An Extreme X-ray Disk Wind in the Black Hole Candidate IGR J17091-3624

0.32 MB·English

by  Ashley L. King

Accepted to ApJLJanuary10,2012
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
AN EXTREME X-RAY DISK WIND IN THE BLACK HOLE CANDIDATE IGR J17091−3624
A. L. King1, J. M. Miller1, J. Raymond2, A. C. Fabian3, C. S. Reynolds4, T. .R. Kallman5, D. Maitra1,
E. M. Cackett3,6, M. P. Rupen7
Accepted to ApJL January 10, 2012
ABSTRACT
Chandra spectroscopy of transient stellar-mass black holes in outburst has clearly revealed accretion disk winds in
2
soft, disk–dominated states, in apparent anti-correlation with relativistic jets in low/hard states. These disk winds
1 are observed to be highly ionized, dense, and to have typical velocities of ∼1000 km/s or less projected along our
0
line of sight. Here, we present an analysis of two Chandra High Energy Transmission Grating spectra of the Galactic
2 blackholecandidateIGRJ17091−3624andcontemporaneousEVLAradioobservations,obtainedin2011. Thesecond
n Chandra observationrevealsanabsorptionline at6.91±0.01keV;associatingthis line with He-likeFe XXV requiresa
a blue-shift of 9300+500 km/s (0.03c, or the escape velocity at 1000 R ). This projected outflow velocity is an order
J −400 Schw
ofmagnitudehigherthanhaspreviouslybeenobservedinstellar-massblackholes,andisbroadlyconsistentwithsome

0 ofthe fastestwindsdetectedinactivegalacticnuclei. Apotentialfeatureat7.32keV,ifdue toFeXXVI, wouldimply
1 a velocity of ∼14600 km/s (0.05c), but this putative feature is marginal. Photoionizationmodeling suggests that the
accretion disk wind in IGR J17091−3624may originate within 43,300 Schwarzschild radii of the black hole, and may
] be expelling more gasthan accretes. The contemporaneousEVLA observationsstrongly indicate that jet activity was
E
indeed quenched at the time of our Chandra observations. We discuss the results in the context of disk winds, jets,
H
and basic accretion disk physics in accreting black hole systems.
h. Subject headings: X-rays: binaries — accretion, accretion disks — black hole physics
p
-
o 1. INTRODUCTION tance because He-like Fe XXV and H-like Fe XXVI
r lines can endure in extremely hot, ionized gas (see, e.g.
t A detailed observational account of how black hole
s accretion disks drive winds and jets remains elusive, Bautista & Kallman2001), andtherefore tracethe wind
a regionclosesttowhereitislaunchedneartheblackhole.
but the combination of high resolution X-ray spec-
[
Studies of some stellar-mass black hole disk winds find
troscopy, improved radio sensitivity, and comparisons
that the gas is too ionized, too dense, and originatestoo
2 across the black hole mass scale hold great potential.
v The broad range in X-ray luminosity covered by tran- close to the black hole to be expelled by radiative pres-
8 sient stellar-mass black holes makes it possible to trace sureorbythermalpressurefromComptonheatingofthe
4 major changes in the accretion flow as a function of disk, requiring magnetic pressure (Miller et al. 2006a,b;
6 the inferred mass accretion rate; this is largely impos- Kubota et al. 2007). Winds that may originate close
3 to the black hole and carry high mass fluxes are also
sible in supermassive black holes. Disk winds and jets,
. observed in AGN (e.g., Kaspi et al. 2002; Chartas et al.
2 for instance,appear to be state-dependentandmutually
1 exclusive in sources such as H 1743−322 (Miller et al. 2002; King et al. 2011b; Tombesi et al. 2010).
1 2006a; Blum et al. 2010), GRO J1655−40 (Miller et al. InthisLetter,wepresentevidenceofaparticularlyfast
diskwindintheblackholecandidateIGRJ17091−3624.
1 2008;Luketic et al.2010;Kallman et al.2009),andGRS
: 1915+105(Miller et al. 2008; Neilsen & Lee 2009). This The current outburst of IGR J17091−3624 was first re-
v ported on 2011 January 28 (Krimm et al. 2011). Our
mayofferinsightsintowhymanySeyfertAGN,whichare
i observations caught IGR J17091−3624 in the high/soft
X wellknownfortheirdiskwinds,aretypicallyradio–quiet
state, but it is important to note that the source has
r (though not necessarily devoid of jets; see King et al.
a 2011a; Jones et al. 2011; Giroletti & Panessa 2009). also showed low/hard state episodes with flaring and
apparent jet activity in radio bands (Rodriguez et al.
The proximity of Galactic black hole binaries (BHB)
2011). X-ray flux variations in IGR J17091−3264 bear
ensures a high flux level and spectra with excellent sen-
similarities to the microquasar GRS 1915+105 (e.g.,
sitivity in the Fe K band. This is of prime impor-
Altamirano et al. 2011).
1Department of Astronomy, University of Michigan, 500 2. OBSERVATIONANDDATAREDUCTION
Church Street, Ann Arbor, MI 48109-1042, USA, ashk- IGR J17091−3624 was first observed with Chandra
[email protected]
2Smithsonian Astrophysical Observatory, 60 Garden Street, on 2011 August 1 (ObsID 12405), starting at 06:59:16
Cambridge,MA,02138, USA (UT), for a total of 30 ksec. The High Energy Trans-
3InstituteofAstronomy,UniversityofCambridge,Madingley missionGratings(HETG)wereusedtodispersetheinci-
Road,Cambridge,CB3OHA,UK
4Department of Astronomy, Universityof Maryland, College dentfluxontotheAdvancedCCDImagingSpectrometer
Park,MD,20742, USA spectroscopicarray(ACIS-S).Topreventphotonpile-up,
5Laboratory for High Energy Astrophysics, NASA Goddard the ACIS-S array was operated in continuous clocking
SpaceFlightCenter,Code662,Greenbelt, MD20771,USA or “GRADED CC” mode, which reduced the nominal
6WayneStateUniversity,DepartmentofPhysicsandAstron-
frame time from 3.2 seconds to 2.85 msec. The zeroth
omy,Detroit,MI,48201,USA
7NationalRadioAstronomyObservatory,Socorro,NM87801, orderfluxisincidentontheS3chip,andframesfromthis
USA chip can be lost from the telemetry stream if a source
2 A. L. King et al.
V)−1 V)−1
e e
s k−1 1 s k−1 1
m −2 m −2
c c
s s
n n
oto 0.5 oto 0.5
h h
P P
V (2 V (2
e e
k k
1.4 1.4
1.2 1.2
o o
rati 1 rati 1
0.8 0.8
5 6 7 5 6 7
Energy (keV) Energy (keV)
Fig. 1.— The second Chandra/HETG spectrum of IGR Fig.2.— The second Chandra/HETG spectrum of IGR
J17019−3624 is shown above, fit with a simple disk blackbody J17019−3624isshownabove,fitwithasimplediskblackbodyplus
plus power-law continuum. The continuum fit excluded the Fe K power-law continuum. A self-consistent photoionization model,
bandtopreventbeingbiasedbylinefeatures. Thelineat6.91keV generatedusingXSTAR,wasusedtomodeltheabsorptioninthe
is clearly apparent in the data/model ratio. Associating this line FeKband. Thedatawerebinnedforvisualclarity.
withHe-likeFeXXVimpliesanoutflowvelocityof9300+−540000km/s. N =7.6×1021cm−2; Dickey & Lockman1990). Dueto
H
Weak evidence of alineat 7.32 keV, plausiblyassociated withFe
this high column that predominantly affects lower ener-
XXVI,wouldimplyanevenhigheroutflowvelocity. Thedatawere
binnedforvisualclarity. gies,theMEGspectrahavecomparativelylowsensitivity
is very bright. We therefore used a gray window over as compared to the HEG, and were therefore excluded.
the zeroth order aimpoint; only one in 10 photons were Thelimitations ofthe HEGandourinstrumentalcon-
telemeteredwithinthisregion. Foralongerdiscussionof figuration enforce an effective lower energy bound of
thismode,pleasesee,e.g.,Milleretal.(2006)andMiller 1.3keV.Inthesecondobservation,theinstrumentalcon-
et al.(2008). The sourcewasobservedfor a secondtime figuration served to enforce an upper limit to the fitting
on 2011 October 6, starting at 11:17:02 (UT), again for range of 7.6 keV. This limit was adopted for the first
a total of 30 ksec. The relatively low flux observed dur- observation as well.
ingthefirstobservationindicatedthattheACIS-Sarray
3.1. The Spectral Continuum
couldbeoperatedinthestandard“timedevent”imaging
mode during this second observation. The HEG spectra were fit with a fiducial spectral
Data reduction was accomplished using CIAO ver- modelincludinganeffectiveHcolumndensity(TBabs),a
sion 4.1 (Fruscione et al. 2006). Time-averaged first- diskblackbodycomponent,andapower-lawcomponent.
order High Energy Grating (HEG) and Medium Energy Thefirstobservation(MJD55775)iswelldescribedby
Grating (MEG) spectra were extractedfrom the Level-2 column density of N =9.9±0.1×1021 cm−2, and a disk
H
event file. Redistribution matrix files (rmfs) were gen- blackbodytemperature of1.3±0.1keV.The resultingfit
erated using the tool “mkgrmf”; ancillary response files gaveaχ2/ν =2657/3156=0.84. Thisspectrumisdom-
(arfs) were generated using “mkgarf”. The first-order inated by the disk black body component, typical of the
HEGspectraandresponseswerecombinedusingthetool high soft state of BHB. A power-law continuum compo-
“add grating orders”. The spectrawerethen groupedto nent is not statistically required. An unabsorbed flux of
require a minimum of 10 counts per bin. All spectral F2−10keV =1.5±0.1×10−9ergscm−2 s−1 wasmeasured.
analyses were conducted using XSPEC version 12.6.0. The second observation (MJD 55841) also had a con-
All errors quoted in this paper are 1σ errors. sistent flux, F2−10keV = 1.9 ±0.5 × 10−9 ergs cm−2
Nearly simultaneous radio observations were made s−1. Again, the column density was large, at N =
H
with the EVLA at each Chandra pointing. The first ra- 1.22±0.07×1022 cm−2. A power-law photon index of
dioepochincludedatwohourintegrationat8.4GHz on Γ=1.7+0.07andadiskblackbodytemperatureof2.3±0.3
2011 August 2 (MJD 55776) at 1:01:04 (UT), while the −0.09
keV were measured. This disk temperature is high but
second was a two hour integration at both 8.4 and 4.8
common in GRS 1915+105 (see, e.g., Vierdayanti et al.
GHz on 2011 October 6 (MJD 55841) at 22:10:16 (UT).
2010). The resulting χ2/ν was 2754/3414=0.81.
Thefluxandbandpasscalibratorwas3C286. Thephase
andgaincalibratorswereJ1720-3552andJ1717-3624,for
3.2. The Line Spectra
the firstandsecondobservations,respectively. The data
are reduced using CASA version 3.2.1 (McMullin et al. In the second HEG spectrum, absorption features are
2007). notedintheFeKband(SeeFigure1),andthesewereini-
tially fit with simple Gaussians. The two strongest lines
3. ANALYSISANDRESULTS arefoundatenergiesof6.91±0.01keVand7.32+0.02. Via
−0.06
A black hole mass has not yet been determined for anF-test, (see Protassovet al. 2002, for some cautions),
J17091−3624; a value of 10 M⊙ is assumed throughout theselinesaresignificantatthe 99.94%and99.67%con-
this work. Preliminary fits to the HETG spectra of IGR fidence levels respectively. Dividing the flux normaliza-
J17091−3624suggestedarelativelyhighcolumndensity, tion of each line by its minus-side error suggests that
inkeepingwith valuespredictedfromradiosurveys(e.g. the feature at 6.91 keV is significant at the 4σ level of
IGR J17091−3624 3
confidence, while the 7.32 keV line is marginal at a 2σ also adopted in a second XSTAR grid (L =3.5×1038
ion
confidence level. ergs/s).
We also modeled the second observation continuum Thedensityoftheabsorbingmaterialwaschosentobe
with a Comptonization model (compTT) instead of the log(n)= 12.0. This is a reasonableassumptionbasedon
disk blackbody and power-law. In general, this gave a the modeling of similar X-ray binaries: GX 13+1, n =
reasonable fit at χ2/ν =2884/3414 = 0.84. This model 1013 cm−3(Ueda et al. 2004), GRO J1655-40, n = 1014
also showed residual absorption features at high energy, cm−3(Miller et al. 2008), H1743-322, n = 1012 cm−3
which again we modeled with Gaussian functions. Rel- (Miller et al. 2006a). A turbulent velocity of 1000 km/s
ative to this continuum, the features at 6.91 keV and was found to provide the best fit after various trials. A
7.32 keV are detected at a higher level of significance coveringfactor of 0.5 was chosen as the absence of emis-
(6σ). sion lines suggests an equatorial wind. Finally, the Fe
Itis reasonabletoassociatethe line at6.91±0.01keV abundancewasassumedtobe twicethe solarvalueafter
withHe-likeFeXXV,whichhasarestenergyof6.70keV initial fits; this characterizesthe Fe K lines but does not
(Verner et al. 1996). This translates into a blue-shift of predict absorption lines, e.g. Si, that are not observed.
9300+500 km/s. This feature clearly signals an extreme The initial, lower luminosity grid was fit to the data
−400
disk wind in IGR J17091−3624. Typical velocities in inXSPECas amultiplicative model; free parametersin-
X-ray Binaries are < 1000 km/s (Miller et al. 2006a,b). cludedthecolumndensity,ionization,andvelocityshifts
If the feature at 7.32+0.02 keV is real and can be asso- of the absorbing gas (see Table 1 and Figure 2). For
−0.06
ciated with H-like Fe XXVI at 6.97 keV, it would cor- thedisk blackbodyandpower-lawcontinuum,anioniza-
respond to a blue-shift of 14600+−2850000. For additional tion parameter of logξ = 3.3+−00..21 is required, as well as
details, see Table 1. Although less likely, the 6.91 keV a wind column density of N = 4.7+1.7×1021 cm−2. Ve-
−1.9
line could also be associatedwith a redshift from the H- locity shifts consistent with simple Gaussian models are
like Fe XXVI line. The corresponding inflowing velocity found using the XSTAR grid.
would be 2600±400km/s. If this is due to gravitational To fit the putative higher energy absorption, a sec-
redshift,thecorrespondingradiuswouldbe1.7+0.3×108 ondoutflowcomponentisrequired. Anadditional,lower
−0.2
cm (60±10 R ). luminosity XSTAR component is significant at the 3σ
Schw
Theabsenceofemissionlinesinthesecondspectrumof level, relative to both continua. The wind column den-
IGRJ17091−3624isnotable,butisonlysuggestiveofan sity was higher at N = 1.7+−10..28 × 1022 cm−2, and the
equatorial wind. Given that disk winds have only been logξ = 3.9+0.5. This system is moving even faster at
−0.3
detected in sources viewed at high inclination angles, 15400±400km/s = 0.05c. (See Table 1 and Figure 2)
andgiventhesimilaritiesbetweenIGRJ17091−3624and Repeating this analysis, but utilizing the higher lumi-
GRS1915+105,itislikelythatIGRJ17091−3624isalso nosityXSTARgrid,wefindthatthetwocomponentsare
viewed at a high inclination. However, there is no evi- againrequired. Infact,the valuesofthe columndensity,
dence ofeclipsesin this source,so inclinationsabove70◦ ionizationandvelocityshiftsarenearlyidenticalandwell
can be ruled out. within 1σ of the previous model.
Absorption lines like those detected in the second ob- Toderiveoneestimatetotheradiuswherethesewinds
servationofIGRJ17091−3624areabsentinthe firstob- are launched, we can estimate the radius at which the
servation. FitstotheFeKbandusingalocalcontinuum observed velocity equals the escape velocity. This con-
model, andnarrowGaussianfunctions with FWHM val- strainstheradiustobeatr ≃2.9×109 cm(970R ).
Schw
uescorrespondingtothosemeasuredinthesecondobser- Using ξ = L/(nr2) and N= nrf, where f is the 1-
vation give 1σ confidence limits of 3 eV or less on lines
dimensional filling factor, we can then derive the filling
in the 6.70–7.32 keV band. This is significantly below
factor and density of the region. Assuming the ionizing
the equivalent widths measured in the second observa- luminosity is 3.5×1037 erg/s,the resulting filling factor
tion. (see Table 1)This may simply be due to a variable is f ≃ 0.0008, and the density is n ≃ 2×1015 cm−3.
absorption geometry in IGR J17091−3624;this has pre- However, if the luminosity is higher (L =3.5 × 1038
viouslybeenobservedinH1743−322andGRS1915+105 ion
erg/s), the filling factor decreases to f ≃ 8×10−5, and
(Miller et al. 2006a,b, 2008; Neilsen & Lee 2009).
the density increases to n≃2×1016 cm−3.
3.3. Photoionization Modeling These density estimates are quite high when com-
pared to other X-ray binaries (e.g., Ueda et al. 2004;
To get a better physical picture of the absorption
Miller et al. 2006a, 2008). However, we can invert the
in the second observation of IGR 17091−3624, we also
previous argument and instead derive the filling factor
fit the data with a grid of self-consistent photoioniza- and radius from an assumed density, i.e. n=1012cm−3.
tion models created with XSTAR (Kallman & Bautista We find a larger filling factor, (f ≃ 0.04), and radius,
2001). The ionizing luminosity for this model was de- (r ≃ 1.3 × 1011 cm, 43,300 R ), if we require a
rived from extrapolating the unabsorbed spectrum from Schw
luminosity of 3.5 × 1037 erg/s. A larger luminosity,
the second observation to 0.0136–30 keV, ensuring cov-
i.e. L =3.5 × 1038 erg/s, reduces the filling factor,
erageabove8.8keV,whichisrequiredtoionizeFeXXV. ion
A distance of 8.5 kpc is first assumed to derive this (f ≃ 0.01) but increases the radius (r ≃ 3×1011 cm,
luminosity (Lion=3.5×1037 ergs/s), owing to the loca- 100,000 RSchw). At these radii the escape velocity is
tion of J17091−3624 within the Galactic bulge. How- much lower than the observed velocity.
ever,Altamirano et al.(2011b)alsosuggestthe possibil- Finally,wecanestimatethemassoutflowrate(m˙ wind)
itythatthissourcecouldbeaccretingathighEddington usingamodifiedsphericaloutflow,whichcanbeapproxi-
fractions but further away,and a distance of 25 kpc was matedasm˙ wind ≃1.23mpLionfvΩ/ξ. Here,weassumea
4 A. L. King et al.
coveringfactorΩ/4π=0.5,andanoutflowingvelocityof APM 08279+5255 there are broad absorption features,
v =9,600km/s. Aluminosity ofL =3.5×1038 erg/s which are likely highly relativistic Fe XXV and/or Fe
ion
and filling factor of f = 8×10−5, gives a lower limit of XXVI lines. In these regards, it bears some similarities
m˙ ≃3.5×1016 (104/ξ) g/s. However,a much larger to the most extreme winds in BHB’s.
wind
outflow rate of m˙ ≃1.7×1018 (104/ξ) g/s is found, Observations of BHB point to an anti-correlation of
wind
if we assume L =3.5×1037 erg/sand filling factor of wind and jet outflows from accretion disks (Miller et al.
ion 2006a, 2008; Blum et al. 2010; Neilsen & Lee 2010).
f =0.04
For comparison, L=ηm˙ c2, where η is an efficiency Windsappeartoonlybedetected,oratleastareconsid-
factortypicallytakentobea1cc0%. ForIGRJ17091−3624, erably stronger, in soft, disk–dominated states, and ab-
m˙ = 3.8×1017 g/s. Using logξ = 3.3 from the disk sentinhardstateswherecompact,steadyjetsareubiqui-
acc tous (Fender 2006). In H 1743−322,in particular, there
blackbody and power-law model, we find that the ob-
is evidence that the absence of winds in hard states is
servedportionof the outflow is likely to carry away0.4–
notanartifactofhighionizationhindering the detection
20 times the amount of accretedgas. Unless a geometri-
of absorption lines, but instead represents a real change
calconsiderationservestobiasourestimates,ahighfrac-
in the column density (and thus the mass outflow rate)
tion of the available gas may not accrete onto the black
in any wind (Blum et al. 2010).
hole. This trend is not only seen in BHB but in Seyferts
It appears that our coordinated Chandra and EVLA
aswell. Blustin et al.(2005)note thatmorethanhalfof
observations of IGR J17091−3624 support this anti-
their observed Seyferts show M˙ /M˙ >0.3 .
out acc correlation. The EVLA observations place very tight
limits on the radio flux when the disk wind is de-
3.4. Radio Non-Detections
tected,ordersofmagnitudebelowthelevelatwhichIGR
The EVLA radio observations at 8.4 GHz were made J17091−3624 was detected in radio during its low/hard
nearlycontemporaneouslywiththeirX-raycounterparts. state only a few months prior (Rodriguez et al. 2011).
Both radio observations were nearly two hours in du- Neilsen & Lee (2009) suggested that the production
ration. Neither observation detected a source at the of winds may be responsible for quenching jets in GRS
location of IGR J17091−3624. The RMS noise level 1915+105. It might then be the case that jets should
for each observation was 0.02 mJy and 0.07 mJy for be observed whenever winds are absent. In our first ob-
the two epochs, respectively. The second observation servation of IGR J17091−3624,however, neither a wind
had extended coverage to 4.8 GHz that also had a non- nor a jet is detected, with tight limits. Instead, the
detection. The RMS for this frequency was 0.13 mJy. apparent dichotomy between winds and jets may sig-
In contrast, IGR J17091−3624 was detected at the 1– nal the magnetic field topology in and above the disk is
2 mJy level during the low/hard state (Rodriguez et al. state-dependent. This is broadly consistent with multi-
2011). This supports prior findings that the radio jet is wavelength studies suggesting synchrotron flares above
absent during the periods when winds are seen in BHB thedisk,butonlyinthehardstate(e.g. GX339−4,XTE
(Miller et al. 2006a, 2008; Neilsen & Lee 2010). J1118+480, Di Matteo et al. 1999; Gandhi et al. 2010).
Itisinterestingtospeculatethatthemagneticfieldmight
4. DISCUSSIONANDCONCLUSIONS
be primarily toroidalin the soft state, where a Shakura-
At ionizations above 103, radiation pressure is inef- Sunyaev disk is dominant, but primarily poloidal in the
ficient, and it is not able to drive these winds (e.g., hard state, when the mass accretion rate is lower (e.g.,
Proga et al. 2000). Thus, although the UV components Beckwith et al. 2008). The type of outflow that is ob-
of disk winds in AGN are driven at least partially by served may also depend greatly on how much mass is
radiation pressure, the wind in IGR J17091−3624likely loaded onto magnetic field lines; that could depend on
cannot be driven in this way. A thermal wind can arise variables including the mass accretion rate through the
at radii greater than 0.2 RC (Woods et al. 1996), where disk.
RC = (1.0 × 1010) × (MBH/M⊙)/TC8, where TC8 is
the Compton temperature of the gas in units of 108 K.
We would like to thank the anonymous referee. We
The spectrum observed in the second observation gives
thank Michael Nowak for his instrumental help as well.
R ≃ 5× 1012 cm. Therefore, if we assume our con-
C ALKgratefullyacknowledgessupportthroughtheNASA
servative estimate of the launching radius, it is pos-
Earth and Space Sciences Fellowship. JMM gratefully
sible for IGRJ17091−3624 to have a thermally driven
acknowledges support through the Chandra Guest Ob-
wind. However,ifthe windoriginatescloserto the black
server program. The National Radio Astronomy Obser-
hole, then it is likely that magnetic processes – either
vatoryisafacilityoftheNationalScienceFoundationop-
pressure from magnetic viscosity within the disk (e.g.,
erated under cooperative agreement by Associated Uni-
Proga 2003) or magneto-centrifugal acceleration (e.g.,
versities, Inc.
Blandford & Payne1982)–mustplayaroleinlaunching
the wind observed in IGR J17091−3624.
Fast X-ray disk winds are not only seen in BHB like
IGR J17091−3624, but also in AGN and quasars (e.g.,
King et al. 2011b; Chartas et al. 2002). The fastest UV
winds observed in AGN are pushed to high velocities
by radiation pressure. It remains to be seen whether
a common driving mechanism works across the black
hole mass scale to drive fast, highly ionized X-ray disk
winds. Chartas et al. (2002) show that in the quasar
IGR J17091−3624 5
TABLE 1
SpectralModeling Parametersof the 2nd HEG
observation
Parameter Model1 Model2 Model3 Model4
diskbb+po (diskbb+po) comptt (comptt)
+Gauss+Gauss ×Xstar×Xstar +Gauss+Gauss ×Xstar×Xstar
NH (1022 cm−2) 1.14±0.06 1.13±0.06 0.475+−00..001178 0.558+−00..002258
-
Tin (keV) 1.53±0.09 1.51+−00..1019 - -
Norm 13.1+−32..65 13.8+−41..07 - -
Γ 1.93+−00..1156 1.91±0.17 - -
Norm 0.35+−00..0078 0.34±0.08 - -
-
T0 (keV) - - 0.58±0.01 0.59±0.01
kT(keV) - - 10.5+−310.7 9.8±0.02
τplasma - - 2.24±0.01 2.28±0.01
Norm - - 0.0584±0.0001 0.062+−00..00052
-
EFeXXV (keV) 6.91±0.01 - 6.91+−00..0021 -
FWHM(keV) 0.091+−00..002429 - 0.13+−00..1094 -
EW(keV) 0.021+−00..000052 - 0.040+−00..000079 -
Norm(10−4) 3.5+−00..86 - 6.0+−11..13 -
v(km/s) 9300+−540000 - 9300+−480000 -
-
EFeXXVI (keV) 7.32+−00..0026 - 7.30±0.02 -
FWHM(keV) 0.081+−00..007297 - 0.25+−00..1031 -
EW(keV) 0.032+−00..001084 - 0.089+−00..001134 -
Norm(10−4) 3.4+−10..94 - 11.8+−11..76 -
v(km/s) 14600+−2850000 - 13800±800 -
-
N(1022 cm−2) - 0.47+−00..1179 - 0.45+−00..3137
logξ(ergscms−1) - 3.3+−00..21 - 3.4+−00..21
v(km/s) - 9600+−450000 - 9600±300
-
N(1022 cm−2) - 1.66+−10..1883 - 1.97+−10..2561
logξ(ergscms−1) - 3.9+−00..53 - 3.7+−00..31
v(km/s) - 15400±400 - 15400+−430000
-
χ2/ν 2725/3408 =0.80 2731/3408 =0.80 2793/3408 =0.82 2761/3408 =0.81
Note. — This Table lists the line detectionsusing Gaussian func-
tionsaswellasmoreself-consistent,photoionizationcomponentscre-
atedwith XSTAR,assumingtwodifferentcontinuummodels. TBabs
isappliedtoallthemodelsandtheerrorsare1σconfidencelevel.
6 A. L. King et al.
REFERENCES
Altamirano,D.,etal.2011,TheAstronomer’sTelegram,3299,1 McMullin,J.P.,Waters,B.,Schiebel,D.,Young,W.,&Golap,
—.2011b,ApJ,submitted K.2007,inAstronomicalSocietyofthePacificConference
Bautista,M.A.,&Kallman,T.R.2001,ApJS,134,139 Series,Vol.376,AstronomicalDataAnalysisSoftwareand
Beckwith,K.,Hawley,J.F.,&Krolik,J.H.2008,ApJ,678,1180 SystemsXVI,ed.R.A.Shaw,F.Hill,&D.J.Bell,127
Blandford,R.D.,&Payne,D.G.1982,MNRAS,199,883 Miller,J.M.,etal.2006a, ApJ,646,394
Blum,J.L.,Miller,J.M.,Cackett, E.,Yamaoka,K.,Takahashi, Miller,J.M.,Raymond,J.,Fabian,A.,Steeghs, D.,Homan,J.,
H.,Raymond,J.,Reynolds,C.S.,&Fabian,A.C.2010,ApJ, Reynolds,C.,vanderKlis,M.,&Wijnands,R.2006b,Nature,
713,1244 441,953
Blustin,A.J.,Page,M.J.,Fuerst,S.V.,Branduardi-Raymont, Miller,J.M.,Raymond,J.,Reynolds,C.S.,Fabian,A.C.,
G.,&Ashton,C.E.2005,A&A,431,111 Kallman,T.R.,&Homan,J.2008,ApJ,680,1359
Chartas,G.,Brandt,W.N.,Gallagher,S.C.,&Garmire,G.P. Neilsen,J.,&Lee,J.C.2009, Nature,458,481
2002, ApJ,579,169 Neilsen,J.,&Lee,J.C.2010, inBulletinoftheAmerican
DiMatteo,T.,Celotti,A.,&Fabian,A.C.1999,MNRAS,304, AstronomicalSociety, Vol.42,AmericanAstronomicalSociety
809 MeetingAbstracts#215,371.04
Dickey,J.M.,&Lockman,F.J.1990,ARA&A,28,215 Proga,D.2003,ApJ,585,406
Fender,R.2006,JetsfromX-raybinaries,ed.Lewin,W.H.G.& Proga,D.,Stone,J.M.,&Kallman,T.R.2000, ApJ,543,686
vanderKlis,M.,381–419 Protassov,R.,vanDyk,D.A.,Connors,A.,Kashyap,V.L.,&
Fruscione,A.,etal.2006, Proc.SPIE,6270 Siemiginowska,A.2002,ApJ,571,545
Gandhi,P.,etal.2010,MNRAS,407,2166 Rodriguez,J.,Corbel,S.,Caballero,I.,Tomsick,J.A.,Tzioumis,
Giroletti,M.,&Panessa,F.2009,ApJ,706,L260 T.,Paizis,A.,CadolleBel,M.,&Kuulkers,E.2011,A&A,533,
Jones,S.,McHardy,I.,Moss,D.,Seymour,N.,Breedt,E., L4
Uttley,P.,K¨ording,E.,&Tudose, V.2011,MNRAS,412,2641 Tombesi,F.,Sambruna,R.M.,Reeves,J.N.,Braito,V.,Ballo,
Kallman,T.,&Bautista,M.2001,ApJS,133,221 L.,Gofford,J.,Cappi,M.,&Mushotzky, R.F.2010,ApJ,719,
Kallman,T.R.,Bautista,M.A.,Goriely,S.,Mendoza,C.,Miller, 700
J.M.,Palmeri,P.,Quinet,P.,&Raymond,J.2009,ApJ,701, Ueda,Y.,Murakami,H.,Yamaoka,K.,Dotani,T.,&Ebisawa,K.
865 2004,ApJ,609,325
Kaspi,S.,etal.2002,ApJ,574,643 Verner,D.A.,Verner,E.M.,&Ferland,G.J.1996,AtomicData
King,A.L.,etal.2011a, ApJ,729,19 andNuclearDataTables,64,1
King,A.L.,Miller,J.M.,&Raymond,J.C.2011b,ApJ, Vierdayanti,K.,Mineshige,S.,&Ueda,Y.2010, PASJ,62,239
submitted Woods,D.T.,Klein,R.I.,Castor,J.I.,McKee,C.F.,&Bell,
Krimm,H.A.,etal.2011, TheAstronomer’sTelegram,3144,1 J.B.1996,ApJ,461,767
Kubota,A.,etal.2007,PASJ,59,185
Luketic,S.,Proga,D.,Kallman,T.R.,Raymond,J.C.,&Miller,
J.M.2010,ApJ,719,515