PI: Matthew Middleton

Our first set of 3D GR-RMHD simulation results for aligned super-critical accretion onto stellar mass black holes was presented in Fragile et al. (2025) (https://ui.adsabs.harvard.edu/abs/2025MNRAS.540.2820F/abstract).

The simulations revealed hitherto unexpected behaviour – the accretion rate onto the black hole itself (which is orders of magnitude less than at very large distances from the black hole due to radiatively driven mass loss), is limited to the Eddington rate. This is distinctly unlike simulation results from other groups where inwards radial advection allows the black hole to grow at super (perhaps even hyper)-Eddington rates. We believe the difference is due to our unique numerical set-up; whilst other groups start with a massive torus in pressure equilibrium at small radii (and allow it to become MRI unstable leading to accretion at very high rates), we instead initialise a disc out to very large radii, turn up the accretion rate and let it evolve for a long time. Our result calls into question the assumption that black holes can grow rapidly via advection, and presents a challenge for explaining the detection of massive supermassive black holes at high redshift (z > 7); our continuing work, made possible by DiRAC will seek to solve this problem.