Direct Collapse to Precursors of Supermassive Black Hole Seeds:Radiation-feedback-generated Outflows
Abstract
We use high-resolution zoom-in cosmological simulations to model outflow triggered by radiation and thermal drivers around the central mass accumulation during direct collapse within the dark matter (DM) halo. The maximal resolution is 1.3× 10-5\,pc, and no restrictions are put on the geometry of the inflow/outflow. The central mass is considered prior to the formation of the supermassive black hole seed at a redshift of z 15.9, and can constitute either a supermassive star (SMS) of 105\,M surrounded by a growing accretion disk or a self-gravitating disk. The radiation transfer is modeled using the ray-tracing algorithm. Due to the high accretion rate of 1\,M\, yr-1 determined by the DM halo, accretion is mildly supercritical, resulting in mildly super-critical luminosity which has only a limited effect on the accretion rate, with the duty cycle of 0.9. We observe a fast development of hot cavities, which quickly extend into polar funnels and expand dense shells. Within the funnels, fast winds, 103\, km\,s-1, are mass-loaded by the accreting gas. We follow the expanding shells to 1\,pc, when the shell velocity remains substantially, 5 times, above the escape speed. The ionization cones formed by the central UV/X-ray completely ionize the cavities. Extrapolating the outflow properties shows that the halo material outside the shell will have difficulty stopping it. We therefore conclude that the expanding wind-driven shell will break out of the central parsec and will reach the halo virial radius. Finally, the anisotropic accretion flow on sub-parsec scales will attenuate the UV/soft X-rays on the H2. Hence, the formation of funnels and powerful outflows around, e.g., SMS, can have interesting observational corollaries.
Turn this paper into a full lesson
ArcXiv compiles a staged curriculum from this paper: 8-12 lessons across beginner → advanced, synthesised section guides, visuals, flashcards, a quiz, exercises, and on-demand deep dives per section. Grounded in the abstract, never invented.