Improved constraints from ultra-faint dwarf galaxies on primordial black holes as dark matter

TL;DR

This study refines constraints on primordial black holes as dark matter candidates by analyzing a large sample of ultra-faint dwarf galaxies with extended PBH mass distributions, ruling out PBHs of 1-100 solar masses as the sole dark matter component.

Contribution

It improves previous constraints by using a larger observational sample and considering extended PBH mass distributions through collisional Fokker-Planck simulations.

Findings

PBHs of about 1 solar mass are only compatible with a subset of UFDGs.

No PBH mass in the 1-100 solar mass range can constitute all dark matter.

Most simulated UFDGs with PBHs are too spatially extended compared to observations.

Abstract

Soon after the recent first ever detection of gravitational waves from merging black holes it has been suggested that their origin is primordial. Appealingly, a sufficient number of primordial black holes (PBHs) could also partially or entirely constitute the dark matter (DM) in the Universe. However, recent studies on PBHs in ultra-faint dwarf galaxies (UFDGs) suggest that they would dynamically heat up the stellar component due to two-body relaxation processes. From the comparison with the observed stellar velocity dispersions and the stellar half-light radii it was claimed that only PBHs with masses $\lesssim10\,M_\odot$ can significantly contribute to the DM. In this work, we improve the latter constraints by considering the largest observational sample of UFDGs and by allowing the PBH masses to follow an extended (log-normal) distribution. By means of collisional Fokker-Planck simulations, we explore a wide parameter space of UFDGs containing PBHs. The analysis of the half-light radii and velocity dispersions resulting from the simulations leads to three general findings that exclude PBHs with masses $\sim\mathcal{O}(1$-$100)\,M_\odot$ from constituting all of the DM: (i) We identify a critical sub-sample of UFDGs that only allows for $\sim\mathcal{O}(1)\,M_\odot$ PBH masses; (ii) for any PBH mass, there is an UFDG in our sample that disfavours it; (iii) the spatial extensions of a majority of simulated UFDGs containing PBHs are too large to match the observed.