The Collapse of Rotating Massive Stars in Three Dimensions

Abstract

Most simulations of the core collapse of massive stars have focused onthe collapse of spherically symmetric objects. If these stars arerotating, this symmetry is broken, opening up a number of effects thatare just now being studied. The list of proposed effects spans a rangeof extremes: from fragmentation of the collapsed iron core tomodifications of the convective instabilities above the core; from thegeneration of strong magnetic fields that then drive the supernovaexplosion to the late-time formation of magnetic fields to producemagnetars after the launch of the supernova explosion. The list of theobservational effects of rotation ranges from modifications in thegamma-ray line spectra, nucleosynthetic yields, and shape of supernovaremnants caused by rotation-induced asymmetric explosions to strongpulsar radiation, the emission of gravitational waves, and alteredr-process nucleosynthetic yields caused by rapidly rotating stars. Inthis paper we present the results of three-dimensional collapsesimulations of rotating stars for a range of stellar progenitors. Wefind that for the most rapidly spinning stars, rotation does indeedmodify the convection above the proto-neutron star, but it is not fastenough to cause core fragmentation. Similarly, although strong magneticfields can be produced once the proto-neutron star cools and contracts,the proto-neutron star does not spin fast enough to generate strongmagnetic fields quickly after collapse, and, for our simulations,magnetic fields will not dominate the supernova explosion mechanism.Even so, the resulting pulsars for our most rapidly rotating models mayemit enough energy to dominate the total explosion energy of thesupernova. However, more recent stellar models predict rotation ratesthat are much too slow to affect the explosion, but these models are notsophisticated enough to determine whether the most recent or paststellar rotation rates are more likely. Thus, we must rely onobservational constraints to determine the true rotation rates ofstellar cores just before collapse. We conclude with a discussion of thepossible constraints on stellar rotation that we can derive fromcore-collapse supernovae.