1: \documentclass[12pt,preprint]{aastex}
2: %\documentclass{emulateapj}
3: \shortauthors{Raichur and Paul}
4: \shorttitle{QPOs in Cen X-3}
5:
6: \slugcomment{To be submitted to The Astrophysical Journal}
7: \received{}
8:
9: \begin{document}
10:
11: \title{Quasiperiodic oscillations in Cen X-3 and the long term
12: intensity variations}
13:
14: \author{Harsha Raichur\altaffilmark{1,2} and Biswajit Paul\altaffilmark{1}}
15:
16: \altaffiltext{1}{Raman Research Institute, Bangalore 560\,080, India}
17: \altaffiltext{2}{Joint Astronomy Programme, Indian Institute of Science, Bangalore 560\,080, India}
18:
19: \begin{abstract}
20:
21: We have investigated properties of the Quasi Periodic Oscillation (QPO)
22: features in the accretion powered X-ray pulsar Cen X-3 over a period of
23: about four years using observations carried out with the Proportional Counter
24: Array (PCA) of the {\it {Rossi X-ray Timing Explorer}}. The observations cover
25: a wide range of X-ray intensity of the source in excess of the binary
26: intensity modulation. We have detected QPOs in
27: 11 out of a total 81 pointings with the PCA
28: with rms intensity fluctuation upto 10\%. The QPO peak
29: frequency shows clustering around 40 and 90 mHz with the QPO
30: frequency having no dependence on X-ray intensity.
31: This indicates that either (a) the observed X-ray
32: luminosity of the source is not related to the mass accretion rate or
33: inner radius of the accretion disk or (b) that the QPO generation
34: mechanism in Cen X-3 is different from the beat frequency model or
35: Keplerian frequency model that is believed to be operational in most
36: other transient and persistent X-ray pulsars. We have also found that,
37: the rms variation in the 40 mHz QPO feature is not dependent on the
38: X-ray energy, indicating that disk absorption related origin for the
39: QPO is unlikely.
40:
41: \end{abstract}
42: \keywords{ Stars: neutron -- (Stars:) pulsars: individual: Cen X-3 --
43: X-rays: stars -- (Stars:) binaries: general -- X-rays: individual: Cen X-3 -- X-rays: binaries
44: }
45:
46:
47: \section{Introduction}
48:
49: The lightcurves of X-ray binary pulsars show periodic intensity variations
50: with the spin of the neutron star and its orbital motion. But a few of the
51: persistent X-ray binary pulsars also show a long term periodic intensity
52: variation with time scales more than an order of magnitude greater than the
53: orbital period of the binary.
54: Periodic superorbital intensity variations are seen in Her X-1 (35 day: Still
55: \& Boyd 2004), LMC X-4 (30.5 day: Paul \& Kitamoto 2002) and 2S 0114+650 (30.7
56: day: Farrell et al 2006). SMC X-1 shows quasi periodic superorbital intensity
57: variations with a 50-60 day cycle (Clarkson et al. 2003).
58: The intensity variations in
59: these systems are understood to be due to obscuration of the central X-ray
60: source by a warped precessing accretion disc. Spectral studies of Her X-1 and
61: LMC X-4 show iron line intensity and equivalent width evolving during the
62: superorbital periods. Also the absorption column density
63: in the line of sight is found to be higher during the low intensity states
64: indicating an excess of absorbing matter in the line of sight during these
65: times (Naik \& Paul 2003).
66:
67: Cen X-3 is a high mass X-ray binary pulsar with very strong but aperiodic
68: long term intensity variations (Figure 1). This is the first X-ray pulsar
69: discovered (Giaconni et al. 1971) and is also the brightest persistent pulsar.
70: It has a spin period of $\sim$4.8 s and an overall spin-up trend with alternate
71: spin-up and spin-down intervals which last from about 10 to 100 days
72: (Finger, Wilson \& Fishman 1994). It has an orbital period of 2.1 days and a
73: companion star
74: of about $20M_{\odot}$ (Avni \& Bahcall 1974). Though Cen X-3 is a persistent
75: pulsar, its binary period averaged X-ray intensity varies by a factor of more
76: than 40 (Paul, Raichur \& Mukherjee 2005).
77: As the long term intensity variation of
78: Cen X-3 does not remotely appear to have any periodic or quasi-periodic nature
79: (Paul et al. 2005), it is natural to assume that the X-ray flux variation
80: is due to changing mass accretion rate.
81: However, using a strong dependence of the orbital modulation and the pulsed
82: fraction of Cen X-3 on its X-ray intensity state we have shown that
83: the long term X-ray intensity variation in this source can be
84: due to change in obscuration by an aperiodically precessing warped accretion
85: disk (Raichur and Paul 2008). In this scenario, as the X-ray intensity
86: decreases, reprocessed and unpulsed X-rays from a relatively large scattering
87: medium progressively dominates the observed X-ray intensity.
88: We further investigate this hypothesis using the Quasi Periodic Oscillations
89: (QPO) in Cen X-3 with respect to its intensity state. In the accretion
90: powered X-ray pulsars, the QPOs are
91: understood to be due to inhomogenities in the inner accretion disk and
92: therefore its frequency is expected to be related to the inner radius of the
93: accretion disk. A correlation between the QPO frequency and X-ray luminosity
94: (and hence mass accretion rate / inner disk radius) has been observed in
95: several transient and persistent X-ray sources which show a large range of
96: X-ray intensity (EXO 2030+375: Angelini, Stella \& Parmar 1989, 3A 0535+262:
97: Finger, Wilson \& Harmon 1996, XTE J1858+034: Mukherjee et al. 2006, 4U
98: 1626--67: Kaur et al. 2008).
99:
100: QPOs are known to be present in all types of accreting X-ray pulsars.
101: In most sources it is a transient phenonema and QPOs have been detected
102: in about a dozen out of about a hundred known accreting X-ray pulsars.
103: With a few exceptions (4U 1748-288: Zhang et al. 1996 and XTE J 0111.2-7317:
104: Kaur et al. 2007) the QPO frequency is ususally in the range of 40-200 mHz,
105: consistent with it being related to the inner radius of the accretion disk
106: around a highly magnetised neutron star in its bright X-ray state.
107: Previous studies of Cen X-3 power spectrum have shown Quasi Periodic
108: oscillation (QPO) at $\sim 40$ mHz (Takeshima et al. 1991).
109:
110: In the present work, we have mainly studied the QPOs of Cen X-3 and its
111: relation to the source intensity if any. For this we have analysed all the
112: available archival data of the {\it{Rossi X-ray Timing Explorer (RXTE)}}
113: proportional courter array (PCA) from the year 1996 to 2000.
114:
115: \section{Observations and data analysis}
116:
117: Cen X-3 was observed extensively by {\it RXTE}-PCA during 1996-1998 and again
118: for some time in 2000. We have analysed X-ray lightcurves from all the data
119: available during this period. A total of 525 ks data was obtained from 81
120: pointings carried out in
121: this period. Very few of the observations were carried out during the eclipse
122: or ingres/egress of eclipse and data collected during these periods were
123: excluded from further analysis. Table 1 gives details of the observations.
124:
125: Light curves were extracted from all the observations using Standard-I data
126: which has a time resolution of 0.125 s. Power spectrum was obtained for
127: each of these observations using the Standard-I lightcurve from data
128: streches of duration 1024 s. Power spectra from all such stretches within
129: one observation pointing were averaged and normalised such
130: that their integral gives the squared rms fractional variability and the
131: expected white noise level was subtracted. We have detected QPO features
132: at different frequencies as described below. However, QPO features were
133: not present in all the data sets. Table 2 lists mid-times of the 11 segments
134: in which QPOs were detected along with the QPO frequencies. We have detected
135: the earlier reported QPOs around 40 mHz in all of the years although not in
136: every observation. The QPO at 90 mHz was seen only in the observations made
137: in 1996 and never again. The two QPOs 40 mHz and 90 mHz are not seen
138: together. In all the power spectra in which the QPOs were detected, the
139: peak associated with pulsar spin period was seen very clearly at $\sim 0.2$
140: Hz along with several harmonics. As reported earlier, pulsations were not
141: detected below a background subtracted count rate of 50 per proportional
142: counter unit with an upper limit of 0.8 \% on the pulsed fraction (Raichur
143: and Paul 2008). Some representative power spectra in different intensity
144: states, with and without the QPO features are shown in Figure 2 and
145: Figure 3.
146:
147: Another feature seen in most of the power spectra is a broadening of the
148: fundamental spin frequency peak. To investigate the reason for this broadening
149: we chose the
150: observation which had the highest RMS of the spin frequency broadening feature
151: (Obs. Id P20104, TJD of observation 10508.012 - 10508.713). The total length of
152: this observation is about 60608 s and it gives 51 intervals of
153: 1024 s each. A final power spectrum using 1024 s of data streches averaged over
154: all the intervals was made and then fitted with the model power spectrum given
155: by Lazzati and Stella (1997). The model power spectrum considers the fact that
156: any aperiodic variability in the emission from accretion column(s)
157: of a magnetic neutron star should be modulated at the X-ray pulsar period,
158: hence giving rise to a coupling between the periodic and aperiodic variability
159: (Burderi et al. 1997; Menna et al. 2003).
160: The coupling parameter R as defined in the model, is a measure of the degree
161: of coupling between the periodic and the aperiodic variabilities. The effects
162: due to a finite length of the lightcurve are built into the model. Fitting
163: this model (equation 5 of Lazzati and Stella 1997)
164: to our power spectrum gives $R \approx 0.64$, similar to the results
165: derived by Lazzati and Stella (1997) using EXOSAT data for Cen X-3. A value of
166: R $\approx$ 1.0 or greater indicates a strong coupling between the aperiodic
167: and periodic variabilities.
168:
169: In the top panel of Figure 4, a plot of the QPO frequency is shown against
170: the 2-30 keV X-ray flux measured with the PCA. To determine the X-ray fluxes
171: we have fitted the X-ray spectra with a simple model consisting of a high
172: energy cut-off power law along with line of sight absorption and iron emission
173: lines. Part of the X-ray light curves that showed pre-eclipse X-ray dips were
174: excluded from X-ray flux determination.
175: The PCA X-ray light curves from which the power spectra have been
176: generated represent the instantaneous measurement of X-ray flux over a
177: small period of a few ks. However, even outside the X-ray eclipse, within one
178: orbital period, the X-ray intensity of Cen X-3 varies smoothly by more than a
179: factor of two and rapidly by a factor of upto 4 during the pre-eclipse dips.
180: In a circular orbit, the orbital intensity variation is due to
181: different visibility of the X-ray source and its reprocessing region rather
182: than changing mass accretion rate. Therefore, we have also looked at the
183: QPO frequency against the orbital phase averaged X-ray intensity using data
184: from the All Sky Monitor (ASM) onboard {\it RXTE}. ASM has three detectors
185: which scan the sky
186: in a series of 90 s dwells in 3 energy bands, namely 1.5-3, 3-5 and 5-12 keV
187: (Levine, et. al.,1996). The combined ASM lightcurve with about 10-20 exposures
188: during each binary orbit of Cen X-3 gives a better estimate of the overall
189: X-ray intensity state of the source. We used the corresponding binary period
190: averaged count rates from ASM lightcurve to see any dependence of the
191: QPO frequency on the X-ray intensity.
192: The ASM count rates given in Table 2 are obtained from orbital period
193: averaged lightcurve after removal of data taken during the eclipse.
194: A plot of the QPO frequency with ASM count rate is shown in the bottom
195: panel of Figure 4. It is very clear from the figure that there is no
196: dependence of QPO frequency with the instantaneous or orbital phase averaged
197: X-ray intensity. The QPO features are clustered around two frequencies,
198: 40 mHz and 90 mHz. Even within each cluster, there is no dependence of
199: QPO frequency with the X-ray intensity.
200: In the top panel of Figure 4 we have also shown a plot of expected QPO
201: frequency as a function of the X-ray flux in the beat frequency model.
202: To calculate the X-ray flux expected by the beat frequency model we have taken
203: a source distance of 8 kpc (Krzeminski 1974), and a magnetic field strength of
204: $3.4 \times 10^{12}$ G (Coburn et. al. 2002).
205:
206: We have also carried out an energy resolved QPO analysis from one of the
207: observations (ID P10132) in which strong QPOs were detected at $\sim 40$ mHz
208: using energy resolved binned mode and event mode data. We extracted
209: lightcurves with time resolution of 0.125 s in energy bands of 2-4.1,
210: 4.1-6.6, 6.6-9.5, 9.5-13.1, 13.1-16.7, 16.7-20.4, 20.4-25.7 and
211: 25.7-34.8 keV. The energy resolved analysis of the 40 mHz QPO, as
212: shown in Figure 5 did not reveal
213: any measurable dependence of the rms fractional variability on energy.
214: Energy resolved analysis for the 90 mHz QPO could not
215: be done as the data from the corresponding observation did not have the
216: required energy and timing resolution in any of the data storage modes.
217:
218: \section{Results and discussions}
219:
220: We can summarize the principal results of our analysis presented in the
221: previous section as follows:
222:
223: 1. Cen X-3 shows intermittent QPOs in different frequency ranges namely 40 mHz, and 90 mHz.
224:
225: 2. The presence of the QPOs or the frequency of the QPOs are not
226: related to the luminosity state of the source. The RMS fluctuations associated
227: with the QPOs are not correlated with the luminosity of the source.
228:
229: 3. RMS fluctuation of the 40 mHz QPO is energy independent.
230:
231: 4. A weak coupling is measured between the low frequency aperiodic
232: variabilities and the spin frequency.
233:
234: In the discussion that follows we argue that the observed QPO properties
235: of Cen X-3 is in agreement with the scenario in which the long term X-ray
236: intensity variation is due to change in obscuration by an aperiodically
237: precessing warped accretion disk (Raichur and Paul 2008).
238:
239: The radius of the inner accretion disk around a magnetised neutron star with a
240: mass of 1.4 $M_{\odot}$ and a radius of 10 km can also be approximately
241: expressed in terms of its magnetic moment and X-ray luminosity as
242: (Frank, King and Raine 1992)
243: \begin{equation}
244: r_M = 3 \times 10^8 L_{37}^{-2/7}\mu_{30}^{4/7}
245: \end{equation}
246: where $L_{37}$
247: is the X-ray luminosity in the units of $10^{37}$ erg and $\mu_{30}$ is the
248: magnetic moment in units of $10^{30}$ G cm$^3 $. For disc accretion,
249: often a scaling factor of 0.5 is used with the above expression of $r_M$.
250: Then the radius of the inner accretion disk would be $R_M = 0.5 r_M$.
251: Coburn et. al. (2002) have
252: estimated the magnetic field of Cen X-3 neutron star to be
253: $B \simeq 3.4 \times 10^{12}$ G (i.e, $\mu_{30} = 3.4$) using the cyclotron
254: absorption line in the X-ray spectrum of the source.
255:
256: The lowest and the highest 3-30 keV X-ray flux at which the 40 mHz QPO feature
257: is seen are $1.1 \times 10^{-9}$ and $1.18 \times 10^{-8}$ erg
258: cm$^{-2}$s$^{-1}$ respectively. Assuming a distance of 8 kpc (Krzeminski 1974),
259: these correspond to X-ray luminosity of $L_{low} = 2.64 \times 10^{37}$ erg
260: s$^{-1}$ and $L_{high} = 2.83 \times 10^{38}$ erg s$^{-1}$. If the observed
261: X-ray luminosity represents the true X-ray luminosity and a proportional
262: mass accretion rate of Cen X-3, the inner accretion disk radius ($R_M$) will
263: approximately vary between $3 \times 10^8$ cm and $1.5 \times 10^8$ cm.
264: We note that the corotation radius of Cen X-3 ($P_{spin} \sim 4.8$ s) is
265: $4.7 \times 10^8 cm$, larger than the inner disk radius for the lowest X-ray
266: luminosity, and the QPO detections are outside a possible propeller regime.
267:
268: The two largely used QPO models are the Beat Frequency model (BFM) and the
269: Keplarian Frequency model (KFM). The BFM explains the QPO as the beat between
270: the spin frequency $\nu_{spin}$ and the keplarian frequency $\nu_k$ of the
271: inner accretion disk $\nu_{QPO} = \nu_{k}-\nu_{spin}$. Thus the radius of
272: the inner accretion disk according the BFM is given as follows.
273: \begin{equation}
274: r_{M,BFM}=\left(\frac{GM_{NS}}{4\pi^2(\nu_{spin}+\nu_{QPO})^2}\right)^{1/3}
275: \end{equation}
276: In KFM, the QPO occurs at the keplarian frequency of the inner accretion
277: disk $\nu_{QPO} = \nu_{k}$. Then radius of the inner accretion disk due to KFM
278: will be
279: \begin{equation}
280: r_{M,KFM}=\left(\frac{GM_{NS}}{4\pi^2\nu_{QPO}^2}\right)^{1/3}
281: \end{equation}
282:
283: However, in the
284: case of Cen X-3, as the $\nu_{spin}$ is larger than the observed QPO
285: frquencies, KFM is not applicable. This is because if the inner accretion disk
286: rotates slower than the neutron star, propeller effect is expected to inhibit
287: accretion of material from the accretion disk. Thus assuming an inner accretion
288: disk origin of the QPOs and using equations 1 and 2, one can experss a
289: relation between the QPO frequency and the X-ray luminosity. In the top
290: panel of Figure 4, we have shown the expected QPO frequency against the
291: measured X-ray for a source distance of 8 kpc. It is obvious from the figure
292: that the QPO frequency of Cen X-3 does not have the flux dependence as
293: expected in the beat frequency model.
294:
295:
296: From a study of the X-ray intensity dependence of the orbital modulation and
297: pulsed fraction in Cen X-3, recently we have proposed that the different flux
298: states of Cen X-3 are primarily due to varying degree of obscuration by an
299: aperiodically precessing warped accretion disk (Raichur and Paul 2008). The
300: nearly constant QPO frequency (ignoring the rare 90 mHz feature) reported here
301: is indeed consistent with this hypothesis. We propose that the mass accretion
302: rate and thus the inner accretion disk radius of Cen X-3 is not highly variable,
303: thus the source produces a nearly constant QPO frequency. We note here that
304: the frequencies predicted by the BFM are significantly larger than the measured
305: ones. However, the expression used here for magnetospheric radius is only approximate
306: and a different prescription for the magnetospheric radius (for example if it is
307: considerably larger than $r_M$ given in equation 1) can explain the observed QPO
308: frequencies at the highest observed X-ray flux state and also be consistent with
309: the proposal that the QPO frequency is insensitive to the measured X-ray flux as
310: the X-ray flux variation is primarily due to disk obscuration.
311:
312: We also note the other possibility that the QPOs in Cen X-3 may
313: not be due to any material inhomogenity in the inner accretion disk as is the
314: case for a few other X-ray binary pulsars like A0535+262 (Finger et al.
315: 1996), EXO 2030+375 (Angelini et al. 1989) XTE J1858+034 (Mukherjee et al.
316: 2006) and 4U 1626--67 (Kaur et al. 2008). In these transient binary X-ray
317: pulsars, the QPO frequency is well or
318: somewhat correlated with the X-ray luminosity of the source and hence the QPO
319: frequency variations are understood to be due to changes in the mass accretion
320: rate and associated changes in the radius of the inner accretion disk.
321:
322: One study which could
323: give us more insight would be to study the emission lines from neutral, H-like
324: and He-like iron atoms in the photoionised circumstellar material. Such
325: spectral observations done during the eclipse ingress and egress of the source
326: would give us knowledge of the distance at which the lines are produced.
327: Measurements carried out at different intensity states of the source will tell
328: us if the observed X-ray intensity is a true measure of the X-ray luminosity
329: and the ionization parameter.
330:
331: \acknowledgements
332: We thank an anonymous referee for many suggestions that helped us to improve
333: the paper.
334: This research has made use of data obtained from the High Energy Astrophysics
335: Science Archive Research Center (HEASARC), provided by NASA's Goddard Space
336: Flight Centre.
337:
338:
339: \begin{thebibliography}{}
340: \bibitem[]{348}
341: Angelini, L., Stella, L., \& Parmar, A.~N.,\ 1989, \apj, 346, 906
342: \bibitem[]{350}
343: Avni, Y., \& Bahcall, J.~N.\ 1974, \apjl, 192, L139.
344: \bibitem[]{337}
345: Burderi, L., Robba, N.~R., La Barbera, N., \& Guainazzi, M.\ 1997, \apj,
346: 481, 943
347: \bibitem[]{352}
348: Clarkson, W.~I., Chales, P.~A., Coe, M.~J., Laylock, S., Tout, M.~D., Wilson, C.~A.,\ 2003, \mnras, 339, 447.
349: \bibitem[]{353}
350: Coburn, W., Heindl, W.~A., Rothschild, R.~E., Gruber, D.~E., Kreykenbohm, Wilms, J., Kretschmar, P., \& Staubert, R.,\ 2002, \apj., 580, 394
351: \bibitem[]{344}
352: Farrell, S.~A., Sood, R.~K., \& O'Neill, P.~M.\ 2006, \mnras, 367, 1457
353: \bibitem[]{354}
354: Finger, M.~H., Wilson, R.~B., \& Fishman, G.~J.,\ 1994, in AIP Conf. Proc. 280, {\emph{Compton Gamma-Ray Observatory}}, ed. M. W. Friedlander \& N. Gehrels (New York: AIP), 304
355: \bibitem[]{356_01}
356: Finger, M.~H., Wilson, R.~B., \& Harmon, B.~A.,\ 1996, \apj, 459, 288
357: \bibitem[]{358_01}
358: Frank, J., King, A. \& Raine, D.,\ 1992, Accretion Power in Astrophysics, (2nd ed,; Cambridge: Cambridge University Press)
359: \bibitem[]{360_01}
360: Giacconi, R., Gursky, H., Kellogg, E., Schreier, E., \& Tananbaum, H.,\ 1971, \apj, 167, 67
361: \bibitem[]{365}
362: Kaur, R., Paul, B., Raichur, H. \& Sagar, R.,\ 2007, \apj, 660, 1409
363: \bibitem[]{356_02}
364: Kaur, R., Paul, B., Kumar, B., \& Sagar, R.,\ 2008, \apj, 676, 1184
365: \bibitem[]{358_02}
366: Krzeminski, W.,\ 1974, \apj, 192, 135
367: \bibitem[]{360_02}
368: Lazzati, D., \& Stella, L.,\ 1997, \apj, 476, 267
369: \bibitem[]{367}
370: Levine, A.~M., Bradt, Hale, Cui, Wei, Jernigan, J. G., Morgan, Edward H., Remillard, Ronald, Shirey, Robert E., \& Smith, Donald A.,\ 1996, \apj, 469, 33
371: \bibitem[]{364}
372: Menna, M. T., Burderi, L., Stella, L., Robba, N., \& van der Klis, M.,\ 2003, \apj, 589, 503
373: \bibitem[]{371}
374: Mukherjee, U., Bapna, S., Raichur, H., Paul, B., \& Jaaffrey,\ 2006, J. A. A., 27, 25
375: \bibitem[]{373}
376: Naik, S., \& Paul, B.,\ 2003, \aap, 401, 265
377: \bibitem[]{375}
378: Paul, B., \& Kitamoto, S.,\ 2002, J.A.A., 23, 33
379: \bibitem[]{377}
380: Paul, B., Raichur, H., \& Mukherjee, U.,\ 2005, \aap, 442, 15
381: \bibitem[]{374}
382: Raichur, H., \& Paul, B. 2008, \mnras (in press, arXiv:astro-ph/0804.1614)
383: \bibitem[]{379}
384: Still, M., \& Boyd, P.\ 2004, \apj, 606, 135.
385: \bibitem[]{381}
386: Tekeshima, T., Dotani, T., Mitsuda, K., \& Naga, F.,\ 1991, \pasj, 43
387: \bibitem[]{383}
388: Zhang, W., Morgan, E.~H., Jahoda, K., Swank, J.~H., Strohmayer, T.~E., Jernigan, G., Klein, R.~I.,\ 1996, \apj, 469, 29
389: \end{thebibliography}{}
390:
391: \clearpage
392:
393: {
394: \begin{table}
395: \caption{List of Observations}
396: ~\\
397: \begin{tabular}{|c|c|c|c|}
398: \hline
399: Year&Obs Ids&No. of Pointings&Total Durations(ks)\\
400: \hline
401: 1996&P10063&3&40\\
402: &P10132&5&18\\
403: &P10133&4&92\\
404: &P10134&7&143\\
405: &P10144&5&32\\
406: \hline
407: 1997&P20104&12&32\\
408: &P20105&3&26\\
409: &P20106&7&14\\
410: \hline
411: 1998&P30083&10&13.6\\
412: &P30084&23&65\\
413: \hline
414: 2000&P40072&2&50\\
415: \hline
416: \end{tabular}
417: \end{table}
418: }
419:
420:
421: \begin{table}
422: {
423: \caption{Details of QPOs detected in Cen X-3}
424: ~\\
425: \begin{tabular}{|c|c|c|c|c|}
426: \hline
427: Mid Time of&ASM Count&QPO Freq&QPO&RMS\\
428: Observation&Rate&(Hz)&width&\\
429: (MJD)&&&&\\
430: \hline
431: 50146.7283&$17.52\pm1.21$&$0.0903\pm0.0013$&$0.0091\pm0.0012$&$10.95\pm1.93$\\
432: 50147.3781&$17.52\pm1.21$&$0.0931\pm0.0015$&$0.0094\pm0.0016$&$ 6.21\pm1.52$\\
433: 50345.6888&$17.12\pm1.35$&$0.0452\pm0.0010$&$0.0068\pm0.0010$&$10.88\pm2.23$\\
434: 50509.5281&$13.60\pm0.92$&$0.0384\pm0.0008$&$0.0113\pm0.0009$&$5.11\pm0.46$\\
435: 50991.4068&$10.09\pm1.80$&$0.0454\pm0.0019$&$0.0125\pm0.0027$&$ 5.71\pm1.19$\\
436: 50998.7648&$17.73\pm1.47$&$0.0403\pm0.0016$&$0.1281\pm0.0026$&$ 6.19\pm1.03$\\
437: 50999.7818&$17.73\pm1.47$&$0.0419\pm0.0019$&$0.0048\pm0.0036$&$ 3.43\pm1.18$\\
438: 51094.9594&$ 1.83\pm1.21$&$0.0431\pm0.0009$&$0.0069\pm0.0010$&$ 5.93\pm1.16$\\
439: 51095.9637&$ 1.83\pm1.21$&$0.0445\pm0.0015$&$0.0064\pm0.0015$&$ 4.93\pm1.57$\\
440: 51578.9780&$ 8.97\pm1.19$&$0.0402\pm0.0016$&$0.0051\pm0.0013$&$ 3.75\pm1.01$\\
441: 51579.9307&$ 7.06\pm1.06$&$0.0502\pm0.0022$&$0.0101\pm0.0018$&$ 6.49\pm1.29$\\
442: \hline
443: \end{tabular}
444: }
445: \end{table}
446:
447: \clearpage
448:
449:
450: \begin{figure}
451: \centering
452: \includegraphics[height=5in, angle=-90]{f1.eps}
453: \caption{The 1.5-12.0 keV band {\it RXTE}-ASM lightcurve of Cen X-3 is shown
454: here with a binsize same as the orbital period.}
455: \end{figure}
456:
457:
458:
459: \begin{figure}
460: \centering
461: \includegraphics[height=5in, angle=-90]{f2.eps}
462: \caption{Representative power spectra of en X-3 with QPOs are shwon here.
463: Top three pannels show the 40 mHz QPO at different intensity states of Cen X-3.
464: The bottom pannel shows the 90 mHz QPO seen only during the high intensity
465: state of Cen X-3 in 1996.}
466: \end{figure}
467:
468: \begin{figure}
469: \centering
470: \includegraphics[height=5in, angle=-90]{f3.eps}
471: \caption{Representative power spectra of Cen X-3 at different intensity state
472: without QPOs are shown here. The first and second plots are multiplied by
473: constant numbers for clarity.}
474: \end{figure}
475:
476: \begin{figure}
477: \centering
478: \includegraphics[height=5in, angle=-90]{f4a.eps}
479: \vskip 1cm
480: \includegraphics[height=5in, angle=-90]{f4b.eps}
481: \caption{Upper pannel plots (points marked with circles) the observed QPO
482: frequency against the instantaneous flux as measured using the PCA spectrum.
483: The dashed line shows the relation of QPO frequency and flux of the source as
484: expected due to the BFM when $r_M$ is calculated using the scaling factor of
485: 0.5 and solid line shows the same relation when $r_M$ is calculated without
486: using the scaling factor. Lower panel plots the QPO frequency against the orbit
487: averaged ASM count rate of Cen X-3 at the time of observation.}
488: \end{figure}
489:
490: \begin{figure}
491: \centering
492: \includegraphics[height=5in, angle=-90]{f5.eps}
493: \caption{RMS fluctuation in the 40 mHz QPO feature is shown here as a function
494: of energy for observation ID P10132. The average RMS value is shown with
495: the dashed line.}
496: \end{figure}
497:
498: \end{document}
499:
500: