Thermonuclear explosion of rotating massive stars could explain core-collapse supernovae

It is widely thought that core-collapse supernovae (CCSNe), the explosions of massive stars following the collapse of the stars' iron cores, is obtained due to energy deposition by neutrinos. So far, this scenario was not demonstrated from first principles. Kushnir and Katz (2014) have recently shown, by using one-dimensional simulations, that if the neutrinos failed to explode the star, a thermonuclear explosion of the outer shells is possible for some (tuned) initial profiles. However, the energy released was small and negligible amounts of ejected $^{56}$Ni were obtained, implying that these one-dimensional collapse induced thermonuclear explosions (CITE) are unlikely to represent typical CCSNe. Here I provide evidence supporting a scenario in which the majority of CCSNe are the result of CITE. I use two-dimensional simulations to show that collapse of stars that include slowly (few percent of breakup) rotating $\sim0.1-10\,M_{\odot}$ shells of mixed helium-oxygen, leads to an ignition of a thermonuclear detonation wave that unbinds the stars' outer layers. Simulations of massive stars with different properties show that CITE is a robust process, and results in explosions with kinetic energies in the range of $10^{49}-10^{52}\,\textrm{erg}$, and $^{56}$Ni yields of up to $\sim\,M_{\odot}$, which are correlated, in agreement with observations for the majority of CCSNe. Stronger explosions are predicted from higher mass progenitors that leave more massive remnants, in contrast to the neutrino mechanism. Neutron stars are produced in weak ($\lt10^{51}\,\textrm{erg}$) explosions, while strong ($\gt10^{51}\,\textrm{erg}$) explosions leave black hole remnants.