On the rates of type Ia supernovae originating from white dwarf collisions in quadruple star systems

TL;DR

This study models the evolution of quadruple star systems to estimate the rates of type Ia supernovae caused by white dwarf collisions, finding significantly lower rates than observed, mainly due to early interactions preventing WD formation.

Contribution

It provides the first detailed simulations of quadruple systems considering secular, stellar, tidal evolution, and encounters, to estimate their contribution to SNe Ia rates.

Findings

Approximately 0.4-0.6 of systems interact before WD formation.

Computed SNe Ia rates are three orders of magnitude lower than observed.

Delay-time distributions are flatter than observed DTDs.

Abstract

We consider the evolution of stellar hierarchical quadruple systems in the 2+2 (two binaries orbiting each other's barycentre) and 3+1 (triple orbited by a fourth star) configurations. In our simulations, we take into account the effects of secular dynamical evolution, stellar evolution, tidal evolution and encounters with passing stars. We focus on type Ia supernovae (SNe Ia) driven by collisions of carbon-oxygen (CO) white dwarfs (WDs). Such collisions can arise from several channels: (1) collisions due to extremely high eccentricities induced by secular evolution, (2) collisions following a dynamical instability of the system, and (3) collisions driven by semisecular evolution. The systems considered here have initially wide inner orbits, with initial semilatus recti larger than 12 AU, implying no interaction if the orbits were isolated. However, taking into account dynamical evolution, we find that $\approx 0.4$ ($\approx 0.6$) of 2+2 (3+1) systems interact. In particular, Roche Lobe overflow can be triggered possibly in highly eccentric orbits, dynamical instability can ensue due to mass-loss-driven orbital expansion or secular evolution, or a semisecular regime can be entered. We compute the delay-time distributions (DTDs) of collision-induced SNe Ia, and find that they are flatter compared to the observed DTD. Moreover, our combined SNe Ia rates are $(3.7\pm0.7) \times 10^{-6} \,\mathrm{M}_\odot^{-1}$ and $(1.3\pm0.2) \times 10^{-6} \,\mathrm{M}_\odot^{-1}$ for 2+2 and 3+1 systems, respectively, three orders of magnitude lower compared to the observed rate, of order $10^{-3} \, \mathrm{M}_\odot^{-1}$. The low rates can be ascribed to interactions before the stars evolve to CO WDs. However, our results are lower limits given that we considered a subset of quadruple systems.