Cosmological Simulations of Early Blackhole Formation: Halo Mergers, Tidal Disruption, and the Conditions for Direct Collapse

TL;DR

This study investigates the formation of supermassive blackholes via direct collapse in the early universe, using cosmological simulations to identify conditions and frequency of such events, highlighting the importance of halo mergers and tidal forces.

Contribution

It combines semi-analytic galaxy formation models with hydrodynamics simulations to assess the viability and rate of direct collapse blackhole formation in a cosmological setting.

Findings

Only 2 out of 42 halos led to direct collapse.

Tidal forces and ram pressure often prevent collapse.

Event rate of direct collapse is lower than previous estimates.

Abstract

Gravitational collapse of a massive primordial gas cloud is thought to be a promising path for the formation of supermassive blackholes in the early universe. We study conditions for the so-called direct collapse (DC) blackhole formation in a fully cosmological context. We combine a semianalytic model of early galaxy formation with halo merger trees constructed from dark matter $N$-body simulations. We locate a total of 68 possible DC sites in a volume of $20\;h^{-1}\;\mathrm{Mpc}$ on a side. We then perform hydrodynamics simulations for 42 selected halos to study in detail the evolution of the massive clouds within them. We find only two successful cases where the gas clouds rapidly collapse to form stars. In the other cases, gravitational collapse is prevented by the tidal force exerted by a nearby massive halo, which otherwise should serve as a radiation source necessary for DC. Ram pressure stripping disturbs the cloud approaching the source. In many cases, a DC halo and its nearby light source halo merge before the onset of cloud collapse. Only when the DC halo is assembled through major mergers, the gas density increases rapidly to trigger gravitational instability. Based on our cosmological simulations, we conclude that the event rate of DC is an order of magnitude smaller than reported in previous studies, although the absolute rate is still poorly constrained. It is necessary to follow the dynamical evolution of a DC cloud and its nearby halo(s) in order to determine the critical radiation flux for DC.