GW190425, GW190521 and GW190814: Three candidate mergers of primordial black holes from the QCD epoch

Abstract

The two recent gravitational-wave events GW190425 and GW190814 from the third observing run of LIGO/Virgo have both a companion which is unexpected if originated from a neutron star or a stellar black hole, with masses $[1.6-2.5]~M_\odot$ and $[2.5-2.7]~M_\odot$ and merging rates $ 460^{+1050}_{-360} $ and $ 7^{+16}_{-6}$ events/yr/Gpc$^3$ respectively, at 90\% c.l.. Moreover, the recent event GW190521 has black hole components with masses 67 and $91~M_\odot$, and therefore lies in the so-called pair-instability mass gap, where there should not be direct formation of stellar black holes. The possibility that all of these compact objects are Primordial Black Holes (PBHs) is investigated. The known thermal history of the Universe predicts that PBH formation is boosted at the time of the QCD transition, inducing a peak in their distribution at this particular mass scale, and a bump around $30-50~M_\odot$. We find that the merging rates inferred from GW190425, GW190521 and GW190814 are consistent with PBH binaries formed by capture in dense halos in the matter era or in the early universe. At the same time, the rate of black hole mergers around $30~M_\odot$ and of sub-solar PBH mergers do not exceed the LIGO/Virgo limits. Such PBHs could explain a significant fraction, or even the totality of the Dark Matter, but they must be sufficiently strongly clustered in order to be consistent with current astrophysical limits.