GRRMHD Simulations of MAD Accretion Disks Declining from Super-Eddington to Sub-Eddington Accretion Rates

TL;DR

This study uses GRRMHD simulations to explore how MAD accretion disks around black holes behave as they transition from super-Eddington to sub-Eddington rates, revealing differences based on black hole spin and jet activity.

Contribution

First detailed GRRMHD simulations of MAD disks across a range of accretion rates for both non-spinning and spinning black holes, highlighting efficiency variations and jet dynamics.

Findings

Non-spinning BHs show increased efficiencies as accretion decreases.

Spinning BHs produce powerful jets that decline with accretion rate.

Magnetic flux eruptions cause variability and quasi-periodic behavior.

Abstract

We present two general relativistic radiation magnetohydrodynamics (GRRMHD) simulations of magnetically arrested disks (MADs) around non-spinning ($a_*=0$) and spinning ($a_*=0.9$) supermassive black holes (BHs). In each simulation, the mass accretion rate is decreased with time such that we sample Eddington-scaled rates over the range $3 \gtrsim \dot{M}/\dot{M}_{\rm{Edd}}\gtrsim 0.3$. For the non-spinning BH model, the total and radiative efficiencies increase as the accretion rate decreases, varying over the range $\eta_{\rm{tot}}\sim9-16\%$ and $\eta_{\rm{rad}}\sim6-12\%$, respectively. This model shows very little jet activity. In contrast, the spinning BH model has a strong relativistic jet powered by spin energy extracted from the BH. The jet power declines with accretion rate such that $\eta_{\rm{jet}}\sim 18-39\%$ while the total and radiative efficiencies are $\eta_{\rm{tot}}\sim 64-100\%$ and $\eta_{\rm{rad}}\sim 45-79\%$, respectively. We confirm that mildly sub-Eddington disks can extract substantial power from a spinning BH, provided they are in the MAD state. The jet profile out to $100\, GM/c^2$ is roughly parabolic with a power-law index of $k\approx0.43-0.53$ during the sub-Eddington evolution. Both models show significant variability in the outgoing radiation which is likely associated with episodes of magnetic flux eruptions. The $a_*=0.9$ model shows semi-regular variations with a period of $\sim2000\, GM/c^3$ over the final $\sim10,000\, GM/c^3$ of the simulation, which suggests that magnetic flux eruptions may be an important source of quasi-periodic variability. For the simulated accretion rates, the $a_*=0$ model is spinning up while the $a_*=0.9$ model is spinning down. Spinup-spindown equilibrium of the BH will likely be achieved at $0.5 < a_{*,{\rm{eq}}} < 0.6$, assuming continuous accretion in the MAD state.