Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 318, 812-818 (1997)

Previous Section Next Section Title Page Table of Contents

1. Introduction

A star is classified as a runaway star when its peculiar velocity [FORMULA] is larger than 30 [FORMULA]. With this definition the observations reveal that [FORMULA] 10% of all O-type stars are runaways (Gies and Bolton, 1986; Gies, 1987).

We know of two processes able to produce runaway stars: close encounters in dense clusters (Poveda et al., 1967; van Albada, 1968; Aarseth and Lecar, 1975; Lada et al., 1984) and the supernova (SN) explosion in massive close binaries (MCBs) (Blaauw, 1961). Gies and Bolton favored the cluster ejection process, based on the fact that a number of O-type runaways are normal non-evolved close binaries. Furthermore in the '80s it was accepted that the SN ejecta did not deviate much from spherical symmetry, and thus most of the runaways produced by the binary scenario were supposed to have a compact companion. The detailed study of Gies and Bolton did not show any signature of the presence of a compact companion in their remaining set of single line O-type runaways.

Following the qualitative scenario for MCB evolution of van den Heuvel and Heise (1972) and the early ideas on the physics of the SN explosion in MCBs, it was expected that a significant number of WR stars hide a compact companion. However Uhuru did not detect any WR hard-X-ray sources and this was a problem for the evolutionary scenario of MCBs (Vanbeveren et al., 1982).

There are 3 (4) binary pulsars known in the Galaxy where both components are neutron stars or black holes. Accounting for the expected lifetime of such binary pulsars, van den Heuvel (1992) estimated a galactic formation rate for these systems of [FORMULA] 2.4 10-5 /year. This value may still be a factor 2 lower (van den Heuvel, 1996, private communication). Again with the early ideas on the SN explosion and its effect on binary parameters, this value was at least a factor 30 smaller than theoretically expected from binary evolution.

Using recent measurements of pulsar proper motions (Harrison et al., 1993) and a new pulsar distance scale (Taylor and Cordes, 1993), Lyne and Lorimer (1994) obtained a 3-D pulsar velocity distribution f([FORMULA]), which can be very well described by the following relation:


If these velocities reflect the kick velocity a compact star may get as a consequence of an asymmetric SN explosion (notice that this distribution implies an average kick velocity of 450 [FORMULA] which is substantially larger than any previous estimate), one might expect that a large number of binaries will become unbound during the SN explosion, contrary to earlier ideas.

In Vanbeveren et al. (1996) (Paper I) we discussed in detail the evolutionary model for MCBs and how to use it in population synthesis. A summary is given in section 2. We will use this model in order to study the effect of the SN explosion, accounting for the distribution of kick velocities given above, on the following questions:

How many O-type stars are runaways as a consequence of a previous binary SN explosion and how many of them have a CC?

How many WR stars are expected to have a CC?

What is the expected formation rate of double pulsars in our Galaxy?

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998