Bondi Flow from a Slowly Rotating Hot Atmosphere

Abstract

A supermassive black hole in the nucleus of an elliptical galaxy at the centre of a cool-core group or cluster of galaxies is immersed in hot gas. Bondi accretion should occur at a rate determined by the properties of the gas at the Bondi radius and the mass of the black hole. X-ray observations of massive nearby elliptical galaxies, including M87 in the Virgo cluster, indicate a Bondi accretion rate \(M_B\) which roughly matches the total kinetic power of the jets, suggesting that there is a tight coupling between the jet power and the mass accretion rate. While the Bondi model considers non-rotating gas, it is likely that the external gas has some angular momentum, which previous studies have shown could decrease the accretion rate drastically. We investigate here the possibility that viscosity acts at all radii to transport angular momentum outwards so that the accretion inflow proceeds rapidly and steadily. The situation corresponds to a giant advection-dominated accretion flow (ADAF) which extends from beyond the Bondi radius down to the black hole. We find solutions of the ADAF equations in which the gas accretes at just a factor of a few less than \(M_B\). These solutions assume that the atmosphere beyond the Bondi radius rotates with a sub-Keplerian velocity and that the viscosity parameter is large, α≥ 0.1, both of which are reasonable for the problem at hand. The infall time of the ADAF solutions is no more than a few times the free-fall time. Thus, the accretion rate at the black hole is closely coupled to the surrounding gas, enabling tight feedback to occur. We show that jet powers of a few per cent of \(M_{B}c^{2}\) are expected if either a fraction of the accretion power is channelled into the jet or the black hole spin energy is tapped by a strong magnetic field pressed against the black hole by the pressure of the accretion flow. We discuss the Bernoulli parameter of the flow, the role of convection and the possibility that these as well as magnetohydrodynamic effects may invalidate the model. If the latter comes to pass, it would imply that the rough agreement between observed jet powers and the Bondi accretion rate is a coincidence and jet power is determined by factors other than the mass accretion rate.