[Paper Review] The Collapse of Atomically-Cooled Primordial Haloes. I. High Lyman-Werner Backgrounds

Samuel Patrick, Daniel J. Whalen+2 more|arXiv (Cornell University)|Dec 21, 2020

International Science and Diplomacy4 citations

TL;DR

This study models the long-term collapse of atomically-cooled primordial haloes under high Lyman-Werner backgrounds using high-resolution hydrodynamic simulations, revealing that fragmentation in accretion disks is nearly universal and leads to the formation of binary or multiple supermassive stars. The simulations confirm that rapid baryon collapse persists long enough for direct-collapse black holes (DCBHs) to form, suggesting such binaries were common in the early universe and could merge via gravitational wave emission detectable by LISA.

ABSTRACT

Pristine, atomically-cooled haloes may be the sites of primordial quasar formation because atomic cooling triggers rapid baryon collapse that can create 10$^4$ - 10$^5$ M$_{\odot}$ black hole seeds. However, no numerical simulation has ever followed the collapse of these haloes for the times required to form supermassive stars and direct-collapse black holes (DCBHs). We have now modeled baryon collapse in atomically-cooled haloes with a wide range of spin parameters and assembly histories for times that are sufficient for DCBH formation. Fragmentation of accretion disks after $\sim$ 500 kyr is nearly ubiquitous in these haloes and in most cases leads to the formation of binary or multiple supermassive stellar systems. They also confirm that rapid baryon collapse proceeds for the times required for these stars to form DCBHs. Our simulations suggest that binary or even multiple DCBH formation was the rule rather than the exception in the primordial Universe.

Motivation & Objective

• To investigate the long-term evolution of baryon collapse in atomically-cooled primordial haloes under high Lyman-Werner (LW) UV backgrounds.
• To determine whether accretion rates remain high over timescales sufficient for direct-collapse black hole (DCBH) formation.
• To examine the role of disk fragmentation in shaping the final outcome of collapse in these haloes.
• To assess the likelihood of forming binary or multiple DCBH systems and their potential detectability via gravitational waves.

Proposed method

• High-resolution cosmological hydrodynamic simulations using the Enzo code to model baryon collapse in atomically-cooled haloes with diverse spin parameters and assembly histories.
• Implementation of sink particles to track accretion and resolve protostellar scales while maintaining numerical stability.
• Use of a pressure floor to prevent artificial fragmentation at small scales and ensure accurate angular momentum transport.
• Incorporation of radiative feedback from forming stars to assess its impact on accretion and fragmentation.
• Simulation of collapse over 600 kyr in multiple haloes to capture disk evolution and fragment formation.
• Analysis of disk structure and fragment dynamics using the ytree software for halo and substructure identification.

Experimental results

Research questions

• RQ1 Does baryon collapse in atomically-cooled haloes persist long enough to form direct-collapse black holes (DCBHs) under high Lyman-Werner backgrounds?
• RQ2 To what extent is disk fragmentation inevitable during the collapse of primordial haloes, and what are the resulting systems (e.g., binaries, multiples)?
• RQ3 How do the final masses and configurations of forming stars and black holes depend on halo spin and assembly history?
• RQ4 What is the likelihood of forming binary or multiple DCBH systems that could merge and emit detectable gravitational waves?
• RQ5 How do the properties of the resulting systems compare to those expected from H2-cooling or Pop III star formation pathways?

Key findings

• Fragmentation in accretion disks occurs nearly ubiquitously after ~500 kyr, leading to the formation of binary or multiple supermassive stellar systems in most simulated haloes.
• Rapid baryon collapse persists for sufficient timescales—over 600 kyr—to allow the formation of direct-collapse black holes (DCBHs) with masses of 10^4–10^5 M☉.
• The simulations confirm that DCBH formation is not a rare event but rather the norm, with binary or multiple systems forming as the rule rather than the exception in the early universe.
• DCBHs form in close proximity (0.1–0.2 pc) and high-density environments (≥10^7 cm⁻³), enabling rapid orbital decay via radiative drag and subsequent mergers detectable by LISA.
• The final systems include SMS–SMS, SMS–DCBH, and DCBH–DCBH binaries, with potential for unique X-ray and near-infrared signatures from tidal disruption events.
• Despite the absence of magnetic fields in the simulations, fragmentation patterns are consistent with expectations, suggesting that large-scale clump formation is robust to magnetic suppression on the scales resolved.