Remnant masses from 1D+ core-collapse supernovae simulations: bimodal neutron star mass distribution and black holes in the low-mass gap

The explosion of core-collapse supernovae (CCSNe) is an extremely challenging problem, and there are still large uncertainties regarding which stars lead to successful explosions that leave behind a neutron star, and which ones will form a black hole instead. In this paper, we simulate 341 progenitors at three different metallicities using spherically symmetric simulations that include neutrino-driven convection via a mixing-length theory. We use these simulations to improve previously derived explosion criteria based on the density and entropy profiles of the pre-supernova progenitor. We also provide numerical fits to calculate the final mass of neutron stars based on either compactness, the location of the Si/Si-O interface, or the Chandrasekhar mass. The neutron star birth mass distribution derived from our 1D+ simulations is bimodal, contrary to what the most popular 1D CCSN simulations have shown so far. We compare the theoretically derived neutron star mass distributions with the observed ones and discuss potential implications for population synthesis studies. We also analyze the black hole mass distribution predicted by our simulations. To be consistent with current models of matter ejection in failed SNe, a large fraction of the envelope must be expelled, leading to small black holes in the low-mass gap. One black hole in this mass region has recently been observed in the GW230529 event by the LIGO-Virgo-KAGRA collaboration. Our results naturally agree with this detection, which the most popular prescriptions for explodability and remnant masses are not able to reproduce. In general, we find that the explosion outcome and mass of the remnant strongly depend on the pre-collapse structure of the progenitor. However, their dependence on the initial mass of the star and the mass of the CO core is highly uncertain and non-linear.