Modelling the fragmentation of protostellar cores with ambipolar diffusion

Abstract

Most stars are observed in multiple star systems, which are thought to result from the fragmentation of a protostellar cores as they collapse. This process has been studied extensively using numerical simulations but, to date, relatively few of these account for the effects of magnetic fields. Of those that do, almost all work under the assumption of ideal magnetohydrodynamics (i.e. the gas is completely ionized). However, the gas in a protostellar core would only be very weakly ionized and would thus tend to drift relative to magnetic field lines, in a process called ambipolar diffusion. In this thesis, an attempt was made to implement ambipolar diffusion in the ZEUS-3D magnetohydrodynamics code, so as to investigate the effects of ambipolar diffusion on the fragmentation process. While the implementation worked well for 1D tests, it was found to produce instabilities when run in 3D while solving for total energy. Simulations could be run without instabilities when solving for internal energy, but ambipolar diffusion required such a small time-step as to make all but the lowest resolution simulations impractical. What simulations could be run indicated that ambipolar diffusion does, indeed, produce noticeably different results from those obtained with ideal MHD or ignoring magnetic fields altogether.