1: \documentclass{pazh_engl}
2: %\input ruschar
3: \usepackage{epsf}
4: \usepackage{flushrt}
5: \usepackage{graphicx}
6: \usepackage{natbib}
7:
8:
9: \begin{document}
10:
11: \sloppypar
12:
13: \title{Constraints on the luminosity of the stellar remnant
14: in SNR1987A}
15:
16: \author{P.Shtykovskiy\inst{1,2}, A.Lutovinov \inst{1,2}, M.Gilfanov\inst{2,1},
17: R.Sunyaev\inst{1,2}}
18:
19: \offprints{pav\_sht@hea.iki.rssi.ru}
20:
21: \institute{Space Research Institute, Russian Academy of Sciences,
22: Profsoyuznaya 84/32, 117997 Moscow, Russia
23: \and
24: Max-Planck-Institute f\"ur Astrophysik,
25: Karl-Schwarzschild-Str. 1, D-85740 Garching bei M\"unchen,
26: Germany
27: }
28: \date{Received 18 November 2004/Accepted}
29:
30: \authorrunning{Shtykovskiy et al.}
31: \titlerunning{Constraints on the luminosity of the stellar remnant
32: in SNR1987A}
33:
34: \abstract{We obtain photometric constraints on the luminosity of the stellar
35: remnant in SNR1987A using XMM-Newton and INTEGRAL data.
36: The upper limit in the 2--10 keV band based on the
37: XMM-Newton data is
38: $L_X\la5\times 10^{34}$erg/s. We note, however, that the optical depth
39: of the envelope is still high in the XMM-Newton band, therefore, this
40: upper limit does not constrain the true unabsorbed luminosity of the
41: central source. The optical depth is expected to be small in the hard
42: X-ray band of the IBIS telescope aboard the INTEGRAL
43: observatory, therefore it provides an unobscured look at the stellar
44: remnant. We did not detect statistically significant emission
45: from SN1987A in the 20-60 keV band with the upper limit of
46: $L_X\la1.1\times 10^{36}$erg/s.
47: We also obtained an upper limit on the mass of radioactive $^{44}$Ti
48: M($^{44}$Ti)$\la10^{-3}$M$_{\sun}$.
49: \keywords{supernova remnants, pulsars}
50: }
51:
52: \maketitle
53:
54: %
55: %________________________________________________________________
56:
57: \section{Introduction.}
58: \label{sec:intro}
59: Owing to its proximity, Supernova SN1987A in the Large
60: Magellanic Cloud provides a unique possibility to study the physics
61: of a supernovae and supernova remnants.
62:
63: Today it reveals itself as an extended object experiencing rapid
64: evolution. The observed X-ray emission of the supernova is dominated by the
65: interaction of the shock wave with the matter produced by the wind
66: from a supergiant phase of its progenitor. As a result of this
67: interaction the supernova was gradually increasing its luminosity
68: during the past decade. This was followed by a strong brightening of
69: the supernova on December 2002 when the shock wave encountered the
70: so-called inner ring (see \citet{chandra} and references therein).
71:
72: Based on the Balmer lines in the optical spectrum, identification of a
73: massive blue supergiant progenitor
74: \citep{progenitor} and the neutrino burst \citep{neutrinoburst} coincident
75: with the SN event, SN1987A was classified as a Type II core-collapse
76: supernova.
77: The mass of the progenitor was estimated to be
78: $\approx15-20$M$_{\sun}$ \citep{progenitormass}.
79: \citet{sunyaevhexe} detected strong very hard X-ray source after about
80: 169 days from the explosion but this emission was connected with
81: comptonisation of gamma-ray lines of Co$^{56}$ decay in the optically
82: thick envelope. The lines themselves became visible later \citep[e.g.][]{gammalines}.
83:
84: The collapse of such a massive star should lead to the formation of a
85: neutron star or a black hole. If a neutron star was born with strong
86: magnetic field and was rapidly rotating, it should appear as a luminous
87: Crab-like X-ray source or as an accreting pulsar, which usually also
88: have very hard X-ray spectrum with maximal contribution to
89: luminosity in the INTEGRAL band.
90: Neutron star without strong magnetic field or a black hole might
91: become a bright X-ray source due to the accretion of the matter
92: of the dense envelope.
93: \citet{chandra} obtained from an imaging analysis of the Chandra data
94: an upper limit L$_X$(2-10 keV)$<1.5\times10^{34}$ erg/s on the
95: luminosity of the central source.
96: However the envelope continues to be optically thick and there is
97: a possibility that its optical depth is still high in
98: the standart X-ray band and does not permit us to
99: observe the central source directly.
100: It is important to check that it is not specially bright in the hard
101: X-ray band, where the envelope is transparent both for the Thomson
102: scattering and photoabsorption.
103: In addition we decided to check the available data obtained by the
104: XMM-Newton observatory, which has higher sensitivity for photons
105: with energies $\ga5$keV than Chandra.
106:
107: Here we present results of analysis of INTEGRAL and XMM-Newton observations of
108: SNR1987A and constrain the luminosity of the stellar remnant embedded
109: in the envelope based on its X-ray spectrum.
110:
111:
112: \section{Data reduction and results}
113: \label{sec:data}
114:
115: \subsection{XMM-Newton}
116:
117: SNR1987A was observed by XMM-Newton in September and November 2000,
118: April 2001 and May 2003. These observations were processed with the
119: Science Analysis System (SAS) v6.0.0. The data were screened for the
120: soft proton flares by removing the time intervals where the count
121: rate above 10 keV significantly exceeded the mean level. For the
122: pointing in April 2001 because of numerous soft proton flares
123: the background threshold was chosen higher than the
124: quiescent background typical for time periods without flaring
125: activity. To be sure that this does not influences our results, we
126: repeated analysis with different thresholds on background level and a
127: background subtraction from different regions on detector and found no
128: significant variations in the obtained results.
129:
130: \begin{figure*}
131: \hbox{
132: \includegraphics[width=0.5\textwidth]{xmmspectrum1.ps}
133: \includegraphics[width=0.5\textwidth]{xmmspectrum2.ps}
134: }
135: \caption{{\em Left:}
136: EPIC PN spectra of the SNR1987A. The upper curve corresponds
137: to pointing on May 2003, the lower to pointing on September 2000.
138: {\em Right:}
139: Deconvolved photon spectrum of SNR1987A on April 2001 with the powerlaw model
140: describing the data on energies above 4keV.
141: }
142: \label{fig:xmmspectra}
143: \end{figure*}
144:
145: The angular resolution of the XMM-Newton, unlike that of the Chandra, is
146: insufficient to resolve the
147: remnant, and it appears as a point-like source on the extracted images.
148: However, XMM-Newton is more sensitive to the
149: photons with energy $>5$ keV and therefore allows to reconstruct the
150: sources spectra to higher energies.
151: The EPIC spectra were extracted from a $30\arcsec-40\arcsec$ circle
152: regions around the SNR. Corresponding background spectra were
153: extracted from the nearby regions.
154:
155:
156: The spectra of the SN1987A remnant obtained with XMM-Newton on September
157: 2000 and May 2003 together with NEI model fit (the plane-parallel
158: shock model \citep{borkowski}
159: where the plasma has not reached the collisional ionization equilibrium)
160: are shown in ~Fig.\ref{fig:xmmspectra}. As is
161: evident from the figure, the luminosity of the remnant increased by
162: a factor of $\approx$3 during three years. This is expected from
163: the interaction of the supernova shock with the matter left from the
164: stellar wind of a supergiant progenitor.
165: It seems that there is some excess of hard X-ray emission in
166: comparison with NEI model for pointing on May 2003.
167: Such a high energy tail could be associated with either a central source
168: or a synchrotron radiation from the shell of the remnant \citep[e.g.][]{synchrotronrem}.
169: We focus on the possible association of the high energy excess
170: with the central source.
171: Detailed analysis of the supernova remnant spectra is not a subject
172: of this paper and could be found in e.g. \citet{chandra,aschenbach}.
173:
174:
175: To obtain a conservative upper limit on the central source luminosity
176: we shall assume that the total flux observed at the high
177: energy part of the remnant's spectrum is produced by the central
178: source. For this analysis we
179: used the data of the early pointings (2000 and 2001 years),
180: as a contribution of the
181: emission originating from the interaction of the shock with the medium
182: would be smaller for them. Assuming the spectrum of the central source to be
183: a powerlaw with a photon index 2.0, we obtained the upper limit
184: (1$\sigma$) for its 2-10 keV luminosity L$\la5\times 10^{34}$erg/s,
185: assuming a distance to the source $d=50$ kpc. The obtained value is
186: in the same range, but somewhat higher than the upper limit obtained by Chandra
187: from the imaging analysis \citep{chandra}.
188:
189:
190:
191:
192: \subsection{INTEGRAL}
193:
194: The international gamma-ray laboratory INTEGRAL \citep{wink03}
195: observed the Large Magellanic Cloud several times on January 2003 with a
196: total exposure about 1 Mln sec. In our analysis we used publicly
197: available data obtained by the ISGRI detector of the IBIS telescope of
198: the observatory. ISGRI is sensitive to the photons in the E$>20$ keV
199: energy band. The data of INTEGRAL/IBIS/ISGRI were processed with the
200: method developed by Eugene Churazov and described in
201: Revnivtsev et al. (2004). Detailed analysis of
202: Crab nebula observations suggests that with the approach and software
203: employed, the conservative estimation of the uncertainty in measurements
204: of absolute fluxes from the sources is about 10\%.
205:
206:
207: To investigate the hard X-ray emission from the SN1987A remnant we
208: analyzed data obtained with INTEGRAL/IBIS/ISGRI in the 20-60 keV
209: energy band, where the sensitivity of the ISGRI detector is
210: maximal. We reconstructed the image of the central part of the Large
211: Magellanic Cloud in this energy band (see Fig. \ref{ima}) and did not
212: find statistically significant emission from the remnant with the
213: upper limit of
214: $F_X\la3.7\times10^{-12}$ erg/s/cm$^2$ (2$\sigma$, assuming
215: a Crab-like spectrum), that corresponds to the luminosity of
216: $L_X\la1.1\times 10^{36}$erg/s.
217:
218: \begin{figure*}
219: \begin{center}
220: \includegraphics[width=0.8\textwidth,clip=true]{sn_ima300.ps}
221: \end{center}
222: \caption{Map of the part of LMC around of the SN1987A position,
223: obtained with INTEGRAL in the 20-60 keV energy band. Position of
224: SN1987A is shown by the cross.} \label{ima}
225: \end{figure*}
226:
227: \begin{figure*}
228: \begin{center}
229: \includegraphics[width=0.8\textwidth,clip=true]{limit.ps}
230: \end{center}
231: \caption{The typical spectra produced by
232: accreting X-ray pulsar, accreting black hole \citep{cygx1} and
233: rotation-powered Crab-like pulsar.
234: All spectra are normalised to the luminosity of $2.5\cdot10^{36}$erg/s in the
235: 1-12 keV energy band.
236: This corresponds to the bolometric optical and infrared luminosity
237: \citep{bolom}, which could be interpreted as an upper limit on the
238: central source luminosity absorbed by the envelope.
239: Upper limits on the luminosity of central source
240: in SNR1987A obtained by the XMM-Newton and INTEGRAL observatories
241: are also shown.
242: } \label{fig:limit}
243: \end{figure*}
244:
245:
246:
247:
248: To compare this result with results of Chandra and XMM-Newton
249: observatories we can extrapolate it to the 2-10 keV energy band.
250: The spectra of pulsars in the supernova remnants are usually
251: approximated by a powerlaw model with a slope of $\sim1.5-2$.
252: For such values of the photon index we can estimate
253: the central source luminosity in the 2-10 keV energy band corresponding to the
254: INTEGRAL upper limit as $L_X\la(0.6-1.6)\times 10^{36}$erg/s.
255:
256:
257: \subsection{Optical depth of the envelope}
258: The envelope of the supernova at early times
259: was optically thick both for photoabsorption and Thomson scattering.
260: This, for example, follows from the observations of Rentgen observatory
261: on the KVANT module of MIR station
262: \citep[][see also \citet{snexpectedem}]{sunyaevhexe}.
263: As the envelope expands and its density decreases, its
264: optical depth decreases with time as $\tau\propto t^{-2}$.
265: Given the \citet{snlateemission} model for the photoabsorption optical
266: depth of the envelope, and assuming t$^{-2}$ behaviour, we obtain
267: $\tau_{photo}\sim7$ for a 5 keV photons at the epoch of XMM-Newton obervations.
268: The photoabsorption optical depth reaches unity only on energies
269: $\sim$11-15 keV.
270: Thus, the optical depth is still high in the energy bands of Chandra
271: and XMM-Newton.
272: Therefore the upper limits provided by these observatories
273: correspond to the small fraction of the central source emission
274: which escaped the envelope,
275: while the real central source luminosity could be much
276: higher.
277:
278: Similar estimate of the optical depth for 40 keV photons give value
279: $\tau_{photo}\sim0.02$, therefore the hard X-ray radiation escapes
280: the envelope without absorption.
281: For the Thomson optical depth we obtained also very low values,
282: $\tau_T\sim0.005-0.1$ depending on the model of matter distribution
283: in the envelope.
284: Was the $\tau_T$ close to unity or even a bit higher, it would not
285: affect the flux in 20-60 keV energy band because of small
286: energy change for one scattering, $\Delta E/E\approx h\nu/m_ec^2\sim0.1$
287: \citep{pozdnyak}.
288: Thus, the INTEGRAL observations offer us a direct look at the
289: central source and its upper limit constraints the real luminosity
290: of the stellar remnant.
291:
292: Absorbed emission in the standart X-ray band is reemitted in the
293: optical and infrared bands and an upper limit on the central source
294: luminosity could be estimated based on a bolometric luminosity of the
295: SNR in these energy bands.
296: For example, \citet{bolom} estimated the bolometric luminosity of the remnant
297: log(L)$\sim36.1-36.4$ after the 3600 days from the explosion.
298: This value could be interpreted as an upper limit on the
299: central source luminosity corrected for the absorption, what is close to
300: a limit obtained from the INTEGRAL data.
301: Obviously, the upcoming Spitzer data will provide an interesting
302: estimates on the luminosity of the central source, absorbed and
303: reradiated in the infrared band by the optically thick for the
304: standart X-rays envelope.
305:
306:
307: \subsection{Limit on the nucleosynthesis of $^{44}$Ti}
308: While the early (except for the first few weeks) bolometric
309: luminosity of the supernova was powered by
310: the $^{56}$Co decay (t$_{1/2}\sim$77 days), at later times
311: elements with longer decay times, like $^{44}$Ti (t$_{1/2}\sim$60 years),
312: give the main contribution.
313: Decay of the $^{44}$Ti gives rise to
314: two gamma photons with energies 67.9 keV and 78.4 keV.
315: Simple estimations on the fluxes of these lines give value
316: $\sim3\cdot10^{-13}$erg/s/cm$^{2}$ \citep[see e.g.][]{titan},
317: assuming mass of titanium M($^{44}$Ti)$=10^{-4}$M$_{\sun}$.
318: To estimate the mass of the synthesed $^{44}$Ti, we calculated
319: fluxes observed by INTEGRAL in the 61-73 keV and 73-88 keV energy bands,
320: F$_X$(61-73 keV)=$4.30\pm1.25\cdot10^{-12}$erg/s/cm$^{2}$
321: and F$_X$(73-88 keV)=$2.50\pm1.60\cdot10^{-12}$erg/s/cm$^{2}$.
322: In the first case the flux
323: has a formal significance of 3.4$\sigma$, however we
324: consider it insufficient to claim a reliable detection of the line emission.
325: Therefore to estimate the mass of $^{44}$Ti, we assume both fluxes to
326: be upper limits and obtain M($^{44}$Ti)$\la10^{-3}$M$_{\sun}$.
327:
328: \section{Conclusion}
329: \label{sec:summary}
330: Based on the data of XMM-Newton and INTEGRAL, we provide broadband
331: spectroscopic constraints
332: on the luminosity of the central source in the SN1987A.
333: Assuming that a hard energy tail of the XMM-Newton spectrum of the
334: remnant is dominated by a central source, we obtained an upper limit
335: L$\la5\times 10^{34}$erg/s for 2-10 keV central
336: source luminosity uncorrected for absorption,
337: which is somewhat higher than the upper limit obtained by Chandra from
338: imaging analysis.
339: INTEGRAL did not detect any statistically significant emission from
340: the remnant and provided an
341: upper limit for 20-60 keV luminosity of $L_X\la1.1\times 10^{36}$erg/s,
342: what corresponds to 2-10 keV luminosity $L_X\la(0.6-1.6)\times
343: 10^{36}$erg/s assuming Crab-pulsar like spectrum.
344: The obtained upper limits on the possible central source luminosity
345: in comparison with the typical spectra of Crab-like pulsar,
346: accreting X-ray pulsar and accreting black hole are shown on Fig.~\ref{fig:limit}.
347:
348: Based on the existing calculations of the photoabsorption optical
349: depth of the envelope, we show that absorption in the standart X-ray band
350: of the XMM-Newton and Chandra is still high and obtained upper limit
351: corresponds to luminosity not corrected for absorption, which
352: could be only a small fraction of a real luminosity.
353: The optical depth in the INTEGRAL energy band is significantly less
354: than unity and obtained upper limit constrain real luminosity of a
355: stellar remnant.
356:
357: We also obtained an upper limit on the mass of $^{44}$Ti
358: M($^{44}$Ti)$\la10^{-3}$M$_{\sun}$.
359:
360:
361:
362:
363: \begin{acknowledgements}
364: The authors thank Eugene Churazov for developing of methods of
365: analysis of the IBIS data and software.
366: The authors also would like to thank the referee S.A. Grebenev
367: for helpful comments. This research has made use of
368: data obtained through the INTEGRAL Science Data Center (ISDC),
369: Versoix, Switzerland, Russian INTEGRAL Science Data Center (RSDC),
370: Moscow, Russia and XMM-Newton science archive.
371: This work was supported by grants of Minpromnauka
372: NSH-2083.2003.2 and the program of Russian Academy of Sciences
373: ``Non-stationary phenomena in astronomy''. AL acknowledge the support
374: of RFFI grant 04-02-17276.
375: \end{acknowledgements}
376:
377: \begin{thebibliography}{}
378:
379: \bibitem[\protect\citeauthoryear{Aschenbach}{2002}]{aschenbach}
380: Aschenbach, B. Proceedings of the 270. WE-Heraeus Seminar on Neutron Stars, Pulsars, and Supernova Remnants. MPE Report 278. Edited by W. Becker, H. Lesch, and J. Truemper. Garching bei Muenchen: Max-Plank-Institut fuer extraterrestrische Physik, 2002., p.13
381:
382: \bibitem[\protect\citeauthoryear{Borkowski et
383: al.}{2001a}]{borkowski} Borkowski, Kazimierz J., Lyerly, William J., \&
384: Reynolds, Stephen P. 2001, ApJ ,548, 820
385:
386:
387: \bibitem[\protect\citeauthoryear{Borkowski et
388: al.}{2001b}]{synchrotronrem}Borkowski, Kazimierz J., Rho, Jeonghee,
389: Reynolds, Stephen P.,\& Dyer, Kristy K. 2001b, ApJ, 550, 334
390:
391:
392: \bibitem[\protect\citeauthoryear{Fransson \& Chevalier}{1987}]{snlateemission}Fransson, C.; Chevalier, R. A. 1987, ApJ, 322L, 15
393:
394:
395: \bibitem[\protect\citeauthoryear{Gilfanov et al.}{2000}]{cygx1}
396: Gilfanov, M., Churazov, E., \& Revnivtsev, M. 2000, MNRAS, 316, 923
397:
398:
399:
400: \bibitem[\protect\citeauthoryear{Grebenev \& Syunyaev}{1987}]{snexpectedem}
401: Grebenev, S. A., \& Syunyaev, R. A. 1987, SvAL, 13, 397
402:
403:
404:
405:
406: \bibitem[\protect\citeauthoryear{Hirata et al.}{1987}]{neutrinoburst}
407: Hirata, K., Kajita, T., Koshiba, M., Nakahata, M., \& Oyama, Y. 1987, PhRvL, 58, 1490
408:
409: %\bibitem[\protect\citeauthoryear{McCray}{1993}]{coldens}
410: %McCray, R. 1993, ARA\&A, 31, 175
411:
412: \bibitem[\protect\citeauthoryear{Motizuki \& Kumagai}{2004}]{titan}
413: Motizuki, Yuko, Kumagai, Shiomi 2004, NewAR, 48, 69
414:
415:
416:
417: \bibitem[\protect\citeauthoryear{Park et al.}{2004}]{chandra}
418: Park, Sangwook, Zhekov, Svetozar A., Burrows, David N., Garmire,
419: Gordon P., McCray, Richard 2004, ApJ, 610, 275
420:
421: \bibitem[\protect\citeauthoryear{Pozdniakov et al.}{1983}]{pozdnyak}
422: Pozdniakov, L. A., Sobol, I. M., Siuniaev, R. A. 1983, ASPRv, 2, 189
423:
424: \bibitem[\protect\citeauthoryear{Revnivtsev et al.}{2004}]{rev04}
425: Revnivtsev, M., Sunyaev, R., Varshalovich, D., et al. 2004, Astr.
426: Lett., 30, 382
427:
428: \bibitem[\protect\citeauthoryear{Sonneborn et al.}{1987}]{progenitor}
429: Sonneborn, George, Altner, Bruce, Kirshner, \& Robert P. 1987, ApJ, 323L, 35
430:
431: %\bibitem[\protect\citeauthoryear{Suntzeff et al.}{1992}]{bolom}
432: %Suntzeff, Nicholas B., Phillips, M. M., Elias, J. H., Walker, A. R.,
433: %Depoy, D. L. 1992, ApJ, 384L, 33
434:
435: \bibitem[\protect\citeauthoryear{Suntzeff}{1997}]{bolom}
436: Suntzeff, Nicholas B., "SN1987A: Ten Years After," The Fifth
437: CTIO/ESO/LCO Workshop, 22-28 Feb 1997, La Serena, Chile, editors
438: M.M. Phillips and N.B. Suntzeff, ASP Conference Series, astro-ph/9707324.
439:
440:
441: \bibitem[\protect\citeauthoryear{Sunyaev et al.}{1987}]{sunyaevhexe}
442: Sunyaev, R. et al. 1987, Nature, 330, 227
443: %10 Aug 1987
444: %23feb1987 neutrino burst
445:
446:
447: \bibitem[\protect\citeauthoryear{Teegarden et
448: al.}{1989}]{gammalines}Teegarden, B. J., Barthelmy, S. D., Gehrels,
449: N., Tueller, J., Leventhal, M. 1989, Nature, 339, 122
450:
451: \bibitem[\protect\citeauthoryear{Winkler et al.}{2003}]{wink03} Winkler, C.,
452: Courvoisier, T., Di Cocco, G., et al. 2003 A\&A, 411, L1
453:
454: \bibitem[\protect\citeauthoryear{Woosley et al.}{1987}]{progenitormass}
455: Woosley, S. E., Pinto, P. A., Martin, P. G., Weaver, Thomas A. 1987,
456: ApJ, 318, 664
457:
458:
459: \end{thebibliography}
460:
461: \end{document}
462:
463: