Faint X-ray Binaries and Their Optical Counterparts in M31

TL;DR

This study investigates the population of X-ray binaries in M31 using extensive Chandra observations, identifying their associations with star clusters and HII regions, and analyzing their properties and potential as high-mass XRB candidates.

Contribution

It provides the first comprehensive analysis of XRBs in M31, linking X-ray sources to star clusters and HII regions, and identifies new high-mass XRB candidates.

Findings

15 star clusters matched to 17 X-ray sources, mainly globular cluster low-mass XRBs.

Brighter and more compact star clusters are more likely to host X-ray sources.

10 X-ray sources matched with HII regions, with some being new high-mass XRB candidates.

Abstract

X-ray binaries (XRBs) are probes of both star formation and stellar mass, but more importantly remain one of the only direct tracers of the compact object population. To investigate the XRB population in M31, we utilized all 121 publicly available observations of M31 totalling over 1 Ms from $\it{Chandra's}$ ACIS instrument. We studied 83 star clusters in the bulge using the year 1 star cluster catalogue from the Panchromatic Hubble Andromeda Treasury Survey. We found 15 unique star clusters that matched to 17 X-ray point sources within 1'' (3.8 pc). This population is composed predominantly of globular cluster low-mass XRBs, with one previously unidentified star cluster X-ray source. Star clusters that were brighter and more compact preferentially hosted an X-ray source. Specifically, logistic regression showed that the F475W magnitude was the most important predictor followed by the effective radius, while color (F475W$-$F814W) was not statistically significant. We also completed a matching analysis of 1566 HII regions and found 10 unique matches to 9 X-ray point sources within 3'' (11 pc). The HII regions hosting X-ray point sources were on average more compact than unmatched HII regions, but logistic regression concluded that neither the radius nor H$\alpha$ luminosity was a significant predictor. Four matches have no previous classification and thus are high-mass XRB candidates. A stacking analysis of both star clusters and HII regions resulted in non-detections, giving typical upper limits of $\approx10^{32}$ erg s$^{-1}$, which probes the quiescent XRB regime.