The formation of a strange or hybrid star from a neutron star progenitor is believed to occur when the central stellar density exceeds a critical value. If the transition from hadron to quark matter is of first order, the event has to release a huge
amount of energy in a very short time and we would be able to observe the phenomenon even if it is at cosmological distance far from us; most likely, such violent quark deconfinement would be associated with at least a fraction of the observed gamma ray bursts. If we allow for temporal variations of fundamental constants like $Lambda_{QCD}$ or $G_N$, we can expect that neutron stars with an initial central density just below the critical value can enter into the region where strange or hybrid stars are the true ground state. From the observed rate of long gamma ray bursts, we are able to deduce the constraint $dot{G}_N/G_N lesssim 10^{-17} {rm yr^{-1}}$, which is about 5 orders of magnitude more stringent than the strongest previous bounds on a possible increasing $G_N$.
Gravitational particle production in time variable metric of an expanding universe is efficient only when the Hubble parameter $H$ is not too small in comparison with the particle mass. In standard cosmology, the huge value of the Planck mass $M_{Pl}
$ makes the mechanism phenomenologically irrelevant. On the other hand, in braneworld cosmology the expansion rate of the early universe can be much faster and many weakly interacting particles can be abundantly created. Cosmological implications are discussed.
