Low-mass X-ray binaries indicate a top-heavy stellar initial mass function in ultra compact dwarf galaxies

TL;DR

This study provides evidence that ultra compact dwarf galaxies have a top-heavy stellar initial mass function, indicated by the high frequency of low-mass X-ray binaries, which explains their high mass-to-light ratios.

Contribution

It demonstrates that UCDs likely have a top-heavy IMF, supported by X-ray binary data and supernova rates, offering a new independent validation of previous IMF hypotheses.

Findings

LMXBs are up to 10 times more frequent in UCDs than expected.

A top-heavy IMF explains both high mass-to-light ratios and LMXB abundance.

Starburst galaxy Arp 220 shows high SNII rates consistent with a top-heavy IMF.

Abstract

It has been shown before that the high mass-to-light ratios of ultra compact dwarf galaxies (UCDs) can be explained if their stellar initial mass function (IMF) was top-heavy, i.e. that the IMF was skewed towards high mass stars. In this case, neutron stars and black holes would provide unseen mass in the UCDs. In order to test this scenario with an independent method, we use data on which fraction of UCDs has a bright X-ray source. These X-ray sources are interpreted as low-mass X-ray binaries (LMXBs), i.e. binaries where a neutron star accretes matter from an evolving low-mass star. We find that LMXBs are indeed up to 10 times more frequent in UCDs than expected if the IMF was invariant. The top-heavy IMF required to account for this overabundance is the same as needed to explain the unusually high mass-to-light ratios of UCDs and a top-heavy IMF appears to be the only simultaneous explanation for both findings. Furthermore, we show that the high rate of type II supernovae (SNII) in the star-burst galaxy Arp 220 suggests a top-heavy IMF in that system. This finding is consistent with the notion that star-burst galaxies are sites where UCDs are likely to be formed and that the IMF of UCDs is top-heavy. It is estimated that the IMF becomes top-heavy whenever the star formation rate per volume surpasses 0.1 Solar masses per year and cubic parsec in pc-scale regions.