Shaping the structure of a GMC with radiation and winds

TL;DR

This study uses advanced 3D simulations to explore how stellar feedback mechanisms like radiation and winds influence the evolution and star formation efficiency of a giant molecular cloud, revealing two distinct evolutionary stages.

Contribution

It introduces a comprehensive simulation incorporating multiple stellar feedback processes and complex chemistry to analyze GMC evolution and star formation.

Findings

Radiation and winds create ionized bubbles and dissociate H2 in the GMC.

Star formation rapidly consumes molecular gas, leading to high UV escape.

H2 is exhausted in 1.6 Myr with a 36% star formation efficiency.

Abstract

We study the effect of stellar feedback (photodissociation/ionization, radiation pressure and winds) on the evolution of a Giant Molecular Cloud (GMC), by means of a 3D radiative transfer, hydro-simulation implementing a complex chemical network featuring ${\rm H}_2$ formation and destruction. We track the formation of individual stars with mass $M>1\,{\rm M}_\odot$ with a stochastic recipe. Each star emits radiation according to its spectrum, sampled with 10 photon bins from near-infrared to extreme ultra-violet bands; winds are implemented by energy injection in the neighbouring cells. We run a simulation of a GMC with mass $M=10^5\,{\rm M}_\odot$, following the evolution of different gas phases. Thanks to the simultaneous inclusion of different stellar feedback mechanisms, we identify two stages in the cloud evolution: (1) radiation and winds carve ionized, low-density bubbles around massive stars, while FUV radiation dissociates most ${\rm H}_2$ in the cloud, apart from dense, self-shielded clumps; (2) rapid star formation (SFR$\simeq 0.1\,{\rm M}_\odot\,{\rm yr}^{-1}$) consumes molecular gas in the dense clumps, so that UV radiation escapes and ionizes the remaining HI gas in the GMC. ${\rm H}_2$ is exhausted in $1.6$ Myr, yielding a final star formation efficiency of 36 per cent. The average intensity of FUV and ionizing fields increases almost steadily with time; by the end of the simulation ($t=2.5$ Myr) we find $\langle G_0 \rangle \simeq 10^3$ (in Habing units), and a ionization parameter $\langle U_{\rm ion} \rangle \simeq 10^2$, respectively. The ionization field has also a more patchy distribution than the FUV one within the GMC. Throughout the evolution, the escape fraction of ionizing photons from the cloud is $f_{\rm ion, esc} < 0.03$.