2012ApJ...757...40B

We examined 134 Chandra observations of the population of X-ray sources associated with globular clusters (GCs) in the central region of M31. These are expected to be X-ray binary systems (XBs), consisting of a neutron star or black hole accreting material from a close companion. We created long-term light curves for these sources, correcting for background, interstellar absorption, and instrumental effects. We tested for variability by examining the goodness of fit for the best-fit constant intensity. We also created structure functions (SFs) for every object in our sample, the first time this technique has been applied to XBs. We found significant variability in 28 out of 34 GCs and GC candidates; the other 6 sources had 0.3-10 keV luminosities fainter than ∼2x10³⁶ erg/s, limiting our ability to detect similar variability. The SFs of XBs with 0.3-10 keV luminosities ∼2-50x10³⁶ erg/s generally showed considerably more variability than the published ensemble SF of active galactic nuclei (AGNs). Our brightest XBs were mostly consistent with the AGN SF; however, their 2-10 keV fluxes could be matched by <1 AGN per square degree. These encouraging results suggest that examining the long-term light curves of other X-ray sources in the field may provide an important distinction between X-ray binaries and background galaxies, as the X-ray emission spectra from these two classes of X-ray sources are similar. Additionally, we identify 3 new black hole candidates (BHCs) using additional XMM-Newton data, bringing the total number of M31 GC BHCs to 9, with 8 covered in this survey.