This review highlights current (and future!) hot topics in astrophysics where atomic or molecular input data are (or will be) essential, with special emphasis on topics relating to nucleosynthesis and cosmochemistry.We first discuss issues (like the abundances of oxygen and iron in the Sun, and that of lithium in post-AGB stars) where the use of poor-quality atomic or molecular data have led to spurious astrophysical puzzles which sparked fancy astrophysical models or theories. We then address issues where the advent of new instruments (like the ultraviolet high-resolution spectrographs-GHRS onboard HST, Keck-HRS or VLT-UVES-or future infrared satellites) calls for new and accurate atomic or molecular data.
The solar-system abundance curve: A clue to nucleosynthesisThe relative abundances of the chemical elements in the solar system have been derived for the first time by Russell in 1929 [114], from the spectroscopic analysis of the solar photosphere, and by Goldschmidt in 1937 [48], from the chemical analysis of primitive meteorites known as carbonaceous chondrites. More recent compilations and revisions of the solar-system abundances were provided by Suess & Urey in 1956 [124],