Supermassive Black Hole Seed Formation at High Redshifts: Long-Term Evolution of the Direct Collapse

TL;DR

This study uses advanced simulations to explore how supermassive black hole seeds form and evolve at high redshifts through direct gas collapse, revealing key processes and disk structures that support rapid SMBH growth.

Contribution

It demonstrates the feasibility of SMBH seed formation via direct collapse, detailing the long-term evolution and disk dynamics with high-resolution cosmological simulations.

Findings

Central seed grows to ~2 million solar masses in 2 Myr.

No fragmentation detected during collapse.

Formation of misaligned inner and outer disks.

Abstract

We use cosmological adaptive mesh refinement (AMR) code Enzo zoom-in simulations to study the long term evolution of the collapsing gas within dark matter (DM) halos at high redshifts. This direct collapse process is a leading candidate for rapid formation of supermassive black hole (SMBH) seeds at high z. To circumvent the Courant condition at small radii, we have used the sink particle method, and focus on the evolution on scales ~0.01-10 pc. The collapse proceeds in two stages, with the secondary runaway happening within the central 10 pc, and with no detected fragmentation. The sink particles form when the collapsing gas requires additional refinement of the grid size at the highest refinement level. Their mass never exceeds ~10^3 Mo, with the sole exception of the central seed which grows dramatically to ~ 2 x 10^6 Mo in ~2 Myr, confirming the feasibility of this path to the SMBH. The time variability of angular momentum axis in the accreted gas results in the formation of two misaligned disks --- a small inner disk, and a more massive, outer disk which is inclined by ~45^o to the inner disk. The self-gravity of these disks is heavily diluted --- both disks lie within the Roche limit of the central seed. While the inner disk is geometrically thin and weakly asymmetric, the outer disk flares due to turbulent motions as a result of the massive inflow along a pair of penetrating filaments. The geometry of inflow via filaments determines the dominant and secondary Fourier modes in this disk --- these modes have a non-self-gravitational origin. We do not confirm that m=1 is a principal mode that drives the inflow in the presence of a central massive object. While the positions of the disks depend on the scale chosen to break the self-similar collapse, the overall configuration appears to be generic, and is expected to form when the central seed becomes sufficiently massive.