1: \documentclass[12pt,preprint]{aastex}
2: \usepackage{emulateapj5}
3:
4: \shorttitle{SIGMA Observations of GRO J1744-28}
5: \shortauthors{Mej\'{\i}a et al.}
6:
7: \input psfig.sty
8:
9: \begin{document}
10:
11: \title{SIGMA Observations of the Bursting Pulsar GRO J1744-28}
12: \author{J. Mej\'{\i}a and T. Villela}
13: \affil{Instituto Nacional de Pesquisas Espaciais, Divis\~ao de Astrof\'{\i}sica,
14: C.P. 515, 12201-970, S\~ao Jos\'e dos Campos, SP, Brazil}
15: \email{mejia@das.inpe.br}
16:
17: \author{P. Goldoni and F. Lebrun}
18: \affil{CEA/DSM/DAPNIA/SAp, CEA-Saclay, F-91191 Gif-sur-Yvette Cedex, France}
19:
20: \author{L. Bouchet, E. Jourdain, J.-P. Roques and P. Mandrou}
21: \affil{Centre d'Etude Spatiale des Rayonnements, 9 Avenue du Colonel Roche, B.P. 4346, 31029
22: Toulouse Cedex, France}
23:
24: \and
25:
26: \author{S. Kuznetsov, N. Khavenson, A. Dyachkov, I. Chulkov, B. Novikov, K. Shuhanov,
27: I. Tserenin and A. Sheikhet}
28: \affil{Space Research Institute, Profsoyuznaya 84/32, Moscow 117810, Russia}
29:
30:
31: \begin{abstract}
32: We present the results of the GRANAT/SIGMA hard X-/soft $\gamma$-ray
33: long-term monitoring of the Galactic Center (GC) region concerning the source
34: GRO J1744-28, discovered on 1995 Dec. 2 by CGRO/BATSE. SIGMA observed
35: the region containing the source in 14 opportunities
36: between 1990 and 1997. In two of these observing sessions, corresponding
37: to March 1996 and March 1997, GRO J1744-28 was
38: detected with a confidence level greater than 5$\sigma$ in the 35-75 keV
39: energy band without detection in the 75-150 keV energy band. For the other
40: sessions, upper limits of the flux are indicated. The particular
41: imaging capabilities of the SIGMA telescope allow us to identify,
42: specifically, the source position in the very crowded GC region, giving
43: us a mean flux of (73.1$\pm$5.5)$\times$10$^{-11}$ and (44.7$\pm$6.4)$\times$10$^{-11}$
44: ergs cm$^{-2}$ s$^{-1}$ in the 35-75 keV energy band, for
45: the March 1996 and March 1997 observing sessions, respectively.
46: Combining the March 1997 SIGMA and BATSE observations, we found evidence pointing to
47: the type-II nature of the source bursts for this period.
48: For the same observing campaigns, spectra
49: were obtained in the 35 to 150 keV energy band. The best fit
50: corresponds to an optically thin thermal Bremsstrahlung with
51: F$_{50 keV}$=3.6($\pm$0.6)$\times$10$^{-4}$ phot cm$^{-2}$ s$^{-1}$ keV$^{-1}$
52: and kT$_{Bremss}$=28$\pm$7 keV, for the first campaign, and
53: F$_{50 keV}$=2.3($\pm$0.7)$\times$10$^{-4}$ phot cm$^{-2}$
54: s$^{-1}$ keV$^{-1}$ and kT$_{Bremss}$=18$^{+12}_{-7}$ keV, for the second.
55: This kind of soft spectrum is typical of binary sources containing a neutron
56: star as the compact object, in contrast to the harder spectra typical of systems
57: containing a black hole candidate.
58:
59: \end{abstract}
60:
61: \keywords{pulsars:individual:GRO J1744-28 - X-rays:bursts }
62:
63: \section{INTRODUCTION}
64:
65: GRO J1744-28, the bursting pulsar, was discovered on 1995 December 2 by
66: CGRO/ BATSE \citep{fis95,kou96}. Timing studies revealed a period of
67: 0.467 s in the persistent flux and an 11.8-day orbital period \citep{fin96}.
68: This source presented two strong outbursts:
69: at the end of 1995 and early 1996, and then almost a year later. At that time,
70: this was the only source known to show, simultaneously, periodic
71: X-ray pulsations and frequent X-ray bursts. The presence of pulsations
72: indicates that the compact object is a neutron star. The X-ray mass
73: function, determined by \citet{fin96} as being f(M$_x$)=1.31$\times$10$^{-4}$
74: M$_{\odot}$, together with a mass canonical value for the compact object of
75: 1.4 M$_{\odot}$, suggest that the neutron star in GRO J1744-28 is
76: being fed by Roche lobe overflow from a low-mass red giant \citep{dau96,mil96,stu96,bil97}.
77:
78: The best position coordinates for the source are those obtained from
79: triangulation of the Ulysses and BATSE observations of the region. With
80: this method, \citet{hur00} determined the coordinates RA=17$^h$44$^m$32$^s$,
81: decl.=-28$^{\circ}$44$^{'}$31$^{''}.$7 (J2000.0). A possible infrared counterpart was
82: reported by \citet{aug96} while a posterior optical/IR
83: observation of the same region by \citet{col97} neither
84: confirmed nor discarded this IR counterpart. Based on the ASCA data, the
85: absorption column, N$_H$, for the position
86: of GRO J1744-28 was determined as being (5-6)$\times$10$^{22}$ cm$^{-2}$,
87: independent of the emission date and phase and, therefore, corresponding to
88: interstellar absorption \citep{dot96,nis99}. This suggests
89: that the source is located near the Galactic Center (GC), at a distance of $\sim$8.0 kpc.
90: The source was also observed with GRANAT/WATCH from January 13 to March 12, 1996
91: \citep{saz97} showing, at the maximum between bursts, a daily average flux
92: of $\sim$3.7 Crab, in the 8-20 keV energy band.
93:
94: Between 1990 and 1997, the SIGMA telescope was pointed in the direction
95: of the GC region two times per year, in the March-April and
96: September-October periods. In 14 opportunities, the region containing
97: GRO J1744-28 was observed, for a total of $\sim$2993 hours.
98: In two of these sessions, corresponding to March 1996 and March 1997,
99: the source was clearly visible, with a confidence level higher than 5$\sigma$. In
100: this paper, the analysis of these observations is presented. Also,
101: upper limits are reported for the remaining observing sessions.
102: Finally, for the March 1997 observing session, a combination of the
103: SIGMA value for the integrated flux of the source with the BATSE value
104: for the burst mean flux allows us to obtain evidence favouring
105: the type-II nature of the bursts, due to spasmodic accretion rather than
106: thermonuclear burning of matter.
107:
108: \section{THE SIGMA TELESCOPE}
109:
110: The French coded-mask telescope SIGMA \citep{pau91} onboard the
111: Russian GRANAT orbital observatory provides high resolution images in the hard
112: X-/soft $\gamma$-ray band from 35 keV to 1300 keV, divided in 95 energy
113: channels. The nominal angular resolution of the instrument is $\sim$15
114: arcmin. The position determination accuracy of the instrument is 3-5
115: arcmin for a 6$\sigma$ source and $<$1 arcmin for a 30$\sigma$ one. The fully
116: coded field of view of the instrument is a 4$^{\circ}$.7 x 4$^{\circ}$.3
117: rectangle while the half sensitivity boundary of the field of view corresponds
118: to an 11$^{\circ}$.5 x 10$^{\circ}$.9 rectangle. The energy resolution of
119: the instrument at 511 keV is $\sim$8$\%$. The typical duration of individual
120: observing sessions is $\sim$20 hours, that provides a 1$\sigma$
121: sensitivity of $\sim$20 mCrab in the 35-150 keV energy domain. For details on
122: the SIGMA in-orbit performances, see \citet{bou01}.
123:
124: \section{THE OBSERVATIONS}
125:
126: The region containing the source GRO J1744-28 was
127: observed by SIGMA in 14 different opportunities, from 1990 to 1997. In two of these
128: observing campaigns, the source was clearly detected in the 35-75 keV energy
129: band, with a confidence level higher than 5$\sigma$, as shown in Figure 1.
130: In contrast, the source was not detected in the 75-150 keV energy band.
131: The first campaign began on 1996, March 15 (MJD 50157.63), 103 days
132: after the discovery of the source by BATSE, and lasted until March 30,
133: corresponding to $\sim$267 hours of effective observing time (see Table 1).
134: In this period, GRO J1744-28 appears as a 13.5$\sigma$ source
135: with a mean detected flux in the 35-75 keV energy band of 85$\pm$6.4 mCrab (7.3($\pm$0.6)$\times$10$^{-10}$
136: ergs cm$^{-2}$ s$^{-1}$),
137: showing a soft spectrum that can be approximated by an optically thin
138: thermal Bremsstrahlung model ($\alpha$=-1.4) with a flux at 50 keV of F$_{50 keV}$=
139: 3.6($\pm$0.6)$\times$10$^{-4}$
140: phot cm$^{-2}$ s$^{-1}$ keV$^{-1}$ and kT$_{Bremss}$=28$\pm$7 keV
141: ($\chi^2$(d.o.f.)=13.0(8)).
142:
143: The second campaign began on 1997, March 14 (MJD 50521.31) and last
144: until March 28, corresponding to an effective observational time of $\sim$148.5
145: hours. In this opportunity, the mean flux of the source in the 35-75 keV
146: energy band was of 52$\pm$7.4 mCrab (4.5($\pm$0.6)$\times$10$^{-10}$ ergs cm$^{-2}$ s$^{-1}$),
147: corresponding to a confidence level of 8$\sigma$. The source showed a very steep
148: spectrum fitting with an optically thin thermal Bremsstrahlung
149: model with a F$_{50 keV}$=2.3($\pm$0.7)$\times$10$^{-4}$ phot cm$^{-2}$ s$^{-1}$
150: keV$^{-1}$ and kT$_{Bremss}$=18$^{+12}_{-7}$ keV ($\chi^2$(d.o.f.) = 8.04(8)).
151: In Table 1, the values of the fluxes
152: corresponding to each one of the sessions of these two campaigns in the 35-75 keV
153: and 75-150 keV energy bands are indicated.
154:
155: \section{THE LIGHT CURVE AND SPECTRA OF GRO J1744-28}
156:
157: In Figure 2, the light curves of the source GRO J1744-28 for the
158: campaigns corresponding to March 1996 and March 1997 are presented. In the
159: first campaign, the source flux in the 35-75 keV energy band was almost
160: constant with an average flux of $\sim$85 mCrab or 7.3$\times$10$^{-10}$
161: ergs cm$^{-2}$ s$^{-1}$. The second campaign does not show significant flux variations;
162: the average flux being $\sim$52 mCrab or 4.5$\times$10$^{-10}$
163: ergs cm$^{-2}$ s$^{-1}$ (see also \citet{tru99}. This last value
164: is a factor 7 lower than that reported by \citet{woo99} of
165: $\sim$3$\times$10$^{-9}$ ergs cm$^{-2}$ s$^{-1}$, as taken from the
166: BATSE observations of the source using the Earth-occultation technique
167: in the 30 to 100 keV energy band.
168: %In this figure, they indicate a value for the persistent flux of
169: %F(persistent)=3x10^-9 ergs cm-2 s-1, for the 10700<TJD<10850.
170:
171: Since GRO J1744-28 is located near the GC, it was within the SIGMA
172: field of view during all the observations constituting the GC survey
173: conducted by SIGMA between 1990 and 1997, which allows us to estimate upper
174: limits for the hard X-ray flux from the source position before and after
175: the 1996 and 1997 outbursts. An analysis of these SIGMA observations
176: was done in order to verify the possibility of source appearances in
177: other epochs than that corresponding to the two outbursts on 1996
178: and 1997. In Figure 3, the light curve of the source is shown,
179: corresponding to all the pointed SIGMA observations of the GC region containing
180: the source, in the 35-75 keV energy band. The resulting 3$\sigma$
181: upper limit for the average source flux during each campaign is $\sim$17 mCrab
182: (1.5$\times$10$^{-10}$ ergs cm$^{-2}$ s$^{-1}$) varying between 9 and 24 mCrab,
183: with an integrated upper limit of $\sim$5 mCrab (4.3$\times$10$^{-11}$ ergs cm$^{-2}$ s$^{-1}$).
184:
185: As said previously, for the two observing sessions in which GRO J1744-28
186: was detected, this source presented a very steep spectrum, reasonably well fitted
187: with an optically thin thermal Bremsstrahlung model, as given by the expression
188:
189:
190: $$\frac{dN}{dE} \propto E^{-1.4} \; exp \left( \frac{-E}{kT} \right).$$
191:
192: In order to have a better coverage of
193: the spectral behaviour of the source, we also analyzed the RXTE archival data obtained
194: on 1996 March 22 and 1997 March 18. The source was very bright during RXTE/PCA observations,
195: therefore any possible contamination by the other sources in the PCA's field of view is
196: negligible. The PCA data reduction was performed with the help of standard tools from
197: LHEASOFT/FTOOLS 5.0.4 package. Both RXTE/PCA and GRANAT/SIGMA spectra were approximated by
198: a model which is typical for pulsars: power law with high-energy cutoff, Gaussian line and
199: photoabsorption. The pure statistical significance of SIGMA data is much less than that of
200: RXTE/PCA. Because of this, the SIGMA spectral points almost do not influence on the parameters
201: of the combined PCA+SIGMA spectral approximation. In order to make the spectral fit more
202: sensitive to the SIGMA data, we artificially increased the uncertainties of the PCA spectral
203: points to the value 5\%. The best fit parameters of the combined PCA+SIGMA spectral fits are
204: (for 1996/1997): photon index of the power law $\alpha$=1.44$\pm$0.12/1.41$\pm$0.07;
205: E$_{Hcutoff}$=14.7$\pm$0.7/16.4$\pm$0.9. In Figure 4 we plot the 1 to 100 keV PCA/SIGMA
206: spectrum corresponding to the March 1996 and March 1997 SIGMA observing periods, along
207: with the SIGMA spectral fit.
208:
209: Our result for the second outburst is compatible with the analysis of the same
210: event of the source made on the BATSE data and presented by \citet{woo99}.
211: For the epoch of the SIGMA observation, they report a temperature
212: %kT $\sim$10.5 keV for the burst emission and
213: kT $\sim$14.5$\pm$1.8 keV for the persistent emission (see Figure 4, therein).
214: However, as it will be shown later, there exists the possibility of a contamination
215: of the BATSE data because of the presence of a transient source in the neighbourhood of GRO J1744-28,
216: namely GRS 1737-31.
217:
218: \section{CONCLUSION}
219:
220: In the period going from 1990 to 1997, observations of the GC region
221: containing the coordinates of GRO J1744-28 were made with the SIGMA
222: telescope in 14 opportunities, in March-April and September-October each
223: year. In two of these opportunities, corresponding to March 1996 and March
224: 1997, GRO J1744-28 was clearly detected
225: well above the 5$\sigma$ flux limit, corresponding to the two outbursts
226: detected by several other instruments.
227:
228: The mean fluxes for these two observations, as seen by SIGMA in the
229: 35-75 keV energy band, were 85 and 52 mCrab. In the 75-150 keV energy
230: band, the source was not detected. The spectra were reasonably well
231: fitted by optically thin thermal Bremsstrahlung with kT$_{Bremss}$=28 and 18
232: keV respectively. These values are compatible with that presented by \citet{woo99}
233: for the second burst. However, the flux seen by SIGMA during this second
234: observation campaign is almost a factor 7 smaller than that based on the BATSE
235: observations of the region using the
236: Earth-occultation technique. In this epoch, the transient source GRS
237: 1737-31 was discovered with SIGMA \citep{sun97} and observed by
238: RXTE and BeppoSAX/WFC \citep{mar97,cui97,hei97}.
239: This source was active for a period of at least 1 month beginning
240: on middle February, 1997, in the 35-150 keV energy band, as reported by
241: \citet{tru99}. We argue that the presence of this source may
242: explain, at least partially, the different flux values obtained with BATSE
243: and SIGMA for the bursting pulsar. For GRS 1737-31, \citet{tru99} have
244: found a flux of $\sim$100 mCrab (1.6$\times$10$^{-9}$ ergs cm$^{-2}$
245: s$^{-1}$) in the 35-150 keV which, added to the GRO J1744-28 flux (4.5$\times$10$^{-10}$
246: ergs cm$^{-2}$ s$^{-1}$), accounts for a representative part of
247: the flux reported from BATSE. It is important to note that the use of
248: coded mask imaging techniques, like that used by SIGMA and other
249: instruments of this kind, allows us to obtain in an unambiguous way the
250: integrated flux and spectrum of the sources even in very crowded fields as in
251: the case of the GC region.
252:
253: With regard to the nature of the bursts, several authors have proposed
254: that they are type-II bursts (see, for instance, Lewin, Rutledge and Kommers 1996 for
255: the first outburst). In their Figure 6, \citet{woo99} have shown that the
256: parameter alpha, the ratio between the persistent flux and the average burst
257: flux taken over the time interval since the previous burst, displayed for
258: almost all the BATSE observations of the second
259: outburst a value greater than 20, which suggests that they are due to
260: thermonuclear burning rather than spasmodic accretion (type-I bursts rather
261: than type-II). However, if we consider the value of the flux obtained
262: from the SIGMA observations as the sum of the persistent and the burst
263: flux, and the value of the burst flux from the BATSE data, the value of
264: alpha for the period of the observation is of the order of the 10,
265: %(6.9 to 12.3 depending on the way the burst flux is obtained),
266: more consistent with the type-II nature of the bursts, as indicated by \citet{lew95}.
267: Once again, the difference
268: could be due to a possible contamination of the BATSE data by GRS 1737-31.
269:
270: \acknowledgments We acknowledge the paramount contribution of the SIGMA
271: Project Group of the CNES Toulouse Space Center to the overall success
272: of the mission. We thank the staffs of the Lavotchine Space Company, of
273: the Babakin Space Center, of the Baikonour Space Center, and the Evpatoria
274: Ground Station for their unfailing support. This research has made use of data
275: obtained through the High Energy Astrophysics Research Center Online Service,
276: provided by the NASA/Goddard Space Flight Center. J. Mej\'{\i}a was supported
277: by CAPES and CAPES grant BEX1224/00-0. T. Villela was partially supported by
278: CNPq under grant 302266/88-7-FA. The Space Research Institute (IKI)
279: co-authors would like to acknowledge partial support of this work by RBRF grant
280: No. 00-15-96649.
281:
282: \begin{thebibliography}{}
283: \bibitem[Augusteijn et al.(1996)]{aug96} Augusteijn, T. et al. 1996, IAU Circ. 6369
284: \bibitem[Bildsten and Brown(1997)]{bil97} Bildsten, L., and Brown, E. F. 1997, \apj, 477, 897
285: \bibitem[Bouchet et al.(2001)]{bou01} Bouchet, L. et al. 2001, \apj, 548, 990
286: \bibitem[Cole et al.(1997)]{col97} Cole, D.M. et al. 1997, \apj, 480, 377
287: \bibitem[Cui et al.(1997)]{cui97} Cui, W. et al. 1997, IAU Circ. 6604
288: \bibitem[Daumerie et al.(1996)]{dau96} Daumerie, P.R. et al. 1996, Nature, 382, 141
289: \bibitem[Dotani et al.(1996)]{dot96} Dotani, T. et al. 1996, IAU Circ. 6337
290: \bibitem[Fishman et al.(1995)]{fis95} Fishman, G.J. et al. 1995, IAU Circ. 6272
291: \bibitem[Finger, Wilson, and van Paradijs(1996)]{fin96} Finger, G.J., Wilson, R.B.,
292: and van Paradijs, J. 1996, IAU Circ. 6286
293: \bibitem[Heise(1997)]{hei97} Heise, J. 1997, IAU Circ. 6606
294: \bibitem[Kouveliotou et al.(1996)]{kou96} Kouveliotou, C. et al. 1996, IAU Circ 6369
295: \bibitem[Hurley et al.(2000)]{hur00} Hurley, K. et al. 2000, \apj, 537, 953
296: \bibitem[Lewin, Rutledge, and Kommers(1996)]{lew96} Lewin, W., Rutledge, R., and Kommers, J.
297: 1996, \apjl, 462, L39
298: \bibitem[Lewin, van Paradijs, and Taam(1995)]{lew95} Lewin, W., van Paradijs, J., and Taam, R.
299: 1995, X-Ray Binaries, W. Lewin, J. van Paradijs, and E. van de Heuvel, Cambridge: Cambridge Univ., 196
300: \bibitem[Marshall and Smith(1997)]{mar97} Marshall, F.E., and Smith, D.M. 1997, IAU Circ. 6603
301: \bibitem[Miller(1996)]{mil96} Miller, G.S. 1996, \apjl, 468, L29
302: \bibitem[Nishiuchi et al.(1999)]{nis99} Nishiuchi, M. et al. 1999, \apj, 517, 436
303: \bibitem[Paul et al.(1991)]{pau91} Paul, J. et al. 1991, Adv. Space Res., 11, 289
304: \bibitem[Sazonov, Sunyaev, and Lund(1997)]{saz97} Sazonov, S.Y., Sunyaev, R.A., and Lund, N.
305: 1997, Astron. Lett., 23, 286
306: \bibitem[Sturner and Dermer(1996)]{stu96} Sturner, S.J., $\&$ Dermer, C.D. 1996, ApJ, 465, L31
307: \bibitem[Sunyaev et al.(1997)]{sun97} Sunyaev, R. et al. 1997, IAU Circ. 6599
308: \bibitem[Trudolyubov et al.(1999)]{tru99} Trudolyubov, S. et al. 1999, A$\&$A, 342, 496
309: \bibitem[Woods et al.(1999)]{woo99} Woods, P.M. et al. 1999, ApJ, 517, 431
310:
311: \end{thebibliography}
312:
313: \clearpage
314:
315: %Figure 1 ---------------------------------------
316: \begin{figure}
317:
318: \centerline{\psfig{figure=f1a.ps,height=75mm,width=75mm}
319: \psfig{figure=f1b.ps,height=75mm,width=75mm,angle=90}}
320:
321: \caption{Contour images of the GC region obtained with SIGMA on March 1996 (left) and
322: March 1997 (right) in the 35-75 keV energy band. Right Ascension and Declination are
323: in degrees. Notice the presence of the transient
324: source GRS 1737-31 in the March 1997 image. Coordinates are for 1950.0. Confidence
325: levels start at 3$\sigma$ with 1.5$\sigma$ steps. \label{Figure1}}
326: \end{figure}
327:
328: \clearpage
329:
330: %Figure 2 ---------------------------------------
331: \begin{figure}
332:
333: \centerline{\psfig{figure=f2a.ps,height=60mm,width=80mm,angle=90}
334: \psfig{figure=f2b.ps,height=60mm,width=80mm,angle=90}}
335:
336: \caption{Light curve of GRO J1744-28 in the 35-75 keV energy band for the March 1996
337: and March 1997 campaigns, as observed with SIGMA. The horizontal axis corresponds to the
338: date since the begining of the SIGMA observations on 1 Dec. 1989. \label{Figure2}}
339: \end{figure}
340:
341: \clearpage
342:
343: %Figure 3 ---------------------------------------
344: \begin{figure}
345:
346: \centerline{\psfig{figure=f3.ps,height=100mm,width=150mm,angle=90}}
347:
348: \caption{Light curve of GRO J1744-28 in the 35-75 keV energy band for all the
349: SIGMA observing campaigns in the direction of the GC region, begining on 1 Dec. 1989. \label{Figure3}}
350: \end{figure}
351:
352: \clearpage
353:
354: %Figure 4 ---------------------------------------
355: \begin{figure}
356:
357: \centerline{\psfig{figure=f4a.ps ,height=90mm,width=80mm}
358: \psfig{figure=f4b.ps ,height=90mm,width=80mm}}
359:
360: \caption{Combined RXTE/PCA and GRANAT/SIGMA spectra of GRO J1744-28 obtained in March 1996
361: and March 1997.
362: The PCA data points were renormalized to the SIGMA flux. The dotted line represents the best
363: fit model for the spectra (see text), with power law photon index $\alpha$$\sim$1.4 and
364: high energy cutoff E$\sim$15 keV in the first case and E$\sim$16 keV in the second. \label{Figure4}}
365: \end{figure}
366:
367: \clearpage
368:
369: \begin{table}
370: \caption{SIGMA March 1996 and March 1997 observation log of the Galactic Center
371: region containing the coordinates of GRO J1744-28.}
372:
373: \label{Table1}
374: \[
375: \begin{array}{cccccc}
376: \hline\hline
377: \noalign{\smallskip}
378: ${\rm Session}$ & ${\rm Date}$ & ${\rm Exposure}$
379: & ${\rm 35-75 keV flux}$ & ${\rm 75-150 keV flux}$ &\\
380: & ${\rm (UT)}$ & ${\rm (hours)}$
381: & ${\rm (mCrab)}$ & ${\rm (mCrab)}$ &\\
382:
383: \noalign{\smallskip}
384: \hline
385: \noalign{\smallskip}
386: & & & ${\rm March 1996 Campaign}$ & &\\
387: \noalign{\smallskip}
388: \hline
389: \noalign{\smallskip}
390: 894 & 15.63-16.68 & 25.25 & 80 \pm 26 & <34 &\\
391: 895 & 16.82-18.23 & 33.84 & 77 \pm 15 & <23 &\\
392: 896 & 18.92-20.38 & 35.06 & 62 \pm 15 & <25 &\\
393: 897 & 20.77-22.48 & 41.14 & 75 \pm 16 & <21 &\\
394: 898 & 23.52-24.52 & 24.03 & 106 \pm 18 & 57 \pm 28 &\\
395: 899 & 24.77-26.48 & 41.14 & 86 \pm 16 & <24 &\\
396: 900 & 27.60-28.66 & 25.43 & 96 \pm 20 & <30 &\\
397: 901 & 28.79-30.50 & 41.14 & 96 \pm 17 & <28 &\\
398: TOTAL & & 267.03 & 85 \pm 6.4 & <28 &\\
399: \noalign{\smallskip}
400: \hline
401: \noalign{\smallskip}
402: & & & ${\rm March 1997 Campaign}$ & &\\
403: \noalign{\smallskip}
404: \hline
405: \noalign{\smallskip}
406: 935 & 14.54-15.59 & 25.30 & 34 \pm 17 & <30 &\\
407: 937 & 18.68-19.65 & 23.45 & 75 \pm 22 & <29 &\\
408: 938 & 19.79-20.69 & 21.49 & 69 \pm 21 & <33 &\\
409: 939 & 22.47-23.58 & 26.66 & 43 \pm 17 & <28 &\\
410: 940 & 23.71-24.71 & 24.00 & 22 \pm 21 & <28 &\\
411: 941 & 26.49-27.61 & 26.81 & 83 \pm 18 & 46 \pm 26 &\\
412: 942 & 27.73-28.63 & 21.58 & 34 \pm 20 & 58 \pm 31 &\\
413: TOTAL & & 148.53 & 52 \pm 7.4 & <33 &\\
414: \noalign{\smallskip}
415: \hline
416: \end{array}
417: \]
418: \noindent 1 mCrab corresponds to $\sim$1.08$\times$10$^{-4}$ and $\sim$4.4$\times$10$^{-5}$
419: photons cm$^{-2}$ s$^{-1}$ ($\sim$8.6$\times$10$^{-12}$ and $\sim$7.3$\times$10$^{-12}$
420: ergs cm$^{-2}$ s$^{-1}$) for the 35-75 keV and 75-150 keV bands respectively.
421: \end{table}
422:
423: \end{document}
424:
425:
426:
427:
428: