Nucleosynthesis in massive stars with improved nuclear and stellar physics

We present the first calculations to follow the evolution of all stable nuclei and their radioactive progenitors in stellar models computed from the onset of central hydrogen burning through explosion as Type II supernovae. Calculations are performed for Population I stars of 15, 19, 20, 21, and 25 M-circle dot using the most recently available experimental and theoretical nuclear data, revised opacity tables, neutrino losses, and weak interaction rates and taking into account mass loss due to stellar winds. A novel ``adaptive`` reaction network is employed with a variable number of nuclei ( adjusted each time step) ranging from similar to700 on the main sequence to greater than or similar to2200 during the explosion. The network includes, at any given time, all relevant isotopes from hydrogen through polonium (Z = 84). Even the limited grid of stellar masses studied suggests that overall good agreement can be achieved with the solar abundances of nuclei between O-16 and Zr-90. Interesting discrepancies are seen in the 20 M-circle dot model and ( so far, only in that model) are a consequence of the merging of the oxygen, neon, and carbon shells about a day prior to core collapse. We find that, in some stars, most of the ``p-process`` nuclei can be produced in the convective oxygen-burning shell moments prior to collapse; in others, they are made only in the explosion. Serious deficiencies still exist in all cases for the p-process isotopes of Ru and Mo.