Supernova Type Ia progenitors from merging double white dwarfs

Using a new population synthesis model^⋆

¹ Department of Astrophysics/IMAPPRadboud University Nijmegen, PO Box 9010 6500 GL, Nijmegen, The Netherlands
e-mail: s.toonen@astro.ru.nl
² Instituut voor Sterrenkunde, KU Leuven, Celestijnenstraat 200D, 3001 Leuven, Belgium
³ Leiden Observatory, Leiden University, PO Box 9513, 2300 RA Leiden, The Netherlands

Received: 6 February 2012
Accepted: 20 July 2012

Abstract

Context. The study of Type Ia supernovae (SNIa) has lead to greatly improved insights into many fields in astrophysics, e.g. cosmology, and also into the metal enrichment of the universe. Although a theoretical explanation of the origin of these events is still lacking, there is a general consensus that SNIa are caused by the thermonuclear explosions of carbon/oxygen white dwarfs with masses near the Chandrasekhar mass.

Aims. We investigate the potential contribution to the supernova Type Ia rate from the population of merging double carbon-oxygen white dwarfs. We aim to develop a model that fits the observed SNIa progenitors as well as the observed close double white dwarf population. We differentiate between two scenarios for the common envelope (CE) evolution; the α-formalism based on the energy equation and the γ-formalism that is based on the angular momentum equation. In one model we apply the α-formalism throughout. In the second model the γ-formalism is applied, unless the binary contains a compact object or the CE is triggered by a tidal instability for which the α-formalism is used.

Methods. The binary population synthesis code SeBa was used to evolve binary systems from the zero-age main sequence to the formation of double white dwarfs and subsequent mergers. SeBa has been thoroughly updated since the last publication of the content of the code.

Results. The limited sample of observed double white dwarfs is better represented by the simulated population using the γ-formalism for the first CE phase than the α-formalism. For both CE formalisms, we find that although the morphology of the simulated delay time distribution matches that of the observations within the errors, the normalisation and time-integrated rate per stellar mass are a factor  ~7−12 lower than observed. Furthermore, the characteristics of the simulated populations of merging double carbon-oxygen white dwarfs are discussed and put in the context of alternative SNIa models for merging double white dwarfs.