Radiation hydrodynamic simulations of massive star formation via gravitationally trapped H ii regions – spherically symmetric ionized accretion flows

Abstract

This paper investigates the gravitational trapping of Hii regions predicted by steadystate analysis using radiation hydrodynamical simulations. We present idealised spherically symmetric radiation hydrodynamical simulations of the early evolution of Hii regions including the gravity of the central source. As with analytic steady state solutions of spherically symmetric ionised Bondi accretion flows, we find gravitationally trapped Hii regions with accretion through the ionisation front onto the source. We found that, for a constant ionising luminosity, fluctuations in the ionisation front are unstable. This instability only occurs in this spherically symmetric accretion geometry. In the context of massive star formation, the ionising luminosity increases with time as the source accretes mass. The maximum radius of the recurring Hii region increases on the accretion timescale until it reaches the sonic radius, where the infall velocity equals the sound speed of the ionised gas, after which it enters a pressure-driven expansion phase. This expansion prevents accretion of gas through the ionisation front, the accretion rate onto the star decreases to zero, and it stops growing from accretion. Because of the time required for any significant change in stellar mass and luminosity through accretion our simulations keep both mass and luminosity constant and follow the evolution from trapped to expanding in a piecewise manner. Implications of this evolution of Hii regions include a continuation of accretion of material onto forming stars for a period after the star starts to emit ionising radiation, and an extension of the lifetime of ultracompact Hii regions.

CMACIONIZE 2.0: a novel task-based approach to Monte Carlo radiation transfer

“…The Bondi test hence tests the coupling between hydrodynamics and external gravity, the use of inflow boundary conditions, and the use of masked-out regions. In principle, the test could be extended to include a trapped Hii region, but because this setup is unstable (Lund et al 2019;Vandenbroucke et al 2019), this does not provide a practical benchmark test.…”

CMacIonize 2.0: a novel task-based approach to Monte Carlo radiation transfer

“…The Bondi test hence tests the coupling between hydrodynamics and external gravity, the use of inflow boundary conditions, and the use of masked-out regions. In principle, the test could be extended to include a trapped Hii region, but since this setup is unstable (Lund et al 2019;Vandenbroucke et al 2019), this does not provide a practical benchmark test.…”