Star Formation Regulation and Self-Pollution by Stellar Wind Feedback

TL;DR

This study uses simulations to show that stellar wind feedback, primarily momentum-driven due to cooling, effectively disperses gas and regulates star formation in turbulent clouds, with self-pollution being significant only in dense, extreme environments.

Contribution

The paper extends previous simulations to demonstrate that wind energy losses are dominated by turbulent mixing, confirming momentum-driven bubbles and quantifying self-pollution in various cloud densities.

Findings

Star formation completes within 2-8 free-fall times.

Wind energy losses are dominated by turbulent mixing layers.

Self-pollution is negligible in moderate-density clouds but significant in dense, super star cluster conditions.

Abstract

Stellar winds contain enough energy to easily disrupt the parent cloud surrounding a nascent star cluster, and for this reason have been considered candidates for regulating star formation. However, direct observations suggest most wind power is lost, and Lancaster21a,b recently proposed that this is due to efficient mixing and cooling processes. Here, we simulate star formation with wind feedback in turbulent, self-gravitating clouds, extending our previous work. Our simulations cover clouds with initial surface density $10^2-10^4$ $M_{\odot} \, {\rm pc}^{-2}$, and show that star formation and residual gas dispersal is complete within 2 - 8 initial cloud free-fall times. The "Efficiently Cooled" model for stellar wind bubble evolution predicts enough energy is lost for the bubbles to become momentum-driven, we find this is satisfied in our simulations. We also find that wind energy losses from turbulent, radiative mixing layers dominate losses by "cloud leakage" over the timescales relevant for star formation. We show that the net star formation efficiency (SFE) in our simulations can be explained by theories that apply wind momentum to disperse cloud gas, allowing for highly inhomogeneous internal cloud structure. For very dense clouds, the SFE is similar to those observed in extreme star-forming environments. Finally, we find that, while self-pollution by wind material is insignificant in cloud conditions with moderate density (only $\lesssim 10^{-4}$ of the stellar mass originated in winds), our simulations with conditions more typical of a super star cluster have star particles that form with as much as 1\% of their mass in wind material.