1: \documentstyle[12pt,aasms4]{article}
2: %\shorttitle{OGLE-III Be/pulsars }
3: %\shortauthors{Schmidtke et al. }
4: \begin{document}
5:
6: \title{An Investigation of Be/X-ray Pulsars with OGLE-III Data }
7:
8: \author{ P.C.\ Schmidtke \& A.P.\ Cowley }
9:
10: \affil{Dept.\ of Physics \& Astronomy, Arizona State Univ., Tempe, AZ,
11: 85287-1504, USA; paul.schmidtke@asu.edu, anne.cowley@asu.edu }
12:
13: \and
14:
15: \author{A. Udalski}
16:
17: \affil{Warsaw University Observatory, Al.\ Ujazdowskie 4, 00-478 Warsaw,
18: Poland; udalski@astrouw.edu.pl}
19:
20: \begin{abstract}
21:
22: We have studied five seasons of OGLE-III data for eight SMC Be/X-ray
23: pulsars for which no other survey data were available. We have determined
24: orbital periods for four of these binary systems, one of which also shows
25: nonradial pulsations. Optical identification of SMC X-2 is reconsidered,
26: but no periods were found for either of the two possible candidates.
27:
28: \end{abstract}
29:
30: \keywords{X-rays: binaries -- stars: Be -- (stars:) pulsars -- stars:
31: variable -- stars: individual:
32: RX~J0052.9$-$7158,
33: 2E~0053.7$-$7227,
34: SMC X-2,
35: RX~J0057.8$-$7202,
36: CXOU~J005750.3$-$720756,
37: RX~J0059.2$-$7138,
38: XTE~J0111.2$-$7317,
39: RX~J0117.6-7330
40: }
41:
42: \section{Introduction}
43:
44: Be/X-ray binary systems account for about half of the high-mass X-ray
45: binaries, but relatively little is known about their properties. We have
46: been studying the photometric behavior of these systems in the Small
47: Magellanic Cloud using the OGLE-II and MACHO databases (Szymanski 2005,
48: Udalski et al.\ 1997, Zebrun et al.\ 2001, Alcock et al.\ 1999) in an
49: attempt to learn more about the component stars. Many of these systems
50: contain X-ray pulsars; lists of such systems are given by Coe et al.\
51: (2005), Liu et al.\ (2005), Haberl \& Pietsch (2004), Yokogawa et al.\
52: 2003, and others. Some of these binaries lie outside the fields of the
53: OGLE-II and MACHO surveys and hence had no available longterm photometry.
54:
55: The OGLE-III project covers a larger area around the SMC and includes some
56: Be/X-ray binaries for which longterm photometry had not previously been
57: obtained. We have undertaken a study of some of these systems, as listed
58: in Table 1. We give the OGLE-III positions of the optical counterparts
59: that are more accurate, in most cases, than previously available
60: positions. Table 1 also lists the catalogue number in the list of
61: H$\alpha$ emission-line stars in the SMC by Meyssonnier \& Azzopardi
62: (1993; denoted as [MA93]) and the SXP number (Coe et al. 2005) which
63: gives the X-ray pulse period for \underbar{S}mall Magellanic Cloud
64: \underbar{X}-ray \underbar{P}ulsars. Finding charts for all systems
65: studied here can be found on M.\ Coe's web
66: site\footnote{http://www.astro.soton.ac.uk/$\sim$mjc/smc}. However,
67: because of confusion about the optical identification of SMC X-2, we
68: present a new finding chart in this paper.
69:
70: \section{Analysis of Data from OGLE-III }
71:
72: The OGLE-III data cover five years with photometry in the $I$ bandpass.
73: Fig.\ 1 displays the longterm $I$ light curves for the sources studied
74: here. For all systems the data were prewhitened (flattened) using a
75: low-order polynomial fit to remove long term variations in brightness. In
76: some cases it was necessary to flatten the data in segments when there
77: were large, irregular brightness changes (e.g. SMC X-2). We use $I^*$ to
78: designate the flattened data. To analyze the $I^*$ data we computed
79: periodograms using the technique described by Horne \& Baliunas (1986).
80: When the variations are very non-sinusoidal we also used the phase
81: dispersion minimization (PDM) technique of Stellingwerf (1978). We
82: searched for photometric periods in the range of 0.25 days to 1000 days.
83: The techniques we used here are more fully described in our earlier papers
84: on Be/X-ray pulsar binaries (e.g. Schmidtke et al.\ 2004, Schmidtke \&
85: Cowley 2005, Schmidtke \& Cowley 2006).
86:
87: \section{Individual Be/X-ray Pulsar Systems Studied with OGLE-III Data}
88:
89: \subsection{ RX~J0052.9$-$7158 = SXP169 }
90:
91: This pulsar is identified with the Be star [MA93]623. Galache et al.\
92: (2005) note that SXP169 and SXP165 are the same pulsar which has been spun
93: up by $\sim$3 sec over 7.5 years.
94:
95: The OGLE-III light curve for SXP169 (Fig.\ 1) shows a small longterm
96: variation with considerable scatter about the mean. Analysis of all
97: seasons of photometry shows two clear periods. The first is the orbital
98: period of P=67.6$\pm$0.3 days, found both in the periodogram and in the
99: PDM variance (Fig.\ 2). The epoch of maximum is JD~2452583.3. The light
100: curve folded on this ephemeris shows an outburst of $\sim$0.02 mag with a
101: rapid rise and more gradual decline. Galache et al.\ (2005) derived a
102: period of 68.6$\pm$0.2 days from their weekly X-ray monitoring program
103: with $RXTE$. We note that both our period and outburst epoch are in
104: reasonable agreement with those of Galache et al.
105:
106: In addition to orbital outbursts, we also have found nonradial pulsations
107: (NRP) of the Be star with a period of P$_{NRP}$=0.755 days. These
108: pulsations are best detected in the out-of-outburst data (i.e. from
109: orbital phases $\sim$0.2-0.9. In Fig.\ 2 we plot the periodogram and
110: folded light curve for these NRP. There is no evidence that the pulsation
111: period changed through the five seasons of OGLE-III observations.
112:
113: \subsection{ SMC X-2 = RX~J0054.5$-$7340 = SXP2.37 }
114:
115: SMC X-2 is one of the earliest X-ray sources found in the SMC (Li,
116: Jernigan, \& Clark 1977), and it was soon noticed to be highly variable
117: (Li \& Clark 1977). Sanduleak \& Philip (1977) identified SMC X-2 with a
118: 14th magnitude OB star, and this was supported by spectroscopic
119: observations by Allen (1977) who found H emission lines. Subsequent
120: spectra taken by Crampton et al.\ (1978) showed the star to be a late Oe
121: star with a velocity appropriate to the SMC. However, Murdin, Morton, \&
122: Thomas (1979) later obtained an improved image of the field and resolved
123: the optical counterpart into a close double with $\sim$0.8 mag difference
124: in brightness. They assigned spectral types of O to the brighter
125: (northern) star and Be to the fainter (southern) star. Because the
126: components are very close together ($\sim2.5^{\prime\prime}$), it is
127: possible that the emission observed by both Allen and by Crampton et al.\
128: came from the fainter component which would have been blended with the
129: spectrum of the brighter star. However, Oe and Be stars have long been
130: known to lose their emission lines from time to time, so we cannot rule
131: out the brighter component as being the X-ray source. The X-ray position
132: is not sufficiently accurate to distinquish between these two close stars.
133: Because the only finding chart in the literature is a low resolution image
134: (Clark et al. 1978), we present the OGLE-III image in Fig.\ 3 which
135: clearly shows the two components. In addition, we give the OGLE-III
136: position of each component in Table 1.
137:
138: Fig.\ 1 shows the light curves, using comparable scales, for both the
139: northern (brighter) and southern (fainter) component. The northern star
140: has a swooping light curve with a range of $\sim$1 mag. This type of
141: variation is commonly seen among the Be/X-ray pulsar binaries. The
142: southern component is 1-2 magnitudes fainter and shows only a small light
143: variation. Because the two components are so close, it is likely that the
144: ``variability" of the southern star is due to contamination with the much
145: brighter northern component whose magnitude is extremely variable.
146:
147: For completeness, we have searched for periodic behavior in both
148: components. The brighter, northern component shows weak periodicities at
149: $\sim$90, 46, and 23 days. All are probably aliases of each other. The
150: folded light curves show very small sinusoidal variation. None is
151: convincing enough to claim as a true orbital period. No periods were
152: found in the analysis of the fainter, southern star. Based on the
153: character of the light curves, we conclude the northern star is more
154: likely to be the optical counterpart of SMC X-2, but an improved X-ray
155: position is needed to confirm this.
156:
157: \subsection{ 2E~0053.7$-$7227 = SXP34.1 }
158:
159: This source was found to be a 34-sec X-ray pulsar by Edge et al.\ (2004)
160: using $Chandra$ data (CXOU~J005527.9$-$721058). The excellent X-ray
161: position makes the optical identification secure. As can be seen in Fig.\
162: 1, the longterm light curve shows a relatively constant mean magnitude of
163: $I$=16.82 with a scatter of $\sim$0.06 mag. We searched the entire data
164: set and each of the individual seasons for possible photometric periods,
165: but found nothing convincing.
166:
167: \subsection{ RX~J0057.8$-$7202 = SXP280 }
168:
169: This source is identified with the Be star [MA93]1036. Its longterm light
170: curve (Fig.\ 1) clearly shows the star undergoes repeated outbursts of
171: $\sim$0.05 mag. The outburst (orbital) period is 127.3$\pm$1.0 days with
172: the epoch of maximum light at JD~2452195.2. The periodogram and folded
173: light curve are plotted in Fig.\ 4. We also searched for shorter periods
174: in the data, using both the entire data set and only the data from phases
175: between outbursts. No clear short periods were identified.
176:
177: \subsection{ CXOU~J005750.3$-$720756 = SXP152 }
178:
179: Macomb et al.\ (2003) found this source to be a 152-sec X-ray pulsar using
180: $Chandra$ data. The position is coincident with the Be star [MA93]1038.
181: Its OGLE-III longterm light curve (Fig.\ 1) shows a gradual rise from
182: $I$=15.38 to 15.32 over four years, with considerable scatter in the last
183: two seasons. Analysis of the flattened light curve shows no clear periods.
184: Some sections of the data have power near 70, 130, and 200 days, but these
185: may be artifacts due to the length of data train or spacing between
186: seasons. We were unable to find any consistent period which fit all of
187: the data.
188:
189: \subsection{ RX~J0059.2$-$7138 = SXP2.76 }
190:
191: The longterm light curve for this short-period pulsar is shown in Fig.\ 1.
192: Even before formal analysis one can detect repeated low-amplitude
193: outbursts. The data were flattened in two segments (first three seasons
194: and final two seasons). Both the individual segments and the full data
195: set give P=82.1$\pm$0.4 days, with the epoch of maximum at JD~2452189.4.
196: The outburst amplitude is $\sim$0.02 mag, with some variation from cycle
197: to cycle. The folded light curve (Fig.\ 5) shows a rapid rise and slower
198: decline. We have also searched for shorter periods using both the full
199: data set and data from out-of-outburst phases, but we found no significant
200: short periods that might be due to NRP.
201:
202: Based on Corbet's (1984) relation between pulse and orbital periods, this
203: source's orbital period is much longer than would be expected. We suggest
204: that the pulsar may have been spun up by accretion of material from the Be
205: star's disk.
206:
207:
208: \subsection{ XTE~J0111.2$-$7317 = SXP31.0 }
209:
210: This X-ray transient was identified by Coe et al.\ (2000) with a B0-2
211: emission-line star in the SMC. They give $I$=15.31 on 1999 January 21,
212: which is in general agreement with the range of magnitudes ($I$=15.1-15.3)
213: observed in the OGLE-III survey. Coe et al.\ found an irregular, small
214: emission nebula surrounding the star which they suggest might be a
215: supernova remnant. Covino et al.\ (2001) obtained spectra and found this
216: star to have hydrogen emission lines, confiming its identification as a Be
217: star.
218:
219: Analysis of the OGLE-III photometry reveals a pronounced orbital period at
220: P=90.4$\pm$0.5 days, with an epoch of maximum at JD~2452808.9. The
221: periodogram and PDM variance are plotted in Fig.\ 6, with both clearly
222: showing this orbital period. However, the PDM variance gives the better
223: value since the light curve is very non-sinusoidal. The folded light
224: curve has an extremely rapid climb to peak outburst and a more gradual
225: decay. We have searched the inter-outburst data for any shorter periods,
226: but found none.
227:
228: \subsection{ RX~J0117.6-7330 = SXP22.1 }
229:
230: Macomb et al.\ (1999) found this X-ray transient to be a 22-sec pulsar. It
231: is identified with the Be Star [MA93]1845. Soria (1999) studied the
232: optical counterpart before the source was know to contain a pulsar. He
233: gives the spectral type as B0.5IIIe with mean magnitudes of
234: $V$=14.9$\pm$0.01 and $I$=14.10$\pm$0.01. We show the longterm OGLE-III
235: light curve in Fig.\ 1. It is what we would call a ``moderate swooper",
236: with longterm variations from $I\sim13.9$ to 14.1 over several years.
237:
238: Analysis of the data from the first three seasons (the part which is most
239: easily flattened) reveals no clear periods that are not related to the
240: length of the data train or to seasonal gaps in the data. The last two
241: seasons were more difficult to prewhiten, but these also did not show any
242: significant periods. Finally, all of the flattened data were combined,
243: but again no photometric periods were found.
244:
245: \section{ Summary}
246:
247: In summary, we have studied the OGLE-III light curves of eight Be/X-ray
248: pulsars in the SMC. We have found orbital periods for four systems. The
249: folded light curves all show asymmetrical outbursts with the rise time
250: being shorter than the decline. One of the sources (SXP169) shows both
251: orbital outbursts and strong nonradial pulsations. For the remaining
252: stars, no clear periodic behavior was found in the range 0.25-1000 days.
253: The light curves of two stars near the position of SMC X-2 indicate that
254: the northern (brighter) object is the likely optical counterpart.
255:
256: In a recent paper on Be/X-ray binaries (Schmidtke \& Cowley 2006) we
257: presented some preliminary results about the photometric behavior of this
258: type of source. The same general trends persist in the present
259: investigation. We find that systems with strong orbital outbursts usually
260: do not exhibit nonradial pulsations of the Be star, and conversely the
261: sources with strong NRP rarely show orbital outbursts. SXP169 is the
262: single exception to this. We also find that systems with ``swooping''
263: light curves may show weak orbital outbursts, but don't have NRP.
264:
265: The strength of the orbital outbursts may be related to the eccentricity
266: of the orbit, if the pulsar penetrates the Be star's disk near periastron.
267: Perhaps the NRP sources are in fairly circular orbits with little
268: interaction between the stars, and hence orbital outbursts are not
269: present. To investigate these ideas, spectroscopic studies are needed.
270: Finally, we note that Corbet's (1984) relation between pulse period and
271: orbital period is generally supported here, but it shows considerable
272: scatter. Of note is SXP2.76 which shows an unexpectedly long orbital
273: period (P$_{orb}$=82.1 days) for its short X-ray pulse period, perhaps
274: indicating that the pulsar has been spun up.
275:
276: \acknowledgments
277:
278: Support for the OGLE project was provided by the following grants: Polish
279: MEiN grant 2P03D02124, NSF grant AST-0204908, and NASA grant NAG5-12212.
280:
281: \newpage
282:
283: \begin{references}
284:
285: \reference{} Alcock, C., et al. 1999, PASP, 111, 1539
286:
287: \reference{} Allen, D. 1977, IAU Circ.\ \#3143
288:
289: \reference{} Clark, G., Doxsey, R., Li, F., Jernigan, J.G., \& van
290: Paradijs, J. 1978, ApJ, 221, L37
291:
292: \reference{} Coe, M.J., Haigh, N.J., \& Reig, P. 2000, MNRAS, 314, 290
293:
294: \reference{} Coe, M.J., Edge, W.R.T., Galache, J.L., \& McBride, V.A.
295: 2005, MNRAS, 356, 502
296:
297: \reference{} Corbet, R.H.D. 1984, A\&A, 141, 91
298:
299: \reference{} Covino, S., et al. 2001, A\&A, 374, 1009
300:
301: \reference{} Crampton, D., Hutchings, J.B., \& Cowley, A.P. 1978, ApJ,
302: 223, L79
303:
304: \reference{} Edge, W.R.T., Coe, M.J., \& McBride, V.A. 2004, ATel, \#217
305:
306: \reference{} Galache, J.L., Corbet, R.H.D., Coe, M.J., Laycock, S.,
307: Markwardt, C.B., \& Marshall, F.E. 2005, ATel, \#674
308:
309: \reference{} Haberl, F., \& Pietsch, W. 2004, A\&A, 414, 667
310:
311: \reference{} Horne, J.H., \& Baliunas, S.L. 1986, ApJ, 302, 757
312:
313: \reference{} Li, F., \& Clark, G. 1977, IAU Circ.\ 3154
314:
315: \reference{} Li, F., Jernigan, G., \& Clark, G. 1977, IAU Circ.\ \#3126
316:
317: \reference{} Liu, Q.Z., van Paradijs, J., \& van den Heuvel, E.P.J. 2005,
318: A\&A, 442, 1135
319:
320: \reference{} Macomb, D.J., Finger, M.H., Harmon, B.A., Lamb, R.C., \&
321: Prince, T.A. 1999, ApJ, 518, L99
322:
323: \reference{} Macomb, D.J., Fox, D.W., Lamb, R.C., \& Prince, T.A. 2003,
324: ApJ, 584, L79
325:
326: \reference{} Meyssonnier, N., \& Azzopardi, M. 1993, A\&AS, 102, 451
327: ~ ([MA93])
328:
329: \reference{} Murdin, P., Morton, D.C., \& Thomas, R.M. 1979, MNRAS, 186, 42p
330:
331: \reference{} Sanduleak, N., \& Philip, A.G.D. 1977, IAU Circ.\ \#3127
332:
333: \reference{} Schmidtke, P.C., \& Cowley, A.P. 2005, AJ, 130, 2220
334:
335: \reference{} Schmidtke, P.C., \& Cowley, A.P. 2006, AJ, submitted
336:
337: \reference{} Schmidtke, P.C., Cowley, A.P., Levenson, L., \& Sweet, K.
338: 2004, AJ, 127, 3388
339:
340: \reference{} Soria, R. 1999, PASA, 16, 147
341:
342: \reference{} Stellingwerf, R.F. 1978, ApJ, 224, 953
343:
344: \reference{} Szymanski, M.K. 2005, Acta Astron., 55, 43
345:
346: \reference{} Udalski, A., Kubiak, M., \& Szymanski, M. 1997, Acta Astron.,
347: 47, 319
348:
349: \reference{} Yokogawa, J., Imanishi, K., Tsujimoto, M., Koyama, K., \&
350: Nishiuchi, M. 2003, PASJ, 55, 161
351:
352: \reference{} Zebrun, K., et al. 2001, Acta Astron., 51, 317
353:
354: \end{references}
355:
356: \newpage
357:
358: \begin{figure}
359: %fig 1
360: \caption{ Longterm OGLE-III light curves of all the Be/X-ray pulsar
361: systems studied in this paper. In the case of SMC X-2, for which the
362: identification is still uncertain, we have included the light curves for
363: both components. The northern star is a ``swooper", a photometric
364: behavior which is found in many sources of this type. }
365: \end{figure}
366:
367: \begin{figure}
368: %fig 2
369: \caption{ (top) PDM variance (upper; right scale) and periodogram (lower;
370: left scale) for RX~J0052.9$-$7158 (SXP169). (second) Light curve
371: folded on the orbital period of 67.6 days and binned in 20 phase bins.
372: (third) Periodogram of non-outburst data revealing the strong NRP at 0.755
373: days. (bottom) Light curve of non-outburst data folded on P=0.755 days. }
374: \end{figure}
375:
376: \begin{figure}
377: %fig 3
378: \caption{ Finding chart for SMC X-2 (SXP2.37) showing two stars which
379: are separated by $\sim2.5^{\prime\prime}$. North is at the top; east is
380: on the left. The figure is $\sim1.3^{\prime}$ on each side. The light
381: curves of both components are shown in Fig.\ 1. See discussion of the
382: optical identification in the text. }
383: \end{figure}
384:
385: \begin{figure}
386: %fig 4
387: \caption{ (top) Periodogram of RX~J0057.8$-$7202 (SXP280) showing the
388: strong period at 127.3 days. The outbursting behavior is clearly visible
389: even without formal analysis (see Fig.\ 1). (bottom) Light curve folded
390: on the orbital period of P=127.3 days.}
391: \end{figure}
392:
393: \begin{figure}
394: %fig 5
395: \caption{ (top) Periodogram of RX~J0059.2$-$7138 (SXP2.76) showing
396: power at P=82.1 days. (bottom) Light curve folded on the orbital period
397: of P=82.1 days.
398: }
399: \end{figure}
400:
401: \begin{figure}
402: %fig 6
403: \caption{ (top) PDM variance (upper; right scale) and periodogram (lower;
404: left scale) for XTE~J0111.2$-$7317 (SXP31.0). Both techniques show
405: power at P=90.4 days which appears to be the orbital period. (bottom)
406: Light curve folded on the orbital period, showing prominent outbursts. }
407: \end{figure}
408:
409: \newpage
410:
411: \begin{deluxetable}{llll}
412: \tablenum{1}
413: \tablecaption{Small Magellanic Cloud Be/Neutron-Star Systems Studied}
414: \tablehead{
415: \colhead{System Name} &
416: \colhead{SXP \#} &
417: \colhead{[MA93]\tablenotemark{a}} &
418: \colhead{2000 Position\tablenotemark{b}}
419: }
420: \startdata
421: RX~J0052.9$-$7158 & SXP169 & 623 & 00:52:55.28 ~ $-$71:58:06.0 \nl
422: SMC X-2 (North) & SXP2.37 & - - - & 00:54:33.47 ~ $-$73:41:01.1 \nl
423: SMC X-2 (South) & SXP2.37 & - - - & 00:54:33.37 ~ $-$73:41:03.5 \nl
424: 2E~0053.7$-$7227 & SXP34.1 & - - - & 00:55:28.46 ~ $-$72:10:56.5 \nl
425: RX~J0057.8$-$7202 & SXP280 & 1036 & 00:57:49.58 ~ $-$72:02:35.7 \nl
426: CXOU~J005750.3$-$720756 & SXP152 & 1038 & 00:57:50.38 ~ $-$72:07:55.9 \nl
427: RX~J0059.2$-$7138 & SXP2.76 & - - - & 00:59:12.74 ~ $-$71:38:44.8 \nl
428: XTE~J0111.2$-$7317 & SXP31.0 & - - - & 01:11:08.53 ~ $-$73:16:46.0 \nl
429: RX~J0117.6-7330 & SXP22.1 & 1845 & 01:17:40.18 ~ $-$73:30:50.5 \nl
430: \enddata
431:
432: \tablenotetext{a}{From catalogue of SMC H$\alpha$ emission-line stars
433: by Meyssonnier \& Azzopardi (1993). }
434:
435: \tablenotetext{b}{2000 positions are from OGLE-III data.}
436:
437: \end{deluxetable}
438:
439: \newpage
440:
441: \begin{deluxetable}{lccl}
442: \tablenum{2}
443: \tablecaption{Photometric Periods for
444: SMC Be/Neutron-Star Systems}
445: \tablehead{
446: \colhead{System Names} &
447: \colhead{P$_{orb}$} &
448: \colhead{NRP\tablenotemark{a}} &
449: \colhead{References} \\
450: \colhead{} &
451: \colhead{(days)} &
452: \colhead{(days)} &
453: \colhead{}
454: }
455: \startdata
456: RX~J0052.9$-$7158 (SXP169) & 67.6 & 0.755 &
457: Galache et al.\tablenotemark{b} \nl
458: RX~J0057.8$-$7202 (SXP280) & 127.3 & - - \nl
459: RX~J0059.2$-$7138 (SXP2.76) & 82.1 & - - \nl
460: XTE~J0111.2$-$7317 (SXP31.0) & 90.4 & - - & Coe et al.\tablenotemark{c} \nl
461:
462: \enddata
463:
464: \tablenotetext{a}{NRP = nonradial pulsations of the Be star. }
465:
466: \tablenotetext{b}{Galache et al.\ (2005) derived P$_{orb}$=68.6 days
467: from recurring X-ray outbursts. }
468:
469: \tablenotetext{c}{Coe et al.\ (2000) found a small emission nebula
470: surrounding this system. }
471:
472: \end{deluxetable}
473:
474: \end{document}
475:
476:
477: