Constraining the final fates of massive stars by oxygen and iron enrichment history in the Galaxy

Recent observational studies of core-collapse supernovae suggest only stars with zero-age main sequence masses smaller than $16$-$18\ M_\odot$ explode when they are red supergiants, producing type IIP supernovae. This may imply that more massive stars produce other types of supernovae or they simply collapse to black holes without giving rise to bright supernovae. This failed supernova hypothesis can lead to significantly inefficient oxygen production because oxygen abundantly produced in inner layers of massive stars with zero-age main sequence masses around $20$-$30\ M_\odot$ might not be ejected into the surrounding interstellar space. We first assume an unspecified population of oxygen injection events related to massive stars and obtain a model-independent constraint on how much oxygen should be released in a single event and how frequently such events should happen. We further carry out one-box galactic chemical enrichment calculations with different mass ranges of massive stars exploding as core-collapse supernovae. Our results suggest that the model assuming that all massive stars with $9$-$100\ M_\odot$ explode as core-collapse supernovae is still most appropriate in explaining the solar abundances of oxygen and iron and their enrichment history in the Galaxy. The oxygen mass in the Galaxy is not explained when assuming that only massive stars with zero-age main sequence masses in the range of 9-17 $M_\odot$, contribute to the galactic oxygen enrichment. This finding implies that a good fraction of stars more massive than $17M_\odot$ should eject their oxygen layers in either supernova explosions or some other mass loss processes.