000,000 target was nt subscribed by may 1693 thus reducing the size of the lottery payoff for the final survivor the investors who had enrolled in the tontine during the six - month subscription period ( starting in november 1692 ) would have the option of converting their 100 tontine shares into a life annuity paying 14 per year . under this alternative ,
the 14% dividend payments were structured as a conventional life annuity with no group survivorship benefits or tontine features .
( think of a single premium life annuity . )
this `` option to choose '' between a tontine and a life annuity is quite intriguing and fodder for further research . alas ,
the funds raised by early may 1963 fell far short of the million - pound target . according to records maintained by the office of the exchequer , stored within the archives of the british library ,
a total of only 377,000 was subscribed and approximately 3,750 people were nominated to the tontine prior to may 1693 .
this then triggered the option to exchange the tontine into a 14% life annuity .
interestingly , it seems that a total of 1,013 nominees ( representing 1,081 tontine shares ) remained in the original 10%/7% tontine , while the other two - thirds elected to convert their tontine shares into a 14% life annuity contingent on the same nominee .
( see milevsky ( 2014 ) for an extensive historical discussion of the 1693 tontine , the characteristics of investors who selected the tontine versus the annuity and some of the empirical factors driving the decisions made by participants .
here we only describe the highlights . ) at first glance , it is rather puzzling why anyone would stay in the tontine pool instead of switching to the life annuity . on a present - value basis ,
a cash flow of 10 for seven years and 7 per annum thereafter is much less valuable compared to 14 per annum for life . as discussed in milevsky ( 2014 ) , the actuarial present value of the 10%/7% combination at the 6% official interest rate was worth approximately 133 at the ( typical nominee ) age of ten , whereas the value of the life annuity was worth almost 185 .
( remember , the original investment was 100 . )
it should come as no surprise that the british government was losing money on these 14% annuities and they were offering these terms to anyone , regardless of how young they were .
why did anyone stay in the tontine pool ? could differing views on mortality and/or risk aversion
explain why some investors switched , while others did nt ?
can a rational explanation for this seemingly irrational choice be found ?
the framework we will introduce in section 3 might help shed light on this decision , or at least rationalize the choice .
motivated by the historical tontine , in the next section we present an economic theory to describe and understand the choice theory illuminating who might elect to participate in a tontine and who might choose a life annuity , as well as the properties of a tontine that are likely to generate the highest lifetime utility .
to pre - empt the result , we will show that although the life annuity clearly dominated the tontine in terms of lifetime utility , a declining tontine payout structure ( as in the 1693 tontine ) is in fact optimal .
so , having a tontine that pays 10% in the first few years ( to the entire pool ) and then 7% for the remaining years
( to the entire pool ) is a historical aberration but will be shown to be based in sound economic theory .
we assume an objective survival function @xmath5 , for an individual aged @xmath6 to survive @xmath7 years .
one purpose of the tontine structure is to insulate the issuer from the risk of a stochastic ( or simply mis - specified ) survival function , but in this paper we assume @xmath5 is given , and applies to all individuals .
we leave for future work the exploration of subjective vs. objective survival rates , or stochastic hazard rates .
we assume that the tontine pays out continuously , as opposed to quarterly or monthly . for simplicity , we assume a constant risk - free interest rate @xmath8 ,
though it would be simple to incorporate a term structure ( all funds contributed are invested at time 0 , so payouts could be funded from a static bond portfolio as easily as from a money market account ) .
our basic annuity involves annuitants ( who are also the nominees ) each paying $ 1 to the insurer initially , and receiving in return an income stream of @xmath9 for life .
the constraint on these annuities is that they are fairly priced , in other words that with a sufficiently large client base , the initial payments invested at the risk - free rate will fund the called - for payments in perpetuity ( later we discuss the implications of insurance loadings ) .
this implies a constraint on the annuity payout function @xmath10 , namely that @xmath11 though @xmath10 is the payout rate per survivor , note that the payout rate per initial dollar invested is @xmath12 .
we will return to this later . letting @xmath13 denote the instantaneous utility of consumption ( a.k.a .
the felicity function ) , a rational annuitant ( with lifetime @xmath14 ) having no bequest motive will choose a life annuity payout function for which @xmath10 maximizes the discounted lifetime utility : @xmath15=\int_0^\infty e^{-rt}{}_tp_x\ , u(c(t))\,dt \label{annuityutilityspecification}\ ] ] where @xmath8 is ( also ) the subjective discount rate ( sdr ) , all subject to the constraint . by the euler - lagrange theorem , this implies the existence of a constant @xmath16 such that @xmath17 in other words , @xmath18 is constant , so provided that utility function @xmath13 is strictly concave , the optimal annuity payout function @xmath10 is also constant .
that constant is now determined by , showing the following : optimized life annuities have constant @xmath19 , where @xmath20^{-1}.\ ] ] this result can be traced back to yaari ( 1965 ) who showed that the optimal ( retirement ) consumption profile is constant ( flat ) and that 100% of wealth is annuitized when there is no bequest motive . for more details and an alternate proof ,
see the excellent book by cannon and tonks ( 2008 ) and specifically the discussion on annuity demand theory in chapter 7 . in practice ,
insurance companies funding the life annuity @xmath10 are exposed to both systematic longevity risk ( due to randomness or uncertainty in @xmath5 ) , model risk ( the risk that @xmath5 is mis - specified ) , as well as re - investment or interest rate risk ( which is the uncertainty in @xmath8 over long horizons ) .
the latter is not our focus here , so we will continue to assume that @xmath8 is a given constant for most of what follows . note that even if re - investment rates were known with certainty , the insurance company would likely pay out less than the @xmath10 implied by equation as a result of required capital and reserves , effectively lowering the lifetime utility of the ( annuity and the ) retiree .
this brings us to the tontine structures we will consider as an alternative , in which a predetermined dollar amount is shared among survivors at every @xmath7 . let @xmath21 be the rate at which funds are paid out per initial dollar invested , a.k.a . the tontine payout function .
our main point in this paper is that there is no reason for tontine payout function to be a constant fixed percentage of the initial dollar invested ( e.g. 4% or 7% ) , as it was historically .
in fact , we can pose the same question as considered above for annuities : what @xmath21 is optimal for subscribers , subject to the constraint that sponsors of the tontine can not sustain a loss ?
note that the natural comparison is now between @xmath21 and @xmath12 , where @xmath10 is the optimal annuity payout found above .
suppose there are initially @xmath0 subscribers to the tontine scheme , each depositing a dollar with the tontine sponsor .
let @xmath22 be the random number of live subscribers at time @xmath7 .
consider one of these subscribers .
given that this individual is alive , @xmath23 .
in other words , the number of other ( live ) subscribers at any time @xmath7 is binomially distributed with probability parameter @xmath5 .
so , as we found for the life annuity , this individual s discounted lifetime utility is @xmath24=\int_0^\infty e^{-rt}{}_tp_x \,e[u\big(\frac{n d(t)}{n(t)}\big)\mid
\zeta > t]\,dt\\ & \qquad=\int_0^\infty e^{-rt}{}_tp_x\sum_{k=0}^{n-1 } \binom{n-1}{k}{}_tp_x^k(1-{}_tp_x)^{n-1-k}u\big(\frac{nd(t)}{k+1}\big)\,dt.\end{aligned}\ ] ] the constraint on the tontine payout function @xmath21 is that the initial deposit of @xmath0 should be sufficient to sustain withdrawals in perpetuity . of course , at some point all subscribers will have died , so in fact the tontine sponsor will eventually be able to cease making payments , leaving a small remainder or windfall . but
this time is not predetermined , so we treat that profit as an unavoidable feature of the tontine .
remember that we do not want to expose the sponsor to any longevity risk .
it is the pool that bears this risk entirely .
our budget or pricing constraint is therefore that @xmath25 so , for example , if @xmath26 is forced to be constant ( the historical structure , which we call a _ flat tontine _ ) , then the tontine payout function ( rate ) is simply @xmath27 ( or somewhat more if a cap on permissible ages is imposed , replacing the upper bound of integration in by a value less than infinity ) .
we are instead searching for the _ optimal _ @xmath21 which we will find is far from constant . by the euler - lagrange theorem from the calculus of variations , there is a constant @xmath16 such that the optimal @xmath21 satisfies @xmath28 for every @xmath7 . note that this expression directly links individual utility @xmath29 to the optimal participating annuity .
recall that a tontine is an extreme case of participation or pooling of all longevity risk .
equation dictates exactly how a risk - averse retiree will trade off consumption against longevity risk . in other words
, we are not advocating an _ ad hoc _ actuarial process for smoothing realized mortality experience .
note that an actual mortality hazard rate @xmath30 does not appear in the above equation
it appears only implicitly , in both @xmath5 and @xmath16 ( which is determined by ) .
therefore , we will simplify our notation by re - parametrizing in terms of the probability : let @xmath31 satisfy @xmath32 substituting @xmath33 into the above equation , it collapses to @xmath34 .
the optimal tontine structure is @xmath35 , where @xmath16 is chosen so holds .
it is feasible ( but complicated ) to solve this once a generic @xmath36 is given .
but in the case of constant relative risk aversion ( crra ) utility it simplifies greatly .
let @xmath37 if @xmath38 , and when @xmath39 take @xmath40 instead .
define @xmath41=\sum_{k=0}^{n-1 } \binom{n-1}{k}p^{k}(1-p)^{n-1-k}\big(\frac{n}{k+1}\big)^{1-\gamma}\ ] ] where @xmath42 .
. then [ crraoptimum ] with crra utility , the optimal tontine has withdrawal rate @xmath44 , where @xmath45^{-1}. \label{d(1)formula}\ ] ] suppose @xmath38 .
then the equation for @xmath46 becomes that @xmath47 the constraint now implies .
a similar argument applies when @xmath48 .
it is worth emphasizing that @xmath49 does not depend on the particular form of the mortality hazard rate @xmath30 , or on the interest rate @xmath8 , but only on the longevity risk aversion @xmath1 and the number of initial subscribers to the tontine pool , @xmath0 .
in other words , the mortality hazard rate and @xmath8 enter into the expression for @xmath46 only via the constant @xmath50 .
we will prove the following in section [ proof1 ] of the appendix : for any @xmath51 and @xmath52 [ betabound ] 1 .
@xmath53 increases with @xmath54 ; 2 .
@xmath55 and @xmath56 ; 3 .
@xmath57 , @xmath58 , @xmath59 ; 4 .
@xmath60 and @xmath61 ; 5 .
for @xmath62 we have that @xmath63 one immediate consequence of corollary [ crraoptimum ] and ( a ) of lemma [ betabound ] is that the optimal tontine payout rate @xmath21 decreases with @xmath7 .
by ( d ) we have that @xmath64 for @xmath65 , so by dominated convergence , @xmath66 as @xmath67 . therefore the ( historical ) flat tontine structure @xmath68 is optimal in the limit as @xmath67 .
likewise @xmath69 for @xmath70 .
renormalizing , this implies that @xmath71 concentrates increasingly near @xmath33 .
in other words , as we approach risk - neutrality , the optimal tontine comes closer and closer to exhausting itself immediately following @xmath72 .
we define a _ natural tontine _ to have payout @xmath73 where @xmath74 is proportional to @xmath54 , just as is the case for the annuity payment per initial dollar invested . comparing the budget constraints and
, we get that @xmath75 , so the natural tontine payout rate agrees with that of the annuity ( which justifies our singling it out ) . by corollary [ crraoptimum ] and ( c ) of lemma [ betabound ]
, we see that the natural tontine is optimal for logarithmic utility @xmath48 .
we will see in section [ numerics ] that the natural tontine is close to optimal when @xmath0 is large .
we therefore propose the natural tontine as a reasonable structure for designing tontine products in practice , rather than expecting insurers to offer a range of products with differing @xmath1 s .
figure [ fig5 ] shows the ratio of @xmath46 to @xmath74 , with gompertz hazard rate @xmath76 , and parametrized by @xmath7 via @xmath77 .
this exponentially - rising hazard rate is the basic mortality assumption in much of the actuarial literature .
the figure provides numerical evidence that higher risk aversion implies a preference for reducing consumption at early ages in favour of reserving funds to consume at advanced ages . to make this statement precise ,
define @xmath78 namely the present value of payouts from the optimal tontine through time @xmath7 , per initial dollar invested ( or equivalently , the proportion of initial capital used to fund payouts till time @xmath7 ) .
we conjecture that @xmath79 in any realistic situation ( specifically , whenever the morality distribution has an increasing hazard rate ) .
we have not yet succeeded in proving this conjecture , but instead will derive a number of partial results that support it .
in fact , numerical evidence suggests that holds regardless of the mortality distribution , provided @xmath80 is only of moderate size .
it is in fact possible to construct pathological mortality distributions for which fails when @xmath80 and @xmath81 are large . specifically , this may actually happen during a low - hazard rate lull between two periods with high hazard rates .
we discuss this at greater length in the appendix , where we will also prove the following instances of the above conjecture .
[ increasinggamma ] holds in the following situations : 1 . for @xmath82 and @xmath83 arbitrary
; 2 . for any fixed @xmath81 , @xmath0 , and @xmath7 , in the limit as @xmath84
; 3 . for some initial period of time @xmath85 , where @xmath86 depends on @xmath87 and @xmath0 .
we will also prove a modified version of , for fixed @xmath87 and @xmath7 , in the limit as @xmath88 ( see theorem [ asymptotictheorem ] ) .
in section [ numerics ] we will provide a variety of numerical examples that illustrate the optimal tontine payout function @xmath21 as a function of longevity risk aversion @xmath1 and the initial size of the tontine pool @xmath0 .
we now compare the annuity and tontine .
[ tontineannuitycagematch ] let @xmath89 denote the utility of the optimal tontine . to compute this ,
suppose @xmath38 , and observe that @xmath90 by .
the utility of the optimal tontine is therefore precisely @xmath91 by and .
consider instead the utility @xmath92 provided by the annuity , namely @xmath93 @xmath94 for any @xmath0 and @xmath52 .
[ utilityinequality ] for @xmath38 this follows from ( e ) of lemma [ betabound ] and the calculations given above .
we show the case @xmath48 in the appendix .
of course , the conclusion is economically plausible , as the tontine retains risk that the annuity does nt .
the insurer offering an annuity will be modelled as setting aside some fraction of the initial deposits to fund the annuity s costs . in other words , a fraction @xmath95 of the initial deposits
are deducted initially , to fund risk management , capital reserves , etc .
the balance , invested at the risk - free rate @xmath8 will fund the annuity .
therefore , with loading , becomes that @xmath96 , which implies that @xmath97 is the optimal payout structure for the annuity .
the utility of the loaded annuity is therefore @xmath98 for @xmath38 , and @xmath99 for @xmath48 . in section [ numerics ]
we will consider numerically the _ indifference loading _
@xmath95 that , when applied to the annuity , makes an individual indifferent between the annuity and a tontine , i.e. @xmath100 .
it turns out that the loading @xmath95 decreases with @xmath0 , in such a way that the _ total _ loading @xmath101 stays roughly stable . in other words , there is at most a fixed amount ( roughly ) that the insurer can deduct from the _ aggregate _ annuity pool , regardless of the number of participants , before individuals start to derive greater utility from the tontine .
we will illustrate this observation , at least for @xmath102 , by proving the following inequality in the appendix : [ loadinginequality ] suppose that @xmath102 . then @xmath103 .
note that @xmath104 , since @xmath105 .
there are several issues we do not address here which we openly acknowledge and leave for future research .
in particular , there is the role of _ credit risk _ as well as the impact of _ stochastic mortality _ which will change the dynamics between tontines and annuities .
the existence of credit risk is a much greater concern for the buyers of life annuities , vs. tontines , given the risk assumed by the insurance sponsor . likewise , under a stochastic mortality model the tontine payout would be more variable and uncertain , which might reduce the utility of the tontine relative to the life annuity . on the other hand , the capital charges added to the market price of the life annuity would be higher in a stochastic mortality model .
we appreciate that these are an unanswered ( or unaddressed ) questions and leave the examination of the robustness of the _ natural tontine _ payout in a full stochastic mortality environment to a subsequent paper .
we note that there is also the question of asymmetric mortality , in which the individual believes that his or her ( subjective ) hazard rate is less than that of the typical tontine investor ( for whom the tontine payout function is optimized ) .
indeed , this is potentially the explanation for the significant fraction of investors in king william s tontine , who did not exercise their option to convert to an annuity .
they ( might have ) thought their nominee was _ much _ healthier than everyone else , and the only way to capitalize on that was to invest in the tontine .
this leads to the idea that the _ mortality loading _ , or level of asymmetric mortality , would induce an individual to prefer the tontine to an annuity .
this , obviously , ties into the issue of stochastic mortality and is being addressed in a follow - up paper by milevsky and salisbury ( 2015 ) . in the next section ,
we focus on the numerical implications of our current model .
figure [ fig1 ] displays the range of possible 4% flat tontine dividends over time , assuming an initial pool of @xmath106 nominees , under a gompertz law of mortality with parameters @xmath107 and @xmath108 .
this mortality basis corresponds to a survival probability of @xmath109 , i.e. from age 65 to age 100 and is the baseline value for several of our numerical examples .
the figure clearly shows an increasing payment stream conditional on survival , which is nt the optimal function .
* figure [ fig1 ] goes here * indeed , in such a ( traditional , historical ) tontine scheme , the initial expected payment is quite low , relative to what a life annuity might offer in the early years of retirement ; while the payment in the final years for those fortunate enough to survive would be both very high and quite variable .
it is not surprising then that this form of tontine is both suboptimal from an economic utility point of view and also is nt very appealing to individuals ( with reasonable risk aversions ) who want to maximize their standard of living over their entire retirement lifespan .
* figure [ fig2 ] goes here * in contrast to figure [ fig1 ] which displays the sub - optimal flat tontine , figure [ fig2 ] displays the range of outcomes from the optimal tontine payout function , under the same interest rate and mortality basis . to be very specific , figure [ fig2 ]
is computed by solving for the value of @xmath50 and then constructing @xmath110 for @xmath111 and @xmath48 .
once the payout function is known for all @xmath7 , the number of survivors at the 10th and 90th percentile of the binomial distribution is used to bracket the range of the payout from age 65 to age 100 .
clearly , the expected payout per survivor is relatively constant over the retirement years , which is much more appealing intuitively .
moreover , the discounted expected utility from a tontine payout such as the one displayed in figure [ fig2 ] is much higher than the utility of the one displayed in figure [ fig1 ] .
* table [ table06 ] goes here * table [ table06 ] displays the optimal tontine payout function for a very small pool of size @xmath112 .
these values correspond to the @xmath110 values derived in section [ theory ] .
notice how the optimal tontine payout function is quite similar ( identical in the first significant digit ) regardless of the individual s longevity risk aversion ( lora ) @xmath1 , even when the tontine pool is relatively small at @xmath112 .
the minimum guaranteed dividend starts off at about 7% at age 65 and then declines to approximately 1% at age 95 .
of course , the actual cash flow payout to an individual , conditional on being alive does not necessarily decline and in reality stays relatively constant .
* table [ table07 ] goes here * table [ table07 ] displays utility indifference loadings @xmath95 for a participant at age 60 .
notice how even a retiree with a very high level of longevity risk aversion ( lora ) @xmath1 , will select a tontine ( with pool size @xmath113 ) instead of a life annuity if the insurance loading is greater than 7.5% ; for less extreme levels of risk aversion , and larger pools , the required loading is much smaller ( e.g. 30 b.p . ) .
recall that this is a one - time charge at the time of initiation , not an annualized value over the duration of the contract ( which would be a significantly smaller value ) .
* table [ table08 ] goes here * table [ table08 ] computes certainty equivalent factors associated with natural tontines .
see section [ moreoncertaintyequiv ] for further discussion of this comparison .
if an individual with lora @xmath114 is faced with a tontine structure that is only optimal for someone with lora @xmath48 ( i.e. logarithmic utility ) the welfare loss is minuscule .
this is one reason we advocate the _ natural tontine _ payout function , which is only precisely optimal for @xmath48 , as the basis for 21st - century tontines .
* figure [ fig3 ] goes here * figure [ fig3 ] compares the optimal tontine payout function @xmath21 for different levels of longevity risk aversion @xmath1 , and shows that the difference is barely noticeable when the tontine pool size is greater than @xmath115 , mainly due to the effect of the law of large numbers .
this curve traces the _ minimum _ dividend that a survivor can expect to receive at various ages ( i.e. when all purchasers survive ) .
the median is ( obviously ) much higher ( see figure [ fig2 ] ) .
* figure [ fig4 ] goes here * figure [ fig4 ] illustrates that the optimal tontine payout function for someone with logarithmic @xmath48 utility starts off paying the exact same rate as a life annuity regardless of the number of participants in the tontine pool , ie @xmath116 .
but , for higher levels of longevity risk aversion @xmath1 and a relatively smaller tontine pool , the function starts off at a lower value and declines at a slower rate .
the perpetuity curve corresponds to @xmath117 .
* figure [ fig5 ] goes here * figure [ fig5 ] shows that if we compare two retirees , the one who is more averse to longevity risk will prefer a higher guaranteed minimum payout rate ( gmpr ) at advanced ages . in exchange
they will accept a lower gmpr at younger ages .
we choose to quantify this by taking the natural tontine payout as our baseline , since this is the product structure we advocate .
we take quite a small pool size here ( @xmath112 ) to illustrate the effect .
the effect persists at larger pool sizes , but is much less dramatic .
* figure [ fig6 ] goes here * figure [ fig6 ] shows an actuarially - fair life annuity that guarantees 7.5% for life starting at age 65 .
it provides more utility than an optimal tontine regardless of longevity risk aversion ( lora ) or the size of the tontine pool .
but , once an insurance loading is included , driving the annuity yield under the initial payout from the optimal tontine , the utility of the life annuity might be lower .
the indifference loading is @xmath95 and is reported in table [ table07 ] .
so here is our main takeaway and idea in the paper , once again .
the historical tontine in which dividends to the entire pool are a constant ( e.g. 4% ) interest rate over the entire retirement horizon are suboptimal because they create an increasing consumption profile that is both variable and undesirable .
however , a tontine scheme in which interest payments to the pool early on are higher ( e.g. 8% ) and then decline over time , so that the few winning centenarians receive a much lower interest rate ( e.g. 1% ) is in fact the optimal design .
coincidently , king william s 1693 tontine had a similar declining structure of interest payments to the pool , which was quite rare historically .
we are careful to distinguish between the guaranteed _ interest _ rate ( e.g. 8% or 1% ) paid to the entire pool , and the expected _ dividend _ to the individual investor in the optimal tontine , which will be relatively constant over time , as is evident from figure [ fig2 ] .
of course , the present value of the interest paid to the entire pool over time is exactly equal to the original contribution made by the pool itself .
we are simply re - arranging and parsing cash flows of identical present value , in a different manner over time .
we have shown that a tontine provides less utility than an actuarially - fair life annuity , which is reasonable given that the tontine exposes investors to longevity risk .
what is striking is that the utility difference from a properly - designed tontine scheme is actually quite small when compared to an actuarially - fair life annuity , which is the workhorse of the pension economics and lifecycle literature .
in fact , the utility from a tontine might actually be higher than the utility generated by a pure life annuity when the insurance ( commission , capital cost , etc . )
one - time loading exceeds 10% .
this result should not negate or be viewed as conflicting with the wide - ranging annuity literature which proves the optimality of life annuities in a lifecycle model .
both tontines and annuities hedge idiosyncratic mortality risk .
in fact , what we show is that it is still optimal to fully hedge the remaining systemic longevity risk , but the instrument that one uses to do so depends on the relative costs . in other words , _ sharing _
longevity risk amongst a relatively small ( @xmath118 ) pool of people does nt create large dis - utilities or welfare losses , at least within a classical rational model .
this finding can also be viewed as a further endorsement of the participating life annuity , which lies in between the tontine and the conventional life annuity .
this is not the place nor do we have the space for a full review of the literature on tontines , so we provide a selected list of key articles for those interested in further reading . for literature and all sources available for the 1693 tontine , we refer to the earlier mentioned historical paper by milevsky ( 2014 ) as well as the book by milevsky ( 2015 ) .
more generally , the original tontine proposal by lorenzo de tonti appears in french in tonti ( 1654 ) ( and in english translation in the cited reference ) .
the review article by kopf ( 1927 ) and the book by odonnell ( 1936 ) are quite dated , but document how the historical tontine operated , discussing its checkered history , and providing a readable biography of some of its earliest promoters in denmark , holland , france and england .
the monograph by cooper ( 1972 ) is devoted entirely to tontines and the foundations of the 19th - century ( u.s . )
tontine insurance industry , which is based on the tontine concept but is somewhat different because of the savings and lapsation component .
in a widely - cited article , ransom and sutch ( 1987 ) provide the background and story of the banning of tontine insurance in new york state , and then eventually the entire u.s .
the comprehensive monograph by jennings and trout ( 1982 ) reviews the history of tontines , with particular emphasis on the french period , while carefully documenting payout rates and yields from most known tontines . for
those interested in the pricing of mortality - contingent claims during the 17th and 18th century , as well as the history and background of the people involved , see alter ( 1986 ) , poitras ( 2000 ) , hald ( 2003 ) , poterba ( 2005 ) , ciecka ( 2008a , 2008b ) , rothschild ( 2009 ) as well as bellhouse ( 2011 ) , and of course , homer and sylla ( 2005 ) for the relevant interest rates .
more recently , the newspaper article by chancellor ( 2001 ) , the book by lewin ( 2003 ) and especially the recent review by mckeever ( 2009 ) all provide a very good history of tontines and discuss the possibility of a tontine revival .
the standard actuarial textbooks , such as promislow ( 2011 ) or pitacco , et al .
( 2009 ) for example , each have a few pages devoted to the historical tontine principal .
more relevant to the modern design of participating annuities and sharing longevity risk in the 21st century , a series of recent papers on pooled annuity funds have resurrected the topic .
for example piggot , valdez and detzel ( 2005 ) , valdez , piggott and wang ( 2006 ) , stamos ( 2008 ) , richter and weber ( 2011 ) , donnelly , guillen and nielsen ( 2013 ) as well as qiao and sherris ( 2013 ) have all attempted to reintroduce and analyze tontine - like structures in one form or another . in these proposed structures ,
investors contribute to an initial pool of capital , which is invested , and then paid out over time to the survivors .
it need not be the case that investments are risk free ( as we have assumed for the tontines we analyze ) , so the control variables are the asset allocation and payout strategy .
the prospectus given to pooled annuity investors should therefore specify both .
donnelly , guillen and nielsen ( 2013 ) consider a different approach , in the context of pooled annuities , to what we above have called the indifference annuity loading .
instead of the loading being payed in a lump sum at the time of purchase ( as we have done ) , they incorporate a variable fee into their guaranteed product , paid continuously over time .
they find that this fee may be structured so the utility of the pooled annuity precisely matches that of the guaranteed product it is being compared to , at all times .
they term this fee structure the `` lifetime breakeven costs '' , and explore its properties and implications for consumption . to further compare with our results ,
let us assume for now that the pooled annuity funds are indeed invested risk - free .
following the argument of stamos ( 2008 ) , one would then obtain the optimal _ proportion _ of the remaining funds that should be paid out at each time , as a function of both time _ and the number of individuals remaining in the pool_. in a tontine , the amount withdrawn may vary with time , but not with the number of survivors . in other words , the prospectus for a tontine should lay out a specific dollar amount to be withdrawn in each year of the contract , so any uncertainty in the amount paid to an individual derives simply from the size of the pool with whom the withdrawal is split .
a pooled annuity has the latter uncertainty , but in addition has uncertainty about the dollar amount to be withdrawn , which will vary in a path - dependent way according to how the mortality experience of the pool evolves . to know how much you will receive in year 10 , it is no longer enough to know how many survivors there are in year 10 , but you also need to know how many survived in each of years one through nine . there are both advantages and disadvantages to pooled annuities versus tontines . on the plus side
, the pooled annuity has an extra variable to control for , so should yield higher utility with risk - free investing we expect its utility to be intermediate between that of a tontine and an annuity .
as we have seen , the latter are in fact very close , so the utility gain ( though real ) is actually modest . on the down side , the path - dependent nature of a pooled annuity ( or the annuity overlay introduced by donnelly et .
al . ( 2014 ) ) makes it more complex to explain to participants , and more difficult for those participants to evaluate the risks and uncertainties of their income stream .
for these reasons , we ( on balance ) advocate tontines .
the simpler design of tontines certainly makes their calculations less computationally intensive . at a mathematical level
, it also allows us to go further in analyzing their properties , which is an important component of the current paper .
the closest other work we can find to our natural tontine payout is that of sabin ( 2010 ) on a fair annuity tontine .
he gives a specific interpretation of `` fair '' and analyzes how to adjust tontine payments for heterogeneous ages and initial contributions . a related fairness question for pooled annuities
is treated in donnelly ( 2015 ) .
our natural tontine is `` fair '' by construction because we do nt mix cohorts or ages . in sum , although research on tontine schemes is scattered across the insurance , actuarial , economic , and history journals , we have come across few , if any , scholarly articles that condemn or dismiss the tontine concept outright .
it is not widely known that in the year 1790 , the first u.s . secretary of the treasury , alexander hamilton proposed one of the largest tontines in history . to help reduce a crushing national debt something that is clearly not a recent phenomenon he suggested the u.s .
government replace high - interest revolutionary war debt with new bonds in which coupon payments would be made to a group , as opposed to individuals .
the group members would share the interest payments evenly amongst themselves , provided they were alive .
but , once a member of the group died , his or her portion would stay in a pool and be shared amongst the survivors .
this process would theoretically continue until the very last survivor would be entitled to the entire interest payment potentially millions of dollars .
this obscure episode in u.s .
history has become known as hamilton s tontine proposal , which he claimed in a letter to george washington would reduce the interest paid on u.s .
debt , and eventually eliminate it entirely .
although congress decided not to act on hamilton s proposal the tontine idea itself never died on american soil .
insurance companies began issuing tontine insurance policies which are close cousins to de tonti s tontine to the public in the mid-19th century , and they became wildly popular . by the start of the 20th century , historians have documented that half of u.s .
households owned a tontine insurance policy , which many used to support themselves through retirement .
the longer one lived , the greater their payments .
this was a personal hedge against longevity , with little risk exposure for the insurance company .
sadly though , due to shenanigans and malfeasance on the part of company executives , the influential new york state insurance commission banned tontine insurance in the state , and by 1910 most other states followed .
tontines have been illegal in the u.s . for over a century and most insurance executives have likely never heard of them .
tontines not only have a fascinating history but have some basis in economic principles , in fact , adam smith himself , writing in the _ wealth of nations _
, noted that tontines may be preferred to life annuities .
we believe that a strong case can be made for overturning the current ban on tontine insurance _ allowing both tontine and annuities to co - exist as they did 320 years ago _ with suitable adjustments to alleviate problems encountered in the early 20th century .
indeed , given the insurance industry s concern for longevity risk capacity , and its poor experience in managing the risk of long - dated fixed guarantees , one can argue that an ( optimal ) tontine annuity is a triple win " proposition for individuals , corporations and governments .
see milevsky ( 2015 ) for further arguments along this line .
it is worth noting that under the proposed ( eu ) solvency ii guidelines for insurer s capital as well as risk management , there is a renewed focus on _ total balance sheet _ risks .
in particular , insurers will be required to hold _ more _ capital against market risk , credit risk and operational risk .
in fact , a recently - released report moody s ( 2013 ) claims that _ solvency ratios will exhibit a more complex volatility under solvency ii than under solvency i , as both the available capital and the capital requirements will change with market conditions . _ according to many commentators this is likely to translate into higher consumer prices for insurance products with long - term maturities and guarantees . and , although this only applies to european companies ( at this point time ) , it is not unreasonable to conclude that in a global market , annuity loadings will increase , making ( participating ) tontine products relatively more appealing to price - sensitive consumers .
moreover , perhaps a properly - designed tontine product could help alleviate the low levels of voluntary annuity purchases a.k.a . the annuity puzzle by gaming the behavioral biases and prejudices exhibited by retirees .
the behavioral economics literature and advocates of cumulative prospect theory have argued that consumers make decisions based on more general _ value functions _ with personalized decision weights . among other testable hypotheses ,
this leads to a preference for investments with ( highly ) skewed outcomes , even when the alternative is a product with the same expected present values .
of course , whether the public and regulators can be convinced of these benefits remains to be seen , but a debate would be informative .
we are not alone in this quest .
indeed , during the last decade a number of companies around the world egged on by scholars and journalists have tried to resuscitate the tontine concept while trying to avoid the bans and taints .
although the specific designs differ from proposal to proposal , all share the same idea we described in this paper : companies act as custodians and guarantee very little .
this arrangement requires less capital which then translates into more affordable pricing for the consumer .
once again the main qualitative insight from our model is that a properly - designed tontine could hold its own in the _ utility arena _
, against an actuarially - fair life annuity and pose a real challenge to a loaded annuity .
the next ( natural ) step would be to examine the robustness of our results in a full - blown stochastic mortality model . 99 alter , g. ( 1986 ) , how to bet on lives : a guide to life contingent contracts in early modern europe , _ research in economic history _ , vol .
10 , pg . 1 - 53 .
barberis , n.c .
( 2013 ) , thirty years of prospect theory in economics : a review and assessment , _ the journal of economic perspectives _ ,
27(1 ) , pg .
173 - 196 bellhouse , d. r. ( 2011 ) , _ de moivre : setting the stage for classical probability and its applications _ , crc press , new york .
bernard , c. , x. he , j .- a .
yan and x. y. zhou ( 2013 ) , optimal insurance design under rank - dependent utility , _ mathematical finance _ , vol . 25 ( 1 ) , pg .
154186 cannon , e. and i. tonks ( 2008 ) , _ annuity markets _ , oxford university press , uk .
chancellor , e. ( 2001 ) , life long and prosper , _ the spectator _ , 24 march .
chung , j. and g. tett ( 2007 ) , death and the salesmen : as people live longer , pension funds struggle to keep up , which is where a new , highly profitable market will come in
one that bets on matters of life and death , _ the financial times _ , february 24 , page 26 .
ciecka , j.e .
( 2008a ) , the first mathematically correct life annuity valuation formula , _ journal of legal economics _
15(1 ) , pg .
59 - 63 .
ciecka , j.e .
( 2008b ) , edmond halley s life table and its uses , _ journal of legal economics _
15(1 ) , pg . 65 - 74 .
cooper , r. ( 1972 ) , _ an historical analysis of the tontine principle with emphasis on tontine and semi - tontine life insurance policies _ , s.s .
huebner foundation for insurance education , university of pennsylvania .
dai , m. , y.k .
kwok and j. zong ( 2008 ) , guaranteed minimum withdrawal benefit in variable annuities , _ mathematical finance _
18(3 ) , pg .
595 - 611 dickson , p.g.m .
( 1967 ) , _ the financial revolution in england a study in the development of public credit , 1688 - 1756_. published by macmillan donnelly , c. , m. guillen and j.p .
nielsen ( 2013 ) , exchanging mortality for a cost , _ insurance : mathematics and economics _ ,
52(1 ) , pg . 65 - 76 .
donnelly , c. , m. guilln and j.p .
nielsen ( 2014 ) , bringing cost transparency to the life annuity market , _ insurance : mathematics and economics _ ,
56 ( 1 ) , pg . 14?27 .
donnelly , c. ( 2015 ) actuarial fairness and solidarity in pooled annuity funds , _ astin bulletin _ , vol .
. 49 - 74 elsgolc , l.d .
( 2007 ) , _ calculus of variations _ ,
dover publications , mineola , new york .
finlaison , j. ( 1829 ) , _ report of john finlaison , actuary of the national debt , on the evidence and elementary facts on which the tables of life annuities are founded _ ,
house of commons , london , uk .
gelfand , i.m . and s.v .
fomin ( 2000 ) , _ calculus of variations _ , translated by r.a .
silverman , dover publications , mineola , new york .
goldsticker , r. ( 2007 ) a mutual fund to yield annuity - like benefits , _ financial analysts journal _ , vol .
63 ( 1 ) , pg .
63 - 67 .
hald , a. ( 2003 ) , _ history of probability and statistics and their applications before 1750 _ , wiley series in probability and statistics , john wiley & sons , new jersey homer , s. and r. sylla ( 2005 ) , _ a history of interest rates _ , 4th edition , john wiley and sons , new jersey .
jennings , r. m. and a. p. trout ( 1982 ) , _ the tontine : from the reign of louis xiv to the french revolutionary era _ ,
huebner foundation for insurance education , university of pennsylvania , 91 pages .
jennings , r. m. , d. f. swanson , and a. p. trout ( 1988 ) alexander hamilton s tontine proposal , _ the william and mary quarterly _ , vol .
45(1 ) , pg . 107 - 115 .
kopf , e. w. ( 1927 ) the early history of the annuity , _ proceedings of the casualty actuarial society _ , vol .
13(28 ) , pg . 225 - 266 .
lewin , c.g .
( 2003 ) , _ pensions and insurance before 1800 : a social history _ , tuckwell press , east
lothian , scotland .
mckeever , k. ( 2009 ) , a short history of tontines , _ fordham journal of corporate and financial law _ , vol .
15(2 ) , pg . 491 - 521 .
milevsky , m.a .
( 2014 ) , portfolio choice and longevity risk in the late 17th century : a re - examination of the first english tontine , _ financial history review _ , vol .
21(3 ) , pg .
225 - 258 milevsky , m.a .
( 2015 ) , _ king william s tontine : why the retirement annuity of the future should resemble its past _ , cambridge university press , new york city .
milevsky , m.a . and
salisbury ( 2015 ) , on the choice between tontines and annuities under stochastic and asymmetric mortality , _ manuscript in preparation _ moody s investor services ( 2013 ) , _
european insurers : solvency ii - volatility of regulatory ratios could have broad implications for european insurers _ , retrieved may 2013 from www.moodys.com odonnell , t. ( 1936 ) , _ history of life insurance in its formative years ; compiled from approved sources _ , american conservation co. , chicago .
pechter , k. ( 2007 ) , possible tontine revival raises worries , _ annuity market news _ , 1 may , sourcemedia .
piggott , j. , e. a. valdez , and b. detzel ( 2005 ) , the simple analytics of a pooled annuity fund , _ the journal of risk and insurance _ , vol .
72(3 ) , pg . 497 - 520 .
pitacco , e. , m. denuit , s. haberman , and a. olivieri ( 2009 ) , _ modeling longevity dynamics for pensions and annuity business _ oxford university press , uk .
poitras , g. ( 2000 ) , _ the early history of financial economics : 1478 - 1776 _ , edward elgar , cheltenham uk .
poterba , j.m .
( 2005 ) , annuities in early modern europe , in _ the origins of value : the financial innovations that created modern capital markets _ , edited by w. n. goetzmann and k. g. rouwenhorst .
new york : oxford university press .
promislow , s.d .
( 2011 ) , _ fundamentals of actuarial mathematics _ , 2nd edition , john wiley & sons , united kingdom .
qiao , c and m. sherris ( 2013 ) , managing systematic mortality risk with group self - pooling and annuitization schemes , _ journal of risk and insurance _ , vol .
80(4 ) , pg . 949 - 974 .
ransom , r. l. and r. sutch ( 1987 ) , tontine insurance and the armstrong investigation : a case of stifled innovation , 1868 - 1905 , _ the journal of economic history _ , vol .
47(2 ) , pg . 379 - 390 .
richter , a. , and f. weber ( 2011 ) , mortality - indexed annuities managing longevity risk via product design , _ north american actuarial journal _ ,
15(2 ) , pg . 212 - 236 .
rotemberg , j. j. ( 2009 ) , can a continuously - liquidating tontine ( or mutual inheritance fund ) succeed where immediate annuities have floundered ?
, _ harvard business school : working paper _ rothschild , c. ( 2009 ) , adverse selection in annuity markets : evidence from the british life annuity act of 1808 , _ journal of public economics _ ,
93(5 - 6 ) , pg .
776 - 784 .
sabin , m.j .
( 2010 ) , fair tontine annuity , _ ssrn abstract # 1579932 _ stamos , m. z. ( 2008 ) , optimal consumption and portfolio choice for pooled annuity funds , _ insurance : mathematics and economics _ , vol .
43 ( 1 ) , pg .
56 - 68 .
steele , j. m. ( 1997 ) , _ probability theory and combinatorial optimization _
, cbms - nsf regional conference series in applied mathematics , vol 69 , siam , philadelphia , pennsylvania tonti , l. ( 1654 ) , _ edict of the king for the creation of the society of the royal tontine _ , translated by v. gasseau - dryer , published in _ history of actuarial science _ , volume v , edited by s. haberman and t.a .
sibbett , published by william pickering , london , 1995 valdez , e. a. , j. piggott , and l. wang ( 2006 ) , demand and adverse selection in a pooled annuity fund , _ insurance : mathematics and economics _ , vol .
39(2 ) , pg . 251 - 266 .
weir , d. r. ( 1989 ) , tontines , public finance , and revolution in france and england , _ the journal of economic history _ , vol .
49 ( 1 ) , pg .
95 - 124 .
yaari , m. e. ( 1965 ) , uncertain lifetime , life insurance and the theory of the consumer , _ the review of economic studies _ , vol .
32(2 ) , pg . 137 - 150 .
the bulk of this section consists of proofs of results from section [ properties ] to simplify notation , we use as few subscripts as possible , and write @xmath119 , @xmath120 , or @xmath121 , as long as the meaning is clear .
we start with some calculations for the binomial distribution @xmath123 & = \sum_{k=0}^{n-1}\binom{n-1}{k}f(k+1)\big[kp^{k-1}(1-p)^{n - k-1}-(n - k-1)p^k(1-p)^{n - k-2}\big]\\ & = \sum_{k=1}^{n-1}\binom{n-1}{k}p^{k-1}(1-p)^{n - k-1}kf(k+1)\\ & \qquad\qquad - \sum_{k=0}^{n-2}\binom{n-1}{k+1}p^{k}(1-p)^{n - k-2}(k+1)f(k+1)\\ & = \frac{1}{p}\sum_{k=1}^{n-1}\binom{n-1}{k}p^k(1-p)^{n - k-1}k[f(k+1)-f(k)].\end{aligned}\ ] ] parts ( b ) and ( c ) are elementary calculations . to see ( a ) , apply lemma [ derivativeofmeans ] to obtain that @xmath126=\frac{1}{n^{\gamma-1}}\big(e[n^{\gamma-1}]+p\cdot\frac{1}{p}e[(n-1)(n^{\gamma-1}-(n-1)^{\gamma-1})]\big)\\ = \frac{1}{n^{\gamma-1}}e[n^\gamma-(n-1)^\gamma]>0.\end{gathered}\ ] ] the first statement of ( d ) holds , because @xmath127\to p(n = n)=p^{n-1}$ ] .
the second part of ( d ) follows from lemma [ reciprocalbound ] .
once this is established , the monotonicity of @xmath133 in @xmath1 immediately implies ( e ) , completing the proof of lemma [ betabound ] .
it is worth noting that for @xmath134 there is a simpler proof of ( e ) , since @xmath135\ge e\big[\frac{n(p)}{n}\big]^{\gamma-1}=\big(\frac{1+(n-1)p}{n}\big)^{\gamma-1}=\big(p+\frac{1-p}{n}\big)^{\gamma-1 } > p^{\gamma-1}\ ] ] by jensen , implying @xmath136 .
let @xmath137 .
define @xmath138 by @xmath139 . by lemma [ reciprocalbound ]
, @xmath138 is a sub - probability .
set @xmath140 . by holder , @xmath141 = e_q\big[\big(\frac{n(p)}{n}\big)^{\gamma_2}\big ] \le e_q\big[\big(\frac{n(p)}{n}\big)^{\gamma_2 b}\big]^{\frac{1}{b}}e_q[1]^{1-\frac{1}{b}}\\ < e_q\big[\big(\frac{n(p)}{n}\big)^{\gamma_1}\big]^{\gamma_2/\gamma_1 } = \big(pe\big[\big(\frac{n(p)}{n}\big)^{\gamma_1 - 1}\big]\big)^{\gamma_2/\gamma_1 } = \beta_{\gamma_1}(p)^{\gamma_2/\gamma_1}.\end{gathered}\ ] ] taking a @xmath142 power and dividing by @xmath54 now shows that @xmath143 . to show the monotonicity properties of @xmath133 with respect to @xmath54 , it suffices to show the same for @xmath144 . as in the proof of ( a ) of lemma [ betabound ] ,
@xmath145\\ = \frac{1}{\gamma\beta(p)n^{\gamma-1}}e\big[n(p)^\gamma-(n(p)-1)^\gamma-\gamma n(p)^{\gamma-1}\big].\end{gathered}\ ] ] by taylor s theorem , @xmath146 for some @xmath147 .
this is @xmath148 when @xmath132 and @xmath149 if @xmath131 , which completes the proof .
turning to conjecture , define @xmath150 so @xmath151 and @xmath152 . for @xmath138 as in the proof of lemma [ rlemma ] ,
an application of holder s inequality ( as in that proof ) shows that @xmath153^{1/\gamma_1}/e_q[(n(p)/n)^{\gamma_2}]^{1/\gamma_2}\ge 1 , \label{rbiggerthanone}\ ] ] when @xmath87 . there is a constant @xmath156 [ depending on @xmath0 , @xmath80 , @xmath81 , and the mortality parameters ] such that @xmath157 for each @xmath54
. therefore @xmath158\,ds.\ ] ] by , this @xmath159 in the limit as @xmath160 , so the integrand must take both positive and negative values .
by we have @xmath161 for each @xmath162 , so @xmath163 . therefore @xmath164 starts out negative , and then increases with @xmath162 , eventually becoming positive . by hypothesis ,
once it becomes positive , it stays so .
there is thus a @xmath86 such that the integrand is negative for @xmath165 and positive for @xmath166 . in other words
, @xmath167 is decreasing on @xmath168 $ ] and increasing on @xmath169 . since @xmath170 and @xmath171 when @xmath160 , we see that @xmath172 for every @xmath83 , as required . to show part ( b ) it suffices to show that @xmath176 is decreasing in @xmath54 .
this follows from ( a ) and ( d ) of lemma [ betabound ] , since @xmath177 .
part ( c ) is implicit in the proof of lemma [ rmonotonicity ] , since the hypothesis of that result is not used in showing that @xmath167 initially decreases .
numerical evidence ( not reported here ) suggests that there is a @xmath178 ( @xmath179 ) such that @xmath154 is decreasing in @xmath54 whenever @xmath180 . by lemma [ rmonotonicity ] , this would imply for these parameters , without any restriction on the underlying mortality distribution .
however , for larger @xmath1 values , @xmath154 can in fact increase over a small range of @xmath54 s .
for example , with @xmath181 , @xmath182 and @xmath183 , it increases over the range @xmath184 $ ] .
this defect in @xmath185 for very small @xmath54 does not appear significant enough to cause to fail under any reasonable mortality distribution .
but it is in fact possible to construct a pathological mortality distribution for which fails , by making @xmath5 spend essentially all of its time in the above range of @xmath54 s .
however , this distribution s hazard rate will not be monotone , since we achieve it by taking the hazard rate low over the above range of @xmath54 s , and high elsewhere .
asymptotics in @xmath0 are problematic , because of the need to estimate @xmath186 for @xmath54 very small .
we circumvent this difficulty by capping tontine payments at some advanced age @xmath187 . for a tontine running @xmath188 years
, the optimal payout @xmath189 will maximize @xmath190 over @xmath191 satisfying the constraint @xmath192 .
the arguments of corollary [ crraoptimum ] now show that @xmath193 we now set @xmath194 , and can prove the following asymptotic version of .
we start by establishing a number of estimates .
fix @xmath52 , @xmath200 , @xmath201 , and @xmath202 .
we may then find @xmath203 such that if @xmath204 and @xmath205 then @xmath206 .
recall azuma s inequality ( see for example steele ( 1997 ) ) , in the form that if @xmath207 then @xmath208 .
observe that @xmath209 , and set @xmath210 , @xmath211 , and @xmath212 .
it follows that if @xmath204 and @xmath205 then @xmath213 \le(1\lor n^{1-\gamma})p\big(\big|\frac{b - kp}{n}\big|>\eta p\big)\\ \le(1\lor n^{1-\gamma})p\big(\big|\frac{b}{k}-p\big|>\eta p\big ) \le
\frac{c_1}{n^{3/2}}. \label{firstbound}\end{gathered}\ ] ] likewise , if @xmath204 and @xmath205 then @xmath214\le \frac{c_1}{n^{3/2}}$ ] for each @xmath215 , so @xmath216= \begin{cases } 1+e_0 , & a=0\\ e_1 , & a=1\\ \frac{kp(1-p)}{n^2}+e_2 , & a=2 \end{cases } \label{secondbound}\ ] ] where in each case @xmath217 . by azuma s inequality
again , @xmath218 & = \int_0^\infty p\big(\big|\frac{n(p)}{n}-\mu\big|^3>q\big)\,dq = \int_0^\infty p\big(\big|\frac{b - kp}{n}\big|>q^{1/3}\big)\,dq\nonumber\\ & \le \int_0^\infty p\big(\big|\frac{b}{k}-p\big|>q^{1/3}\big)\,dq \le 2\int_0^\infty e^{-\frac12kq^{2/3}}\,dq \label{thirdbound}\\ & = \frac{6}{k^{3/2}}\int_0^\infty e^{-\frac12 z^2}z^2\,dz = \frac{3\sqrt{2\pi}}{k^{3/2}}.\nonumber\end{aligned}\ ] ] by taylor s theorem , @xmath219=e\big[\mu^{\gamma-1}+(\gamma-1)\mu^{\gamma-2}\big(\frac{n}{n}-\mu\big ) + \frac{(\gamma-1)(\gamma-2)}{2}\mu^{\gamma-3}\big(\frac{n}{n}-\mu\big)^2\\ + \frac{(\gamma-1)(\gamma-2)(\gamma-3)}{6}\xi^{\gamma-4}\big(\frac{n}{n}-\mu\big)^3 , \big|\frac{n}{n}-\mu\big|\le\eta p\big]\end{gathered}\ ] ] where @xmath147 lies between @xmath220 and @xmath221 . applying , , and yields a constant @xmath222 such that if @xmath204 and @xmath205 then @xmath223 = \mu^{\gamma-1}+\frac{(\gamma-1)(\gamma-2)}{2}\mu^{\gamma-3}\frac{kp(1-p)}{n^2}+e_4,\ ] ] where @xmath224 .
since @xmath212 it is now immediate that for these @xmath0 s and @xmath54 s , also @xmath223 = p^{\gamma-1}+\frac{\gamma(\gamma-1)(1-p)}{2n}p^{\gamma-2}+e_5=p^{\gamma-1}\big(1+\frac{\gamma(\gamma-1)(1-p)}{2np}+\frac{e_5}{p^{\gamma-1}}\big)\ ] ] where @xmath225 for some constant @xmath226 . by passing to some @xmath227 if necessary ,
we obtain finally that @xmath228\big)^{1/\gamma}\\ = p\big(1+\frac{(\gamma-1)(1-p)}{2np}+e_6\big ) = p\big(1-\frac{(\gamma-1)}{2n}\big)+\frac{\gamma-1}{2n}+pe_6 \label{fourthbound}\end{gathered}\ ] ] for @xmath229 and @xmath205 , where @xmath230 , and @xmath231 is a constant depending on @xmath1 and @xmath232 . to apply this , fix @xmath87 , set @xmath233 , and choose @xmath234 and @xmath231 to ensure that holds for both @xmath80 and @xmath81 . for any @xmath235 ,
the inequality @xmath197 is equivalent to the following : @xmath236 set @xmath237 and @xmath238 .
using , we have an upper bound of @xmath239\big[f(t)(1-\frac{(\gamma_2 - 1)}{2n}\big)+g(t)\frac{\gamma_2 - 1}{2n}\big]+\frac{c_5}{n^{3/2}}\ ] ] for the left side of ( for some constant @xmath240 ) , provided @xmath229 .
there is likewise a lower bound @xmath241\big[f(t)(1-\frac{(\gamma_1 - 1)}{2n}\big)+g(t)\frac{\gamma_1 - 1}{2n}\big]-\frac{c_6}{n^{3/2}}\ ] ] for the right side ( for some constant @xmath242 ) , again provided @xmath229 .
thus will hold for @xmath7 and @xmath0 , provided @xmath229 and @xmath243.\ ] ] it is easily seen that @xmath244 , and it follows that @xmath245 on @xmath246 .
it is therefore clear how to choose @xmath247 to ensure that the desired conclusion holds .
as remarked earlier , the case @xmath38 follows immediately from ( e ) of lemma [ betabound ] . when @xmath48 , we have @xmath40 so @xmath248\,dt$ ] , while @xmath249 . therefore @xmath250-\log({}_tp_x)\big)\,dt.\ ] ] the following result now completes the proof . by holder ,
@xmath252^{1/a}$ ] increases with @xmath253 . by lemma
[ reciprocalbound ] and lhpital s rule , @xmath254^{1/a } = e^{\lim_{a\downarrow 0}\frac{1}{a}\log e[(\frac{n}{n(p)})^a]}\\ & = e^{\lim_{a\downarrow 0}e[(\frac{n}{n(p)})^a\log(\frac{n}{n(p)})]/e[(\frac{n}{n(p)})^a ] } = e^{-e[\log(\frac{n(p)}{n})]}.\end{aligned}\ ] ] taking log s shows the result . since @xmath102 , @xmath255 is concave in @xmath256 , so lies below its tangents . therefore @xmath135
\le e\big[p^{\gamma-1}+(\gamma-1)p^{\gamma-2}\big(\frac{n}{n}-p\big)\big ] = p^{\gamma-1}+(\gamma-1)p^{\gamma-2}\frac{1-p}{n}.\ ] ] @xmath257 is also strictly concave in @xmath256 , so in the same way @xmath258 therefore @xmath259 by definition , @xmath260 and @xmath261 is also concave in @xmath256 , so as before @xmath262 for any @xmath1 one can derive ( using the second moment of @xmath263 now , as in the proof of theorem [ asymptotictheorem ] ) the asymptotic result that @xmath264 .
but convergence turns out to be so slow that this precise asymptotic is of limited use .
the slow convergence derives from the observation that , with gompertz mortality , the time @xmath7 till @xmath5 reaches @xmath265 grows only at rate @xmath266 .
( for the same reason , numerical computations are best carried out using @xmath267 rather than @xmath5 . ) for example , with @xmath268 , @xmath269 , age @xmath270 , and gompertz parameters @xmath271 and @xmath272 , we obtain @xmath273 ; but for @xmath274 10 , 100 , or 1000 we only have @xmath275 0.2858 , 0.3377 , and 0.3671 ; in fact , even with @xmath276 tontine participants ( roughly the world s entire population ) we would only reach @xmath277 . asymptotics become more reliable if tontine and annuity payments are capped at some advanced age @xmath187 , as in theorem [ asymptotictheorem ] .
the new loading @xmath278 now satisfies @xmath279 , where @xmath280 is the payout from the capped annuity . with the same parameters as above and capping at age 120 it still takes @xmath281 to give @xmath282 ( well approximating the @xmath283 value of @xmath284 ) .
convergence is more rapid with smaller @xmath188 s : with capping at age 110 we only need @xmath285 to give @xmath286 ( approximating 0.3850 ) and for age 100 capping , even @xmath287 yields 0.2855 ( approximating 0.2897 ) . in table
[ table08 ] we compare the welfare loss experienced by an individual with longevity risk aversion @xmath38 , if they participate in a natural tontine rather than an optimal one . to do
so we calculate the ratio @xmath288 of the certainty equivalents of the two tontines .
this represents the initial deposit into a natural tontine needed to provide the same utility as a $ 1 deposit into an optimal one . in this section
, we discuss this comparison further .
the natural tontine has utility @xmath289 therefore @xmath290^{\frac{1}{1-\gamma } } & = \left[\frac{d^{\text{ot}}_{n,\gamma}(1)^{-\gamma}}{c_0^{1-\gamma}\int_0^\infty e^{-rt}{}_tp_x^{2-\gamma}\theta_{n,\gamma}({}_tp_x)\,dt}\right]^{\frac{1}{1-\gamma}}\\ & = \frac{\big(\int_0^\infty e^{-rt}{}_tp_x\,dt\big)\big(\int_0^\infty e^{-rt}\beta_{n,\gamma}({}_tp_x)^{\frac{1}{\gamma}}\,dt)^{\frac{\gamma}{1-\gamma}}}{\big(\int_0^\infty e^{-rt}{}_tp_x^{2-\gamma}\theta_{n,\gamma}({}_tp_x)\,dt\big)^{\frac{1}{1-\gamma}}}.\end{aligned}\ ] ] when we compute these values numerically , for @xmath291 , we get values quite close to 1 . moreover , if we let @xmath88 , then @xmath292 and @xmath293 , which makes @xmath294 asymptotically , as long as @xmath295 . for @xmath296
this is not a fair comparison , as the integral diverges and @xmath297 .
it is worth understanding why .
the integrand in the numerator involves @xmath298^{1/\gamma}\sim p^{\frac{1}{\gamma}}n^{\frac{1-\gamma}{\gamma}}$ ] which is well behaved as @xmath299 .
the denominator integrand involves @xmath300 however , which can be large while @xmath299 , if @xmath296 .
this means that for @xmath296 the natural tontine utility is unduly influenced by the possibility of living to highly advanced ages .
even if there is only a single survivor , that survivor s payout will have dropped to quite low levels by the time it is actually improbable that anyone will live that long . and
for @xmath296 the negative consequences of the low payout dominate the small probability of surviving that long .
capping tontine payouts at an advanced age @xmath187 , as in theorem [ asymptotictheorem ] , would enable comparisons for @xmath296 .
for example , take age @xmath301 and let us compare with the certainty equivalent @xmath302 at that age from table [ table08 ] ( ie . with @xmath287 and @xmath268 ) .
capping payouts at age 100 , we can achieve @xmath303 for @xmath304 even with @xmath305 , and @xmath306 for @xmath307 by going to @xmath308 . with the capping age raised to 110 and @xmath304
, it instead takes @xmath309 to achieve a comparable @xmath303 lora ( @xmath310 ) & payout age 65 & payout age 80 & payout age 95 + 0.5 & 7.565% & 5.446% & 1.200% + 1.0 & 7.520% & 5.435% & 1.268% + 1.5 & 7.482% & 5.428% & 1.324% + 2.0 & 7.447% & 5.423% & 1.374% + 4.0 & 7.324% & 5.410% & 1.541% + 9.0 & 7.081% & 5.394% & 1.847% + * survival * & @xmath311100% & @xmath31272.2% & @xmath31316.8% + + + lora @xmath1 & @xmath314 & @xmath287 & @xmath315 & @xmath316 & @xmath317 + 0.5 & 72.6 b.p .
& 14.5 b.p . & 2.97 b.p . & 1.50 b.p . & 0.30 b.p .
+ 1.0 & 129.8 b.p . & 27.4 b.p .
& 5.74 b.p . & 2.92 b.p . &
+ 1.5 & 182.4 b.p . & 39.8 b.p . & 8.45 b.p .
& 4.31 b.p . & 0.89 b.p .
+ 2.0 & 231.7 b.p .
& 51.8 b.p . & 11.1 b.p . & 5.68 b.p . & 1.18 b.p .
+ 3.0 & 323.1 b.p . & 75.1 b.p . &
16.3 b.p . & 8.38 b.p . & 1.75 b.p .
+ 9.0 & 753.6 b.p . &
199.8 b.p . & 45.9 b.p . & 23.8 b.p . & 5.09 b.p .
+ + + age @xmath6 & @xmath318 & @xmath48 & @xmath268 + 30 & 1.000018 & 1 & 1.000215 + 40 & 1.000026 & 1 & 1.000753 + 50 & 1.000041 & 1 & 1.001674 + 60 & 1.000067 & 1 & 1.003388 + 70 & 1.000118 & 1 & 1.003451 + 80 & 1.000225 & 1 & 1.009877 + + tontine dividends during the first few decades of retirement is relatively low and predictable for a pool size in the hundreds .
the dividends increase exponentially at later ages and the 80% range is much wider as well . but this is not the only way to construct a tontine.,scaledwidth=90.0% ] utility starts off paying the exact same rate as a life annuity regardless of the number of participants in the tontine pool .
but , for higher levels of longevity risk aversion @xmath1 and a relatively smaller tontine pool , the function starts off at a lower value and declines at a slower rate , scaledwidth=90.0% ] of the optimal tontine payout to the natural tontine payout .
this is flat for @xmath48 since the natural tontine is optimal for logarithmic utility , but for rising @xmath1 we see increased reserving against the risk of living longer than expected.,scaledwidth=90.0% ] ) .
they agree at @xmath72 , but the tontine pays more at time zero once loading is included .
the utility of the loaded life annuity may be lower than that of the tontine , depending on the size of the loading .
, scaledwidth=90.0% ] | tontines were once a popular type of mortality - linked investment pool .
they promised enormous rewards to the last survivors at the expense of those died early . and , while this design _ appealed to the gambling instinct _ , it is a suboptimal way to generate retirement income .
indeed , actuarially - fair life annuities making constant payments where the insurance company is exposed to longevity risk induce greater lifetime utility .
however , tontines do not have to be structured the historical way , i.e. with a constant cash flow shared amongst a shrinking group of survivors . moreover
, insurance companies do not sell actuarially - fair life annuities , in part due to aggregate longevity risk .
we derive the tontine structure that maximizes lifetime utility . technically speaking we solve the euler - lagrange equation and examine its sensitivity to ( i. ) the size of the tontine pool @xmath0 , and ( ii . ) individual longevity risk aversion @xmath1 .
we examine how the optimal tontine varies with @xmath1 and @xmath0 , and prove some qualitative theorems about the optimal payout .
interestingly , lorenzo de tonti s original structure is optimal in the limit as longevity risk aversion @xmath2 .
we define the _ natural tontine _ as the function for which the payout declines in exact proportion to the survival probabilities , which we show is near - optimal for all @xmath1 and @xmath0 .
we conclude by comparing the utility of optimal tontines to the utility of loaded life annuities under reasonable demographic and economic conditions and find that the life annuity s advantage over the optimal tontine is minimal . in sum
, this paper s contribution is to ( i. ) rekindle a discussion about a retirement income product that has been long neglected , and ( ii . )
leverage economic theory as well as tools from mathematical finance to design the next generation of tontine annuities . |
we would like to talk about the virtual photon structure , especially on the mass effects in the polarized photon structure functions . in the last several years
, there has been much interest in the spin - dependent photon structure functions which can be studied in the polarized version of the @xmath8 collider or more directly in the polarized @xmath9 collision in the future linear collider ( figure.1 ) .
now let us consider the two - photon process in the polarized electron - positron collision where both of the photons are off - shell .
we particularly consider the case where the mass square of the probephoton , @xmath10 , is much bigger than that of the targetphoton , @xmath11 , which is in turn much bigger than the square of the qcd scale , @xmath12 .
the advantage for studying the virtual photon target in this kinematical region is that we can calculate whole structure functions @xmath6 and @xmath7 in perturbative qcd up to next - leading - order ( nlo ) , in contrast to the real photon target where there remain uncalculable non - perturbative pieces .
note that the 1st moment of @xmath6 is related to the axial anomaly just like the nucleon case , while the second structure function @xmath7 only exists for the virtual photon target ( @xmath13 ) .
now the possible mass effects are twofold ; target - mass effects and quark - mass effects . here in this
talk we discuss the former , which appear as power - corrections in @xmath5 . for the real photon target ( @xmath14 ) , there is no need to consider target mass corrections .
but when the target becomes off - shell and for relatively low values of @xmath10 , contributions suppressed by powers of @xmath15 may become important .
then we need to take into account these target mass contributions just like the case of the nucleon structure functions .
the consideration of target mass effects ( tme ) is important by another reason . for the virtual photon target ,
the maximal value of the bjorken variable @xmath4 is not 1 but @xmath16 due to the constraint @xmath17 , which is contrasted with the nucleon case where @xmath18 .
the structure functions should vanish at @xmath19 .
however , the nlo qcd result @xcite for @xmath20 ( see also the second paper of @xcite ) shows that the predicted graph does not vanish but remains finite at @xmath19 . in this talk
we discuss the tme for @xmath20 and @xmath21 in the framework of operator product expansion ( ope ) supplemented by the renormalization group ( rg ) method @xcite .
let us consider the structure tensor @xmath22 which is the absorptive part of the forward scattering amplitude @xmath23 ( figure 2 ) : @xmath24 for the target photon with mass squared @xmath25 probed by the photon with @xmath26 .
the antisymmetric part @xmath27 under @xmath28 and @xmath29 , can be decomposed as @xmath30 which gives two spin - dependent structure functions , @xmath31 and @xmath32 . when the target is real photon ( @xmath33 ) , @xmath7 is identically zero , and there exists only one spin structure function , @xmath34 . while , for the off - shell or virtual photon ( @xmath35 ) target , we have two spin - dependent structure functions @xmath6 and @xmath7 . the deep inelastic photon - photon scattering ( figure 2 . )
amplitude relevant for the polarized structure functions is given by @xmath36 for the product of the two electromagnetic currents we apply the ope and obtain for the @xmath37-@xmath38 antisymmetric part @xmath39 where @xmath40 and @xmath41 are the twist-2 and twist-3 operators , respectively , and are both traceless , and @xmath42 and @xmath43 are corresponding coefficient functions .
the twist-2 operators @xmath40 have totally symmetric lorentz indices @xmath44 , while the indices of twist-3 operators @xmath41 are totally symmetric among @xmath45 but antisymmetric under @xmath46 .
for the photon target we evaluate _ matrix elements _ " of the traceless operators @xmath40 and @xmath41 sandwiched by two photon states with momentum @xmath47 , which are written in the following forms : @xmath48\cdots\mu_{n-1}\}}~,\label{mattwist3}\end{aligned}\ ] ] where the subscript amp stands for the amputation of external photon lines , @xmath49 and @xmath50 are reduced photon matrix elements .
the traceless tensors @xmath51 and @xmath52\cdots\mu_{n-1}\}}$ ] satisfy the traceless conditions for @xmath53 : @xmath54 taking the of ( [ currentprodfourier ] ) with the virtual photon states , we obtain the deep - inelastic photon - photon forward scattering amplitude . the basic idea for treating target mass corrections
exactly is to take account of trace terms in the traceless tensors properly @xcite .
we evaluate the contraction between @xmath55 and the traceless tensors without neglecting any of the trace terms .
the results are expressed in terms of gegenbauer polynomials .
now we follow the same procedures as were taken by wandzura @xcite and in ref .
@xcite for the polarized nucleon case , and we obtain the analytic expression of the nachtmann moments @xcite for the twist-2 and twist-3 operators with definite spin @xmath56 . by writing down the dispersion relations for the amplitudes and with the use of orthogonality relations as well as an integration formula for gegenbauer polynomials @xmath57 , we project out @xmath58 and @xmath59 with definite spin @xmath56 , which still include the infinite series in powers of @xmath5 .
we then sum up those infinite series and express them in compact analytic forms @xcite .
then we obtain the nachtmann moments @xcite : @xmath60 \nonumber\\ & & \hspace{-0.5cm}\quad ( n=1,3,\cdots ) , \label{nacht2}\\ & & \hspace{-0.5cm}m_3^n \equiv \sum_i a_{(3)i}^{\gamma , n}e_{(3 ) i}^n(q^2,p^2,g)\nonumber\\ & & \hspace{-0.5cm}=\int_0^{x_{\rm max}}\frac{dx}{x^2}{\xi}^{n+1}\left [ \frac{x}{\xi}g_1^\gamma(x , q^2,p^2)\right.\nonumber\\ & & \hspace{-0.5cm}\left.+\left\{\frac{n}{n-1}\frac{x^2}{\xi^2}+ \frac{n}{n+1}\frac{p^2x^2}{q^2 } \right\}g_2^\gamma(x , q^2,p^2 ) \right ] \nonumber\\ & & \quad ( n=3,5 , \cdots),\end{aligned}\ ] ] where @xmath61 and @xmath62 , the so - called @xmath62-scaling variable @xcite , is given by @xmath63 the allowed range of @xmath62 is @xmath64 , and @xmath65 . in the nucleon case ,
the constraint @xmath66 gives @xmath67 and @xmath68 , leading to the problem of non - vanishing structure function at @xmath69 .
the resolution to this problem was argued in refs .
@xcite , by considering the dynamical higher - twist effects . @xmath70 and @xmath71 are perturbatively calculable .
in fact , the perturbative qcd calculation of @xmath70 has been done in lo @xcite and in nlo @xcite , while the qcd analysis of @xmath71 has been carried out in lo for the flavor non - singlet part in the limit of large @xmath72 @xcite .
once the moments @xmath70 and @xmath71 are known , we can derive @xmath31 and @xmath32 as functions of @xmath4 by inverting @xmath70 and @xmath71 as follows : @xmath73 where @xmath74 and we have introduced the following functions first discussed in ref .
@xcite : @xmath75 in figure 3 , we have shown @xmath31 with tme as a function of @xmath4 ( solid curve ) for @xmath76 gev@xmath77 ( upper ) with @xmath78 gev@xmath77 and for @xmath79 gev@xmath77 ( lower ) with @xmath78 gev@xmath77 . the vertical axis is in units of @xmath80 , where @xmath81 , the qed coupling constant , @xmath82 is the number of active flavors , and @xmath83 with @xmath84 being the electric charge of @xmath85th flavor quark .
also plotted are @xmath31 without tme ( dashed curve ) and the one with tme included up to the first order in @xmath5 ( short - dashed curve ) .
we observe that the target mass effects appear between intermediate @xmath4 and @xmath86 , and that the effects become sizable when the ratio @xmath5 is increased .
the distinction between the behaviors of @xmath6 with and without tme is remarkable near @xmath86 .
we get @xmath87 for @xmath76 gev@xmath77 with @xmath78 gev@xmath77 and @xmath88 for @xmath79 gev@xmath77 with @xmath78 gev@xmath77 .
the graphs of @xmath6 with tme vanish at @xmath86 as they should .
if tme is not taken into account , the polarized virtual photon structure function @xmath31 satisfies the following sum rule @xcite : @xmath89 note for the real photon target we have the vanishing sum rule @xcite .
the right - hand side corresponds to the twist-2 contribution , and actually the first term is the consequence of the qed axial anomaly .
now it will be interesting to see how this is modified when tme is included .
once the target mass corrections are taken into account , the above sum rule is modified to the first nachtmann moment , which reads @xmath90g_1^\gamma(x , q^2,p^2)\nonumber\\ & & \hspace{-0.5cm}+\frac{4}{3}\int_0^{x_{\rm max}}dx \frac{\xi^2}{x^2}\frac{p^2x^2}{q^2}g_2^\gamma(x , q^2,p^2)\nonumber\\ & & \hspace{-0.5 cm } = -\frac{3\alpha}{\pi}\sum_{i=1}^{n_f}e_i^4+{\cal o}(\alpha_s)~.\label{firstg1with}\end{aligned}\ ] ] the 1st moment of @xmath6 with tme for the nucleon target was discussed in refs .
the power - series expansion in @xmath5 gives the first order tme , the difference of lhs s of ( [ firstg1with ] ) and ( [ firstg1without ] ) : @xmath91 the 1st moment for @xmath7 without tme known as the burkhardt - cottingham sum rule @xcite : @xmath92 turns into the one with tme given by @xmath93 where the upper - limit of the integration has changed from 1 to @xmath86 .
to summarize we have studied the target mass effects in the virtual photon s spin structure functions , @xmath31 and @xmath1 , which can be measured in the future experiments of the polarized version of the @xmath8 or @xmath9 colliders .
the evaluation of kinematical target mass effects is considered to be important to extract dynamical higher - twist effects .
we have derived the expressions for @xmath0 and @xmath1 in closed form by inverting the nachtmann moments for the twist-2 and twist-3 operators .
our numerical analysis shows that the target mass effects appear at large @xmath4 and become sizable near @xmath94 , as the ratio @xmath5 increases .
here we have also examined the target mass effects for the first - moment sum rules of @xmath6 and @xmath7 . for the kinematic region
we consider , the corrections to the first moment of @xmath6 turn out to be negligibly small .
the first moment of @xmath7 leads to the burkhardt - cottingham sum rule , where only change exists in the upper limit of integration from 1 to @xmath86 .
there still remain two important subjects to be studied .
the first one is the quark - mass effects in @xmath6 and @xmath7 .
the heavy flavor contribution to @xmath95 for the nucleon has been explored in ref.@xcite .
it would be intriguing to investigate the polarized photon case .
another subject yet to be studied is the transition from real to virtual photon , especially the 1st moment sum rule ; how to reconcile the vanishing sum for the former with the non - vanishing one for the latter . | we discuss target mass effects in the polarized virtual photon structure functions @xmath0 , @xmath1 for the kinematic region @xmath2 , where @xmath3 is the mass squared of the probe ( target ) photon .
we obtain the expressions for the structure functions in closed form by inverting the nachtmann moments for the twist-2 and twist-3 operators .
numerical analysis shows that target mass effects appear at large @xmath4 and become sizable near the maximal value of @xmath4 , as the ratio @xmath5 increases .
target mass effects for the sum rules of @xmath6 and @xmath7 are also discussed . |
the investigation of relativistic cosmological models usually has the energy momentum tensor of matter generated by a perfect fluid . to consider more realistic models one must take into account the viscosity mechanisms , which have already attracted the attention of many researchers @xcite .
the nature of cosmological solutions for homogeneous bianchi type i ( bi ) model was investigated by belinsky and khalatnikov @xcite by taking into account dissipative process due to viscosity . in @xcite
we reinvestigate the problem posed in @xcite in presence of a @xmath0 term . though that murphy @xcite claimed that the introduction of bulk viscosity can avoid the initial singularity at finite past , but belinsky and khalatnikov @xcite showed that viscosity can not remove the cosmological singularity but results in a qualitatively new behavior of the solutions near singularity . to eliminate the initial singularities a self - consistent system of nonlinear spinor and bi gravitational field was considered by us in a series of papers @xcite . for some cases we were able to find field configurations those were always regular .
recently it was found that the introduction of a spinor field into the system may explain the late time acceleration of the universe , hence can be considered as an alternative to dark energy @xcite . given the importance of viscous fluid and spinor field to construct a more realistic model of the universe , we in @xcite we introduced spinor field into the system and solved the system for some special choice of viscosity .
the purpose of this paper is to study an interacting system of spinor and scalar fields within the scope of a bianchi type i cosmological model filled with viscous fluid in presence of a @xmath0 term and clarify the role of viscosity and field interaction in the evolution of the universe .
in this section we derive the fundamental equations for the interacting spinor , scalar and gravitational fields from the action and write their solutions in term of the volume scale @xmath1 defined bellow .
we also derive the equation for @xmath1 which plays the central role here .
we consider a system of nonlinear spinor , scalar and bi gravitational field in presence of perfect fluid given by the action @xmath3 with @xmath4 the gravitational part of the lagrangian is given by a bianchi type i ( bi hereafter ) space - time , whereas @xmath5 describes the interacting spinor and scalar field lagrangian and @xmath6 stands for the lagrangian density of viscous fluid . we choose the interacting spinor and scalar field lagrangian as @xmath7 - m{\bar \psi}{\psi}+ \frac{1}{2}{\varphi}_{,\alpha}{\varphi}^{,\alpha}(1 + \lambda f ) , \label{nlspin}\ ] ] here @xmath8 is the spinor mass , @xmath9 is the coupling constant and @xmath10 with @xmath11 and @xmath12 .
we would like to mention that there are 5 invariants constructed from bilinear spinor from .
using fierz transformation it can be shown that among the five invariants only @xmath13 and @xmath14 are independent as all other can be expressed by them : @xmath15 and @xmath16 @xcite . since the the bilinear identities appear to have been given in literature only partially and the relation between the invariants
are given as problem [ cf .
@xcite ] , we work them out in the appendix below .
therefore , the choice @xmath17 , describes the nonlinearity in the most general of its form @xcite .
note that setting @xmath18 in we come to the case with minimal coupling . as a gravitational field
we consider the bianchi type i ( bi ) cosmological model .
it is the simplest model of anisotropic universe that describes a homogeneous and spatially flat space - time and if filled with perfect fluid with the equation of state @xmath19 , it eventually evolves into a frw universe @xcite .
the isotropy of present - day universe makes bi model a prime candidate for studying the possible effects of an anisotropy in the early universe on modern - day data observations . in view of what has been mentioned above
we choose the gravitational part of the lagrangian in the form @xmath20 where @xmath21 is the scalar curvature , @xmath22 being the einstein s gravitational constant .
the gravitational field in our case is given by a bianchi type i ( bi ) metric @xmath23 with @xmath24 being the functions of time @xmath25 only . here the speed of light is taken to be unity .
let us now write the field equations corresponding to the action .
variation of with respect to spinor field @xmath26 gives spinor field equations [ speq ] @xmath27 where we denote @xmath28 varying with respect to scalar field we find @xmath29 variation of with respect to metric tensor @xmath30 gives the einstein s field equation .
for the bi space - time on account of the @xmath0 term this system has the form [ bid ] @xmath31 where over dot means differentiation with respect to @xmath25 and @xmath32 is the energy - momentum tensor of the material field given by @xmath33 here @xmath34 is the energy - momentum tensor of a viscous fluid having the form @xmath35 , \label{imper}\ ] ] where @xmath36 here @xmath2 is the energy density , @xmath37 - pressure , @xmath38 and @xmath39 are the coefficients of shear and bulk viscosity , respectively . in a comoving system of reference
such that @xmath40 we have @xmath41 in the eqs . and @xmath42 is the covariant derivatives acting on a spinor field as @xcite @xmath43 where @xmath44 are the fock - ivanenko spinor connection coefficients defined by @xmath45 for the metric one has the following components of the spinor connection coefficients @xmath46 the dirac matrices @xmath47 of curved space - time are connected with those of minkowski one as follows : @xmath48 with @xmath49 where @xmath50 are the pauli matrices : @xmath51 note that the @xmath52 and the @xmath53 matrices obey the following properties : @xmath54 where @xmath55 is the diagonal matrix , @xmath56 is the kronekar symbol and @xmath57 is the totally antisymmetric matrix with @xmath58 .
we study the space - independent solutions to the spinor and scalar field equations so that @xmath59 and @xmath60 . here
we define @xmath61 under this assumption from for the scalr field we find @xmath62^{-1 } dt , \qquad c = { \rm const . }
\label{sfsol}\ ] ] the spinor field equation in account of and takes the form @xmath63 setting @xmath64 from one deduces the following system of equations : [ v ] @xmath65 from we also write the equations for the invariants @xmath66 and @xmath67 [ inv ] @xmath68 where @xmath69 , and @xmath70 the eq . leads to the following relation @xmath71 giving the concrete form of @xmath72 from one writes the components of the spinor functions in explicitly and using the solutions obtained one can write the components of spinor current : @xmath73 the component @xmath74 @xmath75,\ ] ] defines the charge density of spinor field that has the following chronometric - invariant form @xmath76 the total charge of spinor field is defined as @xmath77 where @xmath78 is the volume . from the spin tensor
@xmath79 one finds chronometric invariant spin tensor @xmath80 and the projection of the spin vector on @xmath81 axis
@xmath82 let us now solve the einstein equations . to do it we first write the expressions for the components of the energy - momentum tensor explicitly : [ total ] @xmath83 in account of from we find the metric functions
@xcite [ abc ] @xmath84 , \label{a } \\
b(t ) & = & y_2\tau^{1/3}\exp \biggl[\frac{x_2}{3 } \int\,\frac{e^{-2 \kappa \int \eta dt } } { \tau ( t ) } dt \biggr ] , \label{b } \\
c(t ) & = & y_3\tau^{1/3}\exp \biggl[\frac{x_3}{3 } \int\,\frac{e^{-2 \kappa \int \eta dt}}{\tau ( t ) } dt \biggr ] , \label{c}\end{aligned}\ ] ] with the constants @xmath85 and @xmath86 obeying @xmath87 as one sees from , and , for @xmath88 with @xmath89 the exponent tends to unity at large @xmath25 , and the anisotropic model becomes isotropic one .
further we will investigate the existence of singularity ( singular point ) of the gravitational case , which can be done by investigating the invariant characteristics of the space - time . in general relativity
these invariants are composed from the curvature tensor and the metric one . in a 4d riemann space - time
there are 14 independent invariants .
instead of analyzing all 14 invariants , one can confine this study only in 3 , namely the scalar curvature @xmath90 , @xmath91 , and the kretschmann scalar @xmath92 . at any regular space - time point , these three invariants @xmath93 should be finite .
one can easily verify that @xmath94 thus we see that at any space - time point , where @xmath95 the invariants @xmath93 , as well as the scalar and spinor fields become infinity , hence the space - time becomes singular at this point . in
what follows , we write the equation for @xmath1 and study it in details .
summation of einstein equations , , and multiplied by 3 gives @xmath96 which can be rearranged as @xmath97 for the right - hand - side of to be a function of @xmath1 only , the solution to this equation is well - known @xcite .
on the other hand from bianchi identity @xmath98 one finds @xmath99 which in our case has the form @xmath100 this equation can be rewritten as @xmath101 recall that gives @xmath102 in view of that after a little manipulation from we obtain @xmath103 where @xmath104 is the thermal function .
let us now in analogy with hubble constant introduce the quantity @xmath105 , such that @xmath106 then and in account of can be rewritten as [ hve ] @xmath107 .
\label{ve}\end{aligned}\ ] ] thus , the metric functions are found explicitly in terms of @xmath1 and viscosity . to write @xmath1 and components of spinor field as well and scalar one we have to specify the function @xmath72 . in the next section we explicitly solve eqs . and for some concrete value of @xmath72 .
in this section we first solve the spinor field equations for some special choice of @xmath72 , which will be given in terms of @xmath1 .
thereafter , we will study the system in details and give explicit solution for some special cases .
as one sees , introduction of viscous fluid has no direct effect on the system of spinor field equations .
viscous fluid has an implicit influence on the system through @xmath1 .
a detailed analysis of the system in question can be found in @xcite .
here we just write the final results . here
we consider the case when the nonlinear spinor field is given by @xmath109 as in the case with minimal coupling from one finds @xmath110 for components of spinor field we find @xcite @xmath111 with @xmath112 being the integration constants and are related to @xmath113 as @xmath114 here @xmath115 .
for the components of the spin current from we find @xmath116,\quad j^1 = \frac{2}{a\tau } \bigl[c_{1 } c_{4 } + c_{2 } c_{3}\bigr ] \cos ( 2\beta ) , \nonumber \\ j^2 & = & \frac{2}{b\tau } \bigl[c_{1 } c_{4 } - c_{2 } c_{3}\bigr ] \sin ( 2\beta),\quad j^3 = \frac{2}{c\tau } \bigl[c_{1 } c_{3 } - c_{2 } c_{4}\bigr ] \cos ( 2\beta ) , \nonumber\end{aligned}\ ] ] whereas , for the projection of spin vectors on the @xmath117 , @xmath118 and @xmath119 axis we find @xmath120 total charge of the system in a volume @xmath121 in this case is @xmath122 { \mathcal}{v}.\ ] ] thus , for @xmath123 the components of spin current and the projection of spin vectors are singularity - free and the total charge of the system in a finite volume is always finite .
note that , setting @xmath18 , i.e. , @xmath124 in the foregoing expressions one get the results for the linear spinor field .
here we consider the case with @xmath126 in this case we assume the spinor field to be massless .
note that , in the unified nonlinear spinor theory of heisenberg , the massive term remains absent , and according to heisenberg , the particle mass should be obtained as a result of quantization of spinor prematter @xcite . in the nonlinear generalization of classical field equations ,
the massive term does not possess the significance that it possesses in the linear one , as it by no means defines total energy ( or mass ) of the nonlinear field system .
thus without losing the generality we can consider massless spinor field putting @xmath127 then from one gets @xmath128 in this case the spinor field components take the form @xmath129 the integration constants @xmath130 are connected to @xmath131 by @xmath132 here we set @xmath133 . for the components of the spin current from we find
@xmath134,\quad j^1 = \frac{4}{a\tau } \bigl[d_{2 } d_{3 } + d_{1 } d_{4}\bigr ] \cos ( 2 \sigma ) , \nonumber\\ j^2 & = & \frac{4}{b\tau } \bigl[d_{2 } d_{3 } - d_{1 } d_{4}\bigr ] \sin ( 2 \sigma),\quad j^3 = \frac{4}{c\tau } \bigl[d_{1 } d_{3 } - d_{2 } d_{4}\bigr ] \cos ( 2 \sigma ) , \nonumber\end{aligned}\ ] ] whereas , for the projection of spin vectors on the @xmath117 , @xmath118 and @xmath119 axis we find @xmath135 we see that for any nontrivial @xmath1 as in previous case the components of spin current and the projection of spin vectors are singularity - free and the total charge of the system in a finite volume is always finite .
in this subsection we simultaneously solve the system of equations for @xmath1 and @xmath2 . since setting @xmath136 in the equations for @xmath137
one comes to the case when @xmath125 , we consider the case with @xmath72 being the function of @xmath13 only .
let @xmath72 be the power function of @xmath138 , i.e. , @xmath139 . as it was established earlier , in this case @xmath140 , or setting @xmath141 simply @xmath142 . for simplicity we also set @xmath143 .
evaluating @xmath144 in terms of @xmath1 we then come to the following system of equations [ tauve ] @xmath145 , \label{vepn}\end{aligned}\ ] ] or in terms of @xmath105 [ hven ] @xmath146 , \label{hn}\\ \dot { { \varepsilon } } & = & 3 h\bigl(3 \xi h - \omega\bigr ) + 4 \eta \bigl(3 h^2 - { \varepsilon}- \lambda \bigr ) - 4 \eta \bigl [ \frac{m}{\tau } + \frac{\tau^{n-2}}{2(\lambda + \tau^n ) } \bigr ] .
\label{ven}\end{aligned}\ ] ] here @xmath38 and @xmath39 are the bulk and shear viscosity , respectively and they are both positively definite , i.e. , @xmath147 they may be either constant or function of time or energy .
we consider the case when @xmath148 with @xmath149 and @xmath150 being some positive quantities .
for @xmath37 we set as in perfect fluid , @xmath151 . \label{pzeta}\ ] ] note that in this case @xmath152 , since for dust pressure , hence temperature is zero , that results in vanishing viscosity . the system without spinor field have been extensively studied in literature either partially @xcite or as a whole @xcite . here
we try to solve the system for some particular choice of parameters .
let us first consider the case with bulk viscosity alone setting coefficient of shear viscosity @xmath153 . in this case from
and we find the following relation @xmath154 we also demand the coefficient of bulk viscosity be inverse proportional to expansion , i.e. , @xmath155 inserting @xmath153 , and into one finds @xmath156.\label{vecase1}\ ] ] then from we get the following equation for determining @xmath1 : @xmath157 \tau + \frac{3\kappa(1 - \zeta)}{2(1+\zeta)}\frac{c_2 \tau^{1+\zeta } - c_3}{\tau^\zeta } + \frac{3\kappa \lambda n}{4}\frac{\tau^{n-1}}{(\lambda + \tau^n)^2 } \equiv { { \mathcal}f}(q,\tau ) , \label{dettau1}\ ] ] where @xmath158 is the set of problem parameters . as one sees , the right hand side of the eq .
is a function of @xmath1 , hence can be solved in quadrature @xcite .
we solve the eq . numerically .
it can be noted that the eq . can be viewed as one describing the motion of a single particle .
sometimes it is useful to plot the potential of the corresponding equation which in this case is @xmath159 the problem parameters are chosen as follows : @xmath160 , @xmath161 , @xmath162 , @xmath163 , @xmath164 , @xmath165 and @xmath166 . here
we consider the cases with different @xmath0 , namely with @xmath167 , respectively .
the initial value of @xmath1 is taken to be a small one , whereas , the first derivative of @xmath1 , i.e. , @xmath168 at that point of time is calculated from . in fig . [ rfsspoten ]
we have illustrated the potential corresponding to eq . .
it can be immediately seen that independent of the sign of @xmath0 we have always expanding universe . but as is seen from fig .
[ rfsstauall ] a positive @xmath0 results in accelerated mode of expansion , while the negative one causes deceleration .
let us consider more general case .
following @xcite we choose the shear viscosity being proportional to the expansion , namely , @xmath169 in absence of spinor field this assumption leads to @xmath170 it can be shown that the relation in our case can be achieved only for massless spinor field with the nonlinear term being @xmath171 equation for @xmath1 in this case has the form @xmath172 \tau^2 = 0 .
\label{sbvtau}\ ] ] in case of @xmath173 there exists several special solutions available in handbooks on differential equations .
for that reason we rewrite this equation in terms of @xmath105 : @xmath174 . \label{newh}\ ] ] the solution of the foregoing equation can be written in quadrature as @xmath175 with @xmath176 , @xmath177 and @xmath178 .
if the bulk viscosity is taken to be a constant one , i.e. , @xmath179 , then depending on the value of the discriminant @xmath180 there exists three types of solutions , namely @xcite : @xmath181 note that a detailed analysis of these solutions in absence of spinor and scalar field was given in @xcite .
we choose the problem parameters as follows : @xmath182 , @xmath163 and @xmath183 . under this choice
we find @xmath184 .
after that we chose @xmath0 positive , trivial or negative ( in particular we chose @xmath185 ) .
the quantity @xmath39 now is taken such a way that we have @xmath186 for all values of @xmath0 chosen above , whereas , @xmath187 , @xmath188 and @xmath189 . in figs .
[ delnz ] , [ delnp ] and [ delnm ] we plot the evolution of @xmath1 corresponding to a trivial , positive and negative value of @xmath190 , respectively . as is seen , the behavior of @xmath1 mainly depends on @xmath190 and independent to the sign of @xmath0 .
since a negative @xmath190 gives non - periodic mode of evolution we plot the corresponding phase diagram in fig .
we consider the self consistent system of spinor , scalar and gravitational fields within the framework of bianchi type - i cosmological model filled with viscous fluid .
solutions to the corresponding equations are given in terms of the volume scale of the bi space - time , i.e. , in terms of @xmath191 .
the system of equations for determining @xmath1 , energy - density of the viscous fluid @xmath2 and hubble parameter @xmath105 has been worked out .
exact solution to the aforementioned system has been given only for some special choice of viscosity .
it should be noted that the system is far richer and allows a number of mathematically interesting results , though not all of them is physically realizable .
given this fact we plan to review this system and give a detailed analysis and qualitative solutions of the corresponding system in some of our future works .
let us now construct the invariants of spinor field . since @xmath192 and @xmath193 ( complex conjugate of @xmath192 ) has @xmath194 component each , one can construct @xmath195 independent bilinear combinations .
they are scalar , pseudoscalar , vector , axial vector , and tensor denoted , respective , by @xmath199 matrices in the above expressions obey the following algebra @xmath200 and are connected with the flat space - time dirac matrices @xmath52 in the following way @xmath201 where @xmath202 and @xmath203 is a set of tetrad 4-vectors . for the diagonal metric such as bianchi - i @xmath204
we have @xmath205 flat space - time matrices @xmath52 we will choose in the form : @xmath206 defining @xmath207 as follows , @xmath208 we obtain @xmath209 note that @xmath192 is a 4 component function given by , @xmath210 denoting @xmath211 bilinear spinor form in minkowski spacetime and @xmath72 in curve spacetime ( in our case in bi ) we find the following expressions for the non - trivial components of bilinear spinor form : @xmath212 using the above expressions we find @xmath213 .
\end{aligned}\ ] ] @xmath217 \nonumber \\ & = & ( \psi_1^*\psi_1)^2 + ( \psi_2^*\psi_2)^2 + ( \psi_3^*\psi_3)^2 + ( \psi_4^*\psi_4)^2 \nonumber\\ & + & ( \psi_1^*\psi_3)^2 + ( \psi_2^*\psi_4)^2 + ( \psi_3^*\psi_1)^2 + ( \psi_4^*\psi_2)^2 \nonumber\\ & + & 2[\psi_1^*\psi_1\psi_2^*\psi_2 - \psi_1^*\psi_1\psi_4^*\psi_4 - \psi_2^*\psi_2\psi_3^*\psi_3 + \psi_3^*\psi_3\psi_4^*\psi_4 \nonumber\\ & + & \psi_1^*\psi_4\psi_2^*\psi_3 + \psi_3^*\psi_2\psi_4^*\psi_1 - \psi_1^*\psi_3\psi_4^*\psi_2 - \psi_2^*\psi_4\psi_3^*\psi_1]\nonumber\\ & - & 4 ( \psi_1^*\psi_1\psi_3^*\psi_3 + \psi_2^*\psi_2\psi_4^*\psi_4 ) = 2(i_s - i_p).\end{aligned}\ ] ] an alternative proof of @xmath223 and @xmath224 can be given using feirz transformation .
in doing so we recall that any @xmath225 matrix @xmath226 can be presented as a linear combination of @xmath227 : @xmath228 explicitly it can be presented as @xmath229 here @xmath230 the completeness condition gives @xmath231 multiplying the foregoing expression by @xmath232 one gets @xmath233 replacing @xmath234 and @xmath235 and using the expression @xmath236 one gets the other identities . denoting
@xmath237 one finds @xcite @xmath238 after a little manipulations from the foregoing system one comes to desired relations between the invariants of the bilinear spinor fields . | we consider a self - consistent system of spinor and scalar fields within the framework of a bianchi type i gravitational field filled with viscous fluid in presence of a @xmath0 term .
exact self - consistent solutions to the corresponding spinor , scalar and bi gravitational field equations are obtained in terms of @xmath1 , where @xmath1 is the volume scale of bi universe .
system of equations for @xmath1 and @xmath2 , where @xmath2 is the energy of the viscous fluid , is deduced .
some special cases allowing exact solutions are thoroughly studied . |
the process of electromagnetic @xmath0 pair production in heavy - ion collisions plays an essential role in collider experiments .
it has a long history of experimental and theoretical investigations .
the process takes place in two different flavors dubbed as `` free - free '' and `` bound - free '' production , depending on whether the final electron is in the continuous spectrum or in the bound state with one of the nuclei . as for the free - free pair production ,
the pioneering papers @xcite appeared already in 1930-s and dealt with the high - energy asymptotics of the process . in late 1990-s the interest to the process has been revived due to the rhic experiment and approaching launch of the lhc experiment .
in particular , the contribution of the higher orders in the parameters @xmath2 ( the coulomb corrections ) in the high - energy limit has been discussed intensively , see refs . @xcite and the review @xcite .
the interest to the lepton pair production in collisions of slow nuclei appeared long ago in connection with the supercritical regime taking place when the total charge of the nuclei is large enough ( at least larger than 173 ) , see ref .
@xcite and references therein . recently , in ref
@xcite the total born cross section of the free - free pair production has been calculated exactly in the relative velocity @xmath1 of the colliding nuclei .
it turns out that the cross section is strongly suppressed as @xmath7 at @xmath8 .
a natural question arises whether such a suppression also holds for the coulomb corrections ( the higher terms in @xmath9 ) . in the present paper
we show that the coulomb corrections are less suppressed with respect to @xmath1 than the born term .
we assume that @xmath10 and @xmath11 and take into account the higher - order terms in @xmath9 amplified with respect to @xmath1 .
we consider both free - free and bound - free pair production . in the next section we perform calculations in the approximation in which both nuclei have constant velocities , i.e. , we treat the nuclei as infinitely heavy objects and neglect the coulomb interaction between them .
this approach has severe restrictions with respect to values of @xmath1 .
these restrictions are discussed in the third section together with the qualitative modification of the results in the region where the constant - velocity approximation is not valid .
let us first assume that the parameter @xmath12 is sufficiently small to be treated in the leading order . in particular , we assume that @xmath13 .
we neglect the coulomb interaction between the nuclei and work in the rest frame of the nucleus @xmath14 with @xmath15 axis directed along the momentum of the nucleus @xmath16 .
since our primary goal is the total cross section of the process , we find it convenient to use the eigenfunctions of angular momentum as a basis . in this basis
the cross section has the form @xmath17 where @xmath1 is the relative velocity of the nuclei , @xmath18 is the space components of the momentum transfer to nucleus @xmath16 , @xmath19 is the electron energy , @xmath20 is the electron mass , and the corresponding quantities with tildes are related to a positron .
the summation in eq .
( [ eq : sigmaff ] ) is performed over all discrete quantum numbers related to the states of both particles , i.e. , over the total angular momentum @xmath21 , its projection @xmath22 , and two possible values of @xmath23 , related to the parity of the state . within our accuracy ,
the matrix element @xmath22 reads @xmath24 here @xmath25 , @xmath26 is the electron wave function with the energy @xmath27 , the total angular momentum @xmath28 , and @xmath29 .
this wave function is the solution of the dirac equation in the attractive potential @xmath30 .
the function @xmath31 is the negative - energy solution of the dirac equation corresponding to the charge conjugation of the positron wave function , so that @xmath32 in the derivation of eq .
we have used the gauge in which the photon propagator has the form @xmath33 the limit @xmath8 is quite special . from kinematic constraints , it is easy to conclude ( cf .
@xcite ) that the characteristic momentum transfers to both nuclei are of the order of @xmath34 .
a simple estimate @xmath35 justifies using the small-@xmath36 asymptotics of the coulomb wave functions : @xmath37 where @xmath38 , @xmath39 is the spherical spinor , and the radial wave functions read @xmath40}\\ 1-\underline{\frac{r\left(\kappa-\gamma\right)}{\left(2\gamma+1\right)\eta_{b}}\left[\varepsilon\left(2\gamma+2\kappa+1\right)-m\right ] } \end{array}\right .
\label{eq : radialfunctions}\nonumber\\ \left.\begin{array}{c } \tilde{f}\left(r\right)\\ \tilde{g}\left(r\right ) \end{array}\right\ } & = \tilde{c}r^{\tilde{\gamma}-1 } \left\ { \begin{array}{c } -\frac{\tilde{\kappa}-\tilde{\gamma}}{\eta_{b}}+\underline{\frac{r}{2\tilde{\gamma}+1}\left[\tilde{\varepsilon}\left(2\tilde{\gamma}-2\tilde{\kappa}+1\right)+m\right]}\\ 1+\underline{\frac{r\left(\tilde{\kappa}-\tilde{\gamma}\right)}{\left(2\tilde{\gamma}+1\right)\eta_{b}}\left[\tilde{\varepsilon}\left(2\tilde{\gamma}+2\tilde{\kappa}+1\right)-m\right ] } \end{array}\right.\end{aligned}\ ] ] here @xmath41 let us assume for the moment that @xmath42 .
then the underlined terms can be safely neglected due to the estimate @xmath43 . moreover , due to the same estimate , the leading contribution to the sum in eq .
( [ eq : sigmaff ] ) is given by the terms with @xmath44 and @xmath45 .
if we also assume that @xmath46 , then only the contributions of two states with @xmath47 survive .
the underlined terms in eq .
become important for @xmath48 . in this region ,
in addition to the two states mentioned above , the states with @xmath49 equal to @xmath50 also should be taken into account . integrating over @xmath51 in eq . , substituting the result in eq .
( [ eq : sigmaff ] ) , and integrating over @xmath18 , we obtain the cross section @xmath52 of the free - free pair production : @xmath53\bigg\}\,.\label{eq : ffspectrum}\end{gathered}\ ] ] the relative order of the three terms in braces is regulated by the ratio @xmath54 .
when this ratio is small , the last term dominates .
this term coincides with the born result obtained in ref .
@xcite , as should be .
the parameter @xmath54 appears due to the `` accidental '' suppression of the born amplitude of pair production and has nothing to do with the sommerfeld - gamov - sakharov factor .
the bound - free pair production can be treated exactly in the same way as the free - free pair production .
it appears that an electron is produced mostly in @xmath55 states ( @xmath56 ) .
the positron spectrum reads @xmath57\bigg\}\,,\label{eq : bfspectrum}\end{gathered}\ ] ] where the riemann zeta function @xmath58 comes from summation over the principal quantum number .
it is quite remarkable that eq .
( [ eq : bfspectrum ] ) can be obtained from eq .
( [ eq : ffspectrum ] ) by the simple substitution @xmath59 followed by the replacement @xmath60 .
this substitution works because of the factorization of hard - scale @xmath43 and soft - scale @xmath61 contributions .
the total cross sections are obtained by the direct integration over energies ( energy ) ! ) are totally omitted in the annihilation current . ] : @xmath62 the main contribution to the integral is given by the region @xmath63 .
note the cancelation of the terms @xmath64 in braces for both free - free and bound - free cross sections . while this cancellation for the free - free case is a trivial consequence of the charge parity conservation , for the bound - free case it comes as a sort of surprise .
as it concerns the free - free pair production , the results and can be reproduced in a completely independent way .
namely , one can obtain the matrix element of the process in conventional diagrammatic technique taking into account the diagrams shown in fig . 1 . and calculating the contribution of the region where all coulomb exchanges have momenta @xmath65 .
pair production with the account of the first coulomb correction in @xmath66 . ]
this contribution gives the correct amplitude up to the coulomb phase which cancels in the cross sections and .
let us now assume that @xmath67 .
then the higher - order terms in @xmath12 should be treated on the same basis as those in @xmath68 . however , the account of these terms is not reduced to the substitution @xmath69 in eqs.([eq : ffspectrum ] ) , ( [ eq : bfspectrum ] ) , and ( [ eq : totalcrosssectionb ] ) .
speaking of the free - free pair production , the substitution @xmath69 should be taken into account on the level of matrix element , but not in the cross section .
the relative phase between the contributions @xmath70 and @xmath71 to the matrix element can be fixed from the diagrammatic approach mentioned above .
then we obtain @xmath72\bigg\}\,,\nonumber \\ \sigma_{ff } & = \frac{\eta_{a}^{2}\eta_{b}^{2}\beta^{6}}{1050\pi m^{2}}\bigg\{\pi^{2}\left(\eta_{a}+\eta_{b}\right)^{2}+\frac{592}{105}\beta^{2}\bigg\}\,.\end{aligned}\ ] ] for the bound - free pair production we have @xmath73\bigg\}\,,\nonumber\\ \sigma_{bf } & = \frac{16\left(\eta_{a}^{2}\eta_{b}^{5}+\eta_{b}^{2}\eta_{a}^{5}\right)\beta^{6}}{15015m^{2}}\zeta_{3}\left\ { \pi^{2}\left(\eta_{a}+\eta_{b}\right)^{2}+\frac{976}{153}\beta^{2}\right\ } \,.\end{aligned}\ ] ] note that @xmath74 for @xmath75 , in contrast to the statement in ref.@xcite .
it is interesting that in the supercritical case , @xmath76 , the relation between @xmath52 and @xmath77 is opposite ( see e.g. , ref.@xcite ) , since in this case at @xmath78 due to the energy conservation law the electron can be produced in the bound state but not in the free state .
the results and are obtained in the limit @xmath79 .
we show in this section that the account for the finite nuclear mass leads to an essential modification of both free - free and bound - free pair production cross sections at sufficiently small @xmath1 .
one of the sources , which restrict the applicability of eqs . and , is a deviation of the nuclear trajectories from the straight lines due to the coulomb interaction between the nuclei .
this deviation can be neglected if a shift of the minimal distance between the nuclei is smaller than the impact parameter @xmath80 : @xmath81 where @xmath82 and @xmath83 are the masses of the corresponding nuclei .
substituting @xmath84 we come to the constraint @xmath85 where @xmath86 is the proton mass and @xmath87 .
let us discuss qualitatively the modification of the cross section at @xmath88 .
first of all let us consider the dependence of the cross section @xmath89 on the impact parameter @xmath80 at @xmath90 .
similarly to the derivation of eq . , we obtain @xmath91 integrating over @xmath92 we obtain the first term in eq . .
if we consider the classical motion of the nuclei interacting by the coulomb field , we find that the minimal distance @xmath80 between the nuclei and the relative velocity @xmath1 at this point are expressed via the impact parameter @xmath93 and the relative velocity @xmath94 at infinity as @xmath95 then the parameter @xmath96 in eq .
can be written as @xmath97 calculating the minimum value of the quantity @xmath96 with respect to @xmath98 , we find that @xmath99 therefore , it follows from eq . that the cross section @xmath52 is exponentially small at @xmath100 .
the same conclusion is valid for the bound - free cross section .
suppose now that the condition holds .
since the results and are strongly suppressed by the factor @xmath101 , it is natural to ask whether the contributions formally suppressed with respect to @xmath102 may dominate at sufficiently small @xmath1 .
the answer is positive .
let us consider the `` bremsstrahlung '' mechanism of pair production , when the pair is produced by a virtual photon emitted by the scattered nucleus .
we consider the case @xmath103 when the motion of the nuclei is classical .
then the cross section of the @xmath104 pair production can be written as a product of the cross section @xmath105 of bremsstrahlung of virtual photon with the energy @xmath106 and the probability of virtual photon conversion into @xmath104 pair .
we assume that @xmath107 so that @xmath105 can be calculated in the non - relativistic dipole approximation .
we have @xcite @xmath108 where @xmath109 and the functions @xmath110 has the following asymptotic forms @xmath111 it is seen from eq . that @xmath112 is exponentially small for @xmath113 , which is in agreement with our previous statement . for @xmath114 , the main contribution to the integral over @xmath115
is given by the region @xmath116 .
then , taking into account that @xmath117 at @xmath118 , we obtain in the leading logarithmic approximation @xmath119 it is seen that the contribution to @xmath52 starts to dominate over very soon as @xmath1 decreases .
let us discuss our results .
we have calculated the infinite - mass limit of the free - free and bound - free pair production cross sections , eqs . and . as expected , the bound - free pair production cross section is much smaller than the free - free one , with the relative magnitude @xmath120 . in the region
@xmath121 both cross sections essentially deviate from the results obtained in the leading order in @xmath9 .
this is due to the `` accidental '' suppression of the born amplitude . in this connection
, it is interesting to compare eqs . and with the corresponding cross sections @xmath122 and @xmath123 for the production of scalar particles .
using the same technique we easily obtain @xmath124 the result for @xmath122 coincides with the asymptotics in eq .
( 17 ) of ref .
@xcite factor in the denominator . ] .
in contrast to the spinor case , the cross sections do not contain the terms of relative order @xmath125 since the leading - order contribution in @xmath9 is not suppressed by the power of @xmath1 anymore .
we note that the coulomb corrections in eqs . and are still more strongly suppressed in @xmath1 than the coulomb corrections to the corresponding cross sections for scalar particles , though the suppression is only @xmath126 , which is to be compared with @xmath127 for the ratio of the leading terms in @xmath9 .
finally , we have obtained the contribution of the bremsstrahlung mechanism which appears due to the account of the finite nuclear mass .
it turns out that this contribution starts to dominate very soon when @xmath1 decreases .
this severely restricts the region of applicability of the results and .
this work has been supported by russian science foundation ( project no .
14 - 50 - 00080 ) .
it has been also supported in part by rfbr ( grant no .
16 - 02 - 00103 ) . | we derive the spectrum and the total cross section of electromagnetic @xmath0 pair production in the collisions of two nuclei at low relative velocity @xmath1 . both free - free and bound - free @xmath0 pair production
is considered .
the parameters @xmath2 are assumed to be small compared to unity but arbitrary compared to @xmath1 ( @xmath3 are the charge numbers of the nuclei and @xmath4 is the fine structure constant ) . due to a suppression of the born term by high power of @xmath1 , the first coulomb correction to the amplitude appears to be important at @xmath5 .
the effect of a finite nuclear mass is discussed .
in contrast to the result obtained in the infinite nuclear mass limit , the terms @xmath6 are not suppressed by the high power of @xmath1 and may easily dominate at sufficiently small velocities . |
the gamma - ray burst ( grb ) 050525a is a bright brief flash of gamma - ray radiation detected by the _ swift _
burst alert telescope ( bat ) on 25 may 2005 00:02:53 ( ut ) @xcite .
it showed two peaks , with a duration of @xmath5 in the 15 - 350 kev band @xcite .
fitting to the band spectral model yields a low - energy photon index of @xmath6 and a peak energy of @xmath7 kev @xcite .
the fluence in the 15 - 350 kev band is @xmath8 .
this burst was also detected by other on - board instruments such as integral and konus - wind , with peak fluxes of @xmath9 @xcite and @xmath10 @xcite , respectively .
the fluences were @xmath11 in 20 - 200 kev ( for the 12 second integration time ) by integral and @xmath12 by konus - wind in 20 - 1000 kev ( for the 11.5 second integration time ) , respectively .
the time - integrated spectrum shows a peak energy @xmath13 by konus - wind @xcite .
about six minutes later , the rotse iii telescope in namibia was able to obtain images , and detected an optical afterglow of magnitude 14.7 @xcite .
a presumed host galaxy was measured with a redshift of z=0.606 , based on both [ o iii ] 5007 and @xmath14 emission and ca h@xmath15k and ca i 4228 absorption @xcite .
assuming cosmological parameters
@xmath16 , @xmath17 , and @xmath18 , the isotropic - equivalent gamma - ray energy is about @xmath19 .
the follow - up observations of grb 050525a revealed that the optical afterglow decayed as @xmath20 until about 0.01 days after the burst .
subsequently the afterglow showed a major rebrightening starting at @xmath0 days and ending at @xmath1 days , and then decayed as @xmath21 .
this early - time emission feature is similar to the behavior of external reverse shock emission predicted by the standard fireball - shock model @xcite , which was first confirmed by the detections of early prompt optical emission from grb 990123 @xcite .
the reverse shock model was also suggested to interpret early optical emission of grb 021004 @xcite and grb 021211 @xcite .
besides , the prompt optical - infrared emission from grb 041219a was recently suggested to be the internal shock emission , since a strong correlation between @xmath22-ray and optical - infrared signals was detected @xcite , which further confirmed the standard internal - external - shock model of grbs .
mirabal et al .
( 2005 ) reported a break at @xmath23 days in the optical afterglow light curve of grb 050525a . in this paper , we show that the optical afterglow in @xmath23 days after the _ swift _ trigger can be well understood due to the reverse shock emission superposed by the forward shock emission .
the standard model for grbs and their afterglows is the fireball - shock model ( for recent review articles , see mszros 2002 ; zhang & mszros 2004 ; piran 2005 ) .
in this model , the prompt emission of grbs is ascribed to internal shocks and the long - term afterglow to external shocks .
the prompt or very early optical emission is therefore explained as a consequence of the reverse component of the external shocks @xcite . for shock - accelerated slow - cooling electrons with a power - law distribution ,
the typical spectrum of synchrotron emission is described by three power laws @xcite : ( 1 ) @xmath24 for @xmath25 ; ( 2 ) @xmath26 for @xmath27 ; ( 3 ) @xmath28 for @xmath29 , where @xmath30 , @xmath31 and @xmath32 are the typical synchrotron frequency , cooling frequency and peak flux , respectively .
the parameter @xmath33 is the index of the electron energy distribution .
here we ignore the synchrotron self - absorption because its corresponding frequency may be much less than the optical band of interest .
both a forward shock and a reverse shock emerge when an ultrarelativistic cold grb ejecta with initial lorentz factor of @xmath34 sweeps up a stationary cold interstellar medium : the forward shock propagates into the interstellar medium and the reverse shock propagates back into the ejecta @xcite .
the emission of the forward shock is characterized by @xcite @xmath35 @xmath36 @xmath37 @xmath38 where , @xmath39 is the critical times when the break frequency , @xmath40 crosses the observed frequency @xmath41 , @xmath42 is the fraction of the shock energy goes into the magnetic field , @xmath43 is the fraction of the shock energy goes into the electrons , @xmath44 , @xmath45 , @xmath46 @xcite , @xmath47 is the isotropic - equivalent kinetic energy , @xmath48 is the density of interstellar medium in units of 1 @xmath49 , @xmath50 is the redshift of the burst , @xmath51 is the corresponding luminosity distance , and @xmath52 is the observer s time in units of 1 day .
the above equations are valid for @xmath53
. afterglows with the hard electron spectrum of @xmath54 have been discussed by dai & cheng ( 2001 ) . for reverse shocks
, it is possible to get a simple analytic solution in two limiting cases : thin shell and thick shell ( which are corresponding to newtonian and relativistic reverse shock , respectively ; sari @xmath15 piran , 1995 ) .
since the reverse - shocked gas is separated with the forward - shocked gas by a contact discontinuity , which keeps the equality of pressures and velocities in both shocks , we can find the emission properties of reverse shock in the aid of the correlations between forward and reverse shocks at the crossing time @xmath55 @xcite .
first , since the reverse- and forward - shocked gases have the same magnetic field and lorentz factor at @xmath55 , the cooling frequency of the reverse - shocked electrons is equal to that of the forward - shocked electrons , @xmath56 .
second , the electron s random lorentz factor of the reverse shock is smaller than that of the forward shock by a factor of @xmath57 ( where @xmath58 is the lorentz factor of both shocked regions at @xmath55 ) , the typical synchrotron frequency of the reverse shock is by a factor of @xmath59 less than that of the forward shock , @xmath60 .
this is basically valid for the thick shell case .
as for a thin shell , it is more realistic with the expression , @xmath61 , where @xmath62 is the lorentz factor of the shocked ejecta in the rest frame comoving with the unshocked ejecta at the crossing time and @xmath63 @xcite . third , the peak flux @xmath64 of an emission region is proportional to the electron number , the magnetic field , and the lorentz boost . since at @xmath55 ,
the electron number of the ejected shell is larger than that of the swept - up medium by a factor of @xmath65 , we obtain @xmath66 .
therefore , we can directly write down the corresponding characteristic variables of the reverse shock emission , at @xmath67 , @xmath68 @xmath69 @xmath70 for the thin shell ( i.e. newtonian ) case , and @xmath71 @xmath72 @xmath73 for the thick shell ( i.e. relativistic ) case . here , the grb duration t is given by the shell width @xmath74 , according to the internal shock model .
as derived by sari & piran ( 1999a ) and kobayashi ( 2000 ) , the typical synchrotron frequency and peak flux evolve as @xmath75 , @xmath76 at @xmath77 , and @xmath78 , @xmath79 at @xmath80 , for the thin - shell case ; @xmath81constant , @xmath82 at @xmath83 , and @xmath84 , @xmath85 at @xmath86 , for the thick - shell case . before the crossing time
, the cooling frequency evolve as @xmath87 for the thin - shell case and @xmath88 for the thick - shell case ; after the crossing time , @xmath89 turns into the cutoff frequency due to synchrotron cooling of electrons without acceleration , and it evolves the same as @xmath90 for both cases . according to the emission properties shown above , we derive light curves of the emission from the forward- and reverse - shocked regions and thus can constrain the model parameters by fitting the early afterglow of grb 050525a .
from the emission features described by equations ( 1 ) , ( 2 ) , ( 5 ) , ( 6 ) , ( 8) , and ( 9 ) , we see that the optical band is commonly between the typical frequency @xmath30 and cooling frequency @xmath31 in both forward and reverse shocks around the crossing time .
thus , the optical light curve of grb 050525a can be fitted by the superposition of two broken power - laws : ( 1 ) forward shock : @xmath91 for @xmath92 , and @xmath93 for @xmath94 ; ( 2 ) reverse shock : @xmath95 for @xmath96 , and @xmath97 for @xmath98 .
these scaling laws are valid for the thin - shell case , in which the reverse shock decelerates the shell insignificantly @xcite . in figure [ fig1 ] we present the r - band afterglow light curve of grb 050525a between may 25.0062 ut ( @xmath99 ) and may 25.5333 ut ( @xmath100 ) .
we find that , @xmath101 , @xmath102 , and @xmath103 can fit the light curve well for the component of forward shock emission .
given redshift @xmath104 @xcite , the corresponding luminosity distance is @xmath105 . using equations ( 3 ) and ( 4 ) ,
we obtain @xmath106 and @xmath107 furthermore , from the flux inferred by the light curve of the reverse - shock emission ( depicted as a dashed line in figure [ fig1 ] , where we see a flux @xmath108 at @xmath109 ) , using equations ( 5 ) and ( 7 ) , we further obtain @xmath110 by simple algebra from equations ( 11 ) , ( 12 ) and ( 13 ) , we obtain a constraint on the initial lorentz factor of grb ejecta , @xmath111 which is weakly dependent on the isotropic - equivalent kinetic energy @xmath47 and interstellar medium density @xmath48 . assuming the efficiency of energy conversion is @xmath112 , @xmath47 is approximately equal to the detected isotropic - equivalent @xmath22-ray energy , i.e. @xmath113 .
thus , we can obtain a mutual constraint between @xmath42 and @xmath43 from equations ( 11 ) and ( 12 ) . by eliminating the medium density @xmath48
, we have @xmath114 .
moreover , as a natural consequence of @xmath115 , we have @xmath116 , @xmath3 , and @xmath4 .
these constraints are more clear than the cases of grb 990123 @xcite , grb 021004 @xcite and grb 021211 @xcite .
our preliminary fitting to the afterglow observations of grb 050525a update to may 25.5333 ut ( @xmath100 ) is shown in figure [ fig1 ] .
additionally , we also give a constraint on the parameters for the thick - shell case .
we can find from a fitting similar to that shown above that , the crossing time of the reverse shock , i.e. , the grb duration @xmath117 is approximated by @xmath118 which is inconsistent with the observed @xmath119 @xcite .
this implies that grb 050525a seems to be a thin - shell case .
the optical afterglow of grb 050525a rebrightened starting at @xmath0 days and ending at @xmath1 days , which was followed by an initial power - law decay . in this paper , we have shown that this early emission feature can be interpreted as the reverse shock emission superposed by the forward shock emission in an interstellar medium environment . a good fitting to the observed light curve is shown in figure 1 .
we further find the initial lorentz factor of the ejecta @xmath2 .
some other model parameters are also constrained : the magnetic energy fraction @xmath3 and the medium density @xmath4 .
these limits are consistent with those from the other very - early optical afterglows observed so far @xcite and with the values of the forward - shock parameters @xcite .
our model is simplified under some assumptions : first , the reverse- and forward - shocked regions have the same electron and magnetic energy fractions , requiring that the ejecta is not initially magnetized .
this requirement seems to be relaxed for grb 990123 @xcite .
the interaction of a magnetized ejecta with its surrounding medium has been investigated in details @xcite .
second , we neglect the effect of inverse compton scattering ( ics ) on the reverse shock emission in 2 .
as shown by wang , dai , & lu ( 2001a , b ) , this effect not only decreases the cooling frequency but also produces higher - energy emission .
however , this decrease does not affect the limits derived above because the cooling frequency is much larger than the optical band at the times of our interest .
third , the surrounding environment is assumed to be a uniform interstellar medium , in which case two types of light curves are expected for the reverse - forward - shock emission combination : a rebrightening light curve ( type i ) and a flattening light curve ( type ii ) @xcite .
early afterglows in wind environments have been studied by chevalier , & li ( 2000 ) , wu et al .
( 2003 ) , kobayashi & zhang ( 2003b ) , panaitescu & kumar ( 2004b ) , and zou et al .
generally , a flattening light curve in optical band is expected in the wind model @xcite , i.e. , there is only type ii light curve for the reverse - forward - shock emission combination . since the early optical afterglow of grb 050525a shows a rebrightening ( type i ) light curve , the reverse - forward shock model in a wind environment for the early afterglow of grb 050525a is not favored .
the observed break at @xmath120 days in the optical afterglow light curve of grb 050525a @xcite may be due to an ultrarelativistic jet .
this conclusion seems to be strengthened by lulin observatory around 17.7 hours after the burst ( at which the r - band magnitude is about @xmath121 , chiang et al .
if so , the jet s half opening angle @xmath122 ( sari , piran , & halpern 1999 ) and thus the beaming - corrected gamma - ray energy @xmath123 ergs , which is by a factor of a few smaller than the mean energy release found by frail et al .
( 2001 ) for @xmath124 and @xmath125 .
even so , this energy still satisfies the relation of ghirlanda et al .
( 2004 ) , @xmath126^{1.50\pm 0.08}$ ] ( see dai , liang , & xu 2004 ) . finally , what we want to point out is that besides the reverse - shock model discussed in this paper , the other plausible explanations of a rebrightening light curve include the density - jump medium ( dai & lu 2002 ; lazzati et al .
2002 ) , pure poynting - flux injection ( dai & lu 1998 ) , baryon - dominated injection ( rees & mszros 1998 ; granot , nakar & piran 2003 ; nakar , piran & granot 2003 ; bjrnsson et al .
2004 ) , two - component jet ( berger et al . 2003 ; huang et al . 2003 ) , neutron - fed fireball ( beloborodov 2003 ) , temporal fluctuation of @xmath43 and/or @xmath42 and departure of the electron distribution from a power law ( panaitescu & kumar 2004 ) .
multiwavelength observations in the _ swift _ era are expected to distinguish among these possibilities .
we thank y. z. fan and l. j. gou for constructive comments , x. f. wu and y. c. zou for helpful discussions , and y. f. huang and x. y. wang for a careful reading of our manuscript .
this work was supported by the national natural science foundation of china ( grants 10233010 and 10221001 ) and the ministry of science and technology of china ( nkbrsf g19990754 ) .
akerlof , c. , et al .
1999 , , 398,400 band , d. , et al .
2005 , gcn 3466 beloborodov , a. m. 2003 , , 585 , l19 berger , e. , et al .
2003 , , 426 , 154 bjrnsson , g. , gudmundusson , e. h. , & johanesson , g. 2004 , , 615 , l77 blake , c. h. , et al .
2005 , , 435 , 181 chevalier , r. a. , & li , z. y. 2000 , , 536 , 195 chiang , p. s. , huang , k. y. , ip , w. h. , urata , y. , qiu , y. , & lou , y. q. 2005 , gcn 3486 cobb , b. e. , & bailyn , c. d. 2005 , gcn 3506 cummings , j. , et al . 2005 , gcn 3479 dai , z. g. , & cheng , k. s. 2001 , , 558 , l109 dai , z. g. , & lu , t. 1998 , , 81 , 4301 dai , z. g. , & lu , t. 2002 , , 565 , l87 dai , z. g. , liang , e. w. , & xu , d. 2004 , , 612 , l101 fan , y. z. , dai , z. g. , huang , y. f. , & lu , t. 2002 , , 2 , 449 fan , y. z. , wei , d. m. , & wang , c. f. 2004 , , 424 , 477 fan , y. z. , & wei , d. m. 2005 , astro - ph/0506155 foley , r. j. , chen , h. w. , bloom , j. , & prochaska , j. x. 2005 , gcn 3483 fox , d. w. , et al .
2003a , , 422 , 284 fox , d. w. , et al .
2003b , , 586 , l5 frail , d. a. , et al .
2001 , apj , 562 , l55 galama , t. j. , et al . 1999 , , 398 , 394 ghirlanda , g. , ghisellini , g. , & lazzati , d. 2004 , , 616 , 331 golenetskii , s. , et al .
2005 , gcn 3474 gotz , d. , et al .
2005 , gcn 3472 granot , j. , & sari , r. 2002 , , 568 , 820 granot , j. , nakar , e. & piran , t. 2003 , , 426 , 138 homewood , a. , et al .
2005 , gcn 3491 huang , y. f. , wu , x. f. , dai , z. g. , ma , h. t. , & lu , t. 2004 , , 605 , 300 katz , j. i. 1994 , , 422 , 248 klotz , a. , boer , m. , & atteia , j. l. 2005 , gcn 3473 klotz , a. , et al . 2005 , , submitted ( astro - ph/0506259 ) kobayashi , s. 2000 , , 545 , 807 kobayashi , s. , & zhang , b. 2003a , , 582 , l75 kobayashi , s. , & zhang , b. 2003b , , 597 , 455 lazzati , d. , rossi , e. , covino , s. , ghisellini , g. , & malesani , d. 2002 , , 396 , l5 li , w. , et al .
2003 , , 586 , l9 malesani , d. , et al .
2005 , gcn 3469 markwardt , c. , et al .
2005 , gcn 3467 mszros , p. , & rees , m. j. 1997 , 476 , 232 mszros , p. , & rees , m. j. 1999 , 306 , l39 mszros , p. 2002
, , 40 , 137 milne , p. a. , williams , g. g. , & park , h. s. 2005 , gcn 3485 mirabal , n. , bonfield , d. , & schawinski , k. 2005 , gcn 3488 nakar , e. , piran , t. , & granot , j. 2003 , , 8 , 495 nakar , e. , & piran , t. 2005 , , 619 , l147 panaitescu , a. , & kumar , p. 2001
, , 560 , l49 panaitescu , a. , & kumar , p. 2004a
, , 350 , 213 panaitescu , a. , & kumar , p. 2004b
, , 353 , 511 piran , t. 2005 , rev . mod .
, 76 , 1143 rees , m. , & mszros , p. 1998
, , 496 , 1 rykoff , e. s. , yost , s. a. , & swan , h. 2005 , gcn 3465 rykoff , e. s. , yost , s. a. , swan , h. , & quimby , r. 2005 , gcn 3468 sari , r. , & piran , t. 1995 , , 455 , l143 sari , r. , piran , t. , & narayan , r. 1998 , , 497 , l17 sari , r. , & piran , t. 1999a , , 520 , 641 sari , r. , & piran , t. 1999b , , 517 , l109 sari , r. , piran , t. , & halpern , j. p. 1999 , apj , 519 , l17 torii , k. , & bendaniel , m. 2005 , gcn 3470 vestrand , w. t. et al .
2005 , , 435 , 178 wang , x. y. , dai , z. g. , & lu , t. 2001a , , 546 , l33 wang , x. y. , dai , z. g. , & lu , t. 2001b , , 556 , 1010 wei , d. m. 2003 , , 402 , l9 wu , x. f. , dai , z. g. , huang , y. f. , & lu , t. 2003 , , 342 , 1131 yanagisawa , k. , toda , h. , & kawai , n. 2005 , gcn 3489 zhang , b. , kobayashi , s. , & mszros , p. 2003
, , 595 , 950 zhang , b. , & mszros , p. 2004 , int .
a , 19 , 2385 zhang , b , & kobayashi , s. 2005 , , in press ( astro - ph/0404140 ) zou , y. c. , wu , x. f. , & dai , z. g. 2005 , , submitted | the prompt localization of gamma - ray burst ( grb ) 050525a by _ swift _ allowed the rapid follow - up of the afterglow .
the observations revealed that the optical afterglow had a major rebrightening starting at @xmath0 days and ending at @xmath1 days , which was followed by an initial power - law decay .
here we show that this early emission feature can be interpreted as the reverse shock emission superposed by the forward shock emission in an interstellar medium environment . by fitting the observed data
, we further constrain some parameters of the standard fireball - shock model : the initial lorentz factor of the ejecta @xmath2 , the magnetic energy fraction @xmath3 , and the medium density @xmath4 .
these limits are consistent with those from the other very - early optical afterglows observed so far . in principle , a wind environment for grb 050525a is disfavored . |
there is great current interest in quantum and nanoscale systems that convert heat into electric current @xcite .
the simplest such systems are those that exhibit a thermoelectric effect @xcite : they do the conversion in a steady - state ( dc ) manner , and so avoid the need for pumping cycles relying on time - dependent couplings .
quantum dots are particularly promising in this respect , and various applications have been proposed and ( at least partially ) realized experimentally , including thermoelectric engines @xcite , refrigerators @xcite , thermal rectifiers @xcite , and hybrid refrigerator power - sources @xcite . in addition , the simplicity of these thermoelectric quantum dots makes them ideal model systems for the study of quantum thermodynamics .
the most developed theory of non - equilibrium thermodynamics , known as irreversible thermodynamics @xcite , assumes that the shortest lengthscale in the system is that on which the particles thermalize ( the inelastic scattering length ) .
then , any system coupled between two reservoirs at very different temperatures has a well - defined local temperature everywhere within it . however , as we reduce the size of the circuit elements into the nanoscale regime , this ceases to be the case . a system that is much smaller than the electron thermalization length will be in a _ non - thermal state _ whenever it is coupled to reservoirs that are at significantly different temperatures or electrochemical potentials . in other words ,
the distribution of the excitations in the system can not be described in terms of a thermal distribution . in general
, we wish to answer the question : what new physics can emerge in a quantum system operating in this non - thermal regime ?
this work presents a first response to this question , in the context of the thermoelectric quantum dot system sketched in fig .
[ fig : system](a ) . this can be considered as a miniaturized version of the usual macroscopic thermocouple power generator , in which the macroscopic metal reservoir between the two thermoelectrics has been replaced by the quantum dot m. our system is similar to three - terminal energy harvesters considered in refs .
@xcite , which separate the conductor from a heats source with which it exchanges energy but no particles .
see also refs .
@xcite , which consider related models with bosonic heat sources .
however , instead of the dot being coupled only to the heat source ( reservoir h ) , we consider the case when it is also coupled to the cold environment ( reservoir c ) .
since the environment coupling is never negligible , we argue that this is the generic case .
it is certainly the case in conventional thermocouples , where it is well known that the region between the two thermoelectrics is not as hot as the heat - source , because it also exchanges heat with the cold environment . with this in mind
, the simplest case would be when the temperature of the cold environment @xmath1 is the same as the temperature @xmath2 of the electrical circuit in which the thermocouple is inserted ( reservoirs l and r ) .
however , we consider the more general case where @xmath3 is different from @xmath4 , for example due to joule heating in the wires or load ( @xmath5 ) . following what is known about conventional thermocouples
, one would expect the energy available to produce electricity to be equal to the heat that flows in from reservoir h minus the heat that is lost as it flows out into reservoir c. we call this quantity @xmath6 , and study how the power generated by a quantum dot thermocouple , @xmath7 , depends on it . in this work
, we show that the quantum - dot thermocouple can generate power even when the total heat absorbed from the thermal reservoirs ( h and c ) is zero ( @xmath8 )
. instead , the dot extracts heat from the electronic reservoirs ( l and r ) and converts it into electrical power .
it might appear that this `` exotic '' power generation violates the laws of thermodynamics ( since the thermal reservoirs provide no energy to allow the dot to convert heat into work ) , but we show that this is not the case . it can be explained in terms of a non - local heat drag effect , in which the heat flow from h to c can induce heat and charge currents in the circuit ( i.e. through the thermoelectric contacts to reservoirs r and l ) even when @xmath8 .
we show that this exotic power generation only arises because no thermalization occurs within the quantum dot , meaning that the dot is maintained in a non - thermal state by its contact with the hot and cold reservoirs . as such it has no analogue in conventional thermocouples .
we argue that as a result a quantum - dot thermocouple can achieve power outputs larger than any conventional thermocouple ( even one working with carnot efficiency ) over a broad range of parameters .
this is not because our device has an efficiency higher than carnot efficiency ( this is forbidden by the laws of thermodynamics ) , but because it can also extract useful work from the non - local drag effect .
this is something a conventional thermocouple can not do , irrespective of how efficient it is .
this work is organized as follows . in sec .
[ sect : quantum - dot ] , we introduce our model of the quantum dot thermocouple .
[ sect : v=0 ] discusses the basic mechanisms behind the rectification of heat fluctuations in our device , in the simpler case where it generates no power .
[ sect : finite - power ] gives results for the power generation , which sec .
[ sect : interpretation ] explains in terms of non - local heat drag , shows that it obeys the laws of thermodynamics , and discusses potential experimental implementation .
[ sect : thermalize ] shows that the effect is suppressed if relaxation processes cause the state of dot m to become thermal . finally , in sec .
[ sect : thermocouple ] , we compare the power output of the quantum dot thermocouple with that of a conventional thermocouple for a broad range of parameters .
the quantum dot m , shown in fig . [ fig : system](a ) , can exchange energy ( but not charge ) with a hot ( h ) and a cold ( c ) thermal reservoir . in the following we are interested in the resulting amount of electrical power that this quantum dot can produce by driving an electric current between the left ( l ) and right ( r ) electron reservoirs , to which it is tunnel - coupled .
importantly , while the reservoirs are characterized by temperatures , the quantum dot is generally in a _ non - thermal _ state .
this is a crucial point of our paper and an important difference to conventional macroscopic thermocouples , which are large enough such that every element of the thermocouple is able to thermalize .
the energy exchange of the quantum dot m with the thermal reservoirs h and c occurs via a capacitative coupling to thermal fluctuations in these reservoirs .
they are modeled as thermally activated hopping of electrons between the bulk of each of the two thermal reservoirs and a localized state on their surface .
the localized states can be thought of as quantum dots , which enables us to recast the problem as the triple - dot circuit shown in fig .
[ fig : system](b ) . here , the localized states associated with reservoirs h and c are the upper red and blue dots . apart from thermoelectric purposes , similar models have been useful for relaxation times detection @xcite , maxwell demon physics @xcite , and the operation of stochastic logic gates @xcite .
we therefore now consider the three capacitively coupled quantum dots m , h , and c , in contact with four electronic reservoirs h , c , l , and r , as shown in fig .
[ fig : system](b ) .
we assume that the on - site coulomb repulsion is very large , such that doubly occupation of any dot can be neglected . for simplicity
, we neglect the spin degree of freedom of the electrons , which , in this case , does not alter the working principle of the device , see also @xcite ( note however , that in general the spin might play an important role in quantum - dot based heat engines @xcite ) .
the isolated system of three capacitively coupled quantum dots can then be described by the model hamiltonian @xmath9 with the number operator @xmath10 counting the number of electrons ( @xmath11 or @xmath12 ) on each of the three dots , @xmath13 . here , @xmath14 are the energies for single occupation of the different dots and @xmath15 represents the coulomb charging energy that needs to be paid to have both dots @xmath16 and @xmath17 occupied simultaneously .
they play an essential role , as they mediate the energy exchange between the different dots .
the eigenstates of this hamiltonian are given by the eight states @xmath18 with eigenenergies @xmath19 , where the labels @xmath20 can take the values @xmath11 ( empty ) and @xmath12 ( full ) depending on the occupation of the respective dots m , h , and c. transitions between eigenstates of the triple - dot system can occur due to the tunnel coupling of these dots to the electronic reservoirs .
these reservoirs are kept at temperatures @xmath21 and electrochemical potentials @xmath22 . here
we take the temperature of the hot reservoir to be always larger than the temperature of the cold reservoir , @xmath23 .
the left and right reservoirs are assumed to be at the same temperature , @xmath24 for @xmath25 . in this paper , we are interested in the power that the quantum - dot thermocouple can generate , by driving an electrical current against a potential difference ( due to a load ) .
a simple way of implementing this setting is the one where we take the electrochemical potential of the right lead to be @xmath26 ( where @xmath27 is the electron charge ) with respect to the equal electrochemical potentials of the other reservoirs , which we take as the reference energy here , @xmath28 . the coupling between the quantum dots and the reservoirs , as depicted in fig .
[ fig : system](b ) , is characterized by the tunnel coupling strengths , @xmath29 .
we focus on a situation , where the dots are weakly coupled to the reservoirs , @xmath30 .
importantly , since we are interested in the quantum dot m behaving as a thermoelectric , we require that @xmath31 and @xmath32 depend on energy @xcite . because of the capacitive coupling between the dots , this means in turn that they depend on the occupation @xmath33 and @xmath34 of the dots h and c , yielding @xmath35 and @xmath36 .
a strong energy - dependence of the tunnel - coupling can be achieved for example by coupling the quantum dot to the reservoirs via other resonant levels @xcite .
this is not required for the tunnel - coupling between the thermal baths and the nearby quantum dots .
we therefore assume @xmath37 and @xmath38 to be energy - independent .
we are interested in the steady - state response of the system to different temperature in the heat baths , in the presence of a possible voltage drop across the reservoirs l and r , which results in power being generated across the load . in order to calculate the dc charge and heat currents , the occupation probabilities @xmath39 of the states @xmath18 of the triple - dot system are required . in the weak - coupling regime , these probabilities are given by the solution of the steady - state master equation @xcite , @xmath40 with @xmath41 .
all non - diagonal elements of the transition matrix are determined by @xmath42_{\ , mhc}^{\ , m'h'c ' } \ , \label{eq : m - in - terms - of - gammas}\ ] ] namely by the sum over the rates of transition of the three - dot system from state @xmath43 to state @xmath44 due to electron tunnelling into or out of reservoir @xmath45 , @xmath46_{\ , mhc}^{\ , m'h'c'}$ ] .
the latter are given by @xmath47_{\ , mhc}^{\ , m'h'c ' } \!\ ! & = \delta_{m,1-m'}\delta_{h , h'}\delta_{c , c'}\ \gamma_\text{l } \big(h , c\big)f_\text{l}\left({\delta_{mhc}^{m'h'c ' } } \right ) , \quad \\ \big [ \gamma_\text{r } \big]_{\ , mhc}^{\ , m'h'c ' } \!\ ! & = \delta_{m,1-m'}\delta_{h , h'}\delta_{c , c'}\ \gamma_\text{r } \big(h , c\big ) f_r\left({\delta_{mhc}^{m'h'c'}{-}(m{-}m')e v } \right),\quad \ \\ \big [ \gamma_\text{h } \big]_{\ , mhc}^{\ , m'h'c ' } \!\ ! & = \delta_{m , m'}\delta_{h,1-h'}\delta_{c , c'}\ \gamma_\text{h } f_\text{h}\left({\delta_{mhc}^{m'h'c ' } } \right ) ,
\\ \big [ \gamma_\text{c } \big]_{\ , mhc}^{\ , m'h'c ' } \!\ ! & = \delta_{m , m'}\delta_{h , h'}\delta_{c,1-c'}\ \gamma_\text{c}f_\text{c}\left({\delta_{mhc}^{m'h'c ' } } \right ) , \end{aligned}\ ] ] where @xmath48 is a kronecker delta , @xmath49 , and @xmath50 is the fermi function corresponding to lead @xmath45 .
the diagonal elements in eq .
( [ eq : master ] ) are directly found by using the condition that every column of the transition matrix sums to zero to fulfill probability conservation , @xmath51 . for any heat - engine ,
the relevant quantities are the generated power ( a charge current flowing against a potential difference ) and the heat flow into the device .
we therefore consider the charge and heat currents into the dots from reservoir @xmath45 , [ eq : currents ] @xmath52 here the charge current is trivially related to the particle current , @xmath53 , by multiplication with the electronic charge @xmath54 , while the heat current is given by the rate with which energy changes , @xmath55 , with respect to the energy that particles at the electrochemical potential would carry in the respective reservoir .
using the clausius definition of entropy , we see that the rate of change of entropy in reservoir @xmath45 is connected to the heat current by @xmath56 the particle currents and energy currents can be written using the transition rates @xmath57_{\ , mhc}^{\ , m'h'c'}$ ] and the occupation probabilities @xmath39 of the state @xmath18 with the respective energy @xmath19 and occupation number @xmath58 , @xmath59_{\ , mhc}^{\ , m'h'c ' } p_{m'h'c'}\ , \label{eq : particle - current } \\
i^{e}_\alpha & = & \sum_{mhc}\sum_{m'h'c ' } \big(e_{mhc}-e_{m'h'c'}\big)\ \big [ \gamma_\alpha \big]_{\ , mhc}^{\ , m'h'c ' } p_{m'h'c'}\ .
\label{eq : energy - current}\end{aligned}\ ] ] the particle and energy currents in reservoir @xmath45 are defined to be positive when they are directed flowing _ into _ the triple - dot system . consequently , the charge current in reservoir @xmath45 , eq . ( [ eq : charge - current ] ) , is positive when electrons flow into the reservoir , while the heat current in reservoir @xmath45 , eq .
( [ eq : heat - current ] ) , is positive when flowing into the triple - dot system .
particle and energy currents are conserved , so summing @xmath45 over all reservoirs gives @xmath60 charge current is conserved ( so @xmath61 ) .
heat current is not conserved since @xmath62 , which corresponds to the first law of thermodynamics .
the power generated by the quantum - dot thermocouple is @xmath63 these quantities are evaluated using eqs .
( [ eq : currents ] ) , and we will discuss them in detail in the following sections .
for now , we do not consider the probe reservoir p , shown in fig .
[ fig : system ] , however we will discuss it in sections [ sect : thermalize ] and [ sect : thermocouple ] .
we start by analysing the behaviour of the quantum dot thermocouple in fig . [
fig : system ] under the following two assumptions .
( i ) the load s resistance is zero , so the bias @xmath64 remains zero when a current is flowing .
( ii ) the capacitive coupling between dot h and c is strong enough that @xmath65 is very much bigger than all temperatures , so the probability that these two dots are occupied at the same time becomes vanishingly small .
we will relax both these assumption in section [ sect : finite - power ] . for strong @xmath65 , the states ( 0,1,1 ) and ( 1,1,1 ) drop out of the problem . in this limit , the charge fluctuations that lead to transport through the conductor can be separated in two cycles . the first cycle is [ eq : cycles ] @xmath66 and involves an energy exchange @xmath67 between the system and the hot reservoirs .
the second cycle is @xmath68 and involves an energy exchange @xmath69 between the system and the cold reservoir .
when the tunneling events in these sequences involve an electron that tunnels from reservoir l into the dot , and thereafter into reservoir r ( or viceversa ) , it contributes to transport .
the sum of all tunneling processes results in the generation of a finite electric current by the mere conversion of heat , only if a preferred direction is defined both by broken electron - hole and left - right symmetries @xcite .
we impose these conditions on the corresponding tunneling rates , requiring that the asymmetries , @xmath70[\gamma_{\text{l}}(1,0)+\gamma_{\text{r}}(1,0)]}\ , \label{eq : lambda}\ ] ] and @xmath71 ( obtained by replacing @xmath72 in the previous expression ) are finite . solving the master equation ( [ eq : master ] ) ,
we get analytical expressions for the charge and heat currents . despite these expressions being long and cumbersome ( even for @xmath73 )
, we find a remarkably simple relation between them : @xmath74 it is instructive at this point to introduce @xmath6 and @xmath75 , where [ eq : j_in&j_trans ] @xmath76 is the total heat flowing into dot m from the thermal reservoirs ( reservoirs h and c ) , and @xmath77 is the heat that is transferred from reservoir h to c. note that a proportion of @xmath78 may _ transit _ through dot m on its way from reservoir h to c , even when none of it enters the electrical circuit ( i.e. reservoirs l , r ) .
rewriting eq .
( [ iv0 ] ) in terms of @xmath79 and @xmath78 , one sees that @xmath80 contains one part that depends on @xmath79 as one might expect , but it also contains a term that depends on @xmath78 , even though that energy does not go into the circuit .
if the temperature and coupling of the hot and cold reservoirs are such that @xmath81 , one still generates a charge current [ which we here denote by @xmath82 in the circuit equal to @xmath83 whenever the charging energies @xmath84 and @xmath69 are not equal , there is no symmetry reason for @xmath85 to be zero , so a typical system will exhibit a charge current in the circuit even though there is no energy absorbed from the thermal baths ( @xmath81 ) .
in contrast , in the special case of @xmath86 , one can show that @xmath87 , which means that @xmath88 . to show this
, one notes that if @xmath86 , then the tunneling rates into dot m are not sensitive to which of the upper dot is occupied ( h or c ) , so @xmath89 , which implies @xmath87 .
it is thus essential that the two baths act differently on the system .
we achieve this by taking different capacitive couplings .
upper plot : power generated by the quantum - dot thermocouple ( solid curve ) as a function of @xmath90 compared to the maximum power @xmath91 ( dashed line ) any conventional thermocouple could generate ( see sec .
[ sect : thermocouple ] ) .
lower plot : energy currents into the system as a function of @xmath92 . for both panels ,
the system parameters are given by the temperatures @xmath93 , @xmath94 , the dot energy levels @xmath95 , the charging energies @xmath96 , @xmath97 , @xmath98 , and the coupling constants @xmath99 , @xmath100 , while @xmath31 and @xmath32 are given in eq .
( [ eq : gamma_l&r ] ) . ] for a finite power output , the machine must drive a current against a finite potential difference @xmath64 .
this occurs when the load s resistance is non - zero .
we also now consider the general case in which @xmath101 is finite , so the dot dynamics involves more cycles than just those in eqs .
( [ eq : cycles ] ) , which can explore all eight possible states of the three - dot system .
since in this general situation at any @xmath64 , the algebraic solution of the master equation , eq .
( [ eq : master ] ) , and of the resulting charge and heat currents are extremely cumbersome , we refrain from presenting them here . instead , in figs .
[ fig : basic - effect]-[fig : basic - effect - poor ] , we plot the power output generated by our quantum - dot thermocouple as a function of the temperature of the cold reservoir , @xmath3 , for various specific system parameters . for comparison
, we also show the energy currents flowing from reservoir l and r into the dot m , @xmath102 and @xmath103 , as well as the heat current flowing into the system from the hot and cold reservoirs , @xmath6 .
these plots show the possibility of generating finite power when the energy flow into the system from the hot and the cold reservoir vanishes .
importantly , the effect is shown to be generic , since the variation of different system parameters does not alter the basic features . for the plots in this paper , we give all rates and energies in terms of a rate @xmath104 . as a result the power generated @xmath7 and all energy currrents @xmath105
are in units of @xmath106 ( we take planck s contant @xmath107 ) . in all cases , we fix the temperatures of the hot reservoir and the load , @xmath93 , @xmath108 , and vary @xmath109 . in different plots we take different tunneling rates , @xmath110 , but always take them to be smaller than the temperatures , which ensures the applicability of the master equation approach introduced in the previous section
in each plot , for each value of @xmath90 , we find the bias voltage @xmath64 that maximizes the power generated .
we call this optimal bias @xmath111 .
we then plot the power generated and the energy flows at @xmath112 for each @xmath90 .
the plot in figs .
[ fig : basic - effect ] and [ fig : basic - effect - other - parameters ] are for [ eq : gamma_l&r ] @xmath113 this choice of @xmath114 and @xmath115 , ensures that when dot h and c are empty dot m is decoupled from reservoir r , but if either dot h or c is occupied then dot m is decoupled from reservoir l. this means that the asymmetry parameters in eq .
( [ eq : lambda ] ) are maximal , with @xmath116 .
all the other parameters are given in the figure captions .
as mentioned before , our model contains too many parameters to present a systematic study of the full parameter space , however the features we find appear to be generic .
the following two paragraphs highlight this by discussing central points about figs .
[ fig : basic - effect ] and [ fig : basic - effect - other - parameters ] . in fig .
[ fig : basic - effect ] we take a very large @xmath65 ( much larger than @xmath117 ) , which means that the dots h and c are so strongly capacitively coupled that there is an extremely small probability of them being both occupied at the same time ( similar to the limit @xmath118 taken in sec . [ sect : v=0 ] ) .
the power generated , @xmath7 , vanishes quadratically at a certain value of @xmath90 .
this parabolic behaviour of @xmath119 , occurs because the current @xmath120 changes sign ( going linearly through zero ) , and thus the bias generated also changes sign at this point ( @xmath121 for small @xmath120 ) .
however , we clearly see that @xmath7 vanishes at smaller @xmath90 than @xmath79 ; so @xmath7 is finite when @xmath81 ( marked by the filled circle in fig .
[ fig : basic - effect ] ) .
the power generated at this point is purely an effect of the heat flow @xmath78 , which passes from reservoir h to c without being absorbed by the thermocouple circuit .
we call this the _ exotic _ power generation .
when @xmath81 , one has @xmath122 , thus finite @xmath7 implies that there is a net heat current out of reservoirs l and r , meaning that the total entropy of those reservoirs is dropping .
[ sect : interpretation - nonlocal ] explains why this is not a violation of the second law of thermodynamics .
has @xmath123 , however this has no physical consequences as @xmath124 is gauge - dependent ( it depends on our choice of @xmath125 ) .
in contrast , the other quantities considered above , such as @xmath126 , @xmath127 and @xmath105 , are gauge - independent and so have physical meaning . ]
we do not show @xmath78 in fig .
[ fig : basic - effect ] , as it is significantly larger than the other heat flows , it is a positive and monotonically decreasing function of @xmath92 ( it goes from 0.14 at @xmath128 to 0.03 at @xmath129 ) .
upper plot : power generated by the quantum - dot thermocouple ( solid curve ) as a function of @xmath90 compared to the maximum power @xmath91 ( dashed line ) any conventional thermocouple could generate ( see sec .
[ sect : thermocouple ] ) .
lower plot : energy currents into the system as a function of @xmath92 .
for both panels , the system parameters are given by the temperatures @xmath130 , @xmath94 , the dot energy levels @xmath131 , the charging energies @xmath132 , @xmath97 , @xmath133 , and the coupling constants @xmath134 , @xmath135 , while @xmath31 and @xmath32 are given in eq .
( [ eq : gamma_l&r ] ) .
the bias @xmath112 is the one that maximizes the power generated for each @xmath92 . ]
the plots in fig .
[ fig : basic - effect - other - parameters ] show the same basic features as in fig .
[ fig : basic - effect ] , despite the following differences in the choice of parameters .
we no longer take @xmath65 to be effectively infinite ( i.e. much larger than @xmath117 ) .
we take the energy - levels of the dot to be above the chemical potentials of the reservoirs .
the coupling to reservoirs l and r are still given by eq .
( [ eq : gamma_l&r ] ) , ensuring the optimal asymmetry for the thermoelectric effect .
the decreased capacitive coupling @xmath65 ( of the order of temperature ) allows all the states of the three dots to play an active role in the dynamics ( unlike in the limiting case @xmath118 considered in sec .
[ sect : v=0 ] ) . this however appears to have little qualitative effect on the physics ( one can even take @xmath136 without considerably changing the features that interest us ) .
however , the shift of the energy levels with respect to the electrochemical potentials is directly responsible for the only qualitative difference between the results in figs . [
fig : basic - effect - other - parameters ] and [ fig : basic - effect ] , namely that @xmath79 now vanishes at two points in fig .
[ fig : basic - effect - other - parameters ] . this can be explained as follows .
when @xmath90 is large , then @xmath79 is positive , but as one reduces @xmath90 , the heat flow @xmath79 goes to zero and becomes negative .
however , as one reduces @xmath90 even further @xmath137 becomes significantly less than @xmath138 , and so the dynamics of dot c start to freeze out ( thermal fluctuations of dot c become extremely rare , as @xmath139 ) .
thus , when @xmath90 is sufficiently small , reservoir c becomes effectively decoupled from the rest of the system , the only heat flow is that from reservoir h to reservoirs l and r , and @xmath79 becomes positive again .
the power generated remains positive at all @xmath90 because of the exotic contribution coming from @xmath78 .
another consequence of the off - resonant level positions chosen here , is that both the energy current out of the left and the right reservoir are in general non - zero . as required from energy conservation , they are however opposite , @xmath140 , at the point of vanishing @xmath79 .
again @xmath78 is positive and significantly larger than the other heat flows for all @xmath92 , as we show in the inset of fig .
[ fig : entropy ] .
it drops for small @xmath90 , because of the above mentioned freezing of the dynamics of dot c. upper plot : power generated by the quantum - dot thermocouple ( solid curve ) as a function of @xmath90 compared to the maximum power @xmath91 ( dashed line ) any conventional thermocouple could generate ( see sec . [ sect : thermocouple ] ) .
lower plot : energy currents into the system as a function of @xmath92 . for both panels ,
the system parameters are given by the temperatures @xmath130 , @xmath94 , the dot energy levels @xmath141 , the charging energies @xmath142 , @xmath143 , and @xmath98 , and the coupling constants @xmath144 , while @xmath31 and @xmath32 are given in eq .
( [ eq : gamma_l&r - imperfect ] ) .
the bias @xmath112 is the one that maximizes the power generated for each @xmath92 . ]
the plots in fig .
[ fig : basic - effect - poor ] are for an imperfect thermocouple , where dot m is coupled to both reservoir l and reservoir r for all occupations of the of upper dots ( @xmath33 and @xmath34 ) .
we replace eq .
( [ eq : gamma_l&r ] ) with [ eq : gamma_l&r - imperfect ] @xmath145 this means that the coupling strength is only varying by @xmath146 as @xmath33 and @xmath34 change and the asymmetry parameters in eq .
( [ eq : lambda ] ) are @xmath147 .
the other parameters chosen for this plot are given in the figure caption .
the figure shows that the exotic power generation still exists for this imperfect pump , although its magnitude is greatly reduced ( just as the traditional power generation is greatly reduced ) .
the charge current we found in sec . [ sect : v=0 ] for @xmath81 is due to a form of coulomb heat drag , in which the heat flow from reservoir h to reservoir c , _ drags _ a heat and charge currents from reservoir l to r. this is possible due to the capacitive coupling between the quantum dots .
however , in other drag effects @xcite , there is a transfer of energy associated with one current dragging the other , which corresponds to finite @xmath79 .
hence , this drag effect without energy transfer is extremely unusual .
this heat drag effect can occur for any @xmath148 irrespective of the value of @xmath92 , since it only requires a heat flow from reservoir h to reservoir c. the quantum dot thermocouple then produces more work than a classical one ( see the discussion in sec.7 ) .
remarkably , this does not only occur in the configurations where @xmath149 , but also in a relatively wide parameter range around this points , including the case @xmath150 .
the entropy generated in each reservoir ( main plot ) and @xmath78 ( inset ) for the system in fig .
[ fig : basic - effect - other - parameters ] .
the entropy generated in reservoirs l , r and h are small in magnitude and their sum is negative , however the entropy generated in reservoir c is large and positive .
thus , the second law of thermodynamics is not violated .
we see that the rate of entropy production broadly follows @xmath78 ( inset ) .
we do not show plots of entropy production and @xmath78 for the system in fig .
[ fig : basic - effect ] , as the main features are generic to all of them . ] despite the unusualness of the coulomb drag without energy transfer discussed above , the situation is not paradoxical as long as @xmath73 .
the case @xmath73 corresponds to a load with zero resistance , so nothing opposes the charge flow , and so no work is required to generate a finite charge current .
in contrast , as soon as the load s resistance is finite , the charge current @xmath151 must flow against a finite voltage bias , @xmath64 , then work must be performed on the load .
it would be natural to assume that this work must be provided by the thermal reservoirs , i.e. that the charge current @xmath151 can only flow if @xmath152 .
however , something completely different can be observed in the results in sec .
[ sect : finite - power ] : there we show that finite power can be generated even when @xmath81 .
we call this _ exotic _ power generation .
we observe that the power generated is equal to the heat absorbed by dot m from reservoirs l and r , as there is no heat flow into dot m from anywhere else when @xmath81 .
thus the dot is sucking heat from these reservoirs ( which are both at the same temperature ) , to turn it into useful work .
this obeys the first law of thermodynamics ( energy conservation ) , but if we ignore the presence of reservoirs h and c it appears to violate the second law , since it reduces the entropy of reservoirs l and r. however , if we now turn to reservoirs h and c , we see that there is a heat flow from hot to cold , and this increases the total entropy involved . an elegant way to verify this is the prove in ref
. @xcite that any such master equation respects the laws of thermodynamics , thus we know that the four reservoirs and three dots together do produce entropy .
what is more , one can rewrite the total rate of entropy production , @xmath153 , as @xmath154 now , since we know that @xmath155 for our system , see also fig . [ fig : entropy ] , one knows that @xmath156 hence , the second law allows the power generated , @xmath7 , to be finite and positive , even if @xmath81 , as long as @xmath78 is finite and positive .
the entropy increase in reservoir c compensates for the reduction of entropy in the other reservoirs ( see fig . [
fig : entropy ] ) .
thus the apparent paradox is resolved by the fact the power generation has a non - local nature .
the entropy generation caused by the heat flow between one pair of reservoirs ( h and c ) enables the conversion of heat into work at another pair of reservoirs ( l and r ) , even though there is no heat ( or energy ) flow between these two pairs of reservoirs and even though they can be arbitrarily far apart .
a `` poetic '' way of expressing this non - locality is to say that there is a spatial separation of first and second laws of thermodynamics .
the thermocouple manifests the first law by converting heat in reservoirs l and r into work . at the same time ,
some distance away , entropy is being generated by the flow of heat from reservoirs h and c , manifesting the second law . while we have shown that this behaviour is not paradoxical
, it certainly is exotic !
transport through capacitively coupled quantum dots in multiterminal conductors has been measured @xcite .
the effect of large charging energies on the thermoelectric response of small tunnel junctions @xcite and quantum dots @xcite is well - known experimentally @xcite .
our proposed device is close to those very recently realized in experiments @xcite , in which a quantum dot generates a charge current between two reservoirs ( l and r ) by the rectification of thermal fluctuations in a third capacitatively - coupled reservoir ( the heat source ) . in those systems , it is likely that the dot with tunnel - coupling to reservoirs l and r also has at least a weak capacitative coupling to thermal fluctuations in its cold environment ( including reservoirs l and r ) .
thus , these experimental systems have a strong similarity with the four - terminal set - up that we discuss .
however , to unambiguously show the exotic power generation experimentally , it would be desirable to have a more controlled coupling to the cold environment .
this could be done by capacitatively coupling a third dot to the systems in refs .
@xcite , which we believe is achievable with the technologies used in those works .
more challenging will be to show experimentally that @xmath81 , for this it will be crucial to measure the heat currents accurately .
for this , the capacitive coupling to the heat source and sink permits the different charge fluctuations to be resolved experimentally @xcite
. therefore , @xmath157 and @xmath158 can be measured @xcite , so it would be possible to verify that a finite charge current @xmath151 flows when the condition @xmath8 is met .
it is already possible to manipulate locally the temperature of the reservoirs @xcite , thus getting a knob for controlling the heat currents and ensuring @xmath8 .
our quantum dot model differs from that of a conventional thermocouple in a number of ways ; such as the fact that dot m has discrete states and that it is in the coulomb blockade regime .
yet , we claim that the exotic power generation is the consequence of the non - thermal nature of the distribution in dot m , rather than one of the other properties of the dot .
to verify this , we model the thermalization processes within dot m phenomenologically , by considering an additional probe reservoir ( p ) , which is tunnel - coupled to dot m , as indicated in fig .
[ fig : system ] . the working principle of the probe @xcite is analogous to the one of a voltage probe @xcite , with the additional feature of probing the temperature . in other words ,
the probe is treated as an electronic reservoir in thermal equilibrium at a temperature @xmath159 and a chemical potential @xmath160 , which have to be self - consistently determined by requiring that both the charge and heat currents , eq .
( [ eq : currents ] ) , into the probe reservoir are equal to zero , @xmath161 . in order to evaluate the currents into the probe and to extract @xmath159 and @xmath160 that make them vanish
, we treat the probe reservoir on the same footing as the other electronic reservoirs introduced in sec .
[ sect : quantum - dot ] .
the coupling between dot m and the probe p is @xmath162_{\ , mhc}^{\ , m'h'c'}\!\!\!\!\!\ ! & = & \!\!\ ! \delta_{m,1-m'}\delta_{h , h'}\delta_{c , c'}\ \gamma_\text{p}\ f_\text{p}\left(\delta^{m'h'c'}_{mhc } \right ) \ , \end{aligned}\ ] ] with eq .
( [ eq : m - in - terms - of - gammas ] ) s sum over @xmath45 extended to include p. then , @xmath163 and @xmath164 are given by eqs .
( [ eq : currents ] ) with @xmath165 .
if the probe is much more weakly coupled to dot m than the other two electronic reservoirs , ( @xmath166 ) , it does not influence the dynamics of the 3-dots system , i.e. it does not alter the results discussed in sec .
[ sect : v=0 ] and [ sect : finite - power ] .
in contrast , if @xmath167 is comparable with @xmath168 , or even larger , then the probe mimics phenomenologically the process of thermalizing transitions within dot m ( i.e. transitions that take the dot towards thermal equilibrium , without changing its average energy or charge ) . in other words ,
the state of dot m becomes closer to thermal as we increase the coupling to the probe reservoir . to measure how far dot m
is from a thermal state , we first note that if it were in a thermal state corresponding to equilibrium with the probe reservoir , then fluctuations of @xmath169 induced by the probe reservoir in both directions ( @xmath170 and @xmath171 ) would have the same rate , irrespective of the state of the other dots . in other words ,
dot m is in a thermal state in equilibrium with the probe reservoir p if @xmath172_{0hc}^{1hc}p_{1hc } = \big[\gamma_{\rm p}\big]_{1hc}^{0hc}p_{0hc}$ ] for all @xmath173 . a little
algrebra shows that this thermal state corresponds to @xmath174 \ , \end{aligned}\ ] ] for all @xmath173 .
thus , we define the following measure of how far dot m is from thermal equilibrium , @xmath175 -1 \ .
\label{eq : d_hc}\end{aligned}\ ] ] fig .
[ fig : probe - effect](a ) shows how increasing the coupling @xmath167 to the probe tends to reduce @xmath176 for all @xmath173 , indicating that the steady state of dot m becomes closer to that of the thermal state with temperature @xmath177 .
[ fig : probe - effect](b ) shows that as dot m s state becomes closer to the thermal state , the power generated at @xmath8 goes to zero .
thus the exotic power generation that we discuss in this article is a consequence of the fact that dot m is maintained in a non - thermal state by its couplings to reservoirs h and c.
a conventional thermocouple corresponds to the system in fig .
[ fig : system](a ) , with the quantum dot replaced by a macroscopic reservoir m that is larger than the electron s thermalization length , so it equilibrates at a temperature @xmath178 between @xmath179 and @xmath3 .
the role of the coupling to reservoirs h and c is to establish the temperature @xmath178 . in the steady - state
, @xmath178 is the temperature which ensures that the heat - flow out of reservoir m into the thermoelectrics is equal to the heat flow into reservoir m from reservoirs h and c , which we call @xmath79 . given @xmath79 and @xmath178 , the thermocouple s efficiency , @xmath180
, can not exceed carnot efficiency @xmath181 .
thus @xmath182 with the equals sign holding only if the two thermoelectrics are carnot efficient , corresponding to them both having a figure of merit @xmath183 .
we can immediately see that a system like ours , which generates finite @xmath7 when @xmath81 , violates the bound in eq .
( [ eq : naive ] ) . here
, we wish to see over what range of parameters our system violates this bound . to do this
, we need to associate a temperature , @xmath178 , to the non - thermal state of the quantum dot , so we can calculate @xmath184 .
there is no unique way of defining such a temperature , however a natural way to do so is to say it is the temperature that a probe would measure if placed in thermal contact with the dot @xcite . as such , we can use probe reservoir p considered in the previous section in the limit of very weak coupling to the dot ( @xmath185 ) , so it has negligible effect on dot - m s state .
although dot m is in general in a non - thermal state , we can associate to it an _ effective temperature _ @xmath178 , by defining @xmath178 as the temperature of the probe reservoir when there is no charge or heat current flow between dot m and reservoir p , i.e. @xmath186 .
note that ref .
@xcite argued that for having a meaningful definition of @xmath187 , it is crucial that the probe s coupling to the dot is the same at all energies , so the probe is sensitive to the whole energy distribution of the dot .
having defined an effective temperature for dot m , and knowing the heat - flowing into it , @xmath79 , we can plot the upper bound given by eq .
( [ eq : naive ] ) for any given parameters of the model .
this is represented as a dashed curve in all plots , see figs .
[ fig : basic - effect]-[fig : probe - effect ] .
we see that the power generated by our machine ( solid curves ) exceeds this classical bound over a broad range of parameters , including when @xmath90 is equal to or above @xmath4 .
note that @xmath91 will be negative whenever @xmath79 and @xmath188 are of opposite sign .
for example , in fig .
[ fig : basic - effect ] one has @xmath189 for @xmath190 , while @xmath191 only for @xmath192 .
thus for @xmath193 , the dot m is hotter than reservoirs l and r but , since @xmath189 , we know that energy is flowing from reservoirs l and r into the dot . in a macroscopic thermocouple , this can occur only if one provides energy to the system by replacing the load with a power supply that forces the current flowing through the circuit . in this case ,
power is not generated , but absorbed by the system , i.e. @xmath194 .
in contrast , our quantum dot system generates power , @xmath195 , even in this range of parameters , for the reasons discussed in sec .
[ sect : interpretation ] . figs .
[ fig : probe - effect](b ) and ( c ) show that as dot m s state becomes closer to that thermal state , the violation of eq .
( [ eq : naive ] ) goes away , leaving a system with @xmath196 , where @xmath197 is the system s efficiency , and it is less than the carnot efficiency @xmath184 .
for example , for @xmath198 in fig .
[ fig : probe - effect](b ) , @xmath197 is of order @xmath199 .
this confirms our conclusion that a non - thermal steady - state distribution of electrons in dot m is necessary for it to generate more power than any conventional thermoelectric could .
we have investigated the unexpected properties that arise in thermoelectric conductors when they become smaller than the lengthscale in which electrons thermalise . for this , we have considered a quantum dot based thermocouple in a four terminal geometry : two terminals support the electrical current , with the other two being the hot and cold environments with which energy is exchanged .
our system can supply electrical power to a load .
surprisingly , this can be done also in configurations in which no net heat is absorbed from the thermal reservoirs .
this exotic power generation can be explained in terms of non - local coulomb heat drag in which the heat flow from the hot to the cold reservoir induces both a heat and a charge flow in the capacitively coupled electronic circuit .
this effect should be realizable in systems very close to those in recent experiments @xcite , as outlined in sec .
[ sect : expt ] .
we argue that this exotic power generation relies on the fact that quantum dot m is maintained in an non - thermal state , by showing that it disappears when relaxation effects destroy the non - thermal state . as such it can not occur in a conventional macroscopic thermocouple heat - engine .
we compare our quantum dot thermocouple to its macroscopic equivalent , and find that avoiding the thermalization of carriers can help to improve the thermoelectric performance . in particular , the generated power can be larger than that of any macroscopic thermocouple , even when the latter works with carnot efficiency . in the long term , we hope that this observation will be applied in thermoelectric and photovoltaic applications , since in both cases a large part of the heat flowing into the central region of the thermocouple ( or the central region of the p - n junction in photovoltaics ) flows out again into the cold environment , without contributing to the power generation .
it would be a great benefit if a part of this `` lost '' energy flow could be used to make electrical power in the manner presented here .
the same effect could be achieved with non - interacting conductors in magnetic fields , via a nernst effect @xcite .
note that refs .
@xcite assumed that the reservoirs that carry the charge current ( reservoirs l and r in our fig .
[ fig : system ] ) are at temperatures chosen such that there is no heat flow out of them . in this situation ,
energy conservation guarantees that @xmath200 for @xmath81 .
however , for other temperatures , one expects to observe the exotic power generation that is discussed in the present paper .
an open question is how to engineer the exotic power generation when the capacitative coupling is replaced by exchange of photons @xcite or phonons @xcite .
not only is this crucial for thermoelectric and photovoltaic applications , it would help to elucidate the fundamental requirements for the effect . as a step in this direction
, it may be worthwhile studying the linear - response regime in more detail .
the results presented here are valid for arbitrary @xmath179 , @xmath3 and @xmath4 ; while the observed effect is most prominent for very different temperatures , it also exists when the temperature differences are small . in this limit
, one could analyze the effect using a multi - terminal onsager treatment @xcite , which is more universal , but only works when there is a linear relationship between thermodynamic forces and currents .
r. w. and r. s. thank d. weinmann and b. sothmann , respectively , for helpful discussions .
part of this work was carried out thanks to the cost action mp1209 `` thermodynamics in the quantum regime '' ( stsm visit to the institute for quantum information , rwth aachen , and the `` second conference on quantum thermodynamics '' , mallorca , 2015 ) .
we acknowledge financial support from the cnrs peps funding ( r. w. ) , the spanish micinn juan de la cierva program and mat2014 - 58241-p .
( r. s. ) , the knut and alice wallenberg foundation and the swedish vr ( j. s. ) , and the alexander von humboldt foundation ( f. h. ) .
d. g. cahill , p. v. braun,1 g. chen , d. r. clarke , s. fan , k. e. goodson , p. keblinski , w. p. king , g. d. mahan , a. majumdar , h. j. maris , s. r. phillpot , e. pop , and l. shi , appl .
phys . rev . * 1 * , 011305 ( 2014 ) .
c. van den broeck , `` stochastic thermodynamics : a brief introduction '' in _ physics of complex colloids _ , editors : c. bechinger , f. sciortino , and p. ziherl , proc .
internat .
school phys .
enrico fermi * 184 * , 155 ( 2013 ) .
online at ebooks.iospress.nl/volumearticle/33636 .
i. h. chan , r. m. westervelt , k. d. maranowski , and a. c. gossard , appl .
. lett . * 80 * , 1818 ( 2002 ) .
a. hbel , j. weis , w. dietsche , and k.v .
klitzing , appl .
phys . lett . * 91 * , 102101 ( 2007 ) .
d. t. mcclure , l. dicarlo , y. zhang , h .- a .
engel , c. m. marcus , m. p. hanson and a. c. gossard , phys .
. lett . * 98 * , 056801 ( 2007 ) . | we analyze the power output of a quantum dot machine coupled to two electronic reservoirs via thermoelectric contacts , and to two thermal reservoirs one hot and one cold .
this machine is a nanoscale analogue of a conventional thermocouple heat - engine , in which the _ active _ region being heated is unavoidably also exchanging heat with its cold environment .
heat exchange between the dot and the thermal reservoirs is treated as a capacitive coupling to electronic fluctuations in localized levels , modeled as two additional quantum dots .
the resulting multiple - dot setup is described using a master equation approach .
we observe an `` exotic '' power generation , which remains finite even when the heat absorbed from the thermal reservoirs is zero ( in other words the heat coming from the hot reservoir all escapes into the cold environment ) .
this effect can be understood in terms of a non - local effect in which the heat flow from heat source to the cold environment generates power via a mechanism which we refer to as _ coulomb heat drag_. it relies on the fact that there is no relaxation in the quantum dot system , so electrons within it have a non - thermal energy distribution .
more poetically , one can say that we find a spatial separation of the first - law of thermodynamics ( heat to work conversion ) from the second - law of thermodynamics ( generation of entropy ) .
we present circumstances in which this _ non - thermal _ system can generate more power than any conventional macroscopic thermocouple ( with local thermalization ) , even when the latter works with carnot efficiency . ' '' '' -3.5 mm * highlights * -4.1 mm @xmath0 model of a quantum - dot energy - harvester : a quantum analogue of a thermocouple .
@xmath0 two thermal reservoirs ( heat source and cold environment ) and two electric contacts .
@xmath0 non - local coulomb heat drag effects , non - locality of laws of thermodynamics . @xmath0 finite power output without absorbing heat from the thermal reservoirs .
@xmath0 can beat a carnot - efficient conventional thermocouple under equivalent conditions .
quantum thermodynamics , thermocouples , thermoelectricity , quantum transport , energy harvesting , coulomb drag ( a ) a sketch of a quantum dot thermocouple in which the electrons in the central part of the thermocouple ( the quantum dot ) exchange heat with its cold environment ( reservoir c ) as well as the heat source ( reservoir h ) .
the couplings of the dot to reservoirs l and r have an energy dependence chosen to ensure that each one has the opposite thermoelectric response . as a result they form a thermocouple , which can be used to generate electrical power .
the load is taken to be a device that converts the electrical power into some other form of useful work ( it could be a motor converting the electrical work into mechanical work ) .
( b ) a circuit picture of the set - up in ( a ) , in which the exchange of energy with the heat source and the environment are modeled by a capacitive coupling to electronic fluctuations in those reservoirs , which we take to be thermally activated hopping of electrons between the bulk and a localized state , indicated by the two upper circles ( h and c ) .
thus each of the three circles represents a two - state system , with occupations 0 or 1 . ] _ dedicated to markus bttiker : in addition to his human qualities , we remember markus fondly for the inspiring discussions we had .
we think that this work would have led to another lively and enjoyable debate . _ |
an active galactic nucleus ( agn ) is known to generate multi - wavelength radiation by mass accretion onto a super massive black hole ( smbh ) in its central engine .
an important element of the central engine is the standard accretion disk ( e.g. , shakura & suynaev 1973 ; balbus & hawley 1991 ; machida 2000 ) , which is expected to form at the eddington ratio of @xmath111 ( e.g. , abramowicz et al . 1995 ) , and converts appreciable fraction of the gravitational energy release of the accreting matter into optically - thick radiation .
a part of the radiation has been observed as big blue bump " seen in optical spectra from agns that have disk radiation stronger than jet emission , like seyfert galaxies and quasars ( e.g. , elvis et al . 1994 ) .
the central engine requires another element that boosts up these low - energy photons into the observed x - ray signals .
this is usually attributed to some sort of hot maxwellian electrons , or `` corona '' , which compton - upscatter the seed photons into broadband x - ray photons .
however , the configuration of such corona is still under big debates , in contrast to the well - understood standard accretion disk .
some x - ray studies led to an argument that a compact corona is located on a rotation axis of the accretion disk like a `` lamppost '' , producing an extremely relativistically - broadened reflection component by illuminating the inner accretion disk ( e.g. , miniutti & fabian 2004 ; dauser et al .
others argue that an extended corona is present at the inner edge of an accretion disk , sometimes affected by partially - covering absorptions of ionized matters ( e.g. , miller et al . 2008 ; miyakawa et al .
2012 ) . to settle the scenario of the comptonizing corona , we need to focus on primary x - ray spectra and their flux variability .
according to x - ray studies of agns , several kinds of primary x - ray signals with distinct spectral and timing properties were reported .
a flat primary spectrum , which is reproduced by a single power - law ( pl ) continuum with the photon index of @xmath12 , dominates in x - rays from relatively low eddington - ratio agns ( e.g. , terashima et al .
2002 ) . on the other hand ,
a steep primary continuum , which can be explained by a pl model with @xmath13 , is dominant in x - rays from highly - accreting agns like narrow - line type 1 seyferts ( e.g. , laor et al .
1994 ; boller et al .
recently , noda et al . ( 2011a , 2013a , and 2014 ) revealed that both these flat and steep primary continua are simultaneously present in the x - ray emission from multiple seyfert galaxies .
the presence of these different primary x - rays may represent presence of several distinct types of coronae with different electron temperatures and optical depths .
furthermore , in order to explain a soft x - ray excess structure at a low energy band below @xmath14 kev , a soft thermal comtonization continuum has been suggested , invoking yet another corona ( e.g. , mehdipour et al .
2011 ; noda et al .
2011b , 2013b ; jin et al .
the reality of this third corona , however , needs further examination , because the soft excess may alternatively be modeled in terms of relativistically - smeared ionized reflection ( e.g. , fabian et al . 2004 ) or absorption by disk winds ( e.g. , cierlinski & done 2004 ) .
ccccccc[b ] 1 & 2013 apr 9 23:13:20 & apr 11 01:06:16 & 56392.51 & 51 + 2 & 2013 apr 27 00:17:13 & apr 27 10:42:22 & 56409.27 & 19 + 3 & 2013 may 12 00:22:24 & may 13 02:30:23 & 56424.58 & 50 + 4 & 2013 may 23 03:32:08 & may 24 07:05:07 & 56435.72 & 51 + 5 & 2013 may 29 11:02:50 & may 30 15:15:14 & 56442.05 & 54 + 6 & 2013 nov 4 06:15:13 & nov 5 05:10:17 & 56600.74 & 46 + 7 & 2014 apr 7 16:54:26 & apr 8 12:00:24 & 56755.12 & 52 [ tab : xobs ] lcccccc[b ] pirka & 1.6 & msi & @xmath15 & @xmath16 & @xmath1 , @xmath17 & 86 + mitsume & 0.5 & mitsume & @xmath18 & @xmath19 & @xmath20 , @xmath21 , @xmath22 & 6 + kiso schmidt & 1.5 & kwfc & @xmath23 & @xmath24 & @xmath1 , @xmath17 & 32 + nayuta & 2.0 & mint & @xmath25 & @xmath26 & @xmath1 , @xmath17 & 31 + kanata & 1.5 & howpol & @xmath27 & @xmath28 & @xmath1 , @xmath17 & 31 [ tab : optobs ] to clarify geometry of materials around a smbh , correlations and time lags between emissions in different energy bands are useful .
recently , studies of reverberation between different x - ray bands revealed that soft x - rays and fe - k@xmath7 emission lines have positive time lags , by hundreds seconds , against hard x - rays , with an implication that some reprocessing materials are present near the hard x - ray emitters ( e.g. , uttley et al .
2014 ) . likewise ,
x - ray and optical / ultraviolet ( uv ) correlations are useful to investigate coronal geometries around an accretion disk .
because seed photons for the inverse comptonization process are presumably provided via the optical / uv disk black body , variations in these low - energy bands can precede that of x - rays with a lag of @xmath29days ( e.g. , nandra et al .
when fluctuations of the mass accretion rate propagate inward from the accretion disk to the corona , the optical emission is also expected to precede x - rays , but on a viscous or thermalizing time scale which may be much longer than days ( e.g. , uttley & casella 2014 ) .
on the other hand , the disk should be illuminated by the x - rays , so that x - ray variations can cause optical flux changes , producing delays of @xmath29days in the opposite sense ( e.g. , krolik et al . 1991 ) . in order to distinguish these cases , and to better understand the geometry of the coronae and accretion disk
, we need to measure the sign and length of the optical vs. x - ray time lag , and quantify the strength of their correlations .
so far , a large amount of effort has been invested on coordinated x - ray and optical / uv observations of a number of agns .
however , we are far from achieving a unified view . for example , seyfert galaxies including ngc 5548 ( uttley et al . 2003 ; suganuma et al . 2006 ; mchardy et al . 2014 ) , ngc 3783 ( arevalo et al . 2009 ) , and mrk 79 ( breedt et al . 2000 ) exhibited relatively strong correlations with optical lags by several days , while those including ark 564 ( gaskell 2006 ) and ngc 3516 ( maoz et al .
2002 ) showed much poorer correlations .
one possible cause of this variety , we speculate , is that the different x - ray primary continua have different correlations to the optical / uv signals , and the previous coordinated observations mixed up the effects from these multiple x - ray components . to overcome this difficulty , simultaneous observations of agns should be conducted , under a condition wherein the different x - ray components can be clearly identified and separated . to derive the geometrical information from the multi - wavelength correlations
, we performed an x - ray and optical simultaneous monitoring during 20132014 , by utilizing _
( mitsuda et al . 2007 ) and 5 japanese ground - based telescopes .
the target of this campaign is the bright type 1.5 seyfert galaxy ngc 3516 .
it has a redshift of @xmath30 which corresponds to a distance of @xmath31 mpc @xmath32 cm .
its smbh is estimated to have a mass of @xmath33 ( denney et al .
2010 ) , which yields the schwarzschild radius of @xmath34 cm .
the column density of the galactic interstellar absorption toward ngc 3516 is @xmath35 @xmath5 ( dickey & lockman 1990 ) , and the optical galactic extinction is @xmath36 mag ( nasa / ipac extragalactic database based on schlafly & finkbeiner 2011 ) .
although ngc 3516 was previously reported , by maoz et al . ( 2002 ) , to exhibit a poor x - ray and optical correlation , the result will change if we properly decompose its x - ray continuum using the technique of noda et al .
( 2011b , 2013b ) , because ngc 3516 is one of the prototypical objects to which this method has been successfully applied ( noda et al . 2013 ) .
unless otherwise stated , errors shown in tables and figures in the present paper refer 1@xmath37 errors .
the x - ray monitoring observations of ngc 3516 were performed in the _ suzaku _ ao-8 cycle during 20132014 .
specifically , ngc 3516 was observed with _
suzaku _ five times from 2013 april 10 to may 29 , with intervals of @xmath382 weeks .
another pointing was carried out on november 7 , and the last one on 2014 april 4 ; thus , total 7 observations , with various intervals from days to months . in the present paper ,
we call the 1st , 2nd , ... and the 7th observations epoch 1 , epoch 2 , ... and epoch 7 , respectively .
the exposure in each epoch was @xmath39 ksec , except for epoch 2 which has an exposure of @xmath40 ksec .
in all these epochs , the xis ( koyama et al . 2007 ) and hxd ( takahashi et al . 2007 ) were operated in their normal modes , and data of the xis and hxd - pin are utilized in the present paper , except for epoch 7 , in which the hxd - pin count rate was too low to significantly detect the signals .
the data of the xis and hxd - pin were processed via the software version 2.4 . in the xis analysis
, we added the data of xis0 and xis3 , and utilized it as fi data , while we did not use xis1 due to its higher background .
the xis source events were extracted from a @xmath41-radius circle with its center on the source .
background events were accumulated from a surrounding annular region with the inner radius of @xmath42 and the outer radius of @xmath43 . the response and ancillary - response files were prepared via softwares in heasoft 6.14 called _ xisrmfgen _ and _ xissimarfgen _ ( ishisaki et al . 2007 ) , respectively .
the hxd data were prepared by the same way as that of the xis . in the hxd analysis ,
non x - ray background ( nxb ) and cosmic x - ray background ( cxb ) were estimated by standard models described by fukazawa et al .
( 2009 ) and gruber et al .
( 1999 ) , respectively .
they were subtracted from the on - source data .
the optical photometric monitoring observations of ngc 3516 were performed by using the charge - coupled device ( ccd ) cameras installed at the pirka , mitsume , kiso shcmidt , nayuta , and kanata telescopes , of which major parameters are summarized in table [ tab : optobs ] .
simultaneous monitoring observations were performed at the same 7 epochs as the x - ray observations by _ suzaku_. in addition , from 2013 january to 2014 april , these telescopes densely monitored the source even without the simultaneous x - ray coverage , with a typical intervals of @xmath44 day . in the present paper ,
we present the @xmath1- and the @xmath20-bands photometric data , because the optical continuum emission of agns is generally bluer than that of the host galaxy , thus ensuring better sensitivity to the agn signals .
the images were reduced using iraf following the standard procedures of image reduction for ccd detectors .
figure [ fig : xis_lc ] shows light curves in the 23 kev and 310 kev bands in all the 7 epochs . while we can see gradual flux changes , the source did not show short - term variations , on time scales of @xmath29hours , that were observed by _ xmm - newton _ in 2006 ( mehdipour et al .
the 23 kev light curves are clearly synchronized with that in the 310 kev band .
the highest flux was obtained in epoch 4 , while the lowest in epoch 7 , with a peak to peak variation amplitude reaching a factor of @xmath40 . to quantify the flux variability , we made , in figure [ fig : xccp ] , a count - count plot between count rates in the 23 kev and 33.5 kev bands , where these objects generally exhibit rather high variability ( noda et al .
2013a ) . the count - count plot of ngc 3516 derived in 2009 october ( noda et al .
2013c ) is also shown for reference , after correcting the data for a pointing position difference and long - term changes in the detector response .
surprisingly , all the data points in 20132014 line up with those in 2009 , thus defining an almost linear correlation between the two bands .
this means that the variable component in those energy bands observed in 20132014 has nearly the same spectral shape to that detected in 2009 . in energy bands higher than 3.5 kev , a reflection component with a prominent fe - k@xmath7 emission line at @xmath45 kev
is expected to become more dominant than in energies below 3.5 kev .
therefore , this variable component should be separated from the reflection via spectral fitting .
although the reflection component is considered almost stationary within a week ( noda et al .
2013b ) , it can vary on timescales of months , because a broad line region and/or a dusty torus , where the reflection emission possibly originates , are known to locate at several tens
hundreds light days ( e.g. , koshida et al .
accordingly , we treat the reflection signal as a variable component .
cccccccccc[t ] ` wabs ` & @xmath46 & @xmath47 & @xmath48 & @xmath49 & @xmath50 & @xmath51 & @xmath52 & @xmath53 + ` cutoffpl ` & @xmath54 & + & @xmath55 ( kev ) & + & @xmath56 & @xmath57 & @xmath58 & @xmath59 & @xmath60 & @xmath61 & @xmath62 & @xmath63 + & @xmath64 & @xmath65 & @xmath66 & @xmath67 & @xmath68 & @xmath69 & @xmath70 & @xmath71 + ` pexmon ` & @xmath72 & + & @xmath73 ( kev ) & + & @xmath74 & + & @xmath75 ( @xmath76 ) & + & @xmath77 & @xmath78 & @xmath79 & @xmath80 & @xmath81 & @xmath82 & @xmath83 & @xmath84 + & @xmath85 & @xmath86 & @xmath87 & @xmath88 & @xmath89 & @xmath90 & @xmath91 & @xmath92 + & @xmath93 & @xmath94 & @xmath95 & @xmath96 & @xmath97 & @xmath98 & @xmath99 & @xmath100 + & @xmath101/d.o.f . &
[ tab : specfit ] as presented in figure [ fig : xspectra ] , we produced seven time - averaged and background - subtracted spectra , one from each epoch , and performed a simultaneous model fitting to them . because of the presence of diffuse x - ray emissions in the host galaxy ( e.g. , constantini et al . 2000 ; george et al .
2002 ) , and effects of absorption variations ( e.g. , turner et al . 2011 ) , we ignored energy bands lower than 2 kev , and utilized the 245 kev energies in each spectrum .
the employed spectral model is ` wabs * ( cutoffpl + pexmon ) ` in xspec12 , which is consistent with that utilized in noda et al .
( 2013a , c ) . here , ` wabs ` is a model for the photoelectric absorption ( morrison and mccammon 1983 ) , and its column density @xmath102 was allowed to differ among the epochs .
the ` cutoffpl ` model represents the pl continuum with a high energy exponential rolloff , utilized to emulate the inverse comptonization radiation .
the reflection continuum and the fe - k@xmath7 emission line , both produced by this continuum , are reproduced by ` pexmon ` ( nandra et al .
2007 ) . in the fitting ,
the photon index @xmath103 was tied among the epochs ( because of figure 2 ) , and left free .
the cutoff energy @xmath104 was fixed at 150 kev based on the typical value reported by malizia et al .
( 2014 ) , while the normalization @xmath105 was left free in each epoch . for `
pexmon ` , @xmath103 and @xmath104 of the incident pl were tied to those in ` cutoffpl ` , and the fe abundance @xmath75 , inclination angle @xmath106 , and the reflection fraction @xmath107 were fixed at 1 solar , 60@xmath108 , and 1 , respectively .
the normalization @xmath109 of the ` pexmon ` component was allowed to differ among the epochs .
the @xmath109 value is determined almost solely by the fe - k@xmath7 line intensity , because the ratio between the fe - k@xmath7 line intensity and the reflection continuum in ` pexmon ` does not change when @xmath104 , @xmath75 , and @xmath106 are all fixed .
as summarized in figure [ fig : xspectra ] and table [ tab : specfit ] , the simultaneous fitting was successful with @xmath110/d.o.f.=1184.0/1113 .
as expected from figure [ fig : xccp ] , all the spectra have been reproduced by a hard pl continuum having @xmath111 , which is consistent with that in 2009 ( @xmath112 ) , together with a reflection component accompanied by a prominent fe - k@xmath7 emission line .
although @xmath102 slightly varied , the low - energy shapes of the time - averaged spectra were not so strongly affected ( see figure [ fig : xspectra ] ) .
the highest and lowest 210 kev fluxes were @xmath8 erg s@xmath4 @xmath5 in epoch 4 and @xmath113 erg s@xmath4 @xmath5 in epoch 7 , respectively .
they are in a range of just 7%70% of the 210 kev flux of @xmath114
erg s@xmath4 @xmath5 averaged over 19972002 ( _ rxte _ agn timing & spectral database ; maoz et al .
2002 ) ; thus ngc 3516 was in an x - ray faint phase during the present observations .
table [ tab : specfit ] further gives the 210 kev fluxes of the hard pl continuum and the reflection components , calculated separately , after removing the absorption .
the 210 kev light curves of the two spectral components , derived in this way , are presented in figure [ fig : xolc](top ) .
the time of each data point refers to the middle epoch ( in mjd ) of that observation as given in table [ tab : xobs ] .
interestingly , the flux of the reflection component significantly changed by a factor of 2 on a time scale of several months . however , the hard pl component varied almost by an order of magnitude or more in amplitude . clearly , the large intensity change in figure [ fig : xccp ] was mostly carried by the hard pl variation .
the largest flux change of the hard pl emission in 210 kev is @xmath115 erg s@xmath4 @xmath5 in difference , and a factor of @xmath116 between epoch 4 and 7 . for further information about the agn activity in ngc 3516
, we also performed the spectral analysis including the energy band below 2 kev .
for this purpose we selected epoch 7 , where the agn was faintest .
in fact , as shown in figure [ fig : lowenespec ] , the spectrum on this occasion exhibits a prominent soft x - ray excess at @xmath117 kev , which is much less prominent in the other epochs . in order to examine
if the soft excess structure originates from the agn activity , we fitted the 0.510 kev spectrum at epoch 7 by two contrasting models ; ( a ) ` wabs * ( cutoffpl + pexmon ) + apec ` and ( b ) ` wabs * ( cutoffpl + pexmon + kdblur * reflionx ) ` . in model ( a ) , the soft excess is explained by ` apec ` ( smith et al . 2001 ) which represents thin - thermal plasma emission from the host galaxy , while in model ( b ) by a ` kdblur * reflionx ` model ( laor 1991 ; ross & fabian 2005 ) which describes relativistically - smeared reflection continuum . as a result ,
the fit with model ( a ) became successful with @xmath118 ( figure [ fig : lowenespec]a ) , in which the soft excess , apparently involving emission lines , is reproduced by the plasma emission model . on the other hand , the fits with model ( b ) was unsuccessful with @xmath119 , mainly due to lack of the emission lines at @xmath120 kev and a convex data shape in the 25 kev band ( figure [ fig : lowenespec]b ) .
therefore , the spectrum below 2 kev is dominated by the host galaxy emission , at least in epoch 7 .
this justifies us to limit the spectral studies to the energies above @xmath121 kev .
of course , the agn emission , when bright , probably dominates down to @xmath122 kev , but is strongly affected by ionized - absorption features at @xmath123 kev , as shown in figure [ fig : lowenespec](a ) . according to changes of the atmospheric seeing during an observation , the flux of an agn within an aperture changes differently from that of the host galaxy , leading to a significant uncertainty in measuring the flux variation of a faint agn hosted by a bright galaxy . to minimize such uncertainty in the photometry of the ngc 3516 nucleus , we therefore performed difference image photometry ( dip ; crotts 1992 ; tomaney & crotts 1996 ) .
as presented in figure [ fig : ngc3516_b_mint_6x6arcmin_log.ps ] for example , a reference image was created by stacking many images obtained at the same night with good seeing condition .
then , we matched the position , the photometric intensity , and the point - spread function ( psf ) of the reference image to those of each individual image and subtracted the former from the latter .
two nearby field stars , located at ( ra , dec ) @xmath124 ( 11:06:28.51 , + 72:35:46.2 ) and ( 11:07:00.41 , + 72:33:33.3 ) , were used as the reference to match the photometric intensity and the psf . the iraf _ psfmatch _ task was used for matching the psfs .
after that , for each individual image , the residual flux at the center of the galaxy was measured relative to the two nearby reference stars , with a circular aperture of @xmath125full - width at half maximum ( fwhm ) of the psf in diameter .
finally , the residual flux data at the same observing night were averaged to obtain the flux difference of the ngc 3516 nucleus at that epoch with respect to the reference image .
the photometric error was estimated from the ensemble scatter of the dip fluxes and the number of the images obtained at the same night .
these procedures were applied to the data set of each telescope , individually . because the observing epochs of the reference images used in the dip analysis depend on the telescopes , there can be systematic offsets between them .
systematic differences in the scaling factor among them are also possible because of the differences in the filter color term . to match the offset and the scaling factor among the four telescopes , we therefore performed a linear regression analysis to their @xmath1-band flux data sets of the same observing nights .
an example , between two telescopes , is presented in figure [ fig : kwfc_mint_ff.ps ] .
the reduced @xmath110 of the linear regression was much larger than unity when only the photometric errors were incorporated . because this suggests the presence of some systematic errors , we added an error @xmath126 to the photometric errors in quadrature , and regarded it as the systematic error of the photometry . by requiring the reduced @xmath110 to become unity , @xmath127 mjy was obtained .
the @xmath1- and @xmath17-band magnitudes of the two reference stars were calibrated relative to those of the more distant field stars whose magnitudes are presented in sakata et al .
( 2010 ) ; these are based on the wide - field image data of the kwfc , in which both stars were observed simultaneously .
the @xmath20-band magnitudes of the reference stars were estimated from their @xmath1- and @xmath17-band magnitudes according to jordi et al .
( 2006 ) .
the light curves of the ngc 3516 nucleus , thus derived in the optical @xmath1 and @xmath20 bands , are presented in the bottom panel of figure [ fig : xolc ] , and the data are listed in table [ tab : optical_fluxes ] , without correction for the galactic extinction .
they are both the differential fluxes with respect to the reference image obtained from the image data on 2014 april 08 ( utc ) , and the error bars of the data points do not include @xmath126 .
the total numbers of the photometric data points are 180 and 6 for the @xmath1 and the @xmath20 bands , respectively .
as shown in figure [ fig : xolc ] , the @xmath1-band flux varied rather similarly to that of the hard pl component in the 210 kev band .
although the number of the data points are small , the @xmath20-band flux variation also followed them .
the larger relative scatters of the @xmath20-band data points are caused by the small telescope size and the large pixel scale of the camera ( mitsume ) by which the data were obtained .
these results indicate that the optical continuum and the 210 kev pl component varied in a good correlation with each other . between the peak and bottom , the @xmath1-band flux density varied by @xmath128 mjy . after correcting for the galactic extinction and the optical extinction of the ngc 3516 nucleus ( @xmath129 mag in total ) ,
it becomes @xmath130 mjy , corresponding to @xmath131 erg s@xmath4 @xmath5 in @xmath132 units.@xmath133 mag , which can be converted to @xmath134@xmath135 mag , and denney et al .
estimated it at @xmath136@xmath137 mag .
we adopted their average here .
although the @xmath138 @xmath5 derived from the x - ray absorption suggests much larger optical extinction of @xmath139 mag , agns often show smaller optical extinction than that converted from the x - ray @xmath102 ( e.g. , burtscher et al . 2015 ) . ]
lcccc[t ] 56316.544 & p & @xmath1 & 0.738 & 0.035 + 56334.674 & p & @xmath1 & 0.519 & 0.003 + 56382.554 & n & @xmath1 & 0.462 & 0.011 + 56382.588 & h & @xmath1 & 0.133 & 0.144 + 56383.576 & p & @xmath1 & 0.350 & 0.021 + 56384.454 & p & @xmath1 & 0.274 & 0.008 + 56385.448 & k & @xmath1 & 0.278 & 0.007 + 56385.611 & h & @xmath1 & 0.103 & 0.115 + & & & + 56392.487 & k & @xmath1 & 0.303 & 0.011 + 56392.625 & m & @xmath20 & 0.541 & 0.029 + 56392.686 & n & @xmath1 & 0.341 & 0.012 + & & & [ tab : optical_fluxes ] to complete this dip analysis , the agn flux at the faintest phase during our observation was estimated .
we applied an aperture photometry to the stacked @xmath1-band image obtained on 2014 april 7 by the mint attached on the nayuta telescope ( figure 5 ) , because it achieved , on this day , the best seeing among the four .
sakata et al . ( 2010 ) estimated the host galaxy flux in @xmath1 band as @xmath140 mjy , mag according to schlegel , finkbeiner & davis ( 1998 ) , we converted their host galaxy flux to that assuming @xmath141 mag adopted in this paper . ] within an aperture diameter of @xmath142 arcsec with the sky reference of a @xmath143@xmath144 arcsec annulus . by applying the photometry with the same parameters
, we obtained the @xmath1-band flux density of @xmath145 mjy after the correction for the galactic extinction , yielding the agn flux of @xmath146 mjy . considering the various systematic errors in the photometry , the @xmath1-band agn flux at the faintest phase during the observations is estimated as about a few times @xmath147 mjy , with a similar amount of flux error .
as shown in figure [ fig : xolc ] , the optical flux changed almost simultaneously with the 210 kev hard pl flux . in order to examine possible time lags between them , we first focus on their expanded light curves from epoch 1 to 5 , as presented in figure [ fig : xolc1 - 5](a ) .
clearly , the agn became the brightest in both bands at epoch 4 . also plotted in the same figure
is the x - ray light curve shifted by @xmath148 days and @xmath149 days .
apparently , the x - ray light curve with a few days shift shows the closest agreement with the @xmath1-band light curve , suggesting that the variation of the x - rays preceded that of the @xmath1 band by a few days .
figure [ fig : xolc1 - 5](b ) presents the correlation of the x - ray flux with @xmath150 , @xmath148 , and @xmath149 days temporal shifts , against the @xmath1-band flux at the delayed epochs .
the latter was estimated by averaging the @xmath1-band fluxes in close observations within @xmath1511 day .
the correlation between those fluxes appears to be strongest when the x - ray light curve is shifted by @xmath148 days .
this reconfirms that the x - ray variations preceded that of the @xmath1 band by a few days .
below , let us quantify the suggested time lag with two methods .
one method is the interpolated cross - correlation function ( iccf ) method , which has been widely used ( white & peterson 1994 ; peterson et al . 1998 ) .
this is the same as the ordinary cross - correlation function ( ccf ) , but either or both of the time - series data are interpolated , so that the ccf can be calculated even if the two data sets have different samplings , or when either or both are irregularly sampled .
since the monitoring cadence of the @xmath1-band data was much higher than that of the x - ray data , only the @xmath1-band light curve was interpolated , to make it nearly continuous .
the interpolated @xmath1-band light curve was calculated using a fitting code developed by zu et al .
( 2011 ) , which assumes a damped random walk ( drw ) model for the flux variation .
the drw model has been demonstrated to be a good statistical model of flux variations of the uv - optical continuum emission of agns ( e.g. , kelly et al . 2009 ; kozlowski et al .
2010 ; macleod et al .
2010 , 2012 ; zu et al .
2013 ) . in this interpolation , the systematic error @xmath126 of the photometry , either as determined in 3.2 or somewhat changed , was added in quadrature .
figure [ fig : blc_sys ] presents the observed @xmath1-band flux data and their interpolation .
thus , the interpolated light curve does not follow the observed data when @xmath152 mjy as determined in 3.2 is employed : the peak flux of the interpolated light curve was lower , and its flux decrease after the peak was slower .
we generally found that a larger value of @xmath126 made the interpolated light curve smoother and less variable , probably because a larger @xmath126 would work as if applying a stronger low - pass filter .
therefore , we performed the ccf analysis using the @xmath1-band data with @xmath153 mjy in addition to 0.06 mjy to examine the possible uncertainty in lag caused by @xmath126 .
as shown in figure [ fig : blc_sys ] , the interpolated light curves then became to follow the observed data much better when @xmath126 is reduced .
the x - ray data of epoch 15 and the optical data of mjd @xmath154@xmath155 were selected for the ccf analysis , because the most remarkable flux variations were present in those epochs , and also because the @xmath1-band light curve data were sampled densely ( figure [ fig : xolc ] and [ fig : xolc1 - 5 ] ) .
the calculated ccfs , in which the @xmath1-band data were smoothed using @xmath156 mjy and 0.06 mjy , are presented in figure [ fig : cccd ] , in which the time lag @xmath157 is defined to be positive if the optical precede x - rays .
as shown in figure [ fig : cccd ] , the ccf is peaked at about @xmath158 days .
this reconfirms the inference from figure [ fig : xolc1 - 5 ] , and implies that the optical variation is delayed from that in x - rays by @xmath159 days .
the ccf value at the peak is very high , @xmath160 , as indicated by figure [ fig : xolc1 - 5](b ) . as quantitative measurement of @xmath157
, we use the centroid of the ccf peak , @xmath161 , which is computed from all neighboring points around the ccf peak where ccf values are @xmath162 times that of the peak .
the uncertainty of @xmath161 is estimated using the model - independent monte carlo method of flux randomization and random subset sampling ( fr / rss ) introduced by peterson et al .
( 1998 , 2004 ) .
the fr method modifies the observed fluxes in each realization randomly within the errors assigned to the individual data points , and the rss method randomly extracts the same number of data points from the observed light curve allowing for multiple extraction .
then , the ccf and @xmath161 are calculated for each realization in the same way , to produce the cross - correlation centroid distribution ( cccd ) .
the cccds calculated by 5000 realizations with the fr / rss method are presented in figure [ fig : cccd ] , and the derived @xmath161 and its uncertainties are listed in table [ tab : lag ] .
thus , the results with @xmath156 mjy and 0.06 mjy agree well with each other . on average , the lag has been estimated as @xmath163 days ( 1@xmath164 error ) , and the time lag is significantly non - zero , because the @xmath165%-confidence limit is @xmath166 days .
the other method of the lag estimation is the javelin software developed by zu et al .
( 2011 ) , which is widely employed not only in recent reverberation studies for the optical broad emission lines and the thermal dust emission of agns ( e.g. , grier et al . 2012 ; peterson et al . 2014
; koshida et al .
2014 ) , but also in the lag analysis between flux variations of their x - ray emission and the uv - optical continuum emission ( shappee et al . 2014 ; mchardy et al . 2014
; lira et al . 2015 ) .
it explicitly builds a model of a response light curve that is expressed as a convolution of a source light curve by a top - hat transfer function with a certain lag , and fits the model to the data of flux variations in different bands , one as the source light curve and the others as the response light curves .
the source light curve is modeled as a stochastic process using a drw model , and the posterior distributions of the time lag as well as other model parameters are estimated using the bayesian markov chain monte carlo ( mcmc ) method .
ccccccc[t ] @xmath161 & @xmath167 & @xmath168 & @xmath169 & @xmath170 & @xmath171 & @xmath172 + & @xmath173 & @xmath174 & @xmath175 & @xmath176 & @xmath177 & @xmath178 + @xmath179 & @xmath167 & @xmath180 & @xmath181 & @xmath182 & @xmath183 & @xmath184 + & @xmath173 & @xmath185 & @xmath186 & @xmath187 & @xmath188 & @xmath189 [ tab : lag ] we use the @xmath1-band light curve as the source , whereas the x - ray data as the response , again with @xmath190 meaning x - ray delays .
the x - ray data of epoch 15 and the optical data of mjd @xmath154@xmath155 were selected for the javelin analysis , and @xmath156 mjy or 0.06 mjy was added to the @xmath1-band photometric errors in quadrature , just as in the ccf analysis .
the posterior distribution of the time lag calculated by 250000 realizations of the javelin software are presented in figure [ fig : javelin ] , and the resultant @xmath179 and its uncertainties are also listed in table [ tab : lag ] .
the two values of @xmath126 again gave very similar results .
the lag times were estimated as @xmath191 days and @xmath192 days ( 1@xmath164 error ) with @xmath156 and 0.06 , respectively .
again , we can exclude the case of @xmath193 , because the @xmath165%-confidence limit is @xmath194 days on average . in summary
, the two methods have given consistent results .
as shown in table [ tab : specfit ] and figure [ fig : xolc ] ( top ) , we found a significant flux variation in the x - ray reflection component as well , on a time scale of several months , although its variation amplitude is much smaller than that of the primary x - rays .
this indicates that the source region of the reflection component has an extent on a scale of @xmath195 pc .
moreover , the flux variation of the reflection component slightly lags behind that of the hard pl component , on a scale of a week , or possibly larger because the x - ray sampling becomes very sparse after epoch 5 . according to the unified model of agns , the reflection component accompanied by the neutral fe - k@xmath7 line
is supposed to arise from a dust torus , but different origins such as outer accretion disks , and the broad emission - line region are also suggested ( e.g. awaki et al . 1991 ; yaqoob & padmanabhan 2004 ; nandra 2006 ; jian et al .
2011 ; minezaki & matsushita 2015 ; gandhi et al . 2015 ) .
interestingly , the reverberation studies of ngc 3516 yielded the dust lag of @xmath196 days ( koshida et al .
2014 ) , and the lag of broad balmer emission lines of @xmath197 days ( peterson et al . 2004 ; denney et al . 2010 ) .
they are comparable to the time scale of the variation of the x - ray reflection , and its delay behind the hard pl component . in order to further examine the origin of the x - ray reflection component ,
a direct comparison of its variation with that of the dust emission and the broad balmer emission lines would be necessary ; this will be discussed in a forthcoming paper .
in the present study , we performed an x - ray optical simultaneous monitoring of the type 1.5 seyfert galaxy ngc 3516 with _ suzaku _ and japanese ground - based telescopes , the pirka , kiso schmidt , mitsume , nayuta , and kanata telescopes . by applying the spectral fitting to the x - ray data and differential image photometry to the @xmath1-band images , and quantitatively comparing the x - ray and optical flux variations ,
we have obtained the following results . * during our observations , ngc 3516 was in an x - ray - faint state .
it was faintest in epoch 7 , when the 210 kev flux became @xmath198 erg s@xmath4 @xmath5 , which is just 5% of the average flux recorded in 19972002 with _
rxte_. even when brightest ( epoch 4 ) , the 210 kev flux was @xmath199 erg s@xmath4 @xmath5 , which is only 70% of the average in 19972002 .
the flux varied on time scales longer than @xmath29days , without any intraday changes . *
the 245 kev emission mainly consisted of two spectral components ; a variable hard power - law continuum with a photon index of @xmath200 , and a reflection component with a prominent narrow fe - k@xmath7 emission line . throughout the monitoring , the hard x - ray component kept almost the same spectral shape , and exhibited a peak - to - peak flux change by @xmath201 erg s@xmath4 @xmath5 , or by about an order of magnitude .
* the @xmath1-band flux density varied by @xmath29 4.0 mjy in peak to peak , which translates to a flux change by @xmath131 erg s@xmath4 @xmath5 in @xmath132 units , after correcting for the optical extinction . the @xmath1-band flux varied on time scales longer than @xmath29days , like those of the x - rays .
* x - ray and @xmath1-band flux correlation was significantly detected for the first time in ngc 3516 .
the flux changes of the hard x - ray component significantly preceded those in @xmath1-band by @xmath9 days ( @xmath10 error ) . by the previous monitoring of ngc 3516 conducted with _
rxte _ and the israeli _ wise _ telescope in 19972002 , no significant correlations were derived between x - ray and @xmath1-band flux variations , giving a cross correlation coefficient of @xmath202 ( maoz et al .
in contrast , we succeeded in detecting a significant correlation between them with a high cross correlation coefficient of @xmath203 .
what is responsible for the clear difference ?
maoz et al . (
2002 ) argued that the lack of correlation might be due to absorption changes that are independent of the primary x - ray variations .
however , the 210 kev flux varied during their monitoring by a factor of @xmath204 in peak to peak . such a large change would be hardly explained by the so - far recored variations in the column density of optically - thick neutral absorbers ( turner et al .
we hence suggest alternatively that the dominant x - ray variable component changed between the two monitoring campaigns .
noda et al .
( 2013 ) discovered that the x - ray emission of ngc 3516 comprises at least two different primary continua with distinct spectral shape and flux variability ; a flat ( @xmath2051.7 ) and steep ( @xmath206 ) spectral continua , which we hereafter call the hard and gradually - varying primary component ( hgpc ) , and the soft and rapidly - varying primary component ( srpc ) , respectively .
noda et al .
( 2013 ) showed that luminosities of the hgpc and srpc were comparable in a bright state in 2005 , when the absorption - corrected 210 kev total flux was @xmath207 erg s@xmath4 @xmath5 .
in contrast , the 2009 suzaku observation caught ngc 3516 in a faint state with @xmath208 erg s@xmath4 @xmath5 , wherein the variation was carried solely by a @xmath209 pl which can be identified with the hgpc .
thus , there is a certain threshold in between these two flux values , say , at @xmath210 erg s@xmath4 @xmath5 ; above @xmath211 the hgpc and srpc coexist , whereas only the hgpc remains below @xmath211 .
these two primary continua have also been identified in another seyfert ngc 3227 ( noda et al .
2014 ) : their luminosity - dependent behavior was found to be very similar to that observed from ngc 3516 .
namely , the hgpc was always present , whereas the srpc appeared when the source was brighter than a certain threshold , to become co - existent with the other . in the present _ suzaku
_ observations , the absorption - corrected 210 kev flux of ngc 3516 was in the range @xmath212 erg s@xmath4 @xmath5 , i.e. , always below the suggested @xmath211 . furthermore , the 245 kev spectrum of the main variable pl was flat with @xmath111 , and the ccp distributions are smoothly connected to those in 2009 ( figure [ fig : xccp ] ) .
hence , we conclude that ngc 3516 was in the faint phase , with the hgpc dominating just as in the 2009 observation . on the other hand , during the 19972002 monitoring by maoz et al .
( 2002 ) , the averaged 210 kev flux was @xmath213 erg s@xmath4 @xmath5 which is higher than @xmath211 , indicating that ngc 3516 was mostly in the bright phase .
if the srpc has a significantly poorer correlation with the optical than the hgpc does , the correlation between the total x - ray flux and the optical should become worse when ngc 3516 is in the bright phase .
thus , the difference between maoz et al .
( 2002 ) and ours may be explained by presuming that the hgpc is well correlated with the optical while the srpc is not . to generalize the above conclusion , we collected cross correlation coefficients ( cccs ) from peak values of the ccf in previous simultaneous observations of various seyferts . as a result
, sources with high @xmath214 ratios were found to have relatively low cccs ; mcg6 - 30 - 15 with @xmath215 has @xmath216 ( lira et al .
2015 ) , ngc 4051 with @xmath217 has @xmath218 ( breedt et al .
2009 ) , and mr 2251 - 178 with @xmath215 has @xmath219 ( arevalo et al .
in contrast , sources with low @xmath214 ratios possibly have high cccs ; ngc 6814 with @xmath220 has @xmath221 ( troyer et al .
2015 ) ; and ngc 3516 in the present study has @xmath222 wherein @xmath223 is indicated by the value of 0.00612 ( vasudevan & fabian 2009 ) , obtained in 2001 when ngc 3516 showed @xmath224 close to the second highest among ours .
therefore , we speculate that x - ray optical correlation in seyfert galaxies becomes commonly worse when the eddington ratio gets higher , because the srpc start dominating .
according to figure 12(b ) and related descriptions in noda et al .
( 2014 ) , the srpc may originate from patchy coronae heated by magnetic process on the surface of a disk ( e.g. , reynolds & nowak 2003 ) . if covering fraction of the patchy coronae against the disk is small , the srpc generated in the coronae can be related to just small areas of the disk , making x - ray and optical variations independent .
another possible origin of the srpc is a faulty jet at the accretion axis of the smbh ( e.g. , ghisellini et al .
if the jet emission region is located far away from the accretion disk as suggested by noda et al .
( 2014 ) , connection between the srpc and the disk optical emission may become weaker , making their correlation poorer . in any case
, the emergence and disappearance of these srpc - generating regions are likely to be more localized effects than those responsible for the hard vs. soft state transitions seen in black - hole binaries , because the agn state changes considered here take place on time scales of weeks to years , which are much shorter than would be expected ( e.g. , several ten thousand years ) if the typical state - transition time scales , @xmath29days , are scaled to the mass ratios .
the measured optical lag of @xmath9 days strongly supports the x - ray reprocessing model ( e.g. , krolik et al .
1991 ; cackett et al .
2007 ) ; optical brightening occurred , at least in the present case , through irradiation by the increased hard x - ray intensity .
this process is energetically feasible , because the 210 kev x - ray flux increase by @xmath225 erg s@xmath4 @xmath5 is comparable to the @xmath1-band flux increase by @xmath226 erg s@xmath4 @xmath5 .
the optical lag time in ngc 3516 obtained here is similar to those of a few days observed from other seyfert galaxies , including ngc 5548 ( e.g. , suganuma et al .
2006 ; mchardy et al . 2014 ; edelson et al .
2015 ) , ngc 3783 ( arevalo et al .
2009 ) , ngc 4051 ( breedt et al . 2010 ) , and ngc 6418 ( troyer et al . 2015 ) .
therefore , the suggested mechanism , i.e. , the x - ray reprocessing , can be operating commonly among these seyfert galaxies .
let us , then , investigate whether or not the derived optical lag can be explained by the most commonly adopted x - ray reprocessing model ( e.g. , cackett et al .
2007 ) , which assumes that an optically - thick geometrically - thin standard accretion disk continues down to the innermost last - stable circular orbit ( isco ) , located at @xmath227 where @xmath228 is the schwarzschild radius and @xmath229 the black - hole mass , and is illuminated by a lamppost - type x - ray source proximate to the black hole . at radii @xmath230
, the radial temperature profile of the disk is given as ( shakura & sunyaev 1973 ) @xmath231 where @xmath232 is the mass accretion rate , @xmath233 is the gravitational constant , and @xmath37 is the stefan - boltzmann constant . after the delay time @xmath157 ,
the x - ray flux variation that arose close to the black hole will propagate to a radius @xmath234 , and will increase the emissivity there .
let us define a wavelength @xmath235 in such a way that the continuum emission at this radius is peaked at @xmath236 , where @xmath237 is a numerical factor of order unity .
although the wien s displacement law for a blackbody radiation in @xmath238 gives @xmath239 , we here set @xmath240 , according to the detailed calculations of the disk response to continuum emission by cackett et al .
( 2007 ) and collier et al .
( 1999 ) . substituting @xmath241 for @xmath242 , and expressing @xmath243 with @xmath236 , eq .
( 1 ) can be rewritten as @xmath244 by eliminating @xmath245 from eq .
( 1 ) and eq .
( 2 ) , we obtain @xmath246 . further expressing @xmath232 by the bolometric luminosity of the accretion disk , @xmath247 ( assuming that the disk extends to the isco ) , and normalizing it to the eddington luminosity ( @xmath248 ) , the eddington ratio of the source @xmath249 is estimated as @xmath250 equation ( 3 ) , in fact , implies a serious problem .
first of all , it indicates an unrealistic super - eddington " condition .
even putting aside this issue , it means a factor of @xmath251 discrepancy against the likely value of @xmath252 ( 4.2 ) , which characterizes the ngc 3516 nucleus during the present observations .
the difficulty can be more directly stated : by summing up the blackbody contributions from various disk annuli , the continuum spectrum from the accretion disk under consideration can be calculated as @xmath253 where @xmath254 is the disk inclination ( collier et al . 1999 ; cackett et al .
thus , the measured delay predicts very bright optical emission , arising from the inner part of the accretion disk . in other words ,
the smbh would have to be accreting with a very high rate , in order for its accretion disk to emit the @xmath1-band light from such a large distance as @xmath159 light days .
equation ( 4 ) means a two - orders - of - magnitude contradiction to the present optical data , that the peak - to - peak flux variation in the @xmath1 band was @xmath130 mjy , and the minimum @xmath1-band flux during the observation would be less than that .
inversely , if we started from the assumption of @xmath255
@xmath256 , the @xmath1-band lag predicted by eq.(3 ) would become @xmath257 times smaller than was measured .
it is thus extremely difficult to reconcile the measured clear optical delay by @xmath159 days with the observed optical faintness , as long as we assume a standard disk extending down to the isco and an illuminating x - ray source close to the smbh . in the x - ray reprocessing model , we should consider not only the viscous heating but also the heating by the x - ray irradiation .
the former , which underlies eq.(1 ) , is written as @xmath258 whereas the latter is given as @xmath259 where @xmath260 is the x - ray luminosity , @xmath261 is the disk albedo , @xmath262 is the distance from the disk to the x - ray illuminator , and @xmath263 is the x - ray incidence angle onto the disk .
collier et al .
( 1999 ) and cackett et al .
( 2007 ) assumed that the x - ray emitter is located on the rotating axis of the smbh like a `` lamppost '' ( figure [ fig : model]a ) , and its height @xmath264 from the smbh is much smaller than @xmath242 ( @xmath265 ) . under these assumptions ,
we obtain @xmath266 and @xmath267 .
then , @xmath268 and @xmath269 have just the same @xmath242-profile , and the total heating per unit disk face area , @xmath270 , is described as @xmath271 as a result , the radial temperature profile of the disk is now given by @xmath272^{\frac{1}{4}}r^{-\frac{3}{4}}.\ ] ] it exhibits the same @xmath242-dependence as the standard viscous disk , namely , @xmath273 .
furthermore , because of the @xmath274 proportionality , we should use @xmath275 in place of @xmath232 when converting this expression into that of @xmath249 as eq.(3 ) .
consequently , the eddington ratio @xmath249 and the @xmath1-band flux density can be given by the same eq.(3 ) and eq.(4 ) , respectively .
thus , we encounter just the same problem as before , even when the x - ray irradiation is taken into account . according to e.g. , cackett et al .
( 2007 ) , mchardy et al .
( 2014 ) , and edelson et al .
( 2014 ) , typical seyfert galaxies including ngc 5548 , ngc 4051 , and mrk 335 yielded @xmath157 that are a few times larger than those expected from their estimated @xmath249 . according to mogan et al .
( 2010 ) , the accretion disk sizes of gravitationally lensed quasars are larger , by a factor of @xmath204 , than those expected from @xmath249 , just like in the seyfert galaxies .
these subtle but systematic contradictions between the disk size and @xmath249 have also been well known in the x - ray reprocessing scenario ( e.g. , collier et al .
1999 , cackett et al .
recently , troyer et al . (
2015 ) reported that ngc 6814 exhibited @xmath157 which is @xmath276 times larger than that predicted from the estimated @xmath249 . in the present case of ngc 3516 , the measured vs. predicted difference in the optical delay amounts to a factor of @xmath277 , which is considerably larger than those of ngc 5548 , ngc 4051 , and mrk 335 .
these systematic discrepancies imply that the lamppost - type x - ray irradiation geometry shown in figure [ fig : model](a ) has a common problem , which is considered to become more prominent towards lower luminosities , considering that ngc 3516 and ngc 6814 have both relatively low values of @xmath249 .
thus , we need to revise the geometrical assumptions , in order to make @xmath157 and @xmath249 consistent in low eddington - ratio seyfert galaxies . what kind of geometry of the corona and the accretion disk can reconcile @xmath249 and @xmath157 from the present study ?
the contradiction in the lamppost - type x - ray reprocessing model , revealed in 4.3.1 , arises mainly because the observed luminosity is too low for the size of the accretion disk determined by the observed time lag .
conversely , the measured @xmath278 days is too large to be explained by the standard scenario , given the very low luminosity . in this respect
, it has been very essential that the present observations caught the object at a very faint state not only in x - rays but also in the optical band ( see the final paragraph in 3.2 ) .
this indicates that the @xmath279 inconsistency can not be solved even if considering an anisotropically - irradiating lamppost corona ( dabrowski et al .
1997 ) that would make the x - ray flux we observe much lower than that illuminating accretion disk . hence , we need to consider other geometries of the corona and the accretion disk that can reconcile @xmath249 and @xmath157 from the present study .
one possible variant of the lamppost - type scenario is to place the x - ray source at a large height from the smbh , to achieve @xmath280 light days . in this geometry ,
x - ray flux variations can still precede that in the optical .
furthermore , the inner part of the accretion disk can be made rather cool , because @xmath232 can be reduced and @xmath281 gets much smaller than that with the assumption of @xmath265 .
these would make the black body flux low enough to be consistent with the observed optical faintness .
such a lamppost - type corona with @xmath282 was suggested to form at the base of a jet moving away from a smbh with relativistic speeds ( e.g. , lohfink et al .
2013 ; wilkins & gallo 2015 ) .
however , the value of @xmath280 light days , or @xmath283 , which is required to explain the present results , is orders of magnitude larger than invoked in the previous studies
. it would be highly unrealistic to assume that such an emission region forms at a height of @xmath283 in this radio - quiet object , and that the region radiates essentially the _ entire primary x - ray emission _ ( 3.1 ) .
in addition , the x - ray emission mechanism from such a region is problematic .
as observed from bl - lac objects , synchrotron radiation would give an x - ray spectral slope of @xmath284 ( e.g. , tanihata et al .
2004 ) , which is much steeper than we observed ( @xmath111 ) .
comptonization of soft photons from the accretion disk would produce _ x - ray delays _ , contrary to the observed optical delay . in short ,
this variant scenario fails to explain the present results .
as we noted repeatedly , the x - ray reprocessing model which has so far been considered ( figure [ fig : model]a ) assumes the optically - thick accretion disk to extend down to near the isco .
however , it has long been known theoretically ( e.g. , abramowicz et al . 1995 ) that a decreased luminosity , e.g. , to @xmath285 , causes inner regions of such a disk to make a transition into an optically - thin and geometrically - thick hot flow .
this will lead to a configuration shown in figure [ fig : model](b ) ; the disk is truncated at a radius much larger than the isco , and the accretion flow inside that radius may form a radiatively inefficient accretion flow , or riaf ( e.g. , yuan & narayan 2014 ) . such bimodal behavior of accretion disks has been firmly verified through extensive observations of black - hole binaries ( e.g. , done et al .
the truncated disk picture has also been investigated by many authors , both observationally and theoretically , to explain low - luminosity agns ( e.g. ho 2008 ; taam et al .
2012 ; nemmen et al . 2014 ; and references therein ) .
it is hence much more natural to try to explain the present results based on the geometry of figure [ fig : model](b ) .
referring to figure [ fig : model](b ) , let us then consider an accretion disk that is truncated at a radius @xmath286 light days ( or @xmath251 times there schwarzschild radii ) from the smbh .
let us also assume that the accreting matter inside this radius forms a riaf region , where the illuminating hard x - ray photons are produced via comptonization ( noda et al . 2013a , 2014 ; makishima et al .
2008 ; yamada et al . 2013 ) .
in such a hot riaf region , the electron density and temperature are theoretically predicted to increase inwards towards the smbh ( e.g. , esin et al .
1997 ) , whereas the seed photon flux for inverse comptonization obviously increases towards positions closer to the disk .
hence , the x - ray emissivity are considered to become maximum at a location in between the smbh and the inner disk edge .
then , the optical delay of @xmath287 days can be explained simply as the light travel time from the x - ray brightest regions to the inner disk edge .
of course , this scenario requires a relatively low inclination , in order not to produce large differences in the light travel time to us , from the near side and far side of the disk ; ngc 3516 may satisfy the condition , because of a relatively low inclination value of @xmath288 ( wu & han et al .
2001 ) .
the truncated - disk picture can naturally explain , at the same time , the optical faintness , because the large amount of continuum radiation originating in the inner part of the accretion disk is no longer present .
the @xmath1-band to x - ray flux ratio of ngc 3516 in these epochs is @xmath289 .
it is much lower than that of the composite spectrum for radio - quiet quasars ( @xmath290 ; shang et al .
2011 ) , but is consistent with those of the composite and model spectra for low - luminosity agns ( ho 2008 ; nemmen et al . 2014 ) . to be quantitative ,
let us assume , in eq .
( 6 ) , @xmath291 erg s@xmath4 , @xmath292 , @xmath293 light days , and @xmath294 .
then , the illuminating x - ray flux at a distance of 2.0 light days from the x - ray - bright region is calculated as @xmath295 erg s@xmath4 @xmath5 during the peak of the flux variation .
it predicts the disk temperature at the inner radius to be @xmath296k wherein @xmath1-band photons can be produced on the wien side , even when the viscous heating is ignored in eq .
( 8) . incidentally , the disk temperature would increase by a factor of @xmath297 if the disk extended down to the isco . as another merit of the truncated - disk scenario ,
the gradual x - ray variations without any intraday changes can be explained as well by the large volume of the riaf region .
in addition , the observed hard x - ray spectrum with @xmath298 can be naturally explained as arising from the hot riaf region via thermal comptonization , as theoretically predicted ( esin et al .
1997 ) , and observed from black - hole binaries ( e.g. , makishima et al .
moreover , as shown in figure [ fig : lowenespec ] , the soft x - ray band at epoch 7 was dominated by the galactic thin - thermal plasma emission , and no relativistically - smeared spectral components were required , also supporting that the standard accretion disk did not continue down to the isco .
this scenario may be able to solve the @xmath157 vs. @xmath249 conflict in the other low-@xmath249 seyferts as well , including ngc 6814 in particular ( troyer et al .
we hence prefer the picture of the truncated disk and riaf shown in figure [ fig : model](b ) . how does our model compare with other attempts on the same subject ?
recently , gardner & done ( 2016 ) proposed a model , that a geometrically - thick region , called the soft - excess region , is present at the inner edge of an accretion disk , and completely hides the hard x - ray corona .
far - ultraviolet photons , arising from the soft - excess region by the x - ray illumination , are reprocessed by optically - thick clouds distributing above the accretion disk , and produce optical variations .
this model is different from ours in that x - rays from the corona can not directly reach the accretion disk , and the optical variation is mainly due to the clouds rather than the disk .
presumably , our ngc 3516 results can not be explained by the model of gardner & done ( 2016 ) , because the 0.52 kev spectrum at epoch 7 is explained adequately by the galactic emission , without any strong soft excess resulting from the agn activity .
therefore , we prefer our model ( figure [ fig : model]b ) to theirs . for some quasars
, it is claimed that the accretion disks , as measured by the uv - optical microlensing observations , are generally several times larger than those predicted by the standard accretion disk model ( e.g. , morgan et al .
these measurements are thought to be observing the unexpectedly large disk size like those revealed by the x - ray - to - optical continuum reverberation mapping for nearby seyfert galaxies ( e.g. , edelson et al .
2015 , mchardy et al .
dexter & agol ( 2011 ) suggested that the large disk size of quasars from microlensing is due to stochastic and strong local temperature fluctuations on the effective temperature profile of the standard accretion disk .
their inhomogeneous disk model is designed to enhance the flux contribution from outer disk radii , and thus make the disk half - light radius larger than the predictions of the standard accretion disk .
however , their model , which does not consider the x - ray emission and attributes all uv optical variability to the assumed temperature fluctuations , can not explain the strong x - ray vs. optical correlations observed in ngc 3516 and other seyfert galaxies .
furthermore , kokubo ( 2015 ) showed that the disk model in dexter & agol ( 2011 ) can not explain , either , the tight inter - band correlations often observed in the optical light curves of quasars .
therefore , we do not discuss about the model any further . for further refinement of the truncated - disk model , and its demarcation from the large-@xmath264 model , correlations and time lags among different optical bands , including @xmath299 ,
@xmath17 , and @xmath242 bands become important ( e.g. , kokubo et al .
2014 ; edelson et al . 2015 ; fausnaugh et al .
2015 ; kokubo 2015 ) . these will be discussed in elsewhere .
the present coordinated x - ray and optical observations of ngc 3516 found the object in a very faint state in both wavelengths .
a tightly correlated intensity change lasting for @xmath300 days was detected , wherein the optical @xmath1-band variation showed a clear delay by @xmath159 days behind that of the x - ray continuum which is described by a power - law model with @xmath298 .
this optical lag indicates the effect of x - ray reprocessing , but the delay can not be reconciled with the @xmath1-band faintness , as long as we consider the standard reprocessing scenario which invokes a standard accretion disk irradiated by a lamppost - type x - ray source . instead , the observed results can be consistently explained assuming that the disk is truncated at @xmath283 , and turns into a riaf where the illuminating hard x - rays are produced via thermal comptonization .
we thank the referee for his / her valuable comments and suggestions .
this work was partly supported by the grants - in - aid with grant numbers 26800095 ( hn ) , 25287062 ( tm ) , and 15j10324 ( mk ) , from the japan society for the promotion of science ( jsps ) .
hn was supported by the special postdoctoral researchers program in riken .
abramowicz , m. a. , chen , x. , kato , s. , lasota , j .-
p . , & regev , o. 1995 , , 438 , l37 arvalo , p. , uttley , p. , lira , p. , et al .
2009 , , 397 , 2004 arvalo , p. , uttley , p. , kaspi , s. , et al .
2008 , , 389 , 1479 awaki , h. , koyama , k. , inoue , h. , & halpern , j. p. 1991
, , 43 , 195 balbus , s. a. , & hawley , j. f. 1991 , , 376 , 214 boldt , e. , 1987 , phys .
, 146 , 215 breedt , e. , mchardy , i. m. , arvalo , p. , et al . 2010 , , 403 , 605 burtscher , l. , davies , r. i. , gracia - carpio , j. , et al .
2015 , accepted for publication by a&a ( arxiv:1511.05566 ) cackett , e. m. , horne , k. , & winkler , h. 2007 , , 380 , 669 costantini , e. , nicastro , f. , fruscione , a. , et al .
2000 , , 544 , 283 crotts , a. p. s. 1992 , , 399 , l43 dabrowski , y. , fabian , a. c. , iwasawa , k. , lasenby , a. n. , & reynolds , c. s. 1997 , , 288 , l11 dauser , t. , garcia , j. , wilms , j. , et al . 2013 , , 430 , 1694 denney , k. d , peterson , b. m. , pogge , r. w. , et al .
2010 , , 721 , 715 dexter , j. , & agol , e. 2011 , , 727 , l24 dickey , j. m. , & lockman , f. j. 1990 , , 28 , 215 edelson , r. , gelbord , j. m. , horne , k. , et al .
2015 , , 806 , 129 elvis , m. , wilkes , b. j. , mcdowell , j. c. , et al .
1994 , , 95 , 1 esin , a. a. , mcclintock , j. e. , & narayan , r. 1997 , , 489 , 865 fabian , a. c. , miniutti , g. , gallo , l. , et al .
2004 , , 353 , 1071 fausnaugh , m. m , denney , k. d. , barth , a. j. et al . 2015 , arxiv:1510.05648 fukazawa , y. , mizuno , t. , watanabe , s. , et al . 2009 , , 61 , 17 gandhi , p. , hnig , s. f. , & kishimoto , m. 2015 , , 812 , 113 gardner , e. , & done , c. 2016 , arxiv:1603.09564 gaskell , c. m , & peterson , b. m. , 1987 , , 65 , 1 gaskell , c. m. 2006 , astronomical society of the pacific conference series , 360 , 111 george , i. m. , turner , t. j. , netzer , h. , et al .
2002 , , 571 , 265 ghisellini , g. , haardt , f. , & matt , g. 2004 , , 413 , 535 gierliski , m. , & done , c. 2004 , , 349 , l7 grier , c. j. , peterson , b. m. , pogge , r. w. , et al .
2012 , , 755 , 60 gruber , d. e. , matteson , j. l. , peterson , l. e. , & jung , g. v. 1999 , , 520 , 124 ho , l. c. 2008 , , 46 , 475 ishisaki , y. , maeda , y. , fujimoto , r. , et al .
2007 , , 59 , s113 jiang , p. , wang , j. , & shu , x. 2011 , scpma , 54 , 1354 jordi , k. , grebel , e. k. , & ammon , k. 2006 , , 460 , 339 kawabata , k. s. , nagae , o. , chiyonobu , s. , et al .
2008 , in society of photo - optical instrumentation engineers ( spie ) conference series , vol . 7014 , society of photo - optical instrumentation engineers ( spie ) conference serie kokubo , m. , morokuma , t. , minezaki , t. , et al .
2014 , , 783 , 46 kokubo , m. 2015 , , 449 , 94 koshida , s. , minezaki , t. , yoshii , y. , et al .
2014 , , 788 , 159 kotani , t. , kawai , n. , yanagisawa , k. , et al .
2005 , nuovo cimento c geophysics space physics c , 28 , 755 koyama , k. , tsunemi , h. , dotani , t. , et al .
2007 , , 59 , 23 krolik , j. h. , horne , k. , kallman , t. r. , et al .
1991 , , 371 , 541 laor , a. 1991 , , 376 , 90 lira , p. , arvalo , p. , uttley , p. , mchardy , i. m. m. , & videla , l. 2015 , , 454 , 368 lobban , a. p. , reeves , j. n. , porquet , d. , et al .
2010 , , 408 , 551 machida , m. , hayashi , m. r. , & matsumoto , r. 2000 , , 532 , l67 makishima , k. , takahashi , h. , yamada , s. , et al .
2008 , , 60 , 585 maoz , d. , markowitz , a. , edelson , r. , & nandra , k. 2002 , , 124 , 1988 mchardy , i. m. , cameron , d. t. , dwelly , t. , et al .
2014 , , 444 , 1469 mchardy , i. , connolly , s. , peterson , b. , et al .
2016 , arxiv:1601.00215 malizia , a. , molina , m. , bassani , l. , et al .
2014 , , 782 , l25 morgan , c. w. , kochanek , c. s. , morgan , n. d. , & falco , e. e. 2010 , , 712 , 1129 morrison , r. , & mccammon , d. 1983 , , 270 , 119 minezaki , t. , & matsushita , k. 2015 , , 802 , 98 miniutti , g. , & fabian , a. c. 2004 , , 349 , 1435 mitsuda , k. , bautz , m. , inoue , h. , et al .
2007 , , 59 , 1 miyakawa , t. , ebisawa , k. , & inoue , h. 2012 , , 64 , 140 nandra , k. , le , t. , george , i. m. , et al .
2000 , , 544 , 734 nandra , k. 2006 , , 368 , 62 nandra , k. , oneill , p. m. , george , i. m. , & reeves , j. n. 2007 , , 382 , 194 narayan , r. , & yi , i. 1994 , , 428 , l13 nemmen , r.?s .
, storchi - bergmann , t. , & eracleous , m. 2014 , , 438 , 2804 noda , h. , makishima , k. , uehara , y. , et al .
2011a , , 63 , 449 noda , h. , makishima , k. , yamada , s. , et al .
2011b , , 63 , s925 noda , h. , makishima , k. , nakazawa , k. , & yamada , s. 2013a , , 771 , 100 noda , h. , makishima , k. , nakazawa , k. , et al . 2013b , , 65 , 4 noda , h. , makishima , k. , nakazawa , k. , & yamada , s. 2013c , , 84 , 707 noda , h. , makishima , k. , yamada , s. , et al .
2014 , , 794 , 2 ozaki , s. 2005 , annual report of the nishi - harima astronomical observatory ( issn 0917 - 6926 ) , no . 15 , p. 614
( 2005 ) , 15 , 6 pei , l. , barth , a. j. , aldering , g. s. , et al .
2014 , , 795 , 38 peterson , b. m. , wanders , i. , horne , k. , collier , s. , alexander , t. , kaspi , s. , & maoz , d. , 1998 , , 110 , 660 peterson , b. m. et al .
, 2004 , , 613 , 682 peterson , b. m. , ferrarese , l. , gilbert , k. m. , et al .
2004 , , 613 , 682 peterson , b. m. , grier , c. j. , horne , k. , et al .
2014 , , 795 , 149 reis , r. c. , & miller , j. m. 2013 , , 769 , ll7 reynolds , c. s. , & nowak , m. a. 2003 , , 377 , 389 rivers , e. , markowitz , a. , & rothschild , r. 2013 , , 772 , 114 ross , r. r. , & fabian , a. c. 2005 , , 358 , 211 sakata , y. , minezaki , t. , yoshii , y. , kobayashi , y. , koshida , s. , aoki , t. , enya , k. , tomita , h. , suganuma , m. , katsuno uchimoto , y. , & sugawara , s. 2010 , , 711 , 461 sako , s. , aoki , t. , doi , m. , et al .
2012 , , 8446 , 84466l schlafly , e. f. , & finkbeiner , d. p. 2011
, , 737 , 103 schlegel , d. j. , finkbeiner , d. p. , & davis , m. 1998 , , 500 , 525 shakura , n. i. , & sunyaev , r. a. 1973 , , 24 , 337 shang , z. , brotherton , m. s. , wills , b. j. , et al .
2011 , , 196 , 2 shappee , b. j. , prieto , j. l. , grupe , d. , et al .
2014 , , 788 , 48 smith , r. k. , brickhouse , n. s. , liedahl , d. a. , & raymond , j. c. 2001 , , 556 , l91 suganuma , m. , yoshii , y. , kobayashi , y. , et al .
2004 , , 612 , 113 suganuma , m. , yoshii , y. , kobayashi , y. , et al .
2006 , , 639 , 46 taam , r. e. , liu , b. f. , yuan , w. , & qiao , e. , 2012 , , 759 , 65 takahashi , t. , abe , k. , endo , m. , et al .
2007 , , 59 , 35 terashima , y. , iyomoto , n. , ho , l. c. , & ptak , a. f. 2002 , , 139 , 1 tomaney , a. b. , & crotts , a. p. s. 1996 , , 112 , 2872 tanihata , c. , kataoka , j. , takahashi , t. , & madejski , g. m. 2004 , , 601 , 759 troyer , j. , starkey , d. , cackett , e. , et al .
2015 , arxiv:1509.01124 turner , t. j. , george , i. m. , nandra , k. , & mushotzky , r. f. 1997 , , 488 , 164 turner , t. j. , miller , l. , kraemer , s. b. , & reeves , j. n. 2011 , , 733 , 48 uttley , p. , edelson , r. , mchardy , i. m. , peterson , b. m. , & markowitz , a. 2003 , , 584 , l53 uttley , p. , cackett , e. m. , fabian , a. c. , kara , e. , & wilkins , d. r. 2014 , , 22 , 72 uttley , p. , & casella , p. 2014
, , 183 , 453 vasudevan , r. v. , & fabian , a. c. 2009 , , 392 , 1124 watanabe , m. , takahashi , y. , sato , m. , et al .
2012 , in society of photo - optical instrumentation engineers ( spie ) conference series , vol . 8446 , society of photo - optical instrumentation engineers ( spie ) conference series white , r. j. , & peterson , b. m. , 1994 , , 106 , 879 wu , x .- b . ,
& han , j. l. 2001 , , 561 , l59 yamada , s. , makishima , k. , done , c. , et al .
2013 , , 65 , yaqoob , t. , & padmanabhan , u. 2004 , , 604 , 63 yuan , f. , & narayan , r. 2014 , , 52 , 529 zu , y. , kochanek , c. s. , & peterson , b. m. 2011 , , 735 , 80 zu , y. , kochanek , c. s. , kozowski , s. , & udalski , a. 2013 , , 765 , 106 | from 2013 april to 2014 april , we performed an x - ray and optical simultaneous monitoring of the type 1.5 seyfert galaxy ngc 3516 .
it employed _ suzaku _ , and 5 japanese ground - based telescopes , the pirka , kiso schmidt , nayuta , mitsume , and the kanata telescopes .
the _ suzaku _
observations were conducted seven times with various intervals ranging from days , weeks , to months , with an exposure of @xmath0 ksec each .
the optical @xmath1-band observations not only covered those of _ suzaku _ almost simultaneously , but also followed the source as frequently as possible . as a result , ngc 3516 was found in its faint phase with the 210 kev flux of @xmath2@xmath3 erg s@xmath4 @xmath5 .
the 245 kev x - ray spectra were composed of a dominant variable hard power - law continuum with a photon index of @xmath6 , and a non - relativistic reflection component with a prominent fe - k@xmath7 emission line .
producing the @xmath1-band light curve by differential image photometry , we found that the @xmath1-band flux changed by @xmath8 erg s@xmath4 @xmath5 , which is comparable to the x - ray variation , and detected a significant flux correlation between the hard power - law component in x - rays and the @xmath1-band radiation , for the first time in ngc 3516 . by examining their correlation
, we found that the x - ray flux preceded that of @xmath1 band by @xmath9 days ( @xmath10 error ) .
although this result supports the x - ray reprocessing model , the derived lag is too large to be explained by the standard view which assumes a `` lamppost''-type x - ray illuminator located near a standard accretion disk .
our results are better explained by assuming a hot accretion flow and a truncated disk . |
it is well known that every ( anti)baryon except the ( anti)proton decays to a lighter ( anti)baryon , such as the beta decay of the neutron , @xmath6 , which is the simplest baryonic decay . however , up to now , no _ purely _ baryonic decay process , with only baryons involved , has yet been observed @xcite . by virtue of the baryon number conservation , in order to have a purely baryonic decay , at least three baryons in the final state are needed , e.g. , @xmath7 , where @xmath8 and @xmath9 represent heavy and light spin-1/2 baryons , respectively .
it is easy to show that the simplest , and lightest , possible purely baryonic decay process is @xmath10 without breaking any known conservation law .
other examples of such decays are @xmath2 and @xmath3 as well as the corresponding @xmath5 decays . since baryons are the main constituents of our matter world , their production and decay mechanisms should all be explored .
being one of the three conditions for the baryogenesis to explain the puzzle of the matter and antimatter asymmetry in the universe , @xmath11 violation has been a primary topic of study at the @xmath12 factories and at the lhcb experiment , among others . however , the unique physical @xmath11 phase in the cabibbo - kobayashi - maskawa ( ckm ) quark mixing matrix @xcite of the standard model ( sm ) is not sufficient to solve the mystery , leaving room for new physics in the formation of the matter world . on the other hand , for a direct connection to the baryonic contents of the universe ,
one expects to observe @xmath11 violation in purely baryonic processes .
theoretically and experimentally , though , the latter has , to our knowledge , never been studied before . clearly , it is interesting to discuss @xmath11 violating rate asymmetries as well as time - reversal violating spin involved triple correlations due to the rich spin structures in these purely baryonic decays , to test the sm and search for new physics manifestations . in the sm
, the decay @xmath10 , with the tree - level dominated contribution through the @xmath13 quark currents , can be factorized as a color - allowed process , which is insensitive to nonfactorizable effects , where the required matrix elements of the @xmath14 transition and the recoiled @xmath15 pair have been well studied
. a reliable prediction of the branching ratio is hence expected , which should be as large as those in the tree - level @xmath12 decays such as @xmath16 .
in addition , the threshold effect around the @xmath15 invariant mass spectrum , measured as a salient feature in three - body baryonic @xmath12 decays , could also enhance the contribution .
it is interesting to note that in this simplest purely baryonic decay @xmath10 , a possible intermediate resonant state such as @xmath17 is rather suppressed , which is estimated to have @xmath18 @xcite .
while the threshold effect , receiving the dominant contribution from the threshold of @xmath19 , has been commonly observed in the baryon pair production @xcite , the meson resonances or the final state interaction due to the multiparticle exchange , which deviates the baryon pair production from the threshold , should be suppressed . in this article , we concentrate on this simplest purely baryonic decay @xmath10 .
we will give the theoretical estimation of its decay branching ratio , and stimulate a possible measurement by the lhcb collaboration .
possible @xmath11 and @xmath20 violating effects in purely baryonic decays are also discussed .
decay.,title="fig:",width=254 ] decay.,title="fig:",width=240 ] decay.,title="fig:",width=240 ] decay.,title="fig:",width=254 ] in terms of the effective hamiltonian at the quark level for the charmless @xmath21 transition in the sm , the amplitude for @xmath10 in the factorization approach can be written as @xmath22 where @xmath23 is the fermi constant ; @xmath24 are the ckm matrix elements ; @xmath25 stands for @xmath26 ; and @xmath27 for @xmath28odd ( even ) with the effective wilson coefficients @xmath29 defined in ref .
@xcite and the color number @xmath30 , which floats from 2 to @xmath31 to estimate the nonfactorizable effects in the generalized version of the factorization approach .
the amplitude in eq .
( [ amp1 ] ) is dominated by the tree contribution from fig .
[ dia](a ) , while that from fig .
[ dia](c ) is primarily the nonfactorizable effect with its size proportional to @xmath32 .
since the global fittings for @xmath32 indicate a universal value of @xmath33 @xcite , @xmath32 is within the uncertainty of @xmath34 . on the other hand , the penguin contributions in figs . [ dia](b ) and [ dia](d ) have been neglected due to the suppressed values of @xmath35 ( @xmath36 ) . in eq .
( [ amp1 ] ) , the matrix elements for the baryon pair production are well defined , given by @xmath37\gamma_\mu + \frac{f_2}{m_n+m_{\bar p}}(p_{\bar p}-p_n)_\mu\bigg\}v_{\bar p}\;,\nonumber\\ \langle n\bar p|\bar d\gamma_\mu
\gamma_5 u|0\rangle&= & \bar u_n\bigg\{g_a\gamma_\mu+\frac{h_a}{m_n+m_{\bar p}}q_\mu\bigg\}\gamma_5 v_{\bar p}\,,\end{aligned}\ ] ] where @xmath38 is the momentum transfer , @xmath39 , @xmath40 and @xmath41 are the timelike baryonic form factors , and @xmath42(@xmath43 ) is the neutron ( antiproton ) spinor . on the other hand , the matrix elements of the @xmath44 baryon transition in eq .
( [ amp1 ] ) have the general forms : @xmath45 u_{\lambda_b}\,,\nonumber\\ & & \langle p|\bar u \gamma_\mu\gamma_5 b|\lambda_b\rangle= \bar u_p\left[g_1\gamma_\mu+\frac{g_2}{m_{\lambda_b}}i\sigma_{\mu\nu}q^\nu+ \frac{g_3}{m_{\lambda_b}}q_\mu\right]\gamma_5 u_{\lambda_b}\,,\end{aligned}\ ] ] where @xmath46 ( @xmath47 ) ( @xmath48 ) are the form factors , with @xmath49 and @xmath50 resulting from the su(3 ) flavor and su(2 ) spin symmetries @xcite , which agree with the results based on the heavy - quark and large - energy symmetries in ref . @xcite . for the numerical analysis ,
the theoretical inputs of the ckm matrix elements in the wolfenstein parametrization are given by @xcite @xmath51 with @xmath52 .
we adopt @xmath53 in ref .
@xcite , and obtain @xmath54 . for the timelike baryonic form factors , it is adopted that @xmath55^{-\gamma}$ ] , @xmath56 , and @xmath57)$ ] with @xmath58 , @xmath59 gev and @xmath60 in pqcd counting rules @xcite .
the form factors at the timelike region may possess the strong phase with the analytical continuation to the spacelike region @xcite , since it can be derived as an overall factor to all subprocesses in fig .
this strong phase , in fact has no effect and is neglected in our paper .
the form factors with the asymptotic behaviors in pqcd counting rules have been justified to agree with the @xmath61 data at @xmath62 gev@xmath63 @xcite . besides
, they have been used to explain the branching ratios , and the so - called threshold effect in the baryon pair invariant mass spectra , which presents the peak around the threshold of @xmath64 gev@xmath63 and gradually turns out to be flat around @xmath65 gev@xmath63 @xcite , being observed as the common feature in the three - body baryonic @xmath12 decays . for the @xmath44 transition , the @xmath66 is presented as the double - pole momentum dependences : @xmath67 @xcite . since the form factors are associated with the studies of baryonic @xmath12 decays and @xmath4-baryon decays , we use the numerical results in refs .
@xcite to give @xmath68 gev@xmath69 , and @xmath70 . with all theoretical inputs ,
we find @xmath71 where the errors come from the form factors , the nonfactorizable effects , and the ckm matrix elements , respectively . by combining the uncertainties
, we obtain the first prediction , @xmath72 , on the branching ratio of this purely baryonic decay , which is sizable and comparable to the branching ratios of other baryonic @xmath12 decays observed at the @xmath12 factories and lhcb .
figure [ mbb ] displays the invariant mass spectra of @xmath73 and @xmath74 .
the neutron - antiproton invariant mass spectrum in @xmath10 presents the threshold effect due to the form of @xmath75 for the @xmath76-pair production , which is similar to the peak around the threshold area of @xmath19 that enhances the branching ratios in the three - body baryonic @xmath77 decays . on the other hand ,
the @xmath78 distribution is in accordance with the fact that the proton and antiproton are not pair produced .
note that the spectra in fig .
[ mbb ] are partly a consequence of ignoring the contributions from the diagrams in figs .
[ dia](c ) and [ dia](d ) .
future measurements of dibaryon spectra should be able to test our assumptions , in particular the factorization approach .
.,width=254 ] since purely baryonic decays are directly connected to the baryonic contents of the universe , it is worthwhile to have systematic investigations of their decay branching ratios and direct @xmath11 violating rate asymmetries as well as the possible @xmath20-odd triple correlations . besides the simplest mode @xmath79 , example decays are @xmath80 and @xmath3 decays , and decays of other antitriplet @xmath4 baryons such as @xmath5 .
the direct @xmath11 violating rate asymmetry can be defined by @xmath81 if both weak ( @xmath82 ) and strong ( @xmath83 ) phases are nonvanishing , one has that @xmath84 for @xmath20 violation in purely baryonic decays , we can examine the triple product correlations of the @xmath20-odd form @xmath85 where @xmath86 are spins or momenta of the baryons .
explicitly , for this class of decays @xmath7 , there are many @xmath20-odd correlations of @xmath87 and @xmath88 , corresponding to one and two spins , respectively , due to the rich spin structures of the baryons
. the asymmetry depending on these correlations is defined by @xmath89 which results in @xmath90 note that @xmath91 may not indicate the real @xmath11 violating effect in the gauge theory . to obtain the true effect
, one can construct the asymmetry by using @xmath92 where @xmath93 is measured in the @xmath11-conjugate decay process , and consequently one finds that @xmath94 which is in general nonzero as long as @xmath95 , no matter whether the strong phase @xmath83 exists .
we remark that , for the baryonic decays with mesons , such as @xmath96 , @xmath97 , and @xmath98 , their branching ratios , and the @xmath11 and @xmath20 violating asymmetries have been well studied @xcite .
in contrast , the purely baryonic decays are conceptually new species and have not been well explored yet . in the sm , the direct @xmath11 violating asymmetry in eq .
( [ acp1 ] ) and the @xmath20 violating asymmetry in eq .
( [ at12 ] ) are estimated to be both around -4% .
as some new @xmath11 violating mechanism is anticipated , these asymmetries could be sensitive to it .
for example , the asymmetries can be enhanced in the @xmath11 violating models with charged higgs and gauge bosons .
experimentally speaking , the presence of a neutron in the final state gives rise to a signature resembling that of a semileptonic decay .
the lhcb experiment has already published studies of such topologies with two charged tracks and a particle escaping detection , for example @xmath99 @xcite .
the analysis can be seen as a proof of concept for the study of @xmath10 .
the observation of the @xmath10 signal is nevertheless rather challenging given a branching ratio significantly lower than that of a typical semileptonic decay
. decays with a @xmath100 pair in the final state and extra invisible or nonreconstructed particles are potentially dangerous sources of background , with branching ratios in the same range
@xmath101 as the signal .
the @xmath102 candidate can be reconstructed using the so - called _ corrected mass _
@xcite defined by @xmath103 where @xmath74 is the invariant mass of the @xmath104 pair and @xmath105 its momentum transverse to the @xmath102 direction of flight . figure [ ppbarcorrm ] shows the distributions of the corrected @xmath104 mass for the signal and a few typical background decay modes resulting from a toy simulation study .
care has been taken to smear the momentum resolution of tracks according to the average resolution published by the lhcb experiment .
also the relative contributions have been scaled taking into account the experimental branching ratios ( reasonable assumptions on the branching ratio were made when the decay mode has not yet been observed ) and typical misidentification rates in lhcb @xcite .
the bottom figure gives in particular the sum over all contributions .
the signal appears as a shoulder around the region @xmath106 .
it is evident that isolation requirements similar to those implemented e.g. , in ref .
@xcite , need to be exploited in order to control the decays to @xmath107 final states , where @xmath108 represents one or several charged and/or neutral particles .
an interesting alternative is the identification of the signal neutron with the calorimeter of the lhcb experiment .
the authors are aware that no publication from lhcb has ever studied neutrons , which makes the route challenging .
these preliminary studies do indicate , though , that the observation of @xmath0 at lhcb is promising . for @xmath11 and/or @xmath20 violation
, although the current sensitivity of lhcb is unable to reach the level predicted in the sm , it is still worthwhile to explore the @xmath11/@xmath20 violating asymmetries of fully reconstructed baryonic decays
@xmath2 is a prominent example as they could be large in the new @xmath11 violating models .
pair corrected mass for the signal @xmath0 and various sources of background resulting from a toy simulation study .
the top part presents all distributions normalized to the unit area , whereas the bottom part stacks all contributions so as to give a more realistic picture of the kind of spectrum at hand .
refer to the text for further details.,title="fig:",width=384 ] pair corrected mass for the signal @xmath0 and various sources of background resulting from a toy simulation study .
the top part presents all distributions normalized to the unit area , whereas the bottom part stacks all contributions so as to give a more realistic picture of the kind of spectrum at hand .
refer to the text for further details.,title="fig:",width=384 ]
since @xmath11 violation in fully baryonic decay processes is directly related to the matter and antimatter asymmetry of the universe , we have studied the simplest case of @xmath0 to investigate its accessibility at the lhcb experiment . with the predicted @xmath1
, this decay can be the new frontier to test the sm and search for new physics .
one can , and should , study other purely baryonic decay modes : @xmath2 , @xmath3 and other similar decays of antitriplet @xmath4 baryon such as @xmath5 .
e. r. wishes to thank the national center for theoretical sciences at hsinchu for its warm hospitality .
the work was supported in part by national center for theoretical sciences , national science council ( nsc-101 - 2112-m-007 - 006-my3 ) , most ( most-104 - 2112-m-007 - 003-my3 ) and national tsing hua university ( 104n2724e1 ) .
lepage and s.j .
brodsky , phys .
lett . * 43 * , 545(1979 ) [ erratum - ibid . *
43 * , 1625 ( 1979 ) ] ; phys .
d * 22 * , 2157 ( 1980 ) ; s.j .
brodsky , g.p .
lepage and s. a. a. zaidi , phys .
d * 23 * , 1152 ( 1981 ) .
chua , w.s .
hou and s.y .
tsai , phys .
d * 66 * , 054004 ( 2002 ) ; c.k .
chua and w.s .
hou , eur .
j. * c29 * , 27 ( 2003 ) ; c.h .
chen , h.y .
cheng , c.q .
geng and y.k .
hsiao , phys .
d * 78 * , 054016 ( 2008 ) . | though not considered in general , purely baryonic decays could shed light on the puzzle of the baryon number asymmetry in the universe by means of a better understanding of the baryonic nature of our matter world .
as such , they constitute a yet unexplored class of decay processes worth investigating .
we propose to search for purely baryonic decay processes at the lhcb experiment .
no such type of decay has ever been observed . in particular , we concentrate on the decay @xmath0 , which is the simplest purely baryonic decay mode , with solely spin-1/2 baryons involved .
we predict its decay branching ratio to be @xmath1 , which is sufficiently large to make the decay mode accessible to lhcb .
our study can be extended to other purely baryonic decays such as @xmath2 and @xmath3 , as well as to similar decays of antitriplet @xmath4 baryons such as @xmath5 . |
recently lhcb collaboration has measured the unexpectedly large cp violating asymmetries in @xmath1 and @xmath2 decays @xcite : @xmath3\% \;\ ; , \label{1}\ ] ] where @xmath4 and @xmath5 is defined analogously .
this result was later confirmed by cdf collaboration , which obtained @xcite : @xmath6\% \;\ ; . \label{3}\ ] ] the most important question concerning experimental results ( [ 1 ] ) and ( [ 3 ] ) is whether in the standard model the cp - violation ( cpv ) in these decays can be as large as 0.5% - 1% . in the standard model
the cpv in @xmath7 decays originates from the interference of the tree and penguin diagrams shown in fig .
1 . for @xmath8 decays @xmath9-quarks in these diagrams should be substituted by @xmath10-quarks .
decay in the standard model .
a wavy line denotes @xmath11-boson , a curly line gluon . ]
it is convenient to present the penguin diagram contribution to @xmath1 decay amplitude in the following form @xcite : @xmath12 + v_{cb } v_{ub}^ * [ f(m_b ) - f(m_s ) ] \;\ ; , \label{4}\end{aligned}\ ] ] attributing the first term to the tree amplitude and considering the second term only as the penguin amplitude . in the case of @xmath2 decay
the following presentation is useful @xcite : @xmath13 + v_{cb } v_{ub}^ * [ f(m_b ) - f(m_d ) ] \ , , \label{5}\end{aligned}\ ] ] where the first term is attributed to the tree amplitude while the second one is the penguin amplitude .
denoting the absolute values of @xmath14 decay amplitudes by @xmath15 and @xmath16 we get : @xmath17 \;\ ; , \nonumber \\ \bar a_{\pi^+\pi^- } & = & t\left[1+\frac{p}{t } e^{i(\delta+\gamma)}\right ] \;\ ; , \label{6}\end{aligned}\ ] ] where @xmath18 stands for the difference of the strong interaction phases of the tree and the penguin amplitudes , while @xmath19 is the phase of @xmath20 ( the product @xmath21 as well as @xmath22 are practically real in the standard parametrization of the ckm matrix ) . from eq .
( [ 6 ] ) for the cpv asymmetry we obtain : @xmath23 where in the denominator of ( [ 2 ] ) we neglect the terms of the order of @xmath24 and @xmath25 which is a very good approximation because @xmath26 . here
@xmath27 is close to unity and we use this value in what follows .
let us present an argument demonstrating that @xmath18 can also be close to 90@xmath28 .
the tree diagram gives dominant contribution to the @xmath29 decay rates .
the corresponding to it 4-fermion hamiltonian has parts with isospin 1/2 and 3/2 .
that is why the produced @xmath30-meson may have isospin zero or two .
so three decay probabilities , @xmath31 , @xmath32 , and @xmath33 , depend on the absolute values of the decay amplitudes @xmath34 and @xmath35 and their strong phases difference @xmath36 . from the experimentally measured branching ratios @xcite : @xmath37\cdot 10^{-4 } \ ; , \;\ ; { \rm br}(d^0 \to\pi^0 \pi^0 ) = [ 8.0\pm 0.8]\cdot 10^{-4 } \ ; , \ ] ] @xmath38\cdot 10^{-4 } \label{8}\ ] ] we find for the phase difference of the amplitudes with @xmath39 and @xmath40 : @xmath41 in eq .
( [ 7 ] ) @xmath18 stands for the difference of the strong phases of penguin amplitude which has @xmath42 and produces pions with @xmath39 and tree amplitude , which has parts with @xmath42 and @xmath43 and produces pions with @xmath39 and @xmath40 , that is why @xmath44 .
nevertheless eq .
( [ 9 ] ) demonstrates that @xmath18 can be large , and so we substitute @xmath45 into eq .
( [ 7 ] ) . in the limit of @xmath46-spin ( @xmath47 interchange ) symmetry
the tree amplitude of @xmath8 decay differs by sign from that of @xmath7 decay , while the penguin amplitudes of these decays are equal , that is why @xmath48 however since @xcite @xmath49 \cdot 10^{-4 } \label{11}\,,\ ] ] we obtain from eq .
( [ 8 ] ) that @xmath50 and @xmath46-spin symmetry is heavily broken in @xmath0 decays .
nevertheless let us suppose that ( [ 10 ] ) is not badly violated , so finally we get : @xmath51 now let us try to understand if in the standard model we can obtain @xmath52 which is needed to reproduce the average value of the lhcb and cdf results .
though the four - fermion quark hamiltonian responsible for these decays is known , strong interactions does not allow to make an exact calculation of the decay amplitudes .
what can be done is an estimate of the decay amplitudes with the help of factorization .
let us start from the tree diagram shown in fig .
1a which dominates in the decay amplitude : @xmath53^{-2/b } + \frac{1}{3}[\alpha_s(m_c)/\alpha_s(m_w)]^{4/b}\right\ } \;\ ; , \label{14}\end{aligned}\ ] ] where the last factor originates from the summation of the gluon exchanges in the leading logarithmic approximation . substituting into it @xmath54 , @xmath55
, @xmath56 we find that the factor in the curly brackets is close to one , @xmath57 .
factorizing the decay amplitude we obtain : @xmath58 = 1.1 \frac{g_f}{\sqrt 2}v_{cd } f_\pi f_+^\pi(0 ) m_d^2 \ ; , \label{15}\end{aligned}\ ] ] where @xmath59 and @xmath60 are the momenta of the produced @xmath30-mesons , @xmath61 is the @xmath0-meson momentum and we neglect @xmath62 in comparison with @xmath63 .
the value of the @xmath64 transition formfactor at @xmath65 can be found in ref .
@xcite : @xmath66 and for the decay width we obtain : @xmath67 where @xmath68 mev was used . from the branching ratio of the @xmath32 decay ( [ 8 ] ) and @xmath69-meson mean life , @xmath70 s
, we find : @xmath71 so the naive factorization overestimates the decay amplitude by the factor @xmath72 .
calculating the @xmath2 decay probability we obtain : @xmath73 ^ 2 \gamma_{d\to \pi^+ \pi^-}^{\rm theor } = 12.2 \cdot 10 ^ 9 { \rm s}^{-1 } \;\ ; , \label{19}\ ] ] where we substituted @xmath74 and @xmath75 taken from ref . @xcite . from eq .
( [ 11 ] ) it follows : @xmath76 so the factorization overestimates the decay amplitude by the factor @xmath77 .
we see that in the case of the tree diagrams the accuracy of the factorization approximation is very good .
let us make a brief remark on the @xmath78 decay .
at the tree level it proceeds through the diagram with @xmath11-boson exchange in @xmath79-channel , so it should be suppressed .
even more , @xmath80 and @xmath81 amplitudes interfere destructively and in the @xmath46-spin symmetry limit their sum is zero @xcite . according to experimental data @xcite : @xmath82 which is approximately 6 times smaller than @xmath83 .
it means that the decay amplitude is smaller than that to charged kaons by factor 2.5 .
this unexpectedly small suppression may indicate that large distance effects like @xmath84 rescattering can be important . the four - fermion qcd penguin amplitude which describes @xmath85 decay looks like : @xmath86 \frac{8}{9 } \;\ ; , \label{22}\end{aligned}\ ] ] where @xmath87 are the gell - mann su(3 ) matrices and we use the fierz identities : @xmath88 also the identity @xmath89 , where @xmath90 or @xmath91 , was used . calculating the matrix element in the factorization approximation with the help of the equations of motion for quark fields we find : @xmath92 \;\ ; . \label{23}\ ] ] dividing it by the experimental value of the tree amplitude and using eq.([15 ] ) we obtain : @xmath93 \;\ ; . \label{24}\ ] ] substituting @xmath94 , @xmath95 , @xmath96 , @xmath56 , @xmath97 gev , @xmath98 gev , @xmath99 mev we come to : @xmath100 comparing it with eq .
( [ 13 ] ) we see that in order to fit the experimental data on @xmath101 the penguin amplitude should be enhanced by the factor 20 in comparison with what factorization gives . concerning the tree amplitudes
, we have found in this section that factorization result differs from the experimental value by the factor 1.4 in the case of @xmath102 decay and by 1.1 in the case of @xmath2 decay . in the next two sections
we will study how accurate is the factorization approximation to the penguin amplitudes in @xmath103- and @xmath104-meson decays .
@xmath105 decay is described by the penguin amplitude shown in fig .
2 . decay proceeds through the penguin amplitude only . ]
the hamiltonian responsible for this decay looks like : @xmath106 \;\ ; , \label{26}\ ] ] @xmath107 is substituted for @xmath108 ( the contribution of a loop with the virtual @xmath79-quark is negligible ) and @xmath109 where @xmath110 are the color indexes . using the fierz identities as well as @xmath111 identity we obtain : @xmath112 \;\ ; , \label{28}\ ] ] where @xmath113 , @xmath114 . calculating the matrix element in the factorization approximation
we obtain : @xmath115 \;\ ; , \label{29}\ ] ] where in the leading logarithmic approximation the following approximate equation is valid : @xmath116 while at nlo approximation from table 1 of @xcite we obtain : @xmath117 , @xmath118 . substituting @xmath119 mev , @xmath120 gev
we find : @xmath121 where @xmath122 and @xmath123 from @xcite was used .
the experimental result is : @xmath124 so , the factorization result is enhanced by the factor @xmath125 the numerical value of the penguin amplitude is important in the calculation of cp asymmetries in @xmath126 and @xmath127 decays @xcite .
@xmath128 penguin transition changes the isospin by 1/2 in this way explaining the famous @xmath129 rule in @xmath130 decays .
the calculation of the @xmath131 decay amplitude generated by the penguin transition using the factorization underestimates the amplitude by the factor 2 - 3 according to refs .
@xcite . in view of the results for @xmath103 and @xmath104 decays we can cautiously assume that for @xmath102 decay the factorization calculation underestimates the penguin amplitude at most by factor 5 leading to : @xmath132 thus the following alternative emerges : either the experimental results are wrong or new physics is found .
of course we can not determine what kind of new particles and interactions are responsible for large cpv asymmetry in @xmath133 ( @xmath134 ) decays .
however , in the next section we will propose the straightforward generalization of the standard model in which large cpv in @xmath0 decays can be explained .
as it was stated in paper @xcite the introduction of the fourth quark - lepton generation may easily remove standard model upper bound ( [ 34 ] ) matching the experimental results @xcite . in the case of the fourth generation the additional term with the intermediate @xmath135 quark should be added to the expression for the penguin amplitude . in this way expression ( [ 4 ] ) is substituted by : @xmath136 + v_{cb}v_{ub}^ * [ f(m_b)-f(m_s ) ] + v_{cb ' } v_{ub'}[f(m_{b ' } ) -f(m_s ) ] \;\ ; , \label{35}\end{aligned}\ ] ] where the unitarity of 4@xmath1374 quark mixing matrix is used . according to the experimental constraints from the direct searches of the fourth generation quarks
@xmath135 should weigh several hundreds gev , that is why @xmath138 is small and can be neglected just as it is done with @xmath79-quark contribution to @xmath139 penguin , see the remark after eq .
( [ 26 ] ) .
in order to enhance sm contribution to the penguin amplitude we should suppose that the term @xmath140 dominates .
then the enhancement of @xmath141 in the case of the fourth generation is equal to : @xmath142 where in the last equality we use the allowed values of the product @xmath143 taken from fig . 1 of paper @xcite .
) enhancement originates not from the diagram with the intermediate @xmath135 quark but from the term @xmath144 .
] so we see that the enhancement necessary to describe the experimental data on @xmath145 can be achieved in the case of the fourth generation .
if weakly mixed particles exist , then the sphaleron processes can create the baryon asymmetry of the universe @xcite . as it is noted in ref .
@xcite , the long - lived fourth generation particles save baryon asymmetry generated in the early universe from erasure by the sphaleron transitions .
the sphaleron transitions conserve @xmath146 , thus , if in the early universe @xmath147 is generated , then the final baryon and lepton asymmetries being proportional to @xmath146 are completely erased .
if the fourth generation particles weakly mix with three quark - lepton generations of the standard model , then two additional quantities are conserved : @xmath148 and @xmath149 , where @xmath150 and @xmath151 are the densities of baryons and leptons of the fourth generation , while @xmath103 and @xmath152 are the densities of baryons and leptons of three light generations . in ref .
@xcite initial asymmetries @xmath153 and @xmath154 were chosen and since @xmath155 , the total baryonic number density , @xmath156 , being proportional to a linear superposition of conserved quantities is nonzero at the sphaleron freeze - out temperature . after the sphaleron freeze - out @xmath156 is conserved in comoving volume and is equal to the present day baryon density of the universe . however
, if heavy baryons of the 4th generation do not decay prior to big bang nucleosynthesis ( bbn ) , the light baryon number density at bbn could be different from that determined from the angular fluctuations of cmb .
the impact of this effect on the light element abundances is discussed below .
for such a scenario to occur the lifetimes of the fourth generation quarks and leptons should be at least larger than the universe age at the sphaleron freeze - out : @xmath157 s. for the mixing angles in the case of @xmath158 decay it gives @xmath159 @xcite , much smaller than what we need to explain the large cpv in @xmath0-decays , see eq .
( [ 36 ] ) .
so in our case quarks of the fourth generation should be much stronger mixed with quarks of three light generations .
however , let us suppose that leptons of the fourth generation are weakly mixed with the leptons of three light generations .
let us introduce the total baryon density , @xmath160 , and take the initial conditions analogous to those in ref .
@xcite : @xmath161 and @xmath162 .
we can choose four independent chemical potentials as : @xmath163 , @xmath164 , @xmath165 and @xmath166 , which are the chemical potentials for the upper type quarks , @xmath11-bosons , @xmath167 neutrino and sum over all sm neutrino chemical potentials ( see appendix ) .
in the limit @xmath168 the baryon and lepton densities are linear combinations of these chemical potentials with the coefficients which depend on the ratio of masses of the corresponding particles to the temperature .
we will take into account the masses of @xmath11-boson , @xmath79-quark , @xmath169- and @xmath135-quarks of the fourth generation and the fourth generation leptons @xmath170 and @xmath171 , the masses of all the other components of the primeval plasma can be neglected in comparison with @xmath172 . as a function of sphaleron freeze - out temperature @xmath172 for the unmixed fourth generation
is shown by a dashed ( blue ) line .
it is analogous to fig . 2 from @xcite but for @xmath173 gev , @xmath174 gev , @xmath175 gev , @xmath176 gev .
the final baryon asymmetry for the case of the mixed fourth generation quarks and the unmixed fourth generation leptons is shown by a solid ( green ) line . ]
finally we have four equations for four unknown chemical potentials : two quantities are conserved under the sphaleron transitions ; we can choose them as @xmath177 the third equation is that of the electric neutrality of the primeval plasma , @xmath178 , and , finally , the sum of the chemical potentials of all the particles which are converted into nothing by sphaleron ( @xmath179 of each generation ) equals zero .
the values of masses of the 4th generation particles we take from paper @xcite in which the fit to the electroweak observables for higgs mass @xmath180 gev was performed and recent lhc bounds on the masses of @xmath169- and @xmath135-quarks were taken into account : @xmath181 the dashed blue line in fig .
3 corresponds to the case of the unmixed fourth generation particles considered in @xcite .
the results for the case of the strongly mixed fourth generation quarks and the unmixed fourth generation leptons are shown by the solid green line . in order that leptons , @xmath170
, do not decay before the sphaleron freeze - out , which happens at @xmath182 s , the mixing angles of @xmath170 with three light neutrinos should be small : @xmath183 ( @xmath170 decays through four fermion interaction ) . assuming similar bound @xmath184 , the existence of heavy dirac sequential neutrino with @xmath185 gev is compatible with the search at lep ii @xcite .
according to the standard cosmological scenario nonrelativistic matter started to dominate the cosmic energy density at redshift @xmath186 . if we demand that @xmath170 should decay before that epoch , its life - time
should be sufficiently short , @xmath187 s , from which we obtain the lower bound @xmath188 .
( let us note that direct searches exclude @xmath170 as a unique dark matter candidate @xcite ) .
a stronger bound on @xmath189 follows from the equilibrium form of the energy spectrum of cmb . according to ref .
@xcite a large influx of energy into the usual cosmological cosmic background would be thermalized if it took place before @xmath190 .
otherwise the observed black body spectrum of cmb would be noticeably distorted . since the precision of the spectral shape is at the level of @xmath191 , only a very small distortion is permitted .
the condition that @xmath170 decays before or at @xmath192 demands @xmath193 s , or @xmath194 .
if @xmath170 indeed decays before @xmath190 , the contribution from its decay to the energy density of cmb would be not larger than 1% and the ratio of baryon to photon number densities @xmath195 at bbn epoch and at cmb recombination would be slightly different but in principle measurable by the light element abundances .
more interesting and pronounced effect appears if heavy quarks of the 4th generation are long - lived . in this case
we can not explain the large value of cpv in @xmath0 decays but may explain the difference of @xmath196 at bbn epoch ( @xmath197 ) and at the recombination ( @xmath198 ) which is probably requested by the recent data on the light element abundances @xcite .
if heavy baryons of the 4th generation decays after bbn but before the hydrogen recombination , the number of light baryons in the comoving volume at bbn would be different from that at the recombination .
the ratio @xmath199 at these epochs could be either larger or smaller than unity depending upon the value of the baryon asymmetry in the heavy quark sector and the energy influx to cmb from the heavy baryon decays .
so in principle both rise or decrease of @xmath197 is possible and the number of light neutrino species influence nucleosynthesis , the change in the value of @xmath197 can be formulated as an additional ( positive or negative ) number of light neutrino species @xcite . ] . in the limit @xmath200 heavy particles of the fourth generation
are not produced : @xmath201 , @xmath202 . in the physically interesting opposite limit
@xmath203 the value of baryon asymmetry is nonzero since the right - handed neutrinos of three light generations are not produced in the primordial plasma violating symmetry between the leptons of four generations which would occur at @xmath204 .
the characteristic time of the right - handed neutrino to thermalize is @xmath205 and for @xmath206 kev ( which is valid for three light neutrinos ) this time is longer than the universe age , @xmath207 for @xmath208 gev , @xcite .
in introduction we determined what ratio of the penguin to the tree amplitudes of @xmath1 decay is needed to get the observed cp asymmetry . in section 2 we found that the factorization describes the tree amplitude with good accuracy ; concerning the penguin amplitude it appears to be twenty times smaller than one needs to describe the experimental data on @xmath141 . in section 3
we demonstrated that in the case of @xmath209 decay the factorization underestimates the penguin amplitude by factor 2 . in the case of @xmath210 decay the penguin amplitude is enhanced by factor 2 - 3 in comparison with the factorization result .
in section 5 we demonstrated that the fourth quark - lepton generation may enhance the penguin amplitude describing the experimental data .
if the leptons of the fourth generation weakly mix with three light generation leptons , then the baryonic charge generated at high scale escapes the erasure by sphalerons and survives till now according to the results presented in section 6 .
we are grateful to s.i .
blinnikov for the illuminating discussion on the chemical potentials , to v.a .
rubakov for the clarifying discussion on the baryon density in the unbroken electroweak phase , and to j. zupan for the remark concerning @xmath211 decay .
, s.g . , and
acknowledge the support of the grant of the russian federation government 11.g34.31.0047 .
s.g . and m.v .
are partially supported by the grants rfbr 11 - 02 - 00441 , 12 - 02 - 00193 and by the grant nsh-3172.2012.2 .
is partially supported by the grant rfbr 10 - 02 - 01398 .
being interested in the values of the asymmetries at sphaleron freeze - out temperature we should assume that the electroweak phase transition already has occured and the neutral higgs boson condenses .
that is why the higgs boson chemical potential is zero . sometimes in the literature
the baryon density in the electroweak unbroken phase is looked for . in this case
the higgs boson does not condense and its chemical potential is nonzero . to find it an additional equation
is needed .
it is provided by the condition that the density of charges with which the massless bosons interact should be zero , and in an unbroken phase there are two such charges : the hypercharge and the third projection of a weak isospin .
the baryon density in the unbroken phase is analyzed , for example , in book @xcite and it differs from its value in a broken phase . since the right - handed components of quarks and leptons emitting neutral higgs transform to the left - handed components the chemical potential of both components are equal : @xmath212 , @xmath213 , @xmath214 .
the analogous relations are valid for the particles of the second and third families .
the right - handed neutrinos of three light generations are not thermalized and should not be taken into account ( see the end of sect .
the fourth generation right - handed neutrinos , being heavy , rapidly thermalize : @xmath215 .
the chemical potentials of up and down weak isospin components are related by @xmath216 chemical potential : @xmath217 , @xmath218 , @xmath219 . mixing of quarks of four families and leptons of three families equilibrates the chemical potentials of the particles with the identical gauge quantum numbers . as a result four
independent chemical potentials remain : @xmath220 , @xmath221 , @xmath164 and @xmath222 .
the particle number densities depend on their ( fermi or bose ) statistics , temperature , chemical potential , and masses .
the chemical potential of an antiparticle is opposite to that of the particle .
the asymmetries and , hence , chemical potentials are very small . expanding the equilibrium integrals for the asymmetry over @xmath223 we obtain : @xmath224 @xmath225 where @xmath226 is the number of the degrees of freedom of the particle @xmath61 ( @xmath227 , @xmath228 , @xmath229 , @xmath230 ) and @xmath231 .
functions @xmath232 are normalized in such a way that @xmath233 . in what follows
we take into account the nonzero masses of the particles of the fourth generation , of @xmath79-quark , and of @xmath11-boson .
the condition of electroneutrality of the primeval plasma looks as : @xmath234 - 3\,\frac{1}{3}[2(\alpha_d + \alpha_s + \alpha_b + \alpha_{b ' } ) \times \nonumber \\ & \times & ( \mu_w + \mu_u ) ] -2[(\alpha_e + \alpha_\mu + \alpha_\tau)(\mu_w + \mu_\nu ) ] - 2\alpha_e(\mu_w +
\mu_n ) - \nonumber \\ & - & 3\cdot 2 \alpha_w \mu_w = 0 \;\ ; , \label{a.2}\end{aligned}\ ] ] @xmath235 here and below we omit irrelevant factor @xmath236 .
the remaining two equations are two superpositions of @xmath238 , @xmath152 , and @xmath151 conserved under sphaleron transitions thus being equal to their initial values .
the expressions for these quantities look like : @xmath239 @xmath240 @xmath241 = \nonumber \\ & = & 2(2+\alpha_t + \alpha_{t'})\mu_u + 2(3+\alpha_{b'})(\mu_u + \mu_w ) \;\ ; . \label{a.7}\end{aligned}\ ] ] 99 r. aaij et al .
* 108 * ( 2012 ) 111602 .
t. aaltonen et al . , arxiv:1207.2158 ( 2012 ) .
g. isisdori , j. kamenik , z. ligeti , g. perez , phys .
b * 711 * ( 2012 ) 46 .
k. nakamura et al . ( particle data group ) ,
g * 37 * ( 2010 ) 075021 .
j. brod , a.l .
kogan , j. zupan , arxiv:1111.5000 ( 2011 ) .
buras et al . , nucl . phys .
b * 370 * ( 1992 ) 69 . a.b .
kaidalov , m.i .
vysotsky , phys . atom .
* 72 * ( 2009 ) 2126 .
a. vainshtein , hep - ph/9906263 ( 1999 ) .
rozanov , m.i .
vysotsky , pisma v zhetf * 95 * ( 2012 ) 443 .
s. nandi , a. soni , phys .
d * 83 * ( 2011 ) 114510 .
k. dick , m. lindner , m. ratz , d. wright , phys . rev .
* 84 * ( 2000 ) 4039 .
h. murayama , v. rentala , j. shu , t. yanagida , phys .
b * 705 * ( 2011 ) 208 .
o. eberhardt et al . , arxiv:1204.3872 ( 2012 ) .
k. ackerstaff et al .
j. * c1 * ( 1998 ) 45 ; + p. achard et al .
, phys . lett .
* b 517 * ( 2001 ) 75 .
lee , z. liu , a. soni , phys . lett .
* b 704 * ( 2011 ) 30 .
sunyaev , ya.b .
zeldovich , ap . and
space sci .
* 20 * ( 1970 ) 20 ; + a.f .
illarionov , r.a .
sunyaev , astron .
j. * 51 * ( 1974 ) 698 . | since the standard model most probably can not explain the large value of cp asymmetries recently observed in @xmath0-meson decays we propose the fourth quark - lepton generation explanation of it . as a byproduct weakly mixed leptons of the fourth generation make it possible to save the baryon number of the universe from erasure by sphalerons .
an impact of the 4th generation on bbn is briefly discussed . |
in modern astronomy , one is increasingly faced with the problem of analysing large , complicated and multidimensional data sets .
such analyses typically include tasks such as : data description and interpretation , inference , pattern recognition , prediction , classification , compression , and many more .
one way of performing such tasks is through the use of machine learning methods . for accessible accounts of machine learning and its use in astronomy ,
see , for example , @xcite , @xcite and @xcite .
moreover , machine learning software easily used for astronomy , such as the python - based astroml package , or c - based fast artificial neural network library ( fann ) have recently started to become available .
two major categories of machine learning are : _ supervised learning _ and _ unsupervised learning_. in supervised learning , the goal is to infer a function from labeled training data , which consist of a set of training examples .
each example has both ` properties ' and ` labels ' .
the properties are known ` input ' quantities whose values are to be used to predict the values of the labels , which may be considered as ` output ' quantities .
thus , the function to be inferred is the mapping from properties to labels .
once learned , this mapping can be applied to datasets for which the values of the labels are not known .
supervised learning is usually further subdivided into _ classification _ and _ regression_. in classification , the labels take discrete values , whereas in regression the labels are continuous . in astronomy , for example , using multifrequency observations of a supernova lightcurve ( its properties ) to determine its type ( e.g. ia , ib , ii , etc . )
is a classification problem since the label ( supernova type ) is discrete ( see , e.g. , @xcite ) , whereas using the observations to determine ( say ) the energy output of the supernova explosion is a regression problem , since the label ( energy output ) is continuous .
classification can also be used to obtain a distribution for an output value that would normally be treated as a regression problem .
this is demonstrated by @xcite for measuring redshifts in cfhtlens .
a particularly important recent application of regression supervised learning in astrophysics and cosmology ( and beyond ) is the acceleration of the statistical analysis of large data sets in the context of complicated models . in such analyses ,
one typically performs many ( @xmath0 ) evaluations of the likelihood function describing the probability of obtaining the data for different sets of values of the model parameters . for some problems , in particular in cosmology ,
each such function evaluation can take up to tens of seconds , making the analysis very computationally expensive . by performing regression
supervised learning to infer and then replace the mapping between model parameters and likelihood value , once can reduce the computation required for each likelihood evaluation by several orders of magnitude , thereby vastly accelerating the analysis ( see , e.g. , @xcite ) . in unsupervised learning ,
the data have no labels .
more precisely , the quantities ( often termed ` observations ' ) associated with each data item are not divided into properties ( inputs ) and labels ( outputs ) .
this lack of a ` causal structure ' , where the inputs are assumed to be at the beginning and outputs at the end of a causal chain , is the key difference from supervised learning . instead ,
all the observations are considered to be at the end of a causal chain , which is assumed to begin with some set of ` latent ' ( or hidden ) variables .
the aim of unsupervised learning is to infer the number and/or nature of these latent variables ( which may be discrete or continuous ) by finding similarities between the data items .
this then enables one to summarize and explain key features of the dataset .
the most common tasks in unsupervised learning include _ density estimation _ , _ clustering _ and _ dimensionality reduction_. indeed , in some cases ,
dimensionality reduction can be used as a pre - processing step to supervised learning , since classification and regression can sometimes be performed in the reduced space more accurately than in the original space . as an astronomical example of unsupervised learning one
might wish to use multifrequency observations of the lightcurves of a set of supernovae to determine how many different types of supernovae are contained in the set ( a clustering task ) .
alternatively , if the data set also includes the type of each supernova ( determined using spectroscopic observations ) , one might wish to determine which properties , or combination of properties , in the lightcurves are most important for determining their type photometrically ( a dimensionality reduction task ) .
this reduced set of property combinations could then be used instead of the original lightcurve data to perform the supernovae classification or regression analyses mentioned above .
an intuitive and well - established approach to machine learning , both supervised and unsupervised , is based on the use of artificial neural networks ( nns ) , which are loosely inspired by the structure and functional aspects of a brain .
they consist of a group of interconnected nodes , each of which processes information that it receives and then passes this product on to other nodes via weighted connections . in this way ,
nns constitute a non - linear statistical data modeling tool , which may be used to represent complex relationships between a set of inputs and outputs , to find patterns in data , or to capture the statistical structure in an unknown joint probability distribution between observed variables .
in general , the structure of a nn can be arbitrary , but many machine learning applications can be performed using only feed - forward nns . for such networks
the structure is directed : an input layer of nodes passes information to an output layer via zero , one , or many ` hidden ' layers in between .
such a network is able to ` learn ' a mapping between inputs and outputs , given a set of training data , and can then make predictions of the outputs for new input data .
moreover , a universal approximation theorem assures us that we can accurately and precisely approximate the mapping with a nn of a given form .
a useful introduction to nns can be found in @xcite . in astronomy ,
feed - forward nns have been applied to various machine learning problems for over 20 years ( see , e.g. , @xcite ) .
nonetheless , their more widespread use in astronomy has been limited by the difficulty associated with standard techniques , such as backpropagation , in training networks having many nodes and/or numerous hidden layers ( i.e. ` large ' and/or ` deep ' networks ) , which are often necessary to model the complicated mappings between the numerous inputs and outputs in modern astronomical applications . in this paper
, we therefore present the first public release of skynet : an efficient and robust neural network training tool that is able to train large and/or deep feed - forward networks .
skynet is able to achieve this by using a combination of the ` pre - training ' method of @xcite to obtain a set of network weights close to a good optimum of the training objective function , followed by further optimisation of the weights using a regularised variant of newton s method based on that developed for the memsys software package @xcite . in particular ,
second - order derivative information is used to improve convergence , but without the need to evaluate or store the full hessian matrix , by using a fast approximate method to calculate hessian - vector products @xcite .
skynet is implemented in the standard ansi c programming language and parallelised using mpi .
we also note that skynet has already been combined with multinest @xcite , to produce the blind accelerated multimodal bayesian inference ( bambi ) package @xcite , which is a generic and completely automated tool for greatly accelerating bayesian inference problems ( by up to a factor of @xmath1 ; see , e.g. , @xcite ) .
multinest is a fully - parallelised implementation of nested sampling @xcite , extended to handle multimodal and highly - degenerate distributions . in most astrophysical ( and particle physics )
bayesian inference problems , multinest typically reduces the number of likelihood evaluations required by an order of magnitude or more , compared to standard mcmc methods , but bambi achieves further substantial gains by speeding up the evaluation of the likelihood itself by replacing it with a trained regression neural network .
bambi proceeds by first using multinest to obtain a specified number of new samples from the model parameter space , and then uses these as input to skynet to train a network on the likelihood function .
after convergence to the optimal weights , the network s ability to predict likelihood values to within a specified tolerance level is tested .
if it fails , sampling continues using the original likelihood until enough new samples have been made for training to be performed again . once a network is trained that is sufficiently accurate , its predictions are used in place of the original likelihood function for future samples for multinest . on typical problems in cosmology , for example , using the network reduces the likelihood evaluation time from seconds to less than a millisecond , allowing multinest to complete the analysis much more rapidly . as a bonus , at the end of the analysis the user also obtains a network that is trained to provide more likelihood evaluations near the peak if needed , or in subsequent analyses . with the public release of skynet
, we now also make bambi publically available .
the structure of this paper is as follows . in section [ sec : nnstruct ]
we describe the general structure of feed - forward nns , including a particular special case of such networks , called autoencoders , which may be used for performing non - linear dimensionality reduction . in section
[ sec : nntrain ] we present the procedures used by skynet to train networks of these types .
skynet is then applied to some toy machine learning examples in section [ sec : nntoyex ] , including a regression task , a classification task , and a dimensionality reduction task using autoencoders .
we also apply skynet to the problem of classifying images of handwritten digits from the mnist database , which is a widely - used benchmarking test of machine learning algorithms .
the application of skynet to astronomical machine learning examples is presented in section [ sec : nnex_astro ] , including : a regression task to determine the projected ellipticity of a galaxy from blurred and noisy images of the galaxy and of a field star ; a classification task , based on a simulated gamma - ray burst detection pipeline for the swift satellite @xcite , to determine if a grb with given source parameters will be detected ; and a dimensionality reduction task using autoencoders to compress and denoise galaxy images .
finally , we present our conclusions in section [ sec : nndiscuss ] .
a multilayer perceptron feed - forward neural network is the simplest type of network and consists of ordered layers of perceptron nodes that pass scalar values from one layer to the next .
the perceptron is the simplest kind of node , and maps an input vector @xmath2 to a scalar output @xmath3 via @xmath4 where @xmath5 and @xmath6 are the parameters of the perceptron , called the ` weights ' and ` bias ' , respectively . for a 3-layer nn , which consists of an input layer , a hidden layer , and an output layer , as shown in fig .
[ fig : neuralnet ] , the outputs of the nodes in the hidden and output layers are given by the following equations : @xmath7 where @xmath8 runs over input nodes , @xmath9 runs over hidden nodes , and @xmath10 runs over output nodes .
the functions @xmath11 and @xmath12 are called activation functions and must be smooth and monotonic for our purposes .
we use @xmath13 ( sigmoid ) and @xmath14 ; the non - linearity of @xmath11 is essential to allowing the network to model non - linear functions . to expand the nn to include more hidden layers , we iterate for each connection from one hidden layer to the next , each time using the same activation function , @xmath11
. the final hidden layer will connect to the output layer using the relation .
the weights and biases are the values we wish to determine in our training ( described in section [ sec : nntrain ] ) .
as they vary , a huge range of non - linear mappings from inputs to outputs is possible .
in fact , a universal approximation theorem @xcite states that a nn with three or more layers can approximate any continuous function to some given accuracy , as long as the activation function is locally bounded , piecewise continuous , and not a polynomial ( hence our use of sigmoid @xmath11 , although other functions would work just as well , such as @xmath15 ) . by increasing the number of hidden nodes
, one can achieve more accuracy at the risk of overfitting to our training data .
other activation functions have also been proposed , such as the rectified linear function wherein @xmath16 or the ` softsign ' function where @xmath17 .
it has been argued that the former removes the need for pre - training ( as described in section [ sec : nnpretrain ] ) @xcite and serves as a better model of biological neurons .
the ` softsign ' is similar to @xmath15 , but with slower approach to the asymptotes of @xmath18 ( quadratic rather than exponential ) @xcite .
autoencoders are a specific type of feed - forward neural network containing one or more hidden layers , where the inputs are mapped to themselves , i.e. the network is trained to approximate the identity operation ; when more than one hidden layer is used this is typically referred to as a ` deep ' autoencoder .
such networks typically contain several hidden layers and are symmetric about a central layer containing fewer nodes than there are inputs ( or outputs ) .
a basic diagram of an autoencoder is shown in fig . [
fig : autoencoder ] , in which the three inputs @xmath19 are mapped to themselves via three symmetrically - arranged hidden layers , with two nodes in the central layer . ) defines the weight matrices @xmath20 and @xmath21 . ]
an autoencoder can thus be considered as two half - networks , with one part mapping the inputs to the central layer and the second part mapping the central layer values to the outputs ( which approximate as closely as possible the original inputs ) .
these two parts are called the ` encoder ' and ` decoder ' , respectively , and map either to or from a reduced set of ` feature variables ' embodied in the nodes of the central layer ( denoted by @xmath22 and @xmath23 in fig .
[ fig : autoencoder ] ) .
these variables are , in general , non - linear functions of the original input variables .
one can determine this dependence for each feature variable in turn simply by decoding @xmath24 , @xmath25 , and so on , as the corresponding @xmath26 value is varied ; in this way , for each feature variable , one obtains a curve in the original data space .
conversely , the collection of feature values @xmath27 in the central layer might reasonably be termed the feature vector of the input data .
autoencoders therefore provide a very intuitive approach to non - linear dimensionality reduction and constitute a natural generalisation of linear methods such as principal component analysis ( pca ) and independent component analysis ( ica ) , which are widely used in astronomy .
indeed , an antoencoder with a single hidden layer and linear activation functions may be shown to be identical to pca @xcite .
this topic is explored further in section [ sec : nntoy_ae ] .
it is worth noting that encoding from input data to feature variables can also be useful in performing clustering tasks ; this is illustrated in section [ sec : mnist ] .
autoencoders are , however , notoriously difficult to train , since the objective function contains a broad local maximum where each output is the average value of the inputs @xcite .
nonetheless , this difficulty can be overcome by the use of pre - training methods , as discussed in section [ sec : nnpretrain ] .
an important choice when training a nn is the number of nodes in its hidden layers .
the optimal number and organisation into one or more layers has a complicated dependence on the number of training data points , the number of inputs and outputs , and the complexity of the function to be trained .
choosing too few nodes will mean that the nn is unable to learn the relationship to the highest possible accuracy ; choosing too many will increase the risk of overfitting to the training data and will also slow down the training process . using empirical evidence @xcite and theoretical considerations
@xcite , it has been suggested that the optimal architecture for approximating a continuous function is one hidden layer containing @xmath28 nodes , where @xmath29 is the number of input nodes .
@xcite also find empirical support for this suggestion .
such a choice allows the network to model the form of the mapping function without unnecessary work . in practice
, it can be better to over - estimate ( slightly ) the number of hidden nodes required . as described in section [ sec : nntrain ] , skynet performs basic checks to prevent over - fitting , and the additional training time associated with having more hidden nodes is not a large penalty if an optimal network can be obtained in an early attempt . in any case ,
given a particular problem , the optimal network structure , both in terms of the number of hidden nodes and how they are distributed into layers , can be determined by comparing the correlation and error squared of different trained nns ; this is illustrated in section [ sec : nntoyex ] .
in training a nn , we wish to find the optimal set of network weights and biases that maximise the accuracy of the predicted outputs . however , we must be careful to avoid overfitting to our training data , which may lead to inaccurate predictions from inputs on which the network has not been trained .
the set of training data inputs and outputs ( or ` targets ' ) , @xmath30 , is provided by the user ( where @xmath31 counts training items ) .
approximately @xmath32 per cent should be used for actual nn training and the remainder retained as a validation set that will be used to determine convergence and to avoid overfitting .
this ratio of 3:1 gives plenty of information for training but still leaves a representative subset of the data for checks to be made .
it is prudent to ` whiten ' the data before training a network .
whitening normalises the input and/or output values , so that it easier to train a network starting from initial weights that are small and centred on zero .
the network weights in the first and last layers can then be ` unwhitened ' after training so that the network will be able to perform the mapping from original inputs to outputs .
standard whitening transforms each input to a standard distribution by subtracting the mean and dividing by the standard deviation over all elements in the training data , such that [ eq : whitening1 ] @xmath33 an alternative whitening transform is also commonly used , wherein all values are scaled and shifted into the interval @xmath34 $ ] , such that @xmath35 one of these transforms may be chosen by the user if they wish to whiten the inputs of the training data .
the whitening is normally performed separately on each input , but can be calculated across all inputs if they are related .
the mean , standard deviation , minimum , or maximum would then be computed over all inputs for all training data items .
the chosen whitening transform is also used for whitening the outputs . since both transforms consist of subtracting an offset and multiplying by a scale factor , they can easily be performed and reversed . to unwhiten network weights
the inverse transform is applied , with the offset and scale determined by the source input node or target output node .
outputs for a classification network are not whitened since they are already just probabilities ( see below ) .
let us denote the network weights and biases collectively by the network parameter vector @xmath36 .
skynet considers the parameters @xmath36 to be random variables with a posterior distribution given by @xmath37 where @xmath38 is the likelihood , which also depends on a set of hyperparameters @xmath39 that describe the standard deviation of the outputs ( see below ) .
the likelihood encodes how well the nn , characterised by a given set of parameters @xmath36 , is able to reproduce the known training data outputs .
this is modulated by the prior @xmath40 , which is assumed to have the ( logarithmic ) form @xmath41 where @xmath42 is a hyperparameter that plays the role of a regularisation parameter during optimisation since it determines relative importance of the prior and the likelihood .
this prior can also be seen as an @xmath43-norm penalty .
the form of the likelihood depends on the type of network being trained . for regression problems
, skynet assumes a log - likelihood function for the network parameters @xmath36 given by the standard @xmath44 misfit function @xmath45 ^ 2},\end{aligned}\ ] ] where @xmath29 is the number of outputs , @xmath46 is the number of training data examples and @xmath47 is the nn s predicted output vector for the input vector @xmath48 and network parameters @xmath36 .
the hyperparameters @xmath49 describe the standard deviation ( error size ) of each of the outputs . for classification problems ,
skynet again uses continuous outputs ( rather than discrete ones ) , which are interpreted as the probabilities that a set of inputs belongs to a particular output class .
this is achieved by applying the _ softmax _ transformation to the output values , so that they are all non - negative and sum to unity , namely @xmath50}{\sum_{j=1}^n \exp[y_j(\bmath{x}^{(k)};\bmath{a})]}.\ ] ] the classification likelihood is then given by the _ cross - entropy _ of the targets and softmaxed output values @xcite , @xmath51 in this scenario , the true and predicted output values are both probabilities ( which lie in @xmath34 $ ] ) . for the true outputs , all are zero except for the correct output which has a value of unity . for classification networks
, the @xmath39 hyper - parameters do not appear in the log - likelihood .
the training of the nn can be started from some random initial state , or from a state determined from a ` pre - training ' procedure discussed below . in the former case ,
the network training begins by setting random values for the network parameters , sampled from a normal distribution with zero mean and variance of @xmath52 ( this value can be modified by the user ) . in the latter case
, skynet makes use of the pre - training approach developed by @xcite , which obtains a set of network weights and biases close to a good solution of the network objective function .
this method was originally devised with autoencoders in mind and is based on the model of restricted boltzmann machines ( rbms ) .
an rbm is a generative model that can learn a probability distribution over a set of inputs .
it consists of a layer of input nodes and a layer of hidden nodes , as shown in figure [ fig : rbm ] . in this case , the map from the inputs to the hidden layer and then back is treated symmetrically and the weights are adjusted through a number of ` epochs ' , gradually reducing the reproduction error . to model an autoencoder , rbms
are ` stacked ' , with each rbm s hidden layer being the input for the next .
the initial case is the nn s inputs to the first hidden layer ; this is repeated for the first to second hidden layer and so on until the central layer is reached .
the network weights can then be ` unfolded ' by using the transpose for the symmetric connections in the decoding half to provide a decent starting point for the full training to begin .
this is shown in fig .
[ fig : autoencoder ] , where the @xmath20 and @xmath21 weights matrices are defined by pre - training . visible nodes and @xmath53 hidden nodes .
bias nodes are not shown .
image courtesy wikimedia commons . ]
the training is then performed using _ contrastive divergence _ @xcite .
this procedure can be summarised in the following steps , where sampling indicates setting the value of the node to @xmath54 with the probability calculated and @xmath55 otherwise . 1 .
take a training sample @xmath56 and compute the probabilities of the hidden nodes ( their values using a sigmoid activation function ) and sample a hidden vector @xmath57 from this distribution .
2 . let @xmath58 , where @xmath59 is used to indicate the outer product .
3 . using @xmath57 ,
compute the probabilties of the visible nodes and sample @xmath60 from this distribution .
resample the hidden vector from this to obtain @xmath61 .
4 . let @xmath62 . 5 .
@xmath63 for some learning rate @xmath64 .
more details can be found in @xcite and @xcite has useful diagrams and explanations .
this pre - training approach can also be used for more general feed - forward networks .
all layers of weights , except for the final one that connects the last hidden layer to the outputs , are pre - trained as if they were the first half of a symmetric autoencoder .
however , the network weights are not unfolded ; instead the final layer of weights is initialised randomly as would have been done without pre - training . in this way
, the network ` learns the inputs ' before mapping to a set of outputs .
this has been shown to greatly reduce the training time on multiple problems by @xcite .
we note that when an autoencoder is pre - trained , the activation function to the central hidden layer is made linear and the activation function from the final hidden layer to the outputs is made sigmoidal .
general feed - forward networks that are pre - trained continue to use the original activation functions .
both of these are simply the default settings and the user has the freedom to alter them to suit their specific problem .
once the initial set of network parameters have been obtained , either by assigning them randomly or through pre - training , the network is then trained ( further ) by iterative optimisation of the objective function .
first , initial values of the hyperparameters @xmath39 ( for regression networks ) and @xmath42 are set .
the values @xmath39 are set by the user and can be set on either the true output values themselves or on their whitened values ( as defined in section [ sec : nntrain_whiten ] ) .
the only difference between these two settings is the magnitude of the error used .
the algorithm then calculates a large initial estimate for @xmath42 , @xmath65 where @xmath66 is the total number of weights and biases ( nn parameters ) and @xmath67 is a rate set by the user ( @xmath64 , default @xmath68 ) that defines the size of the ` confidence region ' for the gradient .
this expression for @xmath42 sets larger regularisation ( or ` damping ' ) when the magnitude of the gradient of the likelihood is larger .
this relates the amount of ` smoothing ' required to the steepness of the function being smoothed .
the rate factor in the denominator allows us to increase the damping for smaller confidence regions on the value of the gradient .
this results in smaller , more conservative steps that are more likely to result in an increase in the function value but results in more steps being required to reach the optimal weights .
nn training then proceeds using an adapted form of a truncated newton ( or ` hessian - free ' ) optimisation algorithm as described below , to calculate the step , @xmath69 , that should be taken at each iteration .
following each such step , adjustments to @xmath42 and @xmath39 may be made before another step is calculated .
first , @xmath39 can be updated by multiplying it by a value @xmath70 such that @xmath71 .
this serves to assure that at convergence , the @xmath44 value equals the number of unconstrained data points of the problem .
similarly , @xmath42 is then updated such that the probability @xmath72 is maximised for the current set of nn parameters @xmath36 .
these procedures are described in detail by ( * ? ? ?
2.3 & 2.6 ) and ( * ? ? ?
3.6 & appendix b ) . to obtain the step @xmath69 at each iteration
, we first note that one may approximate a general function @xmath73 up to second - order in its taylor expansion by @xmath74 where @xmath75 is the gradient and @xmath76 is the hessian matrix of second derivatives , both evaluated at @xmath36 . for our purposes , the function @xmath73 is the log - posterior distribution of the nn parameters and hence represents a gaussian approximation to the posterior .
the hessian of the log - posterior is the regularised ( ` damped ' ) hessian of the log - likelihood function , where the prior , whose magnitude is set by @xmath42 , provides the regularisation .
if we define the hessian matrix of the log - likelihood as @xmath77 , then @xmath78 , where @xmath79 is the identity matrix . the regularisation parameter @xmath42 can be interpreted as controlling the level of ` conservatism ' in the gaussian approximation to the posterior . in particular
, regularisation helps prevent the optimisation becoming trapped in small local maxima by smoothing out the function being explored .
it also aids in reducing the region of confidence for the gradient information which will make it less likely that a step results in a worse set of parameters .
ideally , we seek a step @xmath69 , such that @xmath80 . using the approximation ,
one thus requires @xmath81 in the standard newton s method of optimisation one simply solves this equation directly for @xmath82 to obtain @xmath83 in principle , iterating this stepping procedure will eventually bring us to a local maximum of @xmath73 .
moreover , newton s method has the important property of being scale - invariant , namely its behaviour is unchanged under any linear rescaling of the parameters .
methods without this property often have problems optimising poorly scaled parameters .
there are , however , some major practical difficulties with the standard newton s method .
first , the hessian @xmath77 of the log - likelihood is not guaranteed to be positive semi - definite .
thus , even after the addition of the damping term @xmath84 derived from the log - prior , the full hessian @xmath85 of the log - posterior may also not be invertible .
second , even if @xmath85 is invertible , the inversion is prohibitively expensive if the number of parameters is large , as is the case even for modestly - sized neural networks . to address the first issue
, we replace the hessian @xmath77 with a form of gauss
newton approximation @xmath86 , which is guaranteed to be positive semi - definite and can be defined both for the regression likelihood and the classification likelihood , respectively @xcite . in particular , the approximation used differs from the classical gauss newton matrix in that it retains some second derivative information .
second , to avoid the prohibitive expense of calculating the inverse in , we instead solve ( with @xmath77 replaced by @xmath86 in @xmath85 ) for @xmath69 iteratively using a conjugate - gradient algorithm , which requires only matrix - vector products @xmath87 for some vector @xmath88 .
one can avoid even the computational burden of calculating and storing the hessian @xmath85 . in principle
, products of the form @xmath87 can be easily computed using finite differences at the cost of a single extra gradient evaluation using the identity @xmath89 this approach is , however , subject to numerical problems .
therefore , we instead calculate @xmath87 products using a stable and efficient procedure applicable to nns @xcite .
this involves an additional forward and backward pass through the network beyond the initial ones required for a gradient calculation .
the combination of all the above methods makes practical the use of second - order derivative information even for large networks and significantly improves the rate of convergence of nn training over standard backpropagation methods .
it has been noted that this method for quasi - newton second - order descent is equivalent to the first - order ` natural gradient ' by @xcite . following each iteration of the optmisation algorithm ,
the posterior , likelihood , correlation , and error squared values are calculated both for the training data and for the validation data ( which were not used in calculating the steps in the optimisation ) .
the correlation of the network outputs is defined for each output @xmath10 as @xmath90 where @xmath91 and @xmath92 are the means of these output variables over all the training data ; the functional dependencies of @xmath93 have been dropped for brevity .
the correlation provides a relative measure of how well the predicted outputs match the true ones . in practice
, the correlations from each output can be averaged together to give an average correlation for the network s predictions .
the average error - squared of the network outputs is defined by @xmath94 ^ 2 } , \label{eq : errsqr_defn}\ ] ] and is complementary to their correlation , since it is an absolute measure of accuracy .
as one might expect , as the optimisation proceeds , there is a steady increase in the values of the posterior , likelihood , correlation , and negative of the error squared , evaluated both for the training and validation data .
eventually , however , the algorithm will begin to overfit , resulting in the continued increase of these quantities when evaluated on the training data , but a decrease in them when evaluated on the validation data .
this divergence in behaviour is taken as indicating that the algorithm has converged and the optimisation in terminated .
the user may choose which of the four quantities listed above is used to determine convergence , although the default is to use the error squared , since it does not include the hyperparameters @xmath95 and @xmath42 in its calculation and is less prone to problems with zeros than the correlation .
we note that the correlation and the error - squared functions discussed above also provide quantitative measures with which to compare the performance of different network architectures , both in terms of the number of hidden nodes and how they are distributed into layers . as network size and complexity
is increased , a point will be reached at which minimal or no further gains may be achieved in increasing correlation or reducing error - squared .
therefore , any network architecture that can achieve this peak performance is equally well - suited . in practice ,
we will wish to find the smallest or simplest network that does so as this minimizes the risk of overfitting and the time required for training .
after training a network , in particular a regression network , one may want to calculate the accuracy of the network s predicted outputs .
a computationally cheap method of estimating this was suggested by @xcite , whereby one adds gaussian noise to the true outputs of the training data and trains a new network on this noisy data . after performing many realisations
, the networks predictions will average to the predictions in the absence of the added noise .
moreover , the standard deviation of their predictions will provide a good estimate of the accuracy of the original network s predictions . since one can train the new networks using the original trained network as a starting point ,
the re - training on noisy data is very fast .
additionally , evaluating the ensemble of predictions to measure the accuracy is not very computationally intensive as network evaluations are simple to perform and can be done in less than a millisecond .
explicitly , the steps of this method are : 1 . start with the converged network with parameters @xmath96 , trained on the original data set @xmath97 .
estimate the noise on the residuals using @xmath98 ^ 2/k$ ] .
2 . define a new data set @xmath99 by adding gaussian noise of zero mean and variance @xmath100 to the outputs ( targets ) in @xmath101 .
3 . train a nn on @xmath99 using the parameters @xmath96 as a starting point .
training should converge rapidly as the new data set is only slightly different from the original .
denote the new network parameters by @xmath102 .
4 . repeat steps ( ii ) and ( iii ) multiple times to obtain an ensemble of networks with parameters @xmath103 .
5 . use each of the networks @xmath103 to make a prediction for a given set of inputs .
the accuracy of the original network s outputs can be estimated as the standard deviation of the outputs of these networks .
in addition to these steps , skynet includes the option for the user to add random gaussian offsets to the parameters @xmath96 before training is performed on the new data set ( step iii ) .
this offset will aid the training in moving the optimisation from a potential local maximum in the posterior distribution of the network parameters , but the size of the offset must be chosen for each problem ; for this , we recommend using a value @xmath104 .
we thus add noise to both the training data and the saved network parameters before training a new network whose posterior maximum will be near to , but not exactly the same as , the original network s . this method may be compared with that described in @xcite for determining the accuracy of the nn predictions for the likelihood used in bambi .
although the method described here requires the overhead time of training additional networks , this is small compared the speed gains possible .
indeed , the new method s accuracy computations require less than a millisecond , as opposed to tenths of a second for the method used previously . consequently , the faster method described here is now incorporated into our new public release version of bambi , leading to around two orders of magnitude increase in speed over that reported in @xcite .
as our first toy example , we consider a simple regression problem . we generate @xmath105 points randomly in the range
@xmath106 $ ] , for which we evaluate the ramped sinc function , @xmath107 and then add gaussian noise with zero mean and a standard deviation of @xmath108 .
the addition of noise makes the regression problem more difficult and prevents any exact solution being possible . to perform the regression , the @xmath105 data items
@xmath109 are divided randomly into @xmath110 items for training and @xmath111 for validation . for this simple problem
, we use a network with a single hidden layer containing @xmath29 nodes ( we denote the full network by @xmath112 ) , and we whiten the input and output data using . the network was not pre - trained .
the optimal value for @xmath29 is determined by comparing the correlation and error - squared for networks with different numbers of hidden nodes .
these results are presented in fig .
[ fig : sincevidence ] , which shows that the correlation increases and the error - squared decreases until we reach @xmath113 hidden nodes , after which both measures level off .
obtained from converged nns with architecture @xmath112 for the ramped sinc function regression problem.,height=215 ] thus , adding additional nodes beyond this number does not improve the accuracy of the network .
for the network with @xmath114 hidden nodes , we obtain a correlation of greater than @xmath115 per cent ; a comparison of the true and predicted outputs in this case is shown in figure [ fig : sincplot ] .
on the training data ( left ) and validation data ( right ) for the ramped sinc function regression problem . ]
we now consider a toy classification problem based on the three - way classification data set created by radford neal for testing his own algorithms for nn training . in this data
set , each of four variables @xmath116 , @xmath117 , @xmath118 , and @xmath119 is drawn @xmath120 times from the standard uniform distribution @xmath121 $ ] .
if the two - dimensional euclidean distance between @xmath122 and @xmath123 is less than @xmath124 , the point is placed in class @xmath55 ; otherwise , if @xmath125 , the class was set to @xmath54 ; and if neither of these conditions is true , the class was set to @xmath126 . note that the values of @xmath118 and @xmath119 play no part in the classification .
gaussian noise with zero mean and standard deviation @xmath127 is then added to the input values .
approximately @xmath32 percent of the data was used for training and the remaining @xmath128 per cent for validation .
we again use a network with a single hidden layer containing @xmath29 nodes , and we whiten the input and output data using . the network was not pre - trained .
the full network thus has the architecture @xmath129 , where the three output nodes give the probabilities ( which sum to unity ) that the input data belong to class 0 , 1 , or 2 , respectively .
the final class assigned is that having the largest probability . the optimal value for @xmath29 is again determined by comparing the correlation and error - squared for networks with different numbers of hidden nodes .
these results are shown in fig .
[ fig : classprobstruct ] , from which we see that the correlation increases and the error - squared decreases until we reach @xmath130 hidden nodes , after which both measures level off . for the network with @xmath131 hidden nodes ,
a total of @xmath132 per cent of training data points and @xmath133 per cent of validation points were correctly classified .
a summary of the classification results for this network is given table [ tab : cdatatrain ] .
.classification results for the converged nn with architecture @xmath134 for the neal data set . [ cols="^,^,^,^,^,^ " , ] these results show that the extra information given to the regression networks trained on @xmath135 feature values from the autoencoder acted as a disadvantage in predicting the galaxy ellipticities . for networks trained on @xmath111 or @xmath136 features , however , the accuracies of the predicted ellipticities were better even than those obtained using the full original images in some cases .
this demonstrates the power of being able to eliminate redundant information and noise , and thereby improve the accuracy of the analysis .
we also observe that adding unnecessary complexity to the nn structure makes it more difficult for the algorithm to find the global maximum .
this same method for dimensionality reduction which also eliminates noise before performing measurements can clearly be applied to a wide range of other astronomical applications .
examples include classification of supernovae by type , or measurements of galaxies and stars by their spectra .
we have described an efficient and robust neural network training algorithm , called skynet , which we have now made freely available for academic purposes .
this generic tool is capable of training large and deep feed - forward networks , including autoencoders , and may be applied to supervised and unsupervised machine learning tasks in astronomy , such as regression , classification , density estimation , clustering and dimensionality reduction .
skynet employs a combination of ( optional ) pre - training followed by iterative refinement of the network parameters using a regularised variant of newton s optimisation algorithm that incorporates second - order derivative information without the need even to compute or store the hessian matrix .
linear and sigmoidal activation functions are provided for the user to choose between .
skynet adopts convergence criteria that naturally prevent overfitting , and it also includes a fast algorithm for estimating the accuracy of network outputs .
we first demonstrate the capabilities of skynet on toy examples of regression , classification , and dimensionality reduction using autoencoder networks , and then apply it to the classic machine learning problem of handwriting classification for determining digits from the mnist database . in an astronomical context , skynet is applied to : the regression problem of measuring the ellipticity of noisy and convolved galaxy images in the mapping dark matter challenge ; the classification problem of identifying gamma - ray bursters that are detectable by the swift satellite ; and the dimensionality reduction problem of compressing and denoising images of galaxies . in each case
, the straightforward use of skynet produces networks that perform the desired task quickly and accurately , and typically achieve results that are competitive with machine learning approaches that have been tailored to the required task .
future development of skynet will expand upon many of the current features and introduce new ones .
we are working to include more activation functions ( e.g. @xmath15 , softsign , and rectified linear ) , pooling of nodes , convolutional nns , diversity in outputs ( i.e. mixing regression and classification ) , and more robust support of recursive nns .
this is all in addition to further improving the speed and efficiency of the training algorithm itself .
however , skynet in its current state is already a useful tool for performing machine learning in astronomy .
the authors thank john skilling for providing very useful advice in the early stages of algorithm development .
we also thank amy lien for providing the data used in seciton [ sec : grb ] .
this work utilized three different high - performance computing facilities at different times : initial work was performed on cosmos viii , an sgi altix uv1000 supercomputer , funded by sgi / intel , hefce and pparc , and the authors thank andrey kaliazin for assistance ; early work also utilized the darwin supercomputer of the university of cambridge high performance computing service ( ` http://www.hpc.cam.ac.uk/ ` ) , provided by dell inc . using strategic research infrastructure funding from the higher education funding council for england
; later work utilised the discover system of the nasa center for climate simulation at nasa goddard space flight center .
pg is currently supported by a nasa postdoctoral fellowship from the oak ridge associated universities and completed a portion of this work while funded by a gates cambridge scholarship at the university of cambridge .
ff is supported by a research fellowship from the leverhulme and newton trusts . 1 andreon s. , gargiulo g. , longo g. , tagliaferri r. , & capuano n. , 1999 , arxiv : astro - ph/9906099 andreon s. , gargiulo g. , longo g. , tagliaferri r. , & capuano n. , 2000 , mnras , 319 , 700716 auld t. , bridges m. , hobson m.p .
, gull s.f . , 2008 , mnras , 376 , l11 auld t. , bridges m. , hobson m.p . , 2008 , mnras , 387 , 1575 ball n.m .
, brunner r.j . , 2010 , int
. j. mod .
, 19 , 1049 bergstra j. , desjardins g. , lamblin p. , & bengio y. , 2009 , technical report 1337 , dpartement dinformatique et de recherche oprationnelle , universit de montral .
bertin e. , arnouts s. , 1996 , a&as supplement , 117 , 393 bridges m. , cranmer k. , feroz f. , hobson m.p . , ruiz de austri r. , trotta r. , 2011 , jhep , 03 , 012 bonnett c. , 2013 , arxiv:1312.1287 [ astro-ph.co ] carreira - perpignan m. a. & hinton . g. e. , 2005 , proceedings of the tenth international workshop on artificial intelligence and statistics , eds . cowell r. g. & ghahramani z. , 3340 ciresan d. c. , meier u. , gambardella l. m. , & schmidhuber j. , 2010 , neural comput . , 22 , 32073220 cybenko g. , 1989 , mathematics of control , signals , and systems , 2 , 303314 erhan d. et al . , 2010 ,
journal of machine learning research , 11 , 625660 fawcett t. , 2006 , pattern recogn .
lett . , 27 , 861 fendt w.a . ,
wandelt b.d . , 2007 , apj , 654 , 2 feroz f. , hobson m.p . , 2008 , mnras , 384 , 449 feroz f. , hobson m.p . ,
bridges m. , 2009 , mnras , 398 , 1601 feroz f. , hobson m. p. , cameron e. , & pettitt a. n. , 2013 , arxiv:1306.2144 [ astro-ph.im ] feroz f. , marshall p. j. , hobson m. p. , 2008 ,
[ astro - ph ] fynbo j. et al . , 2009 ,
apjs , 185 , 526 gehrels n. et al . , 2004 ,
apj , 611 , 1005 geva s. , sitte j. , ieee , 3 , 621 glorot x. & bengio y , 2010 , proceedings of the thirteenth international conference on artificial intelligence and statistics , journal of machine learning research , eds .
teh y. w. & titterington m. , 249256 glorot x. , bordes a. , & bengio y. , 2011 , proceedings of the fourteenth international conference on artificial intelligence and statistics , journal of machine learning research , eds .
gordon g. & dunson d. , 315323 graff p. , feroz f. , hobson m.p .
, lasenby a.n . , 2012 , mnras , 421 , 169 gull s.f .
& skilling j. , 1999 , quantified maximum entropy : memsys 5 users manual .
maximum entropy data consultants ltd .
edmunds , suffolk , uk . ` http://www.maxent.co.uk/ ` hinton g.e . ,
osindero s. , & teh y .- w . , 2006 , neural comput . , 18 , 15271554 hinton g.e . &
salakhutdinov r.r . , 2006 ,
science , 313 , 504 - 507 hobson m. p. , jones a. w. , lasenby a. n. , & bouchet f. r. , 1998 , mnras , 300 , 129 hornik k. , stinchcombe m. & white h. , 1990 , neural networks , 3 , 359 hyvrinen a. , oja e. , 2000 , neural networks , 13 , 411 karpenka n.v . , feroz f. , hobson m.p . , 2013 , mnras , 429 , 12781285 kendall m.g .
, 1957 , a course in multivariate analysis .
griffin , london kitching t. et al .
, 2011 , annals of applied statistics , 5 , 22312263 kitching t. et al . , 2012 , new astronomy reviews , submitted lecun y. , bottou l. , bengio y. , & haffner p. , 1998
, proc . of the ieee , 86 , 22782324 lien a. , sakamoto t. , gehrels n. , palmer d. , graziani c. , 2012 , proceedings of the international astronomical union , 279 , 347 longo g. , tagliaferri r. , & andreon s. , 2001 , mining the sky : proceedings of the mpa / eso / mpe workshop , eds .
banday a. j. , zaroubi s. , bartelmann m. , 379385 mackay d.j.c .
, 1992 , neural computation , 4 , 415447 mackay d.j.c . , 1995 ,
network : computation in neural systems , 6 , 469 mackay d.j.c , 2003 , information theory , inference , and learning algorithms .
cambridge univ . press . ` www.inference.phy.cam.ac.uk/mackay/itila/ ` mandic d. , chambers j. , 2001 , recurrent neural networks for prediction : learning algorithms , architectures and stability .
wiley , new york .
martens j. , 2010 , in frnkranz j. , joachims t. , eds , proc .
machine learning .
omnipress , haifa , p. 735 murtagh f. , 1991 , neural comput . , 2 , 183 pascanu r. & bengio y. , 2013 , arxiv:1301.3584 [ cs.lg ] pearlmutter b.a . , 1994 , neural comput . , 6 , 147 sanger t.d . ,
1989 , neural networks , 2 , 459 schraudolph n.n . , 2002 , neural comput .
, 14 , 1723 serra - ricart m. , calbet x. , garrido l. , & gaitan v. , 1993 , aj , 106 , 1685 skilling j. , 2004 , aip conference series , 735 , 395 tagliaferri r. et al . , 2003a , neural networks , 16 , 297 tagliaferri r. , longo g. , andreon s. , capozziello s. , donalek c. , & giordano g. , 2003b , neural nets : 14th italian workshop on neural nets , eds .
apolloni b. , marinaro m. , & tagliaferri r. , 226234 wanderman d. , piran t. , 2010 , mnras , 406 , 1944 way m.j .
, scargle j.d . , ali k.m . ,
srivastava a.n . , 2012 ,
advances in machine learning and data mining for astronomy .
crc press . | we present the first public release of our generic neural network training algorithm , called skynet .
this efficient and robust machine learning tool is able to train large and deep feed - forward neural networks , including autoencoders , for use in a wide range of supervised and unsupervised learning applications , such as regression , classification , density estimation , clustering and dimensionality reduction .
skynet uses a ` pre - training ' method to obtain a set of network parameters that has empirically been shown to be close to a good solution , followed by further optimisation using a regularised variant of newton s method , where the level of regularisation is determined and adjusted automatically ; the latter uses second - order derivative information to improve convergence , but without the need to evaluate or store the full hessian matrix , by using a fast approximate method to calculate hessian - vector products .
this combination of methods allows for the training of complicated networks that are difficult to optimise using standard backpropagation techniques .
skynet employs convergence criteria that naturally prevent overfitting , and also includes a fast algorithm for estimating the accuracy of network outputs .
the utility and flexibility of skynet are demonstrated by application to a number of toy problems , and to astronomical problems focusing on the recovery of structure from blurred and noisy images , the identification of gamma - ray bursters , and the compression and denoising of galaxy images .
the skynet software , which is implemented in standard ansi c and fully parallelised using mpi , is available at http://www.mrao.cam.ac.uk / software / skynet/. [ firstpage ] methods : data analysis methods : statistical |
the model we consider is shown in fig . [ sketch1 ] . a double dot ( dd ) plays the role of a two - level system : the topmost electron in the dd
can either occupy the upper or the lower dot . a mesoscopic two - terminal conductor ( mc ) serves as a detector :
its conductance is sensitive to the charge on the upper dot .
a set of capacitances @xmath0 describes the coupling between charges of system @xmath1 and detector @xmath2 ( we often use the capacitance in series @xmath3 ) .
single electron movement in such a setup was recently measured @xcite .
three different time scales describe the interaction of system and detector : in the system we distinguish the thermal _ relaxation _ to an equilibrium distribution ( described by a rate @xmath4 ) and the often much faster _ decoherence _ of superpositions of states in the upper and lower dot ( described by a rate @xmath5 ) .
the decoherence depends as well on temperature @xmath6 as on the voltage difference @xmath7 between the terminals of the detector .
applying such a voltage difference permits to measure the state of the system which leads to additional decoherence .
the _ measurement _ process takes some time which is needed to overcome the uncertainty due to shot noise in the detector and is characterized by a third rate @xmath8 .
the decoherence rate @xmath5 at zero temperature is intimately related to the measurement rate @xmath8 and satisfies the inequality @xmath9 @xcite .
an important measure of the quality of the detector is the efficiency @xmath10 .
ideally one would like to find detectors with an efficiency @xmath11 . even in the single channel case an ideal detector @xmath12
is found only under some special conditions ( time - reversal and space - inversion symmetry @xcite ) .
it is shown in ref .
@xcite that an efficient multi - channel detector is subject to a _
third _ condition that links sensitivities and shot noises of different channels @xmath13 where @xmath14 denotes the transmission probability of channel @xmath15 .
a violation of this condition can reduce the efficiency of multichannel detectors @xmath16 drastically in the presence of disorder . in general a large mesoscopic detector is therefore not much faster than a single channel detector . in this publication
we concentrate on another property that lowers the detection speed of multichannel detectors .
the charge fluctuation in one channel can be screened by the charge of other channels .
this effect will depend strongly on the geometry of the detector . in order to be able to describe a large variety of detectors we therefore represent the mc by a scattering matrix @xmath17 that connects in- and outgoing states ( @xmath18 label left and right reservoir ) .
this enables us to treat multi - channel mcs with arbitrary transmission probabilities @xmath19 .
we only consider the case of weak coupling between mc and dd and may thus use a standard master equation ( bloch - redfield approach @xcite ) in lowest order perturbation theory to study the evolution of the reduced density matrix of the dd . on this level of approximation
the dynamics of the dd is influenced only via the charge fluctuation spectrum @xmath20 of the mc . at low frequencies this spectrum
is fully characterized by a generalized wigner - smith time delay matrix @xcite ( @xmath21 label the reservoirs ) @xmath22 the derivative @xmath23 is taken with respect to the electrostatic potential in the mc .
the following constants that appear also in the context of ac - transport @xcite describe the geometry of the detector ( @xmath24 denotes the electron charge ) @xmath25 @xmath26 corresponds to the density of states at fermi energy in the scattering region , @xmath27 is an effective electrochemical capacitance that characterizes the strength of interaction , @xmath28 expresses the equilibrium contribution @xcite to the charge fluctuation spectrum @xmath20 , and @xmath29 the non - equilibrium contribution @xcite .
the fifth constant @xmath30 is of different origin @xcite and can not be expressed by matrix ( [ density elements ] ) only .
it describes the ratio of detector sensitivity and shot noise in the presence of screening .
the two - level system is conventionally represented by the hamiltonian @xmath31 where @xmath32 denote pauli matrices .
the energy difference between upper and lower dot is @xmath33 and @xmath34 accounts for tunneling between the dots .
the full level splitting is thus @xmath35 .
the relaxation , decoherence and measurement rates are given by the following expressions @xcite ( here and in the following we set @xmath36 ) : @xmath37 @xmath38 @xmath39 they have formally the same appearance as the rates given in @xcite .
the geometric structure of the detector is contained in the five parameters given in ( [ four parameters ] ) . in the following
we analyze the behavior of these parameters to discuss the screening properties of the detector .
the parameter @xmath28 does not exceed the range @xmath40 where @xmath41 is the dimension of the scattering matrix .
this demonstrates that the relaxation and decoherence rates @xmath42 do not simply scale with the number of channels through the system . the multichannel result for the relaxation and decoherence rates
can not be obtained as a sum of rates due to each channel .
the constants @xmath28 and @xmath29 are renormalized by a denominator @xmath43 depending on all channels .
this denominator originates from screening . for a large number @xmath41 of open channels
@xmath28 behaves as @xmath44 whereas the electrochemical capacitance @xmath45 tends to a constant .
the constant @xmath29 decreases also like @xmath44 .
we find therefore the somewhat surprising result that relaxation and decoherence decrease in the large channel limit .
this result is a consequence of screening in the mc which reduces the charge fluctuations with increasing channel number @xmath41 .
we briefly explain the derivation of our results .
the coulomb hamiltonian of our model contains three terms @xmath46 its first term contributes to the level splitting of the dd ( @xmath47 is a background charge depending on the applied voltage @xmath48 ) .
the charging energy @xmath49 must be large compared to @xmath50 to allow us to consider only two levels of the dd . the second term @xmath51 couples system and detector . to derive a master equation for the reduced density matrix we assume weak coupling and treat this term perturbatively .
we apply a markov approximation which is strictly speaking only valid at long time scales ( compared to the correlation time of the detector ) .
this permits us to consider only the low - frequency matrix elements of the charge operator that are given by the wigner - smith matrix ( [ density elements ] ) .
the third term affects the fluctuation spectrum @xmath52 of the charge operator @xmath53 .
in contrast to earlier work ( with the exception of ref .
@xcite ) we do not completely disregard this term , but include it on the level of rpa .
this allows us to treat screening in a gaussian approximation valid for geometries in which coulomb blockade effects are weak . to get insight into the fluctuation spectrum , we study the time evolution of the charge @xmath53 on the mc within rpa
this charge is composed of two components : the bare charge @xmath54 that arises from the noninteracting problem and the screening charge which is a response to the self - consistent potential @xmath55 on the mc .
we therefore have in frequency representation @xmath56 in the first equation , @xmath57 is the linear response function .
the second equation expresses the self - consistency condition for the potential @xmath55 .
these two equations can be combined by eliminating the potential @xmath55 . the total charge on the mc is found to be @xmath58 this result shows nicely the two possible effects of the random phase approximation . on the one hand , the prefactor @xmath59 reduces the fluctuations of the bare charge @xmath54 which is compensated partially by a screening charge . on the other hand , we get a back action of the charge @xmath60 of the upper dot on the charge @xmath53 in the mc .
however , this back action is small by a factor @xmath61 and can be omitted in second order perturbation theory .
the examples of this section are not meant to model the geometry of a realistic measuring apparatus .
but they demonstrate nicely the influence of the geometry on the speed and efficiency of the measuring process . as a first generic example we discuss a * short quantum point contact * which is defined by two sharp barriers of distance @xmath62 .
its potential is given by @xmath63 with @xmath64.\ ] ] note that this potential violates inversion symmetry for @xmath65 .
for convenience we introduce the following parameters : the wave vector @xmath66 where @xmath67 is the effective mass of the electron and @xmath68 the fermi energy .
we consider the limit of a mc much shorter than the fermi wave length @xmath69 .
using the general formula ( [ density elements ] ) we calculate the density of state matrix ( we set @xmath36 ) @xmath70 it is now straightforward to obtain the measurement efficiency of the mc at zero temperature @xmath71 for an asymmetric mc with @xmath65 the efficiency is suboptimal . for a * long quantum point contact * with @xmath72 charge screening
gets effective .
the constants @xmath28 and @xmath29 that determine the relaxation and decoherence rates then decrease with increasing number of open channels @xmath41 .
we illustrate this effect using the following toy potential @xmath73 its scattering matrix can be found in @xcite . to get the full density of state matrix ( [ density elements ] ) we replace the potential derivative @xmath23 in ( [ density elements ] ) by a parametric derivative @xmath74 .
the elements of ( [ density elements ] ) can be expressed by digamma - functions . as a confining potential
we choose @xmath75 . as in a two - dimensional electron gas
only the lowest mode in z - direction is occupied .
[ integrable qpc ] compares the cases of ineffective charge screening @xmath76 and effective screening @xmath77 .
it shows the decoherence ( measurement ) rate at zero temperature @xmath6 and finite bias @xmath7 depending on the number of open channels through the constriction .
we change this number by varying the confinement frequency @xmath78 .
the decoherence rate reaches maximal values if one transmission channel @xmath15 is half - open @xmath79 .
this corresponds to a maximum in the sensitivity @xmath80 . in the case of
ineffective screening the decoherence rate does not depend on the number of open channels , whereas screening reduces decoherence in the case of many open channels .
in this work we have investigated the weak measurement of a two state system ( a double quantum dot ) capacitively coupled to a mesoscopic conductor .
we have presented expressions for the relaxation rate , the decoherence rate and the measurement rate in terms of capacitance coefficients and in terms of potential derivatives of the scattering matrix .
the discussion illustrates that scattering theory is useful not only for the discussion of dc - transport processes in mesoscopic physics but in fact has a much wider range of applicability .
we have emphasized screening in the detector : as illustrated in fig .
[ integrable qpc ] the decoherence rate and similarly the relaxation and measurement rate ) are rappidly reduced with increasing channel number .
our discussion is based only on a single fluctuating potential in the detector but in principle a refined calculation which takes into account a more realistic potential distribution is possible @xcite .
this work was supported by the swiss national science foundation . + * note added in proof : * recently an instructive , information theoretical discussion of the problem treated in this work was provided by clerk , girvin , and stone @xcite . for recent experimental advances in the read - out of dynamic states
see : d. vion , a. aassime , a. cottet , p. joyez , h. pothier , c. urbina , d. esteve , and m. h. devoret , science * 296 * , 886 ( 2002 ) . y. yu , s. han , x. chu , s .-
chu , z. wang , science * 296 * , 889 ( 2002 ) ; y. nakamura , yu .
a. pashkin , and j. s. tsai , nature ( london ) * 398 * , 357 ( 2001 ) . | we consider a two - level system coupled to a mesoscopic two - terminal conductor that acts as measuring device . as a convenient description of the conductor
we introduce its scattering matrix .
we show how its elements can be used to calculate the relaxation and decoherence rates of the two - level system .
special emphasis is laid on the charge screening in the conductor that becomes important in the many - channel limit .
finally we give some examples that illustrate charge screening in different limits .
the detection of quantum states by means of mesoscopic detectors is of fundamental importance from the point of view of quantum measurement theory .
it is also important due to the recent interest in different schemes of quantum computation for which the read - out process must be a key step @xcite . a detector ,
typically , has a back action @xcite on the measured system and thus can also be used to introduce decoherence in a controlled way .
here we are concerned with a simple prototype of a detector that consists of a mesoscopic conductor capacitively coupled to the state of a nearby quantum system ( see fig .
( [ sketch1 ] ) ) .
the most basic quantum system is a two state system here taken to be two mesoscopic quantum dots weakly coupled to each other .
recent experiments demonstrated elegantly the effect of a detector on a phase coherent mesoscopic system @xcite .
different aspects of weak measurement were addressed in several theoretical discussions : the relation between detector noise and decoherence @xcite , their connection to measurement time @xcite and to scattering theory @xcite .
master equation approaches have been used to study the time evolution of system and detector @xcite .
these equations have been refined to describe also conditional evolution depending on the outcome of the measurement @xcite .
tunnel contacts and single electron transistors have been identified as candidates for efficient measurement devices @xcite . for such devices
it is natural to ask whether their speed can be improved by increasing their size .
one is tempted to expect that the sensitivity of a detector grows when we add more and more conductance channels .
the bigger the detector the less the uncertainty due to the shot noise compared to the signal of the detector
. however , this effect competes with the charge screening between different conductance channels through the detector that grows as well with increasing channel number .
this screening tends to suppress the sensitivity of the detector .
charge screening is often neglected in mesoscopic physics .
typically investigated quantities are dc - currents and low frequency current noises which are insensitive to displacement currents in the system .
however , there are quantities that depend crucially on screening : ac - conductance measurements pick up screening currents in the system : at contacts which permit the exchange of particles these become noticeable only if the frequency gets of the order of the inverse dwell time in the system ; at nearby capacitors only displacement currents contribute and can be measured at arbitrary low frequencies @xcite . furthermore non - linear current - voltage characteristics depend on the charge distribution inside mesoscopic conductors @xcite . in the case of mesoscopic detectors the quantity of interest is the charge fluctuations in the detector . like the currents induced into a nearby gate this quantity is subject to screening already in the low - frequency limit
! earlier work focused on the effect of screening on the decoherence rate induced by the measurement process under the assumptions that both detector and system are described by a scattering matrix @xcite .
it is the aim of this work to investigate the effect of charge screening in the detector on the dd system shown in fig .
[ sketch1 ] . |
a measurement of the complete breakup of the atomic helium target by xuv radiation was achieved over @xmath3 years ago @xcite . since then , rapid developments in strong xuv light sources and momentum imaging techniques have made it possible to record all the reaction fragments , nuclei and electrons , in double photoionization of the simplest two - electron hydrogen / deuterium molecule by one - photon absorption @xcite , and most recently also for two - photon absorption @xcite . for the double ionization of h@xmath4 by single - photon absorption , only randomly oriented molecules were investigated in earlier experiments ( e.g. @xcite ) . using
fixed - in - space " molecules , more recent experimental efforts include the measurements of energy- and angle - resolved differential cross sections by weber _
et al . _
for either equal energy sharing @xcite or asymmetric energy sharing @xcite , and by gisselbrecht _
@xcite for equal energy sharing at a photon energy of @xmath5 ev .
these experimental studies were at least partially stimulated by the goal of understanding the similarities and differences between the hydrogen molecule and its atomic counterpart helium .
however , all recorded fully differential cross sections to date suffer from some experimental uncertainties regarding the alignment angle of the molecule with respect to the polarization vector of the laser and the emission angles of the photoelectrons . from a theoretical point of view , the hydrogen molecule exhibits a significant complexity compared to helium and , therefore , provides an enormous challenge to a fully _ ab initio _ description inherent in a multi - center , multi - electron system .
the single - center convergent close - coupling method was used to model the double ionization of h@xmath4 by kheifets and bray @xcite .
later mccurdy , rescigno , martn and their collaborators @xcite implemented a formulation based on time - independent exterior complex scaling ( ecs ) in spherical coordinates , with the origin of the coordinate system placed at the center of the molecule , to treat the double photoionization at a photon energy of @xmath0 ev . the radial coordinates of the two electrons are measured from the center , and the radial parts of the wave function were either expanded in @xmath6-splines or using a finite - element discrete - variable representation ( fe - dvr ) . the time - dependent close - coupling ( tdcc ) method @xcite , again in spherical coordinates , was also extended to calculate the triple - differential cross section ( tdcs ) for double photoionization of the h@xmath4 molecule . while the agreement between the published tdcss from the ecs @xcite and tdcc @xcite calculations is basically acceptable
, noticeable discrepancies remained for a few particular geometries . in the parallel geometry , for instance , where the molecular axis @xmath7 is chosen along the laser polarization vector @xmath8 , the coplanar tdcs predictions from the ecs and tdcc calculations differ by up to @xmath9 percent when one of the electrons ( we will refer to it as the `` fixed electron '' below ) is observed along the direction perpendicular to the @xmath8-axis . in some other cases ,
there exists a noticeable wing " structure in the published tdcc predictions for equal energy sharing .
additional tdcc calculations @xcite suggest that the agreement can be systematically improved , albeit the above - mentioned discrepancy still exists at a somewhat reduced level .
another independent approach @xcite to this problem is the very recent time - independent ecs treatment , formulated as in the current work in prolate spheroidal coordinates .
quite surprisingly , the results of that calculation differed from both the earlier ecs and also the tdcc predictions , both obtained in spherical coordinates . specifically , the ecs prolate spheroidal calculations showed differences from the earlier spherical coordinate calculation for the tdcss , at a level of about @xmath10 depending on the details of the energy sharing .
as will be demonstrated below , we have gone to considerable lengths in an attempt to resolve these discrepancies .
however , significant differences between the present results and those of the ecs @xcite still remain .
both the ecs @xcite and the tdcc @xcite calculations made some attempt to deal with the experimental uncertainties in the scattering angles .
given the experimental uncertainties and the differences in the previous calculations , however , it appeared worthwhile to investigate the computational effort required to obtain accurate before averaging over any experimental acceptance angles .
consequently , the present calculation represents an independent implementation of the time - dependent fe - dvr approach in prolate spheroidal coordinates . as in the other approaches mentioned above , the internuclear separation ( @xmath11 )
was held fixed at its equilibrium distance of @xmath12 bohr .
the two - center prolate system , with the foci located on the nuclei , provides a suitable description for the two - center characteristics of the h@xmath4 molecule .
the formulation of the schrdinger equation in prolate spheroidal coordinates for diatomic molecules is not new .
the pioneering work of bates , pik , and poots @xcite for the h@xmath13 ion , which is exactly separable in prolate spheroidal coordinates , already revealed the appealing features of the prolate system .
in particular , the electron - nuclear interaction is rendered benign in this coordinate system .
a partial list of recent applications of prolate spheroidal coordinates to diatomic molecules can be found in @xcite .
as has been demonstrated in a number of recent publications , a grid - based approach provides a very appropriate description of laser - driven atomic and molecular physics when combined with an efficient time - propagation method such as the short iterative lanczos ( sil ) method @xcite . in the present work ,
we employ the fe - dvr / sil approach in prolate spheroidal coordinates to study the correlated response of a two - electron molecule in the double ionization process .
the remainder of this manuscript is organized as follows . after presenting the hamiltonian of the hydrogen molecule in section [ sec.2 ] and providing some details about the discretization of the system in an fe - dvr basis in sec .
[ sec.3 ] , the solution of the two - center poisson equation is presented in sec . [ sec.4 ] .
this is followed by a description of the procedure for extracting the cross sections of interest in sec .
[ sec.5 ] .
the results are presented and discussed in sec .
[ sec.6 ] , before we finish with a summary in sec .
[ sec.7 ] .
the prolate spheroidal coordinates with the two foci separated by a distance @xmath11 are defined by @xmath14 and the azimuthal angle @xmath15 . here
@xmath16 and @xmath17 are the distances measured from the two nuclei , respectively .
these coordinates are specified in the ranges @xmath18 , @xmath19 $ ] , and @xmath20 $ ] .
the volume element is @xmath21 . according to the asymptotic behaviors as @xmath22 , @xmath1 and
@xmath2 approach @xmath23 and @xmath24 , respectively , where @xmath25 and @xmath26 are the standard spherical coordinates .
consequently , @xmath1 is the quasi - radial " coordinate , while @xmath2 is quasi - angular " .
the hamiltonian of a single electron , @xmath27 , ( @xmath28 for the two electrons in h@xmath4 below ) , is written as @xmath29 -\frac{4\xi_q}{r(\xi_q^2-\eta_q^2)}.\end{aligned}\ ] ] we solve the time - dependent schrdinger equation ( tdse ) of the laser - driven h@xmath4 molecule ( with two electrons ) in the dipole length gauge : @xmath30\psi(1,2,t).\ ] ] here @xmath31 is the coordinate of the @xmath32-th electron measured relative to the center of the molecule and @xmath33 is the interelectronic distance . without loss of generality ,
we choose the molecular axis along the @xmath34 axis , and the plane formed by the molecular axis and the polarization vector as the @xmath35 plane .
generally , we can decompose the polarization vector into its two components , @xmath36 , where @xmath37 is the angle between the @xmath38 and @xmath39 axes , and @xmath40 and @xmath41 are the unit vectors along the @xmath34 and @xmath42 axes , respectively .
the dipole interaction is therefore given as @xmath43,\end{aligned}\ ] ] where the rectangular coordinates @xmath42 and @xmath34 and the prolate spheroidal coordinates are related through @xmath44 we expand the wave function for the h@xmath4 molecule in the body - frame as @xmath45 here @xmath46 is the angular function , where @xmath47 and @xmath48 denote the magnetic quantum numbers of the two electrons along the molecular axis .
next , @xmath49 is expanded in a product of normalized radial " @xmath50 and angular " @xmath51 dvr bases : @xmath52 note that the basis is not symmetrized with respect to the coordinates of the two electrons . since we always begin the calculation with a properly symmetrized initial state , however , that symmetry will be preserved in the calculation . to discretize this partial differential equation
, we employ the fe - dvr approach for both the @xmath1 and @xmath2 variables @xcite .
if we normalize the dvr bases according to @xmath53 respectively , the overall @xmath54 dvr basis is not normalized with respect to the volume element in the prolate coordinate system .
this is corrected by defining the two - electron basis @xmath55 which satisfies the desired normalization @xmath56 to expand @xmath49 .
specifically , we have @xmath57 introducing the normalized @xmath54 dvr basis eliminates the complexities of matrix operations related to the overlap matrix at each time , and hence makes the standard sil algorithm directly applicable to study the temporal response of the molecule to laser pulses .
in our current implementation of the fe - dvr approach for the time - dependent wave function in prolate spheroidal coordinates , we have chosen to work directly in the fe - dvr basis .
this differs from what is usually done for atoms in spherical coordinates , where spherical harmonics are used for the angular variables and an fe - dvr for the radial coordinates .
consequently , the boundary conditions in prolate spheroidal coordinates require some more discussion . analyzing the asymptotics
reveals that in the region near the boundaries of @xmath58 and @xmath59 , which correspond to the molecular axis , the single - electron wave function behaves like @xmath60 .
this indicates that the physical wave function is finite for @xmath61 , whereas it goes to zero in the region close to the molecular axis for @xmath62 .
more importantly , the behavior of the wave function for odd @xmath63 contains a square - root factor , giving a decidedly nonpolynomial behavior to the wave function that is impossible to capture in a straightforward fashion using a dvr basis .
the former problem is readily treated by using a gauss - radau quadrature in the first dvr element for @xmath1 , where only the right - most point is constrained to lie on the boundary between the first and second finite element .
the volume element ensures that the integrand is well behaved near the end points and makes it unnecessary to invoke a separate quadrature for different @xmath64 values .
for all the other elements , a gauss - lobatto quadrature is employed .
this allows us to make the fe - dvr basis continuous everywhere and to satisfy the @xmath63-dependent boundary condition . to overcome the nonanalytic behavior of the basis for odd @xmath64 , bachau and collaborators
@xcite explicitly factored out the @xmath60 part before the wave function was expanded in terms of @xmath6-splines in the discretization approach .
we have adopted a similar idea in our fe - dvr treatment of the two - center problem to achieve much faster convergence , as was also done in ref .
@xcite . for the case of even @xmath63 ,
no changes need to be made to define the dvr basis , i.e. , the normalized basis is written as @xmath65 for odd @xmath63 , however , we define the dvr basis as @xmath66 and @xmath67 here @xmath68 and @xmath69 are the weight factors related to the dvr bases @xmath70 and @xmath71 , respectively .
the goal of using a unique set of mesh points , which are @xmath63-independent , to discretize the @xmath54 coordinates has now been achieved in this scheme .
the same technique was employed in recent calculations of one- and two - photon double ionization of h@xmath4 @xcite . in principle , it is also possible to introduce the factors into the dvr bases to circumvent the difficulties related to the nonanalytic behavior near the boundary . however , this results in an @xmath63-dependence of the dvr bases and quadrature points .
this , in turn , leads to a number of unnecessary complications in the practical implementation of the computational methodology .
one might argue that an discretization procedure could be useful for a system in which the magnetic quantum number @xmath64 is conserved .
an example is the h@xmath13 ion in external magnetic fields along the molecular axis @xcite .
however , that is not the situation in the current calculation .
similar to the expansion of the electron - electron interaction in terms of spherical coordinates , a counterpart exists in prolate spheroidal coordinates @xcite through the neumann expansion @xmath72 where @xmath73 .
the two nuclei are located at @xmath74 along the @xmath34 axis and @xmath75 .
both the regular @xmath76 and irregular @xmath77 legendre functions @xcite , which are defined in the region @xmath78 , are involved in the expansion as the radial " part , while the angular " part is only related to @xmath79 .
note that we chose to work in terms of an un - normalized angular " basis rather than the usual spherical harmonics .
the matrix elements of @xmath80 in a traditional basis , for example , a or slater - type basis , can be computed through the well - known mehler - ruedenberg transformation @xcite . due to the discontinuous derivative along the line of @xmath81 in the neumann expansion , the straightforward computation of the matrix element of @xmath80 , using the value of this interaction potential at the mesh points , is very slowly convergent .
we seek a more robust representation of the @xmath80 matrix which retains both the underlying gauss quadrature and the dvr property of all potentials being exactly diagonal with respect to the highly localized dvr basis . in the following ,
we use the simplified notation @xmath82 to denote the basis .
essentially , we need the integral @xmath83 after integrating over @xmath84 and @xmath85 , the matrix element of @xmath80 can be written as @xmath86 where the selection rule @xmath87 has been used .
hence @xmath64 is uniquely determined for a given pair of angular partial waves . above we introduced the reduced ( @xmath88 ) integral @xmath89 since @xmath63
is fixed in the above equation , we omitted it in @xmath90 and will do so in the related quantities below as well .
it is now worthwhile to define the two electron densities @xcite : @xmath91 after truncating the radial integral to the edge of the box , @xmath92 , this yields @xmath93 here a convention for the volume element was made in such a way that , for any function @xmath94 , we define @xmath95 to simplify the notation .
most importantly , the function @xmath96 is defined by @xmath97 instead of evaluating the above integrals directly , we solve the differential equation satisfied by @xmath96 .
as will become apparent later , this equation can be shown to be the radial " poisson equation in the prolate spheroidal coordinate system .
the differential equations satisfied by the legendre functions @xmath76 and @xmath77 suggests that we introduce the operator @xmath98 for given quantum numbers @xmath99 and @xmath64 .
this is the one - dimensional laplacian operator in the @xmath1 coordinate .
after some algebra , we obtain @xmath100 and @xmath101 here the wronskian of the legendre functions @xmath76 and @xmath77 is given by @xmath102 consequently , we obtain @xmath103 this is the second - order inhomogeneous poisson equation satisfied by @xmath104 with the source " term given by @xmath105 after carrying out the integral over @xmath106 via gauss quadrature , we recast the source term as @xmath107 as one might expect , the inhomogeneous equation reduces to the homogeneous one if @xmath108 .
the solution to the poisson equations ( [ poissoneq ] ) and ( [ boundcond ] ) can be uniquely determined by enforcing the boundary conditions @xmath109 at @xmath58 and @xmath110 at @xmath111 , respectively .
there are two important points to realize , namely : first , on the right - hand boundary , the function @xmath96 assumes a nonzero value , which is given by eq .
( [ bdr ] ) , for all possible @xmath63 values .
its asymptotic behavior relies on the function @xmath77 at large @xmath1 , which behaves like @xmath112 .
this indicates that it is _ nonzero _ generally , although it could be small at the large @xmath92 values used in practical calculations .
second , on the left - hand boundary , the situation depends on the value of @xmath63 .
@xmath104 takes a _
value if @xmath61 , while it becomes _ zero _ if @xmath62 .
following the philosophy employed to handle the spherical case @xcite , we first seek a solution , @xmath113 , to the poisson equations ( [ poissoneq])-([bdr ] ) that satisfies the _ zero_-value boundary condition at @xmath92 by using exactly the same @xmath1 mesh points as those for the wave functions . in other words , we have @xmath114 with @xmath115 and @xmath116 .
after substituting the dvr expansion @xmath117 of the solution into the differential equation , we obtain a system of linear equations for the unknown coefficients @xmath118 : @xmath119 the matrix @xmath120 is defined by its elements @xmath121f_{\mu'}(\xi).\end{aligned}\ ] ] therefore , the coefficient @xmath122 can formally be written as @xmath123^{-1}_{\mu i}}{\sqrt{\omega_\xi^i } } ( -1)^{|m|}\frac{(l\!+\!|m|)!}{(l\!-\!|m|)!}a^3\delta_{ii'}\delta_{kk ' } ( \xi_i^2-\eta_k^2)p _ { l } ^ { |m| } ( \eta_k ) , \label{cpc}\ ] ] where @xmath124^{-1}$ ] denotes the inverse of the matrix @xmath125 . in this case
@xmath113 fulfills the left - hand boundary condition @xmath126 , and so does @xmath127 .
recall , however , that its right - hand boundary condition differs from those of @xmath128 .
this suggests that the final answer to the function @xmath96 can be constructed as @xmath129 , i.e. , we add the difference function @xmath130 to @xmath113 .
the function @xmath130 is also a solution to the homogeneous poisson equation , subject to the boundary condition @xmath131 and @xmath132 .
after writing it as a linear combination of @xmath76 and @xmath77 , and imposing the boundary conditions , @xmath130 takes the form @xmath133 we finally arrive at @xmath134^{-1}_{\mu i } f_{\mu}(\xi ) + \delta_{ii'}\delta_{kk'}(\xi_i^2-\eta_k^2)p_{l}^{|m|}(\xi_i ) \\ \times & p_{l}^{|m|}(\eta_k ) \frac{q_{l}^{|m|}(\xi_{\rm max})}{p_{l}^{|m|}(\xi_{\rm max } ) } p_{l}^{|m|}(\xi ) .
\label{solutionu } \end{split}\end{aligned}\ ] ] at this point , the dvr version of the solution @xmath96 is ready for all possible values of @xmath63 , either @xmath62 or @xmath61 . substituting eq .
( [ solutionu ] ) into eq .
( [ intei ] ) allows us to obtain the kernel integral , @xmath135^{-1}_{ji}p_{l}^{|m|}(\eta_k)+ \\ + & p_{l}^{|m|}(\xi_i)p_{l}^{|m|}(\xi_j)p_{l}^{|m|}(\eta_k ) \frac{q_{l}^{|m|}(\xi_{\rm max})}{p_{l}^{|m|}(\xi_{\rm max})}\bigg ] .
\end{split}\end{aligned}\ ] ] the matrix element of @xmath80 can finally be written as @xmath136^{-1}_{ji}+(-1)^{|m| } \frac{(l-|m|)!}{(l+|m| ) ! }
p_{l}^{|m|}(\xi_i ) p_{l}^{|m|}(\xi_j ) \frac{q_{l}^{|m|}(\xi_{\rm max})}{p_{l}^{|m|}(\xi_{\rm max } ) } \bigg ] , \end{split } \label{eq:1/r12}\end{aligned}\ ] ] where we truncated the @xmath99 summation in the neumann expansion to @xmath137 .
the above equation can be converted to the normalized @xmath54 bases with the help of eq .
( [ eq : overall - basis ] ) .
this results in a diagonal representation of the matrix elements of the electron - electron coulomb interaction and thus considerably simplifies the fe - dvr discretization procedure .
the above treatment of the @xmath80 matrix was successfully applied to the two - photon double ionization of h@xmath4 @xcite
. the implementation of this representation will be illustrated below .
the time - dependent laser - driven electronic wave packet in the hydrogen molecule is obtained by solving the tdse on the @xmath54 grid .
launched from the previously determined ground state , the time evolution of the system is achieved by using our recently developed sil method @xcite .
the ground state is determined by relaxing the system in imaginary time from an initial guess of the wave function on the grid points . at each time step
we only need to generate the values of the discretized wave function on the selected grid points .
if desired , the information at arbitrary points within the spatial box can be obtained from the interpolation procedure in terms of the dvr bases .
a few remarks seem appropriate regarding the efficient implementation of the sil algorithm . the highest energy , @xmath138 , which essentially depends on the smallest separation between the @xmath54 mesh points and also on the maximum values of @xmath139 and @xmath140 , determines the largest time step @xmath141 for the propagation in real time . typically , @xmath138 is about @xmath142 atomic units ( a.u . ) in our calculations .
although the chances of electrons populating states with such high energies are practically negligible for short time scales of the laser - molecule interaction , we generally require @xmath143 in order to resolve the most rapid oscillations in the time evolution .
this means that at least a few time steps are needed during one period of @xmath144 .
we refer readers to refs .
@xcite for further details and discussions behind the sil method . in order to ensure that the double - ionization wave packet is sufficiently far away from the nuclei , and also that the two photoelectrons are well separated , we allow the system to evolve for a few more cycles in the field - free hamiltonian , i.e. , after the laser pulse has died off .
this is the wave packet we use to extract the physical information .
the ionization probabilities and the corresponding cross sections are extracted by projecting the time - dependent wave packet onto uncorrelated two - electron continuum states satisfying the standard incoming boundary conditions .
the latter states of h@xmath4 are constructed from the one - electron continuum state of the h@xmath13 ion described in the following subsection .
the field - free wave function @xmath145 of the one - electron molecular ion is completely separable in prolate spheroidal coordinates . for
a given , and conserved , magnetic quantum number @xmath64 , the wave function takes the form @xmath146 , where the azimuthal dependence , is given by @xmath147 .
the radial " part @xmath148 and the angular " part @xmath149 of the wave function satisfy the equations @xmath150t_{mq}(\xi)=0 \label{eq : xi}\ ] ] and @xmath151\xi_{mq}(\eta)=0 , \label{eq : eta}\ ] ] respectively . here
@xmath152 for the continuum state whose momentum vector has the magnitude @xmath153 .
in addition , we need to introduce another quantum number @xmath32 , which denotes the number of nodes of @xmath154 in the region @xmath155 $ ] , to label the states , and finally the separation constant @xmath156 .
when the angular function @xmath157 is discretized in terms of the relevant dvr bases , a few spurious " solutions might be encountered .
this is caused by the residual errors associated with the gauss quadratures .
consequently , we expand the angular function , or spheroidal harmonics " function @xmath158 with @xmath159 instead in terms of spherical harmonics .
these functions are normalized according to @xmath160 after obtaining the separation constant @xmath156 by solving eq .
( [ eq : eta ] ) , the radial " function @xmath148 is again expanded in terms of the dvr bases .
the last dvr point at @xmath111 needs to be kept for the continuum state .
asymptotically , the radial function behaves like @xmath161 \label{eq : asym}\ ] ] as @xmath162 . here
@xmath163 is the two - center coulomb phase shift .
the normalization factor on either the energy or the momentum scale and the coulomb phase shift can be determined by matching the numerical solution of @xmath148 according to its asymptotic behavior given in eq .
( [ eq : asym ] ) .
the plane wave in prolate spheroidal coordinates can be written as @xcite @xmath164 where @xmath165 $ ] in the asymptotic region . note that @xmath166 and @xmath167 are related to the directions of @xmath168 and @xmath169 in spherical coordinates through @xmath170 .
the partial - wave expansion of the plane wave @xmath171 reminds us that the two - center coulomb wave satisfying the incoming boundary condition can be expanded as @xmath172 this function is normalized in momentum space according to @xmath173 , provided the asymptotic solution in eq .
( [ eq : asym ] ) is satisfied .
uncorrelated two - electron continuum states with total spin angular momentum @xmath174 ( @xmath175 in our case ) can generally be constructed as follows : @xmath176 .
\label{eq : two - continuum}\end{aligned}\ ] ] with the help of eq .
( [ eq : con - one ] ) , its partial - wave representation can be written as @xmath177 . \label{eq : two - conti}\end{aligned}\ ] ] here we introduced @xmath178 by representing the radial and angular parts on the @xmath54 grid points : @xmath179 @xmath180 here the exchange symmetry @xmath181 is satisfied .
it has been demonstrated @xcite that using uncorrelated two - electron continuum states is a good approximation in a time - dependent propagation approach , provided the two ejected electrons are well separated from each other .
the probability amplitude of double ionization is then given by @xmath182 where @xmath183 here the exchange symmetry @xmath184 is satisfied .
we also see that the probability amplitude formulated in prolate spheroidal coordinates takes a similar form as for the atomic case in spherical coordinates .
however , a subtle difference from the atomic case is worth pointing out . strictly speaking
, the spheroidal harmonics involved in the probability amplitude generally depend on the magnitude of the momenta @xmath185 and @xmath186 , in addition to their directions .
the energy sharing of the two photoelectrons can be specified by introducing the hyperangle @xmath187 .
this describes the double - ionization reaction with kinetic energies @xmath188 and @xmath189 for the two ionized electrons , respectively . here
@xmath190 is the available excess energy above the double - ionization threshold . in the present work , specifically , @xmath191 ev for absorption of a photon . for double ionization by one - photon absorption ,
the triple differential cross section with respect to one of the kinetic energies and the two solid angles @xmath192 and @xmath193 can be extracted using the same formalism as in the corresponding he case @xcite , i.e. , @xmath194 here @xmath195 and @xmath196 are the central photon energy and the peak intensity of the laser pulse , respectively , while @xmath197 denotes the _ effective _ interaction time between the temporal electric laser field and the electrons in the one - photon absorption process . for a laser pulse of time duration @xmath198 with a sine - squared envelope for the field amplitude , @xmath199 .
note that @xmath200 corresponds to the special case of the generalized @xmath201-photon effective interaction time @xmath202 @xcite for a one - photon reaction .
generalized cross sections for two - photon double ionization of the hydrogen molecule were extracted in the same formalism @xcite . in the tdcc treatment
@xcite , different strategies were employed to describe the linear one - photon and the nonlinear two - photon double ionization processes of atoms and molecules .
for the one - photon case , the cross sections were obtained through the time derivative of the double - ionization probability , @xmath203 .
the laser field does not need to be turned off in this case . on the other hand ,
a true laser pulse was used for the two - photon case and an effective time , defined as the time integral under a flat - top pulse with a smooth turn - on and turn - off , was introduced . in the present work , we employed a unified formulation through an effective interaction time for both one - photon and multi - photon ionization in laser pulses .
for the one - photon double ionization initialized from the lowest @xmath204 state , the two ejected electrons can only populate the final @xmath205 and @xmath206 continuum states , with the specifics depending on the relative orientation of the molecular axis and the laser polarization vector . consequently only partial waves with _ ungerade _ parity [ i.e. , @xmath207 need to be included in eq .
( [ eq : tdcs ] ) .
for the nonsequential double - ionization process induced by one- or two - photon absorption , electronic correlation plays a dominant role , as the two photoelectrons must share the available excess energy @xmath190 .
double ionization by a single photon would not occur at all if the two - electron atom or molecule were approximated by an independent - electron model .
therefore , the quality of the description of electron - electron correlation in a laser - driven system is crucially important for accurate results to be obtained . the coulomb interaction between the two electrons has to be described in a consistent manner for both the initial bound state and the time - evolved wave packet .
before we go any further , it is worth discussing how we prepare the initial @xmath204 state at the equilibrium distance of @xmath208 bohr . as seen from eq .
( [ eq:1/r12 ] ) , the magnetic quantum number in the neumann expansion of the matrix element of @xmath80 is uniquely determined by the angular bases .
however , this is not the case for the index @xmath99 , if we choose to discretize the coordinate @xmath2 , rather then expanding that part of the wave function into spherical harmonics . in practice
, the summation over @xmath99 must be truncated at a finite value of @xmath137 . in principle
, the higher - order expansion terms always guarantee well - converged results . however , as mentioned earlier , we approximate the relevant @xmath2-integrals by using gauss - legendre quadrature . this is the price we have to pay for making the dielectronic coulomb potential diagonal in the dvr bases . as a consequence ,
we need to determine how the approximation introduced in the @xmath2-integrals for the two - electron integrals affects the results for the cross sections of interest . to answer this question
, we first investigate the dependence of the energy obtained for the initial @xmath209 state on the value of @xmath137 used in the neumann expansion .
figure [ fig : energy ] shows the variation of the initial - state electronic energy of the hydrogen molecule with respect to @xmath137 , obtained with a @xmath1 setup of ten elements in the region of @xmath210 , five in the region @xmath211 and another five in @xmath212 .
each element , in turn , contains five dvr points to further discretize the configuration space .
furthermore , we employ dvr points for @xmath2 . for a given number of @xmath2 mesh points ( @xmath213 ) and @xmath214
, we observe that the resulting energy typically exhibits a plateau - like behavior with increasing @xmath137 . for given @xmath213 ,
when @xmath137 is relatively small , the @xmath2-integral can be computed very accurately by using gauss quadrature . however , when @xmath137 is too large , the numerical errors introduced from the gauss quadrature cause the energy value to fluctuate .
this occurs when @xmath137 approaches @xmath215 and is shown by the grey stripes in fig .
[ fig : energy ] . in this region of @xmath137 an unphysically _ low _ energy can be produced . beyond that point ,
the calculated energy increases to the next plateau . ultimately , this is not too surprising , since any gauss quadrature is only reasonably accurate up to a limited polynomial order of the integrand .
consequently , if we want to keep more terms in the neumann expansion , we have to increase @xmath213 correspondingly .
this finding is further substantiated by the dependence of the energy found for @xmath216 and @xmath217 .
the plateaus are indeed extended to the correspondingly larger values of @xmath215 .
most importantly , the amplitude of the energy fluctuation is systematically reduced with increasing @xmath213 .
the error in the energy is lowered from @xmath218 to @xmath219 and finally @xmath220 a.u .
, when @xmath213 increases from @xmath221 to @xmath222 and then @xmath217 for @xmath223 .
we obtained the energy at @xmath208 bohr as @xmath224 a.u . for @xmath225 ,
@xmath226 , and @xmath223 , resulting in a double - ionization potential of @xmath227 ev . keeping the other parameters unchanged
, we obtained an energy of @xmath228 a.u . for @xmath216 .
the benchmark energy in the literature is @xmath229 a.u . at the same @xmath11 @xcite , after we take out the nucleus - nucleus interaction of @xmath230 a.u .
to summarize : unlike for other expansion parameters , it is important to be consistent in the size of the angular quadrature and the largest @xmath137 employed in the neumann expansion in practical calculations , if we discretize the coordinate @xmath2 .
however , this provides a way to examine a potential sensitivity of the physical observables of interest ( here the differential cross sections ) to the ground - state wave functions generated by varying @xmath137 and other parameters .
this will be further discussed below .
before we present our results for the cross sections , let us take a closer look at the survival probability @xmath231 of the aligned h@xmath4 molecule in its ground state @xmath232 .
this is shown in fig .
[ fig : surv ] . for homonuclear molecules , the independent alignment angle @xmath37 between the molecular axis and the polarization vector can be confined to the region from @xmath233 to @xmath234 . in the xuv regime , we observe that the hydrogen molecule shows a larger probability of being ionized or excited ( i.e. , a lower probability of staying in the initial state ) at the end of the pulse in an aligned geometry .
this indicates that the perpendicular component of the temporal electric field exerts more influence on the ionization process due to the larger dipole momentum .
interestingly , at the earlier stages of the time evolution ( e.g. , @xmath235
, when the ionized wave packet is driven back by the change in direction of the electric field , the tilted molecule has a larger probability of staying in its ground state .
this happens near the various minima in @xmath236 .
however , once the electric field has become sufficiently strong ( @xmath237 a.u . ) , the wave packet is driven out and spread into a larger space .
this leads to lower minima in @xmath236 for the tilted molecule .
when the wave packet is driven back to the nuclear region and therefore has a chance to recombine with the h@xmath13 ion , a maximum in @xmath236 appears .
not surprisingly , the parallel geometry always has the largest probability for this to happen .
although the wave packet can also be scattered for the untilted molecule in the plane perpendicular to the molecular axis , the probability is undoubtedly larger if the laser electric field is perpendicular to the molecular axis .
a similar behavior of h@xmath13 in xuv pulses was observed in ref .
@xcite . for most calculations performed in this study ,
we expose the hydrogen molecule to a laser pulse with a peak intensity of @xmath238 w/@xmath239 . looking at fig .
[ fig : surv ] we see that the depletion of the initial ground state can be safely neglected for our typical interaction times . even for @xmath240 , @xmath241 remains very close to unity .
the negligible depletion of the ground state suggests that the concept of cross sections is valid and applicable . on the other hand
, it also presents a numerical challenge to predict the cross sections accurately from a time - dependent treatment , due to the generally small ionization probability . at first glance
, a peak intensity of @xmath238 w/@xmath239 might seem very intense for most atomic and molecular targets . here
, however , we consider an xuv rather than an ir pulse .
for an xuv pulse with central photon energy of @xmath0 ev , such laser fields definitely fall into the weak - field " regime .
the ponderomotive energy in the xuv regime is much smaller than the photon energy of interest .
.[tab : parameters ] the discretization and expansion parameters of the h@xmath4 wave function in prolate spheroidal coordinates . here
@xmath242 stands for the border between the inner and outer regions in the @xmath1 coordinate , while @xmath92 is the size of the @xmath1 box .
in addition , @xmath243 denotes the number of @xmath1 mesh points in each element .
the numbers of @xmath1 elements in the inner and outer region are @xmath244 and @xmath245 , respectively .
these @xmath1 parameters produce the total number of @xmath1 mesh points @xmath246 .
the @xmath1 grid i and @xmath1 grid ii are used to examine the convergence of our results . [ cols="<,^,^,^,^,^,^",options="header " , ] in this work , we are mainly interested in the triple - differential cross section , since it reveals the fine details of possible energy sharings and preferred directions of the ejected electrons in the double - ionization process
. given the discrepancies between results from various calculations found in the literature , we carried out comprehensive convergence tests for our predictions of the tdcss .
these tests are essentially divided into two groups .
the first group concerns the laser parameters , while the second one deals with the discretization and expansion parameters .
an example of two different parameter sets for the @xmath1 grid is given in table [ tab : parameters ] and will be further discussed below . in order to obtain a good handle on the sensitivity of the results to the various parameters and the resulting level of convergence " , we try to only vary a single parameter while keeping all others fixed if possible . for the dependence on the laser parameters , we use the @xmath1 grid
i combined with @xmath247 . for the tests regarding the discretizations and expansions ,
the peak intensity of laser was fixed at @xmath238 w/@xmath239 and a time scale of `` @xmath248 '' optical cycles ( o.c . ) was used .
here `` @xmath248 '' refers to a laser pulse with a sine - squared envelope for the field amplitude , followed by a field - free propagation .
+ + + + + + figures [ fig : tdcs - laser ] and [ fig : tdcs - discret ] show the convergence pattern of our tdcs results for asymmetric energy sharing in the parallel geometry ( @xmath249 ) .
the energy sharing between electron @xmath250 ( observed at the fixed angle @xmath251 ) and electron @xmath252 ( observed at the variable angle @xmath253 ) is @xmath254 .
only the electron that takes away @xmath10 of the excess energy is recorded at fixed positions either parallel or perpendicular to the polarization axis . since the laser pulse is explicitly involved in our time - dependent treatment
, we first have to be sure that the extracted cross sections are essentially independent of the laser intensity and the time scales
. only then are the calculations of cross sections meaningful .
this also allows us to compare the physical information extracted from our time - dependent scenario to that obtained through conventional time - independent treatments , which are effectively equivalent to the weak - field approximation and infinitely " long interaction times . rather than computing the cross sections
, it would be more appropriate to consider ionization rates if the cross sections were found to be sensitive to the laser parameters . in fig .
[ fig : tdcs - laser ] , we display the dependences of our tdcs results upon the laser parameters .
note that the tdcss extracted from @xmath255 w/@xmath239 and @xmath256 w/@xmath239 at fixed time evolution of @xmath248 " cycles are nearly identical and agree with each other to better than the thickness of the line .
when we turn to the dependence of time scales at a fixed intensity of @xmath238 w/@xmath239 , we use the same pulse , but allow the system to freely evolve for a few additional cycles to extract the tdcs .
this corresponds to the time scale of @xmath257 " o.c .
also , we may increase the laser - molecule interaction time , but extract the tdcss at the same cycles of field - free time evolution after the pulse died off .
this gives the scenario of
@xmath258 " o.c .
since the total time durations are the same ( @xmath259 o.c . ) , they allow us to examine the extracted tdcss from different perspectives .
the increased interaction time yields a reduced bandwidth of the photon energy , while the longer field - free propagation ensures that the double - ionization wave packet is further away from the nuclear region @xcite .
the calculated tdcss indeed show a slight , though in our opinion acceptable , sensitivity to the time scales .
not surprisingly , the sensitivity is most visible for the smaller cross sections , when the two ejected electrons travel nearly parallel along the same direction ( c.f .
[ fig : tdcs - laser]@xmath260 ) .
+ + + having confidence in using the current sets of laser parameters , we now turn our attention to the scheme of spatial discretization ( @xmath243 , @xmath92 , @xmath213 ) and the convergence of the expansion ( @xmath137 , @xmath261 ) .
the results are displayed in fig .
[ fig : tdcs - discret ] . for the discretization parameters , we obtain well - converged tdcss by increasing @xmath262 from @xmath263 to @xmath264 , @xmath265 from @xmath221 to @xmath222 , and extending the spatial box of @xmath92 from @xmath266 to @xmath267 .
most importantly , however , we consider two sets of @xmath1 mesh points : @xmath1 grid i and @xmath1 grid ii ( see table [ tab : parameters ] ) .
the principal motivation was to see whether or not we can reproduce the much lower tdcs values ( by about @xmath10 compared to the one - center spherical results ) that were recently obtained in an ecs calculation in two - center elliptical coordinates by tao _
et al . _
@xcite .
we emphasize that these two grids in the radial " @xmath1 coordinate are completely different regarding both the distribution of the elements and the number of grid points per element . in the @xmath1 grid
i , we divide the @xmath1 space into two parts , an inner and an outer region with a border at @xmath268 .
we place a narrow span of elements in the inner region , and then wider elements in the outer region .
in contrast to that , @xmath1 grid ii does not distinguish between inner and outer regions , i.e. , the elements uniformly span the region from @xmath250 to @xmath92 .
the mesh setup in @xmath1 grid ii is the same as that used in ref .
@xcite , except for the complex rotation .
the @xmath1 grid i has a much denser distribution of mesh points than @xmath1 grid ii .
nevertheless , the extracted tdcss from both sets of @xmath1 grids are in excellent agreement with each other , even for the smallest cross sections .
this strongly suggests that the results are well converged at least with regard to the @xmath1 grid .
both @xmath1 sets are good enough to capture the physics of interest .
differences at the @xmath10 level are unlikely to be caused by using different sets of @xmath1 meshes . finally , we discuss the convergence of our results with respect to the expansion parameters , @xmath261 and @xmath137 .
as expected for a one - photon process , @xmath223 produced well - converged results .
recall the discussion above regarding the ground state , especially how the truncated neumann expansion of @xmath80 in our present fe - dvr implementation affects the initial - state energy and therefore the quality of the wave function . for consistency
, we use the same @xmath137 in the real - time propagation and in the ground - state wave function . as seen from figs .
[ fig : tdcs - discret]@xmath269 and [ fig : tdcs - discret]@xmath270 , our truncated neumann expansion has little effect on the calculated tdcs values .
well - converged results can be obtained even with an inappropriately large value of @xmath271 , which yields a slightly higher energy of the ground state ( c.f .
[ fig : energy ] ) .
overall , our detailed convergence tests only reveal a very weak sensitivity of the tdcs results to both the time scales and the values of @xmath137 .
well - converged tdcs results can be obtained by using either @xmath1 grid i or @xmath1 grid ii combined with @xmath247 . in the production calculations for the tdcss shown in the next subsection
, we used the @xmath1 grid i to discretize the two - electron wave packet and a
@xmath248 " sine - squared laser pulse with a peak intensity of @xmath238 w/@xmath239 .
figures [ fig : tdcs - equal - sharing - ele1 - 00 ] , [ fig : tdcs - equal - sharing - ele1 - 90 ] , and [ fig : tdcs - nonequal - sharing - ele1 - 90 ] display the coplanar tdcss of the aligned hydrogen molecule at equal and asymmetric ( @xmath272 ) energy sharing .
the two electrons are detected in the same ( coplanar ) plane defined by the @xmath7 and @xmath8 axes .
the angles @xmath251 , @xmath253 , and @xmath37 are all measured with respect to the laser linear polarization axis .
we compare our tdcs predictions with those obtained in the time - independent one - center spherical ecs calculation @xcite , the time - independent two - center spheroidal ecs model @xcite , and the time - dependent one - center spherical tdcc approach @xcite .
the tdcc numbers were recently recalculated with a bigger box size and differ , in some cases substantially , from those published originally @xcite . except for the recent two - center prolate spheroidal ecs results of tao _ et al . _
@xcite , the agreement between the other three sets of results is very satisfactory .
once again , the largest relative differences occur when the cross sections are small ( see figs .
[ fig : tdcs - equal - sharing - ele1 - 00]@xmath273 and [ fig : tdcs - equal - sharing - ele1 - 90]@xmath273 ) . using spheroidal coordinates
as well , as an illustrative example of their two - center ecs approach , serov and joulakian @xcite recently presented the tdcs at the same photon energy , but only for a single geometry of and for asymmetric energy sharing of @xmath274 .
although not shown here , there is again good agreement between their results , vanroose _ et al . _
s one - center spherical ecs numbers @xcite , and our time - dependent fe - dvr predictions .
it is also interesting to investigate the dominant escape modes for the various scenarios .
these modes are strongly dependent on how the electrons share the excess energy . in an arbitrary geometry ( @xmath275 ) , for example , the back - to - back escape mode @xmath276 is forbidden for equal energy sharing . on the other hand ,
it becomes the dominant mode for significantly asymmetric energy sharing , including the 20%:80% scenario discussed in the present paper ( see fig .
[ fig : tdcs - laser ] ) . these results can be understood from a symmetry analysis @xcite .
equal - energy sharing and back - to - back emission is equivalent to @xmath277 . when we consider the exchange and parity operations simultaneously in eq .
( [ eq : two - continuum ] ) , we have @xmath278 . here
@xmath279 is the parity for the gerade and ungerade states , respectively . for the singlet double - continuum state with _
parity , we therefore must have at any configuration of @xmath280 and @xmath281 .
although the magnitudes of the momenta @xmath282 and @xmath283 are not exactly conserved in the time - dependent picture , the ionization events we collect must satisfy the condition @xmath284 ( because of the @xmath285 function in eq .
( [ eq : tdcs ] ) ) for the equal - energy sharing .
this is the reason behind the forbidden back - to - back ( @xmath286 ) escape mode for the equal energy sharing , as we observed in figs .
[ fig : tdcs - equal - sharing - ele1 - 00 ] and [ fig : tdcs - equal - sharing - ele1 - 90 ] .
since the argument does not involve the relative alignment angles , it is valid for all possible values of @xmath37 . on the other hand , this is not the case when the excess energy is not evenly distributed among the two electrons .
indeed , figs .
[ fig : tdcs - laser]@xmath287 and [ fig : tdcs - laser]@xmath260 show maxima in the back - to - back emission , thereby illustrating the dramatic change in the dominant escape mode . for equal - energy sharing in the parallel geometry ( @xmath249 )
, the electron - electron coulomb repulsion suggests that the tdcs should be dynamically small if the two electrons travel along the same direction .
this is in agreement with the numerically small cross sections ( not exactly zero , however ) at @xmath288 or @xmath289 seen in fig .
[ fig : tdcs - equal - sharing - ele1 - 00]@xmath273 .
recall that the one - photon double - photoionization process in helium @xcite shares the same property . the back - to - back mode is forbidden for equal - energy sharing , and this can be explained by the above argument .
it is one of the similarities between the molecular hydrogen and the atomic helium targets for double photoionization .
however , figs .
[ fig : tdcs - equal - sharing - ele1 - 00 ] , [ fig : tdcs - equal - sharing - ele1 - 90 ] , and [ fig : tdcs - nonequal - sharing - ele1 - 90 ] also reveal significant molecular effects in the tdcs results .
these are missing for the helium atom , not only in the shape of the angular distributions , but also in the magnitudes of the cross sections .
depending on the relative orientation @xmath290 , there is interference between the @xmath291 and @xmath292 symmetries in h@xmath4 .
a nice example of this effect was presented by reddish _
et al . _
@xcite . even without interference (
i.e. , for @xmath249 or @xmath234 ) , the perpendicular geometry shows much larger magnitudes of the tdcs than the parallel geometry .
figure [ fig : tdcs - h2-helium ] shows the three cases of angular distributions : h@xmath4 @xmath293 , h@xmath4 @xmath294 , and he at equal energy sharing . interestingly , in most cases the angular distributions of the perpendicular geometry resemble those of helium .
the molecular effect can definitely not be ignored in the parallel geometry for @xmath295 ( c.f .
[ fig : tdcs - h2-helium]@xmath287 ) .
the _ forward _ escape mode of the second electron is dominant for the h@xmath4 parallel geometry .
in contrast , the _ backward _ mode is dominant for the h@xmath4 perpendicular geometry and also for helium . in fig .
[ fig : tdcs - out - plane ] , we show the tdcs for noncoplanar geometries .
again , all angles are defined with respect to the polarization vector . for the perpendicular geometry ,
[ fig : tdcs - out - plane]@xmath287 depicts the escape modes for the configuration of @xmath296 ( the fixed electron ) and at the same time @xmath297 in the plane perpendicular to the plane formed by @xmath8 and @xmath7 .
figure [ fig : tdcs - out - plane]@xmath298 shows the tdcs after exchanging the directions of @xmath299 and @xmath297 in fig . [ fig : tdcs - out - plane]@xmath287 . with the same directions of @xmath299 and @xmath297 as in fig .
[ fig : tdcs - out - plane]@xmath298 , fig .
[ fig : tdcs - out - plane]@xmath260 is for the case of the molecular axis orientated along the polarization vector . in the parallel case ( @xmath249 ) , we observe that any escape modes of both electrons ejected in the direction perpendicular to @xmath8 are forbidden .
this can be understood by analyzing the spheroidal harmonics in eq .
( [ eq : tdcs ] ) . in this case , only the @xmath205 states can be populated .
hence , only partial waves with @xmath300 and @xmath301 can contribute to the cross sections , since @xmath302 at the angles of @xmath303 vanish in spherical coordinates .
once again , the agreement between our fe - dvr noncoplanar tdcss and the refined tdcc results @xcite is excellent .
we have presented calculations for one - photon double ionization of the hydrogen molecule at a photon energy of @xmath0 ev by solving the time - dependent schrdinger equation in prolate spheroidal coordinates .
the triple - differential cross sections were extracted through the projection of the time - dependent wave packet onto uncorrelated two - electron continuum states , a few cycles of field - free time evolution after the laser pulse died off .
exhaustive convergence studies of the tdcs results were performed with respect to a number of discretization and expansion parameters , as well as the details of the laser field .
these tests provide a strong indication that the results for the triple - differential cross sections presented here are well converged and numerically accurate .
excellent agreement was obtained between the current time - dependent results in prolate spheroidal coordinates , those obtained with the ecs approach in spherical coordinates @xcite and , finally , larger tdcc calculations @xcite than those published earlier @xcite .
the present calculations do not confirm the significant reduction by about @xmath10 in the tdcs results predicted in recent ecs calculations in the two - center prolate spheroidal coordinates @xcite .
furthermore , our results did not show the level of sensitivity to the description of the ground state that was also reported by tao _
et al . _
@xcite .
the detailed analysis reported in this study provides a high level of confidence in the present results .
we hope that they will be used as benchmarks for comparison in future investigations .
tables of these results are available in electronic format from the authors upon request .
we thank drs . t. n. rescigno and j. colgan for sending their results in numerical form and for helpful discussions .
this work was supported by the nsf under grant phy-0757755 ( xg and kb ) and generous supercomputer resources through the nsf teragrid allocation award tg - phy@xmath304 ( kraken at nics , oak ridge national laboratory ) and by the department of energy allocation award mph@xmath305 ( jaguar at nccs , oak ridge national laboratory ) . without these computational resources
it would have been impossible to study the convergence properties of our cross section results on various physical and computational parameters .
99 h. bruning , r. drner , c. l. cocke , m. h. prior , b. krssig , a. s. kheifets , i. bray , a. bruning - demian , k. carnes , s. dreuil , v. mergel , p. richard , j. ullrich , and h. schmidt - bcking , j. phys .
b * 31 * , 5149 ( 1998 ) .
weber , a. czasch , o. jagutzki , a. mller , v. mergel , a. kheifets , j. feagin , e. rotenberg , g. meigs , m. h. prior , s. daveau , a. l. landers , c. l. cocke , t. osipov , h. schmidt - bcking , and r. drner , phys .
. lett . * 92 * 163001 ( 2004 ) .
t. weber , a. o. czasch , o. jagutzki , a. k. mller , v. mergel , a. kheifets , e. rotenberg , g. meigs , m. h. prior , s. daveau , a. landers , c. l. cocke , t. osipov , r. dez muio , h. schmidt - bcking , and r. drner , nature ( london ) * 431 * , 437 ( 2004 ) .
weber , phd .
thesis , universitt frankfurt ( unpublished , 2003 ) .
m. gisselbrecht , m. lavolle , a. huetz , p. bolognesi , l. avaldi , d. p. seccombe , and t. j. reddish , phys .
lett . * 96 * , 153002 ( 2006 ) .
y. h. jiang , a. rudenko , e. plsiat , l. foucar , m. kurka , k. u. khnel , th .
ergler , j. f. prez - torres , f. martn , o. herrwerth , m. lezius , m. f. kling , j. titze , t. jahnke , r. drner , j. l. sanz - vicario , m. schffler , j. van tilborg , a. belkacem , k. ueda , t. j. m. zouros , s. dsterer , r. treusch , c. d. schrter , r. moshammer , and j. ullrich , phys . rev .
a * 81 * , 021401(r ) ( 2010 ) .
j. p. wightman , s. cvejanovi , and t. j. reddish , j. phys .
b * 31 * , 1753 ( 1998 )
. a. s. kheifets , phys .
a * 71 * , 022704 ( 2005 ) .
a. s. kheifets and i. bray , phys .
a * 72 * , 022703 ( 2005 ) .
w. vanroose , f. martn , t. n. rescigno , and c. w. mccurdy , phys .
a * 70 * , 050703(r ) ( 2004 ) .
w. vanroose , d. a. horner , f. martn , t. n. rescigno , and c. w. mccurdy , phys .
a * 74 * , 052702 ( 2006 ) .
w. vanroose , f. martn , t. n. rescigno , and c. w. mccurdy , science * 310 * , 1787 ( 2006 ) . j. colgan , m. s. pindzola , and f. robicheaux , phys . rev . lett . * 98 * , 153001 ( 2007 ) . j. colgan , private communication ( 2010 ) . l. tao , c. w. mccurdy , and t. n. rescigno , phys . rev .
a * 82 * , 023423 ( 2010 ) .
d. r. bates , u. pik , and g. poots , proc . phys .
a * 66 * , 1113 ( 1953 ) .
s. barmaki , s. laulan , h. bachau , and m. ghalim , j. phys .
b * 36 * , 817 ( 2003 ) .
s. barmaki , h. bachau , and m. ghalim , phys .
a * 69 * , 043403 ( 2004 ) .
y. v. vanne and a. saenz , j. phys .
b * 37 * , 4101 ( 2004 ) . g. lagmago kamta and a. d. bandrauk , phys . rev . a * 71 * , 053407 ( 2005 ) .
l. tao , c. w. mccurdy , and t. n. rescigno , phys .
a * 79 * , 012719 ( 2009 ) .
t. j. park and j. c. light , j. chem .
* 85 * , 5870 ( 1986 ) .
d. j. tannor , in _ introduction to quantum mechanics , a time - dependent perspective _ , chap . 11 , p318 ( university science books , sausalito , california 2007 ) .
x. guan , k. bartschat , and b. i. schneider , phys .
a * 82 * , 041404(r ) ( 2010 ) .
x. guan , b. li , and k. t. taylor , j. phys .
b * 36 * , 3569 ( 2003 ) .
p. m. morse and h. feshbach , in _ methods of theoretical physics _ ,
parts i and ii ( mcgraw - hill , 1953 ) . i. a. stegun , in _ handbook of mathematical functions with formulas , graphs , mathematical tables _ , chap . 8 , p331 , eds .
m. abramowitz and i. a. stegun ( 1964 ) .
e. l. mehler and k. ruedenberg , j. chem . phys . *
50 * , 2575 ( 1969 ) . c. w. mccurdy , m. baertschy , and t. n. rescigno , j. phys .
b * 37 * , r137 ( 2004 ) .
x. guan , k. bartschat , and b. i. schneider , phys .
a * 77 * , 043421 ( 2008 ) .
x. guan , c. j. noble , o. zatsarinny , k. bartschat , and b. i. schneider , comp .
comm . * 180 * , 2401 ( 2009 ) .
k. t. taylor , j. s. parker , d. dundas , k. j. meharg , l. r. moore , e. s. smyth , and j. f. mccann , in _ many - particle quantum dynamics in atomic and molecular fragmentation _
9 , p153 , eds .
j. ullrich and v. shevelko ( springer - verlag , berlin heidelberg 2003 ) . c. flammer , in _ spheroidal wave functions _ , ( dover publications , inc . ,
mineola , new york 2005 ) .
j. colgan , m. s. pindzola , and robicheaux , j. phys .
b * 34 * , l457 ( 2001 ) . l. a. a. nikolopoulos and p. lambropoulos , phys . rev .
a * 74 * , 063410 ( 2006 ) .
j. sims and s. hagstrom , j. chem .
phys . * 124 * , 094101 ( 2006 ) .
s. x. hu , l. a. collins , and b. i. schneider , phys . rev .
a * 80 * , 023426 ( 2009 ) .
l. b. madsen , l. a. a. nikolopoulos , t. k. kjeldsen , and j. fernndez , phys .
a * 76 * , 063407 ( 2007 ) .
t. n. rescigno , private communication ( 2010 ) .
v. v. serov and b. b. joulakian , phys .
a * 80 * , 062713 ( 2009 ) .
f. maulbetsch and j. s. briggs , j. phys .
b * 28 * , 551 ( 1995 ) .
x. guan , unpublished ( 2008 ) .
t. j. reddish , j. colgan , p. bolognesi , l. avaldi , m. gisselbrecht , m. lavolle , m. s. pindzola , and a. huetz , phys .
lett . * 100 * , 193001 ( 2008 ) . | we have carried out calculations of the triple - differential cross section for one - photon double ionization of molecular hydrogen for a central photon energy of @xmath0 ev , using a fully _ ab initio _
, nonperturbative approach to solve the time - dependent schrdinger equation in prolate spheroidal coordinates .
the spatial coordinates @xmath1 and @xmath2 are discretized in a finite - element discrete - variable representation .
the wave packet of the laser - driven two - electron system is propagated in time through an effective short iterative lanczos method to simulate the double ionization of the hydrogen molecule . for both symmetric and asymmetric energy sharing
, the present results agree to a satisfactory level with most earlier predictions for the absolute magnitude and the shape of the angular distributions .
a notable exception , however , concerns the predictions of the recent time - independent calculations based on the exterior complex scaling method in prolate spheroidal coordinates [ phys .
rev .
a * 82 * , 023423 ( 2010 ) ] .
extensive tests of the numerical implementation were performed , including the effect of truncating the neumann expansion for the dielectronic interaction on the description of the initial bound state and the predicted cross sections .
we observe that the dominant escape mode of the two photoelectrons dramatically depends upon the energy sharing . in the parallel geometry ,
when the ejected electrons are collected along the direction of the laser polarization axis , back - to - back escape is the dominant channel for strongly asymmetric energy sharing , while it is completely forbidden if the two electrons share the excess energy equally . |
volatility clustering , evaluated through slowly decaying auto - correlations , hurst effect or @xmath0 noise for absolute returns , is a characteristic property of most financial assets return time series @xcite .
statistical analysis alone is not able to provide a definite answer for the presence or absence of long range dependence phenomenon in stock returns or volatility , unless economic mechanisms are proposed to understand the origin of such phenomena @xcite .
whether results of statistical analysis correspond to long range dependence is a difficult question subject to an ongoing statistical debate @xcite .
the agent based economic models @xcite as well as stochastic models @xcite exhibiting long range dependence phenomenon in volatility or trading volume are of grate interest and remain an active topic of research .
the properties of stochastic multiplicative point processes have been investigated analytically and numerically and the formula for the power spectrum has been derived @xcite , later the model has been related with the general form of the multiplicative stochastic differential equation @xcite .
the extensive empirical analysis of the financial market data , supporting the idea that the long - range volatility correlations arise from trading activity , provides valuable background for further development of the long - ranged memory stochastic models @xcite .
the power law behaviour of the auto - regressive conditional duration process @xcite based on the random multiplicative process and its special case the self - modulation process @xcite , exhibiting @xmath1 fluctuations , supported the idea of stochastic modelling with a power law probability density function ( pdf ) and long memory .
a stochastic model of trading activity based on an stochastic differential equation ( sde ) driven poisson - like process has been already presented in @xcite . in the paper @xcite
we proposed a double stochastic model , which generates time series of the return with two power law statistics , i.e. , the pdf and the power spectral density of absolute return , reproducing the empirical data for the one - minute trading return in the nyse . in this contribution
we analyze empirical data from vilnius stock exchange ( vse ) in comparison with nyse and stochastic model proposed in @xcite . at the same time
we demonstrate the scaling of statistical properties with longer time window of return .
recently we proposed the double stochastic model of return in financial market @xcite based on the nonlinear sde .
the main advantage of proposed model is its ability to reproduce power spectral density of absolute return as well as long term pdf of return . in the model
proposed we assume that the empirical return @xmath2 can be written as instantaneous @xmath3-gaussian fluctuations @xmath4 with a slowly diffusing parameter @xmath5 and constant @xmath6@xmath7 q - gaussian distribution of can be written as follows:@xmath8 the parameter @xmath5 serves as a measure of instantaneous volatility of return fluctuations .
see @xcite , for the empirical evidence of this assumption . here
@xmath9 is defined in the selected time interval @xmath10 as a difference of logarithms of asset prices @xmath11:@xmath12 - \ln[p(t ) ] \right| .\ ] ] in this paper we consider dimensionless returns normalized by its dispersion calculated in the whole length of realization .
it is worth to notice that @xmath13 is an additive variable , i.e. , if @xmath14 , then @xmath15 , or in the continuous limit the sum may be replaced by integration .
we do propose to model the measure of volatility @xmath5 by the scaled continuous stochastic variable @xmath16 , having a meaning of average return per unit time interval . by the empirical analyses of high frequency trading data on nyse @xcite we introduced relation:@xmath17 where @xmath18 is an empirical parameter and the average return per unit time interval @xmath19
can be modeled by the nonlinear sde , written in a scaled dimensionless time @xmath20:@xmath21 \frac{\left ( 1+x^2 \right)^{\eta-1}}{(\epsilon \sqrt{1+x^2 } + 1)^2 } x \rmd t_s + \frac{\left ( 1+x^2 \right)^{\frac{\eta}{2}}}{\epsilon \sqrt{1+x^2 } + 1 } \rmd w_s.\ ] ] here are five more empirically defined parameters : @xmath22 - exponent of multiplicativity , @xmath23 - power law exponent of @xmath16 long range pdf , @xmath24 - parameter dividing diffusion into two areas : stationary and excited one , @xmath25 - time scale adjustment parameter and @xmath26 - the upper limit of diffusion .
the term @xmath27 excludes divergence of @xmath16 to the infinity .
seeking to discover the universal nature of financial markets we consider that all these parameters are universal for all stocks traded on various exchanges . in this paper
we analyze empirical data from very different exchanges new york , one of the most developed with highly liquid stocks , and vilnius , emerging one with stocks traded rarely .
we solve numerically introducing variable steps of dimensionless time @xmath28:@xmath29 where @xmath30 is precision parameter of numerical calculations , which should be less than 1 .
then sde , , can be replaced by iterative equation:@xmath31 x_k + \kappa \sqrt{1+x_k^2 } \zeta_k , \ ] ] where @xmath32 is a normally distributed random variable with zero mean and unit variance .
in paper @xcite we analyzed the tick by tick trades of 24 stocks , abt , adm , bmy , c , cvx , dow , fnm , ge , gm , hd , ibm , jnj , jpm , ko , lly , mmm , mo , mot , mrk , sle , pfe , t , wmt , xom , traded on the nyse for 27 months from january , 2005 , recorded in the trades and quotes database .
the parameters of stochastic model presented in were adjusted to the empirical tick by tick one minute returns .
an excellent agreement between empirical and model pdf and power spectrum was achieved , see fig . 3 in @xcite .
the same empirical data and model results with slightly changed values of parameters are given in ( a , b ) .
noticeable difference in theoretical and empirical pdfs for small values of return @xmath9 are related with the prevailing prices of trades expressed in integer values of cents .
we do not account for this discreteness in our continuous description . in the empirical power spectrum one - day resonance
- the largest spike with higher harmonics - is present . this seasonality - an intraday activity pattern of the signal - is not included in the model either and this leads to the explicable difference from observed power spectrum .
provided that we use scaled dimensionless equations derived while making very general assumptions , we expect that proposed model should work for various assets traded on different exchanges as well as for various time scales @xmath10 .
we analyze tick by tick trades of 4 stocks , apg1l , ptr1l , srs1l , ukb1l , traded on vse for 50 months since may , 2005 , trading data was collected and provided for us by vse .
stocks traded on vse in comparison with nyse are less liquid mean inter - trade time for analyzed stocks traded on vse is 362 s , while for stocks traded on nyse mean inter - trade time equals 3.02 s. the difference in trading activity exceeds 100 times .
this great difference is related with comparatively small number of traders and comparatively small companies participating in the emerging vse market .
do these different markets have any statistical affinity is an essential question from the theoretical point of market modeling .
first of all we start with returns for very small time scales @xmath33 . for the vse up to 95% of one minute trading time intervals
elapse without any trade or price change .
one can exclude these time intervals from the sequence calculating pdf of return . with
such simple procedure calculated pdf of vse empirical return overlaps with pdf of nyse empirical return ( see ( a ) ) . ; @xmath34 ; @xmath35 ; @xmath36 ; @xmath37 ; @xmath38 ; @xmath39 .
pdf of normalized absolute returns is given on ( a),(c),(e ) and psd on ( b),(d),(f ) .
( a ) and ( b ) represents results with @xmath33 ; ( c ) and ( d ) @xmath40 ; ( e ) and ( f ) @xmath41 .
empirical data from nyse is averaged over 24 stocks and empirical data from vse is averaged over 4 stocks.,title="fig : " ] ; @xmath34 ; @xmath35 ; @xmath36 ; @xmath37 ; @xmath38 ; @xmath39 .
pdf of normalized absolute returns is given on ( a),(c),(e ) and psd on ( b),(d),(f ) .
( a ) and ( b ) represents results with @xmath33 ; ( c ) and ( d ) @xmath40 ; ( e ) and ( f ) @xmath41 .
empirical data from nyse is averaged over 24 stocks and empirical data from vse is averaged over 4 stocks.,title="fig : " ] + ; @xmath34 ; @xmath35 ; @xmath36 ; @xmath37 ; @xmath38 ; @xmath39 . pdf of normalized absolute returns is given on ( a),(c),(e ) and psd on ( b),(d),(f ) .
( a ) and ( b ) represents results with @xmath33 ; ( c ) and ( d ) @xmath40 ; ( e ) and ( f ) @xmath41 .
empirical data from nyse is averaged over 24 stocks and empirical data from vse is averaged over 4 stocks.,title="fig : " ] ; @xmath34 ; @xmath35 ; @xmath36 ; @xmath37 ; @xmath38 ; @xmath39 .
pdf of normalized absolute returns is given on ( a),(c),(e ) and psd on ( b),(d),(f ) .
( a ) and ( b ) represents results with @xmath33 ; ( c ) and ( d ) @xmath40 ; ( e ) and ( f ) @xmath41 .
empirical data from nyse is averaged over 24 stocks and empirical data from vse is averaged over 4 stocks.,title="fig : " ] + ; @xmath34 ; @xmath35 ; @xmath36 ; @xmath37 ; @xmath38 ; @xmath39 .
pdf of normalized absolute returns is given on ( a),(c),(e ) and psd on ( b),(d),(f ) .
( a ) and ( b ) represents results with @xmath33 ; ( c ) and ( d ) @xmath40 ; ( e ) and ( f ) @xmath41 .
empirical data from nyse is averaged over 24 stocks and empirical data from vse is averaged over 4 stocks.,title="fig : " ] ; @xmath34 ; @xmath35 ; @xmath36 ; @xmath37 ; @xmath38 ; @xmath39 .
pdf of normalized absolute returns is given on ( a),(c),(e ) and psd on ( b),(d),(f ) .
( a ) and ( b ) represents results with @xmath33 ; ( c ) and ( d ) @xmath40 ; ( e ) and ( f ) @xmath41 .
empirical data from nyse is averaged over 24 stocks and empirical data from vse is averaged over 4 stocks.,title="fig : " ] one should use full time sequence of returns calculating the power spectrum . nevertheless , despite low vse liquidity , psd of vse and nyse absolute returns almost overlap .
difference is clearly seen only for higher frequencies , when @xmath33 , and is related with low vse market liquidity contributing to the white noise appearance .
the different length of trading sessions in financial markets causes different positions of resonant spikes . one can conclude that even so marginal market as vse retains essential statistical features as developed market on nyse . at the first glance
the statistical similarity should be even better for the higher values of return time scale @xmath10 .
further we investigate the behavior of returns on nyse and vse for increased values of @xmath40 and @xmath41 with the specific interest to check whether proposed stochastic model scales in the same way as empirical data . apparently , as we can see in ( d ) and ( f ) psds of absolute returns on vse and on nyse overlap even better at larger time scale ( @xmath42 seconds and @xmath43 seconds ) .
this serves as an additional argument for the very general origin of long range memory properties observed in very different , liquidity - wise , markets .
the nonlinear sde is an applicable model to cache up observed empirical properties .
pdfs of absolute return observed in both markets ( see ( c ) and ( e ) ) are practically identical , though we still have to ignore zero returns of vse to arrive to the same normalization of pdf .
we proposed a double stochastic process driven by the nonlinear scaled sde reproducing the main statistical properties of the absolute return , observed in the financial markets .
seven parameters of the model enable us to adjust it to the sophisticated power law statistics of various stocks including long range behaviour .
the scaled no dimensional form of equations gives an opportunity to deal with averaged statistics of various stocks and compare behaviour of different markets .
all parameters introduced are recoverable from the empirical data and are responsible for the specific statistical features of real markets . seeking to discover the universal nature of return statistics we analyse and compare extremely different markets in new york and vilnius and
adjust the model parameters to match statistics of both markets .
the most promising result of this research is discovered increasing coincidence of the model with empirical data from the new york and vilnius markets and between markets , when the time scale of return @xmath10 is growing .
observable specific features of different markets could be a subject of another research based on the proposed model .
for example , it is clear that parameter @xmath26 should be relevant to the maximum number of active traders in the market and consequently should be specific for the every market .
further analyses of empirical data and proposed model reasoning by agent behavior is ongoing . 99 willinger , w. taqqu , m. and teverovsky , v. 1999 .
_ stock market prices and long - range dependence_. finance stochast . 3 : 113 cont , r. _ long range dependence in financial markets_. springer - fractals in engineering , ( e. lutton and j. vehel , eds . ) , ( 2005 ) 159 - 180 mikosch , t. and starica , c. _ long - range dependence effects and arch modeling _ , birkhauser boston - theory and applications of long - range dependence , ( 2003 ) 439459 kirman , a. and teyssiere , g. 2002 .
_ microeconomic models for long - memory in the volatility of financial time series _ , studies in nonlinear dynamics and econometrics , 5 : 281302 lux , t. and marchesi , m. 2000 . _ volatility clustering in financial markets : a microsimulation of interacting agents _ , int .
finance , 3 : 675702 6 .
borland , l. 2004 .
_ on a multi - timescale statistical feedback model for volatility fluctuations _ , arxiv : cond - mat/0412526 duarte queiros , s.m
_ on a generalised model for time - dependent variance with long - term memory_. epl , 80 : 30005 gontis , v. kaulakys b. and ruseckas , j. 2008 .
_ trading activity as driven poisson process : comparison with empirical data_. j. physica a 387 : 3891 - 3896 gontis , v. ruseckas j. and kononoviius a. 2010 .
_ a long - range memory stochastic model of the return in financial markets_. j. physica a 389 : 100 - 106 gontis v. and kaulakys , b. 2004 .
_ multiplicative point process as a model of trading activity_. physica a 343 : 505 - 514 kaulakys , b. ruseckas , j. gontis , v. and alaburda , m. 2006 .
_ nonlinear stochastic models of @xmath1 noise and power - law distributions_. physica a 365 : 217 - 221 .
kaulakys , b. and alaburda , m. 2009 .
_ modeling scaled processes and @xmath0 noise using nonlinear stochastic differential equations_. j. stat .
plerou , v. gopikrishnan , p. gabaix , x. et al . 2001 . _ price fluctuations , market activity , and trading volume_. quant .
finance 1 : 262 - 269 .
gabaix , x. gopikrishnan , p. plerou , v. and stanley , h.e .
_ a theory of power - law distributions in financial market fluctuations_. nature 423 : 267 - 270 sato , a .- h
_ explanation of power law behavior of autoregressive conditional duration processes based on the random multiplicative process_. phys .
e 69 : 047101 .
takayasu , m. takayasu , h. 2003 . _ self - modulation processes and resulting generic @xmath1 fluctuations_. phys . a 324 : 101 | we scale and analyze the empirical data of return from new york and vilnius stock exchanges matching it to the same nonlinear double stochastic model of return in financial market . |
since the critical point symmetries @xcite of the nuclear shape phase transitions were proposed , many experimental and theoretical efforts were made to find the nuclei described by the new symmetries . while at the beginning the x(5 )
@xcite candidates were found in the mass region of @xmath3 @xcite , recently a new region has been suggested for os and pt isotopes with @xmath4 @xcite . in refs .
@xcite data for the isotopes @xmath5os were analyzed with the sextic and spheroidal approach ( ssa ) @xcite , the davidson and spheroidal approach ( dsa ) @xcite , the infinite square well and spheroidal approach ( iswsa ) @xcite and the results were compared with those of the coherent state model ( csm ) @xcite and x(5 ) . according to our analysis these isotopes present features for the @xmath6 shape phase transition with the critical point reached for @xmath7os and @xmath8os .
on the other hand , applying the sextic and mathieu approach ( sma ) @xcite to @xmath9os , one points out that the isotope @xmath10os is a good candidate for the critical point of the phase transition between the prolate and the oblate shapes through the triaxial shape corresponding to @xmath11 . encouraged by the results for the os isotopes , we consider the above mentioned models also for the even - even @xmath2pt isotopes .
we aim not only at determining the energy spectra and the electric transition probabilities but also at showing the static deformation of each isotope in both the ground and excited states .
new features like the shape coexistence or a transition from the prolate to oblate shapes through a triaxial deformation are expected to show up . keeping in mind that the sma , the iswma and the z(5 ) @xcite are suitable for the description of the triaxial nuclei lying close to @xmath11 , a comparison of their predictions represent a challenging task .
iswma is the inedited model proposed of this paper .
recently , in ref . @xcite it was shown that the isotope @xmath12pt has some of the x(5 ) features . according to the interacting boson model-1 ( ibm-1 ) @xcite and the general collective model @xcite , this isotope manifests shape coexistence and it is close to the critical point of the @xmath6 shape phase transition . evidences for shape coexistence
were also presented for @xmath13pt @xcite , @xmath14pt @xcite , @xmath15pt @xcite and @xmath8pt @xcite , which suggests that this behavior is a specific feature for pt isotopes .
some investigations where the ground state shape evolution in pt isotope chain from the prolate towards the oblate shapes were performed in refs .
@xcite .
the objectives formulated above are achieved according to the following plan . in section
ii , a short presentation of the formalisms used for the description of the pt even - even isotopes is given .
numerical results and their comparison with the corresponding experimental data are discussed in section iii .
the final conclusions are drawn in section iv .
the formalisms x(5 ) , z(5 ) , iswsa , iswma , ssa and sma are derived by a set of approximations applied to the bohr - mottelson hamiltonian @xcite , @xmath16+c\frac{\beta^{2}}{2 } , \label{has}\ ] ] amended with a potential @xcite @xmath17 the form of the @xmath0 and @xmath1 potential allows to separate the @xmath0 variable from the @xmath1 and the three euler angles @xmath18 , @xmath19 and @xmath20 . here , @xmath21 s denote the angular momentum components in the intrinsic reference frame .
a full separation may be however achieved by expanding the rotor term in power series of @xmath1 around either of @xmath22 or of @xmath23 and , moreover , by replacing the factor @xmath24 multiplying the @xmath1-dependent term with its average value , denoted hereafter by @xmath25 .
the resulting equations are : @xmath26f(\beta)=\varepsilon_{\beta}f(\beta ) , \label{eqbeta}\ ] ] @xmath27\phi(\gamma)=\tilde{\varepsilon}_{\gamma}\phi(\gamma ) , \label{eqgamma}\ ] ] where the following notations are used : @xmath28 @xmath29 and @xmath30 are the contributions coming from the rotor term and their expressions depend on the order of the @xmath1 series truncation . for the sake of fixing the notations and defining the main ingredients , in what follows the above mentioned approaches will be briefly described . for details
we advise the reader to consult refs .
@xcite . in eq . ( [ eqbeta ] ) , the x(5 ) , z(5 ) , iswsa and iswma models use a common potential in @xmath0 , namely an infinite square well @xmath31 with such a choice eq .
( 2.3 ) admits the bessel functions of irrational order @xmath32 , as solutions : @xmath33 @xmath34 denotes the normalization factor , @xmath35 the bessel function zeros , while @xmath36 is the total intrinsic angular momentum .
by contrast the ssa and sma , use in eq .
( [ eqbeta ] ) a sextic oscillator plus a centrifugal barrier potential @xcite , @xmath37 here , @xmath38 has two different values , one for @xmath36 even and other for @xmath36 odd , while @xmath39 are constants which are fixed such that the two potentials @xmath40 and @xmath41 have the same minimum energy .
( [ eqbeta ] ) , with @xmath42 and the potential given by eq .
( [ v1debeta ] ) , is quasi - exactly solved , the solutions being of the form : @xmath43 where @xmath44 is the normalization factor , while @xmath45 are polynomials of order @xmath46 in @xmath47 .
concerning eq .
( [ eqgamma ] ) , the x(5 ) and z(5 ) chose an oscillator and a shifted oscillator potential , respectively : @xmath48 indeed , for x(5 ) @xmath49 and the solutions of eq . (
2.4 ) are the generalized laguerre polynomials , @xmath50 : @xmath51 the quantum number k corresponds to the angular momentum projection on the intrinsic z - axis . as for z(5 ) , @xmath23 and the corresponding equation ( 2.4 ) is obeyed by the hermite polynomials @xmath52 : @xmath53 both models , the x(5 ) and the z(5 ) , consider in eq .
( [ eqgamma ] ) a zeroth order of approximation for the rotor term .
this is not the case for the iswsa , iswma , ssa and sma models , where a second order power expansion of both the rotor term and the periodic potential @xmath54 is used , which results of having the spheroidal ( @xmath55 ) and mathieu ( @xmath56 ) functions as solutions of the resulting differential equations , respectively : @xmath57 the expressions of @xmath58 and @xmath59 will be specified below .
the advantages of the mathieu and spheroidal functions consist of that they are periodic , defined on a bound interval and normalized to unity with the integration measure of @xmath60 , preserving in this way the hermiticity of the initial hamiltonian .
note that the other approaches do not satisfy these conditions .
the total energy of the system is obtained by summing the eigenvalues of the @xmath0 and @xmath1 equations : @xmath61 the excitation energies yielded by the formalisms used in the present paper , are as follows : @xmath62 with @xmath63 , @xmath64 and @xmath65 arbitrary parameters . in our calcultions the parameter @xmath66
is fitted .
@xmath67 @xmath68,\nonumber\\ & & \lambda_{m_{\gamma},n_{\gamma}}=\frac{1}{9}\left[\tilde{\varepsilon}_{\gamma}-\frac{u_{1}}{2}-\frac{11}{27}d+\frac{1}{3}l(l+1)\right],\,c^{2}=\frac{1}{9}\left(\frac{u_{1}}{2}+u_{2}-\frac{2}{27}d\right),\nonumber\\ & & m_{\gamma}=\frac{k}{2},\;\;d = l(l+1)-k^{2}-2,\,f=\frac{1}{\langle\beta^{2}\rangle}\frac{\hbar^{2}}{2b},\end{aligned}\ ] ] @xmath69,\\ & & q=\frac{1}{36}\left(\frac{10}{9}l(l+1)-\frac{13}{12}r^{2}+u_{1}-\frac{9}{4}\right),\,a_{n_{\gamma}}=\frac{1}{9}\left(\tilde{\varepsilon}_{\gamma}+\frac{3}{4}r^{2}+\frac{5}{2}\right)-2q,\nonumber\end{aligned}\ ] ] @xmath70+f\left[9\lambda_{m_{\gamma},n_{\gamma}}(c)+\frac{u_{1}}{2}+\frac{11}{27}d - l(l+1)\right],\nonumber\\ & & \lambda_{m_{\gamma},n_{\gamma}}=\frac{1}{9}\left[\tilde{\varepsilon}_{\gamma}-\frac{u_{1}}{2}-\frac{11}{27}d+\frac{1}{3}l(l+1)\right]+\frac{2l(l+1)}{27},\,g=\frac{\hbar^{2}}{2b},\end{aligned}\ ] ] @xmath71+f\left[9a_{n_{\gamma}}(l , r)+18q(l , r)-\frac{3}{4}r^{2}-\frac{5}{2}\right]\nonumber,\end{aligned}\ ] ] where @xmath72 satisfy the equation : @xmath73p_{n_{\beta},l}^{(m)}(\beta^{2})=\lambda_{n_{\beta}}^{(m)}p_{n_{\beta},l}^{(m)}(\beta^{2 } ) .
\label{qoper}\ ] ] the specific @xmath0 and @xmath1 potentials of the six approaches used in the present paper are collected , for comparison , in table [ tab-0 ] .
the potentials in the @xmath0 variable are to be amended by a centrifugal term due to the rotor component of the starting hamiltonian . .
here we list the @xmath0 and @xmath1 potentials used by the approaches having as acronyms iswsa , ssa , iswma and sma . for comparison
the potentials characterizing x(5 ) and z(5 ) are also given . [ cols="^,<,<",options="header " , ] [ tab-1 ] each approach involves a number of free parameters for energies as well as for b(e2 ) values .
these are fixed by fitting some particular experimental data concerning either the excitation energies or the reduced transition probabilities .
the results of the fitting procedure adopted are listed in tables [ tab-2]@xmath74[tab-3 ] .
as seen from these tables , the number of parameters used for fitting the spectra in @xmath75 , @xmath76 and @xmath77 are 2 , 4 , 6 , respectively , while in the fitting of @xmath78 values 1 , 2 , 2 parameters are used , respectively . also ,
from table iii we notice that the number of parameters used for fitting the spectra in z(5 ) , iswma and sma are 1 , 3 , 5 respectively , while in fitting the @xmath78s 1 , 2 , 2 parameters are used respectively .
[ tab-2 ] [ tab-3 ] numerical results for the excitation energies of the ground , @xmath0 and @xmath1 bands , as well as for the quadrupole electric transitions between states of these bands are compared with the corresponding experimental data in tables [ tab-4 ] and [ tab-5 ] , respectively . for each formalism
the agreement between the calculation results and the corresponding experimental data is quantitatively appraised by the r.m.s values of the deviations . from table
[ tab-4 ] , one can see that spectra of the isotopes @xmath79pt , @xmath12pt and @xmath14pt are better described by ssa and iswsa than by x(5 ) .
the best approach seems to be ssa .
moreover , the x(5 ) failure in explaining the data from the @xmath0 band is removed by ssa , and that happens especially for @xmath12pt . concerning the @xmath1 band , all three formalisms , ssa , iswsa and x(5 ) , encounter difficulties in explaining the band head energy .
a possible explanation would be that the state @xmath80 , does not actually belong to the @xmath1 band .
in this context we mention the fact that two alternative interpretations have been studied in ref .
@xcite with a related description appearing in ref.@xcite .
a similar situation is met in @xmath15pt . in @xmath8pt , however , all three bands considered here are realistically described by ssa .
the comparison of the numerical results yielded by sma , iswma and z(5 ) with experimental data for the even - even isotopes @xmath81pt , is made also in table [ tab-4 ] , with the result in favor of sma and iswma . .
[ tab-4 ] the electromagnetic transition probabilities , calculated with eq .
( [ redtrans ] ) , are included in tables [ tab-5 ] . analyzing the r.m.s .
values for each model , one may conclude that ssa and iswsa describe the experimental data batter than x(5 ) , while sma and iswma better than z(5 ) .
an explanation for this picture could be that x(5 ) and z(5 ) use only the zero order approximation of the harmonic part of the transition operator ( [ transop ] ) . indeed ,
as shown in table [ tab-5 ] , for @xmath79pt the results obtained by ssa and iswsa using only the harmonic transition operator are almost identical with those of x(5 ) .
by contrast for @xmath82pt where the anharmonic contributions were included , the results of ssa and iswsa are better than those of x(5 ) .
it is worth noticing that the r.m.s .
associated to z(5 ) for @xmath10pt and @xmath83pt are smaller than those provided by iswma .
this situation might be caused by the fact the two approached considered for the @xmath1 band different descriptions .
indeed , in the framework of z(5 ) the states of @xmath1 band are characterized by @xmath84 , while the iswma @xmath1 states have @xmath85 .
[ tab-5 ] [ tab-6 ] in table vii we list the results for branching ratios of few states from the @xmath1 and @xmath0 bands obtained by ssa , iswsa , sma and iswma approaches , respectively .
they are compared with the experimental data of ref.[40 ] .
for @xmath86pt we list also the results yielded by the z(5 ) formalism .
the parameters determining the transition operator were fixed as follows . for @xmath8pt and @xmath87pt
we kept @xmath88 as given in tables iii and iv respectively , while @xmath89 was fixed by a least square procedure .
the results for @xmath89 are also listed in table vii . as for the rest of isotopes from the mentioned table , the parameters @xmath88 , @xmath89 are as listed in tables iii ,
another objective of the present work is to determine the isotope shape in ground and excited states , within both the ssa and the sma .
indeed , it is interesting to see how the shape changes when one passes from one isotope to another and moreover whether this picture is state dependent.we expect to visualize the shape phase transition and also possible shape coexistence .
the static shape is defined by the values of the intrinsic variables @xmath0 and @xmath1 for which the probability density ( the probability in the volume unit of @xmath90 ) , @xmath91 reaches a maximum value . in fig.1 ,
the contour plots are represented in the coordinates @xmath92 . in order to save the space we chose two representatives for ssa , @xmath79pt and @xmath8pt , and one for sma , @xmath87pt .
indeed , the graphs corresponding to @xmath93pt are similar to that of @xmath79pt and those of @xmath94pt resemble that of @xmath87pt .
we may ask ourself why to make such plots once we know that the power expansion in @xmath1 was performed around @xmath95 and @xmath96 we notice that the density maxima are met not in the same point where the potential is minimum .
the reason is that the density accounts also for the kinetic energy and moreover includes a factor defining the measure of the integration in the @xmath0 and @xmath1 coordinates .
these figures reflect the structure of the wave functions . indeed , since the @xmath1 dependent function depends on @xmath97 and the spheroidal functions are symmetric with respect to the space reflection transformation , the graphs exhibit the symmetry @xmath98 . concerning sma
the mentioned symmetry is caused by the fact the potential in @xmath1 is function of @xmath99 .
also , the node of the @xmath0 function causes a doublet maxima with the same @xmath1 . for @xmath8pt
we notice equal density curves which surround two maxima of identical beta .
this situation is specific to the shape coexistence .
it is worth mentioning that such transition is showing up despite the fact that for all isotopes @xmath100pt we used a power expansion in @xmath1 around @xmath101 .
that means that the transition is caused not only by the potential shape but also by the structure coefficients involved in the associated differential equations . actually , we calculated the spectroscopic properties of pt isotopes with @xmath102 also with a power expansion in @xmath1 around @xmath103 .
however , the results of sma are characterized by a smaller r.m.s values for the deviations of the predictions from the experimental data . it is interesting to note that although we changed the description when we passed from @xmath8pt to @xmath87pt the probability density undergoes a smooth transition .
the maxima surrounded by equidensity curves merge in one maximum at @xmath104 for ground and @xmath0 band states , while for @xmath1 band states the doublets are well separated .
how this picture is modified when additional degrees of freedom like octupole @xcite or single particle @xcite will be analyzed elsewhere .
note that for @xmath87pt the considered excited state in the @xmath0 band is @xmath105 and not @xmath106 as happens for other isotopes .
the reason is that , as seen from table v , the highest spin state for which energies in nuclei with @xmath107 , calculated with sma , is @xmath105 . ,
@xmath108 , @xmath109 , @xmath110 , @xmath80 , @xmath111 , @xmath112 and @xmath113 of @xmath114pt and @xmath87pt , calculated with ssa and sma , respectively .
the steps used in the contour plots are 30 , 10 , 20 for @xmath79pt , @xmath8pt and @xmath87pt , respectively .
exceptions are @xmath115 for @xmath8pt and @xmath116 for @xmath87pt , where the steps are 12 and 25 , respectively . , width=529,height=793 ]
in the previous section we described some even - even isotopes of pt by four solvable models emerging from the generalized bohr mottelson hamiltonian .
indeed , for the isotopes with @xmath117 the approaches are those abbreviated by ssa and iswsa , respectively , while for the rest of nuclei , @xmath118 , the sma and iswma are alternatively used .
it is worth mentioning that the approach called iswma was used for the first time in the present paper .
since the first set exhibits some features of the x(5 ) `` symmetry '' we compared the results of our calculations with those obtained with the x(5 ) formalism , if they are available .
as for the other isotopes the results were compared with the z(5 ) results .
one concludes that our results are slightly better than those obtained with x(5 ) and z(5 ) methods regarding both the excitation energies and reduced e2 transition probabilities . the wave function structure is nicely reflected in the contour plots for the probability density .
it is suggested that due to the hamiltonian symmetries the wave functions might be suitable for accounting for shape evolution as well as for possible shape coexistence . | energies of the ground , @xmath0 and @xmath1 bands as well as the associated b(e2 ) values are determined for each even - even isotope of the @xmath2pt chain by the exact solutions of some differential equations which approximate the generalized bohr - mottelson hamiltonian .
the emerging approaches are called the sextic and spheroidal approach ( ssa ) , the sextic and mathieu approach ( sma ) , the infinite square well and spheroidal approach ( iswsa ) and the infinite square well and mathieu approach ( iswma ) , respectively .
while the first three methods were formulated in some earlier papers of the present authors , iswma is an inedited approach of this work .
numerical results are compared with those obtained with the so called x(5 ) and z(5 ) models . a contour plot for the probability density as function of the intrinsic dynamic deformations
is given for a few states of the three considered bands with the aim of evidencing the shape evolution along the isotope chain and pointing out possible shape coexistence . |
the standard model of particle physics considers neutrinos to be massless .
sudbury neutrino observatory@xcite@xcite gave evidence of neutrino oscillations which was further confirmed by kamland experiment @xcite .
this landmark research assigned mass to the neutrinos and gave a clear indication of new physics beyond the standard model .
a simple stretch in the standard model was able to stand up with the mass of neutrino . in neutrino physics
the standard three flavour neutrino oscillations can be explained with the help of six parameters namely @xmath8 , @xmath9 , @xmath10 , @xmath11 , @xmath12 and @xmath13 . amongst these six parameters , solar parameters(@xmath14 , @xmath11 ) and atmospheric parameters ( @xmath15 )
have been measured with high precision .
furthermore , daya bay and reno reactor experiments have strongly constrained the value of mixing angle @xmath16 .
now we are in need of such neutrino experiments which can impose tight constraints on the value of @xmath17 and mass hierarchy .
some anomalies popped up while observing appearance channel and disappearance channel of @xmath18 at lsnd experiment . while observing @xmath19 appearance channel ,
lsnd @xcite @xcite @xcite @xcite @xcite @xcite was the first experiment to publish evidence of a signal at @xmath20 . later in 2002
, miniboone @xcite @xcite checked the lsnd result for @xmath21 ( @xmath22 ) appearance channel . in miniboone experiment , while observing the ccqe events rate through @xmath23 above 475 mev energy , no excess events were found but for energies @xmath24 475 mev @xmath25 excess events were observed . in this way
, miniboone supported the lsnd result .
the lep data@xcite@xcite advocates the number of weakly interacting light neutrinos , that couple with the z bosons through electroweak interactions , to be 2.984 @xmath26 0.008 ; thus closing the door for more than three active neutrinos .
hence , the heavy neutrino announced by lsnd group should be different from these three active neutrinos .
this higher mass splitting in the standard three active neutrino model was accommodated by introducing sterile neutrinos .
sterile neutrinos carry a new flavor which can mix up with the other three flavors of standard model but they do not couple with w and z bosons . the number of sterile neutrinos can vary from minimum one to any integer n. + some cosmological evidences like cmb anisotropies @xcite @xcite @xcite @xcite @xcite and big bang nucleosynthesis @xcite @xcite also stood up with the lsnd data .
the results reported by the combined analysis @xcite of baryonic acoustic oscillations(bao ) ` @xmath27+plasz+shear+rsd ' indicated the presence of sterile neutrinos by stipulating the number of effective neutrinos @xmath28 @xmath29 , @xmath30 and giving preference for @xmath31 at 1.4 @xmath6 level and non zero mass of sterile neutrino at 3.4 @xmath6 level .
the gallium solar neutrino experiments(gallium anomaly ) gallex @xcite , sage @xcite and the antineutrino reactor experiments ( @xmath32 ) like bugey-3 , bugey-4 , gosgen , kransnogark , iil @xcite ( reactor anomaly ) indicated that electron neutrinos and antineutrinos may disappear at short baselines .
such disappearance can be explained by the presence of at least one massive neutrino ( of the order of 1 ev ) .
thus , these experiments also indicated the presence of sterile neutrino and supported the lsnd results .
some constraints imposed by the combined fit of reactor , gallium , solar and @xmath33 scattering data are @xmath34 and @xmath35 at 95% cl @xcite .
few atmospheric neutrino experiments such as icecube @xcite , minos @xcite @xcite @xcite , ccfr @xcite have also imposed strong constraints on sterile parameters .
+ the four flavors of neutrino can be studied in either of the two different neutrino mass schemes , 3 + 1 or 2 + 2 schemes@xcite . for our work
we have selected ( 3 + 1 ) four flavor neutrino mass scheme . in this framework ,
maki - nakagawa - sakata ( mns ) mixing matrix ( @xmath36 ) , includes six mixing angles @xmath37 , three dirac phases and three majorana phases . in our analysis majorana phases are not taken into consideration .
+ neutrino factory @xcite@xcite provides excellent sensitivity to the standard neutrino oscillation parameters and therefore seems to be one of the promising option to explore and reanalyze the global fits for sterile neutrino parameters too .
to mention , it provides a platform to constrain one of the most searched cp violation in leptonic sector @xcite@xcite .
hence neutrino factory seems to provide a promising environment for the study of t and cpt violation .
the neutrino factory set up considered here is based on the international design study of neutrino factory [ ids - nf ] @xcite@xcite . from the measure of
@xmath38 we can not directly constrain cp phase because the value of @xmath38 in the presence of matter will contain in itself some cp odd effects even in the absence of cp phase . therefore ,
instead of checking @xmath39 , variation in @xmath40 can be studied to probe extent of true cp violation .
we have observed t violation through @xmath41 golden channel and @xmath42 discovery channel and cpt violation along @xmath43 disappearance channel .
+ our work is organized as follows . in section 2 ,
we illustrate 3 + 1 neutrino matrix parametrization . in next section , t violating effects
are checked for different channels . in section 4 ,
bounds on cpt violating terms are checked in presence of sterile neutrino . in the last section ,
we have summarized our study and discussed the results observed .
the 3 ( active ) + 1 ( sterile ) neutrino scheme can be looked upon as 3 + 1 or 2 + 2 scheme depending on the selection of mass ordering of the neutrinos . to check t and cpt violation
we have selected 3 + 1 scheme for our analysis . in this scheme
the flavor eigenstates @xmath44 and mass eigenstates @xmath45 are related by the given unitary transformation equation + @xmath46 + n unitary matrix contains n(n-1)/2 mixing angles and ( n-1)(n-2)/2 dirac type cp violating phases .
it will also contain ( n-1 ) number of additional majorana phases if the neutrinos are considered as majorana particles . ] + here unitary matrix ( u ) can be parametrized in terms of six mixing angles(@xmath47 ) , three dirac phases @xmath48 ( @xmath49 ) and three majorana phases .
majorana phases are neglected in our study as they do not affect the neutrino oscillations in any realistically observable way . in principle
, there are different parametrization schemes for the neutrino mixing matrix as their order of sub - rotation is arbitrary . our selection for parametrization of neutrino mixing matrix is @xmath50 where @xmath51 are the complex rotation matrices in the ij plane , defined as @xmath52_{pq}=\begin{cases } cos{\theta}_{ij } & p = q = i ,
j\\ 1 & p = q \neq i , j\\ sin{\theta}_{ij } e^{-i\delta_{l } } & $ p = i , q = j$\\ -sin{\theta}_{ij } e^{i\delta_{l } } & $ p = j , q = i$\\ 0 & \text{otherwise } \end{cases}\ ] ] the order of rotation between 14 and 23 is arbitrary since these matrices commute .
when neutrinos pass through the earth matter , the charge current interactions ( cc ) of @xmath18 and neutral current interactions ( nc ) of @xmath53 with the matter give rise to a cc and nc potentials @xmath54 and @xmath55 respectively . while studying the sterile neutrinos , potential @xmath56 can not be neglected .
the effective cpt violating hamiltonian ( @xmath57 ) of neutrinos can be expressed as @xmath58 \end{split}\ ] ] here @xmath59 , @xmath60 and @xmath61 .
@xmath62 is the fermi constant , @xmath63 and @xmath64 are the number density of electrons and neutrons respectively with @xmath65 in earth matter .
the @xmath66 s are cpt violating terms .
different angular values for unitary matrix @xmath67 are checked in @xcite . in our work
we have considered @xmath68 .
hamiltonian @xmath69 can be diagonalized to @xmath70 by an unitary matrix @xmath71 .
this can be expressed as @xmath72 the matrix elements @xmath73_{ii}$ ] will represent the eigenvalues of @xmath69 .
full analytical expressions for neutrino oscillation probabilities are developed in this work by using time independent perturbation theory . in an attempt to apply perturbation
we have defined few oscillation parameters in terms of perturbative parameter @xmath74 , where @xmath75 .
the neutrino oscillation parameters can be rewritten as + @xmath76 + @xmath77 + @xmath78 + @xmath79 + @xmath80 + + we treat @xmath81 .
now the hamiltonian @xmath69 can be written as @xmath82\ ] ] where @xmath27 , @xmath83 and @xmath84 are the hamiltonians corresponding to zeroth , first and second order in @xmath74 respectively . the evolution equation for neutrino oscillation probability
is defined as @xmath85 where @xmath86 is the evolution matrix of neutrino which is also called oscillation probability amplitude @xmath87 the evolution matrix of neutrinos in terms of eigenvalues of @xmath88 can be written as @xmath89 where @xmath90 + from equation ( 7 ) the neutrino oscillation probability @xmath91 from flavor @xmath92 to flavor @xmath93 can be written as @xmath94 this is the general form of equation for neutrino oscillation probability .
in neutrino oscillations the flavor conversion probabilities from flavor @xmath92 to flavor @xmath93 can be written as @xmath95 { \sin}^{2}{{\delta}_{ij } } + 2 \displaystyle\sum_{i > j } im\left[{\tilde{u}}_{\alpha i } { \tilde{u}}_{\alpha j}^ { * } { \tilde{u}}_{\beta i}^ { * } { \tilde{u}}_{\beta j}\right ] \sin 2{\delta}_{ij}\ ] ] redefining the above probability equation as sum of @xmath96 and @xmath97 terms @xmath98 cp even terms are cp conserving and can be written as @xmath99 cp odd term are cp violating and can be written as @xmath100 assuming cpt to be conserved , the magnitude of cp violation @xmath101 will be equal to the magnitude of t violation @xmath102 , i.e. @xmath103 therefore we can write @xmath104 when neutrinos passes through the earth matter , the interaction of neutrinos with matter gives rise to an extra potential .
this potential is positive for neutrinos and negative for antineutrinos leading to different eigenvalues of hamiltonian for them .
further , this difference in hamiltonian for @xmath105 s and @xmath106 s give rise to fake(extrinsic ) cp violation .
hence , check on t violation appears to be a better choice in the presence of matter . from equation
( 16 ) the t violation can be looked upon as @xmath107 if we consider @xmath108 and @xmath109 the above equation becomes @xmath110
the term @xmath111 is known as jarlskog factor and @xmath112 are energy eigenvalues of hamiltonian in matter . @xmath113 the energy eigenvalues in matter
can be connected to the energy eigenvalues in vacuum by the relations @xmath114 , @xmath115 , @xmath116 and @xmath117 .
the terms @xmath118,@xmath119 , @xmath120 , @xmath121 in matter can be calculated with the help of the terms @xmath122 , @xmath123,@xmath124 , @xmath125 in vacuum ( where @xmath126 ) using the following expressions @xcite .
@xmath127 @xmath128 @xmath129 @xmath130 where @xmath131 @xmath132 @xmath133 is the diagonal element of the matter potential matrix in four neutrino scheme and a is the diagonal element of matter potential matrix in three neutrino scheme . since t violating effects can only be studied in appearance channels so @xmath134 . in an effort to put constraints on @xmath135
we have studied two appearance channels .
these are @xmath136 ( golden channel ) and @xmath42 ( discovery channel ) .
+ the t violation probability difference expression for the golden channel can be expressed as @xmath137[s_{14}c_{14}s_{24}\frac{a_{e}}{2e}\delta m_{31}^{2}/2e + ( \frac{a_{e}}{2e } + \delta m_{31}^{2}/2e)r^{e\mu } + s^{e\mu}]\sin{\delta m_{31}^{2}l/2e } \\ & \quad + \frac{4}{(-\delta m_{31}^{2}/2e)\frac{a_{e}}{2e}(\frac{a_{e}}{2e}-\delta m_{31}^{2}/2e)^{2 } } \times\\ & \quad [ \frac{a_{e}}{2e}r^{e\mu } + s^{e\mu}][(\delta m_{31}^{2}/2e)r^{e\mu } + s^{e\mu}]\sin{(\delta m_{31}^{2}/2e-\frac{a_{e}}{2e})l } + \\ & \quad \frac{4}{(-\delta m_{31}^{2}/2e)(\frac{a_{e}}{2e}-\delta m_{31}^{2}/2e)(\frac{a_{e}}{2e})^{2}}[s_{14}c_{14}s_{24}\frac{a_{e}}{2e}\delta m_{31}^{2}/2e + ( \frac{a_{e}}{2e } + \delta m_{31}^{2}/2e)r^{e\mu } \\ & \quad+ s^{e\mu } ] \times[(\delta m_{31}^{2}/2e ) r^{e\mu } + s^{e\mu}]\sin{\frac{a_{e}}{2e}l } \end{split}\ ] ] since large value of @xmath138 gives rise to rapid oscillations , hence @xmath138 terms can be averaged out . solving the above expression up to the power @xmath139
we get @xmath140 @xmath141 \end{split}\ ] ] @xmath142 is matter dependent term .
the change in @xmath143 will change the value of @xmath144 .
+ further we have developed equation of @xmath40 for discovery channel .
the discovery channel is not very useful in the standard three neutrino flavor framework , nevertheless while studying physics beyond three active neutrino flavor framework , it becomes very important . for discovery channel(@xmath42 )
the probability difference is given as @xmath145[c_{14}^{2}c_{24}s_{24}s_{34}\frac{a_{e}}{2e}\delta m_{31}^{2}/2e + ( \frac{a_{e}}{2e } + \delta m_{31}^{2}/2e)r^{\tau\mu } + s^{\tau\mu}]\sin{\delta m_{31}^{2}l/2e } \\
& \quad + \frac{4}{(-\delta m_{31}^{2}/2e)\frac{a_{e}}{2e}(\frac{a_{e}}{2e}-\delta m_{31}^{2}/2e)^{2 } } \times \\ & \quad [ \frac{a_{e}}{2e}r^{\tau\mu } + s^{\tau\mu}][(\delta m_{31}^{2}/2e)r^{\tau\mu } + s^{\tau\mu}]\sin{(\delta m_{31}^{2}/2e-\frac{a_{e}}{2e})l } + \\ & \quad \frac{4}{(-\delta m_{31}^{2}/2e)(\frac{a_{e}}{2e}-\delta m_{31}^{2}/2e)(\frac{a_{e}}{2e})^{2}}[c_{14}^{2}c_{24}s_{24}s_{34}\frac{a_{e}}{2e}\delta m_{31}^{2}/2e + ( \frac{a_{e}}{2e } + \delta m_{31}^{2}/2e)r^{\tau\mu } \\ & \quad + s^{\tau\mu } ] \times[(\delta m_{31}^{2}/2e ) r^{\tau\mu } + s^{\tau\mu}]\sin{\frac{a_{e}}{2e}l } \end{split}\ ] ] solving the above expression up to the power @xmath139 we get , @xmath146 @xmath147 \end{split}\ ] ] the @xmath40 for three neutrino framework @xcite is given by @xmath148 keeping the best fit values of neutrino oscillation parameters and assigning maximum value to dirac phases i.e keeping mod of sin of dirac phases to be unity will lead us to maximum value of @xmath40 .
this assumption will render the maximum limit on the bounds which can be imposed on t violation arising due to the presence of dirac phases if all other oscillation parameters are known with utmost accuracy . from equation ( 32 )
gives the value of @xmath149 for three neutrino flavor framework @xcite .
this value is independent of the selection of probing channel and presence of matter effects .
whereas in 4 flavor framework it will depend on the selection of channel through which we want to probe cp or t violation and it will vary with matter effects too . within 4 flavor neutrino framework
the magnitude of @xmath40 will depend on active flavor neutrino mixing angles ( known with accuracy ) , sterile neutrino mixing angles ( still needs better bounds ) , matter effects , baseline , energy and dirac phases ( not known ) .
imposition of constraints on dirac phase ( three neutrino flavor ) or phases ( four neutrino flavor ) is still in research phase . with baseline for two different appearance channels ] with baseline for two different appearance channels ] the probability differences @xmath150 in neutrino sector are represented by equations ( 28 ) and ( 31 ) for @xmath151 and @xmath152 channels respectively .
the values of sterile parameters used in the above mentioned equations are taken from + ( i ) long baseline neutrino oscillation experiments @xcite + @xmath153 , @xmath154 and @xmath155 + ( ii ) short baseline reactor and atmospheric experiments @xcite @xcite + @xmath156 , @xmath157 and @xmath158 + the variations in @xmath159 and @xmath160 are checked along the baseline for two different energies i.e 4 gev and 50 gev . plot ( a ) of figure 1 reflects a very small variation in @xmath161 in comparison to @xmath162 for golden channel when it is checked with two different energies and two sets of sterile parameter values . plot ( b ) of figure 1 reflects a reasonable variation in @xmath161 in comparison to @xmath162 for discovery channel when it is checked for 50 gev energy and sterile parameters are taken from reactor+atmospheric experiments .
the @xmath163 comes up as the promising channel to observe the signatures of t violation . from the analysis
we conclude that neutrino factory operating at 50 gev has the potential to capture the signatures of t violation through @xmath164 channel if true sterile parameter values are equal to that taken from reactor+atmospheric experiments . at the same time
if the upcoming neutrino experimental setups captures @xmath40 value above 0.137 then we can stipulate presence of some new physics beyond three active flavor neutrino physics which is responsible for the enlargement observed in @xmath40 value .
cpt invariance is one of the most fundamental symmetries of nature .
cpt conservation indicates the invariance in the properties of physical quantities under the discrete transformations such as charge conjugation ( c ) , parity inversion ( p ) and time reversal ( t ) along with the invariance under lorentz transformation .
cpt invariance is one of the symmetries of local quantum field theory which implies that there is an important relation between cpt invariance and lorentz invariance .
if cpt invariance is violated , lorentz invariance must violate but if lorentz invariance is violated it is not necessary that cpt invariance must violate . in our work @xmath165 disappearance channel
is probed to check cpt violation .
the cpt violating probability difference can be written as @xmath166 the intrinsic cpt violation arises due to the violation of cpt invariance theorem .
a hamiltonian @xmath57 containing cpt violating terms is defined by equation ( 4 ) and the general form of neutrino oscillation probability is mentioned in equation ( 10 ) .
the terms @xmath167 and @xmath168 of the expression ( 10 ) are the elements of the unitary matrix @xmath169 .
the construction of unitary matrix @xmath169 with the help of eigenvalues and eigenvectors of hamiltonian @xmath170,@xmath171 and @xmath172 is mentioned in the appendix .
after the formation of unitary matrix @xmath173 , we have developed the neutrino oscillation probability equations up to second order in @xmath74 . since @xmath138 is large
, so we average out the effects produced due to @xmath138 in the probability equations .
neutrino oscillation probabilities for different oscillation channels containing cpt violating parameters can be developed as @xmath174 @xmath175 @xmath176 \end{split}\ ] ] @xmath177\\ & \quad+ \theta_{24}\theta_{34}\sin{\delta_{3}}\sin{2(\delta_{31}+ { \delta c}_{31}l/2 ) } \end{split}\ ] ] @xmath178 for small angles ( @xmath179 ) these oscillation probabilities can be written as @xmath180 @xmath181 @xmath182 \end{split}\ ] ] @xmath183 + s_{24}s_{34}\sin{\delta_{3}}\sin{2(\delta_{31}+ { \delta c}_{31}l/2 ) } \end{split}\ ] ] @xmath184 in order to analyse cpt violation at probability level in 4 flavor neutrino framework the value of @xmath185 considered in our analysis is given by @xmath186_{\delta c_{ij}\neq 0 } - [ ( p_{\alpha \beta})_{4\nu}]_{\delta c_{ij } = 0}\ ] ] with baseline and energy for normal hierarchy whereas ( c ) and ( d ) illustrates the same for inverted hierarchy .
for the oscillographs ( a ) and ( c ) the values of sterile parameters are selected from long baseline experiments while for ( b ) and ( d ) the values of sterile parameters are taken from reactor and atmospheric experiments . ] with baseline and energy for normal hierarchy whereas ( c ) and ( d ) illustrates the same for inverted hierarchy .
for the oscillographs ( a ) and ( c ) the values of sterile parameters are selected from long baseline experiments while for ( b ) and ( d ) the values of sterile parameters are taken from reactor and atmospheric experiments . ] with baseline and energy .
the values of sterile parameters are selected from reactor+atm experiments . left and right oscillographs are for normal and inverted hierarchies respectively . ] with baseline and energy .
the values of sterile parameters are selected from reactor+atm experiments . left and right oscillographs are for normal and inverted hierarchies respectively . ]
neutrino factory setup considered for analysing cpt violation is taken from the references @xcite@xcite@xcite@xcite and @xcite .
the experimental setup and detector specifications considered in our analysis are mentioned below .
neutrino factory setup consist of @xmath187 useful muon decays per polarity , with parent muon energy @xmath188 .
we have done our analysis for 10 years running of neutrino factory . in particle physics
meaningful observations always demands a detector with very good energy and angular resolutions . this view point lead us to select liquid argon detector for particle detection .
the energy resolution of the detector for muon is @xmath189 . +
a near detector is placed at a distance 20 m from the end of the decay straight of the muon storage ring .
effective baseline ( @xmath190 ) is used in place of baseline ( l ) , which is calculated using @xmath191 @xcite .
fiducial mass of near detector is 200 tons .
presence of near detector will minimize the systematic uncertainties in our observations . +
a 50 kt far detector is placed at a distance of 7500 km .
the systematic uncertainties considered for this analysis are given in table 1 .
an uncertainty of 5% on matter density @xcite@xcite is also considered in our work .
the simulated environment of the neutrino factory is created with the help of globes @xcite @xcite .
the analytical equations for four flavor neutrino conversion probabilities derived in this work are defined in the probability engine of the software .
the best fit values of oscillation parameters @xcite @xcite are mentioned in table 2 .
sterile parameter values mentioned in the table 2 represents best fit values for @xmath192 = 0.1 @xmath193
. .systematics [ cols="^,^,^,^ " , ] [ table : nonlin ] the constraints on cpt violating parameters @xmath194 and @xmath195 within two and three neutrino frameworks are mentioned in references @xcite @xcite @xcite @xcite @xcite @xcite and @xcite . in present work
we are trying to check the neutrino factory potential to capture cpt violating signatures in presence of sterile neutrino .
as the mass hierarchy determination is yet in research phase therefore in an attempt to make this work relevant we have analysed cpt invariance for both the mass hierarchies .
initially cpt violating signatures are checked at probability level .
the value of @xmath185 is estimated by substituting equation ( 41 ) in equation ( 44 ) for the channel @xmath196 .
the variation in @xmath197 and @xmath198 with baseline and energy is shown in figure 2 and figure 3 oscillographs respectively . the total cpt violation captured by any experiment
will be the sum of extrinsic cpt violation ( cpt violation arising due to matter effects ) and intrinsic or genuine cpt violation(which we are probing in present work ) . in an endeavour to constraint intrinsic cpt violating parameters
we must look for places where extrinsic cpt violation is negligible or very less . with three active neutrinos
the extrinsic cpt violation is checked in reference @xcite whereas with 3 ( active ) + 1 ( sterile ) neutrinos it is checked in reference @xcite .
these references conclude that extrinsic cpt violation for energies 4 gev- 6 gev is negligible at shorter baselines , roughly less than 2000 km .
whereas for longer baselines this effect decreases with energy .
equation ( 44 ) of our work will check the presence of pure cpt violation arising in the presence of sterile neutrino at probability level .
the values of cpt violating parameters considered while plotting oscillographs are @xmath199 gev and @xmath200 gev . looking at normal and inverted hierarchy
oscillographs(figure 2 ) we can observe the presence of pure cpt violating signatures at shorter baselines i.e. from 1300 km-2000 km for 4 gev-6 gev energies .
the references @xcite @xcite , which speak about extrinsic cpt violation have recorded very weak or almost negligible signatures of extrinsic cpt violation at the above mentioned energies and baselines .
hence baselines from 1300 km-2000 km with neutrino energies in the range 4 gev-6 gev are favourable for probing cpt violation with neutrino factory . in figure 2
while looking at normal hierarchy oscillographs we observe @xmath197 = -0.05 along baselines 4000 km-7500 km for energies 12 gev-30 gev .
the inverted hierarchy oscillographs of the same figure captures @xmath197=0.25 and 1 for sterile parameters taken from long baseline experiments and reactor+atmospheric experiments respectively .
this probability difference can be observed for baselines 4000 km-7500 km and energies 20 gev-50 gev .
+ after examining the presence of pure cpt violation at probability level we go ahead to observe the signatures of the same with realistic proposed neutrino experiments i.e. neutrino factory .
the specifications of neutrino factory considered in our work are mentioned earlier .
liquid argon detector seems a reasonable choice to grab the signatures of leptons in the considered energy range .
the rate(event ) level analysis depends on mathematical formulation ( oscillation probability ) , physics(types of interactions ) and r & d ( source properties and detector properties ) of the experiment .
looking at equations ( 34 ) to ( 44 ) we found that @xmath194 appears with @xmath201 term and @xmath195 term appears with @xmath202 term . the solar and atmospheric mass square difference ( @xmath201 and @xmath203 )
are of the order of @xmath204 and @xmath205 respectively .
hence , any change in mass terms due to the presence of cpt violating parameter will be better observed in @xmath206 term .
+ to hook cpt violating impression with neutrino factory we have investigated some observable parameters like r , @xmath207 and asymmetry factor .
these terms are defined by equations ( 45 ) and ( 46 ) .
the ratio r and ratio difference @xmath208 are examined as @xmath209 where @xmath210 denotes number of muon neutrinos reaching at detector as muon neutrinos and producing a @xmath211 lepton and @xmath212 denotes number of anti muon neutrinos reaching at detector as anti muon neutrinos and producing a @xmath213 lepton . in @xmath207
, @xmath214 denotes the ratio r in presence of cpt violating terms and @xmath215 denotes the ratio r in absence of cpt violating terms . ; ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] ; ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] with energy in the energy range 4 - 15 gev .
values of sterile parameters considered in plot ( a ) and plot ( b ) are selected from long baseline experiments and reactor+atm experiments respectively .
these observations are made for different values of cpt violating parameter @xmath206 ; ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] with energy in the energy range 4 - 15 gev .
values of sterile parameters considered in plot ( a ) and plot ( b ) are selected from long baseline experiments and reactor+atm experiments respectively .
these observations are made for different values of cpt violating parameter @xmath206 ; ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] with energy in the energy range 15 - 50 gev .
values of sterile parameters considered in plot ( a ) and plot ( b ) are selected from long baseline experiments and reactor+atm experiments respectively .
these observations are made for different values of cpt violating parameter @xmath206 ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] with energy in the energy range 15 - 50 gev .
values of sterile parameters considered in plot ( a ) and plot ( b ) are selected from long baseline experiments and reactor+atm experiments respectively .
these observations are made for different values of cpt violating parameter @xmath206 ( i ) @xmath216 ( setting cpt violating parameter to zero ) ( ii ) @xmath217 gev ( iii ) @xmath218 gev ( iv ) @xmath199 gev .
for all the observations @xmath200 gev ] in presence of matter the observable r will not be equal to one , even if pure cpt violation is absent .
it will be equal to a numerical value representing the ratio of neutrino and antineutrino interaction cross - sections .
if we want to analyse the extent of deviation produced by pure cpt violation , we have to hide or filter out the deviation produced by any other phenomenon . in an attempt to filter out pure cpt violating contribution from the total observed deviation we take into record a new observable @xmath208 .
this parameter is defined in the equation ( 45 ) .
figures 4 and 6 demonstrate the variation in r with energy whereas figures 5 and 7 exhibit variation in @xmath208 with energy for baseline 7500 km . we observe that at long baselines pure cpt violating effects get smaller with increase in energy . the presence of cpt violation signatures can be observed with neutrino factory and it can be checked by looking r and @xmath208 plots ( figure 4- figure 7 ) for different values of cpt violating parameter @xmath195 . as we know that in presence of sterile neutrino the manifestation of pure cpt signatures depends on the values of sterile parameters , hence the entire analysis is performed with two sets of best fit values of sterile parameters which were examined by different neutrino experiments .
the results from neutrino factory with sterile parameter values obtained from reactor+atmospheric experiments exhibit larger deviation in observables r and @xmath208 in comparison to the results obtained with sterile parameter values taken from long baseline experiments .
these observables are checked for both mass hierarchies . from the figures 4,5,6 and 7
we comprehend that after 15 gev there is a flip in sign of the observables for both the hierarchies . at the same time
the amount of deviation measured for pure cpt violating effects are different for nh and ih for the same energy and baseline .
+ the next observable asymmetry factor @xmath219 is defined as @xmath220 this ratio is determined by using far and near detectors . the variations in asymmetry factor with energy for baseline 7500 km are shown by figures 8 and 9 .
these figures reflect the variations in observable @xmath219 for different values of @xmath195 ( cpt violating parameter ) and for both mass hierarchies .
the @xmath195=0 will reflect @xmath219 values without any contribution from cpt violating terms .
the asymmetry factor increases with the increase in the value of @xmath195 .
an enhancement in magnitude of asymmetry factor is also observed with the increase in values of sterile angles .
hence more stringent bounds on sterile parameters are required to check the extent of cpt violation .
( e ) as a function of energy in energy range 4 - 15 gev .
values of sterile parameters considered in plot(a ) and plot(b ) are taken from long baseline experiments and reactor+atm experiments respectively ] ( e ) as a function of energy in energy range 4 - 15 gev .
values of sterile parameters considered in plot(a ) and plot(b ) are taken from long baseline experiments and reactor+atm experiments respectively ] ( e ) as a function of energy in energy range 15 - 50 gev .
values of sterile parameters considered in plot(a ) and plot(b ) are taken from long baseline experiments and reactor+atm experiments respectively ] ( e ) as a function of energy in energy range 15 - 50 gev .
values of sterile parameters considered in plot(a ) and plot(b ) are taken from long baseline experiments and reactor+atm experiments respectively ] -@xmath194 plane for 90% c.l . with cpt violating terms @xmath206 = @xmath194 = @xmath221 gev taking energies 15 gev,25 gev and 50 gev respectively . in top two plots
sterile parameters values are taken from reactor+atmospheric experiments while for the bottom plot these values are taken from both the reactor+atmospheric and long baseline experiments.,scaledwidth=70.0% ] -@xmath194 plane for 90% c.l . with cpt violating terms @xmath206 = @xmath194 = @xmath221 gev taking energies 15 gev,25 gev and 50 gev respectively . in top two plots sterile parameters values
are taken from reactor+atmospheric experiments while for the bottom plot these values are taken from both the reactor+atmospheric and long baseline experiments.,scaledwidth=70.0% ] -@xmath194 plane for 90% c.l . with cpt violating terms @xmath206 = @xmath194 = @xmath221 gev taking energies 15 gev,25 gev and 50 gev respectively . in top two plots sterile parameters values
are taken from reactor+atmospheric experiments while for the bottom plot these values are taken from both the reactor+atmospheric and long baseline experiments.,scaledwidth=70.0% ] -@xmath194 plane for 90% c.l . with cpt violating terms @xmath206 = @xmath222 and @xmath194 = @xmath221 gev taking energies 15 gev,25 gev and 50 gev respectively . in top two plots
sterile parameters values are taken from reactor+atmospheric experiments while for the bottom plot these values are taken from both the reactor+atmospheric and long baseline experiments.,scaledwidth=70.0% ] -@xmath194 plane for 90% c.l . with cpt violating terms @xmath206 = @xmath222 and @xmath194 = @xmath221 gev taking energies 15 gev,25 gev and 50 gev respectively . in top two plots
sterile parameters values are taken from reactor+atmospheric experiments while for the bottom plot these values are taken from both the reactor+atmospheric and long baseline experiments.,scaledwidth=70.0% ] -@xmath194 plane for 90% c.l . with cpt violating terms @xmath206 = @xmath222 and @xmath194 = @xmath221 gev taking energies 15 gev,25 gev and 50 gev respectively . in top two plots
sterile parameters values are taken from reactor+atmospheric experiments while for the bottom plot these values are taken from both the reactor+atmospheric and long baseline experiments.,scaledwidth=70.0% ] as a function of @xmath195 is shown .
the green curve will represents inverted hierarchy and red curve represents normal hierarchy as true hierarchy . ] in our work we have imposed bounds on cpt violating parameters @xmath195 and @xmath194 at 90% c.l .
figures 10 and 11 demonstrate contours in @xmath195 and @xmath194 plane with true value of cpt violating parameters as @xmath195 = @xmath223 gev,@xmath194 = @xmath223 gev and @xmath195 = @xmath224
gev,@xmath194 = @xmath223 gev respectively .
each figure consists of three plots at three different energies 15 gev , 25 gev and 50 gev for baseline 7500 km .
these plots illustrate bounds on cpt violating parameters at mentioned energies .
the selection of three different energies are based on the results of previous observations ( i.e r , @xmath208 and @xmath219 ) . at selected energy 15 gev , change in sign ( + ve to -ve )
is observed in the observables while studying effects of cpt violation for considered baselines and energies .
this observation makes 15 gev energy important for studying cpt violating effects .
a proposal of neutrino factory producing neutrino beam of 25 gev muons is described in reference @xcite whereas in a different proposal we have a 50 gev muon beam for the production of neutrinos in neutrino factory @xcite .
therefore , by checking at the extent of bounds imposed on cpt violating parameters with energies 25 gev and 50 gev we want to check that by what order the results will improve if we move towards higher energies . by looking at different energy contours
we conclude that amongst the three selected energies , 50 gev energy is the best suited energy to constrain cpt violating parameter @xmath195 , if nature allows nh to be true hierarchy . at the same time
we observe that for long baseline experiment ih will be favourable hierarchy for determination of bounds on cpt violating parameters for energies less than 15 gev . + as discussed earlier that , out of two parameters @xmath195 and @xmath194 considered in our analysis , the variation in @xmath195 will produce larger variation in the detectable observables which are used in our work for checking cpt violation .
figure 12 shows value of @xmath225 as a function of cpt violating parameter @xmath195 .
it is plotted by marginalizing over oscillation parameters @xmath226 in 3@xmath6 range of their best fit values and @xmath227 from 0 to 2@xmath228 .
looking at figure we observe that the presence of cpt violation can be detected for @xmath229 gev with neutrino factory for nh within 3@xmath6 limit .
neutrino factory will provide us a potential setup for observing t violation and setting significant bounds on cpt violation in neutrino sector . in four(3 + 1 )
neutrino flavor framework the angular mixing parameters of three active neutrinos are well constrained while the sterile parameters still needs better bounds on them . with the change in the value of sterile parameters a notable variation in bounds on cpt violating parameter and on the extent of t violation
is captured by neutrino factory .
hence , well constrained values of sterile parameters will allow any neutrino experiment to impose better constraints on t violation and cpt violating parameters . amongst two
selected sets of values of sterile parameters i.e. from long baseline experiments and reactor+atmospheric experiments we observed that neutrino factory potential for investigating t and cpt violation enhances when the sterile parameters values are equal to those which are constrained by reactor and atmospheric experiments .
neutrino factory with 50 gev energy is sensitive to probe t violation when true values of sterile parameters will be equal to those predicted by reactor+atmospheric experiments .
we stipulate that a pure cpt violating effects can be observed along short baseline i.e 1300 km-2000 km with energies 4 gev to 6 gev where extrinsic cpt violation is negligible . on the other hand at long baselines we can observe these effects with energies in the range 20 gev- 40 gev along baselines 4000 km-7500 km .
cpt violating parameters @xmath229 gev for nh and @xmath230 gev for ih will make neutrino factory capable to capture signatures of cpt violation at 3 @xmath6 level .
using the time independent perturbation theory we calculate the eigenvalues and eigenvectors of hamiltonian @xmath69 up to the order of @xmath231 correctly .
+ eigenvalues of @xmath27 are given by 100 q. r. ahmad et al .
( sno collaboration ) , phys .
rev.lett . 89 ( 2002 ) 011302 .
s. n. ahmed et al .
( sno collaboration ) , phys .
92 ( 2004 ) 181301 , arxiv : nucl - ex/0309004 .
k. eguchi et al .
( kamland collaboration ) , phys .
lett . 90 ( 2003 ) 021802 .
c. athanassopoulos et al .
( lsnd collaboration ) , phys rev .
77 ( 1996 ) 3082 , arxiv : nucl - ex/9605003 . c. athanassopoulos et al .
( lsnd collaboration ) , phys .
rev.lett . 75 ( 1995 ) 2650 .
c. athanassopoulos et al .
( lsnd collaboration ) , phys rev .
81 ( 1998 ) 1774 , arxiv : nucl - ex/9709006 . c. athanassopoulos et al .
( lsnd collaboration ) , phys .
c 58 ( 1998 ) 2489 , arxiv : nucl - ex/9706006 .
a. aguilar et al.(lsnd collaboration ) , phys .
d 64 ( 2001 ) 112007 , arxiv : hep - ex/0104049 . c. athanassopoulos et al .
( lsnd collaboration ) , phys . rev lett 77 ( 1996 ) 3082 , arxiv : nucl - ex /9605003 ] .
miniboone , a. a. aguilar - arevalo et al .
105,(2010 ) 181801 , arxiv:1007.1150 .
janet m. conrad , arxiv : hep - ex/1306.6494v1 .
w.m yao et al .
( particle data group ) , j. phys .
g 33 , 1 ( 2006 ) .
k. nakamura et al . ( particle data group ) ,
g37 ( 2010 ) 075021 [ spires ] .
b. a. reid , l. verde , r. jimenez and o. mena , jcap 1001 ( 2010 ) 003 , arxiv : hep - ph/0910.0008 .
m. c. gonzalez - garcia , m. maltoni and j. salvado , jhep 1008 , 117 ( 2010 ) , arxiv : hep - ph/1006.3795 .
j. hamann , s. hannestad , g. g. raffelt , i. tamborra and yvonne y. y. wong , phys .
( 2010 ) 181301 , arxiv : astro - ph/1006.5276 .
e. giusarma , m. corsi , m. archidiacono , r. de putter , a. melchiorri , o. mena and s. pandolfi , phys .
d 83 ( 2011 ) 115023 , arxiv : astro - ph/1102.4774
. z. hou , r. keisler , l. knox , m. millea and c. reichardt , arxiv : astro - ph/1104.2333 . y. i. izotov and t. x. thuan , astrophys .
j. 710 ( 2010 ) l67 , arxiv : astro - ph/1001.4440 .
e. aver , k. a. olive and e. d. skillman , jcap 1005 ( 2010 ) 003 , arxiv : astro - ph/1001.5218 ] .
jing - fei zhang , yun - he li , and xin zhang , arxiv : astro - ph/1408.4603v2 .
f. beutler , c. blake , m. colless , d. h. jones , et al.,mnras 416 , 3017 ( 2011 ) , arxiv : astro - ph/1106.3366 . c. blake , s. brough , m. colless , c. contreras ,
et al.,mnras 425 , 405 ( 2012 ) , arxiv : astro - ph/1204.3674 . n. padmanabhan , x. xu , d. j. eisenstein , r. scalzo , et al . , mnras 427 , 2132 ( 2012 ) , arxiv : astro - ph/1202.0090 . l. anderson , e. aubourg , s. bailey , d. bizyaev , et al.,mnras 427 , 3435 ( 2012 ) , arxiv : astro - ph/1203.6594 . c. l. bennett , d. larson , j. l. weiland , n. jarosik , et al .
, arxiv e - prints ( 2012 ) , arxiv : astro - ph/1212.5225 . kaether f , hampel w , heusser g , kiko j and kirsten t 2010 phys . lett .
b685 4754 , arxiv : math.ag/1001.2731 .
abdurashitov j n et al .
( sage collaboration ) 2009 phys .
c80 ( 2009 ) 015807 arxiv : nucl - ex/0901.2200 . c. guinti et al .
, arxiv : hep - ph/1210.5715 . c. guinti , doi:10.5506/aphyspolbsupp.6.667 .
v. barger , y. gao , d. marfatia , arxiv : hep - ph/1109.6748v2 ( 2011 ) .
p. adamson et al .
( minos collaboration ) , phys .
( 2011 ) 011802 , arxiv : hep - ex/1104.3922 .
d. hernandez and a. y. smirnov , phys .
b 706 ( 2012 ) 360 , arxiv : hep - ph/1105.5946 . c giunti and m laveder , arxiv : hep - ph/1109.4033 .
i. e. stockdale et al . , phys .
52 , 1384 ( 1984 ) .
caldwell and r.n .
mohapatra , phys .
d 48 ( 1993 ) 3259 .
s. geer , phys .
d 57 ( 1998 ) 6989 ( erratum - ibid .
d 59 ( 1999 ) 039903 ) , arxiv : hep - ph/9712290 .
a. de rujula , m. b. gavela and p. hernandez , nucl .
b 547 ( 1999)21 . c. albright , g. anderson , v. barger et al . , arxiv : hep - ex/0008064 .
a. blondel et al .
instrum . meth .
a 467(2000)102 . m. apollonio et al.(cern working group on oscillation physics at neutrino factory)(2002 ) , arxiv : hep - ph/0210192 .
a. bandyopadhyay et al .
( iss physics working group ) rept .
72 , 106201 ( 2009 ) , arxiv : hep - ph/0710.4947 .
animesh chatterjee , raj gandhi , jyotsna singh , arxiv:1402.6265v1[hep - ph](2014 ) .
osamu yasuda , arxiv:1004.2388v1 .
fermilab - conf-13 - 300 .
stephen j. parke and thomas j. weiler , phys . lett .
b 501 ( 2001 ) 106 , arxiv : hep - ph/0011247v2 .
davide meloni , jian tang and walter winter , phys .
d 82 ( 2009 ) 093008 , arxiv : hep - ph/1007.2419v2 .
arman esmaili , francis halzen , o. l. g. peres , jcap b07 ( 2013 ) 048 , arxiv : hep - ph/1303.3294v2 .
a. donini , m. maltoni , d. meloni , p. migliozzi and f.terranova , jhep 0712 , 013(2007 ) [ arxiv:0704.0388 [ hep - ph ] ]
. international design study of the neutrino factory , http://www.ids-nf.org .
p. huber , m. lindner , m. rolinec and w. winter , phys . rev .
d 74 , 073003 ( 2006 ) , arxiv : hep - ph/0606119 .
p. huber , m. lindner and w. winter , nucl .
b 6453 ( 2002 ) , arxiv : hep - ph/0204352 .
e. ables et al.(minos ) fermilab - proposal - p-875 c. h. albright et al .
[ neutrino factory / muon collider collaboration ] , arxiv : physics/0411123 .
jian tang , walter winter , phys .
d 80 ( 2009 ) 053001 , arxiv:0903.3039v2 .
r. j. geller and t. hara , nucl .
instrum . meth .
a 503 ( 2003 ) 187 , arxiv : hep - ph/0111342 . t. ohlsson and w. winter , phys . rev .
d 68 , 073007 ( 2003 ) , arxiv : hep - ph/0307178 .
p. huber , m. lindner and w. winter , comput .
commun . 167,195 ( 2005 ) ,
de / lin / globes/ ; arxiv : hep - ph/0407333 .
p. huber , j. kopp , m. lindner , m. rolinec and w. winter comput .
177 , 432 ( 2007 ) , arxiv : hep - ph/07071187 .
m. c. gonzalez - garcia , m. maltoni , t. schwetz , arxiv : hep - ph/1512.06856v1 .
anindya datta , raj gandhi , poonam mehta and s uma sankar , phys .
b 597 ( 2004 ) 356 , arxiv : hep - ph/0312027v2(2004 ) j. n. bahcall , v. barger and d. marfatia , phys .
b 534 , 120 ( 2002 ) , arxiv : hep - ph/0201211 . v. d. barger , s. pakvasa , t. j. weiler and k. whisnant , phys . rev . lett . 85 , 5055 ( 2000 ) , arxiv : hep - ph/0005197 .
a. dighe and s. ray , phys .
d 78 , 036002 ( 2008 ) , arxiv : hep - ph/0802.0121 .
a. samanta , phys .
b 693 , 296 ( 2010 ) , arxiv : hep - ph/1005.4851 . m. c. gonzalez - garcia and m. maltoni , phys .
d 70 , 033010 ( 2004 ) , arxiv : hep - ph/0404085 .
magnus jacobson and tommy ohlsson , phys rev .
d 69 , 013003 , arxiv : hep - ph/0305064v3 sujata diwakar , jyotsna singh , yogita pant , r.b .
singh , `` extrinsic cpt violation with sterile neutrino'',volume : 05 issue : 04 , submitted to esatjournals.net/ijret study of low - energy neutrino factory at the fermilab to dusel baseline , inspirehep - net / record/829747/files / fermilab - fn-0836-apc.pdf the study of a european neutrino factory complex , cern / ps/2002 - 080(pp ) | one of the authors sujata diwakar is thankful to university grant commission , india for giving financial support under the rajiv gandhi national fellowship scheme .
* @xmath0 , @xmath1 , @xmath2 , @xmath3 * @xmath4department of physics , university of lucknow , lucknow 226007 india 1.2 cm * abstract * post lsnd results , sterile neutrinos have drawn attention and motivated the high energy physics , astronomy and cosmology to probe physics beyond the standard model considering minimal 3 + 1 ( 3 active and 1 sterile ) to 3+n neutrino schemes . the analytical equations for neutrino conversion probabilities
are developed in this work for 3 + 1 neutrino scheme . here ,
we have tried to explore the possible signals of t and cpt violations with four flavor neutrino scheme at neutrino factory .
values of sterile parameters considered in this analysis are taken from two different types of neutrino experiments viz .
long baseline experiments and reactor+atmospheric experiments . in this work golden and discovery channels are selected for the investigation of t violation .
while observing t violation we stipulate that neutrino factory working at 50 gev energy have the potential to observe the t violation signatures for the considered range of baselines(3000 km-7500 km ) .
the ability of neutrino factory for constraining cpt violation is enhanced with increase in energy for normal neutrino mass hierarchy(nh ) .
neutrino factory with the exposure time of 500 kt - yr will be able to capture cpt violation with @xmath5 gev at 3@xmath6 level for nh and for ih with @xmath7 gev at 3@xmath6 level .
* _ keywords : _ * sterile neutrinos , t violation , cpt violation , neutrino factory |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.