1: \documentclass[12pt,preprint]{aastex}

2: 

3: \newcommand{\rxte}{{\it RXTE}}

4: \newcommand{\xte}{{\it RXTE}}

5: \newcommand{\cxo}{{\it CXO}}

6: \newcommand{\cc}[1]{\multicolumn{1}{c}{#1}}

7: 

8: \newcommand{\saxsource}{SAX~J1748.9$-$2021}

9: 

10: \def\lapp{\ifmmode\stackrel{<}{_{\sim}}\else$\stackrel{<}{_{\sim}}$\fi}

11: \def\gapp{\ifmmode\stackrel{>}{_{\sim}}\else$\stackrel{>}{_{\sim}}$\fi}

12: 

13: \def\nuddd{\ifmmode\stackrel{\bf \,...}{\textstyle \nu}\else$\stackrel{\,...}{\textstyle \nu}$\fi}

14: \def\nudddd{\ifmmode\stackrel{\bf \,....}{\textstyle \nu}\else$\stackrel{\,....}{\textstyle \nu}$\fi}

15: 

16: \begin{document}

17: \title{Discovery of 442-Hz Pulsations from an X-ray  Source in the Globular

18: Cluster NGC~6440}

19: \author{Fotis P. Gavriil\altaffilmark{1}, Tod E. Strohmayer, Jean H. Swank, Craig B. Markwardt\altaffilmark{2,3}}

20: \altaffiltext{1}{NPP Fellow; Oak Ridge Associated Universities, Oak Ridge, TN}

21: \altaffiltext{2}{CRESST}

22: \altaffiltext{3}{Department of Astronomy, University of Maryland, College Park, MD 20742.}

23: \affil{NASA Goddard Space Flight Center, Astrophysics Science Division, Code 662, X-ray Astrophysics Laboratory, Greenbelt, MD 20771, USA} 

24: 

25: \begin{abstract}

26: We report on the serendipitous discovery of a 442-Hz pulsar during a

27: \textit{Rossi X-ray Timing Explorer} (\xte) observation of the

28: globular cluster NGC~6440.  The oscillation is detected following a

29: burst-like event which was decaying at the beginning of the

30: observation. The time scale of the decay suggests we may have seen the

31: tail-end of a long-duration burst.  Low-mass X-ray binaries (LMXBs)

32: are known to emit thermonuclear X-ray bursts that are sometimes

33: modulated by the spin frequency of the star, the so called burst

34: oscillations.  The pulsations reported here are peculiar if

35: interpreted as canonical burst oscillations.  In particular, the pulse

36: train lasted for $\sim$500 s, much longer than in standard burst

37: oscillations. The signal was highly coherent and drifted down by

38: $\sim$2$\times10^{-3}$Hz, much smaller than the $\sim$Hz drifts

39: typically observed during normal bursts.  The pulsations are

40: reminiscent of those observed during the much more energetic

41: ``superbursts'', however, the temporal profile and the energetics of

42: the burst suggest that it was not the tail end nor the precursor

43: feature of a superburst.  It is possible that we caught the tail end

44: of an outburst from a new `intermittent'' accreting X-ray millisecond

45: pulsar, a phenomenon which until now has only been seen in

46: HETE~J1900.1$-$2455 \citep{gmk+07}.  We note that \citet{kzht03}

47: reported the discovery of a 409.7~Hz burst oscillation from

48: \saxsource, also located in NGC~6440.  However, \textit{Chandra X-ray

49:   Observatory} imaging indicates it contains several point-like X-ray

50: sources, thus the 442~Hz object is likely a different source.

51: \end{abstract}

52: 

53: \keywords{--- stars: neutron --- X-rays: bursts --- globular clusters: individual (NGC~6440)}

54: %%%  --- X-rays: binaries

55: \section{Introduction}

56: \label{sec:intro}

57: %%%

58: The discovery of millisecond spin periods of neutron stars in Low Mass

59: X-ray Binaries (LMXBs) with the \textit{Rossi X-ray Timing Explorer}

60: (\xte) has helped elucidate the nature of these sources.

61: %%%

62: Neutron star LMXBs consist of a neutron star accreting from a low mass

63: companion. As material (mostly H and He) is accreted onto the star and

64: gets compressed, it eventually ignites and burns unstably

65: \citep[see][]{sb07}. This phenomenon is observed as a Type I X-ray

66: Burst. Type I X-ray bursts have been observed from over $\sim$70 LMXBs

67: \citep[see][and references therein]{lph06}. The recurrence time of these

68: bursts varies but in some cases it can be as frequent as every few

69: hours \citep[see][for examples]{gm07}. Occasionally it is possible to

70: observe the spin of the neutron star modulating the burst emission --

71: the so called ``burst oscillations'' \citep{szs+96}. Burst

72: oscillations have been observed from $\sim$18 LMXBs \citep[see][for

73:   examples]{gm07}.

74: 

75: X-ray bursts from LMXBs have been discovered which are $\sim$1000

76: times longer, and thus, that much more energetic, than canonical Type

77: I X-ray bursts. They are aptly named ``superbursts''. Superbursts are

78: believed to occur by the unstable burning of carbon

79: \citep{sb02,cb01}. \citet{sm02} discovered highly coherent pulsations

80: during a superburst from 4U~1636$-$53. The pulse train lasted for

81: $\sim$800~s, as opposed to the $\sim$10~s long pulse trains observed

82: in Type I X-ray bursts. Superbursts are much rarer than Type I X-ray

83: bursts. Thus far ten have been observed from eight LMXBs

84: \citep[see][]{zcc04}.

85: 

86: %% Accreting millisecond pulsars

87: %% Intermittent pulsars

88: 

89: \section{Observations}

90: \label{sec:observations}

91: 

92: Between 2005 March 7 and 2005 July 21 the PCA Galactic Bulge Scan

93: Survey\footnote{\texttt{http://lheawww.gsfc.nasa.gov/users/craigm/galscan/}}

94: discovered an outburst from the direction of the Globular Cluster

95: NGC~6440. A follow up 1.8-ks long pointed \xte\ observation

96: (observation identification number 91050-03-07-00) was initiated on

97: 2005 June 14.  The data presented here were acquired from the

98: Proportional Counter Array (PCA) on board \xte.  The PCA consists of

99: five identical and independent proportional counter units (PCUs). Each

100: PCU is a 90\% Xenon/10\% Methane gas filled proportional counter.

101: Each PCU has a collimated 1$\degr$$\times$1$\degr$ field-of-view, 256

102: spectral channels in the 2--60~keV range, and a limiting temporal

103: resolution of $\sim$1~$\mu$s.  Only two PCUs (PCU~0 and PCU~2) were

104: operational throughout the span of the observation.  The data were

105: taken in \texttt{E\_125us\_64M\_0\_1s} mode, which returns events to a

106: limiting resolution of 125~$\mu$s and with moderate (64 channels as

107: opposed to the full 256 channels) spectral resolution. This mode was

108: used because it is not as susceptible to buffer overflows during high

109: countrate data as compared to the less restrictive modes

110: (e.g. \texttt{GoodXenon}).  Using the pointing of the observation

111: (17$^{\mathrm{h}}$48$^{\mathrm{m}}$52\fs8, -20\degr 21\arcmin

112: 32\arcsec) and the planetary ephemeris DE200, the times of raw events

113: were corrected to the solar system barycenter.  Binning the events

114: into a 1-s resolution lightcurve reveals a quickly decaying burst

115: (Fig.~\ref{fig:lc}).

116: 

117: \begin{figure}

118: \plotone{fig1.eps}

119: \caption{Burst lightcurve binned with 1-s resolution and using the

120:   full PCA bandpass. Inset: ``Leahy'' normalized \citep{lde+83} power

121:   spectrum of the entire observation. Notice the significant peak at

122:   442 Hz. \label{fig:lc}}

123: \end{figure}

124: 

125: % The sharp decay of the burst indicates that we caught the tail-end of

126: % a much longer, and thus more energetic, event.

127: 

128: \section{Analysis and Results}

129: \label{sec:results}

130: 

131: % Here we describe the timing and spectral properties of the

132: % burst. Based on our analysis we believe that this events is in fact

133: % much more interesting than a canonical burst oscillation.

134: 

135: 

136: This burst was reminiscent of canonical Type~I X-ray bursts from

137: LMXBs, therefore we searched this event for burst oscillations.  We

138: rebinned the raw events using the full spectral band pass into a time

139: series with 0.5/1024~s resolution, which yields an equivalent Nyquist

140: frequency of 1024~Hz. A Leahy normalized \citep[see][]{lde+83} power

141: density spectrum (PDS) is displayed in Figure~\ref{fig:lc} (inset). A

142: prominent peak is clearly seen at 442~Hz. The probability of this peak

143: occurring by chance after accounting for the number of trials (the

144: total number of frequency bins in our PDS) is $\sim$$2\times 10^{-9}$.

145: 

146: \subsection{The NGC~6440 Field} 

147: \label{sec:FOV}

148: 

149: NGC~6440 harbors the bright X-ray transient \saxsource, which

150: exhibited a 409.7~Hz burst oscillation \citep{kzht03}. However, there

151: are many other X-ray sources in the cluster (see Fig~\ref{fig:chandra

152:   image}). \textit{Chandra X-ray Observatory} (\textit{CXO}) imaging by \citet{plv+02} revealed 24 X-ray

153: sources in NGC~6440, and they concluded that 4--5 of these sources are

154: likely quiescent LMXBs. Thus, the phenomenon we discovered is most

155: likely from a different object in the cluster.

156: 

157: \begin{figure}

158: \includegraphics[width=\columnwidth]{fig2.eps}

159: \caption{Archival \textit{CXO} image of

160:   NGC~6440. Notice that the cluster contains many X-ray point sources.

161: \label{fig:chandra image}}

162: \end{figure}

163: 

164: \subsection{Timing Analysis}

165: \label{sec:timing}

166: To study  the frequency evolution of the pulse we  calculated a

167: dynamic $Z^2$ statistic .The $Z^2$ statistic is analogous to the Fourier power

168: spectrum, with the advantage that the data need not be binned, thus

169: allowing us to oversample our data.

170: The $Z^2$ statistic is defined as:

171: \begin{equation}

172: %Z^2_j = \frac{2}{N_{\gamma}}\sum_{k=0}^{N_{\mathrm{harm}}} \sum_{i=0}^{N_{\gamma}} \sin^2( 2\pi k \nu_j t_i ) + cos^2(2\pi k \nu_j t_i ),

173: Z^2_{N_{\mathrm{harm}}} = \frac{2}{N_{\gamma}}\sum_{k=0}^{N_{\mathrm{harm}}} \sum_{i=0}^{N_{\gamma}} \left| e^{i 2\pi k \nu_j t_i }\right|^2

174: \end{equation}

175: where $N_\gamma$ is the total number of photons in each interval,

176: $N_{\mathrm{harm}}$ is the total number of harmonics that one deems

177: significant, $\nu_j$ is the frequency searched over, and $t_i$ is the

178: event time. The factor in front of the summations normalizes the $Z^2$

179: statistics in an analogous way to Leahy normalizing a PDS.

180: We calculated the dynamic $Z^2$ power spectrum using a 200-s long

181: window, which was translated across the observation with a step size

182: of 16~s.  Our dynamic $Z^2$ power spectrum is displayed in

183: Fig.~\ref{fig:frequency evolution}.  Notice that the pulsations where

184: highly significant for 576~s. As is common for canonical X-ray bursts,

185: the frequency drifts; however, unlike the large $\sim$1-2~Hz drifts

186: seen in those, the pulsation here drifts down in frequency only by

187: $\sim$2.1$\times 10^{-3}$~Hz in 576~s.

188: 

189: \begin{figure*}

190: \centerline{\includegraphics[height=1.5\columnwidth, angle=270]{fig3.eps}}

191: %%\plotone{Z2.eps} 

192: \figcaption{ Pulse Frequency Evolution. Top:

193:   The burst lightcurve as shown in Fig.~\ref{fig:lc}. Middle: Dynamic

194:   $Z^2$ power spectrum. Notice that the pulsations remain highly

195:   significant (with $Z_1^2>16$) for 576~s. The solid line represents our

196:   best-fit frequency model consisting of just a frequency and a

197:   frequency derivative term. Bottom: Phase residuals after subtracting

198:   our best fit frequency model. Notice that the pulsations are highly

199:   coherent.

200: \label{fig:frequency evolution}}

201: \end{figure*}

202: 

203: To further quantify this frequency evolution we determined how the

204: phase of the pulsations varies as a function of time.  For each interval

205: with $Z_1^2>16$ we generated a pulse profile by folding the events in

206: that interval on the frequency determined from the $Z^2$ statistic. We

207: then cross correlated these pulse profiles with a sinusoid of fixed

208: phase to determine the pulse phase as a function of time. We can model

209: the pulse phase ($\phi$) at a given time ($t$) by the following Taylor

210: expansion:

211: \begin{equation}

212: \phi(t) = \phi(t_0) + \nu(t_0) (t - t_0) + \frac{1}{2} \dot{\nu}(t_0)

213: (t-t_0)^2 + \cdots,

214: \label{eq:phase}

215: \end{equation}

216: where $\nu$ is the barycentric frequency, $\dot{\nu}$ is the frequency

217: derivative, and $t_0$ is some reference epoch. We were able to

218: “whiten” our phase residuals with just a frequency derivative (see

219: Fig.~\ref{fig:frequency evolution}, bottom). We find

220: $\nu=442.33850(5)$~Hz, $\dot{\nu}=3.7(2)\times 10^{-6}$~Hz~s$^{-1}$ at

221: $t_0$=53535.463215~MJD~(UTC). The phase residuals after subtracting

222: the model given by Eq.~\ref{eq:phase} are shown in

223: Fig.~\ref{fig:frequency evolution}~(bottom panel).  Notice that the

224: pulse train lasts for 576~s; this is quite long when compared to those

225: of Type I X-ray bursts.  We folded all the events within the 576~s

226: pulse train using our best fit $\nu$ and $\dot{\nu}$, and the resulting

227: pulse profile is shown in Fig.~\ref{fig:profile}. Notice that the pulse

228: profile is highly sinusoidal. Fitting a sinusoid to the pulse profile

229: returned a reduced $\chi^2$ value of 0.76 for 13 degrees of freedom,

230: affirming the absence of any harmonic content. From the fit we obtain a fractional pulse amplitude of 2.1$\pm$0.1\%.

231: 

232: \begin{figure}

233: \plotone{fig4.eps} 

234: \figcaption{ The pulse profile obtained by

235:   folding all events in the 576-s long pulse train (i.e., the interval

236:   where the pulsations were highly significant as determined from the

237:   $Z^2$ power spectrum) on the ephemeris determined in our timing

238:   analysis. The pulse shape is highly sinusoidal. The solid line

239:   represents the best-fit sinusoid. The fit yielded $\chi^2$=0.76 for

240:   13 degrees of freedom, affirming the absence of any harmonic

241:   content.

242: \label{fig:profile}}

243: \end{figure}

244: 

245: 

246: It is possible that the frequency drift we have observed

247: is entirely due to orbital motion. The observed spin frequency of a

248: neutron star in a circular orbit is given by

249: \begin{equation}

250: \nu = \nu_0 \left [ 1 - \frac{v_{\mathrm{ns}}\sin i}{c}\sin \left(

251:   \frac{2\pi t}{T} + \phi_0 \right) \right],

252: \label{eq:orbit}

253: \end{equation}

254: where, $\nu_0$ is the barycentric frequency at time $t=t_0$,

255: $v_{\mathrm{ns}}\sin i$ is the projected velocity of the neutron star,

256: $T$ is the orbital period, and $\phi_0$ is the orbital phase at time

257: $t=t_0$. Unfortunately we do not have a long enough data set to place

258: interesting constraints on $T$ or $v_{\mathrm{ns}}\sin i$; however, we

259: can determine whether orbital modulation is consistent with the

260: observed frequency drift.  For example, if we assume, for solely

261: demonstrative purposes, that this system has a projected velocity

262: comparable to the one found by \citet{sm02} for 4U~1636$-$53

263: ($v_{\mathrm{ns}}\sin i=136$~km~$^{-1}$), then fitting Eq.~\ref{eq:orbit} to our frequency

264: time series yields a good fit with a reasonable orbital period, i.e.

265: $T=$7$\pm$1~hr. Hence, binary motion can account for the observed

266: frequency drift. Observing another outburst  would be

267: essential in ultimately constraining the orbital parameters of this

268: system.

269: 

270: \subsection{Spectral Analysis}

271: \label{sec:spectral}

272: 

273: The timing analysis seems to suggest that this event is in fact very

274: different from a canonical burst oscillation. To study it further we

275: performed a spectral analysis.  We analyzed the spectral evolution of

276: the burst by breaking the burst up into ten intervals. The exposure of

277: each interval was selected by demanding that each one contained an

278: equal number of photons. Using the last interval as an estimator of

279: the persistent emission, we fit each interval to a simple

280: photoelectrically absorbed blackbody model while holding the column

281: density fixed to the value found by \citet{plv+02} for the optical

282: reddening.  We find only subtle evidence for spectral softening,

283: significant only at the 1-$\sigma$ level.

284: 

285: Using the distance estimate to the cluster found by \citet{obb94},

286: $d=8.5\pm0.4$~kpc, we were able to calculate the luminosity of the

287: burst.  At the start of the observation the flux was

288: $\sim$2$\times$$10^{37}$~erg~s$^{-1}$, which is $\sim$0.1$L_{Edd}$,

289: where $L_{Edd}$ is the Eddington luminsosity for a neutron star.  The

290: burst lightcurve was well fit by an exponential with a decay timescale

291: of 22~s. Unfortunately the observation only caugth the tail end of the

292: burst-like event, we could not therefore characterize the peak

293: luminosity of the burst.

294: 

295: \section{Discussion}

296: \label{sec:discussion}

297: 

298: \subsection{Super Burst?}

299: 

300: The timing properties of the pulsations are reminiscent of those

301: observed during superbursts. In particular the highly coherent

302: pulsations and the long pulse trains.  The superburst interpretation

303: would be very exciting given the rarity of such events. However, we

304: could not unambiguously determine this is the case given that we only

305: caught the tail end of the burst. To study this phenomenon further we

306: extrapolated our luminosity model determined in our spectral analysis

307: before the start time of the observation. We find that the burst

308: reaches the Eddington luminosity $\sim$50~s before the start of the

309: observation.  Now, if we assume that we caught the tail of the

310: outburst it is very surprising that it dropped from $\sim$$L_{Edd}$

311: down to $\sim$.01$L_{Edd}$ in $\sim$77~s.  This is much too fast for a

312: superburst-like event. If, on the other hand, this is a precursor

313: events, such as those seen in superbursts, than it is unclear why the

314: persistent flux did not rise back to $\sim$$L_{Edd}$ shortly after as

315: observed in other superbursts. Thus, the spectral analysis of this

316: event suggests it is not a superburst.

317: 

318: \subsection{Intermittent Pulsar?}

319: The timing properties of the pulsations  share all the hallmarks of those observed during superbursts, but the

320: spectral analysis makes this interpretation unlikely.

321: %The timing and spectral properties tend to imply that this event might

322: %have been a superburst. Although intriguing as this is possibility is,

323: %it is not the only plausible interpretation for what he have

324: %observed. 

325: If this event is not a superburst, then could it be that these highly

326: coherent pulsations have eminated from an accretion-powered

327: millisecond pulsar?  Recently \citet{gmk+07} discovered

328: ``intermittent'' pulsations from the accretion-powered millisecond

329: pulsar (AMP) HETE~J1900.1$-$2455. They found that the properties of

330: this pulsar differed from those of the other six known AMPs. For

331: example, the pulsations were only present in the first few months of

332: its outburst as opposed to other AMPs which show pulsations throughout

333: \citep[see][for a review]{wij04}. In addition, the pulse amplitude

334: increased at three points in time that were almost coincident with the

335: times of thermonuclear bursts. The pulsed amplitude subsequently

336: decaded after the bursts.  Thus, it is plausible that we have observed

337: a similar phenomenon. Following the theoretical work of \citet{czb01}

338: and \citet{pm06}, \citet{gmk+07} suggest that the reason why the pulsations in

339: HETE~J1900.1$-$2455 sometimes ``switch off'' is because the magnetic

340: field is burried by accreted material, thus it cannot channel the flow

341: which gives rise to the pulsations in the first place

342: 

343: If the event we have discovered is another example of these

344: intermittent pulsations from an unidentified source in NGC~6440, then

345: a future outburst should reveal more long pulse trains, and if they

346: happen to be correlated with the times of Type-I X-ray bursts then

347: this would solidify the connection between this source and

348: HETE~J1900.1$-$2455.

349: 

350: %\citet{} reported at 409.7-Hz periodicity during an X-ray burst from the

351: %LMXB \saxsource, which is also in NGC 6440. Although

352: %burst-oscillations tend to drift, for reasons which are currently not

353: %well understood, a change in frequency by more than a few Hz has never

354: %been observed from any source which exhibits burst oscillations.

355: %Based on this, and given that it is not unlikely for globular clusters

356: %to harbor more than on LMXB, we are led to believe that the

357: %pulsations are from another LMXB in NGC 6440. Unfortunately the

358: %enormous 1 degree field-of-view does not permit us to

359: %spatially-discriminate between the two sources. However, \citet{} and

360: %\citet{} observed NGC 6440 using the Chandra X-ray observatory (\cxo),

361: %and their imaging observations of the cluster revealed several

362: %point-like X-ray sources. Furthermore, based on the XXX properties of

363: %the X-ray sources in this cluster \citet{} concluded that 4-5 of them

364: %are most probably LMXBs.

365: 

366: \section{Conclusions}

367: We discovered a 442~Hz pulsation during a burst from an X-ray source

368: in NGC~6440.  We could not establish the energetics of the burst as

369: \textit{RXTE} only caught the tail end of the event, but based on the

370: timing properties of the pulsation we conclude that it was not a

371: canonical burst oscillation. In particular we find that: The pulse

372: train lasted for 576~s, as opposed to the $\sim$10~s long pulse trains

373: observed in Type I X-ray bursts.  The signal was highly coherent and

374: drifted down only by $2.1\times10^{-3}$~Hz, as opposed to Type-I X-ray

375: burst which exhibit drifts which are three orders of magnitude

376: larger. The pulsations share all the properties of those observed

377: during superbursts; however, the complete energetics of this event

378: make this interpretation implausible. We conclude that the long pulse

379: train is most likely an ``intermittent'' pulsation from an accreting

380: millisecond pulsar such as those seen only in HETE~J1900.1-2455 thus

381: far \citep{gmk+07}. We note that NGC~6440 is known to harbor a

382: transient X-ray source with a 409.7~Hz spin frequency \citep{kzht03}

383: However, X-ray imaging of the cluster by \citet{plv+02} has shown that

384: NGC~6440 contains many X-ray sources. Thus, the burst presented here

385: most likely emanated from a different object in the cluster.

386: 

387: 

388: \acknowledgements

389: FPG is supported by the NASA Postdoctoral Program

390: administered by Oak Ridge Associated Universities at NASA Goddard

391: Space Flight Center. This research has made use of data obtained

392: through the High Energy Astrophysics Science Archive Research Center

393: Online Service, provided by the NASA/Goddard Space Flight Center.

394: 

395: \begin{thebibliography}{}

396: 

397: %\bibitem[{Cackett} {et al.}(2005)]{cwh+05}

398: %{Cackett}, E.~M., {Wijnands}, R., {Heinke}, C.~O., {Edmonds}, P.~D., {Lewin}, W.~H.~G., {Pooley}, D., {Grindlay}, J.~E., {Jonker}, P.~G., \& {Miller}, J.~M. 2005, ApJ, 620, 922

399: 

400: %\bibitem[{Casares} {et al.}(2006)]{ccs+06}

401: %{Casares}, J., {Cornelisse}, R., {Steeghs}, D., {Charles}, P.~A., {Hynes}, R.~I., {O'Brien}, K., \& {Strohmayer}, T.~E. 2006, MNRAS, 373, 1235

402: 

403: \bibitem[{Cumming \& Bildsten}(2001)]{cb01}

404: {Cumming}, A. and {Bildsten}, L. 2001, ApJ, 559, L127

405: 

406: \bibitem[{Cumming} {et al.}(2001)]{czb01}

407: {Cumming}, A., {Zweibel}, E., \& {Bildsten}, L. 2001, ApJ, 557, 958

408: 

409: \bibitem[{Galloway} {et al.}(2007)]{gmk+07}

410: {Galloway}, D.~K., {Morgan}, E.~H., {Krauss}, M.~I., {Kaaret}, P., {Chakrabarty}, D. 2007, ApJ, 654, L73

411: 

412: \bibitem[{Galloway \& Muno}(2007)]{gm07}

413: {Galloway}, D.~K., {Muno}, M.~P. 2007, ApJ, in press (arXiv:astro-ph/0608259)

414: 

415: %\bibitem[{Giles} {et al.}(2002)]{ghsc02}

416: %{Giles}, A.~B., {Hill}, K.~M., {Strohmayer}, T.~E., \& {Cummings}, N. 2002, ApJ, 568, 279 

417: 

418: 

419: \bibitem[{Kaaret} {et al.}(2003)]{kzht03}

420: {Kaaret}, P., {Zand}, J.~J.~M.~i., {Heise}, J., \& {Tomsick}, J.~A. 2003, ApJ, 598, 481

421: 

422: \bibitem[{in't~Zand} {et~al.}(2004)]{zcc04}

423: {in't Zand}, J.~J.~M., {Cornelisse}, R., \& {Cumming}, A. 2004, A\&A, 426, 257

424: 

425: \bibitem[{Leahy} {et~al.}(1983)]{lde+83}

426: {Leahy}, D.~A.,{Darbro}, W.,{Elsner}, R.~F., {Weisskopf}, M.~C., {Kahn}, S., {Sutherland}, P.~G., \& {Grindlay}, J.~E. 1983, ApJ, 266, 160

427: 

428: \bibitem[{Liu} {et al.}(2006)]{lph06}

429: {Liu}, Q.~Z.,  {van Paradijs}, J., \& {van den Heuvel}, E.~P.~J. 2006, A\&A, 455, 1165

430: 

431: %\bibitem[{Mu{\~n}oz-Darias} {et~al.}(2005)]{mcm05}

432: %{Mu{\~n}oz-Darias}, T., {Casares}, J., {Mart{\'{\i}}nez-Pais} I.~G., 2005, ApJ, 635, 502

433: 

434: \bibitem[{Ortolani} {et al.}(1994)]{obb94}

435: {Ortolani}, S., {Barbuy}, B., {Bica}, E. 1994, A\&AS, 108, 653

436: 

437: \bibitem[Payne \& Melatos(2006)]{pm06}

438: {Payne}, D.~J.~B., {Melatos}, A. 2006, ApJ, 652, 597

439: 

440: \bibitem[{Pooley} {et al.}(2002)]{plv+02}

441: {Pooley}, D., {Lewin}, W.~H.~G., {Verbunt}, F., {Homer}, L., {Margon}, B., {Gaensler}, B.~M., {Kaspi}, V.~M., {Miller}, J.~M., {Fox}, D.~W., \& {van der Klis}, M. 2002, ApJ, 573, 184

442: 

443: \bibitem[Strohmayer \& Bildsten(2007)]{sb07} 

444: {Strohmayer}, T.~E. \& {Bildsten}, L. 2003,  http://xxx.lanl.gov/abs/astro-ph/0301544

445: 

446: \bibitem[Strohmayer \& Brown(2002)]{sb02}

447: {Strohmayer}, T.~E. \& {Brown}, E.~F. 2002, ApJ, 566, 1045

448: 

449: \bibitem[Strohmayer \& Markwardt(2002)]{sm02}

450: {Strohmayer}, T.~E. \& {Markwardt}, C.~B. 2002, ApJ, 577, 337

451: 

452: \bibitem[{Strohmayer} {et al.}(1996)]{szs+96}

453: {Strohmayer}, T.~E., {Zhang}, W., {Swank}, J.~H., {Smale}, A., {Titarchuk}, L., {Day}, C., \& {Lee}, U. 1996, ApJ, 469, L9

454: 

455: \bibitem[{Wijnands} (2004)]{wij04}

456: {Wijnands}, R. 2004, Nucl. Phys. B, 132, 496

457: 

458: \end{thebibliography}

459: 

460: 

461: \end{document}

462: 

463: 

464: