1: %\documentclass[12pt,preprint]{aastex}
2: %% manuscript produces a one-column, double-spaced document:
3: %% preprint2 produces a double-column, single-spaced document:
4: %\documentclass[preprint2]{aastex}
5: \documentclass{emulateapj} %for page limit
6: %\documentclass[12pt,manuscript]{aastex} %for submission
7: %\usepackage{apjfonts}
8:
9:
10:
11: %\usepackage{graphics,psfrag}
12:
13: \shorttitle{V5116 Sgr, an eclisping supersoft post-outburst nova?}
14: \shortauthors{Sala, Hernanz, and Ferri}
15:
16: \begin{document}
17:
18: \title{V5116 Sgr, an eclipsing supersoft post-outburst nova?}
19:
20: \author{G. Sala\altaffilmark{1},
21: M. Hernanz\altaffilmark{2},
22: C. Ferri\altaffilmark{2},
23: J. Greiner\altaffilmark{1}}
24:
25: \altaffiltext{1}{Max-Planck-Institut f\"ur extraterrestrische Physik,
26: PO Box 1312, D-85741 Garching b.M., Germany; gsala@mpe.mpg.de, jcg@mpe.mpg.de}
27: \altaffiltext{2}{Institut de Ci\`encies de l'Espai (CSIC-IEEC),
28: Campus UAB, F. Ci\`encies,
29: 08193 Bellaterra, Spain; hernanz@ieec.uab.es, ferri@ieec.uab.es}
30:
31: \begin{abstract}
32: V5116 Sgr (Nova Sgr 2005 No. 2), discovered on 2005 July 4,
33: was observed with XMM-Newton in March 2007, 20 months after the optical outburst.
34: The X-ray spectrum shows that the nova had evolved to a pure supersoft X-ray source,
35: with no significant emission at energies above 1~keV.
36: The X-ray light-curve shows abrupt decreases and increases
37: of the flux by a factor $\sim8$.
38: It is consistent with a periodicity of 2.97~h,
39: the orbital period suggested by \cite{dob07},
40: although the observation lasted just a little more than a whole period.
41: We estimate the distance to V5116~Sgr to be $11\pm3\,kpc$.
42: A simple blackbody model does not fit correctly the EPIC spectra, with
43: $\chi^2_{\nu}>4$. In contrast, ONe rich white dwarf atmosphere models provide
44: a good fit, with
45: $N_{\rm H}= 1.3(\pm0.1)\times 10^{21}\,cm^{-2}$,
46: $T=6.1(\pm0.1)\times10^5\,K$, and
47: $L=3.9(\pm0.8)\times10^{37}(D/10~\rm{kpc})^2{\rm erg}\,{\rm s}^{-1}$
48: (during the high-flux periods).
49: This is consistent with residual hydrogen burning in the white dwarf envelope.
50: The white dwarf atmosphere temperature is the same both in the low and the high
51: flux periods, ruling out an intrinsic variation of the X-ray source as the origin of the flux changes.
52: We speculate that the X-ray light-curve may result from a partial coverage by an asymmetric
53: accretion disk in a high inclination system.
54:
55: \end{abstract}
56:
57: \keywords{stars: novae, cataclysmic variables --- X-rays: individual (V5116 Sgr)}
58:
59: \section{Introduction}
60:
61: Classical novae occur in close binary systems of the cataclysmic variable type,
62: when a thermonuclear runaway results in explosive
63: hydrogen burning on the accreting white dwarf (WD). It is theoretically predicted that
64: novae return to hydrostatic equilibrium after the ejection of a fraction of the
65: accreted envelope.
66: Soft X-ray emission arises in some post-outburst novae
67: as a consequence of residual hydrogen burning on top of the WD.
68: As the envelope mass is depleted, the photospheric radius
69: decreases at constant bolometric luminosity
70: (close to the Eddington value) with an increasing effective temperature.
71: This leads to a hardening of the spectrum from optical through
72: UV, extreme UV and finally soft X-rays, with
73: the post-outburst nova emitting as a Supersoft Source (SSS) with a
74: hot WD atmosphere spectrum.
75: The duration of this SSS is related to
76: the nuclear burning timescale of the remaining H-rich envelope
77: and depends, among other factors, on the WD mass \citep{TT98,SH05a,SH05b}.
78:
79: All post-outburst novae are expected to undergo SSS emitting phase,
80: provided a H-rich envelope is left.
81: Nevertheless, the number and duration of SSS states observed in Galactic novae is small.
82: In a systematic search for the X-ray emission from novae in the ROSAT archive,
83: \cite{Ori01a} found only three SSS novae from a total of 39 novae observed less
84: than ten years after the outburst (GQ Mus, V1974~Cyg, N~LMC~1995).
85: After the ROSAT era, observations by Beppo-SAX and Chandra revealed SSS X-ray emission for
86: some more post-outburst novae: V382~Vel \citep{Ori02,Bur02},
87: V1494~Aql, with a puzzling light curve showing a short burst
88: and oscillations \citep{Dra03}, and V4743~Sgr, also with a very variable light-curve
89: \citep{Nes03,Pet05}.
90: The SSS phase of the recurrent
91: nova RS~Oph in the 2006 outburst was monitored with Swift, XMM-Newton and Chandra,
92: and the end of the SSS state occurred less than
93: 100 days after outburst \citep{Bode06,Hachi07}.
94: XMM-Newton has contributed with some more observations of novae.
95: Nova~LMC 1995 showed the SSS emission 5 years after outburst \citep{Ori03}.
96: Other post-outburst novae were detected with a harder spectrum
97: associated to the expanding ejecta, as
98: Nova LMC 2000 \citep{Gre03} and V4633 Sgr \citep{HS07},
99: or reestablished accretion flow \cite[V2487~Oph,][]{HS02,fer07}.
100: More recently, the SSS XMMSL1~J070542.7-381442
101: was identified as a nova \citep{rea07a,rea07b,tor07}.
102:
103: X-ray observations of the central area of M31,
104: with its moderate absorption, offer a good chance to monitor the SSS phase of novae:
105: \cite{pie05} reported 21 X-ray counterparts for novae in M31
106: -- mostly identified as SSS by their hardness
107: ratios -- and two in M33. Following that work, XMM-Newton and Chandra
108: monitoring of M31 between July 2004 and February 2005 provided the
109: detection of eleven out of 34 novae within a year after optical outburst \citep{pie07}.
110: This suggests that the low fraction of SSS found by \cite{Ori01a} was due to selection effects
111: and/or too poor sampling.
112:
113: X-ray observations of post-outburst novae
114: provide crucial information:
115: soft X-rays yield a unique insight
116: into the remaining burning envelope on top of the WD,
117: while hard X-rays reveal shocks in the nova ejecta.
118: The properties of ``quiescent novae'', once they have turned-off and once
119: accretion is reestablished (which can occur before or after the turn-off),
120: are revealed both in hard and soft X-rays;
121: they emit then as ``standard'' cataclysmic variables.
122: In view of the scarcity of objects observed and the diversity of
123: behaviors detected, only the monitoring of as many novae as possible,
124: with large sensitivity and spectral resolution (as those offered
125: by XMM-Newton and Chandra) can help to understand post-outburst novae.
126:
127: V5116 Sgr (Nova Sgr 2005 No. 2) was one of the targets included in our X-ray monitoring programme
128: of post-outburst Galactic novae with XMM-Newton.
129: It was discovered by \cite{lil05} on 2005 July 4.049~UT,
130: with magnitude $\sim$8.0, rising to mag 7.2 on July 5.085.
131: The expansion velocity derived from a sharp P-Cyg profile
132: detected in a spectrum taken on July 5.099 was $\sim$1300~km/s.
133: IR spectroscopy on July 15 showed emission lines with FWHM~$\sim$2200~km/s \citep{rus05}.
134: Photometric observations obtained during 13 nights in the period August-October 2006
135: show the optical light curve modulated with a period of $2.9712\pm0.0024$~h \citep{dob07},
136: which the authors interpret as the orbital period. They propose that the
137: light-curve indicates a high inclination system with an irradiation effect on the secondary star.
138: A first X-ray observation with Swift/XRT (0.3--10~keV) in August 2005 yielded
139: a marginal detection with 1.2($\pm$1.0)$\times10^{-3}$~cts/s \citep{nes07a}. Two years later,
140: on 2007 August 7, Swift/XRT showed the nova as a SSS,
141: with 0.56$\pm$0.1~cts/s \citep{nes07b}. A first fit with a blackbody indicated
142: $T\sim4.5\times10^5K$. More recently, a 35~ks Chandra spectrum obtained
143: on 2007 August 28 was fit with a WD atmospheric
144: model with N$_{\rm H}=4.3\times10^{21}$~cm$^2$ and $T=4.65\times10^5K$ \citep{NelOri07}.
145:
146:
147: Here we report on the X-ray light-curve and broad-band spectra of V5116 Sgr
148: as observed by XMM-Newton in March 2007, 609 days after outburst.
149:
150: \section{Observations and data analysis}
151:
152: V5116 Sgr was observed with XMM-Newton \citep{Jan01} on 2007 March 5
153: \cite[ observation ID: 0405600201, PI: Hernanz]{sal07}.
154: The exposure times were 12.7~ks for the European Photon Imaging Cameras (EPIC)
155: MOS1 and MOS2 \citep{Tur01}, 8.9~ks for the EPIC-pn \citep{Str01},
156: 12.9~ks for the Reflection Grating Spectrometers \cite[RGS,][]{Her01},
157: and 9.2~ks for the Optical Monitor with U filter \cite[OM,][]{Mas01}.
158:
159: The observation was affected by an X-ray storm, which produced
160: moderate background in the EPIC cameras for most of the exposure time.
161: Fortunately enough, our target was at least a factor 10 brighter than the background.
162: In addition, the source spectrum is very soft and little affected by solar flares.
163: We therefore do not exclude any time interval of our exposures, but we pay
164: special attention to the background subtraction of both spectra and light-curves.
165: Data were reduced using the XMM-Newton Science Analysis System
166: (SAS~7.1.0). Standard procedures described in the SAS documentation
167: \citep{sas05,abc04} were followed.
168: We applied flag filter FLAG=0, and only events with pattern smaller than 12~(MOS) and 4~(pn),
169: were selected for the analysis of the EPIC data.
170:
171: The high flux of the source resulted in pile-up in the three EPIC cameras (used in
172: large partial window mode). We minimize the pile-up effect by extracting both spectra and
173: light-curves from annular regions that exclude the central, piled-up pixels.
174: We determined the optimal extraction region by extracting the source spectra
175: using annular regions with increasing inner
176: radius until we found that the pile-up effect was minimized, keeping a
177: reasonable signal to noise. For the extraction of the source light-curve we use an
178: annular region with inner and outer radii 25 and 38~arcsec (for pn),
179: and 10 and 80~arcsec (for MOS) respectively for the whole exposure time.
180: Background for the MOS cameras was extracted from an annulus region centered on the source,
181: with an external radius of 140~arcsec and an internal radius of 115~arcsec.
182: In the case of the pn,
183: the CCD where the source was detected had no large area without source photons, so
184: we used two box regions in different CCDs, free of sources, for the background light-curve
185: determination.
186: Nevertheless, we could not extract a spectrum completely free of pile-up for the pn camera
187: and we consider only the MOS cameras for broad-band spectral analysis.
188: We increased the statistics for the MOS spectra during the low-flux periods by
189: reducing the inner radii of the source extraction regions down to 5~arcsec.
190: For the high-flux state we increased the inner radii to 40~arcsec to
191: avoid pile-up as much as possible.
192: We have also reduced the RGS data, not affected by pile-up, using the
193: standard procedures. We also find two well differentiated states, coincident
194: in time with the high and low rate periods of the EPIC cameras. The RGS spectra,
195: while compatible with the broad band models used to fit the MOS spectra in this paper,
196: show a number of absorption/emission features that require a detailed
197: spectral analysis, to be presented in a separate work.
198:
199:
200:
201: \section{Light-curve}
202:
203: Panel A in Figure~\ref{fig:lcs} shows the background subtracted light-curves
204: of the three EPIC cameras in the 0.2--1.0~keV band. There is no significant flux above 1.0~keV.
205: EPIC-pn experienced full scientific buffer at the beginning of the observation, and started to
206: collect data later than the other instruments. The Y-axis in the EPIC light-curves is only
207: in relative units, since the annulus extraction regions used do not provide
208: the total source rate. For comparison purposes,
209: the 0.2--1.0~keV count-rates have
210: been renormalized for the three EPIC cameras to 10~cts/s (close to the actual
211: maximum count-rate using the annulus extraction region,
212: 10.9~cts/s for MOS1, 10.2~cts/s for MOS2, and 13.1~cts/s in pn).
213:
214:
215: The X-ray light-curve shows clearly two states: a high-flux state, during the first 600s, and
216: again starting 8300~s after the start of the MOS exposures and lasting for 3000~s; and a low-flux state
217: between 600~s and 8300~s, and during the last 1200~s of the observation.
218: A small flare is observed during the last low-flux period.
219: The count-rate during the low-flux period between 600~s and 8300~s
220: does not show significant variations (25\% rms variability).
221: In contrast, the count-rate during the high-state shows flaring activity.
222: If the first decline after the initial high-flux period corresponds to the end of
223: a 3-ks burst like the one starting at 8300~s, this would point to
224: a periodicity in the X-ray light-curve of 10.5--10.7~ks (2.92--2.97~h).
225: This is consistent with the periodicity found in the optical light-curve in August--October 2006
226: by \cite{dob07} ($2.9712\pm0.0024$~h), which was suggested by the authors to be the orbital period.
227:
228: Panels B and C in Figure~\ref{fig:lcs} show the count-rates in two energy bands,
229: ~0.2--0.4~keV and ~0.4--0.6~keV respectively, for the three EPIC cameras.
230: EPIC-pn shows a softer spectral distribution than the two MOS cameras,
231: especially during the high-flux state. This indicates that pile-up is still
232: affecting the EPIC-pn data (and therefore we do not use them in the spectral analysis).
233: The source is very soft with low signal above 0.6~keV,
234: insufficient to define a third energy band above 0.6~keV.
235: The hardness ratio (Figure~\ref{fig:lcs}, panel~D) indicates that there are
236: no significant variations of the spectral properties during the whole observational period,
237: even when the rate is increasing by a factor 8--10 in the high-flux phases.
238:
239: Panel E in Figure~\ref{fig:lcs} shows the light-curve of the OM exposures in the U-band
240: in instrumental magnitudes. The light-curve of two stars in the field of view close
241: to V5116 Sgr are shown for comparison. The OM light-curve is
242: compatible with the 2.97~h periodicity observed by \cite{dob07},
243: but its minimum is not correlated either with
244: the maximum or the minimum in X-rays. This points
245: to different origins for the X-ray and UV emission.
246:
247:
248:
249: \begin{figure}
250: \bigskip
251: \epsscale{1.1}
252: \plotone{f1.eps}
253: \caption{V5116 Sgr X-ray and UV light-curves during the XMM-Newton obsevation on 2007 March 5.
254: Time units are TJD (JD-2440000.5) and the time bin is 50~s.
255: X-ray lightcurves are background substracted.
256: {\bf A:} EPIC MOS1 (black), MOS2 (red), and EPIC-pn (green) count-rates,
257: in the 0.2--1.0~keV energy range, normalized to 10 cts/s (see text for details).
258: {\bf B and C:} EPIC MOS1, MOS2 and pn count-rates in the bands 0.2--0.4~keV and
259: 0.4--0.6~keV. {\bf D:} Hardness ratio for the three EPIC cameras.
260: {\bf E:} OM U-band instrumental magnitudes for V5116 Sgr and for two comparison stars
261: in the OM FOV.}
262: \label{fig:lcs}
263: \end{figure}
264:
265:
266:
267: \begin{figure}
268: \bigskip
269: \epsscale{1.1}
270: \plotone{f2.eps}
271: \caption{XMM-Newton MOS1 (black) and MOS2 (red) spectra of V5116 Sgr.
272: {\bf A:} Low-flux spectra fit with a CO WD atmosphere model.
273: {\bf B:} Low-flux spectra fit with an ONe WD atmosphere model.
274: {\bf C:} High-flux MOS1 and MOS2 spectra fit with an ONe WD atmosphere model.}
275: \label{fig:spec}
276: \end{figure}
277:
278:
279: \section{Spectral analysis}
280:
281: The X-Ray Spectral Fitting Package \cite[XSPEC~11.3,][]{xsp03} was used
282: for spectral analysis. We used only MOS data for the spectral analysis, since
283: EPIC-pn is too affected by pile-up. Even for the EPIC-MOS cameras,
284: some residual pile-up is likely responsible for a component at energies twice
285: the maximum of the emission, i.e., 0.8-1.0~keV, caused by double events at 0.4-0.5~keV
286: counted as single photons with twice the energy. These lost photons are only a small
287: fraction of the total at the maximum of emission (and cause a small effect in the 0.3--0.5~keV band),
288: but they contribute notably at 0.8--1.0~keV,
289: where the intrinsic flux of the source is very small, if any.
290: This implies that we can trust the spectral analysis only up to 0.7~keV and we can not
291: provide any constraint on a possible reestablished accretion disk.
292: In our spectral fits, we use the {\it phabs} model with \cite{wil00}
293: abundances for the foreground absorption.
294:
295: Figure~\ref{fig:spec} shows the MOS1 and MOS2 spectra during the low-flux period.
296: A first attempt to fit a simple absorbed blackbody model to the data yields an
297: unacceptable fit, with $\chi^2_{\nu}>4$.
298: Therefore we used \cite{MV91} WD atmosphere emission models, gently
299: provided by Jim MacDonald, and
300: built tables to be read as external models in XSPEC.
301: Previous studies have shown the importance of including
302: WD atmosphere models instead of simple blackbodies
303: for a correct broad-band spectral analysis of novae in the SSS phase \citep{Bal98,BK01,OG99}.
304: Two sets of models are available, corresponding to CO and ONe compositions of
305: the WD.
306: As shown in Figure~\ref{fig:spec} (panel A), CO atmosphere models do not fit well the data.
307: The main difference between the two types of nova atmospheres at these temperatures is
308: the hard energy tail of ONe models, which is suppressed in CO models
309: due to the C ionization edges (CV at $\sim$0.4~keV and CVI at $\sim$0.5~keV).
310: The best fit is obtained with the ONe models
311: (Figure~\ref{fig:spec}, panel B), with
312: $N_{\rm H}=1.3(\pm0.1)\times 10^{21}cm^{-2}$,
313: $T=6.1(\pm0.1)\times10^5K$, and a luminosity (determined from the normalization constant)
314: $L=4.9(\pm0.6)\times 10^{36}(D/10~kpc)^2erg\,s^{-1}$.
315: The observed $N_{H}$ is consistent with the average interstellar absorption towards the source
316: \cite[$1.34\times 10^{21}cm^{-2}$,][]{LAB}.
317:
318:
319: The spectrum during the high-flux state (bottom panel in Figure~\ref{fig:spec})
320: is well fit with the same ONe WD atmosphere model, with
321: $N_{\rm H}= 1.5(\pm0.2)\times 10^{21}cm^{-2}$, $T=6.1(\pm0.1)\times10^5$K,
322: and a normalization constant a factor $\sim8$ larger,
323: corresponding to a luminosity
324: $L=3.9(\pm0.8)\times10^{37}(D/10~kpc)^2\,erg\,s^{-1}$.
325:
326:
327:
328:
329: \section{Discussion}
330:
331: V5116 Sgr was detected as a bright SSS by XMM-Newton 20 months after the nova outburst.
332: The temperature of the WD atmosphere indicates that residual
333: hydrogen burning is occurring in the WD envelope.
334: To further confirm the origin of the emission and determine the size of the emitting surface,
335: we need an estimation of the distance.
336: Fortunately enough, the time and magnitude at maximum are quite well determined for
337: V5116~Sgr, with a pre-maximum observation only 24~hours before the maximum.
338: \cite{lil05} reported $m_V=7.15$ at maximum, and $t_{2}$ (time required
339: to decline two magnitudes from maximum) was 6.5$\pm$1.0~days \citep{dob07}.
340: Using the empirical relation for novae between $M_V$ and $t_2$ \citep{del95} the
341: observed $t_2$ implies an absolute magnitude at maximum $M_V=-8.8(\pm0.4)$
342: (we add a 5\% error to account for the scatter in the $M_V-t_2$ relation).
343: The photometry at maximum indicates $B-V=+0.48$ \citep{gil05}, and for novae
344: at maximum, intrinsic $B-V=0.23\pm0.06$ \citep{ber87}. This implies
345: $A_{V}=3.1\,E_{B-V}=0.8\pm0.2$.
346: With all this, we estimate the distance to V5116~Sgr to be $d=11\pm3\,kpc$.
347: For simplicity, we scale the distance dependent parameters to 10~kpc.
348: It is worth noticing that the $N_{\rm H}$ obtained from our fits indicates $A_{V}=0.7$
349: \cite[$N_{\rm H}=5.9\times10^{21}E_{B-V}\,\mbox{cm}^{-2}$,][]{zom07},
350: consistent with the value obtained from the observed colors.
351:
352: The luminosity during the high flux periods, together with the atmosphere temperature,
353: indicate a radius of the emitting object
354: $R=6.2(\pm0.9)\times10^8(D/10kpc)\,cm$. This corresponds to
355: the whole WD surface emitting during the high flux periods and supports residual
356: H-burning as the origin of the SSS emission, rather than any feature related to the
357: accretion stream (boundary layer, hot spot, or hot accretion poles).
358:
359: The most remarkable feature in V5116~Sgr is the SSS light-curve.
360: Although the observation lasted just a bit longer than an orbital period,
361: we obtain a period compatible with the orbital period
362: of 2.97~hours found by \cite{dob07}.
363: The fact that the temperature is the same for both the high and the low flux phases
364: indicates that the change in flux cannot be caused by an intrinsic change
365: in the H-burning envelope. The luminosity
366: during the high flux periods also indicates that the whole WD
367: is visible in this state. The decrease by a factor $\sim8$
368: in the flux could be caused by a partial eclipse of the WD.
369: The long duration of the low flux phase and short duration of the high flux phase cannot
370: be reconciled with a partial eclipse by the secondary star. Alternatively,
371: some asymmetric accretion disk could do the job, being responsible for a partial eclipse
372: during most of the orbital period, with a 3~ks window during which the whole WD would be visible.
373: Since no change is observed in the hydrogen absorption column and it is compatible
374: with the average interstellar value both in the low and the high flux spectra,
375: the portion of accretion disk producing the partial eclipse should be optically thick
376: to the soft X-rays. In the particular case of V5116 Sgr, we have
377: checked that an absorbing column of $N_{H}>2\times10^{22}\,cm^{-2}$ would make it
378: completely opaque to the observed SSS emission.
379: Longer X-ray observations covering at least two periods
380: together with a good timing with the OM camera would be required
381: to disentangle the origin of the X-ray light curve behaviour.
382:
383: \acknowledgments
384: We thank Jim MacDonald for kindly providing his WD atmosphere
385: models. This work is based on observations obtained with XMM-Newton, an ESA science mission with
386: instruments and contributions directly funded by ESA Member States and NASA,
387: and supported by BMWI/DLR (FKZ 50 OX 0001) and the Max-Planck Society.
388: This research has been funded by the grants
389: AYA2004-06290-C02-01 from MEC, 2005-SGR00378 from AGAUR, and by FEDER
390: funds. GS is supported through DLR (FKZ 50 OR 0405) and CF through an
391: AGAUR FI fellowship from the Generalitat de Catalunya.
392:
393:
394:
395: \begin{thebibliography}{}
396: \bibitem[Arnaud \& Dorman(2003)]{xsp03} Arnaud \& Dorman 2003, XSPEC v11.3.x (NASA/GSFC, MD)
397: \bibitem[Balman et al.(1998)]{Bal98}Balman, S., Krautter, J. \& \"Ogelman, H. 1998, ApJ, 499, 395
398: \bibitem[Balman \& Krautter(2001)]{BK01}Balman, S. \& Krautter, J. 2001, MNRAS, 326, 1441
399: \bibitem[Bode et al.(2006)]{Bode06}Bode, M.F., O'Brien, T.J., Osborne, J.P. et al. 2006, ApJ, 652, 629
400: \bibitem[Burwitz et al.(2002)]{Bur02}Burwitz, V, Starrfield, S. et al. 2002, AIP Conf.Proc., 637, 386
401: \bibitem[Della Valle \& Livio(1995)]{del95} Della Valle, M. \& Livio, M. 1995, ApJ, 452, 704
402: \bibitem[den Herder et al.(2001)]{Her01}den Herder, J.W. et al. 2001, A\&A, 365, L7
403: \bibitem[Dobrotka et al.(2007)]{dob07} Dobrotka, A., Retter, A.\& Liu, A. 2007 (arxiv:0710.5701)
404: \bibitem[Drake et al.(2003)]{Dra03}Drake, J., Wagner, R., Starrfield, S. et al. 2003, ApJ, 584, 448
405: \bibitem[Ferri et al.(2007)]{fer07} Ferri, C., Hernanz, M. \& Sala, G. 2007, ASP Conf.Ser., 372, 519
406: \bibitem[Gilmore \& Kilmartin(2005)]{gil05}Gilmore, A.C. \& Kilmartin, P.M. 2005, IAUC 8559
407: \bibitem[Greiner et al.(2003)]{Gre03}Greiner, J., Orio, M. \& Schartel, N. 2003, A\&A, 405, 703
408: \bibitem[Hachisu et al.(2007)]{Hachi07}Hachisu, I., Kato, M. \& Luna, G.J.M. 2007, ApJ, 659, L153
409: \bibitem[Hernanz \& Sala(2002)]{HS02}Hernanz, M. \& Sala, G. 2002, Science, 298, 393.
410: \bibitem[Hernanz \& Sala(2007)]{HS07}Hernanz, M. \& Sala, G. 2007, ApJ, 664, 467.
411: \bibitem[Jansen et al.(2001)]{Jan01}Jansen, F. et al. 2001, A\&A, 365, L1
412: \bibitem[Kalberla et al.(2005)]{LAB} Kalberla,P., Burton, W. et al. 2005, A\&A, 440, 775
413: \bibitem[Liller(2005)]{lil05}Liller, W. 2005, IAUC 8559
414: \bibitem[Loiseau(2005)]{sas05}Loiseau, N. 2005, SAS User Guide v3.2 (ESA/XMM SOC, Madrid)
415: \bibitem[MacDonald \& Vennes(1991)]{MV91}MacDonald, J. \& Vennes, S. 1991, ApJ, 373, L51
416: \bibitem[Mason et al.(2001)]{Mas01}Mason, K.O. et al. 2001, A\&A, 365, L36
417: \bibitem[Nelson \& Orio(2007)]{NelOri07} Nelson, T \& Orio, M. 2007, ATel\#~1202
418: \bibitem[Ness et al.(2003)]{Nes03}Ness, J.U., Starrfield, S., Burwitz, V. et al. 2003, ApJ, 594, L127
419: \bibitem[Ness et al.(2007a)]{nes07a}Ness, J.U., Schwarz, G., Retter, A. et al. 2007a, ApJ, 663, 505
420: \bibitem[Ness et al.(2007b)]{nes07b}Ness, J.U., Starrfield, S., Schwarz, G. et al. 2007b, CBET\#~1030
421: \bibitem[Orio \& Greiner(1999)]{OG99}Orio, M. \& Greiner, J. 1999, A\&A, 344, L13
422: \bibitem[Orio et al.(2001)]{Ori01a}Orio, M., Covington, J. \& \"Ogelman, H. 2001, A\&A, 373, 542
423: \bibitem[Orio et al.(2002)]{Ori02}Orio, M., Parmar, A., Greiner, J. et al. 2002, MNRAS, 333, L11
424: \bibitem[Orio et al.(2003)]{Ori03}Orio, M., Hartmann, E., Still, M.\& Greiner, J. 2003, ApJ, 594, 435
425: \bibitem[Petz et al.(2005)]{Pet05}Petz,A., Hauschildt,P., Ness,J.U., Starrfield,S. 2005, A\&A, 431, 321
426: \bibitem[Pietsch et al.(2005)]{pie05} Pietsch, W., Fliri, J., Freyberg, M. et al. 2005, A\&A, 442, 879
427: \bibitem[Pietsch et al.(2007)]{pie07} Pietsch, W., Haberl, F., Sala, G. et al. 2007, A\&A, 465, 375
428: \bibitem[Read et al.(2007a)]{rea07a}Read, A., Saxton, R. \& Esquej, P. 2007a, ATel\#~1282
429: \bibitem[Read et al.(2007b)]{rea07b}Read, A., Saxton, R.~et~al. 2007b, ATel\#~1301
430: \bibitem[Russell et al.(2005)]{rus05}Russell, R., Lynch, D. \& Rudy, R. 2005, IAUC 8579
431: \bibitem[Sala \& Hernanz(2005a)]{SH05a}Sala, G. \& Hernanz, M. 2005a, A\&A, 439, 1057.
432: \bibitem[Sala \& Hernanz(2005b)]{SH05b}Sala, G. \& Hernanz, M. 2005b, A\&A, 439, 1061
433: \bibitem[Sala et al.(2007)]{sal07}Sala, G., Hernanz. M. \& Ferri, C. 2007, ATel\#1184
434: \bibitem[Snowden et al.(2004)]{abc04}Snowden, S. et al. 2004, XMM ABC v2.01 (NASA/GSFC, MD)
435: \bibitem[Str\"uder et al.(2001)]{Str01}Str\"uder, L. et al. 2001, A\&A, 365, L18
436: \bibitem[Torres et al.(2007)]{tor07}Torres, M., Jonker, P., Challis, P. et al. 2007, ATel\#~1285
437: \bibitem[Tuchman \& Truran(1998)]{TT98}Tuchman, Y. \& Truran, J.W. 1998, ApJ, 503, 381
438: \bibitem[Turner et al.(2001)]{Tur01}Turner, M.J.L. et al. 2001, A\&A, 365, L27
439: \bibitem[van den Bergh \& Younger(1987)]{ber87} van den Bergh, F. \& Younger, P.S. 1987, A\&AS, 70, 125
440: \bibitem[Wilms et al.(2000)]{wil00} Wilms, J., Allen, A. \& McCray, R. 2000, ApJ, 542, 914
441: \bibitem[Zombeck (2007)]{zom07} Zombeck,~M.V. 2007, Handbook of Space Astronomy and Astrophysics, Cambridge Univ. Press, 2007, 3rd ed.
442:
443:
444: \end{thebibliography}
445:
446:
447: \end{document}
448: