Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 329, L37-L40 (1998)

Previous Section Next Section Title Page Table of Contents

3. Discussion

There are two types of X-ray bursts with durations less than a few minutes (see review by Lewin, Van Paradijs & Taam 1995). Type II bursts have only been seen from the rapid burster and GRO J1744-28 (Kouveliotou et al. 1996) and have spectra similar to the underlying persistent emission. They are thought to be due to accretion instabilities. Type I bursts are attributed to thermonuclear flashes on or near a neutron star surface. Detection of type I bursts is, therefore, a strong indicator for a neutron star. One diagnostic clearly distinguishes type I from type II bursts: spectral softening is only seen in type I bursts. The two bursts reported here are identified as type I bursts. The average black body temperature during both bursts is consistent with such measurements in other type I bursts. We conclude that there is strong evidence for the neutron star nature of this X-ray source. The inferred size of the emitting region for the black body radiation is consistent with a neutron star interpretation.

The ROSAT spectrum (see Sect. 1) translates into a 2 to 8 keV intensity of 2.8 mCrab. This is consistent with the low level emission detected with BeppoSAX-WFC. This means that, if NGC 6652 is an X-ray transient source, either since 1990 one happened to catch it three times during an on state or it is a long-duration transient. The latter is not uncommon for bright LMXB galactic transients as exemplified by GRO J1655-40 and KS 1731-260 which have been in an on state for years (e.g., Chen et al. 1997).

In the galactic disk, about 100 low-mass X-ray binaries are known, 60 of which are permanently bright. Type I X-ray bursts have been detected in about 20 of the permanent and 13 of the transient sources in the galactic disk according to the tabulation by Van Paradijs (1995). The brighter permanent sources do not show type I X-ray bursts: in these the thermonuclear fusion of helium into carbon is continuous rather than in bursts (Lewin et al. 1995). With the discovery of X-ray bursts in NGC 6652 X-ray bursts have been detected in ten out of twelve bright X-ray sources in globular clusters. Thus the fraction of sources in which an type I X-ray burst has been detected is higher in globular clusters than in the disk; this may be explained, partially by the larger observational interest in globular cluster sources, and partially through the fact that none of the sources in the globular clusters are in the high-luminosity range where type I bursts do not occur.

Amongst the about 40 transient low-mass X-ray binaries in the galactic disk, some 7 have been shown to harbour compact stars with masses deemed too high for a neutron star. These systems are thought to harbour black holes, and indeed no type I X-ray burst has been detected in any of them. Various other transients are candidate black holes on the basis of the spectral properties of their X-ray outbursts; as many as 70% of the X-ray transients might harbour a black hole (see Table 4 in Chen et al. 1997). If the statistics in globular clusters were the same, three or four black hole transients could be expected in this optimistic estimation. The probability of finding zero would then be about 5%, a 2- [FORMULA] indication that the distribution in globular clusters differs from that in the galactic disk. These statistics are subject to uncertainties. For instance, the estimate that 70% of the X-ray transients in the galactic disk harbour a black hole may be an over-estimate, since the spectral characteristics are suggestive evidence, but no proof for a black hole.

The formation of low-mass X-ray binaries is different in globular clusters than in the disk. Tidal capture and exchange encounters favour the capture of more massive stars, and thus of black holes above neutron stars. One may thus predict a larger presence of black holes in the X-ray binaries in globular clusters than in the galactic disk. The formation mechanisms of low-mass X-ray binaries in globular clusters are not understood well enough to allow reliable estimates of this effect but it is interesting to note that it is contrary to what observations suggest.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: December 8, 1997