Stellar evolution on the borderline of white dwarf and neutron star formation

Abstract

This thesis is about the evolution of stars, specifically about the final fate of stars at the borderline between the formation of white dwarfs and neutron stars. It is well known that the mass and the metallicity are the two determining factors in stellar evolution, and for a given initial chemical composition, the mass essentially defines the final fate: Lower mass stars produce white dwarfs, while higher mass stars produce neutron stars and supernovae. The transition region between white dwarf and neutron star formation, however, is not well studied, although it may contain almost half of the total number of supernovae. First, we study the evolution of stars in the transition region at solar metallicity. These stars resemble a configuration similar to asymptotic giant branch (AGB) stars that are lower in mass, with a massive envelope surrounding a degenerate Oxygen-Neon core, hence we call them Super-AGB stars. Depending on the competition between the core growth rate and the mass loss rate, the core grows to a critical mass and explodes due to electron capture on Mg24 or Ne20, or the star looses its massive envelope, leaving a massive white dwarf. We find that SAGB stars populate the mass range between 7.5 and 9.25 solar mass, which depends, however, significantly on the treatment of semiconvective mixing and convective overshooting. The initial mass range that produces electron-capture supernovae is found to be between 9 and 9.25 solar mass, which accounts for 4% of all supernovae in the local universe. Second, we investigate the possible supernova channel from massive AGB stars at lower metallicities. We find that the channel widens from 0.25 solar mass at solar metallicity to almost 2 solar masses at 1/1000 of solar metallicity. A corresponding shift of the minimum initial mass for electron capture supernovae from 9 solar mass to 6.3 solar mass at 1/1000 of solar metallicity implies a doubling of the total supernova rate, with an electron capture supernova fraction of about 50%. We discuss observational consequences of solar metallicity and metal poor electron capture supernovae, their mass loss history and pre-supernova luminosity, and in particular the relevance of these SAGB stars to understand observed s- and r-enhancements in extremely metal poor stars, and the large number of neutron stars found in Galactic globular clusters. Third we study the evolution of SAGB stars to calculate their chemical yields and their relevance to the chemical evolution of the universe, which we conclude to be not important. Fourth, we investigate the possibility and consequences of an electron capture supernova in a binary system. We find that the mass range for which electron capture supernovae can occur will be likely much wider than for single stars, and we suggest that the collapse of the core leads to a prompt or fast explosion, rather than a very slow, delayed neutrino-driven explosion, and that this naturally produces neutron stars with low-velocity kicks. This leads to a dichotomous distribution of neutron star kicks, which might explain the observed dichotomous distribution observed in the local universe.