The Formation of the First Stars. I. The Primordial Star-forming Cloud

, , and

© 2002. The American Astronomical Society. All rights reserved. Printed in U.S.A.
, , Citation Volker Bromm et al 2002 ApJ 564 23 DOI 10.1086/323947

0004-637X/564/1/23

Abstract

To constrain the nature of the very first stars, we investigate the collapse and fragmentation of primordial, metal-free gas clouds. We explore the physics of primordial star formation by means of three-dimensional simulations of the dark matter and gas components, using smoothed particle hydrodynamics, under a wide range of initial conditions, including the initial spin, the total mass of the halo, the redshift of virialization, the power spectrum of the DM fluctuations, the presence of HD cooling, and the number of particles employed in the simulation. We find characteristic values for the temperature, T ~ a few 100 K, and the density, n ~ 103-104 cm-3, characterizing the gas at the end of the initial free-fall phase. These values are rather insensitive to the initial conditions. The corresponding Jeans mass is MJ ~ 103 M. The existence of these characteristic values has a robust explanation in the microphysics of H2 cooling, connected to the minimum temperature that can be reached with the H2 coolant, and to the critical density at which the transition takes place between levels being populated according to non-LTE (NLTE), and according to LTE. In all cases, the gas dissipatively settles into an irregular, central configuration that has a filamentary and knotty appearance. The fluid regions with the highest densities are the first to undergo runaway collapse due to gravitational instability, and to form clumps with initial masses ~103 M, close to the characteristic Jeans scale. These results suggest that the first stars might have been quite massive, possibly even very massive with M* ≳ 100 M. After a gas element has undergone runaway collapse, and has reached densities in excess of 108 cm-3, a sink particle is created. This procedure allows us to follow the evolution of the overall system beyond the point where the first nonlinear region would otherwise force the calculation to a halt. These later evolutionary stages, during which the clumps grow in mass due to accretion and merging with other clumps, are quite sensitive to the initial conditions. The key process in building up very massive clumps, with masses up to a few times 104 M, is merging between clumps. Since the merging rate sensitively depends on the density of the gas, halos with the highest degree of central concentration are able to assemble the most massive clumps. Among these are halos with a low spin (λ ≃ 0.01), and with DM fluctuations imprinted according to a white-noise spectrum.

Export citation and abstract BibTeX RIS

Please wait… references are loading.
10.1086/323947