Delayed radio emission in tidal disruption events from collisions of outflows driven by disk instabilities

TL;DR

This work tackles the origin of delayed radio emission in tidal disruption events (TDEs) by proposing collisions of disk-instability–driven outflows as a core mechanism. It develops a hydrodynamic and radiative framework for two collision channels—Flare+Flare and Flare+CNM—characterized by outflow masses $M_{\rm fl}$, velocity range $v_{\min}$–$v_{\max}$, launch delay $\Delta t$, and circumnuclear medium slope $\alpha$, and computes synchrotron emission with SSA and FFA, using $p=3$, $\epsilon_e=\epsilon_B=0.1$, and $\gamma_{\min}=1$. The model yields shock speeds $v_{\rm sh}\sim 0.05$–$0.3c$ at $\sim 10^{17}$ cm and peak radio luminosities $L_{\nu,obs}\sim 10^{27}$–$10^{30}$ erg s$^{-1}$ Hz$^{-1}$ at 6 GHz, consistent with observed delayed radio emission; Flare+Flare scenarios produce rapid, multi-peak light curves, while Flare+CNM scenarios yield longer-lived evolution depending on the CNM density profile. Comparisons with a sample of TDEs show that several events are well matched in both light curves and SEDs, supporting the idea that delayed outflows from disk instabilities and state transitions can power late-time radio emission. The results also highlight the need for more late-time, multi-frequency radio data to distinguish among flare-collision, CIO, CNM interaction, and jet-related scenarios, and to constrain disk-physics and CNM properties in TDE environments.

Abstract

Delayed radio emission has been associated with a growing proportion of tidal disruption events (TDEs). For many events, the radio synchrotron emission is inferred to originate from the interaction of mildly-relativistic outflows, launched with delay times of $\sim 100$--$1000$ d after the TDE optical peak. The mechanism behind these outflows remains uncertain, but may relate to instabilities or state transitions in the accretion disk formed from the TDE. We model the radio emission powered by the collision of mass outflows ("flares") from TDE accretion disks, considering scenarios in which two successive disk flares collide with each other, as well as collisions between the outflow and the circumnuclear medium (CNM). For flare masses of $\sim 0.01$-$0.1 M_{\odot}$, varied CNM densities, and different time intervals between ejected flares, we demonstrate that the shocks formed by the collisions have velocities $0.05c$-$0.3c$ at $\sim 10^{17}$ cm and power bright radio emission of $L_ν \sim 10^{27}$-$10^{30}$ erg s$^{-1}$ Hz$^{-1}$, consistent with the properties inferred for observed events. We quantify how the typical peak timescale and flux varies for different properties of our models, and compare our model predictions to a selection of TDEs with delayed radio emission. Our models successfully reproduce the light curves and SEDs for several events, supporting the idea that delayed outflows from disk instabilities and state transitions can power late-time radio emission in TDEs.

Figures (11)

• Figure 1: Evolution of the shock velocity and density ahead of the forward shock as a function of shock radius, for a set of flares with flare mass $M_{\rm fl} = 0.01\, M_{\odot}$ and flare velocity range of $v_{\rm min} = 0.04 c$, $v_{\rm max} = 0.4 c$. The time interval $\Delta t$ between two flares is varied in the first column (Flare+Flare), and in the last two columns, the normalization of the circumnuclear medium (CNM) density $n_0$ is varied.
• Figure 2: Same as Figure \ref{['fig:M0p01_shellevol']}, for flare mass $M_{\rm fl} = 0.1\, M_{\odot}$.
• Figure 3: Light curves of radio emission at 6 GHz for the models shown in Figure \ref{['fig:M0p01_shellevol']} (top row) and Figure \ref{['fig:M0p1_shellevol']} (bottom row).
• Figure 4: Peak properties of our model light curves at 6 GHz. The left panel shows the peak luminosity versus the rise time, and the right panel shows the peak luminosity versus the decay time; these are defined in Section \ref{['sec:peakprops']}. The scatter points correspond to Flare+Flare in green, Flare+$\alpha2.5$CNM in blue, and Flare+$\alpha0$CNM in red; circles are $M_{\rm fl}=0.01\, M_{\odot}$ models, while squares are $M_{\rm fl} =0.1\, M_{\odot}$ models. Darker shades indicate higher CNM density for red and blue points, or larger $\Delta t$ for green points. To show the extent of parameter space, we include more models than are listed in Table \ref{['tab:allmodels']} or shown in Figures \ref{['fig:M0p01_shellevol']}--\ref{['fig:allmodelLCs']}. Estimates of the peak flux and rise or decay time are also shown for a selection of observed TDEs, listed in the legend in the right panel. Lower limits indicate events where the peak flux is not constrained, and upper or lower limits to the rise and decay times are also shown for events where the time to reach 10% of the peak flux is uncertain; however, the length of the error bar is not representative of the actual uncertainty and is shown for illustrative purposes only.
• Figure 5: Top: Comparison of a selected model with radio emission from the TDE ASASSN-15oi. The model light curves shown are at 3 GHz. Bottom: Comparison of a selected model with radio emission from the TDE AT2020vwl. The model light curves shown are at 6 GHz.