Gas infall into atomic cooling haloes: on the formation of protogalactic discs and supermassive black holes at z > 10

Abstract

We have performed hydrodynamical simulations from cosmological initial conditions using the AMR code RAMSES to study atomic cooling haloes (ACHs) at z = 10 with masses in the range 5 × 10 7 M < ∼ M < ∼ 2 × 10 9 M . We assume the gas has primordial composition and H 2 -cooling and prior star-formation in the haloes have been suppressed. We present a comprehensive analysis of the gas and DM properties of 19 haloes at a spatial resolution of ∼ 10 (proper) pc, selected from simulations with a total volume of ∼ 2000 (comoving) Mpc 3 . This is the largest statistical hydro-simulation study of ACHs at z > 10 to date. We examine the morphology, angular momentum, thermodynamical state, and turbulent properties of these haloes, in order to assess the prevalence of disks and massive overdensities that may lead to the formation of supermassive black holes (SMBHs). We find no correlation between either the magnitude or the direction of the angular momentum of the gas and its parent DM halo. Only 3 of the haloes form rotationally supported cores. Two of the most massive haloes, however, form massive, compact over-dense blobs, which migrate to the outer region of the halo. These blobs have an accretion rate ∼ 0.5M yr −1 (at a distance of 100 pc from their center), and are possible sites of SMBH formation. Our results suggest that the degree of rotational support and the fate of the gas in a halo is determined by its large-scale environment and merger history. In particular, the two haloes that form over-dense blobs are located at knots of the cosmic web, cooled their gas early on (z > 17), and experienced many mergers. The gas in these haloes is thus lumpy and highly turbulent, with Mach numbers M > ∼ 5. In contrast, the haloes forming rotationally supported cores are relatively more isolated, located midway along filaments of the cosmic web, cooled their gas more recently, and underwent fewer mergers. As a result, the gas in these haloes is less lumpy and less turbulent (Mach numbers M < ∼ 4), and could retain most of its angular momentum. The remaining 14 haloes have a diverse range of intermediate properties. If verified in a larger sample of haloes and with additional physics to account for metals and star-formation, our results will have implications for observations of the highest-redshift galaxies and quasars with JWST.

Supermassive Black Hole Formation at High Redshifts via Direct Collapse: Physical Processes in the Early Stage

“…Our results can be compared to some extent with the concurrent works available in the literature, using ENZO (Wise et al 2008;Latif et al 2013) and RAMSES (Prieto et al 2013). All these works used cosmological initial conditions which provide a more realistic setting for the gravitational collapse, but provide less leverage when studying its detail, and are also more timeconsuming, limiting the number of models run.…”

“…Based on our model Dmod (Section 4.4), we expect that cosmological initial conditions will lead to the formation of a rotationally supported disk at somewhat smaller radii than in our models. This would explain why Prieto et al (2013) missed this runaway stage altogether. Latif et al (2013) has imposed a specific subgrid turbulence model of Schmidt et al (2006).…”

“…The outcome of these models agrees generally with our results of rotationally dominated disks, and turbulence-suppressed or delayed fragmentation. Prieto et al (2013), obtains rotationally supported "cores" in only 3 out of 19 cases. This can be simply explained by the maximal resolution of their models limited to 8 pc (but typically larger, e.g., 14, 15, and 22 pc).…”

“…An appealing alternative is that early SMBHs formed via direct collapse of the halo gas at z ∼ 10-20 (e.g., Oh & Haiman 2002;Bromm & Loeb 2003;Volonteri & Rees 2005;Begelman et al 2006;Wise et al 2008;Levine et al 2008;Regan & Haehnelt 2009;Begelman & Shlosman 2009;Mayer et al 2010;Schleicher et al 2010;Hosokawa et al 2011;Johnson et al 2011;Prieto et al 2013). The direct collapse paradigm assumes that an SMBH seed, M • 10 5 M , forms from the cold halo gas.…”

Supermassive black hole seed formation at high redshifts: long-term evolution of the direct collapse

“…Luminous quasars have been observed up to redshifts z ∼ 7.1 (e.g., Fan et al 2003;Mortlock et al 2011;Wu et al 2015), and have been estimated to host SMBHs in excess of 10 9 M⊙, just 700 Myr after the Big Bang. Unless the SMBHs are primordial, they must have formed by the accumulation of matter during the epoch of galaxy formation -either as remnants of the Population III stars (e.g., Haiman & Loeb 2001;Abel et al 2002;Bromm & Larson 2004; ⋆ E-mail: shlosman@pa.uky.edu † E-mail: jhchoi@astro.as.utexas.edu 2006; Li et al 2007;Pelupessy et al 2007), or the end products of gas collapse into dark matter (DM) halos with virial temperatures > ∼ 10 4 K (e.g., Haehnelt & Rees 1993;Oh & Haiman 2002;Bromm & Loeb 2003;Begelman et al 2006;Begelman & Shlosman 2009;Milosavljević et al 2009;Mayer et al 2010;Schleicher et al 2010;Hosokawa et al 2011;Johnson et al 2011;Choi et al 2013Choi et al , 2015Latif et al 2013;Prieto et al 2013). Accounting for radiative feedback during formation of Pop III stars lowers their main sequence masses to be more in line with those of normal stars.…”