YOUNG RADIO PULSARS IN GALACTIC GLOBULAR CLUSTERS

Abstract

Currently three isolated radio pulsars and one binary radio pulsar with no evidence of any previous recycling are known in 97 surveyed Galactic globular clusters. As pointed out by Lyne et al., the presence of these pulsars cannot be explained by core-collapse supernovae, as is commonly assumed for their counterparts in the Galactic disk. We apply a Bayesian analysis to the results from surveys for radio pulsars in globular clusters and find the number of potentially observable non-recycled radio pulsars present in all clusters to be < 3600. Accounting for beaming and retention considerations, the implied birth rate for any formation scenario for all 97 clusters is < 0.25 pulsars per century assuming a Maxwellian distribution of velocities with a dispersion of 10 km s^{-1}. The implied birth rates for higher velocity dispersions are substantially higher than inferred for such pulsars in the Galactic disk. This suggests that the velocity dispersion of young pulsars in globular clusters is significantly lower than those of disk pulsars. These numbers may be substantial overestimates due to the fact that the currently known sample of young pulsars is observed only in metal-rich clusters. We propose that young pulsars may only be formed in globular clusters with metallicities with log[Fe/H] > -0.6. In this case, the potentially observable population of such young pulsars is 447^{+1420}_{-399} (the error bars give the 95% confidence interval) and their birth rate is 0.012^{+0.037}_{-0.010} pulsars per century. The mostly likely creation scenario to explain these pulsars is the electron capture supernova of a OMgNe white dwarf.