Fast giant flares in discs around supermassive black holes

TL;DR

This work addresses the origin of disruptive nuclear outbursts by examining the thermal stability of non-self-gravitating turbulent $α$-discs around SMBHs. The authors compute disc equilibria and show that, under an ionisation-driven change in the turbulent parameter $α$, a cold disc ring can heat into a hot, advection-dominated state, triggering fast, giant flares for SMBHs with $M_ ext{BH}\sim 10^6-10^8\,M_\odot$. The mechanism relies on the overlap of hydrogen-ionisation and radiation-pressure instabilities to produce a direct transition to a geometrically thick, super-Eddington flow, accompanied by anisotropic outflows and UV/optical emission approaching several $L_\text{Edd}$. If confirmed by numerical simulations, these giant flares could resemble TDEs in energetics and light-curve timescales but would offer a distinct outflow-dominated observational signature and provide a new channel for energetic transients in galactic nuclei.

Abstract

We studied the thermal stability of non-self-gravitating turbulent $α$-discs around supermassive black holes (SMBHs) to test a new type of high-amplitude galactic nucleus flares. By calculating the disc structures, we computed the critical points of equilibrium curves for discs around SMBHs, which cover a wide range of accretion rates and resemble the shape $ξ$. We find that a transition of a disc ring from a recombined cold state to a hot, fully ionised, advection dominated, geometrically thick state is possible. Such a transition can trigger a giant flare for SMBHs with masses $\sim 10^6-10^8\, M_\odot$ if the prior geometrically thin and optically thick disc surrounded a central radiatively inefficient accretion flow. An increase in the viscosity parameter $α$ is a necessary condition for this scenario. This increase may be related to the fact that the magnetic Prandtl number increases and exceeds 1 during ionisation. When self-gravity effects in the disc are negligible, the duration and power of the flare exhibit a positive correlation with the prior truncation radius of the geometrically thin disc. According to our estimates, the mass of about $\sim 4-3000\, M_\odot$ can be involved in the giant flare lasting 1 to 400 years if the flare is triggered somewhere between $60$ and $600$ gravitational radii from the SMBH of $10^7\, M_\odot$. The accretion rate on the SMBH peaks about 10 times faster at the potentially super-Eddington level. An optically thick outflow leads to anisotropy of the emission. At the beginning of the giant flare, the region near the truncation radius is heated to $\sim 10^5\,$K, and its UV/optical luminosity is at least $\sim 0.3-4 \,L_\mathrm{Edd}$ depending on the SMBH mass. The sudden heating of a cold disc around a SMBH can trigger a massive outburst, similar in appearance to what is proposed to occur after a tidal disruption event.

Figures (14)

• Figure 1: General S-curve at some radius in accretion disc. The limit cycle instability is shown schematically: on the lower branch the mass accumulates, on the upper branch the mass flows out of a ring.
• Figure 2: S-curves at different radii $r_1<r_2<r_3$, shown schematically for ionisation instability. When the accretion rate is very low, the whole disc is stable and cold (the bottom line on the plot). If the surface density exceeds the critical value at the innermost radius $r_1$, a heating wave, indicated by arrows, starts outwards, potentially reaching $r_3$.
• Figure 3: Equilibrium curves for discs around stellar-mass BH (left panel) and SMBH (right panel). In each panel, the $\xi$-curves are constructed for two values of the turbulent parameter $\alpha$: 0.1 and 0.01. The solid lines are calculated by AlphaDisc and their upper dotted tails correspond to $P_{\rm rad}>P_{\rm gas}$. Analytic relations in the radiation-pressure regime (A-zone, dashed, $\propto \Sigma^{-1}$) and the advective regime (dash-dotted, $\propto \Sigma$) are calculated using Eqs. \ref{['eq.A_zone']} and \ref{['eq.Adv_zone']}, respectively. The dots mark critical points for the geometrically thin states: the minimum and the maximum surface density, $\Sigma^+$ and $\Sigma_A$, in the ionised state and the maximum surface density $\Sigma^-$ in the recombined state. The arrows schematically show the direction of ring heating due to ionisation instability. At the same time, $\alpha$ increases.
• Figure 4: S-curves for $10^7 \, M_{\odot}$ at $r=100\,R_\mathrm{S}$ for two values of turbulent parameter $\alpha_\mathrm{cold}=0.01$ (blue) and $\alpha_\mathrm{hot}=0.1$ (orange). The branches 'A-zone' and 'Advection' were calculated using the code from Lipunova1999.
• Figure 5: Critical values of surface density \ref{['eq.Sigma_R']} for $M=10^7\, M_{\odot}$. Viscous parameter is $\alpha_\mathrm{cold}=0.01$ for $\Sigma^-$ (the blue line) and $\alpha_\mathrm{hot}=0.1$ in all other dependencies (orange lines). Relations $\Sigma_\mathrm{A}$ and $\Sigma_\mathrm{A}'$ show results of different codes, see Sect. \ref{['s.points']}.