Finding and understanding the rarest and brightest types of stellar explosions
All stars begin their lives fusing hydrogen nuclei to helium, but what happens after that depends on the mass of the star. The more massive the star, the greater the pressure and temperature in its core, allowing it to keep fusing nuclei further along the periodic table. For stars more massive than about 8 times the mass of the Sun, this fusion proceeds all the way to iron, which is the most stable nucleus available.
After this point, fusion becomes endothermic, and the star is no longer able to generate the pressure needed to balance its own gravity. The iron core collapses to a neutron star, while the outer layers bounce off and (with a little help from neutrinos) escape at thousands of kilometres per second. Thermal emission from this shock-heated ejecta, sustained by energy from the decays of freshly synthesized radioactive elements, power an incredibly bright optical display, reaching typical luminosities exceeding 10^8 Suns before slowly fading over weeks to months.
There are a few basic types of core-collapse supernovae, observationally characterised in a sequence from Type II (shows hydrogen lines) through Type Ib (no hydrogen lines, but helium still detected) to Type Ic (neither hydrogen nor helium lines). Physically, this has long been understood as a consequence of seeing explosions from stars that have lost successively deeper layers of the stellar envelope prior to explosion (Type Ia supernovae, which show silicon in place of hydrogen/helium, are actually thermonuclear explosions of accreting or merging white dwarfs, but reach similar energy scales to core-collapse supernovae). The light curves (luminosity as a function of time) of these events are well explained by the recombination of hydrogen atoms (if present) along with the decay of radioactive 56-nickel to 56-cobalt to 56-iron.
The supernova rate is only 1-2 per century in a Milky Way-sized galaxy; however supernovae are visible across vast cosmic distances and are now detected at a rate of several thousand per year by current robotic telescopes. This increase in detection rate has upended the long-standing picture of supernova diversity by revealing entirely new classes of explosion, making up a small percentage by number but offering unique and unexpected clues to understanding the processes that shape the lives and deaths of stars.
We now see supernovae that appear to be interacting with a diverse range of circumstellar environments, in terms its composition, geometry and velocity, showing that the final stages of stellar evolution are much messier than previously thought. Even more surprising are the ‘superluminous supernovae’: hydrogen-poor explosions that are up to 100 times brighter than normal, and extremely hot. They almost always occur in metal-poor dwarf galaxies, and their copious emission cannot be explained by the standard 56-nickel paradigm, since the nickel mass required would exceed the total mass of stellar debris in the explosion.
Our group has been instrumental to this supernova revolution. We discovered many of the first superluminous supernovae through our work in the Pan-STARRS survey from 2010-2014, and have since been improving our search algorithms to increase the detection rate of these extreme events (especially the most nearby ones, where we can get the best data), mapping out the diversity in their properties, and understanding the underlying physics through multiwavelength time-series observations and detailed model fitting.
We have learned that these events seem to form a ‘central engine’ that drives the explosion. For neutron stars, or ‘magnetars’, with spins less than a few millisecond and magnetic fields of 10^13-14 Gauss, the energy from dipole spin-down provides a sustained heating source that can match superluminous supernova light curves. Forming such systems seems to require rapidly rotating progenitor stars with initial masses greater than 20 solar masses, helping to explain why these events are so rare. We have also discovered profound similarities in the properties of superluminous supernovae and long-duration gamma-ray bursts, supporting the central-engine model and suggesting a close connection between the two most energetic death throes in Nature.
Our Group's Related Work
Nicholl et al. (2017) An Ultraviolet Excess in the Superluminous Supernova Gaia16apd Reveals a Powerful Central Engine, ApJL.
Nicholl et al. (2016) SN 2015BN: A Detailed Multi-wavelength View of a Nearby Superluminous Supernova, AJ.