astro-ph0611341/ms.tex
1: \def\IGR{IGR~J17497$-$2821}
2: \def\SWIFT{Swift~J1753.5$-$0127}
3: \def\ergcms{erg cm$^{-2}$ s$^{-1}$ }
4: \def\integral{{\it{INTEGRAL}}}
5: \def\rxte{{\it{RXTE}}}
6: 
7: \documentclass[]{aastex}
8: %\documentclass[12pt,preprint]{aastex}
9: \usepackage{emulateapj5}
10: %\usepackage{amssymb}
11: %\usepackage{natbib}
12: %\usepackage{epsfig}
13: %\usepackage{epsf}
14: \usepackage{color}
15: \definecolor{red}{rgb}{0.7,0,0}
16: \definecolor{black}{rgb}{0,0,0}
17: \def\correc#1{{\textcolor{black}{#1}}}
18: 
19: \bibliographystyle{apj}
20: 
21: \submitted{Accepted in ApJ Letters}
22: \shorttitle{\IGR\ observed by \rxte\ and ATCA}
23: \shortauthors{Rodriguez et al.}
24: %\received{2002 July 13}
25: \begin{document}
26: \title{The discovery outburst of the X-ray transient \IGR\ observed with \rxte\ and ATCA}
27: 
28: 
29: \author{J\'er\^ome Rodriguez\altaffilmark{1}, 
30: Marion Cadolle Bel\altaffilmark{2,1}, John A. Tomsick\altaffilmark{3,4}, 
31: St\'ephane Corbel\altaffilmark{1}, Catherine Brocksopp\altaffilmark{5}, 
32: Ada Paizis\altaffilmark{6}, Simon E. Shaw\altaffilmark{7,8}, 
33:  Arash Bodaghee\altaffilmark{8,9}}
34: 
35: 
36: \altaffiltext{1}{AIM Astrophysique Interaction Multi-\'echelles  
37: CEA Saclay, DSM/DAPNIA/SAp, F-91191 Gif sur Yvette, Cedex France}
38: \altaffiltext{2}{European Space Astronomy Centre, Villafranca del Castillo, 28629
39: Madrid Spain}
40: \altaffiltext{3}{Space Sciences Laboratory, 7 Gauss Way,
41: University of California, Berkeley, CA 94720,  USA}
42: \altaffiltext{4}{Center for Astrophysics and Space Sciences,
43: 9500 Gilman Drive, Code 0424, University of California at San Diego, La Jolla, CA 92093, USA}
44: \altaffiltext{5}{Mullard Space Science Laboratory, University College London, Holmbury St. Mary, Dorking, Surrey RH5 6NT, UK}
45: \altaffiltext{6}{IASF Milano--INAF, Via Bassini 15, 20133 Milano, Italy}
46: \altaffiltext{7}{School of Physics and Astronomy, University of Southampton, SO17 1BJ, UK}
47: \altaffiltext{8}{INTEGRAL Science Data Centre, Chemin d'Ecogia, 16, 1290 Versoix, Switzerland}
48: \altaffiltext{9}{Observatoire Astronomique de l'Universit\'e de Gen\`eve, Chemin des Maillettes 
49: 51, CH-1290 Sauverny, Switzerland}
50: 
51: 
52: 
53: \begin{abstract}
54: We report the results of a series of \rxte\ and ATCA observations of the 
55: recently-discovered X-ray transient  \IGR. Our 3-200 keV PCA+HEXTE spectral analysis shows
56: very little variations over a period of $\sim$10 days around the maximum of the outburst. 
57: \IGR\ is found in a typical Low Hard State (LHS) of X-ray binaries (XRB), well represented 
58: by an absorbed Comptonized spectrum with an iron edge at about 7 keV.
59: The high value of the absorption ($\sim4\times 10^{22}$ cm$^{-2}$) suggests 
60: that the source is located at a large  distance, either close to the Galactic center 
61: or beyond. The timing analysis shows no particular features, while the shape of the 
62: power density spectra is also typical of LHS of XRBs, with $\sim$36\% RMS variability. 
63: No radio counterpart is found down to a limit of  0.21 mJy at 4.80 GHz  and 8.64 GHz. 
64: \correc{Although the position of \IGR\ in the radio to X-ray flux diagram is well 
65: below the correlation usually observed in the LHS of black holes,} the comparison 
66: of its X-ray properties with those of other sources leads us to suggest that it 
67: is  a black hole candidate.
68: \end{abstract}
69: \keywords{accretion --- black hole physics --- stars: individual 
70: (\IGR, Cyg X-1, XTE J1550$-$564, \SWIFT)}
71: 
72: 
73: \section{Introduction}
74: Transient X-ray Binaries (XRB) are known to show a series of different X-ray 
75: spectral states during their outbursts. A certain number of these 
76: objects have, however, undergone outbursts during which they remained in the Low Hard 
77: State (LHS) \cite[e.g.][]{hynes00,brocksopp04,rodriguez06b, cadolle06a}. A strong correlation
78: exists between the radio and X-ray fluxes during this state \citep{corbel00,gallo03}.
79: This may indicate some influence of the radio jet in 
80: the X-ray domain. The study of the LHS should, therefore, reveal important clues 
81: to the connection between the accretion and ejection processes. Interestingly, 
82: black hole (BH) transients are more radio loud than neutron stars (NS) by a factor of 
83: $\sim$30 \citep{miggliari06} \correc{in the LHS}.\\
84: \indent \IGR\ was discovered with ISGRI on-board \integral\ on Sept. 17, 2006
85:  as a new hard XRB \citep{soldi06} \correc{at Galactic coordinates l=0.97$^\circ$, 
86: b=-0.46$^\circ$.} \correc{Given that the line of sight to the source passes close to 
87: the Galactic Center, and assuming a  distance of 8 kpc, \citet{kuulkers06} 
88: estimated a 2-100 keV (unabsorbed) luminosity of $\sim10^{37}$ erg/s. The position and  
89: luminosity strongly indicate \IGR\ is a Galactic XRB.
90: The preliminary spectral analysis of their \integral\ data led \citet{kuulkers06} to further 
91: suggest the source was in a LHS typical of BH and NS XRBs}. A follow up 
92: observation with  the Chandra X-ray Observatory allowed a fine X-ray position to 
93: be given at \correc{$\alpha=17^h 49^m 38.037^s$, $\delta=-28^\circ 21' 17.37''$ 
94: ($\pm0.6''$, at 90\% confidence, Paizis et al. 2006). This refined X-ray position 
95: permitted \citet{paizis06}
96: to identify the probable optical/IR counterpart, and classify it as a red giant K-type
97: star making \IGR\ a Low Mass X-ray Binary (LMXB).}\\
98: \indent Soon after the discovery, we triggered our \rxte\ program P92016 as well as radio 
99: observations with the Australian Telescope Compact Array 
100: (ATCA), aiming to identify the nature of the system. 
101: We report here the results of those campaigns.
102: 
103: \section{Observations and Data Analysis}
104: \subsection{X-ray observations}
105: Our\rxte\ program has been divided into seven pointings, the details 
106: of which are reported in Table \ref{tab:log}. Fig. \ref{fig:Lite} shows the 
107: 3--30 keV \rxte/PCA and 15--50 keV Swift/BAT light curves of \IGR\ over the outburst.
108: The PCA data were reduced in the same  way as in \citet{rodriguez03}, 
109: although with the latest version of the {\tt{HEASOFT}} software package. 
110: Since the source is located in a crowded 
111: area, standard background subtraction is not sufficient to remove the contribution
112: of other sources and the Galactic Bulge (about 10 mCrab between 2-10 keV, 
113: \citet{markwardt06}). In order to better estimate the sky background at the source 
114: position, we analyzed an \rxte\ observation of the nearby
115: BH Candidate XTE J1748-288 made at the end of its 1998 outburst, when the 
116: source flux was comparable with that of the Galactic background (Obs. Id 30185-01-20-00).
117: The best fitted model of this observation was then used as an extra background 
118: component for the spectral fittings of all our observations.\\
119: \indent For HEXTE, we extracted spectral data from the Cluster B 
120: unit only, since Cluster A  no longer obtains a background measurement because the rocking
121: has \correc{ceased}. We checked that ``plus'' and ``minus''
122: offset pointings gave similar spectra (indicating that no active sources 
123: contributed to the background) and combined the two pointings to obtain background 
124: spectra. The remaining processes 
125: for HEXTE reduction are similar to the procedures presented in \citet{rodriguez03}.
126: The PCA and HEXTE spectra of each observation were fitted simultaneously in 
127: {\tt{XSPEC}} v11.3.2t between 3 and 200 keV. A normalization constant 
128: was introduced to account for calibration uncertainties between 
129: PCA and the single HEXTE cluster.\\
130: \indent  High resolution light curves were extracted from Good Xenon and Event 
131: data with a time resolution of $\sim122\mu$s allowing us to study the temporal properties 
132: of the source up to 4096 Hz. In order to restrict the background effects at low and high 
133: energies, we restricted the extraction to the $\sim5$--$40$ keV range. Since QPOs and 
134: coherent pulsations are usually stronger at these energies 
135: \citep[e.g.][]{morgan97,rodriguez04,rodriguez06a}, this ensures a higher sensitivity to any
136: feature that might exist.  
137: We produced power density spectra (PDS) from each of the individual 
138: light curves between $\sim0.02$ and 4096 Hz. 
139: \subsection{Radio observations}
140: \IGR\ was observed on  Sept. 26 and 27, 2006, with the ATCA, 
141: located at Narrabri, Australia. The array was in the H75 configuration 
142: (allowing a baseline as long as 6 km)  with antennas 1 and 5 off-line 
143: due to maintenance. Observations 
144: were carried out simultaneously at 4.80 GHz (6 cm) and 8.64 GHz (3.5 cm), 
145: with a continuum bandwidth of 128 MHz, for a total of about 12.3 hours on source. 
146: The amplitude and band-pass calibrator was PKS~1934$-$638, and the antennas' 
147: gain and phase calibration, as well as the polarization leakage, were 
148: derived from regular observations of the
149: nearby calibrator TXS 1748$-$253. The editing, calibration, Fourier transformation, 
150: deconvolution, and image analysis 
151: were performed using the MIRIAD software package (Sault \& Killen 1998). 
152: No radio emission from \IGR\ was detected 
153: with ATCA on either date. Combining the two sets of observations 
154: resulted in the three sigma upper limits for \IGR\ of 0.21 mJy at 
155: both frequencies.
156: 
157: \section{Results of the X-ray analysis}
158: Fig. \ref{fig:Lite} shows the light curves seen by Swift/BAT and \rxte/PCA. 
159:  In about 5 days the 15-50 keV BAT flux increased by a factor greater than 5, and  
160: remained around 0.02 cts/s until MJD 54006. The BAT flux is not purely constant though, 
161: since it showed  the presence of two peaks.  After MJD 54006 
162: a significant decrease is seen (Fig. \ref{fig:Lite}). From the quality of the data, 
163: it is difficult to say whether the shape of the BAT light curve 
164: is FRED-like,  or if the decay is purely linear. In fact,  
165: fitting the decay with an exponential leads to an e-folding time of 16.9 days. \\
166: \indent The spectra were first fitted with a simple absorbed 
167: power law model. In all cases, the power law photon index was hard ($\sim1.56$).
168: Although the reduced $\chi^2$ was relatively acceptable (1.5 for 96 
169: degrees of freedom (DOF) for Obs. 1), a slight deviation 
170: was visible at high energy. Adding a cutoff improved the fit. An F-test
171: yielded a probability of 2.8$\times$10$^{-3}$ that the improvement 
172: was purely due to chance. In all observations except 4, 5 and 6 the cutoff 
173: was required at a high level of significance.  This may indicate that 
174: it was either truly absent in obs. 4, 5 and 6, or that the cutoff energy 
175: had increased closer to the upper boundary of our spectral analysis. Note that these 
176: three observations occur between the two peaks of the BAT light curve (Fig. \ref{fig:Lite}).\\
177: \indent Since a hard power law with an exponential cutoff is usually interpreted 
178: in XRBs as being due to thermal Comptonization, the pure phenomenological model was 
179: replaced by the {\tt{comptt}} model \citep{titarchuk94}. The same model was applied to all
180: observations to ease the comparison. 
181: In all cases, a hint for an iron edge at $\sim7$ keV was visible. Adding such a feature greatly improved 
182: the fits (1.7$\times$10$^{-9}$ chance improvement). The mention by \citet{itoh06} of such a 
183: feature in the {\it{Suzaku}} spectrum of the source lent further credibility to this detection.  
184: The optical depth of the Edge ranges from 0.08 to 0.12, while the 
185: reduced $\chi^2$ ranges from 0.97 to 1.14 for 93 DOF. All other  
186: fit parameters are reported in Table \ref{tab:fits}, while Fig. \ref{fig:spectrum} shows 
187: the   spectrum from Obs. 7.  The source spectral parameters were 
188: relatively stable
189: over the period of observations. Slight differences are found only for Obs. 1 \& 7, for which 
190: the electron temperatures seem higher, although poorly constrained.\\
191: \indent The 1-s bin PCA light curves extracted from the 7 observations showed no particular differences. 
192:  No obvious dips, eclipses, or X-ray bursts were 
193: visible\footnote{In the PCU 3 light curve of Obs. 2, however a large flare was 
194: detected. This was not confirmed while inspecting all other active PCUs. This flare 
195: seems to be purely instrumental.}. 
196: To quantify the degree of variability, we inspected each PDS 
197: individually. The variability of \IGR\ is dominated by frequencies lower than 1 Hz 
198: (Fig. \ref{fig:pds} for Obs. 1). We modeled the PDSs with the help 
199: of broad Lorentzians \citep{belloni02}. As in typical PDSs of XRBs, three such features
200:  are needed \citep{vanderklis06}, two with their 
201: centroid frozen to 0 and one representing the so called 
202: low frequency hump (LFH). Broad Lorentzians mimic a flat top component 
203: and power law decay above a break frequency. The analogue of the break frequency is 
204: given by $\Delta=\sqrt{(\sigma/2)^2+\nu_{0}^2}$, with $\sigma$ the full width at half
205:  maximum, and 
206: $\nu_0$ the centroid frequency \citep{belloni02}. The PDSs from the different observations
207: showed very little variations. The 3 components  were compatible (within the errors) 
208: with being constant through the outburst. As a reference the parameters we obtained  
209: for Obs. 1 
210: are $\Delta=0.06\pm0.01$ Hz, RMS$\sim13\pm2\%$, for the main component $\Delta=1.8\pm0.7$ 
211: Hz, RMS$=10\pm2\%$
212: for the second, and $\nu_0=0.30\pm0.07$ Hz, $\Delta=0.23\pm0.07$ Hz, RMS$=13\pm2\%$ for 
213: the LFH. 
214: The  \correc{total} RMS variability is then $\sim$36$\%$ RMS. The  $3\sigma$ upper 
215: limit for a 2 Hz FWHM QPO is 2\%. \\
216: \indent \correc{We searched for X-ray pulsations between $\sim0.2$ ms and 256 s in the 
217: unbinned PDSs. We do not detect any coherent pulsations in any of them. While 
218: below 1 Hz the significant level of aperiodic noise  renders our search less 
219: sensitive to the detection of a given feature, above 1 Hz a periodic signal would be more 
220: readily detectable out of the white noise. 
221: From the PDS of  Obs. 1, and with a 
222: frequency resolution of 3.9$\times10^{-3}$ Hz, we 
223: could calculate $3\sigma$ upper limits of $0.9\%$ between 1 and 4096 Hz. Below 1 Hz, 
224: using a similar approach (assuming here that a pulsation should appear out of 
225: the level of aperiodic noise, and using the ``flat top'' level below 0.3 Hz), 
226: we could calculate that the limit for a coherent pulsation 
227: ranges from $\sim2.4 \%$ at $0.4$ mHz to $\sim1\%$ at 1 Hz.}
228: 
229: \section{Discussion}
230: We analyzed a set of \rxte\ observations of the newly-discovered source \IGR.
231: The PDSs of \IGR\ are typical of XRBs in the LHS \cite[e.g.][]{vanderklis06}. Although 
232: the general
233: shape of the PDSs makes it difficult to discriminate a NS from a BH \citep{vanderklis06}, 
234: the typical frequencies are believed to be lower in the case of BH (the ms
235: accreting pulsar IGR J00291+5934 being a recent counter example). In fact \citet{sunyaev00} 
236: observed that in the LHS for BH systems no significant signal is 
237: detected above $\sim50$ Hz, contrary to weakly magnetized NS. In \IGR\  
238: the lack of significant signal above $\sim$10 Hz, \correc{above which the level of variability 
239: is compatible with being purely due to Poisson noise}, and the high level of 
240: the flat top noise together with the low frequency of the break 
241: \correc{in the first Lorentzian 
242: component (0.06 Hz)} are reminiscent of BH systems. 
243: Generally speaking, and although this is not a proof, none of the characteristics that 
244: would make \IGR\ a definite NS (kHz QPOs, X-ray bursts, coherent pulsations) are observed. \\
245: \indent The source spectrum is well represented by a power law (with $\Gamma\sim 1.57$) 
246: convolved by interstellar absorption and a high-energy cutoff \correc{starting at  
247: $E_{\mathrm{cut}}\sim 50$ keV with a folding energy $E_{\mathrm{fold}}\sim190$ keV}, which are  
248: all typical of XRBs in the LHS.  
249: This is usually interpreted as due to thermal Comptonization of soft X-ray photons 
250: on a population of energetic electrons. Whether these electrons form a ``corona'' or the base of 
251: a compact jet is subject to debate \correc{\citep[e.g.][]{markoff05}}. 
252: A thermal Comptonization model represents the spectra well.  
253: The high value of the electron temperature and the optical depth, suggest 
254: that the compact object  probably  a BH. Indeed, with $E_{cut}=3\times kT_{e}$, we obtain 
255: an  equivalent cutoff energy of about 100-120 keV which is similar to e.g. Cyg X-1 while in the
256: LHS \citep{cadolle06}, or XTE J1550$-$564 during its 2002 outburst \citep{belloni02b}), two well
257: known BHs. In addition, even if some extremely hard NS Atoll sources can show similar spectra,  
258: the values 
259: of $\tau$ obtained in those cases are systematically higher \citep{barret01} than those of \IGR.\\
260: \indent The (unabsorbed) 1-20 and 20-200 keV luminosities are 
261: $1.1\times10^{37}\times (d/8kpc)^2$erg/s and 
262: $2.1\times10^{37}\times (d/8kpc)^2$erg/s. At the distance of the Galactic Center,  these 
263: luminosities place the source outside the so-called burster box \citep{barret00}, 
264: in fact very close to Cyg X-1 in its LHS. 
265: Although several NS sources have been seen to lie outside this box, they all have their
266:  20--200 keV 
267: luminosity lower than $\sim10^{37}\times$erg/s. 
268: In this respect, if \IGR\ is located at a large distance, as possibly 
269: suggested by the high value of the equivalent absorption column density, its hard luminosity
270: strongly suggests that it contains a BH. \\
271: \indent  Fig. \ref{fig:correl} shows
272: the position of \IGR\ w.r.t. the radio-X-ray correlation. \IGR\ lies 
273: below the expected correlation for BHs.
274: Fig \ref{fig:correl} \citep[see][]{corbel04,cadolle06a} shows that some BH(C)s have been 
275: observed to lie below the expected correlation. The 
276: non-detection of \IGR\ in radio is therefore not an argument against a BH primary. 
277: Given the results of our X-ray spectral analysis, we conclude that \IGR\ is a BHC in its
278: LHS. \correc{The observations of sources for which the correlation between the radio 
279: and X-ray fluxes 
280: is not observed suggest that, contrary to other BHs in the LHS, the formation and/or 
281: emission from the jet may be prevented for some unknown reasons.} \\
282: \indent Recently several sources have been observed to remain in the LHS during the whole
283:  duration 
284: of their outburst. These sources can either \correc{host} a NS (e.g. Aql X-1), or a BH(C) 
285: (XTE J1550$-$564, \SWIFT, GRO J0422+32, GRO J1719-24, and XTE J1118+480; 
286: \citet{sturner05,cadolle06a, brocksopp04}). Interestingly Aql X-1 and XTE J1550$-$564 
287: are also known to undergo major outbursts. The existence of LHS outbursts in XRBs brings 
288: interesting 
289: questions regarding the physics of accretion. In XTE J1550$-$564, \citet{sturner05} have 
290: suggested
291: that this outburst could correspond to a discrete accretion event. \citet{rodriguez06b} 
292: have suggested 
293: the same could also occur in Aql X-1. In both these cases, the outbursts had a short duration 
294: ($\sim$30-40 days) compared to their previous outbursts. The similar duration of the 
295: outburst of
296: \IGR\ could by comparison indicate that a similar event took place. As pointed out 
297: by \citet{sturner05} 
298: the LHS outburst in GRO J0422+32, GRO J1719-24, and XTE J1118+480 could (by opposition to at 
299: least XTE J1550$-$564) be explained by a lower reservoir of material, these sources having 
300: a short orbital periods. \correc{Although we cannot know
301: whether IGR J17497-2821 will have a major outburst in the future, the fact that 
302: the companion of \IGR\ is a red giant K-type star \citep{paizis06} 
303: rules out a system with such a short orbital period.}
304: 
305: \begin{acknowledgements}
306: We are especially grateful to J. H. Swank and the RXTE mission planers 
307: for having 
308: accepted our ToO, and for their very rapid reaction to plan our observations. 
309: The Swift/BAT transient monitor results are kindly provided by the Swift/BAT team.
310: J.R. would like to thank S. Chaty for useful discussions.
311: This research has made use of data obtained through the High Energy 
312: Astrophysics Science Archive Center Online Service, provided by the 
313: NASA/Goddard Space Flight Center.
314: \end{acknowledgements}
315: 
316: 
317: %\bibliography{ms}
318: 
319: \begin{thebibliography}{21}
320: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
321: \bibitem[Barret(2001)]{barret01} Barret, D. 2001, Adv. Space Res, 28, 307
322: \bibitem[Barret et al.(2000)]{barret00} Barret, D., Olive, J.-F., Boirin, L., et al.  2000, ApJ, 533, 329
323: \bibitem[Belloni et al(2002a)]{belloni02}Belloni, T., Psaltis, D., van der Klis, M. 2002a, ApJ, 72, 392
324: \bibitem[Belloni et al(2002b)]{belloni02b}Belloni, T.,Colombo, A.P., Homan, J., Campana, S., van der Klis, M. 2002b, A\&A, 390, 199
325: \bibitem[Brocksopp et al.(2004)]{brocksopp04}Brocksopp, C., Bandyopadhyay, R.M., Fender, R.P.  2004, NewA., 9, 249
326: \bibitem[Cadolle Bel et al.(2006a)]{cadolle06a} Cadolle Bel, M., Rib\'o, M., Rodriguez, J., et al., 2006a, submitted to ApJ
327: \bibitem[Cadolle Bel et al.(2006b)]{cadolle06} Cadolle Bel, M.,  et al. 2006b, A\&A, 446, 591
328: \bibitem[Corbel et al.(2003)]{corbel00} Corbel, S., Nowak, 
329: M.~A., Fender, R.~P., Tzioumis, A.~K., \& Markoff, S.\ 2003, \aap, 400, 
330: 1007 
331: \bibitem[Corbel et al.(2004)]{corbel04} Corbel, S., Fender, R.P., Tomsick, J.A, et al. 2004, ApJ, 617, 1272
332: \bibitem[Dickey \& Lockman(1990)]{dickey}Dickey, J.M. \& Lockman, F.J. 1990, ARA\&A, 28, 215
333: \bibitem[Gallo et al.(2003)]{gallo03} Gallo, E., Fender, R.P., Pooley, G.G. 2003, MNRAS, 344, 60
334: \bibitem[Hynes et al.(2000)]{hynes00} Hynes, R.~I., Mauche, C.~W., Haswell, C.~A., Shrader, C.~R., Cui, W., \& Chaty, S.\ 2000, \apjl, 539, L37 
335: \bibitem[Itoh et al.(2006)]{itoh06} Itoh, T., Kokubun, M., Yuasa, T., et al. 2006  ATel 914
336: \bibitem[van der Klis(2006)]{vanderklis06} van der Klis 2006, in Compact Stellar X-ray sources, Eds. Lewin \& van der Klis, Cambridge University Press
337: \bibitem[Kuulkers et al.(2006)]{kuulkers06} Kuulkeers, E., Chenevez, J., Shaw, S., et al. 2006, ATel 888
338: \bibitem[Maccarone(2003)]{maccarone} Maccarone, T.J. 2003, A\&A, 409, 697
339: \bibitem[Markoff et al.(2005)]{markoff05} Markoff, S., Nowak, M.~A., \& Wilms, J.\ 2005, \apj, 635, 1203 
340: \bibitem[Markwardt \& Swank(2006)]{markwardt06}Markwardt, C.B. \& Swank, J.H. 2006 ATel 891
341: \bibitem[Migliari \& Fender(2006)]{miggliari06}Migliari, S. \& Fender, R.P. 2006, MNRAS, 366, 79
342: \bibitem[Morgan, Remillard \& Greiner(1997)]{morgan97} Morgan, E.H., Remillard, R.A \& Greiner, J. 1997, ApJ, 482, 993 
343: \bibitem[Paizis et al.(2006)]{paizis06}Paizis, A., Nowak, M., Chaty, S., Rodriguez, J., Courvoisier, T. J.-L. et al. 2006, submitted to ApJ, astro-ph 0611344
344: \bibitem[Rodriguez et al.(2003) Rodriguez, Corbel \& Tomsick]{rodriguez03} Rodriguez, J., Corbel, S. \& Tomsick, J.A. 2003, ApJ, 595, 1032
345: \bibitem[Rodriguez et al.(2004)]{rodriguez04} Rodriguez, J., Corbel, Hannikainen, D.C., Belloni, T. Paizis, A. \& Vilhu, O. 2004, ApJ, 615, 416
346: \bibitem[Rodriguez et al.(2006a)]{rodriguez06b} Rodriguez, J., Shaw, S.E. \& Corbel, S.  2006a, A\&A, 451, 1045
347: \bibitem[Rodriguez et al.(2006b)]{rodriguez06a} Rodriguez, J., Bodaghee, A., Kaaret, P., et al.  2006b, MNRAS, 366, 274
348: \bibitem[Soldi et al.(2006)]{soldi06} Soldi, S., Walter, R., Eckert, D., et al. 2006, ATel 885
349: \bibitem[Sturner \& Shrader(2005)]{sturner05}Sturner, S.J., Schrader, C.R. 2005, ApJ, 625, 923
350: \bibitem[Sunyaev \& Revnivtsev(2000)]{sunyaev00}Sunyaev, R. \& Revnivtsev, M. 2000 A\&A, 358, 617
351: \bibitem[Titarchuk(1994)]{titarchuk94} Titarchuk, L. 1994, ApJ, 434, 313
352: \end{thebibliography}
353: 
354: \begin{table}[htbp]
355: \caption{Journal of the \rxte\ observations analyzed in this paper. MJD 54000 is Sept. 22., 2006 
356:  The last two columns indicate the net count 
357: rates from PCA/PCU2 (3-30 keV) and HEXTE/ClusterB (18-200 keV).}
358: \begin{tabular}{cccccc}
359: \hline
360: Obs.& Obs. Id & MJD  & GTI & \multicolumn{2}{c}{Count rates}\\
361: \#  & (P92016) & (d)  & (s) & PCA & HEXTE \\
362: \hline
363: \hline
364: 1 & 01-01-00 & 53998.2 & 6384 &  61.4 & 13.3\\
365: 2 & 01-02-00 & 54000.2 & 3200 & 67.5 & 14.5\\
366: 3 & 01-01-02 & 54001.3 & 3200 & 68.4& 14.3\\
367: 4 & 01-02-01 & 54002.0 & 2368 & 66.7& 14.3\\
368: 5 & 01-02-02 & 54003.2 & 3216  & 64.5& 15.3\\
369: 6 & 01-01-01 & 54005.0 & 3232 & 60.8 & 13.0\\
370: 7 & 01-03-00 & 54007.1 & 6480 &53.8& 12.2\\
371: \end{tabular}
372: \label{tab:log}
373: \end{table}
374: \begin{table}[htbp]
375: \caption{Best fit parameters obtained from the combined PCA+HEXTE spectra.
376: The spectral model consists of {\tt{phabs*edge*comptt}}. In all cases the reduced 
377: $\chi^2$ is around 1.
378: The seed photon temperature for Comptonization was frozen to 0.1 keV in all cases. 
379: Errors are given at the 90\% confidence level. $^\dagger$ $\times 10^{22}$ cm$^{-2}$.
380: $^\star$3-20 keV unabsorbed  $\times10^{-9}$\ergcms. }
381: \begin{tabular}{cccccc}
382: \hline
383: \hline
384: Obs.& N$_{\mathrm{H}}^\dagger$ & Edge& kT$_e$ & $\tau$ & flux$^\star$\\
385:    \#       &  & keV& keV & &  \\
386: \hline
387: 1       & 4.1$\pm0.4$ & 6.8$\pm0.3$ & 37$_{-7}^{+118}$ & 1.6$_{-1.0}^{+0.3}$ & 1.0\\
388: 2        & 3.8$\pm0.4$ & 6.6$\pm0.3$ & 34$_{-6}^{+21}$ & 1.8$_{-0.6}^{+0.3}$ & 1.14 \\
389: 3        & 3.9$\pm0.4$ & 6.7$\pm0.3$ & 33$_{-5}^{+12}$ & 1.8$_{-0.4}^{+0.2}$ & 1.16\\
390: 4        & 3.8$\pm0.5$ & 7.1$_{-0.3}^{+0.4}$ & 33$_{-5}^{+16}$ & 1.9$_{-0.5}^{+0.3}$  & 1.13\\
391: 5       &3.3$\pm0.4$ & 6.8$_{-0.3}^{+0.4}$ & 37$_{-6}^{+25}$ & 1.7$_{-0.5}^{+0.3}$ & 1.08 \\
392: 6       & 3.5$\pm0.4$ & 6.8$_{-0.3}^{+0.4}$ & 40$_{-8}^{+22}$ & 1.7$_{-0.5}^{+0.3}$ & 1.02 \\
393: 7        &3.6$\pm0.4$ & 6.8$\pm0.3$ & 38$_{-7}^{+88.0}$ & 1.7$_{-1.1}^{+0.3}$ & 0.91 \\
394: \end{tabular}
395: \label{tab:fits}
396: \end{table}
397: 
398: \clearpage
399: 
400: \begin{figure}
401: \epsscale{0.4}
402: \plotone{f1.eps}
403: \caption{15--50 keV Swift/BAT (open circles) and 3--30 keV \rxte/PCA (crosses) light curves
404:  of \IGR\ during the outburst studied in this paper. The vertical arrows represent 
405: the dates of our ATCA observations. For BAT 1 Crab $\sim0.23$ cts/s and $\sim1840$ cts/s for PCA.}
406: \label{fig:Lite}
407: \end{figure}
408: 
409: \begin{figure}
410: \epsscale{0.4}
411: \plotone{f2.eps}
412: \caption{3-200 keV \rxte\ spectrum of Obs. 7. The best model is superimposed 
413: as a line.}
414: \label{fig:spectrum}
415: \end{figure}
416: 
417: \begin{figure}
418: \epsscale{0.4}
419: \plotone{f3.ps}
420: \caption{White noise corrected PDS from Obs. 1. The continuous lines represents the 3 broad Lorentzians 
421: used to model the PDS. The dashed lines is the best model.}
422: \label{fig:pds}
423: \end{figure}
424: 
425: \begin{figure}
426: \epsscale{0.5}
427: \plotone{f4.ps}
428: \caption{Plot of the radio vs. X-ray fluxes for the BH XTE J1650$-$500 \citep{corbel04}, the BHC 
429: Swift J1753.5$-$0127 \citep{cadolle06a}, and \IGR\, with their (assumed) distances. All fluxes 
430: are normalized to a distance of 1 kpc. The range of correlated radio/X-ray fluxes is indicated 
431: by the dashed dot lines, while the continuous line indicates the best position. The vertical dotted 
432: lines indicate 2\% and 5\% of the Eddington luminosity.}
433: \label{fig:correl}
434: \end{figure}
435: 
436: \end{document}
437: