Dynamics and Observational Signatures of Core-Collapse Supernovae with Central Engines: Hydrodynamics Simulations with Monte Carlo Post-Processing

TL;DR

The paper examines how long-lived central engines alter core-collapse supernova dynamics by inflating a central bubble that becomes Rayleigh-Taylor unstable and can rupture, accelerating outer ejecta and creating low-density channels. Using 2D hydrodynamics with magnetar- and accretion-powered energy injection and time-dependent Monte Carlo radiative transfer, it links engine properties to observable light curves and spectra, predicting rapid rise, high luminosities, and spectral evolution from hot, featureless early phases to broad-line Ic-like features, with possible narrow lines from central engine material. A key finding is that the observables are strongly shaped by the dimensionality and breakout geometry, enabling connections between fast blue optical transients (FBOTs), superluminous supernovae (SLSNe), and SNe Ic-BL, while highlighting the potential role of radio emission from high-velocity engine outflows. The work emphasizes that multidimensional effects and channelized breakout can significantly modify diffusion times and spectral evolution, motivating further 3D, NLTE, and radiation-hydrodynamics studies to fully capture the diversity of engine-powered CCSNe.

Abstract

A long-lived central engine embedded in expanding supernova ejecta can alter the dynamics and observational signatures of the event, producing an unusually luminous, energetic, and/or rapidly-evolving transient. We use two-dimensional hydrodynamics simulations to study the effect of a central energy source, varying the amount, rate, and isotropy of the energy deposition. We post-process the results with a time-dependent Monte Carlo radiation transport code to extract observational signatures. The engine excavates a bubble at the centre of the ejecta, which becomes Rayleigh-Taylor unstable. Sufficiently powerful engines are able to break through the edge of the bubble and accelerate, shred, and compositionally mix the entire ejecta. The breakout of the engine-driven wind occurs at distinct rupture points, and the outflowing high-velocity gas may eventually give rise to radio emission. The dynamical impact of the engine leads to faster rising optical light curves, with photon escape facilitated by the faster expansion of the ejecta and the opening of low-density channels. For models with strong engines, the spectra are initially hot and featureless, but later evolve to resemble those of broad-line Ic supernovae. Under certain conditions, line emission from ionized, low-velocity material near the centre of the ejecta may be able to escape and produce narrow emission similar to that seen in interacting supernovae. We discuss how variability in the engine energy reservoir and injection rate could give rise to a heterogeneous set of events spanning multiple observational classes, including the fast blue optical transients, broad-line Ic supernovae, and superluminous supernovae.

Figures (19)

• Figure 1: 1D angle-averaged density, radial velocity, and pressure profiles (from top to bottom) for our 1ms simulation run, plotted at different time points.
• Figure 2: Density maps showing the time-evolution of the 1ms simulation. The energy injection from the central engine inflates a bubble in the centre of the ejecta, filled with high-pressure gas. Rayleigh-Taylor instabilities develop at the edge of the bubble, eventually causing it to rupture at discrete points along its surface. The gas contained in the bubble vents out through these rupture points and leaves behind low-density channels in the remnant.
• Figure 3: Shock position (in velocity space) vs. time for our 1ms model. The solid lines show the median shock position and maximum shock position across all angles; the difference between the two is a result of the growth of instabilities and the eventual breakout of the gas from the central cavity (the top line corresponding to the maximum shock position tracks the initial breakout). The dashed line shows a power-law fit to the shock position using data from $t < 0.5~t_{\rm eng}$. The dotted line is the analytical prediction for a 1D model given by Equation \ref{['eq:R_sh_dimensionless']}.
• Figure 4: Angular variation in estimated column density (assuming pure hydrogen composition) for the 1ms simulation, rescaled for each time point to ease comparison. The column density is calculated along rays of length equal to the radial extent of the domain. As the breakout is approximately symmetric, we only show variation over the top half of the domain. We can see that strong variation in column density (by several orders of magnitude) develops due to gas breakout out of the central cavity and forming channels in the outer ejecta.
• Figure 5: Distribution of total energy with velocity for our 1ms model at 4 different time points (including the initial and final step). This demonstrates the acceleration of the outer ejecta due to the gas venting out of the central bubble. The interior of the central cavity is excluded when calculating the distribution.