Runaway stars as progenitors of supernovae and gamma-ray bursts

John J. Eldridge^1, Norbert Langer^2,3 & Christopher A. Tout^1

(1) Institute of Astronomy, The Observatories, University of Cambridge, Madingley Road, Cambridge, CB3 0HA.
(2) Argelander-Institut f"ur Astronomie, Bonn University, Auf dem H"ugel 71, 53121 Bonn, Germany
(3) Astronomical Institute, University of Utrecht, Postbus 80000, 3508 TA Utrecht, The Netherlands

When a core collapse supernova occurs in a binary system, the surviving star as well as the compact remnant emerging from the supernova, may reach a substantial space velocity. With binary population synthesis modelling at solar and one fifth of solar metallicity, we predict the velocities of such runaway stars or binaries. We compile predictions for runaway OB stars, red supergiants and Wolf-Rayet stars, either isolated or with a compact companion. For those stars or binaries which undergo a second stellar explosion we compute their further evolution and the distance travelled until a Type~II or Type~Ibc supernova or a long or short gamma-ray burst occurs. We find our predicted population of OB runaway stars broadly matches the observed population of stars but, to match the fastest observed Wolf-Rayet runaway stars, we require that black holes receive an asymmetric kick upon formation. We find that at solar metallicity Type~Ic supernova progenitors travel shorter distances than the progenitors of other supernova types because they are typically more massive and thus have shorter lifetimes. Those of Type~IIP supernovae can fly farthest about 48,pc {em on average} at solar metallicity, with about 8 per cent of them reaching 100,pc. In considering the consequences of assuming that the progenitors of long gamma-ray bursts are spun-up secondary stars that experience quasi-homogeneous evolution, we find that such evolution has a dramatic effect on the population of runaway Wolf-Rayet stars and that some 30 per cent of GRBs could occur a hundred parsecs or more from their initial positions. We also consider mergers of double compact object binaries consisting of neutron stars and/or black holes. We find the most common type of visible mergers are neutron star--black hole mergers that are roughly ten times more common than neutron star--neutron star mergers. All compact mergers have a wide range of merger times from years to Gyrs and are predicted to occur three hundred times less often than supernovae in the Milky Way. We also find that there may be a population of low-velocity neutron stars that are ejected from a binary rather than by their own natal kick. These neutrons stars need to be included when the distribution of neutron
star kicks is deduced from observations.

Reference: MNRAS in press
Status: Manuscript has been accepted