Constraints on Thermal X-ray Radiation from SAX J1808.4-3658 and Implications for Neutron Star Neutrino Emission

0.13 MB·English

by  C. O. Heinke

DRAFTVERSIONFEBRUARY5,2008
PreprinttypesetusingLATEXstyleemulateapjv.10/09/06
CONSTRAINTSONTHERMALX-RAYRADIATIONFROMSAXJ1808.4–3658ANDIMPLICATIONSFORNEUTRON
STARNEUTRINOEMISSION1
C.O.HEINKE2,3,P.G.JONKER4,5,6,R.WIJNANDS7,ANDR.E.TAAM2
DraftversionFebruary5,2008
ABSTRACT
ThermalX-ray radiation from neutron star soft X-ray transients in quiescence providesthe strongest con-
straints on the cooling rates of neutron stars, and thus on the interior composition and properties of matter
inthe coresofneutronstars. We analyzenew(2006)andarchival(2001)XMM-Newtonobservationsofthe
accretingmillisecond pulsarSAX J1808.4–3658in quiescence, which providethe most stringentconstraints
7
todate. TheX-rayspectrumofSAXJ1808.4–3658inthe2006observationisconsistentwithapower-lawof
0
0 photonindex1.83±0.17,withoutrequiringthepresenceofa blackbody-likecomponentfroma neutronstar
2 atmosphere. Our2006observationshowsaslightlylower0.5–10keVX-rayluminosity,atalevelof68+- 1153%
thatinferredfromthe2001observation.SimultaneousfittingofallavailableXMMdataallowsaconstrainton
n thequiescentneutronstar(0.01–10keV)luminosityofL <1.1×1031ergs- 1. Thislimitexcludessomecur-
a NS
rentmodelsofneutrinoemissionmediatedbypioncondensates,andprovidesfurtherevidenceforadditional
J
coolingprocesses,suchasneutrinoemissionviadirectUrcaprocessesinvolvingnucleonsand/orhyperons,in
4
thecoresofmassiveneutronstars.
2
Subjectheadings:binaries: X-rays—densematter—neutrinos—stars: neutron

2
v
1. INTRODUCTION of the lowest quiescent thermal luminosities yet measured
2
fromanyaccretingNS(Campanaetal.2002;Wijnandsetal.
3 The X-ray transient SAX J1808.4–3658 (hereafter 1808)
2002), along with 1H1905+000 (Jonkeretal. 2006). Tran-
2 has provided many fundamental breakthroughs in the study
2 of accreting neutron stars (NSs). It was discovered in 1996 siently accreting NSs in quiescence are usually seen to have
1 byBeppoSAX’sWideFieldCameras,andtypeIX-raybursts soft, blackbody-like X-ray spectra, often accompanied by
6 wereseen,identifyingitasanaccretingNSandconstraining a harder X-ray component generally fit by a power-law of
0 photon index 1–2 (Campanaetal. 1998). The harder com-
the distance (in’tZandetal. 1998; Galloway&Cumming
/ ponent is of unknown origin; an effect of continued accre-
h 2006). Coherent millisecond X-ray pulsations, the first
tion, or a shock from a pulsar wind have been suggested
p discovered in accreting systems, were identified during an
(Campanaetal. 1998). The blackbody-like component is
- outburst using the Rossi X-ray Timing Explorer (RXTE;
o generallyunderstoodastheradiationofheatfromtheNSsur-
Wijnands&vanderKlis 1998). Burst oscillations have also
r face. This heat is produced by deep crustal heating during
t been seen at 1808’s 401 Hz spin frequency, confirming that
s accretion, and is radiated by the crust on a timescale of 104
thermonuclear burst oscillations in low-mass X-ray binaries
a years, producing a steady quiescent thermal NS luminosity
: (LMXBs)representthespinperiodoftheNS(in’tZandetal.
v 2001;Chakrabartyetal.2003). Apairofkilohertzquasiperi- (Brownetal. 1998; Campanaetal. 1998; Haensel&Zdunik
i 1990). Measurementoftheblackbody-likecomponentispar-
X odic oscillations (QPOs) were seen from 1808, with a fre-
ticularlyimportantbecauseitindirectlyconstrainstheinterior
quency difference equal to one-half of the spin period,
r structureofNSs.
a forcing a revision of the most popular models for QPOs
WhatfractionofthisheatescapestheNSasneutrinosrather
(Wijnandsetal. 2003). Optical observations of the brown
than photons depends on the physical conditions (i.e. com-
dwarfcompanionwhile1808wasinquiescenceshowedasi-
position, density, and pressure) of the NS interior. If the
nusoidalopticalmodulationattributedtoheatingofthecom-
outburst history (fluence, recurrence time) and distance of
panion(Homeretal.2001).However,therequiredirradiating
a NS are reasonably well-known, then the determination of
luminosityislargerthantheavailableX-rayluminosity,giv-
the quiescent thermal NS luminosity constrains the neutrino
ingrisetospeculationthataradiopulsarmechanismisactive
vs. photon emission, and thus models for the NS interior
duringquiescence(Burderietal.2003;Campanaetal.2004).
(Yakovlev&Pethick 2004; Levenfish&Haensel 2006). For
In addition to these discoveries, 1808 has provided one
example, the transient Cen X-4 has been identified as hav-
1Based on observations obtained with XMM-Newton, an ESA science ingaratherlowquiescentX-rayluminositycomparedtodeep
missionwithinstrumentsandcontributionsdirectlyfundedbyESAMember crustal heating predictions. This suggests that Cen X-4 has
StatesandNASA enhanced neutrino emission, produced in the high density
2NorthwesternUniversity,Dept.ofPhysics&Astronomy,2145Sheridan core of a relatively high mass NS (Colpietal. 2001). Many
Rd.,Evanston,IL60208;[email protected]
3LindheimerPostdoctoralFellow otherLMXBs have also shown low quiescentthermalX-ray
4SRON, Netherlands Institute for Space Research, Sorbonnelaan 2, luminosities,indicatingeitherenhancedneutrinoemissionor
3584CA,Utrecht,theNetherlands extremelylongquiescentintervals(e.g.Wijnandsetal.2001;
5Harvard–SmithsonianCenterforAstrophysics,60GardenStreet,Cam-
Jonkeretal. 2004b; Tomsicketal. 2004; Jonkeretal. 2006).
bridge,MA02138,Massachusetts,U.S.A.
6Astronomical Institute, Utrecht University, PO Box 80000, 3508 TA, Thecoolestoftheseprovidethestrongestconstraintstodate
Utrecht,theNetherlands onneutrinocoolingfromNScores,asabroaderrangeofcool-
7Astronomical Institute "Anton Pannekoek", University of Amsterdam, ingratesisnecessarytoexplainthedatathanforyoungcool-
Kruislaan403,1098SJ,TheNetherlands ingpulsars(Pageetal.2004;Yakovlev&Pethick2004).
2 Heinkeetal.
1808 is a particularly interesting system due to its
known distance and constrained mass transfer rate
(Bildsten&Chakrabarty 2001). Campanaetal. (2002)
observed 1808 in quiescence with XMM in 2001, finding a
klopwc)luanmdinaorseiltayti(vLeXl(y0h.5a-r1d0spkeeVct)r=um5,×w1i0th31leesrsgsthsa- n1,1f0o%rdo=f2th.5e V−1 10−3
e
k
X-ray flux attributable to a possible blackbody-like compo- s −1
nent. We haveobtaineda deeperXMM observationof1808 unts 10−4
in quiescencein 2006to place more stringentconstraintson o
c
neutronstarcoolingprocesses. ed
z
ali
m
2. DATAREDUCTION nor10−5
We observed 1808 on September 14, 2006
1
(obs-id0400230401) for 54 ksec with XMM’s EPIC
2 0
camera, using two MOS CCD detectors (Turneretal. 2001) cS
with medium filters and one pn CCD detector (Strüderetal. D −1
2001) with a thin filter. We also downloaded the March
0.2 0.5 1 2 5 10
24, 2001 XMM observation (obs-id0064940101, reported
Energy (keV)
by Campanaetal. 2002) from the HEASARC archive. All
data were reduced using FTOOLS and SAS version 7.0.0. FIG. 1.— Toppanel: XMMX-rayspectra(dataandbest-fitpower-law
We used only the MOS data from 2001, since the pn data model)ofSAXJ1808.4–3658.Solidlinewithcrosses:2006pndata,model.
Dashedlinewithtriangles: 2006MOSdata,model. Solidlinewithcircles:
weretakenintimingmode,andthetargetwastoofainttobe
2001MOSdata,model.Bottompanel:residualstothefitinunitsofχ2.
detected in this mode. Intervals of flaring backgroundwere
excludedby excludingtimes when the totalMOS countrate eV(90%confidence8)ispermitted,thusplacingalimitonthe
exceeded 5 0.2–12 keV counts per second, and times when NS’s thermal0.01–10keV (essentially bolometric)luminos-
the total pn count rate exceeded 50 0.2–12 keV counts per ityofL <2.4×1031ergs- 1. Theinclinationofthissystem
NS
second. Thisleft36.6,47.0,48.1,and39.3ksecinthe2001 is knownto be low (Bildsten&Chakrabarty2001), andit is
MOS data, the 2006 MOS1 data, the 2006 MOS2 data, and rare for LMXBs to show higher N in quiescence than out-
H
the 2006 pn data, respectively. Event grades higher than 12 burst (e.g. Jonker&Nelemans 2004), so we regarda higher
were also excluded. We extracted spectra from a 10′′ circle N asveryunlikely. Thetotal0.5–10keVunabsorbedlumi-
H
around the position of 1808, correcting the fluxes for the nosityisL =7.6+1.7×1031ergs- 1.
X - 1.5
fraction of photons collected within the extraction radius, For the 2006 data, we find a similar spectral shape, and
and combining the pairs of simultaneous MOS spectra and therefore fit similar models to the pn and MOS data simul-
responsesusingFTOOLS.We generatedresponseandeffec- taneously. Asimplepower-lawfitsthedataadequately,with
tive area files using the SAS tasks rmfgen and arfgen, and aphotonindexofΓ=1.83±0.17andanunabsorbedL (0.5–
X
produced background spectra from 90′′ circular source-free 10keV)=5.2±0.7×1031ergs- 1. The2006fluxappearsless
regions on the same CCD. The spectra were grouped to than the 2001flux. We test this by fitting the spectra simul-
>15 counts per bin for the MOS data, and >30 counts per taneously and tying their power-law slopes together, finding
bin for the pn data (other choices gave similar results). To that the 2001 0.5–10 keV unabsorbed flux is higher at 97%
assess variability within the 2006 and 2001 observations, confidence;the2001fluxis1.28+0.24 (90%conf.) thatofthe
- 0.21
background-subtracted lightcurves were produced within 2006observation.Ifthepower-lawslopesareallowedtovary
SAS and analyzed using HEASARC’s XRONOS software. (1.61±0.3for2001, 1.83±0.17for2006), thenthe best-fit
KS and χ2 tests on the last 15 ksec of 0.2–12 keV pn data fluxratiois1.47+0.35.
- 0.27
(unaffected by background flaring) revealed no evidence of Fitting the 2006 and 2001 data allows a tighter constraint
variability. onthepresenceofaNSatmospherecomponentthanthe2001
data alone, requiringa NS kT <34 eV and a thermal0.01–
3. SPECTRALANALYSIS 10 keV NS luminosity L <1.1×1031 erg s- 1. Choosing
NS
Our fitting includes photoelectric absorption (XSPEC a NS radius of 12 km, or a mass of 2.0 M , varies this
⊙
model phabs), with a hydrogen column density, N , fixed constraint by only 3%. The rather tight distance limits of
H
at the value derived from observations in outburst (1.3× Galloway&Cumming (2006) (3.5±0.1 kpc) produce only
1021 cm- 2; note this is equal to the value derived from a 6% uncertainty. Allowing the N to float freely permits a
H
Dickey&Lockman1990). We also testedmodelswith pho- thermal 0.01–10 keV NS luminosity L <1.0×1032 ergs
NS
toelectric absorptionas a free parameter,in all cases finding s- 1(forN =1.7×1021cm- 2).
H
N consistent with the outburst value. Quoted errors are at
H
90%confidence. 4. RAMIFICATIONS
We fit the combined 2001 MOS spectrum to a power- Wehaveestimatedthetime-averagedmasstransferratesfor
law model, finding Γ = 1.72±0.28 (see Table 1). Fits 1808, and severalother transient LMXBs (Aql X-1, Cen X-
using only a hydrogen-atmosphere model, the NSATMOS 4, 4U 1608–52, KS1731–260, RX 1709–2639,MXB 1659–
modelofHeinkeetal.(2006a)(similartotheNSAmodelof 29, XTE 2123–058,SAX 1810.8–2609,and those in Terzan
Zavlinetal. 1996) gave poor fits (χ2ν >4.8). We then per- 5 and NGC 6440), fromthe RXTE All-SkyMonitor (ASM)
formedfitswithNSATMOSplusapower-law,fixingthetrue record (1996 to Nov. 2006) under the assumption that the
NSradiusto10km,thegravitationalmassto1.4M⊙,andthe time-averaged mass accretion rate over the last 10 years re-
distanceto3.5kpc(Galloway&Cumming2006).Nothermal
componentisrequired,butathermalcomponentwithkT <42 8kT istheredshiftedtemperature,orkT∞
ThermalX-rayRadiationfromSAXJ1808.4-3658 3
(Bildsten&Chakrabarty 2001). We note that the true mass
transferrate cannotbe lessthan7×10- 12 M /yearfora NS
⊙
mass ≥1.4 M⊙ under the assumption of an index n=- 1/3
forthe donor’smass-radiusrelation, orless than3.5×10- 12
M /year for an index n= 1 mass-radius relation. It is un-
⊙
clearwhethertheentropyofthedonorcanbemaintainedby
thelowquiescentluminosityoftheneutronstar(Homeretal.
2001; Burderietal. 2003); we plan future observations and
modelingtoaddressthisissue.
We have plotted the cooling curves calculated by
Yakovlev&Pethick(2004)foravarietyofmodelsinFig. 2.
Low-massNSswillcoolslowly(dottedlineinFig.2);inthe
model of Yakovlev&Pethick (2004), only through photon
emission and neutron-neutron neutrino bremsstrahlung pro-
cesses, whilemodifiedUrca neutrinoemissionis suppressed
by proton superfluidity. Other slow cooling models (invok-
ing, e.g., neutrino emission through Cooper pair formation)
givesimilarresults(e.g.Pageetal.2004). Higher-massNSs
should have higher central densities, sufficient to promote
more rapid direct Urca neutrino cooling processes involv-
ing nucleons and/or hyperons, or direct Urca-like processes
mediated by pions, kaons, or quark matter, in their cores.
Medium-massmodelscanproduceintermediatecoolingrates
if proton superconductivity is important at low densities, as
FIG. 2.— Cooling curves for various NS interior neutrino emission itsdecayatmoderatedensitiescanallowasmoothtransition
scenarios, compared with measurements (or 95% conf. upper limits) of between fast and slow cooling rates (Yakovlevetal. 2003;
the quiescent 0.01–10 keV NS luminosity and time-averaged mass trans- Levenfish&Haensel 2006). Thus NSs of different masses
ferrateforseveralNStransients (seeTable2). Coolingcurvestakenfrom
shouldliebetweenthetopcurveandoneofthelowercurves
Yakovlev&Pethick (2004); the dotted curve represents a low-mass NS,
whilethelowercurvesrepresenthigh-massNSswithkaonorpionconden- in Fig. 2, where the lower curve is the maximum neutrino
sates, ordirectUrca(Durca)processesmediated bynucleons orhyperons. cooling curve. We note a possible trend that NSs with low
LimitsonthequiescentNSluminosityofSAXJ1808.4–3658aregivenfor mass transfer rates seem to have particularly low quiescent
the2001and2006observations. Theeffectofadistanceerroraslargeasa
luminosities, well below the “standard cooling” predictions.
factor1.5isalsoindicated(upperleft).
Thismightbe explainedthroughbinaryevolution;NSs with
lowmasstransferratesmaybeveryoldsystems,whichmay
flectsthetime-averagedmasstransferrate(Table2). Weuse
have accreted significant mass. In enhanced neutrino emis-
PIMMS and a power-law of photon index 2 to convert the
ASM countrates during outbursts into 0.1–20 keV fluxes9. sion scenarios, these massive NSs would then have higher
neutrinoandlowerphotonluminositiesthanyoungersystems.
This is, of course, a rough approximation, as the spectral
Our constraint on the quiescent thermal 0.01–10 keV NS
shapes of LMXBs in outburst vary substantially. Additional
luminosity of 1808 from the 2006 observations thus seems
sources of potential error include poor ASM time coverage
to rule out some models of direct Urca neutrino emission
ofsomeoutbursts,uncertaintyintheNSmassandradius(af-
via pion condensates, favoringdirectUrca processesinvolv-
fecting the energy released per accreted gram, and thus the
ing nucleons and/or hyperons. An extremely large distance
conversionfromL tomassaccretionrate),variabilityinthe
X
uncertainty of 50% (see Fig. 2; cf. Galloway&Cumming
masstransferrate,anduncertaindistances(whichwillequally
2006) would be required to bring 1808’s thermal luminos-
affectthe quiescentluminosity). We plotanarbitraryuncer-
ity up to the pion condensate predictions. The 2001 obser-
tainty of 50% in both mass transfer rate and quiescent lu-
vation, by itself, rules out some models of direct Urca neu-
minosity for each point in Fig. 2. For Cen X-4 we use the
trinoemissionfromkaoncondensates.OthermodelersofNS
lowestmeasuredquiescentluminosity,andthe mass transfer
cooling have suggested that medium effects (Blaschkeetal.
rate limit inferred if Cen X-4 undergoes outbursts every 40
2004) or diquark condensates (Grigorianetal. 2005) could
years with a fluence similar to its 1969 outburst(Chenetal.
provide a wide range of NS cooling rates. These models
1997). The NS component flux for Aquila X-1 is some-
may also be sufficientto explainthe data on 1808presented
what uncertain and possibly variable (Rutledgeetal. 2002;
here.Ourresultsagreewiththeprincipalconclusionsof,e.g.,
Campana&Stella 2003). We assumethatalloutburstsfrom
Yakovlev&Pethick(2004),Levenfish&Haensel(2006),and
NGC6440since1971havebeendetected. ForKS1731-260
Pageetal.(2006),andprovideafirmerobservationalbasisfor
we assume thatthe averageflux seen with RXTE-ASMdur-
futurestudies.
ingoutburstwas theaverageflux duringthe entire12.5year
outburst. ForKS1731–260andthetransientinTerzan1(for
whichwetakea12-yearoutburst)wetakeaminimumrecur-
WethankC.J.Deloye,E.F.Brown,andA.W.Steinerfor
rencetimeof30years.
useful discussions, and the referee for a rapid and construc-
For 1808 we derive a time-averaged mass transfer rate of
1.0×10- 11 M⊙/year, an excellent match to the prediction tivereport. RXTEASMresultsprovidedbytheASM/RXTE
of general relativity of 0.95×10- 11(M2/0.05 M⊙) M⊙/year tpeoarmtfsroamt MthITe LaninddhNeAimSAer’sPGosStdFoCc.toCrOalHFealclokwnoswhilpedagteNsosrutph--
9 Wehaveverifiedthatthisconversioniscorrecttowithin50%forout- westernUniversity,andNASAXMMgrantNNX06AH62G.
burstsofthetransientsEXO1745-245andAquilaX-1. PGJ acknowledges support from the Netherlands Organiza-
4 Heinkeetal.
tionforScientificResearch.
REFERENCES
Bildsten,L.,&Chakrabarty,D.2001, ApJ,557,292 Jonker,P.G.,Galloway,D.K.,McClintock,J.E.,Buxton,M.,Garcia,M.,&
Blaschke,D.,Grigorian,H.,&Voskresensky,D.N.2004, A&A,424,979 Murray,S.2004a,MNRAS,354,666
Brown,E.F.,Bildsten,L.,&Rutledge,R.E.1998,ApJ,504,L95 Jonker,P.G.,&Nelemans,G.2004,MNRAS,354,355
Burderi, L., Di Salvo, T., D’Antona, F., Robba, N. R., & Testa, V. 2003, Jonker,P.G.,Wijnands,R.,&vanderKlis,M.2004b,MNRAS,349,94
A&A,404,L43 Levenfish,K.P.,&Haensel,P.2006,astro-ph/0611115
Cackett,E.M.etal.2005, ApJ,620,922 Page,D.,Geppert,U.,&Weber,F.2006,NuclearPhysicsA,777,497
Cackett,E.M.etal.2006a,MNRAS,369,407 Page,D.,Lattimer, J.M.,Prakash,M.,&Steiner, A.W.2004,ApJSupp,
Cackett,E.M.,Wijnands,R.,Linares,M.,Miller,J.M.,Homan,J.,&Lewin, 155,623
W.H.G.2006b,MNRAS,372,479 Rutledge, R.E.,Bildsten, L.,Brown,E.F.,Pavlov,G.G.,&Zavlin,V.E.
Campana, S., Colpi, M., Mereghetti, S., Stella, L., & Tavani, M. 1998, 1999, ApJ,514,945
A&ARev.,8,279 —.2001a, ApJ,559,1054
Campana,S.etal.2004,ApJ,614,L49 —.2001b, ApJ,551,921
Campana,S.,&Stella,L.2003, ApJ,597,474 —.2002, ApJ,577,346
Campana,S.etal.2002,ApJ,575,L15 Strüder,L.etal.2001, A&A,365,L18
Chakrabarty, D.,Morgan,E.H.,Muno,M.P.,Galloway, D.K.,Wijnands, Tomsick,J.A.,Gelino,D.M.,Halpern,J.P.,&Kaaret,P.2004, ApJ,610,
R.,vanderKlis,M.,&Markwardt,C.B.2003,Nature,424,42 933
Chen,W.,Shrader,C.R.,&Livio,M.1997, ApJ,491,312 Tomsick,J.A.,Gelino,D.M.,&Kaaret,P.2005, ApJ,635,1233
Colpi,M.,Geppert,U.,Page,D.,&Possenti,A.2001,ApJ,548,L175 Turner,M.J.L.etal.2001, A&A,365,L27
Dickey,J.M.,&Lockman,F.J.1990,ARAA,28,215 Wijnands,R.etal.2005, ApJ,618,883
Galloway,D.,&Cumming,A.2006, ApJ,652,559 Wijnands,R.,Kuiper,L.,Zand,J.i.,Dotani,T.,vanderKlis,M.,&Heise,J.
Grigorian, H., Blaschke, D., & Voskresensky, D. 2005, Phys. Rev. C, 71, 2002, ApJ,571,429
045801 Wijnands,R.,Miller,J.M.,Markwardt,C.,Lewin,W.H.G.,&vanderKlis,
Haensel,P.,&Zdunik,J.L.1990, A&A,227,431 M.2001,ApJ,560,L159
Heinke,C.O.,Rybicki,G.B.,Narayan,R.,&Grindlay,J.E.2006a, ApJ, Wijnands,R.,&vanderKlis,M.1998,Nature,394,344
644,1090 Wijnands, R., van der Klis, M., Homan, J., Chakrabarty, D., Markwardt,
Heinke, C.O.,Wijnands, R.,Cohn, H.N.,Lugger, P.M.,Grindlay, J.E., C.B.,&Morgan,E.H.2003,Nature,424,44
Pooley,D.,&Lewin,W.H.G.2006b,ApJ,651,1098 Yakovlev,D.G.,Levenfish,K.P.,&Haensel,P.2003, A&A,407,265
Homer,L.,Deutsch,E.W.,Anderson,S.F.,&Margon,B.2001,AJ,122, Yakovlev,D.G.,&Pethick,C.J.2004,ARAA,42,169
2627 Zavlin,V.E.,Pavlov,G.G.,&Shibanov,Y.A.1996, A&A,315,141
in’tZand,J.J.M.,Verbunt,F.,Heise,J.,&etal.1998, A&A,329,L37
in’tZand,J.J.M.etal.2001, A&A,372,916
Jonker, P. G., Bassa, C. G., Nelemans, G., Juett, A. M., Brown, E. F., &
Chakrabarty,D.2006,MNRAS,368,1803
ThermalX-rayRadiationfromSAXJ1808.4-3658 5
TABLE1
SPECTRALFITSTOSAXJ1808.4–3658
Epoch NH Γ χ2ν/dof LX kT LNS
(1022cm- 2) (ergs- 1) (eV) (ergs- 1)
2001 (0.13) 1.61±0.3 0.51/9 7.6+- 11..75×1031 <42 <2.4×1031
2006 (0.13) 1.83±0.17 0.86/45 5.2±0.7×1031 <35 <1.2×1031
2001&2006 (0.13) 1.83±0.16 0.79/55 5.2±0.7×1031 <34 <1.1×1031
2001&2006 0.15±0.04 1.93+- 00..3279 0.78/54 5.2±1.0×1031 <61 <1.0×1032
NOTE.—Spectralfitswithpower-lawplusNSATMOSmodeltoSAXJ1808.4–3658. Errorsare90%confidenceforasingleparameter. NH is
heldfixedinthefirstthreerows.LXfor0.5–10keVrange,LNSfor0.01–10keV.
TABLE2
LUMINOSITIESANDMASSTRANSFERRATES
Source NH kT D Outbursts Years M˙ LNS Refs
(1022cm- 2) (eV) (kpc) (M⊙yr- 1) (ergs- 1)
AqlX-1 4.2×1021 ∼94 5 8 10.7 4×10- 10 5.3×1033 1,2,3,4
CenX-4 5.5×1020 76 1.2 - - <3.3×10- 11 4.8×1032 5,3
4U1608–522 8×1021 170 3.6 4 10.7 3.6×10- 10 5.3×1033 6,3,4
KS1731–260 1.3×1022 70 7 1 30 <1.5×10- 9 5×1032 7,4
MXB1659–29 2.0×1021 55 ∼10? 2 10.7 1.7×10- 10 2.0×1032 7,4
EXO1747–214 4×1021 <63 <11 - - <3×10- 11 <7×1031 8
Terzan5 1.2×1022 <131 8.7 2 10.7 3×10- 10 <2.1×1033 9,10,4
NGC6440 7×1021 87 8.5 3 35 1.8×10- 10 3.4×1032 11,4
Terzan1 1.4×1022 74 5.2 - - <1.5×10- 10 <1.1×1033 12
XTE2123–058 6×1020 <66 8.5 1 10.7 <2.3×10- 11 <1.4×1032 3,4
SAXJ1810.8–2609 3.3×1021 <72 4.9 1 10.7 <1.5×10- 11 <2.0×1032 13,3,4
RXJ1709–2639 4.4×1021 122 8.8 2 10.7 1.8×10- 10 2.2×1033 14,15,4
1H1905+000 1.9×1021 <50 10 - - <1.1×10- 10 <4.8×1031 16,15
SAXJ1808.4–3658 1.3×1021 <34 3.5 5 10.7 1.0×10- 11 <1.1×1031 17,4,15
NOTE. —Estimatesofquiescentthermalluminositiesfromneutronstartransients,andmasstransferrates(inferredfromRXTEASMobservationsforsystemswith
RXTE-eraoutbursts). QuiescentthermalluminositiesarecomputedfortheunabsorbedNScomponentinthe0.01-10keVrange. Outburstsandyearscolumnsgivethe
numberofoutburstsandthetimebaselineusedtocomputeM˙,ifthiscalculationwasperformedinthiswork(indicatedbyreferringtoreference4).Referencesasfollows:
1:Rutledgeetal.(2001a),2:Campana&Stella(2003),3:Tomsicketal.(2004),4:Masstransferratecomputedinthiswork,5:Rutledgeetal.(2001b),6:Rutledgeetal.
(1999),7: Cackettetal.(2006b),8: Tomsicketal.(2005),9: Wijnandsetal.(2005),10: Heinkeetal.(2006b),11: Cackettetal.(2005),12: Cackettetal.(2006a),13:
Jonkeretal.(2004b),14:Jonkeretal.(2004a),15:Quiescentbolometricluminositycomputedinthiswork,16:Jonkeretal.(2006),17:Galloway&Cumming(2006).