Understanding r-process Nucleosynthesis in the Milky Way

Abstract

Elements heavier than Fe are formed by neutron capture processes when fusion becomes energetically unfavourable; the slow s-process and site are reasonably well understood, but the rapid r-process site is still a highly debated topic. In this thesis I will discuss the current best understanding of both the s-process and r-process, including potential sites. I also discuss the modelling of galaxies using smoothed particle hydrodynamics simulations with the inclusion of nucleosynthesis models to simulate the chemical evolution of galaxies. I then present the results of such chemodynamical simulations including nucleosynthesis yields for neutron star mergers, magneto-rotational supernovae, electron capture supernovae and neutrino driven winds. Using the [Eu/(Fe, α)] - [Fe/H] relation I show the neutron star mergers are unlikely to be able to drive r-process enrichment in the early universe but that magneto-rotational supernovae, or a combination of sources including them, may be able to. I then include a metallicity dependence in the magneto-rotational supernova model, and show that a combined model with neutron star mergers and electron-capture supernovae gives an excellent match to observations of [(Eu, Nd, Dy, Er, Zr)/(Fe, α)]. Finally I discuss the effects of supernova feedback on chemical evolution. I compare four models: a thermal model, a thermal model with a kinetic component, a stochastic model and a mechanical model and show that the kinetic, stochastic and mechanical models can suppress the star formation within isolated dwarf disc galaxies when using optimal parameters and that this has little effect on the fraction of metals ejected from the galaxy.