Hans Bethe worked out the proton-proton chain in 1939 — the sequence of nuclear reactions by which four hydrogen nuclei in a star's core fuse step by step into one helium nucleus, releasing the energy that powers the Sun. Eighteen years later Margaret and Geoffrey Burbidge, William Fowler, and Fred Hoyle worked out most of the rest in B²FH (Synthesis of the Elements in Stars, Reviews of Modern Physics, 1957), the paper that gave nuclear astrophysics its name and established that essentially every element in the periodic table heavier than helium was produced in the interior of a star or in the death of one. The program it laid out has been confirmed at every step by half a century of subsequent observation.
The full sequence runs from the Big Bang to the death of stars. Big Bang nucleosynthesis in the first three minutes produced essentially only hydrogen and helium with traces of deuterium and lithium; the universe before stars was chemically featureless. Hydrogen burning on the main sequence converts H to He through the proton-proton chain (stars like the Sun) or the CNO cycle (more massive stars, where carbon, nitrogen, and oxygen serve as catalysts). Helium burning in red-giant cores converts three alpha particles to carbon through the triple-alpha process; further alpha captures produce oxygen and neon. Advanced burning stages in the cores of massive stars then run, in succession, carbon, oxygen, neon, magnesium, silicon, and finally iron — the most-bound nucleus per nucleon, beyond which fusion no longer releases energy. Stellar fusion accounts for most of the periodic-table middle, from carbon to iron.
Elements heavier than iron have less binding energy per nucleon and so are made by neutron capture, not fusion. The s-process (slow neutron capture) runs in red-giant atmospheres at modest neutron flux and produces about half the elements heavier than iron, including barium and lead. The r-process (rapid neutron capture) requires extreme neutron flux and produces gold, platinum, the lanthanides, the actinides, and most of the rest. For decades the r-process was assumed to occur primarily in core-collapse supernovae; the 2017 GW170817 event — the first directly observed neutron-star merger — produced a kilonova whose spectrum showed unmistakable r-process signatures, settling that neutron-star mergers are at least a major contributor and possibly the dominant one. The gold in your wedding ring came, in part, from collisions of neutron stars billions of years ago — now an empirically grounded statement, not a poetic one.
Stellar archaeology — using the chemical compositions of ancient nearby stars to reconstruct the nucleosynthesis history of the early universe — is now a substantial research program. The 4MOST survey at Paranal (2025+) and the Milky Way Mapper of SDSS-V are observing millions of stars to build the chemical-evolution history of the galaxy element by element. Multi-messenger gravitational-wave observation has elevated kilonova nucleosynthesis from theoretical inference to direct measurement, and third-generation detectors planned for the 2030s (Einstein Telescope, Cosmic Explorer) are expected to detect ~10⁵ neutron-star mergers per year. Every solid object in the solar system, including this hand and the screen these words appear on, is a piece of somebody else's funeral.