paper

A three-dimensional deflagration model for Type Ia supernovae confronted with observations

arXiv:0707.1024 · doi:10.1086/521347

Abstract

A simulation of the thermonuclear explosion of a Chandrasekhar-mass C+O white dwarf, the most popular scenario of a type Ia supernova (SN Ia), is presented. The underlying modeling is pursued in a self-consistent way, treating the combustion wave as a turbulent deflagration using well tested methods developed for laboratory combustion and based on the concept of `large eddy simulations' (LES). Such consistency requires to capture the onset of the turbulent cascade on resolved scales. This is achieved by computing the dynamical evolution on a 1024 moving grid, which resulted in the best-resolved three-dimensional SN Ia simulation carried out thus far, reaching the limits of what can be done on present supercomputers. Consequently, the model has no free parameters other than the initial conditions at the onset of the explosion, and therefore it has considerable predictive power. Our main objective is to determine to which extent such a simulation can account for the observations of normal SNe Ia. Guided by previous simulations with less resolution and a less sophisticated flame model, initial conditions were chosen that yield a reasonably strong explosion and a sufficient amount of radioactive nickel for a bright display. We show that observables are indeed matched to a reasonable degree. In particular, good agreement is found with the light curves of normal SNe Ia. Moreover, the model reproduces the general features of the abundance stratification as inferred from the analysis of spectra. This indicates that it captures the main features of the explosion mechanism of SNe Ia. However, we also show that even a seemingly best-choice pure deflagration model has shortcomings that indicate the need for a different mode of nuclear burning at late times, perhaps the transition to a detonation at low density.

8 pages, 5 figures, accepted for publication in the ApJ

References in corpus (31)

Cited by in corpus (106)