Modelling massive star feedback with Monte Carlo radiation hydrodynamics: photoionization and radiation pressure in a turbulent cloud

Abstract

We simulate a self-gravitating, turbulent cloud of 1000 M with photoionization and radiation pressure feedback from a 34 M star. We use a detailed Monte Carlo radiative transfer scheme alongside the hydrodynamics to compute photoionization and thermal equilibrium with dust grains and multiple atomic species. Using these gas temperatures, dust temperatures, and ionization fractions, we produce self-consistent synthetic observations of line and continuum emission. We find that all material is dispersed from the (15.5 pc) 3 grid within 1.6 Myr or 0.74 free-fall times. Mass exits with a peak flux of 2 × 10 −3 M yr −1 , showing efficient gas dispersal. The model without radiation pressure has a slight delay in the breakthrough of ionization, but overall its effects are negligible. 85 per cent of the volume, and 40 per cent of the mass, become ionized -dense filaments resist ionization and are swept up into spherical cores with pillars that point radially away from the ionizing star. We use free-free emission at 20 cm to estimate the production rate of ionizing photons. This is almost always underestimated: by a factor of a few at early stages, then by orders of magnitude as mass leaves the volume. We also test the ratio of dust continuum surface brightnesses at 450 and 850 µm to probe dust temperatures. This underestimates the actual temperature by more than a factor of 2 in areas of low column density or high line-of-sight temperature dispersion; the H ii region cavity is particularly prone to this discrepancy. However, the probe is accurate in dense locations such as filaments.

Photoionization feedback in turbulent molecular clouds

“…A number of works within the past few years has tackled the effect of photoionization within molecular clouds and it is interesting to draw a comparison between results. Some of the previous work included only photoionization (Mellema et al 2006;Dale et al 2012;Walch et al 2012;Boneberg et al 2015), whereas others have studied photoionization combined with another feedback mechanism, such as radiation pressure (Ali et al 2018;Kim et al 2018) or winds (Dale et al 2014), or considered extra physics, like Geen et al (2015).…”

“…In many of these works, the turbulent molecular cloud is completely destroyed. However, different studies arrive at very different results, with molecular clouds being sometimes completely dispersed by the ionizing radiation of one B-star (Geen et al 2015), one O-star and a few B stars (Ali et al 2018) or a few clusters of stars (Kim et al 2018), and dispersal time scales for the molecular cloud varying from less than a Myr (Ali et al 2018) to a few Myrs (Kim et al 2018).…”

“…This work varies from the previous studies in this respect. Firstly, we consider a turbulence-in-a-box type of scenario, similar to Mellema et al (2006), but which significantly differs from most of the recent work (Geen et al 2015;Ali et al 2018;Kim et al 2018). Usually, a cloud is simulated by adding a turbulent velocity field to a well defined dense region (often spherical at the start), which represents the molecular cloud and then embedding it in a uniform lower density medium.…”

Numerical Methods for Simulating Star Formation

“…Last, we note the developments made to capture ionizing radiation feedback from massive stars: Dale et al (2007) used a Strömgren volume method in SPH, Peters et al (2010) ray tracing in the FLASH code, Kuiper and Hosokawa (2018) an hybrid ray-tracing and FLD irradiation module in the PLUTO code, Geen et al (2015) the M1 moment method in the RAMSES code, and Harries (2015) Monte Carlo radiative transfer in the AMR code TORUS. These schemes are applied first in isolated collapse calculations (Kuiper and Hosokawa, 2018) and, mostly, in cluster formation studies (Dale and Bonnell, 2011;Peters et al, 2011;Dale et al, 2012;Geen et al, 2015Geen et al, , 2018Ali et al, 2018).…”

Shape analysis of H ii regions – II. Synthetic observations

“…This allows us to test observational diagnostics that are well defined in the simulations, and hence produce bespoke models for direct interpretation of the observational data. In this work, we use the statisti-cal shape analysis method of HII regions, developed in Campbell-White et al (2018) to directly compare realistic SOs of an HII region produced by the numerical simulations in Ali et al (2018) to radio continuum observational data. HII regions are the result of photoionisation from massive stars (> 8 M ).…”