Triggering star formation by both radiative and mechanical active galactic nucleus feedback

TL;DR

This study uses hydrodynamic simulations to demonstrate that combined radiative and mechanical feedback from active galactic nuclei can trigger significant star formation by compressing the interstellar medium into dense, unstable shells that fragment into star-forming regions.

Contribution

First to incorporate both mass outflow and radiative feedback in simulations showing their combined role in triggering star formation in AGN environments.

Findings

Star formation rate can be very high in compressed shells.

Star formation rate density exceeds previous estimates due to neglect of stellar disruption.

Mass outflow feedback is dominant in gas-poor scenarios.

Abstract

We perform two dimensional hydrodynamic numerical simulations to study the positive active galactic nucleus feedback which triggers, rather than suppresses, star formation. Recently, it was shown by Nayakshin et al. and Ishibashi et al. that star formation occurs when the cold interstellar medium is squeezed by the impact of mass outflow or radiation pressure, respectively. Mass outflow is ubiquitous in this astrophysical context, and radiation pressure is also important if the AGN is luminous. For the first time in this subject, we incorporate both mass outflow feedback and radiative feedback into our model. Consequently, the ISM is shocked into shells by the AGN feedback, and these shells soon fragment into clumps and filaments because of Rayleigh-Taylor and thermal instabilities. We have two major findings: (1) the star formation rate can indeed be very large in the clumps and filaments. However, the resultant star formation rate density is too large compared with previous works, which is mainly because we ignore the fact that most of the stars that are formed would be disrupted when they move away from the galactic center. (2) Although radiation pressure feedback has a limited effect, when mass outflow feedback is also included, they reinforce each other. Specifically, in the gas-poor case, mass outflow is always the dominant contributor to feedback.