Multiphase, non-spherical gas accretion onto a black hole

TL;DR

This study uses 3D simulations to explore how gas accretion onto supermassive black holes varies with X-ray luminosity, revealing stable, unstable, and outflowing regimes driven by thermal instability and non-spherical structures.

Contribution

It demonstrates the role of thermal instability in creating non-spherical, filamentary accretion flows and outflows in a multiphase gas environment around black holes, extending previous 1D models.

Findings

Low L_X leads to stable spherical accretion.

High L_X causes gas heating and outflows.

Intermediate L_X induces thermal instability and filament formation.

Abstract

(Abridged) We investigate non-spherical behavior of gas accreting onto a central supermassive black hole performing simulations using the SPH code GADGET-3 including radiative cooling and heating by the central X-ray source. As found in earlier 1D studies, our 3D simulations show that the accretion mode depends on the X-ray luminosity (L_X) for a fixed density at infinity and accretion efficiency. In the low L_X limit, gas accretes in a stable, spherically symmetric fashion. In the high L_X limit, the inner gas is significantly heated up and expands, reducing the central mass inflow rate. The expanding gas can turn into a strong enough outflow capable of expelling most of the gas at larger radii. For some intermediate L_X, the accretion flow becomes unstable developing prominent non-spherical features, the key reason for which is thermal instability (TI) as shown by our analyses. Small perturbations of the initially spherically symmetric accretion flow that is heated by the intermediate L_X quickly grow to form cold and dense clumps surrounded by overheated low density regions. The cold clumps continue their inward motion forming filamentary structures; while the hot infalling gas slows down because of buoyancy and can even start outflowing through the channels in between the filaments. We found that the ratio between the mass inflow rates of the cold and hot gas is a dynamical quantity depending on several factors: time, spatial location, and L_X; and ranges between 0 and 4. We briefly discuss astrophysical implications of such TI-driven fragmentation of accreting gas on the formation of clouds in narrow and broad line regions of AGN, the formation of stars, and the observed variability of the AGN luminiosity.