The dipping low-mass X-ray binary 4U 1915-05 was observed by Suzaku on 2007 November 8 for a net exposure of 39 ksec. It was detected by the XIS with a 0.8-10 keV signal rate of 9.84+-0.01 cts/s per camera, and HXD-PIN with a 12-45 keV signal rate of
0.29+/-0.01 cts/s. After removing the periodic dips and an X-ray burst, the 0.8 - 45 keV continuum was successfully described by an optically thick disk emission with an inner-disk temperature ~ 0.7 keV and a neutron-star blackbody emission with a temperature ~ 1.3 keV, on condition that the blackbody component, or possibly the disk emission too, is significantly Comptonized. This successful modeling is consistent with 4U 1915-05 being in a high-soft state in this observation, and implies that its broadband spectrum can be interpreted in the same scheme as for many non-dipping Low-mass X-ray binaries in the soft state. Its bolometric luminosity (~ 0.02 times the Eddington limit) is relatively low for the soft state, but within a tolerance, if considering the distance and inclination uncertainties. As a high-inclination binary, this source exhibited stronger Comptonization effect, with a larger Comptonizing y-parameter, compared to low and medium inclination binaries. This suggests that the Comptonizing coronae of these objects in the soft state is in an oblate (rather than spherical) shape, extending along the accretion disk plane, because the y-parameter would not depend on the inclination if the corona were spherical.
For the first time, we have systematically explored the population of discrete X-ray sources in the outskirts of early-type galaxies. Based on a broad sample of 20 galaxies observed with Chandra we detected overdensity of X-ray sources in their outsk
irts. The overdensity appears as halos of resolved sources around the galaxies. These halos are broader than the stellar light, extending out to at least ~ 10 Re (Re is the effective radius). These halos are composed of sources fainter than ~5E38 erg/s, whereas the more luminous sources appear to follow the distribution of the stellar light, suggesting that the excess source population consists of neutron star binaries. Dividing the galaxy sample into four groups according to their stellar mass and specific frequency of globular clusters, we find that the extended halos are present in all groups except for the low-mass galaxies with low globular cluster content. We propose that the extended halos may be comprised of two independent components, low-mass X-ray binaries (LMXBs) located in globular clusters (GCs), which are known to have a wider distribution than the stellar light, and neutron star (NS) LMXBs kicked out of the main body of the parent galaxy by supernova explosions. The available deep optical and X-ray data of NGC 4365 support this conclusion. For this galaxy we identified 60.1+/-10.8 excess sources in the 4-10 Re region of which ~ 40% are located in GCs, whereas ~ 60% are field LMXBs. We interpret the latter as kicked NS LMXBs. We discuss the implications of these results for the natal kick distributions of black holes and neutron stars.
We investigate the dependence of the low-mass X-ray binary (LMXB) population in early-type galaxies on stellar age, by selecting 20 massive nearby early-type galaxies from the Chandra archive occupying a relatively narrow range of masses and spanning
a broad range of ages, from 1.6 Gyr to more than 10 Gyrs, with the median value of 6 Gyrs. With the ~ 2000 X-ray point sources detected in total, we correlated the specific number of LMXBs in each galaxy with its stellar age and globular cluster (GC) content. We found a correlation between the LMXB population and stellar age: older galaxies tend to possess about ~50% more LMXBs (per unit stellar mass) than the younger ones. The interpretation of this dependence is complicated by large scatter and a rather strong correlation between stellar age and GC content of galaxies in our sample. We present evidence suggesting that the more important factor may be the evolution of the LMXB population with time. Its effect is further amplified by the larger GC content of older galaxies and correspondingly, the larger numbers of dynamically formed binaries in them. We also found clear evolution of the X-ray luminosity function (XLF) with age, that younger galaxies have more bright sources and fewer faint sources per unit stellar mass. The XLF of LMXBs in younger galaxies appears to extend significantly beyond E39 erg/s. Such bright sources seem to be less frequent in older galaxies. We found that 6 out of ~ 12 (ultra-) luminous sources are located in GCs.
Based on the archival data from the Chandra observations of nearby galaxies, we study different sub populations of low-mass X-ray binaries (LMXBs) - dynamically formed systems in globular clusters (GCs) and in the nucleus of M31 and (presumably primo
rdial) X-ray binaries in the fields of galaxies. Our aim is to produce accurate luminosity distributions of X-ray binaries in different environments, suitable for quantitative comparison with each other and with the output of population synthesis calculations. Our sample includes seven nearby galaxies (M31, Maffei 1, Centaurus A,M81, NGC 3379, NGC 4697, and NGC 4278) and the Milky Way, which together provide relatively uniform coverage down to the luminosity limit of E35 erg/s. In total we have detected 185 LMXBs associated with GCs, 35 X-ray sources in the nucleus of M31, and 998 field sources of which ~ 365 are expected to be background AGN. We combine these data, taking special care to accurately account for X-ray and optical incompleteness corrections and the removal of the contamination from the cosmic X-ray background sources, to produce luminosity distributions of X-ray binaries in different environments to far greater accuracy than has been obtained previously. We found that luminosity distributions of GC and field LMXBs differ throughout the entire luminosity range, the fraction of faint (log(Lx) < 37) sources among the former being ~ 4 times less than in the field population. The X-ray luminosity function (XLF) of sources in the nucleus of M31 is similar to that of GC sources at the faint end but differs at the bright end, with the M31 nucleus hosting significantly fewer bright sources. We discuss the possible origin and potential implications of these results.
