1: %%

2: %% ApJ letter on RXTE Eclipse Observations of EXO 0748-676. 

3: %%

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

5: \bibliographystyle{apj}

6: %% \slugcomment{\today}

7: 

8: \newcommand{\gapprox}   {\lower.4ex\hbox{$\;\buildrel >\over{\scriptstyle\sim}\;$}}

9: \newcommand{\lapprox}   {\lower.4ex\hbox{$\;\buildrel <\over{\scriptstyle\sim}\;$}}

10: \newcommand{\order}[1]{\mbox{$\cal{O}$ ({#1})}}

11: 

12: \newcommand{\etal}{\mbox{\it et~al.}}

13: \newcommand{\suph}{\mbox{$^{\rm h}$}}

14: \newcommand{\supm}{\mbox{$^{\rm m}$}}

15: \newcommand{\sups}{\mbox{$^{\rm s}$}}

16: 

17: \def\arcmin{\hbox{$^\prime$}}

18: \def\arcsec{\hbox{$^{\prime\prime}$}}

19: \def\degs          {$^\circ$}

20: \def\ergps         {ergs s$^{-1}$}

21: \def\fluxunit      {ergs cm$^{-2}$s$^{-1}$}

22: \def\phspecunit    {photons cm$^{-2}$s$^{-1}$keV$^{-1}$}

23: \def\massfluxunit  {g cm$^{-2}$s$^{-1}$}

24: \def\numden        {cm$^{-3}$}

25: \def\colden        {cm$^{-2}$}

26: \def\massden       {g cm$^{-3}$}

27: \def\pcuflux       {counts s$^{-1}$ PCU$^{-1}$}

28: 

29: \def\green{f_{_{\rm G}}}

30: \def\columntotal{\Phi_\epsilon}

31: \def\greencolumn{\Phi_\epsilon^{\rm G}}

32: \def\greenphoton{\dot N_\epsilon^{\rm G}}

33: \def\photonparticular{\dot N_\epsilon}

34: \def\column{\Phi_\epsilon}

35: 

36: \def\colrad{r_0}

37: \def\starad{R_*}

38: \def\starmass{M_*}

39: \def\mdot{\dot M}

40: \def\tauperp{\tau_\perp}

41: \def\tauscale{\tau}

42: \def\taushift{\tau_*}

43: \def\sigpar{\sigma_{_{||}}}

44: \def\sigperp{\sigma_\perp}

45: \def\sigbar{\overline\sigma}

46: 

47: \newcommand{\begeq}     {\begin{equation}}

48: \newcommand{\fineq}     {\end{equation}}

49: \newcommand{\msun}      {M_\odot} 

50: \newcommand{\epsmin}    {\epsilon_{\rm min}}

51: \newcommand{\epsmax}    {\epsilon_{\rm max}}

52: \newcommand{\Tmound}    {T_{\rm th}}

53: \newcommand{\rhomound}  {\rho_{\rm th}}

54: \newcommand{\vmound}    {v_{\rm th}}

55: \newcommand{\zmound}    {z_{\rm th}}

56: \newcommand{\taumound}  {\tau_{\rm th}}

57: \newcommand{\epsilonabs} {\epsilon_{\rm abs}}

58: \newcommand{\chiabs}    {\chi_{\rm abs}}

59: \newcommand{\zmax}      {z_{\rm max}}

60: \newcommand{\taumax}    {\tau_{\rm max}}

61: \newcommand{\vff}       {v_{\rm ff}}

62: \newcommand{\vdag}      {(v)^\dagger}

63: \newcommand{\deltapar}  {\delta}

64: \newcommand{\sig}       {\sigma_{_{\rm T}}}

65: \newcommand{\units}[1]  {\mbox{$\rm\,#1$}}

66: \newcommand{\kpc}       {\mbox{\rm\,kpc}}

67: \newcommand{\Hz}        {\mbox{\rm\,Hz}}

68: \newcommand{\kHz}       {\mbox{\rm\,kHz}}

69: \newcommand{\MHz}       {\mbox{\rm\,MHz}}

70: \newcommand{\GHz}       {\mbox{\rm\,GHz}}

71: \newcommand{\erg}       {\mbox{\rm\,erg}}

72: \newcommand{\Jy}        {\mbox{\rm\,Jy}}

73: \newcommand{\mJy}       {\mbox{\rm\,mJy}}

74: \newcommand{\uJy}       {\mbox{$\,\mu\rm{Jy}$}}

75: \newcommand{\K}         {\mbox{\rm\,K}}

76: \newcommand{\KperJy}    {\mbox{$\rm\,K\,Jy^{-1}$}}

77: \newcommand{\dm}        {\mbox{$\rm\,pc\,cm^{-3}$}}

78: \newcommand{\kms}       {\mbox{$\rm\,km\,s^{-1}$}}

79: \newcommand{\yr}        {\mbox{\rm\,y}}

80: \newcommand{\s}         {\mbox{\rm\,s}}

81: \newcommand{\ms}        {\mbox{\rm\,ms}}

82: \newcommand{\us}        {\mbox{$\,\mu\rm{s}$}}

83: \newcommand{\cm}        {\mbox{\rm\,cm}}

84: \newcommand{\m}         {\mbox{\rm\,m}}

85: \newcommand{\degyr}     {\mbox{$\,^\circ\,{\rm y}^{-1}$}}

86: \newcommand{\degree}    {\mbox{$^\circ$}}

87: \newcommand{\gauss}     {\mbox{\rm\,G}}

88: \newcommand{\accel}     {\mbox{$\rm\,m\,s^{-2}$}}

89: \newcommand{\Msun}      {\mbox{$\,M_{\mathord\odot}$}}

90: \newcommand{\Mearth}    {\mbox{$\,M_{\mathord\oplus}$}}

91: \newcommand{\Lsun}      {\mbox{$\,L_{\mathord\odot}$}}

92: \newcommand{\Rsun}      {\mbox{$\,R_{\mathord\odot}$}}

93: \newcommand{\OMC}		{\mbox{${\rm O}-{\rm C}$}}

94: \newcommand{\omczero}   {\mbox{${\rm O}-{\rm C}$}\,\,=\,\,0}

95: 

96: \newcommand{\Porb}{\mbox{$P_{\rm orb}$}}

97: \newcommand{\Ppul}{\mbox{$P_{\rm pul}$}}

98: \newcommand{\Jorb}{\mbox{$J_{\rm orb}$}}

99: \newcommand{\Porbdot}{\mbox{$\dot P_{\rm orb}$}}

100: \newcommand{\Ppuldot}{\mbox{$\dot P_{\rm pul}$}}

101: \newcommand{\Porbddot}{\mbox{$\ddot P_{\rm orb}$}}

102: \newcommand{\Ppulddot}{\mbox{$\ddot P_{\rm pul}$}}

103: \newcommand{\Rltwo}{\mbox{$R_{L2}$}}

104: \newcommand{\Rlone}{\mbox{$R_{L1}$}}

105: \newcommand{\Rltwodot}{\mbox{$\dot R_{L2}$}}

106: \newcommand{\Rlonedot}{\mbox{$\dot R_{L1}$}}

107: 

108: \newcommand{\exo}{\mbox{EXO\,0748$-$676}}

109: \newcommand{\xonesixfiveeight}{\mbox{X1658$-$298}}

110: \newcommand{\xoneighttwozero}{\mbox{X1820$-$303}}

111: \newcommand{\xoneighttwotwo}{\mbox{X1822$-$371}}

112: \newcommand{\xtwoonetwoseven}{\mbox{X2127$+$119}}

113: \newcommand{\smcxone}{\mbox{SMC\,X$-$1}}

114: \newcommand{\herxone}{Her~X-1}

115: \newcommand{\cenxthree}{\mbox{Cen\,X$-$3}}

116: \newcommand{\grojoneseven}{\mbox{GRO\,J1744$-$28}}

117: \newcommand{\voph}{\mbox{V2301\,Oph}}

118: \newcommand{\amher}{\mbox{AM\,\,Her}}

119: \newcommand{\xper}{\mbox{X\,\,Per}}

120: 

121: \begin{document}

122: 

123: \title{Possible Magnetic Activity in the 

124: Low Mass X-ray Binary EXO~0748$-$676}

125: 

126: \author{Michael T. Wolff\altaffilmark{1},

127: Kent S. Wood\altaffilmark{2}}

128: \and

129: \author{ Paul S. Ray\altaffilmark{3}}

130: 

131: \altaffiltext{1}{Space Science Division, 

132: Naval Research Laboratory, Washington, DC 20375; Michael.Wolff@nrl.navy.mil}

133: \altaffiltext{2}{Space Science Division, 

134: Naval Research Laboratory, Washington, DC 20375; Kent.Wood@nrl.navy.mil}

135: \altaffiltext{3}{Space Science Division, 

136: Naval Research Laboratory, Washington, DC 20375; Paul.Ray@nrl.navy.mil}

137: 

138: \begin{abstract}

139: 

140: We report evidence of magnetic activity associated 

141: with the secondary star in the \exo\ low mass X-ray binary system. 

142: An analysis of a sequence of five consecutive X-ray 

143: eclipses observed during December 2003 with the {\it RXTE} satellite 

144: brings out a feature occurring during ingress we 

145: interpret as the X-ray photoelectric absorption shadow, 

146: as seen by an observer at Earth, 

147: of a plasma structure suspended 

148: above the surface of the secondary star.

149: The light curve feature consists of an initial drop 

150: in count rate to near zero (the absorption shadow) with a very 

151: short rebound to a significant fraction of the 

152: pre-ingress count rate and then a final plunge to totality over a

153: total time scale of $\sim$25 s.

154: The ingress feature persists 

155: for at least 5 consecutive orbital 

156: periods (a total of $\sim$19 hr), and possibly up to 5 days in our data.

157: Our data also show significant post-egress dipping during this 

158: eclipse sequence, unusual for this source, 

159: indicating possible secondary star mass ejection 

160: during this episode.

161: 

162: \end{abstract}

163: 

164: \keywords{X-rays: binaries, binaries: eclipsing, stars: individual (\exo), stars: magnetic fields}

165: 

166: \section{Introduction}

167: 

168: \exo\ is the only persistently active low mass X-ray 

169: binary (LMXB) that displays full X-ray eclipses.

170: In order to observe orbital period variations in a 

171: LMXB we have conducted a long-term observational 

172: program \citep[see][and references therein]{whw+02} 

173: of monitoring the X-ray eclipses of \exo\ ($\Porb = 3.82$ hr) with 

174: the {\it Rossi X-Ray Timing Explorer} ({\it RXTE}) satellite. 

175: By monitoring the mid-eclipse times we can investigate changes 

176: in the orbital period and constrain the magnitudes of 

177: the physical processes that cause the observed changes, allowing

178: estimation of their long-term evolutionary effects.

179:  

180: In this letter we report on a chance observational circumstance 

181: that has allowed us to identify a possible X-ray signature of magnetic 

182: activity in the secondary star in \exo.

183: Magnetic activity has been observed for some time in both

184: the white dwarfs and secondary stars of cataclysmic variable 

185: systems \citep[e.g.,][]{warn95,ws01}.

186: To our knowledge, however, magnetic activity associated with 

187: the secondary stars of neutron star or black hole candidate 

188: LMXB systems has yet to be reported.

189: Indeed, the secondary star in the \exo\ system has never been reliably 

190: observed because of the dominance of the optical emission by the

191: accretion disk \citep{sc87,phs+06}.

192: Magnetic fields associated with the companion star may lead to 

193: modulation of rates of angular momentum exchange and mass transfer 

194: between the binary components and changes in the structure of 

195: the secondary star \citep{tv06}.

196: Estimates of the magnitude of magnetic effects on LMXB evolution 

197: have been hampered by the lack of observations 

198: of magnetic activity in real systems that can be 

199: tied directly to the observable properties of the secondary star.

200: This is because there are few systems where distinct evolutionary 

201: processes can be directly observed (e.g., observing orbital

202: evolution by timing eclipses) and even in systems where 

203: such observations are possible the nature of these processes 

204: is controversial \citep[e.g., see discussion of several eclipsing LMXBs in][]{whw+02}.

205: Finally, in analogy with the 11-year solar cycle, 

206: magnetic fields associated with the secondary star in a LMXB  

207: could show activity cycles \citep{lrr98}, and may also 

208: cause observable transient phenomena.

209: 

210: \section{{\it RXTE} Observations}

211: 

212: The observations we report here were all made with the

213: Proportional Counter Array (PCA) on the {\it RXTE}.

214: The PCA is an array of five large-area X-ray proportional counters

215: (Proportional Counter Units or PCUs) with microsecond timing accuracy 

216: \citep{jmr+06}.  

217: From the beginning of the {\it RXTE} mission (1996) we have 

218: monitored \exo\ to time X-ray eclipses with 

219: sub-second accuracy.

220: This monitoring is organized into ``campaigns'' consisting 

221: of 4$-$6 separate observations of \exo\ over roughly one day with 

222: one or two month spacing between campaigns.

223: Our timing accuracy for each eclipse is limited by 

224: counting statistics in the X-ray flux and any intrinsic variability in 

225: the X-ray emission.  

226: 

227: Eclipse cycle ($N$) is determined by the

228: numbering system of \citet{psvc91} with the updated ephemeris:

229: $T_0 ({\rm TDB;MJD}) = 46111.0751910$ and $\Porb = 0.15933778478$ days. 

230: We record full photon event data using {\it GoodXenon} mode.

231: For each eclipse light curve we select only layer 1 photon events from 

232: those PCUs that are on during the entire observation.

233: This allows us to minimize the noise in background-subtracted light 

234: curves at very low count rates (e.g., at eclipse totality where 

235: the source count rate is near zero). 

236: However, below we will have occasion to refer to the hardness of

237: the source spectrum and in those situations we select events from

238: all layers allowing us to bring out more clearly the spectral character 

239: at energies above 7 keV of the ingress feature we study.

240: Because of improvements in the PCA background model we extract events

241: in the energy range 2$-$20 keV, rather than 

242: 2$-$12 keV as done in \citet{whw+02}

243: We utilize the combined background model released on

244: August 6, 2006 available at the {\it RXTE} web 

245: site\footnote{PCA X-ray background model information is 

246: found at the web site http://rxte.gsfc.nasa.gov/docs/xte/pca\_news.html.}.

247: Each {\it RXTE} observed eclipse is processed using 

248: the FTOOLS data analysis package. 

249: See \citet{whw+02} for further details of our analysis procedure.

250: 

251: \section{Observed Eclipse Profiles}

252: 

253: Of primary interest to us here are five sequential 

254: eclipses, $N = 43095$ through $N = 43099$, observed 

255: by {\it RXTE} on December 4$-$5, 2003. 

256: These are listed in Table~\ref{tbl:humpeclipses} where we 

257: also include for comparison eclipses occurring both before and after 

258: this sequence.

259: The ingress profiles for these five eclipses are shown 

260: in Figure~\ref{fig:sixpanel} along with, for comparison, 

261: the ingress of an eclipse observed roughly 93 days later ($N = 43684$).

262: All five eclipse ingresses ($N = 43095 - 43099$) have an 

263: unusual structure, namely, as X-ray eclipse first contact is approached, 

264: the count rate first goes down, then rebounds to 

265: a significant fraction of the pre-ingress level (we refer to this 

266: as the ingress ``spike''), then drops to  

267: a count rate of essentially zero as real totality begins.

268: Contrast this behavior with the eclipse shown in the bottom panel of 

269: Figure~\ref{fig:sixpanel} ($N = 43684$).

270: In that case the ingress is sharp, lasting only about 2$-$3 seconds 

271: whereas in each of the $N = 43095 - 43099$ eclipses the complete 

272: ingress structure lasts at least 25 s.

273: This spike feature lasts for at least five orbits ($\sim$19 hr) as shown

274: in Figure~\ref{fig:sixpanel}.

275: Also, the ingress spikes occur at the same phase of the orbit relative 

276: to the occulting edge of the secondary for each 

277: $N = 43095 - 43099$ eclipse. 

278: The physical cause of the light curve spike 

279: occurs on the same side of the secondary for each of the five orbits.

280: Furthermore, examining eclipse $N = 43065$ we

281: see the ingress comes $\sim$10 s early compared to a local

282: constant period ephemeris, a possible sign of the emergence of the 

283: loop from the stellar interior before its geometrical 

284: structure allows X-rays to partially transit through it on 

285: their way to the Earth. 

286: 

287: The energy distribution of the photons in the 

288: ingress spike is ``hard'' in that

289: the ratio of hard ($E > 7$ keV) to soft ($E < 7$ keV) count rate 

290: goes up in the spike feature and the extended ingress, as can again 

291: be seen in Figure~\ref{fig:sixpanel}.

292: Indeed, above 7 keV the eclipse ingress, while somewhat disorganized, 

293: {\it occurs on the totality-side of the spike}.

294: In other words, the initial count rate drop occurs because

295: of soft photons being removed from the X-ray beam along our line of sight.

296: This is a signature of photoelectric absorption by material moving ahead of  

297: the occulting edge of the secondary star. 

298: 

299: Another feature of this set of eclipses is a persistent

300: level of {\it post-egress dipping} in the PCA energy band 

301: shown in Figure~\ref{fig:lc70048-13-16-00}.

302: Based on our experience from the eclipse monitoring program, 

303: even though the PCA energy band is relatively hard, 

304: dipping before ingress is not unusual.

305: In fact, dipping at all phases of the \exo\ orbit has been 

306: reported by several authors \citep[e.g.,][]{cbda98,spo05} 

307: but dipping of this magnitude at the immediate post-egress

308: orbit phase is relatively rare.

309: Examining the spectral fits to the post-egress PCA data shows,

310: assuming solar abundances, 

311: the inferred hydrogen column density is $n_H \sim 10^{23}$ \colden\ 

312: for these post-eclipse data.

313: \citet{tcs+97} found from {\it ASCA} observations of \exo\

314: that $n_H$ varied from $4 \times 10^{21}$ \colden\ outside of 

315: any dipping to a high of $10^{23}$ \colden\ during deep dips.

316: After the five eclipse sequence of interest the observed column 

317: density in our post-egress spectral fits goes down by more than 

318: an order of magnitude and the strong post-egress dipping ceases.

319: 

320: \section{Discussion}

321: 

322: The spiked ingress behavior described above can occur if, as the 

323: line of sight from the neutron star to the Earth comes close to the 

324: occulting edge of the secondary star, a high-density 

325: plasma structure suspended above the surface of the secondary and in 

326: the plane of the line of sight, absorbs X-ray photons out of 

327: the beam. 

328: Then, a few seconds later the X-ray line of sight passes 

329: through a lower density region between the suspended structure 

330: and the true occulting edge of the secondary star, 

331: creating the brief spike in the light curve.

332: This eclipse geometry is schematically shown in Figure~\ref{fig:cartonpicture}.

333: Such a structure might be similar to a magnetic loop in the solar

334: atmosphere that is anchored at two foot points of opposite magnetic 

335: polarity \citep[e.g.,][]{bll+02}.

336: 

337: Structures of the sort we propose may last for several rotations 

338: of the secondary star because the secondary's magnetic field along

339: with the relatively gentle gravitation potential gradient 

340: will maintain the mechanical rigidity of the structure.

341: Utilizing Kepler's law and the \exo\ orbital period of \Porb = 3.82 hr, 

342: we obtain

343: $ A \,\, = \,\, 1.4 \left( \frac{M_1}{1.4 \Msun} \right) ( 1 + q )^{1/3} \, \Rsun, $

344: for the binary separation, where $q = M_2/M_1$ is the binary 

345: mass ratio and $M_1$ is the compact object mass.

346: We assume component (1) is a neutron star and set $M_1 = 1.4 \Msun$.

347: If $M_2 = 0.4 \Msun$ \citep{pwgg86} then $q = 0.286$ 

348: and $A = 1.5$ \Rsun.

349: The radius of the secondary star, $R_2$, is

350: the radius of the Roche lobe surrounding it \citep{egg83},  

351: %

352: \begin{equation}

353: R_2 \,\, = \,\, \frac{0.49 q^{2/3}}{0.6 q^{2/3} + 

354: \ln ( 1 + q^{1/3})} \, A \, \sim \, 0.42 \, \Rsun,

355: \label{eq:roche}

356: \end{equation}

357: %

358: for the above $A$ and $q$.

359: An estimate of the width of the feature is

360: $w \sim \frac{10 s}{\frac{1}{2} \Delta T_{ec}} \times R_2 \sim 0.0084 R_2 \sim 5.9 \times 10^8 \, \mathrm{cm}$

361: where $\Delta T_{ec}$ is the average duration of eclipse 

362: totality (495 s), 10 s is the time scale of the drop 

363: before the spike feature,

364: and where we have assumed that the system 

365: inclination is $i \geq 75^{\circ}$ \citep{pwgg86}. 

366: %% can be ignored [sin($80^{\circ}$) = 0.98]. 

367: From Figure~\ref{fig:sixpanel} the count rate goes down 75\% 

368: during the initial drop before the spike. 

369: Simulating the \exo\ pre-ingress spectrum to determine the necessary 

370: column density to make the 2$-$7 keV flux drop by this amount 

371: yields a column density $N_H \sim 2.5 \times 10^{23}$ \colden\ implying

372: a hydrogen density $n_H \sim N_H / w \sim 4 \times 10^{14}$ \numden\ for 

373: the loop structure.

374: This density is higher than the corresponding densities in

375: solar magnetic loops \citep[$n_H \sim 10^9$\ \numden;][]{bll+02}

376: and even higher than coronal densities observed in the 

377: G5III flare system Capella 

378: \citep[$n_H \sim 4 \times 10^{11} - 10^{13}$\ \numden;][]{dbd+93}.

379: The secondary mass we use above ($M_2 = 0.4 \Msun$) corresponds to

380: either a late K or early M type main sequence star. 

381: Such stars can have magnetic activity if they rotate and

382: if they have significant convective envelopes \citep{hn87}. 

383: The fact that \exo\ secondary rotates very fast (assuming it 

384: rotates synchronously) and that stars in this mass range are 

385: thought to have significant convective layers does support 

386: our magnetic loop hypothesis.

387: 

388: Figure~\ref{fig:lc70048-13-16-00} shows that post-egress dipping is 

389: prominent during the five eclipse sequence we are studying.

390: This post-egress dipping could indicate significant mass ejection 

391: that moves away from the secondary accompanying the magnetic activity.

392: Such mass ejections could be similar to coronal mass ejections that 

393: sometimes accompany solar magnetic activity.

394: Because we observe only a total column density, however,

395: we can not establish whether this material is associated 

396: with the accretion disk inside the neutron star Roche lobe,

397: or it is outside the neutron star Roche lobe and moving 

398: away radially from the secondary.

399: 

400: In the simple model of \citet{ll02}, in order for the structures to

401: last in the solar corona they must be in hydrostatic equilibrium, 

402: internal velocities within the loop must be negligible, pressure 

403: gradients should be small, and energy dissipation processes 

404: within the loop must be uniform.

405: Because the loop structure resides near the surface of the 

406: secondary star it will be near a saddle point in the Roche 

407: gravitational potential function.

408: Thus, the effective force on a plasma structure will be 

409: small near the surface of the secondary. 

410: Second, if the forces on the loop are small then, unless there is 

411: significant non-uniformity in the energy dissipation or heating 

412: processes along the loop, 

413: velocities should be negligibly 

414: small and any pressure gradients should also be small.

415: However, heating and cooling processes within the loop are 

416: difficult to characterize.

417: The loop may have a hydrogen-poor abundance reflecting 

418: the abundance of the secondary star.

419: Also, the loop will be bathed in 

420: substantial X-rays from the accretion onto the neutron star, 

421: the radiation from an optically thin coronal plasma 

422: (if one exists), and from the radiation emerging from the 

423: stellar photosphere.

424: At a density of $n_H \sim 4 \times 10^{14}$ \numden\ the loop may  

425: be optically thick, making its energy balance particularly difficult 

426: to determine. 

427: 

428: The set of anomalous eclipses runs from cycle 

429: $N = 43095$ to $N = 43099$.

430: Based on the \OMC\ residuals there is considerable intrinsic 

431: period jitter of the type studied in \citet{hwc97} 

432: and \citet{whw+02} prior to the spike eclipses.

433: Other than this jitter, however, both before and after this set of 

434: observations the eclipses appear to be normal in that both the ingresses 

435: and egresses are well behaved although some eclipses 

436: have longer than average duration.

437: However, the eclipses running from $N = 43684$ to $N = 44363$,

438: in particular the set $N = 43684- 43687$ 

439: (Table~\ref{tbl:humpeclipses}), after the ingress feature disappears, 

440: are among the most stable eclipses we have observed from \exo\ 

441: with {\it RXTE} over our entire monitoring program.

442: The emergence of the ingress magnetic loop and significant 

443: post-eclipse dipping may signify a dramatic transient event followed 

444: by quiescence in the system.

445: The stable eclipses $N = 43684 - 43687$ occur during the first of 

446: a series of luminosity excursions observed by the {\it RXTE} 

447: All Sky Monitor that have continued on and off to the present day.

448: During those excursions the \exo\ X-ray brightness can increase from its

449: average of $\sim 8$ mcrab by factors of $2-5$ for weeks at a time.

450: Magnetic activity in the secondary star may be modulating 

451: the mass accretion rate onto the neutron star and affecting

452: the system X-ray luminosity.

453: 

454: \section{Conclusions}

455: 

456: We have reported on a sequence of five consecutive X-ray 

457: eclipses observed by {\it RXTE} during December 2003 from 

458: the \exo\ LMXB system that show a repeating feature during 

459: ingress transitions.

460: This feature is an anomalous X-ray peak during ingress we 

461: interpret as X-rays emitted from near the neutron star passing 

462: {\it underneath} a rigid plasma structure suspended above 

463: the surface of the secondary. 

464: This loop structure may be similar to magnetic loops 

465: suspended above the solar photosphere.

466: This conclusion is based on the ingress feature lasting for at least five 

467: orbit periods and perhaps as long as five days, although this latter 

468: conclusion depends on the interpretation of one ingress profile.

469: The occurrence of this ingress feature may be an indication of a 

470: transient event of significant magnitude in the secondary star 

471: since the \OMC\ residuals around the time of the event

472: show the mid-eclipse timings wandering before the ingress feature 

473: appears but settling down after the ingress feature disappears.

474: During this sequence of eclipses we also observed significant 

475: {\it post}-egress dipping which, as we noted, is unusual.

476: This suggests that significant mass may have been liberated from 

477: the secondary star (similar to a coronal mass ejection) as part

478: of this transient event.  

479: The occurrence of the X-ray spike on the limb of the secondary at

480: the same orbit phase in sequential eclipses 

481: is consistent with the conventional picture of synchronous rotation 

482: as suggested by a number of investigations \citep[e.g.,][]{vp95}.

483: Magnetically maintained features playing a role in 

484: determining the eclipse occulting edge geometry 

485: could lead to significant departures from symmetry between 

486: ingress and egress.

487: This may be the cause of the intrinsic jitter identified by

488: \citet{hwc97}\ and \citet{whw+02}.

489: If true then the magnetic field of the secondary may

490: perturb the structure of the secondary star, moderating both mass transfer 

491: and angular momentum exchange.

492: 

493: \begin{acknowledgements}

494: We thank Drs. Paul Hertz, Gerald Share, and J. Martin Laming 

495: for useful discussions and Evan Smith for help 

496: with {\it RXTE} scheduling.

497: We thank an anonymous referee for critically reading the manuscript

498: and many helpful suggestions.  

499: This work was supported by the NASA {\it RXTE} Guest Observer Program 

500: and the Office of Naval Research.

501: \end{acknowledgements}

502: 

503: %%\bibliography{journapj,usarefs}

504: 

505: \begin{thebibliography}{20}

506: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi

507: 

508: \bibitem[{{Brkovi{\'c}} {et~al.}(2002){Brkovi{\'c}}, {Landi}, {Landini},

509:   {R{\"u}edi}, \& {Solanki}}]{bll+02}

510: {Brkovi{\'c}}, A., {Landi}, E., {Landini}, M., {R{\"u}edi}, I., \& {Solanki},

511:   S.~K. 2002, A\&A, 383, 661

512: 

513: \bibitem[{Church {et~al.}(1998)Church, Balucinska-Church, Dotani, \&

514:   Asai}]{cbda98}

515: Church, M.~J., Balucinska-Church, M., Dotani, T., \& Asai, K. 1998, ApJ, 504,

516:   516

517: 

518: \bibitem[{{Dupree} {et~al.}(1993){Dupree}, {Brickhouse}, {Doschek}, {Green}, \&

519:   {Raymond}}]{dbd+93}

520: {Dupree}, A.~K., {Brickhouse}, N.~S., {Doschek}, G.~A., {Green}, J.~C., \&

521:   {Raymond}, J.~C. 1993, ApJ, 418, L41+

522: 

523: \bibitem[{{Eggleton}(1983)}]{egg83}

524: {Eggleton}, P.~P. 1983, ApJ, 268, 368

525: 

526: \bibitem[{{Hartmann} \& {Noyes}(1987)}]{hn87}

527: {Hartmann}, L.~W., \& {Noyes}, R.~W. 1987, Ann. Rev. Astr. Astrphys., 25, 271

528: 

529: \bibitem[{Hertz {et~al.}(1997)Hertz, Wood, \& Cominsky}]{hwc97}

530: Hertz, P., Wood, K.~S., \& Cominsky, L.~R. 1997, ApJ, 486, 1000

531: 

532: \bibitem[{{Jahoda} {et~al.}(2006){Jahoda}, {Markwardt}, {Radeva}, {Rots},

533:   {Stark}, {Swank}, {Strohmayer}, \& {Zhang}}]{jmr+06}

534: {Jahoda}, K., {Markwardt}, C.~B., {Radeva}, Y., {Rots}, A.~H., {Stark}, M.~J.,

535:   {Swank}, J.~H., {Strohmayer}, T.~E., \& {Zhang}, W. 2006, ApJS, 163, 401

536: 

537: \bibitem[{{Landini} \& {Landi}(2002)}]{ll02}

538: {Landini}, M., \& {Landi}, E. 2002, A\&A, 383, 653

539: 

540: \bibitem[{{Lanza} {et~al.}(1998){Lanza}, {Rodono}, \& {Rosner}}]{lrr98}

541: {Lanza}, A.~F., {Rodono}, M., \& {Rosner}, R. 1998, MNRAS, 296, 893

542: 

543: \bibitem[{Parmar {et~al.}(1991)Parmar, Smale, Verbunt, \& Corbet}]{psvc91}

544: Parmar, A.~N., Smale, A.~P., Verbunt, F., \& Corbet, R. H.~D. 1991, ApJ, 366,

545:   253

546: 

547: \bibitem[{Parmar {et~al.}(1986)Parmar, White, Giommi, \& Gottwald}]{pwgg86}

548: Parmar, A.~N., White, N.~E., Giommi, P., \& Gottwald, M. 1986, ApJ, 308, 199

549: 

550: \bibitem[{{Pearson et~al.}(2006)}]{phs+06}

551: {Pearson et~al.} 2006, ApJ, 648, 1169

552: 

553: \bibitem[{{Schmidtke} \& {Cowley}(1987)}]{sc87}

554: {Schmidtke}, P.~C., \& {Cowley}, A.~P. 1987, Astron. J., 93, 374

555: 

556: \bibitem[{Sidoli {et~al.}(2005)Sidoli, Parmar, \& Oosterbroek}]{spo05}

557: Sidoli, L., Parmar, A.~N., \& Oosterbroek, T. 2005, A\&A, 429, 291

558: 

559: \bibitem[{{Tauris} \& {van den Heuvel}(2006)}]{tv06}

560: {Tauris}, T.~M., \& {van den Heuvel}, E.~P.~J. 2006, in Compact Stellar X-ray

561:   Sources, ed. W.~H.~G. Lewin \& M.~van~der Klis (Cambridge University Press),

562:   623

563: 

564: \bibitem[{Thomas {et~al.}(1997)Thomas, Corbet, Smale, Asai, \& Dotani}]{tcs+97}

565: Thomas, B., Corbet, R., Smale, A.~P., Asai, K., \& Dotani, T. 1997, ApJ, 480,

566:   L21

567: 

568: \bibitem[{Verbunt \& Phinney(1995)}]{vp95}

569: Verbunt, F., \& Phinney, E.~S. 1995, A\&A, 296, 709

570: 

571: \bibitem[{{Warner}(1995)}]{warn95}

572: {Warner}, B. 1995, {Cataclysmic variable stars} (Cambridge Astrophysics Series,

573:   Cambridge, New York: Cambridge University Press)

574: 

575: \bibitem[{{Washuettl} \& {Strassmeier}(2001)}]{ws01}

576: {Washuettl}, A., \& {Strassmeier}, K.~G. 2001, A\&A, 370, 218

577: 

578: \bibitem[{Wolff {et~al.}(2002)Wolff, Hertz, Wood, Ray, \&

579:   Bandyopadhyay}]{whw+02}

580: Wolff, M.~T., Hertz, P.~L., Wood, K.~S., Ray, P.~S., \& Bandyopadhyay, R.~M.

581:   2002, ApJ, 575, 384

582: 

583: \end{thebibliography}

584: 

585: %%

586: %%	Figures:

587: %%

588: \begin{figure}

589: \centerline{\includegraphics[height=5.9in,angle=0.0]{f1.ps}}

590: \caption{Six panels that show the light curves (crosses) and hardness

591: ratios (triangles) of the five spiked eclipse ingresses  

592: (top five panels), and one particularly 

593: sharp eclipse ($N = 43684$) observed 93 days later. 

594: Phase zero is set by the mid-eclipse 

595: time for eclipse $N = 43684$ and then applied to the other eclipses

596: so that all ingress points can be seen phased to the same

597: {\it local} ephemeris.

598: The light curve points are the PCA count rate per PCU in

599: 0.5 s time bins for the 2$-$20 keV energy band and are connected 

600: with lines to allow them to be easily distinguished.

601: The hardness is the ratio of the count rate in the 7$-$20 keV energy

602: band to that in the 2$-$7 keV band.

603: Inside eclipse totality hardness points are not plotted.

604: The spike we discuss is centered at phase $\phi = -0.0185$ (downward arrow) 

605: while the real X-ray ingress appears to start 8 s later 

606: at phase $\phi = -0.0179$.

607: \label{fig:sixpanel}}

608: \end{figure}

609: 

610: \begin{figure}

611: \centerline{\includegraphics[height=6.0in,angle=0.0]{f2.ps}}

612: \caption{The light curves for the 2$-$20 keV energy band for the region 

613: around the eclipses $N = 43095$ (top panel), and $N = 43684$ (bottom panel).

614: Eclipse $N = 43095$ is the first of the eclipses with

615: ingress spikes and eclipse $N = 43684$ occurs three months later.

616: The binary orbit phasing is arbitrary, only layer 1 events are 

617: included with time binning 0.5 s.

618: Note the strong dipping after $N = 43095$ eclipse egress whereas

619: the post-egress count rate is steady for the $N = 43684$ eclipse. 

620: Our experience is that this level of post-egress dipping is rare 

621: during our monitoring observations.

622: The count rate in each time bin is plotted with $\pm 1 \sigma$ error bars

623: and the time scale is compressed relative to Figure~\ref{fig:sixpanel} 

624: hence the ingress spike feature shown in Figure~\ref{fig:sixpanel} does not 

625: stand out as strongly in the upper panel of this figure. 

626: \label{fig:lc70048-13-16-00}}

627: \end{figure}

628: 

629: \begin{figure}

630: \centerline{\includegraphics[height=3.0in,angle=0.0]{f3.eps}}

631: \caption{A schematic drawing of our model for the ``spike'' eclipses

632: from the point of view of an observer at Earth. 

633: The spike in the ingress profile corresponds to when the apparent 

634: neutron star position is between the loop and the edge of the secondary star and 

635: shinning through the lower density material under the loop.

636: True eclipse totality begins as the apparent position of the neutron star moves 

637: completely behind the secondary star and lasts until neutron star re-emergence. 

638: \label{fig:cartonpicture}}

639: \end{figure}

640: 

641: %%%

642: %%%  TABLES:

643: %%%

644: %

645: % Table 1:EXO 0748-676 Eclipses Involved in this paper:

646: %

647: \begin{deluxetable}{cccc}

648: \tablecolumns{4}

649: \tabletypesize{\scriptsize}

650: \tablecaption{\bf{{\it RXTE} Timing of Selected Full \exo\ X-ray Eclipses}

651: \label{tbl:humpeclipses}}

652: \tablewidth{420pt}

653: \tablehead{\colhead{{\it RXTE} ObsID} &\colhead{Eclipse Cycle} 

654: &\colhead{Fitted Mid-Eclipse Time\tablenotemark{1}}

655:  &\colhead{Comment}\\  &  & (MJD;TDB) & }

656: \startdata

657: %% 80040-01-03-00   &42618  &52901.732868(3)  &Normal eclipse.\\

658: %% 80040-01-04-00   &42632  &52903.963581(4)  &Burst partially obscures post-egress light curve.\\

659: %% 80040-01-04-01   &42644  &52905.875627(5)  &Some post-egress dipping.\\

660: %% 70048-13-11-00   &42715  &52917.188590(5)  &Normal eclipse.\\

661: 70048-13-12-00   &42716  &52917.347911(3)  &Normal eclipse.\\

662: 70048-13-13-00   &42717  &52917.507251(5)  &Burst partially obscures post-egress light curve.\\

663: 70048-13-14-00   &42718  &52917.666590(6)  &Normal eclipse.\\

664: 70048-13-15-00   &42719  &52917.825925(5)  &Normal eclipse; Some post-egress dipping.\\

665: 80040-01-07-00   &43065  &52972.956764(5)  &Apparent eclipse duration 505 s.\\

666: 70048-13-16-00   &43095  &52977.736972(7)\tablenotemark{2} &Double-peaked ingress; Post-egress dipping.\\

667: 70048-13-17-00   &43096  &52977.89628(1)\tablenotemark{2}  &Double-peaked ingress; Post-egress dipping.\\

668: 70048-13-18-00   &43097  &52978.055689(4)\tablenotemark{2} &Double-peaked ingress; Post-egress dipping.\\

669: 70048-13-19-00G  &43098  &52978.214950(7)\tablenotemark{2} &Double-peaked ingress; Post-egress dipping.\\

670: 70048-13-20-00   &43099  &52978.374306(2)\tablenotemark{2} &Double-peaked ingress.\\

671: 80040-01-06-00   &43342  &53017.093350(3)  &Normal eclipse.\\

672: %% 90059-01-01-00   &43414  &53028.565701(3)  &Eclipse duration 500 s.\\

673: %% 90059-01-02-00   &43415  &53028.725027(3)  &Normal eclipse.\\

674: %% 90059-01-03-00   &43416  &53028.884375(4)  &Normal eclipse.\\

675: %% 90059-01-04-00   &43417  &53029.043728(3)  &Normal eclipse.\\

676: 90059-02-01-00   &43684  &53071.5869083(9)  &Eclipse features sharp; High count rate.\\

677: 90059-02-02-00   &43685  &53071.746244(3)  &Eclipse features sharp; High count rate.\\

678: 90059-02-03-00   &43686  &53071.9055842(6)  &Eclipse features sharp; High count rate.\\

679: 90059-02-04-00   &43687  &53072.064922(2)  &Eclipse features sharp; High count rate.\\

680: \enddata

681: \tablenotetext{1}{The fitted mid-eclipse times given to the first 

682: uncertain digit; The errors are in parenthesis for that digit.}

683: \tablenotetext{2}{No adjustment is made in the fitted mid-eclipse 

684: time for the double-peaked ingress profile.}

685: \end{deluxetable}

686: 

687: %%%

688: %%%  End of document:

689: %%%

690: \end{document}

691: