Effects of neutrino-driven kicks on the supernova explosion mechanism

TL;DR

This work investigates whether anisotropic neutrino emission during core-collapse can impart natal kicks to the proto-neutron star and influence the explosion energy. Using 3D simulations of a $15\,M_\odot$ progenitor with an artificial kick mechanism, the authors show that, in marginal cases, neutrino-driven kicks can revive convection and cause asymmetric explosions with stronger ejecta in the kick direction. The findings contrast with ejecta-driven kicks, which would produce opposite ejecta motion, and yield observable signatures such as pulsar spin-kick alignment and directional mixing, with potential gravitational-wave signals as further tests. The study provides a framework to connect neutrino physics (including oscillations and sterile neutrinos) to supernova explosion outcomes and remnant morphologies, offering concrete diagnostics for future observations.

Abstract

We show that neutrino-driven pulsar kicks can increase the energy of the supernova shock. The observed large velocities of pulsars are believed to originate in the supernova explosion, either from asymmetries in the ejecta or from an anisotropic emission of neutrinos (or other light particles) from the cooling neutron star. In this paper we assume the velocities are caused by anisotropic neutrino emission and study the effects of these neutrino-driven kicks on the supernova explosion. We find that if the collapsed star is marginally unable to produce an explosion, the neutrino-driven mechanisms can drive the convection to make a successful explosion. The resultant explosion is asymmetric, with the strongest ejecta motion roughly in the direction of the neutron star kick. This is in sharp contrast with the ejecta-driven mechanisms, which predict the motion of the ejecta in the opposite direction. This difference can be used to distinguish between the two mechanisms based on the observations of the supernova remnants.

Figures (8)

• Figure 1: Comparison of the symmetric "HVisc0" (left) and kick "HVisc100" (right) models 90 ms after bounce. These plots show a slice of data centered on the z=0 plane with the kick in the positive x direction. Shading denotes entropy (dark is low, light is high) and the direction and length of arrows denote the direction and magnitude of the velocity. Note that the kicked model has developed some strong convection which is pushing out the accretion shock. It ultimately developes into a strong explosion (Fig. 3). Primarily because of the low resolution, convection does not develop in the symmetric model and this model does not explode.
• Figure 2: Entropy versus radius 40 ms after bounce for models using the coupled equation of state from Her94 (light particles) compared with those using the Lattimer-Swesty Lat91 equation of state down to low densities (dark particles). The vertical lines correspond to the positions of the accretion shock for these two models. Note that although the entropy from the coupled equation of state is lower than that using Lattimer-Swesty down to low densities, the entropy gradient out to the shock is much higher. This is more conducive to convection. The entropy gradient across the shock is also much more gradual in the case of the coupled equation of state. This is because nuclear dissociation and burning is playing a strong role in determining the entropy.
• Figure 3: Radial velocity versus radius for our symmetric (HVisc0) and kicked (HVisc100) models, showing clearly the lack of convection in the symmetric simulation in stark contrast to the convection that has developed in the kicked model. For all practical purposes, the symmetric model is behaving as one might expect in a 1-dimensional simulation.
• Figure 4: Slice of the kicked simulation "HVisc100" 160 ms after bounce. An explosion has been launched and, where it is strongest, has now nearly reached 2000 km. The explosion ejecta is strongest in roughly the same direction as the neutron star. Because the explosion ejecta is driven by convection, and the seeds for this convection are building on small asymmetries in the collapsing core, the fastest moving ejecta is not exactly aligned with the motion of the kick. Ejecta driven kicks predict the exact opposite - the ejecta moves in the opposite direction of the neutron star.
• Figure 5: Possible causes for the ejecta asymmetry: i) asymmetric neutrino heating or ram-heating caused by the motion of the proto-neutron star, ii) weakened pressure at the accretion shock, or iii) convection seeded by the motion of the proto-neutron star (possibly caused by minor effects of the previous two effects). Our analysis suggests that this latter cause is indeed the cause of the ejecta asymmetry in our exploding models.