1: % File draft date:  9 Dec 2003

2: % \documentstyle[12pt,../aasms]{article}

3: %\documentstyle[12pt,emulateapj5]{article}

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

5: %\doublespace

6: 

7: %\received{2004 January 23}

8: \begin{document}

9: \newcommand{\etal}{{\it et al.}}

10: 

11: \include{idl}

12: 

13: \title{{\it CHANDRA} Observations of V407 Vul: Confirmation of the

14: Spin-up} \author{Tod E. Strohmayer} \affil{Laboratory for High Energy

15: Astrophysics, NASA's Goddard Space Flight Center, Greenbelt, MD 20771;

16: stroh@clarence.gsfc.nasa.gov}

17: 

18: \begin{abstract}

19: 

20: V407 Vul (RX J1914.4+2456) is a candidate double-degenerate binary

21: with a putative 1.756 mHz (9.5 min) orbital frequency. In a previous

22: timing study using archival ROSAT and ASCA data we reported evidence

23: for an increase of this frequency at a rate consistent with

24: expectations for gravitational radiation from a detached ultracompact

25: binary system.  Here we report the results of new {\it Chandra} timing

26: observations which confirm the previous indications of spin-up of the

27: X-ray frequency, and provide much tighter constraints on the frequency

28: derivative, $\dot\nu$. We obtained with {\it Chandra} a total of 90

29: ksec of exposure in two epochs separated in time by 11.5 months. The

30: total time span of the archival ROSAT, ASCA and new {\it Chandra} data

31: is now $\approx 10.5$ years. This more than doubles the interval

32: spanned by the ROSAT and ASCA data alone, providing much greater

33: sensitivity to a frequency derivative. With the addition of the

34: Chandra data an increasing frequency is unavoidable, and the mean

35: $\dot\nu$ is $7.0 \pm 0.8 \times 10^{-18}$ Hz s$^{-1}$. Although a

36: long-term spin-up trend is confirmed, there is excess variance in the

37: phase timing residuals, perhaps indicative of shorter timescale torque

38: fluctuations or phase instability associated with the source of the

39: X-ray flux. Power spectral searches for periods longward of the 9.5

40: minute period do not find any significant modulations, however, the

41: sensitivity of searches in this frequency range are somewhat

42: compromised by the dithering of the Chandra attitude. The observed

43: spin-up is of a magnitude consistent with that expected from

44: gravitational radiation decay, however, the factor of $\approx 3$

45: variations in flux combined with the timing noise could conceivably

46: result from accretion-induced spin-up of a white dwarf.  Continued

47: monitoring to explore correlations of torque with X-ray flux could

48: provide a further test of this hypothesis.

49: 

50: \end{abstract}

51: 

52: \keywords{Binaries: general - Stars: individual (RX J1914.4+2456, V407

53: Vul) - Stars: white dwarfs - cataclysmic variables - X-rays: stars -

54: X-rays: binaries}

55: 

56: \section{Introduction}

57: 

58: Ultracompact binary systems could provide a promising means to

59: observe directly the influence of gravitational radiation on orbital

60: evolution. Moreover, such systems would be ideal sources for detection

61: with spaced based gravitational radiation observatories (such as the

62: planned NASA/ESA LISA mission), opening up the possibility for

63: detailed studies of compact interacting binaries.

64: 

65: In recent years a pair of candidate ultracompact systems; V407 Vul

66: (also known as RX J1914.4+2456) and RX J0806+1527 (hereafter J0806)

67: have been proposed.  These objects were first discovered by ROSAT

68: (Motch et al.  1996; Israel et al. 1999; Beuermann et al. 1999), and

69: initially were suggested to be members of a ``soft'' class of

70: Intermediate Polars (IPs), with the X-ray periods of 569 and 321 s,

71: respectively, representing the putative spin periods of the accreting

72: white dwarfs.  Since their discovery extensive follow-up observations

73: have identified the optical counterparts (Ramsay et al. 2000; Israel

74: et al. 2002; Ramsay, Hakala \& Cropper 2002). Their soft X-ray

75: spectra, the shape of the X-ray modulation, the phasing of the X-ray

76: and optical modulations, the lack of additional longer periods, and

77: the lack of strong optical emission lines have all called into

78: question their IP credentials (for a discussion see Cropper et

79: al. 2003). However, Norton, Haswell \& Wynn (2004) have argued that an

80: IP interpretation is still plausible if the systems are stream-fed,

81: pole-switching accretors (ie. no accretion disk), and are viewed from

82: a nearly face-on geometry. 

83: 

84: It was Cropper et al. (1998) who first suggested that V407 Vul might

85: be a double-degenerate compact binary.  They proposed a synchronized,

86: magnetic accretor (polar-like) model with accretion powering the X-ray

87: flux. A non-magnetic variant was subsequently proposed by Marsh \&

88: Steeghs (2002).  In this Algol-like model, the accretion stream

89: impacts directly onto a non-magnetic primary, and the spins are not

90: necessarily synchronized with the orbit.  An interesting alternative

91: not requiring accretion was proposed by Wu et al. (2002). They

92: suggested a unipolar inductor model, analogous to the Jupiter - Io

93: system (Clarke et al.  1996). If these systems are indeed compact, and

94: thus the observed X-ray period is the orbital period, then an

95: important discriminating factor is the magnitude and sign of the

96: orbital evolution. If the systems are accreting stably from degenerate

97: donors, the expected evolution is for the orbit frequency to

98: decrease. In a previous study we (Strohmayer 2002) used archival ROSAT

99: and ASCA data to explore the evolution of the 1.756 mHz X-ray

100: frequency of V407 Vul over an $\approx 5$ yr time period, and found

101: evidence for a positive frequency derivative, $\dot\nu$, with a

102: magnitude consistent with simple expectations for gravitational

103: radiation induced decay of a circular orbit. Since a measurement of

104: the frequency evolution places severe constraints on possible models,

105: it is crucial to confirm the initial indications of orbital decay and

106: place tighter constraints on $\dot\nu$.  In this paper we present the

107: results of new {\it Chandra} observations which confirm an increase in

108: the X-ray frequency and allow us to place much tighter limits on

109: $\dot\nu$. In \S 2 we describe the {\it Chandra} observations and the

110: data extraction and analysis.  In \S 3 we discuss our phase coherent

111: timing study, and we show that the inclusion of the {\it Chandra} data

112: conclusively indicates a positive $\dot\nu = 7.0 \pm 0.8 \times

113: 10^{-18}$ Hz s$^{-1}$. We also discuss the flux variability of the

114: source and excess variance (timing noise) in the phase residuals. In

115: \S 4 we discuss the implications of our findings for the nature of

116: V407 Vul. We conclude in \S 5 with a brief summary and goals for

117: future observations.

118: 

119: \section{Data Extraction}

120: 

121: We obtained a total of $\approx 90$ ksec of exposure on V407 Vul with

122: {\it Chandra} in February, November, December 2003, and January, 2004.

123: A summary of these observations is given in Table 1.  We observed with

124: ACIS-S in Continuous Clocking (CC) mode in order to mitigate the

125: effects of pile-up. To maximize the soft photon response the aimpoint

126: was on the backside illuminated (S3) chip. Preliminary spectral

127: analysis shows that virtually all the source photons have energies

128: less than 1 keV. For the purposes of our work reported here we used

129: only $< 1$ keV photons.  Details of the {\it Chandra} spectroscopy

130: will be presented in a future paper.

131: 

132: To prepare the data for a precise timing analysis we first corrected

133: the detector read out times to arrival times following the CXC

134: analysis thread on timing with CC mode data. We then corrected the

135: arrival times to the solar system barycenter using the CIAO tool {\it

136: axbary}. We used the same source coordinates employed for analysis of

137: the ROSAT and ASCA data (see Strohmayer 2002). This produces a set of

138: photon arrival times in the barycentric dynamical time system

139: (TDB). The ACIS/CC mode produces a one-dimensional ``image'' of the

140: portion of the sky exposed to the detectors. A realization of this

141: image for the February, 2003 observations is shown in Figure 1. V407

142: Vul is the strong peak in the ``image.'' We extracted only events from

143: within the source peak for our timing study.  Figure 2 shows a portion

144: of the lightcurve produced from the source extracted events, and

145: demonstrates that {\it Chandra} easily detects individual pulses from

146: the source. We carried out this procedure on all the data and obtained

147: a total of 29,383 {\it Chandra} events for our timing analysis.

148: 

149: \section{Results}

150: 

151: \subsection{Coherent Timing Analysis}

152: 

153: We performed our coherent timing studies using the $Z_n^2$

154: statistic. Since the method is described in our earlier paper we do

155: not repeat all the details here (see Strohmayer 2002).  As a first

156: step we calculated two $Z_3^2$ power spectra. The first uses just the

157: earlier ROSAT and ASCA data, while the second uses only the new {\it

158: Chandra} data. This procedure allows us to compare the frequencies

159: measured at two widely spaced epochs. A comparison of the two then

160: determines the nature of any long term trend in the frequency

161: evolution. This is perhaps the simplest possible analysis that one can

162: do to search for a frequency change. The results are shown in Figures

163: 3 and 4.  Figure 3 shows the $Z_3^2$ power spectra for the two

164: different epochs, ROSAT + ASCA (top), and {\it Chandra} (bottom). In

165: each case we plot the difference, $\Delta Z_3^2$, between the peak

166: value of $Z_3^2$ and the values at other frequencies. In the absence

167: of a signal, the $Z_3^2$ statistic has noise properties given by the

168: $\chi^2$ distribution with $2\times 3 = 6$ degrees of freedom

169: (dof). The ``forest'' of minima in each plot results from the

170: relatively sparse data sampling. Note, however, that for each dataset

171: the best frequency {\it can} be identified unambiguously (the one with

172: $\Delta Z_3^2 = 0$). This is crucially important, as correctly pointed

173: out by Woudt \& Warner (2004), who cautioned that frequency evolution

174: claims for RX J0806.3+1527 and V407 Vul (see Hakala et al. 2003;

175: Strohmayer 2003; Strohmayer 2002) might be compromised by an inability

176: to identify the correct frequency at a given epoch. For the ROSAT +

177: ASCA data the next-best alias peak is at $\Delta Z_3^2 \approx 1300$

178: from the best frequency. This is an enormously significant difference

179: in $\Delta Z_3^2$. The probability that this alias could be the true

180: frequency, compared to a frequency with a $Z_3^2$ value larger by

181: 1300, is vanishingly small. The same is true for the {\it Chandra}

182: data, for which the nearest alias peak has a $\Delta Z_3^2 \approx

183: 696$. Although a smaller difference than for the ROSAT + ASCA data,

184: the probability of this being the true frequency is still incredibly

185: small. As an example, the probability to obtain a difference in

186: $\Delta Z_3^2$ of 100.0 (much smaller than the observed values) purely

187: by chance is $ < 3 \times 10^{-19}$.

188: 

189: Having shown that at each epoch we can unambiguously identify the

190: correct frequency, we can now compare these two frequencies and test

191: whether the frequency has changed. Figure 4 shows a zoom-in around the

192: best ROSAT + ASCA (dashed) and {\it Chandra} (solid) frequencies, and

193: clearly demonstrates that the frequency is higher in the {\it Chandra}

194: epoch. The increase in frequency is about $2$ nHz. A simple estimate

195: of the mean frequency derivative can be obtained by simply dividing

196: the measured frequency increase by the time interval between the mean

197: epoch of the two datasets.  This gives a value $<\dot\nu> = 2 \times

198: 10^{-9} \ {\rm Hz} / 7.6 \ {\rm yr} = 8.34 \times 10^{-18}$ Hz

199: s$^{-1}$, which is consistent with the limits on $\dot\nu$ deduced

200: from our earlier study as well as the more detailed analysis described

201: below.

202: 

203: We next performed a coherent, phase connected timing analysis to

204: jointly constrain $\nu_0$ and $\dot\nu$. As in our previous study we

205: carried out a grid search by calculating $\chi^2$ at each $\nu_0$,

206: $\dot\nu$ pair (see Strohmayer 2002 for details).  The results are

207: summarized in Figures 5, 6 and 7.  Figure 5 shows two pairs of

208: $\Delta\chi^2$ confidence contours.  The large dashed contours

209: represent the constraints derived only from the ROSAT + ASCA data, and

210: the solid contours represent the best solution obtained with the

211: inclusion of the new {\it Chandra} data. The analysis definitively

212: favors a positive $\dot\nu = 7 \pm 0.8 \times 10^{-18}$ Hz s$^{-1}$.

213: The parameters of our best timing ephemeris are given in Table 1.

214: 

215: Although the presence of a long term spin up trend in the data is now

216: indisputable, the minimum $\chi^2$ value of 390 (with 98 degrees of

217: freedom) for our best fit is formally high.  Because of this we first

218: scaled the $\chi^2$ values for the whole dataset (including {\it

219: Chandra}) by the ratio $390/98 \approx 4$ before plotting the

220: confidence contours. This gives a conservative estimate of the

221: confidence region for $\nu$ and $\dot\nu$. Figure 6 shows two

222: representations of the phase measurements. The phase residuals

223: obtained with $\dot\nu = 0$ and our best constant frequency, $\nu_0$,

224: for all the data are shown in the top panel.  In this representation a

225: positive $\dot\nu$ (spin-up) will show up as a quadratic trend with

226: the parabola opening downward. One can clearly pick out such a trend

227: with the eye. The phase residuals computed with our best ephemeris

228: (ie. including the best non-zero value of $\dot\nu$) are shown in the

229: bottom panel. Figure 7 is essentially identical to Figure 6 except the

230: time gaps have been removed so that all the individual residuals can

231: be more easily seen. Note the much higher statistical quality of the

232: Chandra data.

233: 

234: The excess $\chi^2$ value is indicative of some additional ``timing

235: noise'' in the system. Although the rms deviations are qualitatively

236: similar for all the data, the much higher statistical quality of the

237: {\it Chandra} data contributes substantially to the excess. The

238: presence of timing noise could in principle result from fluctuations

239: in the spin up torque, or perhaps from modest changes in the profile

240: of the pulsed emission. By monitoring the character of such

241: fluctuations it should be possible to further constrain models for the

242: spin up torque. We discuss this further below. Since fluctuations of

243: both sign appear, and over a range of timescales, it seems extremely

244: unlikely that the overall spin up trend could somehow result from a

245: random superposition of fluctuations with just the right amplitude and

246: phasing.Using our best timing ephemeris we have calculated folded

247: pulse profiles for the ROSAT + ASCA and new {\it Chandra} data. The

248: results are shown in Figure 8. The {\it Chandra} profile has been

249: displaced vertically by 600 counts for clarity.  Phase zero in the

250: plot corresponds to MJD 49257.533373137 (TDB). The pulse profile

251: appears to be rather stable over the past decade.

252: 

253: \subsection{Flux Variability}

254: 

255: An important constraint on models for V407 Vul is the present lack of

256: any detectable periodicities with timescales longer than the 1.756 mHz

257: X-ray modulation.  If the system were in fact an IP, then one might

258: expect that the longer, unseen orbital period could eventually be

259: detected with sensitive observations. The {\it Chandra} data are

260: sensitive enough to explore variability on timescales longer than the

261: X-ray pulsation.  We did this by computing a lightcurve using our best

262: fit period as the time bin size.  The resulting lightcurve is shown in

263: Figure 9. In this plot we have suppressed the time gaps. Individual

264: observations are separated by the vertical dotted lines. Epochs for

265: each of the five observations can be found in Table 1. The data show

266: evidence for both long term variations in the flux as well as some

267: pulse to pulse variations.  For example, the flux during our

268: next-to-last pointing was a factor of 3 higher than in the initial

269: pointing.  The longest pointing (2nd panel from the left) shows

270: evidence for $30 - 50 \%$ variations in the flux within a few pulse

271: periods.

272: 

273: Our 35.4 ksec pointing at V407 Vul is the longest contiguous

274: observation of the source to date. This, combined with the higher

275: signal to noise ratio in the {\it Chandra} data compared with ROSAT or

276: ASCA, allows, in principle, more sensitive searches for a putative IP

277: orbital period.  To search for periods longward of the 1.756 mHz

278: frequency we computed Fourier power spectra of each of our

279: observations.  We do not find any robust periodicities which are

280: present in all of our observations.  In the power spectrum of the

281: longest pointing a weak sideband structure accompanies the fundamental

282: and first two harmonics of the 1.756 mHz frequency. The low frequency

283: region of this power spectrum is shown in Figure 10. The strongest of

284: these peaks are the upper sidebands to the fundamental and first

285: harmonic peaks at 1.756 mHz and 3.512 mHz. These sidebands are

286: separated in frequency from their parent peaks by 0.4825 mHz, and have

287: Leahy normalized powers of 74.8 and 52.5. There also appear to be

288: weaker lower sidebands accompanying the first and second harmonic

289: peaks at 3.512 mHz and 5.28 mHz respectively. These lower sidebands

290: are separated in frequency by 0.273 mHz from their respective

291: harmonic peaks.  The upper sideband frequency separation is close to,

292: but significantly different from, twice the lower sideband separation.

293: 

294: Although such a sideband structure could conceivably be intrinsic to

295: the source, and thus could suggest a longer, previously unseen period

296: in the system, we think a more plausible explanation is that these are

297: spurious signals most likely introduced by the dithering of the {\it

298: Chandra} pointing position. {\it Chandra} dithers in a Lissajous

299: pattern with different periods in the horizontal and vertical

300: directions. For ACIS, the dither periods are 1000 and 707 seconds in

301: the Y and Z directions, respectively. Indeed, there is a modest peak

302: at a 1000 s period in the power spectrum of our longest

303: observation. It seems likely that beating between the strong 1.756 mHz

304: modulation and the dither periods could introduce sideband

305: modulations. Moreover, since the strength of power spectral features

306: is proportional to the counting rate, we would expect an astrophysical

307: signal, if robust, to produce a greater signal in our brightest

308: observation, yet there is no evidence for the sideband structure in

309: the observation at the highest counting rate.  Based on these

310: arguments we do not think there is yet sufficient evidence to claim

311: detection of a longer period.  Because of the systematic issues

312: associated with dithering {\it Chandra}, it seems likely that XMM

313: observations will provide more definitive results with regard to

314: searches for an unseen orbital period in V407 Vul.

315: 

316: \section{Implications for the Nature of V407 Vul}

317: 

318: Our {\it Chandra} data definitively shows that the 1.756 mHz X-ray

319: modulation of V407 Vul is increasing at a mean rate of $\approx 7

320: \times 10^{-18}$ Hz s$^{-1}$. This rate is consistent with what would

321: be expected if the X-ray modulation is the orbital period of an

322: ultracompact system and it's orbit were decaying via emission of

323: gravitational radiation (see Strohmayer 2002).  Based on the observed

324: spin-up there would appear to be two remaining alternatives regarding

325: the nature of the source. 1) An ultracompact but detached system

326: powered by electrical energy, as described by Wu et al. (2002), or 2)

327: a low-inclination IP system in which the observed X-ray modulation is

328: effectively the white dwarf spin, and the X-ray flux and observed

329: spin-up are accretion driven (Norton, Haswell \& Wynn 2004). Other

330: ultracompact scenarios appear to be ruled out by the observed spin-up

331: and/or the short orbital period (see Strohmayer 2002; Cropper et

332: al. 2003).

333: 

334: In the IP scenario the white dwarf accretion is stream-fed and

335: switches from pole to pole and back again on the synodic period. An

336: accretion disk would form if the magnetospheric radius is less than

337: the circularization radius for the accretion stream. In order not to

338: form a disk and for V407 Vul to be stream-fed the orbital period must

339: satisfy $P_{orb} < 2.37$ hr (Norton et al. 2004).  If this scenario is

340: correct, then the secular spin up is due to accretion onto the white

341: dwarf, and other aspects of the observations, for example, the long

342: term flux variability and the phase timing noise, may be more easily

343: understood in terms of variations in the mass accretion rate. This

344: identification would also preclude the need to invoke models perceived

345: by some as ``exotic'' (such as the Wu et al. (2002) electric star

346: scenario).

347: 

348: However, some intriguing questions remain, for example, the soft X-ray

349: spectrum, the absence (or at least weakness) of optical emission

350: lines, and the infrared photometry have all been used to argue against

351: an IP identification (see Cropper et al. 2003 for a discussion).

352: Norton et al. (2004) have argued that the relatively high mass

353: accretion rates, of order $10^{-9} M_{\odot}$ yr$^{-1}$, required for

354: accretion to power the observed X-ray flux, as well as the nature of

355: stream-fed accretion may mitigate most of these perceived

356: difficulties.  For example, a highly optically thick stream may mask

357: the emission lines, and if the accretion is ``buried'' in the white

358: dwarf surface layers, then it may manifest itself with predominantly

359: soft blackbody emission. Deeper optical spectroscopy would likely

360: provide important new insight.

361: 

362: Accretion-induced spin-up should be testable with continued phase

363: coherent monitoring.  Comparison of torque fluctuations with X-ray

364: flux (ie. mass accretion rate) could yield a signature of magnetically

365: controlled accretion (see for example, Ghosh \& Lamb 1979). Moreover,

366: if the frequency derivative changes sign that would strongly support

367: the IP scenario.

368: 

369: \section{Summary}

370: 

371: Our new {\it Chandra} data confirm that the 1.756 mHz X-ray frequency

372: of RX V407 Vul is increasing at a rate of $\approx 7 \times 10^{-18}$

373: Hz s$^{-1}$.  Although this rate is consistent with that expected from

374: gravitational radiation losses in a detached ultracompact binary, an

375: IP interpretation in the context of accretion onto a spinning white

376: dwarf cannot yet be strongly ruled out.

377: 

378: In some ways the new {\it Chandra} results have only deepened the

379: mystery surrounding V407 Vul (and by implication, its sister source,

380: RX J0806.3+1527). After a decade of observations we still do not know

381: the nature of the source with any certainty.  However, a number of

382: future observations could help provide the solutions. Deep pointings

383: with XMM could probe more sensitively for an unseen orbital period,

384: and further timing observations with {\it Chandra} will establish

385: constraints on the dependence of torque fluctuations on X-ray

386: flux. Finally, if these do not suffice to crack the mystery, a

387: detection with a spaced-based gravitational radiation observatory

388: (such as the NASA/ESA LISA mission) would provide definitive evidence

389: for an ultracompact system.

390: 

391: \acknowledgements

392: 

393: We thank Richard Mushotzky, Zaven Arzoumanian, Jean Swank and Craig

394: Markwardt for many helpful comments and discussions. We also thank the

395: referee for his/her detailed comments on the original manuscript, that

396: helped us to significantly improve the paper. This work made use of

397: data obtained from the High Energy Astrophysics Science Archive

398: Research Center (HEASARC) at Goddard Space Flight Center.

399: 

400: \centerline{\bf References}

401: 

402: %\noindent{} Armstrong, J. W., Estabrook, F. B. \& Tinto, M. 2001, Class. and

403: %Quantum Gravity, 18, 4059

404: 

405: %\noindent{} Arzoumanian, Z., Fruchter, A. S. \& Taylor, J. H. 1994, ApJ, 

406: %426, L85

407: 

408: %\noindent{} Applegate, J. H. 1992, ApJ, 385, 621

409: 

410: %\noindent{} Applegate, J. H. \& Shaham, J. 1994, ApJ, 436, 312

411: 

412: %\noindent{} Barrett, P.,  O'Donoghue, D., \&  Warner, B. 1988, MNRAS, 233, 

413: 759 

414: 

415: \noindent{} Beuermann, K., Thomas, H. -C., Reinsch, K., Schwope,

416: A. D., Trumper, J. \& Voges, W. 1999, A\&A, 347, 47.

417: 

418: %\noindent{} Buccheri, R. et al. 1983, A\&A, 128, 245

419: 

420: \noindent{} Clarke, J. T. et al. 1996, Science, 274, 404

421: 

422: \noindent{} Cropper, M., Ramsay, G., Wu, K. \& Hakala, P. 2003, ASP

423: Conference Series to be published in Proc. Cape Town Workshop on

424: magnetic CVs, held Dec 2002, (astro-ph/0302240).

425: 

426: \noindent{} Cropper, M. et al. 1998, MNRAS, 293, L57

427: 

428: \noindent{} Ghosh, P. \& Lamb, F. K. 1979, ApJ, 232, 259.

429: 

430: \noindent{} Hakala, P. et al. 2003, MNRAS, 343, 10L.

431: 

432: %\noindent{} Evans, C. R., Iben, I. \& Smarr, L. 1987, ApJ, 323, 129

433: 

434: %\noindent{} Hall, D. S. 1991, ApJ, 380, L85

435: 

436: %\noindent{} Hils, D. \& Bender, P. L.\ 2000, ApJ,, 537, 334

437: 

438: %\noindent{} Hirayama, M. et al. 1996, ASCANews, \#4, ``Time Assignment of ASCA

439: %GIS.'' http://heasarc.gsfc.nasa.gov/docs/asca/newsletters/Contents4.html

440: 

441: %\noindent{} Hynes, R. I. et al. 2001, MNRAS, 324, 180.

442: 

443: %\noindent{} Iben, I. \& Tutukov, A. V. 1991, ApJ, 370, 615

444: 

445: \noindent{} Israel, G. L. et al. 1999, A\&A, 349, L1.

446: 

447: \noindent{} Israel, G. L. et al. 2002, A\&A, 386, L13.

448: 

449: \noindent{} Kuulkers, E., Norton, A., Schwope, A. \& Warner, B. 2003, in 

450: ``Compact Stellar X-ray Sources,'' ed. M. van der Klis \& W. H. G. Lewin,

451: (Cambridge Univ. Press: Cambridge, UK).

452: 

453: \noindent{} Marsh, T. R. \& Steeghs, D. 2002, MNRAS, 331, L7

454: 

455: \noindent{} Motch, C. et al. 1996, A\&A, 307, 459

456: 

457: %\noindent{} Nelemans, G., Portegies Zwart, S. F., Verbunt, F. \& Yungelson, 

458: %L. R. 2001a, A\&A, 368, 939

459: 

460: %\noindent{} Nelemans, G., Yungelson, L.~R., \& Portegies Zwart, S.~F.\ 2001b, 

461: %A\&A, 375, 890

462: 

463: \noindent{} Norton, A. J., Haswell, C. A. \& Wynn, G. A. 2004, A\&A,

464: in press, (astro-ph/0206013)

465: 

466: %\noindent{} Paczynski, B. 1967, Acta Astron., 17, 287

467: 

468: \noindent{} Ramsay, G., Cropper, M., Wu, K., Mason, K.~O., \& Hakala, P.\ 

469: 2000, MNRAS, 311, 75

470: 

471: \noindent{} Ramsay, G., et al. 2002, MNRAS in press, (astro-ph/0202281)

472: 

473: \noindent{} Ramsay, G. Hakala, P. \& Cropper, M. 2002, MNRAS, 332, L7.

474: 

475: %\noindent{} Rappaport, S., Joss, P. C. \& Webbink, R. F. 1982, ApJ, 254, 616

476: 

477: \noindent{} Ritter, H. \& Kolb, U. 2003,, A\&A, 404, 301.

478: 

479: %\noindent{} Saito, Y. et al. 1997, ASCANews, \#5, ``Accuracy of the GIS Time

480: %Assignment.'' 

481: %http://heasarc.gsfc.nasa.gov/docs/asca/newsletters/Contents5.html

482: 

483: %\noindent{} Savonije, G. J., de Kool, M. \& van den Heuvel, E. P. J. 1986, 

484: %A\&A, 155, 51

485: 

486: %\noindent{} Schutz, B. F. 1996, Class. and Quantum Gravity, 13, A219.

487: 

488: %\noindent{} Standish, E. M., Newhall, X. X., Williams, J. G. \& Yeomans, D. K

489: %            1992, in Explanatory Supplememt to the Astronomical Almanac, ed. 

490: %            P. K. Seidelmann (Mill Valley: University Science), 239

491: 

492: \noindent{} Strohmayer, T. E., 2003, ApJ, 593, 39L.

493: 

494: \noindent{} Strohmayer, T. E., 2002, ApJ, 581, 577.

495: 

496: %\noindent{} Strohmayer, T. E., \& Markwardt, C. B., 2002, ApJ, in press, 

497: %(astro-ph/0205435)

498: 

499: %\noindent{} Strohmayer, T. E., \& Markwardt, C. B., 1999, ApJ, 516, L81

500: 

501: %\noindent{} Taylor, J. H. \& Weisberg, J. M. 1989, ApJ, 345, 434

502: 

503: %\noindent{} Tutukov, A., \& Yungelson, L. 1996, MNRAS, 280, 1035

504: 

505: \noindent{} Warner, B. 1995, {\it Cataclysmic Variable Stars}, Cambridge

506: Univ. Press, Cambridge UK.

507: 

508: \noindent{} Warner, B. 1986, MNRAS, 219, 347.

509: 

510: \noindent{} Woudt, P. A. \& Warner, B. 2003, in Proc. Cape Town Workshop on 

511: Magnetic CVs, (astro-ph/0310494).

512: 

513: \noindent{} Wu, K., Cropper, M., Ramsay, G. \& Sekiguchi, K. 2002,

514: MNRAS, 331, 221

515: 

516: 

517: \clearpage

518: 

519: \begin{figure}

520: \begin{center}

521:  \includegraphics[width=6in, height=6in]{f1.ps}

522: \end{center}

523: Figure 1: {\it Chandra} 1-d image of V407 Vul from ACIS-S Continuous

524: Clocking mode data. The sharp peak contains photons from V407 Vul.

525: The inset panel shows an expanded view of the peak.

526: \end{figure}

527: \clearpage

528: 

529: \begin{figure}

530: \begin{center}

531: \includegraphics[width=6in, height=6in]{f2.ps}

532: \end{center}

533: Figure 2: A lightcurve of a portion of the {\it Chandra} ACIS-S data

534: from V407 Vul. The binsize is 16 s, and nine individuals pulses are

535: shown.

536: \end{figure}

537: \clearpage

538: 

539: \begin{figure}

540: \begin{center}

541:  \includegraphics[width=6in, height=5in]{f3.ps}

542: \end{center}

543: Figure 3: Best constant frequency measurements for V407 Vul at two

544: different epochs; the combined ROSAT + ASCA data (top), and the new

545: {\it Chandra} data (bottom). Shown are the values of $\Delta Z_3^2 =

546: {\rm max}(Z_3^2) - Z_3^2$ as a function of frequency.  In this

547: representation the minima denote the best frequency values for each

548: dataset.  Multiple minima are present due to the inclomplete sampling

549: of the datasets. In each case, however, the correct frequency peak can

550: be identified unambiguously (the peaks centered near $\nu - \nu_0 =

551: 0$). All other peaks are at significantly high values of $\Delta

552: Z_3^2$ to be strongly excluded (see the text for further details). The

553: reference frequency, $\nu_0$, is $1.7562475 \times 10^{-3}$ Hz.

554: \end{figure}

555: \clearpage

556: 

557: \begin{figure}

558: \begin{center}

559:  \includegraphics[width=6in,

560:  height=6in]{f4.ps}

561: \end{center}

562: Figure 4: Same as Figure 3 but now showing an expanded view of the

563: region around $\nu - \nu_0 = 0$. The best frequencies from the ROSAT +

564: ASCA (dashed) and Chandra (solid) epochs are shown. The frequency

565: measured at the more recent {\it Chandra} epoch is clearly higher.

566: This establishes beyond doubt that the frequency of V407 Vul has

567: increased over the last 11 years (see the text for further details).

568: \end{figure}

569: \clearpage

570: 

571: \begin{figure}

572: \begin{center}

573:  \includegraphics[width=6in,height=6in]{f5.ps}

574: \end{center}

575: Figure 5: Constraints on $\nu$ and $\dot\nu$ from our phase timing

576: analysis of the ROSAT, ASCA and new {\it Chandra} data for V407

577: Vul. Shown are the 90 and 99\% confidence regions for the ROSAT + ASCA

578: data (dashed), and the combined dataset (ROSAT, ASCA and {\it

579: Chandra}).  In plotting the confidence contours for all the data

580: (solid) we first scaled the $\chi^2$ values by the factor 390/98 (see

581: text for discussion). The inclusion of the {\it Chandra} data

582: definitively requires a positive $\dot\nu = 7 \pm 0.8 \times 10^{-18}$

583: Hz s$^{-1}$.

584: \end{figure}

585: \clearpage

586: 

587: \begin{figure}

588: \begin{center}

589:  \includegraphics[width=6in, height=6in]{f6.ps}

590: \end{center}

591: Figure 6: Phase timing residuals for V407 Vul plotted as a function of

592: time using a constant frequency phase model (ie. $\dot\nu \equiv 0$

593: (top), and with the best fitting solution including a positive $\dot

594: \nu = 7 \times 10^{-18}$ Hz s$^{-1}$ (bottom). In the top panel a

595: parabolic shape with the parabola opening downward is indicative of

596: the need for a positive $\dot\nu$. Such a trend can just about be

597: discerned with the eye. The {\it Chandra} data are those points beyond

598: year 9.

599: \end{figure}

600: \clearpage

601: 

602: \begin{figure}

603: \begin{center}

604:  \includegraphics[width=6in, height=6in]{f7.ps}

605: \end{center}

606: Figure 7: Same as Figure 6 except the time gaps have been removed so

607: that the individual phase residuals can be more easily seen. The rms

608: deviation of the residuals is also shown (dotted lines). The {\it

609: Chandra} data are those beyond measurment number 58 (note the smaller

610: error bars).

611: \end{figure}

612: \clearpage

613: 

614: \begin{figure}

615: \begin{center}

616:  \includegraphics[width=6in, height=6in]{f8.ps}

617: \end{center}

618: Figure 8: Folded pulse profiles for the ROSAT + ASCA (bottom) and new

619: {\it Chandra} data (top) using our best timing ephemeris (see Table

620: 1). The {\it Chandra} profile has been displaced vertically by 600

621: counts for clarity. Phase zero corresponds to MJD 49257.533373137

622: (TDB).

623: \end{figure}

624: \clearpage

625: 

626: \begin{figure}

627: \begin{center}

628:  \includegraphics[width=6in, height=6in]{f9.ps}

629: \end{center}

630: Figure 9: Flux variability of V407 Vul seen with {\it Chandra} at five

631: different epochs. The time binsize was set equal to one cycle of the

632: 1.756 mHz pulsation. There is clear evidence for large, long term flux

633: modulations as well as shorter timescale fluctuations.

634: \end{figure}

635: \clearpage

636: 

637: \begin{figure}

638: \begin{center}

639:  \includegraphics[width=6in, height=6in]{f10.ps}

640: \end{center}

641: Figure 10: Low frequency power spectrum of our longest {\it Chandra}

642: pointing. The power spectrum is normalized such that a pure poisson

643: noise process would have a mean of 2 and be distributed as $\chi^2$

644: with two degrees of freedom.  The three peaks that go off-scale are

645: the fundamental, first and second harmonic of the 1.756 mHz

646: modulation.  Upper sidebands to the fundamental and first harmonic,

647: and lower sidebands to the first and second harmonics are present. The

648: horizontal dotted line indicates the power level with a single trial

649: chance probability of $\approx 2 \times 10^{-9}$. We suspect the

650: sidebands are an artifact introduced by the {\it Chandra} dither (see

651: the text for a detailed discussion).

652: \end{figure}

653: \clearpage

654: 

655: 

656: \begin{table*}

657: \begin{center}{Table 1: {\it Chandra} Observations of V407 Vul}

658: \begin{tabular}{cccccc} \\

659: \tableline

660: \tableline

661:  &  OBSID     &  Instrument    &  Start UTC    &  Stop UTC  &  Exp (ksec) \\

662: \tableline

663: 1 & 300095  &  ACIS-S (CC-mode)  & Feb 18, 2003:11:16:30 & Feb 18, 

664: 2003:15:23:39 & 13.4 \\

665: 2 & 300095  &  ACIS-S (CC-mode)  & Feb 19, 2003:07:49:17 & Feb 19, 

666: 2003:17:58:49 & 35.4 \\

667: 3 & 300096  &  ACIS-S (CC-mode)  & Nov 11, 2003:19:33:05 & Nov 11, 

668: 2003:23:06:22 & 11.2 \\

669: 4 & 300111 &   ACIS-S (CC-mode)  & Dec 27, 2003:02:36:24 & Dec 27, 

670: 2003:07:30:14 & 15.4 \\

671: 5 & 300112 &   ACIS-S (CC-mode)  & Jan 31, 2004:21:20:12 & Feb 01, 

672: 2004:02:05:30 & 15.1 \\

673: \tableline

674: \end{tabular}

675: \end{center}

676: \end{table*}

677: \clearpage

678: 

679: \begin{table}

680: \begin{center}

681: {Table 2: Timing Solution for V407 Vul}

682: \end{center}

683: \begin{center}

684: \begin{tabular}{cc} \\

685: \tableline

686: \tableline

687:  Model Parameter &  Value \\

688: \tableline

689:  $t_0$ (TDB) & MJD 49257.533373137 \\

690: $\nu_0$ (Hz) & 0.0017562462(2)  \\

691: $\dot\nu$ (Hz s$^{-1}$) & $7.0(8) \times 10^{-18}$ \\ 

692: \tableline

693: \end{tabular}

694: \end{center}

695: \end{table}

696: \clearpage

697: 

698: 

699: 

700: %%%%%%%%%%%%%%%%%%%%%%%%%%%

701: %%%%% End of document %%%%%

702: %%%%%%%%%%%%%%%%%%%%%%%%%%%

703: 

704: \end{document}

705: 

706: 

707: 

708: \pagebreak

709: \centerline{\bf Figure Captions}

710: 

711: \figcaption[f1.ps]{{\it Chandra} 1-d image of V407 Vul from ACIS-S

712: Continuous Clocking mode data. The sharp peak contains photons from

713: V407 Vul.  The inset panel shows an expanded view of the peak.

714: \label{fig1}}

715: 

716: \figcaption[f2.ps]{A lightcurve of a portion of the {\it Chandra}

717: ACIS-S data from V407 Vul. The binsize is 16 s, and nine individuals

718: pulses are shown.

719: \label{fig2}}

720: 

721: \figcaption[f3.ps]{Best constant frequency

722: measurements for V407 Vul at two different epochs; the combined ROSAT

723: + ASCA data (top), and the new {\it Chandra} data (bottom). Shown are

724: the values of $\Delta Z_3^2 = {\rm max}(Z_3^2) - Z_3^2$ as a function

725: of frequency.  In this representation the minima denote the best

726: frequency values for each dataset.  Multiple minima are present due to

727: the inclomplete sampling of the datasets. In each case, however, the

728: correct frequency peak can be identified unambiguously (the peaks

729: centered near $\nu - \nu_0 = 0$). All other peaks are at significantly

730: high values of $\Delta Z_3^2$ to be strongly excluded (see the text

731: for further details). The reference frequency, $\nu_0$, is $1.7562475

732: \times 10^{-3}$ Hz.

733: \label{fig3}}

734: 

735: \figcaption[f4.ps]{Same as Figure 3

736: but now showing an expanded view of the region around $\nu - \nu_0 =

737: 0$. The best frequencies from the ROSAT + ASCA (dashed) and {\it

738: Chandra} (solid) epochs are shown. The frequency measured at the more

739: recent {\it Chandra} epoch is clearly higher.  This establishes beyond

740: doubt that the frequency of V407 Vul has increased over the last 11

741: years (see the text for further details).

742: \label{fig4}}

743: 

744: \figcaption[f5.ps]{Constraints on $\nu$ and $\dot\nu$

745: from our phase timing analysis of the ROSAT, ASCA and new {\it

746: Chandra} data for V407 Vul. Shown are the 90 and 99\% confidence

747: regions for the ROSAT + ASCA data (dashed), and the combined dataset

748: (ROSAT, ASCA and {\it Chandra}).  In plotting the confidence contours

749: for all the data (solid) we first scaled the $\chi^2$ values by the

750: factor 390/98 (see text for discussion). The inclusion of the {\it

751: Chandra} data definitively requires a positive $\dot\nu = 7 \pm 0.8

752: \times 10^{-18}$ Hz s$^{-1}$.

753: \label{fig5}}

754: 

755: \figcaption[f6.ps]{Phase timing residuals for

756: V407 Vul plotted as a function of time using a constant frequency

757: phase model (ie. $\dot\nu \equiv 0$ (top), and with the best fitting

758: solution including a positive $\dot \nu = 7 \times 10^{-18}$ Hz

759: s$^{-1}$ (bottom). In the top panel a parabolic shape with the

760: parabola opening downward is indicative of the need for a positive

761: $\dot\nu$. Such a trend can just about be discerned with the eye. The

762: {\it Chandra} data are those points beyond year 9.

763: \label{fig6}}

764: 

765: \figcaption[f7.ps]{Same as Figure 6 except the

766: time gaps have been removed so that the individual phase residuals can

767: be more easily seen. The rms deviation of the residuals is also shown

768: (dotted lines). The {\it Chandra} data are those beyond measurment

769: number 58 (note the smaller error bars).

770: \label{fig7}}

771: 

772: \figcaption[f8.ps]{Folded pulse profiles for the

773: ROSAT + ASCA (bottom) and new {\it Chandra} data (top) using our best

774: timing ephemeris (see Table 1). The {\it Chandra} profile has been

775: displaced vertically by 600 counts for clarity. Phase zero corresponds

776: to MJD 49257.533373137 (TDB).

777: \label{fig8}}

778: 

779: \figcaption[f9.ps]{Flux variability of V407 Vul seen

780: with {\it Chandra} at four different epochs. The time binsize was set

781: equal to one cycle of the 1.756 mHz pulsation. There is clear evidence

782: for large, long term flux modulations as well as shorter timescale

783: fluctuations.

784: \label{fig9}}

785: 

786: \figcaption[f10.ps]{Low frequency power spectrum of our longest {\it

787: Chandra} pointing. The power spectrum is normalized such that a pure

788: poisson noise process would have a mean of 2 and be distributed as

789: $\chi^2$ with two degrees of freedom.  The three peaks that go

790: off-scale are the fundamental, first and second harmonic of the 1.756

791: mHz modulation.  Upper sidebands to the fundamental and first

792: harmonic, and lower sidebands to the first and second harmonics are

793: present. The horizontal dotted line indicates the power level with a

794: single trial chance probability of $\approx 2 \times 10^{-9}$. We

795: suspect the sidebands are an artifact introduced by the {\it Chandra}

796: dither (see the text for a detailed discussion).

797: \label{fig10}}

798: 

799: \end{document}