Simulating the growth of Intermediate Mass Black Holes

TL;DR

This study uses radiation-hydrodynamic simulations to explore how primordial gas accretes onto direct collapse black holes in early dark matter halos, leading to the growth of intermediate mass black holes with super-Eddington luminosities.

Contribution

It provides the first detailed numerical model of the accretion process and growth of IMBH seeds at high redshift, highlighting their intermittent accretion and super-Eddington emission phases.

Findings

Average accretion rate is about 1.35 times the Eddington rate.

Black hole grows to approximately 7 million solar masses.

System emits super-Eddington luminosities during most of the evolution.

Abstract

Theoretical models predict that a population of Intermediate Mass Black Holes (IMBHs) of mass $M_\bullet \approx 10^{4-5} \, \mathrm{M_{\odot}}$ might form at high ($z > 10$) redshift by different processes. Such objects would represent the seeds out of which $z > 6$ Super-Massive Black Holes (SMBHs) grow. We numerically investigate the radiation-hydrodynamic evolution governing the growth of such seeds via accretion of primordial gas within their parent dark matter halo of virial temperature $T_{vir} \sim 10^4 \, \mathrm{K}$. We find that the accretion onto a Direct Collapse Black Hole (DCBH) of initial mass $M_0=10^5 \, \mathrm{M_{\odot}}$ occurs at an average rate $\dot{M}_{\bullet} \simeq 1.35 \, \dot{M}_{Edd} \simeq 0.1 \, \mathrm{M_{\odot} \, yr^{-1}}$, is intermittent (duty-cycle $ < 50\%$) and lasts $\approx 142 \, \mathrm{Myr}$; the system emits on average at super-Eddington luminosities, progressively becoming more luminous as the density of the inner mass shells, directly feeding the central object, increases. Finally, when $\approx 90\%$ of the gas mass has been accreted (in spite of an average super-Eddington emission) onto the black hole, whose final mass is $\sim 7 \times 10^6 \, \mathrm{M_{\odot}}$, the remaining gas is ejected from the halo due to a powerful radiation burst releasing a peak luminosity $L_{peak}\sim 3\times 10^{45} \, \mathrm{erg \, s^{-1}}$. The IMBH is Compton-thick during most of the evolution, reaching a column density $N_H \sim 10^{25} \, \mathrm{cm^{-2}}$ in the late stages of the simulation. We briefly discuss the observational implications of the model.