Type Ia supernova diversity: white dwarf central density as a secondary parameter in three-dimensional delayed detonation models

Delayed detonations of Chandrasekhar mass white dwarfs (WDs) have been very successful in explaining the spectra, light curves and the width-luminosity relation of spectroscopically normal Type Ia supernovae (SNe Ia). The ignition of the thermonuclear deflagration flame at the end of the convective carbon 'simmering' phase in the core of the WD is still not well understood, and much about the ignition kernel distribution remains unknown. Furthermore, the central density at the time of ignition depends on the still uncertain screened carbon fusion reaction rates, the accretion history and cooling time of the progenitor, and the composition. We present the results of 12 high-resolution three-dimensional delayed detonation SN Ia explosion simulations that employ a new criterion to trigger the deflagration to detonation transition (DDT). The simulations fall into three ignition categories: relatively bright SNe with five ignition kernels and a weak deflagration phase (three different central densities); relatively dim SNe with 1600 ignition kernels and a strong deflagration phase (three different central densities) and intermediate SNe with 200 ignition kernels (six different central densities). All simulations trigger our DDT criterion and the resulting delayed detonations unbind the star. We find a trend of increasing iron group element (IGE) production with increasing central density for all three categories. The total Ni-56 yield, however, remains more or less constant, even though increased electron captures at high density result in a decreasing Ni-56 mass fraction of the IGE material. We attribute this to an approximate balance of Ni-56 producing and destroying effects. The deflagrations that were ignited at higher density initially have a faster growth rate of subgrid-scale turbulence. Hence, the effective flame speed increases faster, which triggers the DDT criterion earlier, at a time when the central density of the expanded star is higher. This leads to an overall increase of IGE production, which offsets the percental reduction of Ni-56 due to neutronization.