Black hole candidate XTE J1752-223: Swift observations of canonical states during outburst
P.A. Curran, T. J. Maccarone, P. Casella, P.A. Evans, W. Landsman, H.A. Krimm, C. Brocksopp, M. Still
aa r X i v : . [ a s t r o - ph . H E ] J u l Mon. Not. R. Astron. Soc. , 1–8 () Printed 15 November 2018 (MN L A TEX style file v2.2)
Black hole candidate XTE J1752-223:
Swift observations ofcanonical states during outburst
P.A. Curran ⋆ , T. J. Maccarone , P. Casella , P.A. Evans , W. Landsman ,H.A. Krimm , , C. Brocksopp , M. Still Mullard Space Science Laboratory, University College of London, Holmbury St Mary, Dorking, Surrey RH5 6NT, UK School of Physics and Astronomy, University of Southampton, Southampton, Hampshire, SO17 1BJ, UK Department of Physics and Astronomy, University of Leicester, University Road, Leicester LE1 7RH, UK Adnet Systems, NASA/Goddard Space Flight Center, Code 667, Greenbelt MD 20771, USA NASA/Goddard Space Flight Center, Astrophysics Science Division, Code 661, Greenbelt, MD 20771, USA Universities Space Research Association, Columbia, MD 21044, USA NASA Ames Research Center, Moffett Field, CA 94035, USA
Accepted/ Received;
ABSTRACT
We present
Swift broadband observations of the recently discovered black hole can-didate, X-ray transient, XTE J1752-223, obtained over the period of outburst fromOctober 2009 to June 2010. From
Swift -UVOT data we confirm the presence of an op-tical counterpart which displays variability correlated, in the soft state, to the X-rayemission observed by
Swift -XRT. The optical counterpart also displays hystereticalbehaviour between the states not normally observed in the optical bands, suggestinga possible contribution from a synchrotron emitting jet to the optical emission in therising hard state. We offer a purely phenomenological treatment of the spectra as anindication of the canonical spectral state of the source during different periods of theoutburst. We find that the high energy hardness-intensity diagrams over two separatebands follows the canonical behavior, confirming the spectral states. Our XRT timinganalysis shows that in the hard state there is significant variability below 10 Hz whichis more pronounced at low energies, while during the soft state the level of variabil-ity is consistent with being minimal. These properties of XTE J1752-223 support itscandidacy as a black hole in the Galactic centre region.
Key words:
X-rays: binaries – X-rays: bursts – Binaries: close – Stars: individual:XTE J1752-223
Low mass X-ray binaries are for the majority of the time ina state of quiescensce with faint or non-detected X-ray emis-sion, though optical or near-infrared (nIR) counterparts maybe visible due to emission from the donor star, or possiblythe jet, hot spot, or outer accretion disk. They are oftenonly discovered when they enter an active state of outburstwhen – powered by an increased level of accretion onto thecentral, compact object (black hole or neutron star) – thereis a dramatic increase of the X-ray, optical/nIR and radioflux. During these outbursts the systems have been observedto go through a number of high energy spectral states be- ⋆ Present address: AIM, CEA/DSM - CNRS - Universit´e ParisDiderot, Irfu/SAP, Centre de Saclay, Bat. 709, FR-91191 Gif-sur-Yvette Cedex, France; e-mail: [email protected] fore returning to a quiescent state, usually on times scalesof weeks, months or even longer. These states are a gener-ally low intensity, power-law dominated, hard state followedby a usually, higher intensity, thermal-dominant , soft statewhich decreases in flux, via a late hard state, over time.Additionally, the hard states are associated with aperiodicvariability of the light curve not present in the soft state (seeMcClintock & Remillard 2006 for a fuller description of thevarious possible states).XTE J1752-223, a new X-ray transient and black holecandidate (Mu˜noz-Darias et al. 2010b; Markwardt et al.2009a; Shaposhnikov et al. 2009) in the Galactic cen-ter region, was detected on 2009-10-23 at 19:55 UT(MJD 55128.33) by RXTE and on 2009-10-24 at14:18:50 UT by the BAT instrument on board the Swift satellite (Markwardt et al. 2009b). The high en-ergy, variable emission of the source was confirmed in c (cid:13) RAS
P.A. Curran et al. the following days by
Swift /XRT (Markwardt et al.2009a) and RXTE (Remillard & The ASM Team at MIT2009; Shaposhnikov et al. 2009) as well as byMAXI/GSC (Nakahira et al. 2009) and
Fermi /GBM(Wilson-Hodge et al. 2009). An optical and nIR counter-part was proposed by Torres et al. (2009a,b) while a radiosource coincident with the X-ray position (Markwardt et al.2009a) was observed by the Australia Telescope CompactArray (ATCA; Brocksopp et al. 2009). The source was ob-served to have undergone a state transition in mid-January2010 (MJD ∼ & . ∼ Swift – Burst Alert Tele-scope, X-ray Telescope and Ultraviolet/Optical Telescope– monitoring observations of XTE J1752-223, obtained overthe period of the outburst. Based on these data, we identifythe periods of the various states and compare the behaviorof the major photometric, spectral and timing parametersduring these states to those expected from black hole X-raybinaries. In section 2 we introduce the observations and re-duction methods, while in section 3 we present the results ofour photometric, spectral and timing analyses of the data.We summarise our findings in section 4.
Swift (Gehrels et al. 2004) observed XTE J1752-223 withits narrow field instruments, the Ultraviolet/Optical Tele-scope (UVOT; Roming et al. 2005) and the X-ray Telescope(XRT; Burrows et al. 2005), on 55 epochs from October 25to November 03 2009 and from February 04 2010 to June5 2010. These observations (under Target IDs 31532, 31640and 31688) coincide with the period of the outburst. Obser-vations with the narrow field instruments were interruptedfor 3 months due to the source becoming constrained bythe position of the Sun. Wide field instruments also becamesun constrained but at a later date and for a shorter periodof time, depending on the satellite and instrument in ques-tion (e.g. MAXI was constrained from December 15 to 24).During the period of the outburst, as it had been on pre-vious occasions, the field was also being observed (thoughconstrained from December 13 to 28) by
Swift ’s Burst AlertTelescope (BAT; Barthelmy et al. 2005) which, in combina-tion with XRT and UVOT, affords quasi-simultaneous ob-servations from 0.002 to 150 keV. Reduction of
Swift datawas carried out using version 3.4 of the
Swift software, re-leased as part of HEASOFT 6.7. All uncertainties are quotedat the 1 σ confidence level. UVOT data have been pre-processed at the
Swift
Data Cen-ter (see Breeveld et al. 2010) and require only minimum userprocessing. As the pointing of
Swift is only accurate to a fewarcseconds, the standard pre-process pipeline provides an as-trometric correction by matching sources with the USNO-Bcatalog. However, in this case the automatic matching failedbecause the field differs significantly between the ultravioletand the USNO-B bandpass. To overcome this, we createdour own astrometric reference catalog by selecting blue andbright field stars in the GSC-2.3 catalog (Lasker et al. 2008)to determine astrometric corrections for the data.The image data of each filter, from each observationsequence, i.e., with a given observation ID, were summedusing uvotimsum . All sequences were then combined in anindividual fits file, using fappend , which was again summedusing uvotimsum to obtain a deep image of the field ineach filter. From these images, two possible optical counter-parts of XTE J1752-223 are identified, consistent with theX-ray position (Markwardt et al. 2009a); source A, consis-tent with the counterpart proposed by Torres et al. (2009a),is within 1 σ and source B is within 2 σ of the X-ray position.The positions (J2000; 0.31 ′′ error) of the possible counter-parts were derived from such a deep (1853 s) v band imagesummed between MJD 55130 and 55138 (Figure 1), usingthe uvotdetect command, as:A 17:52:15.08 − − uvotmaghist , using an extraction region ofradius 4 ′′ and XSPEC compatible spectral files for source Acreated with the same region, using uvot2pha . Magnitudesare based on the UVOT photometric system (Poole et al.2008). Source B shows no variation over the time of theobservations, while source A shows significant variabilityin the v , b and u bands (Figure 2) and is undetected inthe individual UV band ( uw um uw
2) images, exceptfor a few uw E ( B − V ) = 1 . Swift filters (Poole et al. 2008) and the parameteri-zation of Pei (1992), the extinctions is the
Swift bands are: A v ≤ . A b ≤ . A u ≤ . A uvw ≤ . A uvw ≤ . A uvm ≤ .
0, though these should be treated with cautionas estimates of the extinction so close to the Galactic plane( < The XRT Windowed Timing mode (WT; Hill et al. 2004)event data [0.3-10 keV] were analysed using the tools pre- c (cid:13) RAS, MNRAS , 1–8 lack hole candidate XTE J1752-223:
Swift observations of canonical states during outburst Figure 1.
UVOT 1853 s v band 130 ′′ × ′′ (MJD 55130-55138)image with XRT 1 σ error circle marked along with optical sourcesA and B. Table 1.
UVOT magnitudes and 3 σ limits of sources A and Bas well as the background limit of source A, limit , from summedimages taken between MJD 55130 and 55138. Uncorrected forinterstellar extinction of E ( B − V ) . . limitv ± ± b
703 18.10 ± ± u ± ± uw ± ± um > > uw ± > sented by Evans et al. (2009). These correct for pile up andthe affect of bad CCD columns. Version v011 response ma-trices were used for the spectral analysis. The best XRTposition of 17:52:15.14 − ′′ error) is aspreviously reported (Markwardt et al. 2009a). Backgroundsubtracted XRT count rate light curves are binned per ob-serving sequence (Figure 3; combined with MAXI observa-tions). Unbinned source light curves (not background sub-tracted), for each sequence, were used to produce powerdensity spectra (PDS) and to calculate the RMS variabil-ity with the FTOOLs, powspec and lcstats respectively.Within each of these the binning was set to be N times theminimum bin size (1.766 ms) where N=1 for the PDS andN=8 (14.128 ms) for the RMS variability, giving frequencyranges of 0.02-570 Hz and 0.02-70 Hz respectively. The av-erage PDS were calculated, within powspec , from an inputlist of the light curves of the required sequences, using thesame binning. Lags between the standard XRT low [0.3-1.5 keV] and high [1.5-10 keV] energy band unbinned lightcurves were calculated with crosscor and a bin size set toN=8 (14.128 ms). BAT detector plane histograms (DPHs) were obtained insurvey mode; these are 2 dimensional images with full spec- M a gn it ud e MJD − 50,000 v b u uvw1
Figure 2.
UVOT v (crosses), b (triangles), u (squares) and uw σ limits) band light curves for source A, which showvariability; dimming by 1 magnitude or more from the start tothe end of observations. The up-pointing arrows signify the datesbetween which the source is transitioning from the soft state tothe late, hard state (MJD 55280, 55328). tral data and coarse time binning. Spectra and light curveswere extracted from these DPHs using mask weighting (alsoknown as ray-tracing) to account for the position of thesource on the coded aperture mask. The data were ini-tially processed with the FTOOL, batsurvey which ap-plies data quality cuts and uses the tools batfftimage and batcelldetect to extract 8-channel spectra for the source.We then used custom software to format the spectra for XSPEC , and batdrmgen to generate the response files. Thelight curve was derived using a similar method. BAT, as asurvey telescope, had been observing the field of XTE J1752-223 since the start of
Swift operations in January 2006,though not monitoring the source. The extracted BAT lightcurve (15-150 keV) of the source from January 12 2006through to the present, displays no activity until the 2009outburst.
The joint XRT-BAT spectra were fit within
XSPEC(12.5.1) , using C-statistics and interstellar absorptionmodeled by tbabs ; initially we fit an absorbed power law( pow ) to the spectra before adding a thermal component( diskbb ). We note that these are phenomenological mod-els for the respective spectral states, not a suggestion thatthese are necessarily the best physical models; they shouldnot be used to deduce physical constraints but only as anindication of the spectral state of the source.The early data (MJD 55131-55138) are well describedby the power law of average photon index, Γ = 1 . ± . N H = 0 . ± . × cm − , greater than the Galactic value of N H =0 . × cm − (Kalberla et al. 2005) but lower than the N H = 0 . +0 . − . × cm − found by Mu˜noz-Darias et al. c (cid:13) RAS, MNRAS , 1–8
P.A. Curran et al. − − . . B A T ( c t s / c m / s ) . . M AX I ( c t s / c m / s ) X R T ( c t s / s ) . . X R T H R X R T % R M S MJD − 50,000
Figure 3.
High energy (
Swift -BAT [15-150 keV], MAXI [4-10 keV],
Swift -XRT [0.3-10 keV]) light curves during the periodof outburst. Also shown are the XRT hardness ratio, HR [1.5-10 keV / 0.3-1.5 keV] and percentage RMS. The absence of dataaround MJD ∼ (2010b). Data observed from MJD 55234-55280 additionallysupport a soft state thermal component decreasing from atemperature of ∼ . ∼ . . ± .
02. The power law in-dices and the hardness ratios at these late times fail to returnto the original, rising hard state values (Γ = 1 . ± . Γ . . . T ( k e V ) T no r m ( ) MJD − 50,000
Figure 4.
Best fit parameters of the joint XRT-BAT spectralfits: power law photon index, Γ, thermal component tempera-ture, T , and thermal component normalization, T norm , in unitsof 10 . Note that the photon indices, in the period where thereis a thermal component, are not reliable. The absence of dataaround MJD ∼ original hard state or that the late hard state has marginallydifferent spectral properties. An optical and nIR counterpart was proposed byTorres et al. (2009a,b) based on a comparison between theirimages and those from the Digitized Sky Survey, 2MASSand UKIDSS. We have confirmed UVOT source A as thisoptical counterpart on the basis of a significant, though lowamplitude variability in the v , b and u bands (Figure 2). Fur-thermore, these optical magnitudes, or count rates, exhibit apower-law correlation with the X-ray count rate, as observedby XRT at times after MJD 55232 (Figure 5), i.e., when thesource is in the soft state and later, during the transitionto the late hard state. The earlier, hard state data does notfit on this correlation, but displays hysteretical behaviourand for the given X-ray count rate it spans has a signifi-cantly higher magnitude in all bands. This is similar to thehysteretical behaviour observed in the nIR for a number oftransients (Russell et al. 2007), most notably XTE J1550-564, where the additional hard state emission is attributedto optically thin synchrotron emission from a jet (Jain et al.2001; Homan et al. 2005; Hynes et al. 2006), which wouldbe quenched in the soft state (e.g., Fender et al. 2004), andweak at low X-ray luminosities, leading to the hysteresis.Though this hysteresis is not normally observed in the opti-cal bands, the optical data presented here is in good agree-ment with the synchrotron emitting jet making a significantcontribution to the optical emission in the rising hard state.While the observed v band brightening from MJD 55292corresponds to the suggested turn-on, in nIR, of a compactjet between MJD 55293 and 55298 (Buxton et al. 2010), itis not corroborated by the other bands and is not statis-tically significant, though it may indicate the turn-on of a c (cid:13) RAS, MNRAS , 1–8 lack hole candidate XTE J1752-223:
Swift observations of canonical states during outburst UVO T m a gn it ud e XRT count rate [0.3−10keV] v b u
Figure 5.
UVOT magnitude in three filters versus XRT countrate. Data points in grey indicate observations before MJD ∼ jet and the transition to the hard state; at such low X-rayluminosity levels, the jet contribution to the optical wouldbe minimal.From a simultaneous fit of the v , b and u bands the cor-relation is found to be Rate optical ∝ Rate X − ray 0 . ± . withobserved colours of u − v = 2 . ± .
09 and u − v = 1 . ± . χ ν = 1 . . ± .
04, 0 . ± .
03 and 0 . ± .
02 respectively, hinting atpossible spectral evolution. Such a correlation is more usu-ally associated with the hard state (Russell et al. 2006) butwe cannot say if the early, hard state data follows anothercorrelation (expected to be Rate optical ∝ Rate X − ray 0 . ;Russell et al. 2006) due to the very limited range of mag-nitudes and X-ray count rates over this period. Though wefind a correlation between the optical and X-ray, we cannotimply causality or say which leads without a study of thephase lag between the two bands, which is not possible giventhe time resolution of the optical data. Similarly a timinganalysis to investigate variability of the optical data is notpossible. The correlation may be due to reprocessing of theX-rays or through both originating from the accretion disk;it is also possible that optical emission in the soft state origi-nates from a jet, the state of which traces the spectral states( ? ) The quality of the UVOT spectral data does not sup-port any spectral evolution in the optical, nor does it enableus to specify the most appropriate spectral model for thesource such as a power law associated with the above sug-gested jet emission. Hence, we model the data phenomeno-logically in XSPEC , by a power law (Flux ∝ ν β ) and a dust ex-tinction component ( zdust ), though we note that this is nota particularly good fit. Given that an unknown amount ofGalactic extinction, E ( B − V ) ≤ .
554 (Schlegel et al. 1998),effects the data we can only place limits on the spectral in-dex of − . . β optical . .
0, though the nominal best fit isat β optical = − . ± . E ( B − V ) = 0 . ± .
04. Given theuncertainties of the spectral slope and the lack of phase lag information, we are unable to make a meaningful compari-son to the predictions of the various models.
The Hardness-Intensity diagrams (HID) for both the XRTdata [1.5-10 keV / 0.3-1.5 keV] and the quasi-simultaneous(same day) BAT/MAXI data starting at the same time areshown in Figure 6. Due to the fact that the BAT/MAXI dataare only quasi-simultaneous, the hardness ratio (HR) of BAT[15-150 keV] with respect to MAXI [4-10 keV] at low BATcount rates ( . . The RMS variation of the light curves exhibit a similarevolution to hardness ratio (Figure 3), from average RMS= 53 . ± . .
12 percent during thesoft state observations and increasing to the original levelsat later times. Examining the light curves in the standardXRT low [0.3-1.5 keV] and high [1.5-10 keV] energy bandswe find that, during the rising hard state this variability ismore pronounced in the low band with an average ratio be-tween the two of 0 . ± .
02. This suggests an energy depen-dency of RMS variation, in contrast to Mu˜noz-Darias et al.(2010b) who find a flat RMS spectrum (Vaughan et al. 2003;Gierli´nski & Zdziarski 2005) from XTE data at energies inthe range 2-10 keV. The lag between the two bands, calcu-lated via a cross correlation, is consistent with zero in allcases though is not tightly constrained.Examining the average power density spectra of thestates (Figure 7) we see that the hard state has a high levelof power in low frequency ( .
10 Hz) variability, though thereis no significant sign of quasi-periodic oscillations (QPOs).The soft state, on the other hand, has a much lower leverof power and only at frequencies . . c (cid:13) RAS, MNRAS , 1–8
P.A. Curran et al. . . . . . C oun t r a t e [ M AX I] HR [BAT/MAXI] C oun t r a t e [ . − e V ] HR [1.5−10keV/0.3−1.5keV]
Figure 6.
Hardness-Intensity diagrams (HID) for bothBAT/MAXI ( upper ) and XRT ( lower ). The grey lines, at highcount rates in the XRT and BAT plots show the time duringwhich the XRT was unable to observe (MJD 55139-55233), whilethe circles show the start and end of the transition from the softstate at MJD 55280 and 55328. Note that at low count rates( . .
1) the BAT/MAXI HR is not reliable; the later, low countrate, data points are shaded for clarity of the plot. by the .
12 percent limit of RMSs during this period (MJD55234-55289). After MJD 55321, where the XRT count ratedrops below 10 cts/s, the light curves and PDSs are heav-ily affected by background counts (Figure 8) and hence theRMS values for these observations are unreliable and notplotted in Figure 3.The excess of low energy variability in the rising hardstate is similar to the behaviour of the hard state vari-ability in SWIFT J1753.5-0127 and GX 339-4 as reportedby Wilkinson & Uttley (2009). Within the framework oftheir model, enhanced low energy variability (at frequen-cies . . − − . F r e qu e n c y * P o w e r (r m s H z − ) − − . Frequency (Hz)
Figure 7.
The average power density spectrum (PDS) for XRTlight curve in the hard state (MJD 55130-55138; upper ) exhibitsthe aperiodic variability of the light curve not present in the av-erage soft state PDS (MJD 55234-55289; lower ). duced in Wilkinson & Uttley (2009) are required to test thispossibility. Data, over a greater energy range, which betterconstrains column density and photon index may also benecessary. XTE J1752-223 was observed by
Swift in a relatively lowintensity, hard state from 2009-10-26 to 2009-11-03 (MJD55131-55138) which was signified by hard power law energyspectra, and high levels ( ∼
50 per cent) of RMS variabilityin the X-ray light curve. XTE J1752-223 then became con-strained by the position of the Sun and was unobservableuntil 2010-02-04 (MJD 55234) by which time it had experi-enced a ten fold increase in the flux observed by XRT [0.3-10.0 keV], which proceeded to decrease to the original levelsover a period of 2 months. The spectra during this periodclearly exhibited a strong thermal component and the vari-ability and hardness ratios of the light curves had droppedsignificantly, consistent with the source being in a soft stateuntil 2010-03-25 (MJD 55280). Observations from 2010-03-28 (MJD 55283) to 2010-05-12 (MJD 55328) do not supporta thermal component but do display a decreasing photonindex along with increasing hardness and RMS variability.This suggests a transitional state before reverting to thehard state, with a relatively constant photon index, in which c (cid:13) RAS, MNRAS , 1–8 lack hole candidate XTE J1752-223:
Swift observations of canonical states during outburst . . F r e qu e n c y * P o w e r (r m s H z − ) . . Frequency (Hz)
Figure 8.
The power density spectra (PDS) for the XRT sourcelight curve, at low count rates ( upper ; example observations onMJD 55340 where the count rate is 1.7 cts/s), is heavily affectedby that of the background light curve at the same time ( lower ). the system is observed until the end of observations on 2010-06-05 (MJD 55352). The high energy hardness-intensity dia-grams over two separate bands follow the canonical behaviorassociated with a black hole binary, further confirming thespectral states of the system during the outburst. Our tim-ing analysis shows that in the hard state there is significantvariability below 10 Hz which is more pronounced at low en-ergies, while during the soft state the level of variability isconsistent with being minimal.We are able to confirm that the UVOT source at theposition of the proposed optical and near-infrared counter-part is associated with XTE J1752-223 due to the observedlow amplitude variability and, in the soft state and later,correlation with the X-ray emission as measured by XRT.However, we cannot state that the optical flux is causallyconnected to the X-ray emission via reprocessing of the X-rays or through both originating from the accretion disk;given the uncertainties of the spectral slope and the lack ofphase lag information, we are unable to make a meaningfulcomparison to the predictions of the various models. Theoptical counterpart also displays hysteretical behaviour notnormally observed in the optical bands; for a given X-raycount rate, the magnitude in the rising hard state is signif-icantly higher than that in the soft state. This is similar tothe hysteretical behaviour observed in the nIR for a numberof transients where the additional hard state emission is at-tributed to optically thin synchrotron emission from a jet,which would be quenched in the soft state and weak at lowX-ray luminosities. Though this hysteresis is not normallyobserved in the optical bands, it is in good agreement withthe synchrotron emitting jet making a significant contribu-tion to the optical emission in the rising hard state. ACKNOWLEDGEMENTS
We thank the referee for constructive comments and PaulKuin for useful discussions on UVOT. PAC, PAE and CBacknowledge support from STFC. TJM thanks the EU FP7for support through grant number ITN 215212 ‘Black Hole Universe’. PC acknowledges funding via a EU Marie CurieIntra-European Fellowship under contract no. 2009-237722.This research has made use of: Swift data supplied by theUK Swift Science Data Centre at the University of Leices-ter; and MAXI data provided by RIKEN, JAXA and MAXIteams.
REFERENCES
Barthelmy, S. D., et al. 2005, Space Science Reviews, 120,143Belloni, T., ed. 2010, Lecture Notes in Physics, BerlinSpringer Verlag, Vol. 794, The Jet ParadigmBreeveld, A. A., et al. 2010, MNRAS, 874Brocksopp, C., Corbel, S., Tzioumis, T., & Fender, R. 2009,ATel, 2278Burrows, D. N., et al. 2005, Space Science Reviews, 120,165Buxton, M., Dincer, T., Kalemci, E., & Tomsick, J. 2010,The Astronomer’s Telegram, 2549, 1Evans, P. A., et al. 2009, MNRAS, 397, 1177Fender, R. P., Belloni, T. M., & Gallo, E. 2004, MNRAS,355, 1105Gehrels, N., et al. 2004, ApJ, 611, 1005Gierli´nski, M., & Zdziarski, A. A. 2005, MNRAS, 363, 1349Hill, J. E., et al. 2004, in Society of Photo-Optical In-strumentation Engineers (SPIE) Conference Series, Vol.5165, Society of Photo-Optical Instrumentation Engineers(SPIE) Conference Series, ed. K. A. Flanagan & O. H. W.Siegmund, 217Homan, J. 2010, The Astronomer’s Telegram, 2387, 1Homan, J., & Belloni, T. 2005, Ap&SS, 300, 107Homan, J., Buxton, M., Markoff, S., Bailyn, C. D., Nespoli,E., & Belloni, T. 2005, ApJ, 624, 295Homan, J., Wijnands, R., van der Klis, M., Belloni, T.,van Paradijs, J., Klein-Wolt, M., Fender, R., & M´endez,M. 2001, ApS, 132, 377Hynes, R. I., et al. 2006, ApJ, 651, 401Jain, R. K., Bailyn, C. D., Orosz, J. A., McClintock, J. E.,& Remillard, R. A. 2001, ApJ, 554, L181Kalberla, P. M. W., Burton, W. B., Hartmann, D., Arnal,E. M., Bajaja, E., Morras, R., & P¨oppel, W. G. L. 2005,A&A, 440, 775Lasker, B. M., et al. 2008, AJ, 136, 735Markwardt, C. B., Barthelmy, S. D., Evans, P. A., &Swank, J. H. 2009a, ATel, 2261Markwardt, C. B., et al. 2009b, ATel, 2258McClintock, J. E., & Remillard, R. A. 2006, Black holebinaries, ed. Lewin, W. H. G. & van der Klis, M. 157Miller-Jones, J. C. A., et al. 2010, ApJ, 716, L109Mu˜noz-Darias, T., Motta, S., Belloni, T., & Homan, J.2010a, The Astronomer’s Telegram, 2518, 1Mu˜noz-Darias, T., Motta, S., Pawar, D., Belloni, T. M.,Campana, S., & Bhattacharya, D. 2010b, MNRAS, 404,L94Nakahira, S., et al. 2009, ATel, 2259Nakahira, S., et al. 2010, ASJ, 62Pei, Y. C. 1992, ApJ, 395, 130Poole, T. S., et al. 2008, MNRAS, 383, 627Remillard, R. A., & The ASM Team at MIT. 2009, ATel,2265 c (cid:13) RAS, MNRAS , 1–8
P.A. Curran et al.
Roming, P. W. A., et al. 2005, Space Science Reviews, 120,95Russell, D. M., Fender, R. P., Hynes, R. I., Brocksopp,C., Homan, J., Jonker, P. G., & Buxton, M. M. 2006,MNRAS, 371, 1334Russell, D. M., Maccarone, T. J., K¨ording, E. G., &Homan, J. 2007, MNRAS, 379, 1401Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ,500, 525Shaposhnikov, N., Markwardt, C. B., & Swank, J. H. 2009,ATel, 2269, 1Torres, M. A. P., Jonker, P. G., Steeghs, D., Yan, H.,Huang, J., & Soderberg, A. M. 2009a, ATel, 2263Torres, M. A. P., Steeghs, D., Jonker, P. G., Thompson, I.,& Soderberg, A. M. 2009b, ATel, 2268Tudose, V., Fender, R. P., Linares, M., Maitra, D., & vander Klis, M. 2009, MNRAS, 400, 2111Vaughan, S., Edelson, R., Warwick, R. S., & Uttley, P.2003, MNRAS, 345, 1271Wilkinson, T., & Uttley, P. 2009, MNRAS, 397, 666Wilson-Hodge, C. A., Camero-Arranz, A., Case, G., Chap-lin, V., & Connaughton, V. 2009, ATel, 2280 c (cid:13) RAS, MNRAS000