SUPERNOVAE, the explosions that mark the death of ageing stars, are among the most violent and spectacular events in space. Yet in the latest computer simulation of their inner workings the dying stars just puff up before fading in a gentle whimper.
The problem has got physicists worried. 鈥淪omething is missing,鈥� says Thomas Janka, who wrote the simulation programme with colleagues at the Max Planck Institute for Astrophysics in Garching, Germany. It shows we don鈥檛 fully understand what makes stars explode, he says.
The simulation was supposed to be a rigorous test of supernova theory. Chris Fryer, who models supernovae at Los Alamos National Laboratory in New Mexico, says he had not expected to see calculations of such a calibre used in simulations for a decade or two.
Advertisement
The new simulation embodies the most accurate calculations yet of how neutrinos behave in stars. Although neutrinos barely interact with matter and have almost no mass, these ghostly particles are thought to play a key role in supernovae.
When stars burn out they collapse, forcing vast numbers of particles out of the star鈥檚 core and into space. But the shock wave produced by the collapse would rapidly stall were it not for high-energy neutrinos pouring out of the core. Although neutrinos only interact faintly with other particles, theorists have calculated that they may dump enough energy into the gases behind the shock wave to heat them up and trigger an explosion.
Support for the theory came in 1987, when a star 20 times as massive as the Sun exploded in a nearby galaxy. The pulse of neutrinos that heralded its appearance was picked up by the Kamiokande neutrino detector in Japan.
Because data from real supernovae is so scarce, astrophysicists rely on computer simulations to test their understanding of the mechanisms involved. Previous simulations have shown stars exploding as expected. But these did not include the serious number crunching needed to accurately model the production of neutrinos 鈥� the way they move and their interactions with other particles in the star, for example.
Janka hoped his more advanced model would remove any doubt that neutrinos trigger the explosions. Instead he found that the shock wave simply faded out when he ran the simulation.
Does this mean the neutrino-driven mechanism is wrong? Tony Mezzacappa, at Oak Ridge National Laboratory in Tennessee, who heads the Terascale Supernova Initiative, says it is too soon to draw final conclusions. Supernovae are sensitive to many things other than neutrinos, such as convection currents, composition, magnetic fields and the star鈥檚 rotation. These factors may interact in ways we do not yet understand.