Research Abstracts Online
January 2010 - March 2011
University of Minnesota Twin Cities
College of Science and Engineering
School of Physics and Astronomy
PI: Alexander Heger
Very Massive Stars, Supernovae, and Superbursts
These researchers study astrophysical explosions of different varieties. The main project involves multi-dimensional simulations of pair instability supernovae, the nuclear burning that powers them, and the hydrodynamic instabilities that occur during the explosion. Pair-instability supernovae occur in stars with initial masses of more than hundred times the mass of the sun and they explode with up to hundred times the explosion energy of normal supernovae. Such massive stars are currently thought to be common among the first generation of stars. The researchers are studying models with rotation that naturally break spherical symmetry on large scales. After a series of two-dimensional models to identify the most interesting cases, they will study several models in three dimensions (one octant) in preparation for a full-star (4 pi) run on an external machine.
They are also studying, in multi-dimensional simulations, a new class of "ginormous” supernovae from stars 50,000 times the mass of the sun that explode with 10,000 times the explosion energy of a normal supernova. Though they may be very rare in nature, they are most likely the largest thermonuclear explosions. For this purpose the researchers already have implemented post-Newtonian corrections into the code.
Another topic of study is shell burning of a degenerate oxygen-neon core in some of the least massive stars that still make core collapse supernovae. The group is currently running some first-test simulations in two dimensions on a local workstation, but want to move to highly resolved three-dimensional simulations. A very similar physics setup is found in x-ray superbursts, the thermonuclear runaway of a carbon layer on the surface of an accreting neutron star in a binary star system. All these simulations will be done using the well-tested explicit adaptive mesh refinement code CASTRO (SciDAC, LLNL, LBNL).
Vincent E. Beckner, High Performance Computing Research, Lawrence Berkeley National Laboratory, Berkeley, California
Ke-Jung (Ken) Chen, Graduate Student
Hou Chen, Graduate Student
Laurens Keek, Research Associate
Charles McEachern, Graduate Student
Meng-Ru Wu, Graduate Student