1: %\documentclass[12pt,preprint]{aastex}
2: \documentclass{emulateapj}
3:
4:
5: \newcommand{\expnt}[2]{\ensuremath{#1 \times 10^{#2}}} % scientific notation
6: \newcommand{\gsim}{\gtrsim}
7: \newcommand{\lsim}{\lesssim}
8:
9:
10: \newcommand{\xmm}{\textit{XMM}}
11: \newcommand{\hr}{\ensuremath{^{\rm h}}}
12: \newcommand{\mn}{\ensuremath{^{\rm m}}}
13: \newcommand{\ergs}{\ensuremath{{\rm ergs \; s^{-1}}}}
14: \newcommand{\ergcms}{\ensuremath{{\rm ergs \; cm^{-2} \; s^{-1}}}}
15: \newcommand{\cmsq}{\ensuremath{{\rm cm}^{-2}}}
16: \newcommand{\rxj}{RX~J0720.4\ensuremath{-}3125}
17: \newcommand{\cxo}{\textit{CXO}}
18: \newcommand{\chandra}{\textit{Chandra}}
19: \newcommand{\rosat}{\textit{ROSAT}}
20:
21: \newcommand{\secsec}{\ensuremath{\mbox{s s}^{-1}}}
22: \newcommand{\Hzsec}{\ensuremath{\mbox{Hz s}^{-1}}}
23:
24: \defcitealias{kkvkm02}{Paper I}
25:
26: \newcommand{\fdot}{\ensuremath{\dot \nu}}
27: \newcommand{\fdotdot}{\ensuremath{\ddot \nu}}
28:
29: \slugcomment{To appear in ApJL}
30: \shorttitle{A Timing Solution for RX~J0720.4$-$3125}
31: \shortauthors{Kaplan \&\ van~Kerkwijk}
32: \begin{document}
33:
34:
35: \title{A Coherent Timing Solution for the Nearby Isolated Neutron Star
36: RX~J0720.4$-$3125}
37: \author{D.~L.~Kaplan\altaffilmark{1,2} and M.~H.~van Kerkwijk\altaffilmark{3}}
38:
39: \altaffiltext{1}{Pappalardo Fellow}
40: \altaffiltext{2}{Center for Space
41: Research, Massachusetts Institute of Technology, 77 Massachusetts Avenue,
42: 37-664D, Cambridge, MA 02139, USA; dlk@space.mit.edu}
43: \altaffiltext{3}{Department of Astronomy and Astrophysics, University
44: of Toronto, 60 St.\ George Street, Toronto, ON M5S 3H8, Canada;
45: mhvk@astro.utoronto.ca}
46:
47: \begin{abstract}
48: We present the results of a dedicated effort to measure the spin-down
49: rate of the nearby isolated neutron star \rxj. Comparing arrival
50: times of the 8.39-sec pulsations for data from \chandra\ we
51: derive an unambiguous timing solution for \rxj\ that is accurate to
52: $<0.1$~cycles over $>5$~years. Adding data from \xmm\ and \rosat, the final
53: solution yields $\dot P=\expnt{(6.98\pm0.02)}{-14}\,\secsec$; for
54: dipole spin-down, this implies a characteristic age of 2~Myr and a
55: magnetic field strength of $\expnt{2.4}{13}$~G. The phase residuals
56: are somewhat larger than those for purely regular spin-down, but do
57: not show conclusive evidence for higher-order terms or a glitch. From
58: our timing solution as well as recent X-ray spectroscopy, we concur
59: with recent suggestions that \rxj\ is most likely an off-beam radio
60: pulsar with a moderately high magnetic field.
61: \end{abstract}
62: \keywords{pulsars: individual (RX J0720.4$-$3125)
63: --- stars: neutron
64: --- X-rays: stars}
65:
66:
67: \section{Introduction}
68:
69: One of the interesting results from \rosat\ All-Sky Survey \citep{rbs}
70: was the discovery of seven objects that appear to be nearby,
71: thermally-emitting neutron stars that have little if any
72: magnetospheric emission (see \citealt{haberl04} for a review). These
73: objects, known most commonly as ``isolated neutron stars,'' are
74: distinguished by their long periods ($\gsim 3$~s,
75: when measured), largely thermal spectra with cool temperatures ($kT
76: \lsim 100$~eV), faint optical counterparts (when detected), and lack
77: of radio emission.
78:
79: It is not yet clear what sets the isolated neutron stars apart from
80: the nearby, relatively young rotation-powered pulsars that also have
81: cool thermal emission --- sources like PSR~B0656+14 and
82: PSR~B1055$-$52--- which tend to have short ($<1$~s) spin periods,
83: $\sim 10^{12}$-G magnetic fields, non-thermal (i.e.\ power-law)
84: components in their X-ray
85: spectra, and radio pulsations \citep[e.g.,][]{pz03,kaplan04}. The isolated
86: neutron stars are known to have longer periods, but their spin-down
87: rates (and hence magnetic fields ) are unknown, largely because it has
88: not yet been possible to determine a reliable, coherent timing
89: solution (\citealt{kkvkm02}, hereafter \citetalias{kkvkm02};
90: \citealt{zhc+02}).
91:
92: In this \textit{Letter} we report a new analysis of the variations of
93: the 8.39-s period of the second brightest source of the group, \rxj\
94: \citep{hmb+97}. We describe our analysis of \chandra\ data obtained
95: specifically for timing purposes, as well as archival \rosat,
96: \chandra, and \xmm\ data, in \S~\ref{sec:obs}. In
97: \S~\ref{sec:timing}, we show that with the new data, we can avoid the
98: pitfalls of the previous phase-coherent timing analyses and obtain a
99: reliable timing solution. We discuss possible timing noise in
100: \S~\ref{sec:noise} and the implications of our result in
101: \S~\ref{sec:discuss}.
102: \ \\
103:
104: \section{Observations}\label{sec:obs}
105:
106: Our primary data were eight observations with the Advanced CCD Imaging
107: Spectrometer \citep[ACIS;][]{gbf+03} aboard the \textit{Chandra X-ray
108: Observatory} \citep[\cxo;][]{wtvso00}. These were
109: designed for timing accuracy, consisting of two sets of four exposures
110: geometrically spaced over a period of about two weeks and separated
111: by about half a year. We combined these with data from other
112: \chandra\ observations, as well as from observations with
113: \textit{XMM-Newton} \citep{jla+01} and \rosat\ \citep{trumper93}. A
114: log of all observations is given in Table~\ref{tab:obs}.
115:
116: For the \chandra\ data, we processed the level-1 event lists to the
117: level-2 stage following standard procedures and the latest calibration
118: set (CALDB version 3.0.0). For the ACIS continuous-clocking data,
119: this includes correcting the recorded event times for readout, dither,
120: and spacecraft motion --- corrections that used to require additional
121: steps \citep{zpst00}. We extracted events within $1\arcsec$ of the
122: source, and then applied a clock correction of $284.7\,\,\mu{\rm s}$
123: to the arrival times \citep*{dhm03}; the arrival times should now be
124: accurate to $\lsim 6\,\,\mu{\rm s}$. For the HRC-S/LETG data, we
125: extracted zeroth-order events from a circle with radius 10 pixels
126: ($1\farcs3$), and first-order events using the standard LETG spectral
127: extraction windows, but limited to $10\leq\lambda\leq65\mbox{\AA}$.
128: Finally, we used the \texttt{axbary} program to barycenter all of the
129: events (using the optical position: $\alpha_{\rm
130: J2000}=07\hr20\mn24\fs96$, $\delta_{\rm
131: J2000}=-31\degr25\arcmin50\farcs2$; \citealt{kvkm+03})
132:
133:
134: \begin{deluxetable*}{lrcrrc}[t]
135: %\tabletypesize{\scriptsize}
136: \tablewidth{0pt}
137: \tablecaption{Log of Observations and Times of Arrival\label{tab:obs}}
138: \tablehead{
139: \colhead{Instrument\tablenotemark{a}}&
140: \colhead{ID\tablenotemark{b}}&
141: \colhead{Date}&
142: \colhead{Exp.}&
143: \colhead{Counts}&
144: \colhead{TOA\tablenotemark{c}}\\
145: &&&\colhead{(ks)}&& \colhead{(MJD)}\\[-2.2ex]
146: }
147: \startdata
148: PSPC\dotfill & 338 & 1993-09-27 & 3.2 & 5800&49257.2547031(25)\\
149: HRI\dotfill & 884 & 1996-11-03 & 33.7 & 12662&50391.3006530(16)\\
150: HRI\dotfill & 944 & 1998-04-20 & 8.1 & 3074&50925.6881172(36)\\
151: HRC\dotfill & 368 & 2000-02-01 & 5.4 & 3472&51575.3026910(46)\\
152: HRC\dotfill & 745 & 2000-02-02 & 26.1 & 15149&51576.2804856(27)\\
153: HRC\dotfill & 369 & 2000-02-04 & 6.1 & 3667&51578.7722735(65)\\
154: PN/ff/thin\dotfill & 78-S3& 2000-05-13 & 21.1 &144104&51677.2260789(\phn5)\\
155: MOS2/thin\dotfill & 78-S2& 2000-05-13 & 43.9 & 73915&51677.4127431(\phn7)\\
156: PN/ff/med\dotfill & 175-S3& 2000-11-21 & 25.7 &153037&51869.8413358(14)\\
157: MOS1/open\dotfill & 175-U2& 2000-11-21 & 6.8 & 17762&51869.8433759(14)\\
158: MOS2/open\dotfill & 175-U2& 2000-11-21 & 7.2 & 21084&51869.9571032(\phn6)\\
159: ACIS-CC\dotfill &2774 & 2001-12-04 & 15.0 & 31831&52247.7881789(11)\\
160: ACIS-CC\dotfill &2773 & 2001-12-05 & 10.6 & 22847&52248.2835843(13)\\
161: ACIS-CC\dotfill &2772 & 2001-12-06 & 4.1 & 8790&52249.6286894(26)\\
162: PN/ff/thin\dotfill & 533-S3& 2002-11-06 & 28.3 &199841&52584.9260561(\phn5)\\
163: PN/ff/thin\dotfill & 534-S3& 2002-11-08 & 30.2 &212177&52587.0013053(\phn4)\\
164: MOS1/open\dotfill & 622-U2& 2003-05-02 & 7.6 & 17629&52761.6222174(14)\\
165: MOS2/open\dotfill & 622-U2& 2003-05-02 & 7.5 & 18788&52761.6226056(12)\\
166: PN/sw/thick\dotfill& 622-U2& 2003-05-02 & 72.8 &210160&52761.9950589(\phn5)\\
167: PN/sw/thin\dotfill & 711-S7& 2003-10-27 & 18.1 &112876&52939.8228751(\phn9)\\
168: PN/sw/thick\dotfill& 711-S8& 2003-10-27 & 25.0 &138689&52939.8228774(\phn8)\\
169: MOS1/open\dotfill & 711-U2& 2003-10-27 & 13.8 & 33323&52939.8506513(\phn5)\\
170: MOS2/open\dotfill & 711-U2& 2003-10-27 & 13.8 & 35636&52940.1162720(\phn5)\\
171: ACIS-CC\dotfill &4666 & 2004-01-06 & 10.1 & 19048&53010.2635608(14)\\
172: ACIS-CC\dotfill &4667 & 2004-01-07 & 4.8 & 8938&53011.2639869(20)\\
173: ACIS-CC\dotfill &4668 & 2004-01-11 & 5.2 & 9334&53015.5407400(19)\\
174: ACIS-CC\dotfill &4669 & 2004-01-19 & 5.2 & 9391&53023.1274147(23)\\
175: HRC\dotfill &5305 & 2004-02-27 & 35.7 & 21597&53062.4142490(27)\\
176: PN/ff/thin\dotfill & 815-S1& 2004-05-22 & 31.6 &219855&53147.6811948(\phn4)\\
177: ACIS-CC\dotfill &4670 & 2004-08-03 & 10.1 & 17432&53220.9975987(14)\\
178: ACIS-CC\dotfill &4671 & 2004-08-05 & 5.1 & 8051&53222.2171299(21)\\
179: ACIS-CC\dotfill &4672 & 2004-08-09 & 5.1 & 8556&53226.2443808(25)\\
180: ACIS-CC\dotfill &4673 & 2004-08-23 & 5.1 & 7133&53240.1824669(24)\\
181: HRC\dotfill &5581 & 2005-01-23 & 67.7 & 44801&53393.6657119(18)\\
182: \enddata
183: \tablenotetext{a}{PSPC: Position Sensitive Proportional Counter
184: \citep{bp95} aboard \rosat.
185: HRI: High-Resolution Imager \citep{zdhk95} aboard \rosat.
186: HRC: High-Resolution Camera for spectroscopy aboard \chandra\ (HRC-S;
187: \citealt{kck+97}), used with the Low-Energy Transmission Grating.
188: ACIS: \chandra's Advanced CCD Imaging Spectrometer \citep{gbf+03},
189: used in continuous clocking mode.
190: EPIC-pn: \xmm's European Photon Imaging Camera with PN detectors
191: \citep{sbd+01}, used in full-frame (ff) or small window (sw)
192: mode, with thin, medium, or thick filter.
193: EPIC-MOS1/2: European Photon Imaging Cameras with MOS detectors
194: aboard \xmm\ \citep{taa+01}, used in small-window mode with thin
195: or no (open) filter.}
196: \tablenotetext{b}{Observation identifier (\cxo, \rosat) or revolution
197: number and exposure identifier (\xmm).}
198: \tablenotetext{c}{The TOA is defined as the time of maximum light
199: closest to the middle of each observation, and is given with
200: 1-$\sigma$ uncertainties.}
201: \end{deluxetable*}
202:
203:
204: For the \xmm\ data, we used the standard procedures \texttt{emchain}
205: and \texttt{epchain} (XMMSAS version 6.1.0) to reprocess the
206: observations. One additional step was necessary for the PN data set
207: 622-U2, for which we found a small number of duplicate events (frames
208: 963685--963719); we removed these before processing. Next, we
209: extracted all single-pixel events within 37\farcs5 of the source
210: position, and used \texttt{barycen} to convert the arrival times to
211: the solar-system barycenter\footnote{Some portions of the 2000 and
212: 2002 \xmm/PN observations were affected by a known processing
213: problem that rejected significant portions of the observations; see
214: \url{http://xmm.vilspa.esa.es/sas/documentation/watchout/lost\_events.shtml}.
215: This should not introduce any systematic error, though it means that
216: our TOAs for these observations are not as precise as possible with
217: all events. However, since the present TOA uncertainties are smaller than
218: the timing noise (\S~\ref{sec:noise}), we decided not to
219: try to remedy this problem.}.
220:
221:
222: The reduction of the \rosat\ data followed that in
223: \citetalias{kkvkm02}, except that we properly corrected the event
224: times to the Barycentric Dynamical Time (TDB) system instead of the
225: Coordinated Universal Time (UTC) system returned by the
226: \texttt{FTOOLS} barycentering tasks (see \citealt{chzz04}). We
227: used the corrections supplied in \citet[][p.\ 14]{allen}.
228:
229:
230: \ \\
231: \section{Timing Analysis}
232: \label{sec:timing}
233: Our goal was to use times-of-arrival (TOAs) to infer a phase-coherent
234: timing solution involving the spin period and its derivative, where
235: each cycle of the source was counted. To measure TOAs we needed an
236: initial reference period, something which we determined using a
237: $Z_1^2$ test \citep[Rayleigh statistic;][]{bbb+83} on the combined
238: 2004 January ACIS and 2004 February HRC data. We find
239: $P=8.3911159(10)$~s (here and below, numbers in parentheses indicate
240: twice the formal 1-$\sigma$ uncertainties in the last digit unless otherwise
241: indicated), which is consistent with our earlier value
242: (\citetalias{kkvkm02}) but much more accurate because we could
243: coherently connect observations over a much longer (52~day)
244: time span.
245:
246: Using this period, we constructed binned light curves (with 16 phase
247: bins) and determined the TOAs by fitting a sinusoid (appropriate for
248: the sinusoidal pulsations of \rxj; \citealt{hmb+97}); the
249: uncertainties were calculated from the uncertainties in the phases of
250: the fitted sinusoids (we verified that we obtain consistent results if
251: we change the binning or measure TOAs using cross-correlation
252: instead). Here, the TOA is defined as the time of maximum light
253: closest to the middle of the observation, a choice which minimizes
254: co-variance with small changes in period. We present TOAs for all of
255: the data in Table~\ref{tab:obs}.
256:
257: We then determined a timing solution for only the \chandra\ data
258: using an iterative procedure. We first used the reference period to
259: determine cycle counts for the five above-mentioned observations, as
260: well as the next observation closest in time. We fit these cycle
261: counts to
262: \begin{equation}
263: \phi(t) = \phi_0 + \nu (t-t_0) + \frac{1}{2} {\dot \nu}(t-t_0)^2 + \frac{1}{6}{\ddot \nu}(t-t_0)^3\ldots,
264: \label{eqn:phi}
265: \end{equation}
266: where $\phi_0$ is the cycle count plus phase at reference time $t_0$,
267: $\nu$ is the spin-frequency, $\dot \nu$ is its derivative, $\ddot \nu$
268: is the second derivative. We then
269: iterated, using the improved solution to determine the cycle count for
270: the next observation, etc. We started with $\fdot=\fdotdot=0$, but
271: left $\fdot$ free once that significantly improved the fit; $\fdotdot$
272: was not required (cf., \S~\ref{sec:noise} below).
273:
274: The final ephemeris listed in Table~\ref{tab:ephem} has small,
275: $\leq0.1$~cycle, residuals, and fits the \chandra\ data well:
276: $\chi^2_\nu\equiv\chi^2/N_{\rm dof}=1.06$ (with $N_{\rm dof}=13$
277: degrees of freedom). To test the uniqueness of our ephemeris, we
278: tried changing the cycle counts (adding or subtracting one or more
279: cycles) at the least unambiguous points, but
280: found that the resulting solutions were very poor (e.g., altering the
281: cycles between the 2001 ACIS and 2000 HRC observations gave
282: $\chi^2_{\nu}=89.26$).
283:
284:
285:
286: \begin{figure}
287: \plotone{f1.eps}
288: \caption{Phase residuals for \rxj. The top panel shows the residuals
289: for each TOA compared to a linear ($\fdot=0$) model. The solid
290: curve gives the best-fit quadratic ($\fdot \neq 0$, $\fdotdot=0$)
291: ephemeris for all data (Tab.~\ref{tab:ephem}, with a 0.27-s
292: systematic uncertainty added in quadrature). The vertical dashed
293: line indicates the reference time $t_0$. The bottom panel shows the
294: residuals relative to the quadratic model. We also show a best-fit
295: cubic model ($\fdotdot \neq 0$; dotted line) and a model that
296: includes a glitch in frequency near MJD~52800 ($\Delta f\approx
297: \expnt{1}{-9}$; dashed curve).}
298: \label{fig:resid}
299: \end{figure}
300:
301:
302: To improve and extend the ephemeris, we added the \xmm\ and \rosat\
303: data into the solution (see Table~\ref{tab:ephem}). As for the
304: \chandra\ data alone, the cycle counts are unambiguous and, as can be
305: seen in Fig.~\ref{fig:resid}, the residuals remain below $0.1$~cycles,
306: lending additional credence to our results. For verification, we also
307: examined the fit using just the \chandra\ and \xmm\ data, which avoids
308: the large gaps between the \rosat\ observations. We found the same
309: cycle counts, but a statistically significant difference in the
310: inferred parameters [$\nu=0.11917366887(12)$~Hz and
311: $\fdot=-\expnt{9.69(4)}{-16}\,\,\Hzsec$, while the fit to all data
312: gave $\nu=0.1191736700(2)$ and $\fdot=-\expnt{9.915(14)}{-16}\,\,\Hzsec$].
313:
314:
315: While we are confident in our fit in general, the above discrepancy is
316: puzzling. Furthermore, the $\chi^2_\nu$ values for the fits including
317: the \xmm\ data are poor, with $\chi^2_\nu=6.16$ for \chandra$+$\xmm\
318: ($N_{\rm dof}=28$; ${\rm rms}=0.34$~s) and $\chi^2_{\nu}=10.25$ for
319: all of the data ($N_{\rm dof}=31$; ${\rm rms}=0.36$~s), while the
320: solution for \chandra$+$\rosat, although somewhat ambiguous, is
321: tolerable ($\chi^2_\nu=1.40$) and similar to the \chandra-only
322: solution. The $\chi^2$ values for the full fits are dominated by the
323: contribution of the \xmm\ data, which have very small formal
324: uncertainties (but are not always entirely consistent from one
325: instrument to another for the same observation; see
326: Tab.~\ref{tab:obs}). The \xmm\ data also cause the values of $\fdot$
327: to differ by 3\% and the \chandra$+$\xmm\ ephemeris to be a poor match
328: to the \rosat\ points. We discuss these deviations in more detail
329: below; here, we note that for the estimates of the values and
330: uncertainties in Table~\ref{tab:ephem}, we ensured $\chi^2_\nu\simeq1$
331: by adding in quadrature an additional uncertainty of 0.27~s to all
332: TOAs. We stress, however, that our overall solution is robust, and
333: the inferred spin-down rate should be reliable at the $<\!3$\% level.
334:
335:
336:
337: \begin{deluxetable}{l c c}
338: \tablewidth{0pt}
339: \tablecaption{X-ray Timing Ephemerides for \rxj\label{tab:ephem}}
340: \tablehead{
341: \colhead{Quantity\tablenotemark{a}} & \colhead{\cxo\ only} &
342: \colhead{All data\tablenotemark{b}}\\[-2.2ex]
343: }
344: \startdata
345: Dates (MJD) \dotfill & 51575--53394 & 49257--53394\\
346: $t_{0}$ (MJD)\dotfill & 53010.2635605(18) & 53010.2635637(24)\\
347: $\nu$ (Hz) \dotfill & 0.11917366926(46) & 0.11917366908(38)\\
348: $\dot \nu$ (\Hzsec) \dotfill & $\expnt{-9.97(11)}{-16}$ & $-\expnt{9.918(30)}{-16}$\\
349: TOA rms (s) \dotfill & 0.18 & 0.31\\
350: $\chi^2$/DOF \dotfill & 13.8/13=1.06 & 30.7/31=0.99\\[1ex]
351: $P$ (s)\dotfill & 8.391115305(32) & 8.391115532(26)\\
352: $\dot P$ (\secsec)\dotfill & $\expnt{7.019(80)}{-14}$ & $\expnt{6.983(22)}{-14}$\\
353: $\dot E$ ($\mbox{erg s}^{-1}$)\dotfill& $\expnt{4.7}{30}$ & $\expnt{4.7}{30}$\\
354: $B_{\rm dip}$ (G) \dotfill & $\expnt{2.5}{13}$ & $\expnt{2.4}{13}$\\
355: $\tau_{\rm char}$ (yr)\dotfill& $\expnt{1.9}{6}$ & $\expnt{1.9}{6}$\\
356: \enddata
357: \tablecomments{Uncertainties quoted are twice the formal 1-$\sigma$
358: uncertainties in the fit.}
359: \tablenotetext{a}{$\dot E=10^{45}I_{45}4\pi^2\nu\dot\nu$ is the
360: spin-down luminosity (with $I=10^{45}I_{45}\,\,{\rm g}\,\,{\rm
361: cm}^{2}$ the moment of
362: inertia); $B_{\rm dip}=\expnt{3.2}{19}\sqrt{P{\dot P}}$ is the
363: magnetic field inferred assuming spin-down by dipole radiation;
364: $\tau_{\rm char}=P/2{\dot P}$ is the characteristic age,
365: assuming an initial spin period $P_0\ll P_{\rm now}$.}
366: \tablenotetext{b}{A 0.27-s systematic uncertainty has been added in
367: quadrature to these data to account for timing noise. See
368: \S~\ref{sec:noise}.}
369: \end{deluxetable}
370:
371:
372: \section{Deviations from Regular Spin Down?}
373: \label{sec:noise}
374: There could be several reasons for the relatively poor fit of the
375: timing solutions to our full set of arrival times. First, \rxj\ may
376: show some rotational instabilities or ``timing noise,'' as seen in
377: radio pulsars. We can estimate the magnitude of the long-term timing
378: noise by fitting the phase residuals with a third-order
379: ($\fdotdot\neq0$) solution like that shown in Fig.~\ref{fig:resid}.
380: This does in fact improve the fit --- giving $\chi^2_\nu=6.02$ for
381: $N_{\rm dof}=30$ with an rms of 0.32~s --- though it is still formally
382: unacceptable. We find $\fdotdot\approx \expnt{2.4(4)}{-25}\mbox{ Hz
383: s}^{-2}$ (this also changes $\fdot$ by 3\% from the value in
384: Tab.~\ref{tab:ephem}), giving a timing noise measurement of
385: $\Delta_8\equiv \log_{10}(|{\ddot \nu}|(10^8)^3/6\nu)= -0.5$. This
386: value is on the high side of, but not outside the range expected from
387: relations between $\dot P$ and $\Delta_8$ found for radio pulsars
388: \citep{antt94} and magnetars \citep{wkf+00,gk02}. We do not believe
389: it is likely that the third-order solution represents a real long-term
390: change in $\fdot$, since the value of $\fdotdot$ changes significantly
391: if one uses just the \chandra, \chandra$+$\xmm, or all of the
392: data\footnote{From Fig.~\ref{fig:resid}, it may appear that one could
393: obtain a better fit by reducing the cycle count for the 1998 HRI
394: point by one; however, doing that, the other \rosat\ points cannot be
395: reproduced any more.} [8(6), 18(4), or $\expnt{2.4(4)}{-25}$
396: $\mbox{Hz s}^{-2}$, with $\chi^2_\nu=0.64$, $2.81$, or $6.02$
397: respectively].
398:
399: A second possibility is a sudden change in rotation --- a glitch, as
400: proposed (and largely rejected) by \citet{dvvmv04} to account for
401: variations in spectral shape (differences of 10\% in the inferred
402: temperature) and pulse shape (changes of 50--100\% in pulsed fraction)
403: observed between 2000 and 2003. The systematic increase in the
404: residuals after MJD~53000 in Figure~\ref{fig:resid} might indeed
405: indicate a glitch. We tried fitting a simple glitch model, in which
406: we assumed that only the frequency changed, that the recovery time was
407: longer than the span of our observations, and that the glitch occurred
408: on 2003 July 1 (MJD~52821), in between the two
409: \xmm\ observations that showed the largest spectral change
410: \citep{dvvmv04}. We find a reasonable fit for a glitch with $\Delta
411: f=\expnt{1.3}{-9}$~Hz and a recovery time $>3$~yr, giving
412: $\chi^2_\nu=7.7$ (see Fig.~\ref{fig:resid}). This would be a small
413: glitch, with $\Delta f/f=\expnt{1.1}{-8}$ compared to values of
414: $10^{-9}$ to $10^{-6}$ for radio pulsars \citep*[with smaller values
415: more typical for larger magnetic fields;][]{lss00}. It also implies
416: that energetically, the putative glitch would be insignificant: the
417: change in kinetic energy of $\sim 10^{36}$~ergs \citep{vrem91} would
418: only be noticeable if dissipated in $<1$~day given the
419: bolometric luminosity of \rxj\ of $\expnt{2}{32}d_{300}^2\mbox{ ergs
420: s}^{-1}$ (where $d=300d_{300}$~pc is the distance to \rxj;
421: \citealt{kvkm+03}). In principle, however, it could still have
422: altered the light curve and spectrum of \rxj\ through realignment of
423: the magnetic field relative to the spin axis.
424:
425: Finally, a more mundane explanation for the relatively poor fit is
426: that the data are from different instruments with different energy
427: responses (even among a single instrument aboard \xmm, the changing
428: filters alter the response). The pulse profile of \rxj\ is known to
429: depend on energy (\citealt{czr+01}; \citetalias{kkvkm02};
430: \citealt{hztb04}) and to change over time \citep{dvvmv04}. While the
431: changes to the shape are small, some systematic offsets are expected between the
432: pulse profiles as measured by different instruments or at different
433: times. We hope to investigate this in more detail in the near
434: future.
435:
436: At present, we cannot distinguish between the various possibilities
437: for the relatively poor fits. The predicted future behavior is
438: different, however, and thus further observations of \rxj\ (some of
439: which are in progress) should be able to distinguish between these
440: models.
441:
442: \section{Discussion \& Conclusions}
443: \label{sec:discuss}
444:
445: From our timing solution, we infer a spin-down rate $\dot
446: P=6.98(2)\times10^{-14}~\secsec$. This is consistent with the
447: limits derived in \citetalias{kkvkm02} and by \citet{zhc+02}, but
448: inconsistent with the tentative solution of \citet{chzz04}, who found
449: $\dot P=\expnt{(1.4\pm0.6)}{-13}\,\,\secsec$ at 99\% confidence (but
450: who noted that elements of their solution were inconsistent with each
451: other and that their analysis was subject to confusing aliases). In
452: Table~\ref{tab:ephem}, we list derived parameters --- rotational
453: energy loss rate, magnetic field, and characteristic age (the latter
454: two under the assumption of magnetic dipole spin-down).
455:
456: The values of $P$ and $\dot P$ place \rxj\ well above above most of
457: the radio-pulsar ``death-lines'' proposed so far
458: \citep*[e.g.,][]{ymj99} and in a region populated by radio pulsars in
459: $P$-$\dot P$ diagrams like that in \citetalias{kkvkm02} (its
460: parameters are approximately between those of PSR~J1830$-$1135, with
461: $P=6.2$~s, $\dot P=\expnt{5}{-14}\,\secsec$, and PSR~J1847$-$0130,
462: with $P=6.7$~s, $\dot P=\expnt{1.3}{-12}\,\secsec$). Hence, \rxj\ may
463: well be a radio pulsar itself, but one whose radio beam(s) do not
464: intersect our line of sight. Its inferred magnetic field,
465: $B=\expnt{2.4}{13}$~G, is not exceptional; the Parkes Multi-beam
466: Survey \citep{mlc+01} in particular has discovered a fair number of
467: radio pulsars with $B\gsim 10^{13}$~G
468: \citep[e.g.,][]{ckl+00,mhl+02,msk+03}, and it is now clear the
469: distribution of magnetic fields is flatter than previously assumed
470: \citep{vml+04}.
471:
472: With $\dot E=\expnt{4.7}{30}\mbox{ ergs s}^{-1}$, \rxj\ is not
473: expected to have much non-thermal X-ray emission: from the relation of
474: \citet{bt97}, one estimates $L_{\rm X,\,non-th}\sim 10^{-3}\dot E
475: \simeq \expnt{5}{27}\mbox{ ergs s}^{-1}$, much smaller than the
476: thermal emission, $L_{\rm X,\,therm}\simeq \expnt{2}{32}d_{300}\mbox{
477: ergs s}^{-1}$. This is consistent with limits from \chandra\ and
478: \xmm\ (\citealt{pmm+01}; \citealt*{pzs02}; \citealt{kvkm+03}).
479:
480: What is somewhat puzzling is the inferred age of 2~Myr. Tracing \rxj\
481: back to OB associations where it might have been born put it close to
482: the Trumpler~10 association $\sim\!0.7$~Myr ago (\citealt*{mzh03};
483: \citealt{kaplan04}). Similarly, based on its estimated temperature
484: and luminosity, most standard cooling models (modified URCA for
485: $1.4~M_{\odot}$ neutron stars) put \rxj\ at $\lsim1$~Myr
486: (\citealt{hh98}; \citetalias{kkvkm02}; \citealt{chzz04}).
487:
488: It is of course possible that \rxj\ was born with a long period and/or
489: had significantly non-dipole braking, such that the spin-down age is
490: not a good estimate of its true age. However, no case with as long a
491: birth period as would be required for \rxj\ is known among radio
492: pulsars \citep[cf.][]{klh+03,gkr04}.
493:
494: Another possible explanation is that \rxj\ was ejected from a binary
495: system with a massive companion $\sim 0.7$~Myr ago, either when the
496: companion underwent a supernova or during a binary exchange interaction.
497: In this case, a relatively long period is expected: if the neutron
498: star accreted matter from its companion, its spin period would have
499: tended toward the equilibrium period, $P_{\rm{}eq}\approx
500: 5{\rm\,s}\,(B/10^{13}{\rm\,G})^{6/7}(\dot M/\dot M_{\rm Edd})^{-3/7}$
501: (where $\dot M$ is the accretion rate and $\dot M_{\rm Edd}$ is the
502: Eddington rate). A relatively short cooling age would also be
503: consistent with this model: the accretion and accompanying steady
504: hydrogen burning could reheat the neutron star (or keep it hot). Of
505: course, it remains to be seen that a suitable evolutionary scenario
506: can be found. In any case, the prediction for the model with two
507: supernovae is that there may well be another $\la\!1~$Myr old neutron
508: star whose proper motion traces back to the same origin as \rxj\
509: \citep[cf.][]{vcc04}.
510:
511: Finally, we can compare the magnetic field strength of
512: $\expnt{2.4}{13}$~G with what is inferred from the broad absorption
513: feature in the spectrum (observed to be at 0.3~keV, which corresponds
514: to 0.39~keV at the surface for a gravitational redshift of 0.3;
515: \citealt{hztb04,vdvmv04}). If due to a proton cyclotron line, one
516: infers $B\simeq\expnt{6}{13}$~G \citep{hztb04}, which is substantially
517: larger than inferred from the spin-down. This may simply reflect the
518: inadequacy of the dipole spin-down model, or the presence of higher
519: order multipoles. On the other hand, based on a comparison with other
520: sources, \citet{vkkd+04} suggested that the absorption feature was due
521: to the transition from the ground state to the second excited tightly
522: bound state of neutral hydrogen, which would require
523: $B\simeq\expnt{2}{13}$~G and matches the spin-down value nicely. If
524: that is the case, higher signal-to-noise spectra should reveal the
525: transition to the first excited state at $\sim\!0.15~$keV.
526:
527: \acknowledgements We thank an anonymous referee for useful comments,
528: and Kaya Mori, George Pavlov, Saul Rappaport, Deepto Chakrabarty, and
529: Peter Woods for helpful discussions. DLK was partially supported by a
530: fellowship from the Fannie and John Hertz Foundation. We acknowledge
531: support through Chandra grant GO4-5082X.
532:
533:
534: %\bibliographystyle{apj}
535: %\bibliography{magrefs,xray,psrrefs,myrefs,snrs}
536:
537: \begin{thebibliography}{10}
538:
539: \bibitem[{Arzoumanian} {et~al.}(1994){Arzoumanian}, {Nice}, {Taylor}, \& {Thorsett}]{antt94}
540: {Arzoumanian}, Z., {Nice}, D.~J., {Taylor}, J.~H., \& {Thorsett}, S.~E. 1994, \apj, 422, 671
541:
542: \bibitem[{Becker} \& {Tr\"{u}mper}(1997){Becker} \& {Tr\"{u}mper}]{bt97}
543: {Becker}, W. \& {Tr\"{u}mper}, J. 1997, \aap, 326, 682
544:
545: \bibitem[{Briel} \& {Pfeffermann}(1995){Briel} \& {Pfeffermann}]{bp95}
546: {Briel}, U.~G. \& {Pfeffermann}, E. 1995, \procspie, 2518, 120
547:
548: \bibitem[{Buccheri} {et~al.}(1983){Buccheri}, {Bennett}, {Bignami}, {Bloemen}, {Boriakoff}, {Caraveo}, {Hermsen}, {Kanbach}, {Manchester}, {Masnou}, {Mayer-Hasselwander}, {Ozel}, {Paul}, {Sacco}, {Scarsi}, \& {Strong}]{bbb+83}
549: {Buccheri}, R., {et al.} 1983, \aap, 128, 245
550:
551: \bibitem[{Camilo} {et~al.}(2000){Camilo}, {Kaspi}, {Lyne}, {Manchester}, {Bell}, {D'Amico}, {McKay}, \& {Crawford}]{ckl+00}
552: {Camilo}, F., {Kaspi}, V.~M., {Lyne}, A.~G., {Manchester}, R.~N., {Bell}, J.~F., {D'Amico}, N., {McKay}, N.~P.~F., \& {Crawford}, F. 2000, \apj, 541, 367
553:
554: \bibitem[{Cox}(2000){Cox}]{allen}
555: {Cox}, A.~N. 2000, Allen's {A}strophysical {Q}uantities, 4th edn. (New York: AIP Press/Springer)
556:
557: \bibitem[{Cropper} {et~al.}(2004){Cropper}, {Haberl}, {Zane}, \& {Zavlin}]{chzz04}
558: {Cropper}, M., {Haberl}, F., {Zane}, S., \& {Zavlin}, V.~E. 2004, \mnras, 351, 1099
559:
560: \bibitem[{Cropper} {et~al.}(2001){Cropper}, {Zane}, {Ramsay}, {Haberl}, \& {Motch}]{czr+01}
561: {Cropper}, M., {Zane}, S., {Ramsay}, G., {Haberl}, F., \& {Motch}, C. 2001, \aap, 365, L302
562:
563: \bibitem[Davis {et~al.}(2003)Davis, Holmes, \& Myers]{dhm03}
564: Davis, W., Holmes, J., \& Myers, R. 2003, in The 2003 Chandra Calibration Workshop
565:
566: \bibitem[{de Vries} {et~al.}(2004){de Vries}, {Vink}, {M{\' e}ndez}, \& {Verbunt}]{dvvmv04}
567: {de Vries}, C.~P., {Vink}, J., {M{\' e}ndez}, M., \& {Verbunt}, F. 2004, \aap, 415, L31
568:
569: \bibitem[{Garmire} {et~al.}(2003){Garmire}, {Bautz}, {Ford}, {Nousek}, \& {Ricker}]{gbf+03}
570: {Garmire}, G.~P., {Bautz}, M.~W., {Ford}, P.~G., {Nousek}, J.~A., \& {Ricker}, G.~R. 2003, \procspie, 4851, 28
571:
572: \bibitem[{Gavriil} \& {Kaspi}(2002){Gavriil} \& {Kaspi}]{gk02}
573: {Gavriil}, F.~P. \& {Kaspi}, V.~M. 2002, \apj, 567, 1067
574:
575: \bibitem[{Gavriil} {et~al.}(2004){Gavriil}, {Kaspi}, \& {Roberts}]{gkr04}
576: {Gavriil}, F.~P., {Kaspi}, V.~M., \& {Roberts}, M.~S.~E. 2004, Advances in Space Research, 33, 592
577:
578: \bibitem[{Haberl}(2004){Haberl}]{haberl04}
579: {Haberl}, F. 2004, in XMM-Newton EPIC Consortium meeting, Palermo, 2003 October 14-16 (astro-ph/0401075)
580:
581: \bibitem[{Haberl} {et~al.}(1997){Haberl}, {Motch}, {Buckley}, {Zickgraf}, \& {Pietsch}]{hmb+97}
582: {Haberl}, F., {Motch}, C., {Buckley}, D. A.~H., {Zickgraf}, F.-J., \& {Pietsch}, W. 1997, \aap, 326, 662
583:
584: \bibitem[{Haberl} {et~al.}(2004){Haberl}, {Zavlin}, {Tr{\" u}mper}, \& {Burwitz}]{hztb04}
585: {Haberl}, F., {Zavlin}, V.~E., {Tr{\" u}mper}, J., \& {Burwitz}, V. 2004, \aap, 419, 1077
586:
587: \bibitem[{Heyl} \& {Hernquist}(1998){Heyl} \& {Hernquist}]{hh98}
588: {Heyl}, J.~S. \& {Hernquist}, L. 1998, \mnras, 298, L17
589:
590: \bibitem[{Jansen} {et~al.}(2001){Jansen}, {Lumb}, {Altieri}, {Clavel}, {Ehle}, {Erd}, {Gabriel}, {Guainazzi}, {Gondoin}, {Much}, {Munoz}, {Santos}, {Schartel}, {Texier}, \& {Vacanti}]{jla+01}
591: {Jansen}, F., {et al.} 2001, \aap, 365, L1
592:
593: \bibitem[{Kaplan}(2004){Kaplan}]{kaplan04}
594: {Kaplan}, D.~L. 2004, Ph.D.~Thesis, California Institute of Technology
595:
596: \bibitem[{Kaplan} {et~al.}(2002){Kaplan}, {Kulkarni}, {van Kerkwijk}, \& {Marshall}]{kkvkm02}
597: {Kaplan}, D.~L., {Kulkarni}, S.~R., {van Kerkwijk}, M.~H., \& {Marshall}, H.~L. 2002, \apjl, 570, L79
598:
599: \bibitem[{Kaplan} {et~al.}(2003){Kaplan}, {van Kerkwijk}, {Marshall}, {Jacoby}, {Kulkarni}, \& {Frail}]{kvkm+03}
600: {Kaplan}, D.~L., {van Kerkwijk}, M.~H., {Marshall}, H.~L., {Jacoby}, B.~A., {Kulkarni}, S.~R., \& {Frail}, D.~A. 2003, \apj, 590, 1008
601:
602: \bibitem[{Kraft} {et~al.}(1997){Kraft}, {Chappell}, {Kenter}, {Kobayashi}, {Meehan}, {Murray}, {Zombeck}, {Fraser}, {Pearson}, {Lees}, {Brunton}, {Barbera}, {Collura}, \& {Serio}]{kck+97}
603: {Kraft}, R.~P., {et al.} 1997, \procspie, 3114, 53
604:
605: \bibitem[{Kramer} {et~al.}(2003){Kramer}, {Lyne}, {Hobbs}, {L{\" o}hmer}, {Carr}, {Jordan}, \& {Wolszczan}]{klh+03}
606: {Kramer}, M., {Lyne}, A.~G., {Hobbs}, G., {L{\" o}hmer}, O., {Carr}, P., {Jordan}, C., \& {Wolszczan}, A. 2003, \apjl, 593, L31
607:
608: \bibitem[{Lyne} {et~al.}(2000){Lyne}, {Shemar}, \& {Smith}]{lss00}
609: {Lyne}, A.~G., {Shemar}, S.~L., \& {Smith}, F.~G. 2000, \mnras, 315, 534
610:
611: \bibitem[{Manchester} {et~al.}(2001){Manchester}, {Lyne}, {Camilo}, {Bell}, {Kaspi}, {D'Amico}, {McKay}, {Crawford}, {Stairs}, {Possenti}, {Kramer}, \& {Sheppard}]{mlc+01}
612: {Manchester}, R.~N., {et al.} 2001, \mnras, 328, 17
613:
614: \bibitem[{McLaughlin} {et~al.}(2003){McLaughlin}, {Stairs}, {Kaspi}, {Lorimer}, {Kramer}, {Lyne}, {Manchester}, {Camilo}, {Hobbs}, {Possenti}, {D'Amico}, \& {Faulkner}]{msk+03}
615: {McLaughlin}, M.~A., {et al.} 2003, \apjl, 591, L135
616:
617: \bibitem[{Morris} {et~al.}(2002){Morris}, {Hobbs}, {Lyne}, {Stairs}, {Camilo}, {Manchester}, {Possenti}, {Bell}, {Kaspi}, {Amico}, {McKay}, {Crawford}, \& {Kramer}]{mhl+02}
618: {Morris}, D.~J., {et al.} 2002, \mnras, 335, 275
619:
620: \bibitem[{Motch} {et~al.}(2003){Motch}, {Zavlin}, \& {Haberl}]{mzh03}
621: {Motch}, C., {Zavlin}, V.~E., \& {Haberl}, F. 2003, \aap, 408, 323
622:
623: \bibitem[{Paerels} {et~al.}(2001){Paerels}, {Mori}, {Motch}, {Haberl}, {Zavlin}, {Zane}, {Ramsay}, {Cropper}, \& {Brinkman}]{pmm+01}
624: {Paerels}, F., {et al.} 2001, \aap, 365, L298
625:
626: \bibitem[{Pavlov} \& {Zavlin}(2003){Pavlov} \& {Zavlin}]{pz03}
627: {Pavlov}, G.~G. \& {Zavlin}, V.~E. 2003, in Texas in Tuscany. XXI Symposium on Relativistic Astrophysics, ed. R.~Bandiera, R.~Maiolino, \& F.~Mannucci (Singapore: World Scientific Publishing), 319--328 (astro-ph/0305435)
628:
629: \bibitem[{Pavlov} {et~al.}(2002){Pavlov}, {Zavlin}, \& {Sanwal}]{pzs02}
630: {Pavlov}, G.~G., {Zavlin}, V.~E., \& {Sanwal}, D. 2002, in Neutron Stars, Pulsars, and Supernova Remnants, ed. W.~Becker, H.~Lesch, \& J.~Tr\"{u}mper (Garching: MPE Rep.~278), 273 (astro-ph/0206024)
631:
632: \bibitem[{Str{\" u}der} {et~al.}(2001){Str{\" u}der}, {Briel}, {Dennerl}, {Hartmann}, {Kendziorra}, {Meidinger}, {Pfeffermann}, {Reppin}, {Aschenbach}, {Bornemann}, {Br{\" a}uninger}, {Burkert}, {Elender}, {Freyberg}, {Haberl}, {Hartner}, {Heuschmann}, {Hippmann}, {Kastelic}, {Kemmer}, {Kettenring}, {Kink}, {Krause}, {M{\" u}ller}, {Oppitz}, {Pietsch}, {Popp}, {Predehl}, {Read}, {Stephan}, {St{\" o}tter}, {Tr{\" u}mper}, {Holl}, {Kemmer}, {Soltau}, {St{\" o}tter}, {Weber}, {Weichert}, {von Zanthier}, {Carathanassis}, {Lutz}, {Richter}, {Solc}, {B{\" o}ttcher}, {Kuster}, {Staubert}, {Abbey}, {Holland}, {Turner}, {Balasini}, {Bignami}, {La Palombara}, {Villa}, {Buttler}, {Gianini}, {Lain{\' e}}, {Lumb}, \& {Dhez}]{sbd+01}
633: {Str{\" u}der}, L., {et al.} 2001, \aap, 365, L18
634:
635: \bibitem[{Tr\"{u}mper}(1993){Tr\"{u}mper}]{trumper93}
636: {Tr\"{u}mper}, J. 1993, Science, 260, 1769
637:
638: \bibitem[{Turner} {et~al.}(2001){Turner}, {Abbey}, {Arnaud}, {Balasini}, {Barbera}, {Belsole}, {Bennie}, {Bernard}, {Bignami}, {Boer}, {Briel}, {Butler}, {Cara}, {Chabaud}, {Cole}, {Collura}, {Conte}, {Cros}, {Denby}, {Dhez}, {Di Coco}, {Dowson}, {Ferrando}, {Ghizzardi}, {Gianotti}, {Goodall}, {Gretton}, {Griffiths}, {Hainaut}, {Hochedez}, {Holland}, {Jourdain}, {Kendziorra}, {Lagostina}, {Laine}, {La Palombara}, {Lortholary}, {Lumb}, {Marty}, {Molendi}, {Pigot}, {Poindron}, {Pounds}, {Reeves}, {Reppin}, {Rothenflug}, {Salvetat}, {Sauvageot}, {Schmitt}, {Sembay}, {Short}, {Spragg}, {Stephen}, {Str{\" u}der}, {Tiengo}, {Trifoglio}, {Tr{\" u}mper}, {Vercellone}, {Vigroux}, {Villa}, {Ward}, {Whitehead}, \& {Zonca}]{taa+01}
639: {Turner}, M.~J.~L., {et al.} 2001, \aap, 365, L27
640:
641: \bibitem[{van Kerkwijk} {et~al.}(2004){van Kerkwijk}, {Kaplan}, {Durant}, {Kulkarni}, \& {Paerels}]{vkkd+04}
642: {van Kerkwijk}, M.~H., {Kaplan}, D.~L., {Durant}, M., {Kulkarni}, S.~R., \& {Paerels}, F. 2004, \apj, 608, 432
643:
644: \bibitem[{van Riper} {et~al.}(1991){van Riper}, {Epstein}, \& {Miller}]{vrem91}
645: {van Riper}, K.~A., {Epstein}, R.~I., \& {Miller}, G.~S. 1991, \apjl, 381, L47
646:
647: \bibitem[{Vink} {et~al.}(2004){Vink}, {de Vries}, {M{\' e}ndez}, \& {Verbunt}]{vdvmv04}
648: {Vink}, J., {de Vries}, C.~P., {M{\' e}ndez}, M., \& {Verbunt}, F. 2004, \apjl, 609, L75
649:
650: \bibitem[{Vlemmings} {et~al.}(2004){Vlemmings}, {Cordes}, \& {Chatterjee}]{vcc04}
651: {Vlemmings}, W.~H.~T., {Cordes}, J.~M., \& {Chatterjee}, S. 2004, \apj, 610, 402
652:
653: \bibitem[{Voges} {et~al.}(1996){Voges}, {Aschenbach}, {Boller}, {Brauninger}, {Briel}, {Burkert}, {Dennerl}, {Englhauser}, {Gruber}, {Haberl}, {Hartner}, {Hasinger}, {Kurster}, {Pfeffermann}, {Pietsch}, {Predehl}, {Rosso}, {Schmitt}, {Tr\"{u}mper}, \& {Zimmermann}]{rbs}
654: {Voges}, W., {et al.} 1996, \iaucirc, 6420, 2
655:
656: \bibitem[{Vranesevic} {et~al.}(2004){Vranesevic}, {Manchester}, {Lorimer}, {Hobbs}, {Lyne}, {Kramer}, {Camilo}, {Stairs}, {Kaspi}, {D'Amico}, {Possenti}, {Crawford}, {Faulkner}, \& {McLaughlin}]{vml+04}
657: {Vranesevic}, N., {et al.} 2004, \apjl, 617, L139
658:
659: \bibitem[{Weisskopf} {et~al.}(2000){Weisskopf}, {Tananbaum}, {Van Speybroeck}, \& {O'Dell}]{wtvso00}
660: {Weisskopf}, M.~C., {Tananbaum}, H.~D., {Van Speybroeck}, L.~P., \& {O'Dell}, S.~L. 2000, \procspie, 4012, 2
661:
662: \bibitem[{Woods} {et~al.}(2000){Woods}, {Kouveliotou}, {Finger}, {G\"{o}\u{g}\"{u}\c{s}}, {Scott}, {Dieters}, {Thompson}, {Duncan}, {Hurley}, {Strohmayer}, {Swank}, \& {Murakami}]{wkf+00}
663: {Woods}, P.~M., {et al.} 2000, \apjl, 535, L55
664:
665: \bibitem[{Young} {et~al.}(1999){Young}, {Manchester}, \& {Johnston}]{ymj99}
666: {Young}, M.~D., {Manchester}, R.~N., \& {Johnston}, S. 1999, \nat,
667: 400, 848
668:
669: \bibitem[{Zane} {et~al.}(2002){Zane}, {Haberl}, {Cropper}, {Zavlin}, {Lumb}, {Sembay}, \& {Motch}]{zhc+02}
670: {Zane}, S., {Haberl}, F., {Cropper}, M., {Zavlin}, V.~E., {Lumb}, D., {Sembay}, S., \& {Motch}, C. 2002, \mnras, 334, 345
671:
672: \bibitem[{Zavlin} {et~al.}(2000){Zavlin}, {Pavlov}, {Sanwal}, \& {Tr{\"u}mper}]{zpst00}
673: {Zavlin}, V.~E., {Pavlov}, G.~G., {Sanwal}, D., \& {Tr{\"u}mper}, J. 2000, \apjl, 540, L25
674:
675: \bibitem[{Zombeck} {et~al.}(1995){Zombeck}, {David}, {Harnden}, \& {Kearns}]{zdhk95}
676: {Zombeck}, M.~V., {David}, L.~P., {Harnden}, F.~R., \& {Kearns}, K. 1995, \procspie, 2518, 304
677:
678: \end{thebibliography}
679:
680:
681:
682:
683: \end{document}
684: