Progenitor Models

Nowadays the most accounted models for the GRB progenitors can be divided into two classes: the first includes the core collaps of very massive stars (``collapsar'' or ``hypernova'' models, [Woosley, 1993], [Paczynski, 1998]), and the ``supranova'' model, concerning the formation of a BH from a rapidly spinning and decelerating NS, left over by a previous SN explosion ([Vietri & Stella, 1998]); the second class includes mergers of compact stars (NS, BH, or even white dwarfs, provided that at least one mergee is either a NS or a BH; see [Eichler et al., 1989], [Belczynski et al., 2001]). Among the remaining models, here we only cite the cannonball model proposed by Dado, Dar & De Rújula, 2001, which reproduces all the light curves of the afterglows in association with underlying SNe, that up to now seem to agree with observations.

The ``collapsar'' model deals with a rotating massive star, whose Fe core collapses, producing a Kerr BH surrounded by a $0.1-1 M_{\odot}$ torus, whose matter is accreted at a very high rate; the energy can be extracted in two ways: first, by accretion of the disk matter by the BH; second, from the rotational energy of the BH via the Blandford-Znajek process ([Blandford & Znajek, 1977]). The so released energy amounts to $\sim 10^{54}$ ergs; according to this model, a fireball with a luminosity $\sim$ 300 times greater than that of a normal SN is produced. In this case, the GRB would be produced in a dense environment, near star forming regions.

The ``supranova model'' requires a supra-massive NS, which is rapidly spinning down, untill it implodes to a BH; during this implosion the surrounding matter, previously produced by the SN explosion, is swept up, leading to a baryon-clean environment. The presence of a $\sim 0.1 M_{\odot}$ torus is predicted, while the energy extraction occurs via the conversion of the Poynting flux into a magnetized relativistic wind.

Finally, the merging of compact stars, like in the NS-NS case, has the following characteristics: since their typical life should be of the order of $\sim 10^9$ years, and given that such systems usually have large escape velocities, they are likely to be found away from star forming regions.

Figure: GRB-Host Galaxy Offset Distribution. It seems to favour a progenitor population associated with star forming regions. (From [Djorgovski et al., 2001]).

In particular, this property does not seem to agree with the observed distribution of GRB-host galaxy offsets (fig. ): this looks like following the light of their hosts, that is roughly proportional to the density of star formation (especially for the high-$z$ galaxies). However, this piece of evidence does not suffice for discarding this model yet. The result of such a merging is a Kerr BH, with an energy release of $\sim 10^{54}$ ergs. Also in this case, it is possible that a $\sim 0.1 M_{\odot}$ accretion disk is formed and that is accreted within a few dozen seconds, then producing the internal shocks responsible for the burst. Eventually, it is worth mentioning that it has been suggested that the short duration bursts might be produced by such compact binary systems.