University of Leicester
Nayakshin_2012_ApJ_753_15.pdf (229.88 kB)

The observed m-σ relations imply that super-massive black holes grow by cold chaotic accretion

Download (229.88 kB)
journal contribution
posted on 2012-10-24, 08:56 authored by Sergei Nayakshin, Chris Power, Andrew R. King
We argue that current observations of M-σ relations for galaxies can be used to constrain theories of super-massive black holes (SMBHs) feeding. In particular, assuming that SMBH mass is limited only by the feedback on the gas that feeds it, we show that SMBHs fed via a planar galaxy-scale gas flow, such as a disk or a bar, should be much more massive than their counterparts fed by quasi-spherical inflows. This follows from the relative inefficiency of active galactic nucleus feedback on a flattened inflow. We find that even under the most optimistic conditions for SMBH feedback on flattened inflows, the mass at which the SMBH expels the gas disk and terminates its own growth is a factor of several higher than the one established for quasi-spherical inflows. Any beaming of feedback away from the disk and any disk self-shadowing strengthen this result further. Contrary to this theoretical expectation, recent observations have shown that SMBHs in pseudobulge galaxies (which are associated with barred galaxies) are typically under- rather than overmassive when compared with their classical bulge counterparts at a fixed value of σ. We conclude from this that SMBHs are not fed by large (100 pc to many kpc) scale gas disks or bars, most likely because such planar flows are turned into stars too efficiently to allow any SMBH growth. Based on this and other related observational evidence, we argue that most SMBHs grow by chaotic accretion of gas clouds with a small and nearly randomly distributed direction of angular momentum.



Astrophysical Journal, 2012, 753 (1)


  • VoR (Version of Record)

Published in

Astrophysical Journal


American Astronomical Society, IOP Publishing





Copyright date


Available date


Publisher version



Usage metrics

    University of Leicester Publications