SIMBAD references

The X-ray emission from normal elliptical galaxies has two major components: soft (kT~0.2-1 keV) emission from diffuse gas and harder (kT~6 keV) emission from populations of accreting (low mass) stellar X-ray binaries. If the low-mass X-ray binary (LMXB) population is intimately tied to the field stellar population in a galaxy, its aggregate X-ray luminosity is expected to be simply proportional to the optical luminosity of the galaxy. However, recent ASCA and Chandra X-ray observations show that the global luminosities of LMXB components in elliptical galaxies exhibit significant scatter (a factor of ∼4) at a given optical luminosity. This scatter may reflect a range of evolutionary stages among X-ray binary populations in elliptical galaxies of different ages. If so, the ratio of the global LMXB X-ray luminosity to the galactic optical luminosity, L_LMXB/L_opt, may in principle be used to determine when the bulk of stars were formed in individual elliptical galaxies. To test this we compare variations in L_LMXB/L_opt for LMXB populations in elliptical galaxies to optically derived estimates of stellar ages in the same galaxies. We find no correlation, which suggests that variations in L_LMXB/L_opt are not a good age indicator for elliptical galaxies. Alternatively, LMXBs may be formed primarily in globular clusters (through stellar tidal interactions) rather than comprise a subset of the primordial binary star population in a galactic stellar field. Since elliptical galaxies exhibit a wide range of globular cluster populations for a given galactic luminosity, this may induce a dispersion in the LMXB populations of elliptical galaxies with similar optical luminosities. Indeed, we find that L_LMXB/L_opt ratios for LMXB populations are strongly correlated with the specific globular cluster frequencies in elliptical galaxies. This suggests that most LMXBs were formed in globular clusters. If so, Chandra observations of central dominant galaxies with unusually large globular cluster populations should find proportionally excessive numbers of LMXBs.