Cold Filamentary Accretion and the Formation of Metal Poor Globular Clusters and Halo Stars

TL;DR

This paper proposes that cold filamentary accretion in high-redshift massive galaxies can lead to the formation of metal-poor globular clusters and halo stars, supported by cosmological simulations and analytical models.

Contribution

It introduces a novel scenario where cold streams induce star formation and globular cluster formation in galaxy halos without dark matter sub-structure, supported by simulations and analytical modeling.

Findings

Cold streams fragment into star-forming clumps in galaxy halos.

Conditions for globular cluster formation are met at high redshift in turbulent regions.

Approximately 30% of metal-poor globular clusters in Milky Way-like halos could form via this mechanism.

Abstract

We propose that cold filamentary accretion in massive galaxies at high redshift can lead to the formation of star-forming clumps in the halos of these galaxies without the presence of dark matter sub-structure. In certain cases, these clumps can be the birth places of metal poor globular-clusters (MP GCs). Using cosmological simulations, we show that narrow streams of dense gas feeding massive galaxies from the cosmic web can fragment, producing star-forming clumps. We then derive an analytical model for the properties of streams as a function of halo mass and redshift, and assess when these are gravitationally unstable, when this can lead to collapse and star-formation in the halo, and when it may result in the formation of MP GCs. For stream metalicities >~1% solar, this is likely to occur at z>4.5. At z~6, the collapsing clouds have masses of (5-10)x10^7 M_{sun} and the average stream pressure is ~10^6 K cm^{-3}. The conditions for GC formation are met in the extremely turbulent "eyewall" at ~0.3 R_{vir}, where counter-rotating streams can collide, driving very large densities. Our scenario can account for the observed kinematics and spatial distribution of MP GCs, the correlation between their mass and metalicity, and the mass ratio between the GC system and the host halo. For MW mass halos, we infer that ~30% of MP GCs could have formed in this way, the remainder likely accreted in mergers. Our predictions for GC formation along circumgalactic filaments at high-redshift are testable with JWST.