X‐Ray Luminosity Function and Total Luminosity of Low‐Mass X‐Ray Binaries in Early‐Type Galaxies

We have derived bias-corrected X-ray luminosity functions (XLFs) of sourcesdetected in a uniformly selected sample of 14 E and S0 galaxies observed withChandra ACIS-S3. The entire sample yields 985 point-like X-ray sources, withtypical detection of 30-140 sources per galaxy. After correcting forincompleteness, the individual XLFs are statistically consistent with a singlepower-law of a (differential) XLF slope b = 1.8 - 2.2 (with a typical error of0.2-0.3). A break at or near LX,Eddington, as reported in the literature forsome of these galaxies, is not required in any case. Although the combined XLFis marginally consistent with a single power-law (with b = 2.1 +- 0.1), abroken power-law gives an improved fit. The best-fit slope is b = 1.8 +- 0.2 inthe low-luminosity range LX = a few x 1037 - 5 x 1038 ergs s-1. At higherluminosities, the slope is steeper, b = 2.8 +- 0.6. The break luminosity is 5+- 1.6 x 1038 ergs s-1 (with an error at 90%), which may be consistent with theEddington luminosity of neutron stars with the largest possible mass (3 Mo), ofHe-enriched neutron star binaries, or low-mass stellar mass black holes. If thechange in XLF slope at high luminosities is real, our result would imply adifferent population of high luminosity sources, instead of a beaming effect.We confirm that the total X-ray luminosity of LMXBs is correlated with theoptical and more tightly with the near-IR luminosities, but in both cases thescatter exceeds that expected from measurement errors. We find that the scatterin LX(LMXB)/LK is marginally correlated with the specific frequency of globularclusters.Comment: 31 pages, 3 tables, 8 figures, final version accepted by ApJ, slightly revised with more rebust error calculatio