Credit: NASA/A. SMITH & J. MORSE

A landmark paper in nuclear astrophysics turns 50 next month. Written by Margaret Burbridge, Geoffrey Burbridge, Willy Fowler and Fred Hoyle — now referred to as 'B2FH' — the paper showed that all of the elements from carbon to uranium can be created inside stars using hydrogen and helium produced from the Big Bang (Rev. Mod. Phys. 29, 547–650; 1957). Alongside work by Al Cameron, it brought stellar nucleosynthesis to the fore in astrophysics. Synthesized elements are spread through space when a star ends its life in a spectacular explosion, or supernova, and with each supernova discovery — now hundreds each year — some aspect of the current models for nucleosynthesis is tested and refined.

Certain supernovae stand out. SN 1987A, the first supernova that was observed in 1987, was the brightest in 400 years. As well as being the first supernova to be confirmed as a source of neutrinos, its progenitor star turned out to be a blue supergiant instead of an assumed red one. Then came SN 1993J, which was unusual in that it underwent significant mass loss before the explosion. And SN 2004dj provided more direct evidence for a non-spherical explosion that might be generic to type-II supernovae.

In 2006 appeared SN 2006gy, the brightest supernova ever recorded. It took 70 days to reach peak luminosity, when it shone brighter than 50 billion Suns, and remained more luminous than any known supernova for more than 100 days. Nathan Smith and co-workers propose that radioactive decay of 56Ni may be responsible for the brightness, although the amount of 56Ni required exceeds what is allowed in the usual core-collapse model (Astrophys. J., in the press; preprint at <http://arxiv.org/abs/astro-ph/0612617>; 2007). Instead, they consider a pair-instability supernova from a massive progenitor star, one with a mass more than 100 times greater than that of the Sun.

In such a high-mass star, core gamma rays with energy greater than the rest mass of two electrons would be able to create electron–positron pairs, setting up a feedback loop that effectively heats the core and generates more high-energy gamma rays and, therefore, more pairs. This mechanism would concentrate energy at the core until the outer layers collapse inwards; the resulting compression would start a thermonuclear explosion of the core. The ensuing blast obliterates the star, leaving no black hole behind. The authors suggest that the progenitor star would be similar to η Carinae (pictured) in our own Galaxy, which could go nova at anytime.