Astrophysical Thermonuclear Flashes

The aim of the Center on Astrophysical Thermonuclear Flashes is to solve the long-standing problem of thermonuclear flashes on the surfaces of compact stars such as neutron stars and white dwarfs, and in the interior of white dwarfs. This problem is remarkable for the breadth of physical phenomena involved, ranging from accretion flow onto the surfaces of these compact stars, to shear flow and Rayleigh-Taylor instabilities on the stellar surfaces, ignition of nuclear burning under conditions leading to convection, and either deflagration or detonation, stellar envelope expansion, and the possible creation of a common envelope binary star system. Indeed, few -- if any -- astrophysical problems present a substantially greater level of physical complexity. Because virtually every aspect of this problem represents a computational Grand Challenge, large-scale numerical simulations are at the heart of its resolution. Given the complexity of the physics, the most ambitious attacks on this problem to date have been in two dimensions; the Cray T3E at the Pittsburgh Supercomputing Center provides a unique opportunity to attack the full problem in three dimensions.

Click on the image to view a larger version.

 

The figure shows results from a 2-dimensional simulation of an X-ray burst on the surface of a neutron star, showing a thermonuclear burning front at about 17 microseconds. The vertical axis is a 500 meter section through the upper layers of the neutron star, with the dark blue line tracking the surface. The horizontal axis is an 800 meter portion along the surface of the neutron star, which is about 1% of the total circumference of the star. Different colors represent different temperatures, with the hottest material appearing the darkest. The light green line tracks regions of pure helium (the initial fuel), while the light blue line tracks regions of pure nickel (the final product of the burning). Overlayed is the velocity field, represented by the gray arrows, which shows material erupting from the neutron star.

The thermonuclear flash problem is inherently three-dimensional, yet no fully three-dimensional calculations have been done to date, and in fact, with the exception of a few recent studies, the calculations have been one-dimensional. The reason is that astrophysical thermonuclear flashes involve a variety of distinct and complex nonlinear physical processes which have widely disparate length and time scales. Advances in massively parallel processing programming techniques to overcome problems such as scalability, parallel I/O performance and adaptive mesh refinement allow the Pittsburgh Supercomputing Center's Cray T3E to lead the way to the first empirically verifiable simulations in this important field.

Figure provided by F. X. Timmes, Michael Zingale, Kevin Olson, B. A. Fryxell, and Don Lamb. The Center includes leading computer scientists and physicists in the fields of nuclear astrophysics, condensed matter physics, statistical physics and complexity theory, structure and evolution of compact stars, and astrophysical (computational) hydrodynamics and convection. This core group includes scientists at the University of Chicago and Argonne National Laboratory. In addition, computer scientists at the Rensselaer Polytechnic Institute will join us to strengthen the partnership still further.