Astrophys. J. 757, 91 (2012)

A star shines, loses energy and then collapses to a compact remnant, which could be a white dwarf, a neutron star or a black hole. Given that stellar masses form a continuum, it's reasonable to expect a similarly smooth distribution of remnant masses — but that is not the case. There is a marked gap between the heaviest observed neutron stars (at two solar masses) and the lightest black holes (five solar masses). Krzysztof Belczynski and co-workers use this curious fact in their modelling of supernova explosions.

During gravitational collapse, the formation of a rigid proto-neutron star can stall the process, as infalling material 'bounces' off the core and sends out shockwaves. But neutrinos within the core heat up the turbulent layer around the proto-neutron star. The resulting temperature and density gradients lead to a Rayleigh–Taylor instability that violently mixes the layers, thus starting the supernova engine.

The timing is critical, say Belczynski et al., if the observed mass gap is to be created. A sufficiently rapid growth time of the instability (10–20 milliseconds) and sufficient energy lead to a supernova explosion within 100–200 milliseconds; this ejects most of the star and leaves behind a lower-mass neutron star. Otherwise, the star collapses into a heavier black hole.