1999ApJS..125..439I

The major uncertainties involved in the Chandrasekhar mass models for Type Ia supernovae (SNe Ia) are related to the companion star of their accreting white dwarf progenitor (which determines the accretion rate and consequently the carbon ignition density) and the flame speed after the carbon ignition. We calculate explosive nucleosynthesis in relatively slow deflagrations with a variety of deflagration speeds and ignition densities to put new constraints on the above key quantities. The abundance of the Fe group, in particular of neutron-rich species like ⁴⁸Ca, ⁵⁰Ti, ⁵⁴Cr, ^54,58Fe, and ⁵⁸Ni, is highly sensitive to the electron captures taking place in the central layers. The yields obtained from such a slow central deflagration, and from a fast deflagration or delayed detonation in the outer layers, are combined and put to comparison with solar isotopic abundances. To avoid excessively large ratios of ⁵⁴Cr/⁵⁶Fe and ⁵⁰Ti/⁵⁶Fe, the central density of the ''average'' white dwarf progenitor at ignition should be as low as ≲2x10⁹ g.cm^–3. To avoid the overproduction of ⁵⁸Ni and ⁵⁴Fe, either the flame speed should not exceed a few percent of the sound speed in the central low Y_e layers or the metallicity of the average progenitors has to be lower than solar. Such low central densities can be realized by a rapid accretion as fast as M{dot}≳1x10^–7 M_☉.yr^–1. In order to reproduce the solar abundance of ⁴⁸Ca, one also needs progenitor systems that undergo ignition at higher densities. Even the smallest laminar flame speeds after the low-density ignitions would not produce sufficient amount of this isotope. We also found that the total amount of ⁵⁶Ni, the Si-Ca/Fe ratio, and the abundance of some elements like Mn and Cr (originating from incomplete Si burning), depend on the density of the deflagration-detonation transition in delayed detonations. Our nucleosynthesis results favor transition densities slightly below 2.2x10⁷ g.cm^–3.