The initial mass-final luminosity relation of type II supernova progenitors. Hints of new physics?

We revise the theoretical initial mass-final luminosity relation for progenitors of type IIP and IIL supernovae. The effects of the major uncertainties, as those due to the treatment of convection, semiconvection, rotation, mass loss, nuclear reaction rates and neutrinos production rates are discussed in some details. The effects of mass transfer between components of close-binary systems are also considered. By comparing the theoretical predictions to a sample of type II supernovae for which the initial mass of the progenitors and the pre-explosive luminosity are available, we conclude that stellar rotation may explain a few progenitors which appear brighter than expected in case of non-rotating models. In the most extreme case, SN2012ec, an initial rotational velocity up to 300 km s$^{-1}$ is required. Alternatively, these objects could be mass-loosing components of close binaries. However, most of the observed progenitors appear fainter than expected. This occurrence seems to indicate that the Compton and pair neutrino energy-loss rates, as predicted by the standard electro-weak theory, are not efficient enough and that an additional negative contribution to the stellar energy balance is required. We show that axions coupled with parameters accessible to currently planned experiments, such as IAXO and, possibly, Baby-IAXO and ALPS II, may account for the missing contribution to the stellar energy-loss.