Partnership Center for High-Fidelity Boundary Plasma Simulation
Dr. C.S. Chang, Lead Principal Investigator (PPPL*)
Dr. L. Chacón (LANL)
Dr. S. Klasky (ORNL)
Prof. S. Parker (U. Colorado Boulder)
Dr. M. Adams (LBNL)
Dr. D. Hatch and R. Moser (U. Texas at Austin)
Prof. R. Moser (U. Texas at Austin) and
Dr. J. Hittinger (LLNL)
The boundary region of the plasma in a magnetic fusion device plays a critical role in the device’s performance. The boundary plasma first acts as an insulator for the burning fusion plasma in the core, much like the walls of a Thermos bottle. But, the boundary plasma also interacts with the surrounding solid walls, potentially damaging the wall surfaces and producing in the process impurity atoms that can cool and dilute the core plasma, slowing the fusion reactions.
Progress towards understanding and predicting the boundary plasma has been elusive because the dynamics of individual plasma particles must be taken into account and the small-scale physics at the individual particle motion level interact with the large scale collective physics at the device scale level; simpler approaches in which the plasma is characterized only by average properties have not fared well. This challenge is exacerbated by the boundary region’s complex geometry.
The XGC suite of advanced high fidelity kinetic simulation codes, run on extreme-scale computers, is designed to address the physics challenges arising in this boundary region. XGC uses first-principles based equations to maximize physics fidelity, with the degree of fidelity increasing as the capabilities of leadership class computers continue to grow. This project is essential to understanding existing tokamak experiments and to predicting with greater confidence the operation and performance of the ITER international fusion reactor, under construction in France, and of future fusion reactors. To address these challenges, we have assembled a team of world experts in physics, computer science, applied mathematics, and data management to collaboratively keep XGC at the cutting edge of computational and algorithmic development, as well as of physics fidelity.
*The PPPL members include the following subcontractors: V. Carey (U. Colorado Denver), D. Curreli (U. Illinois), M. Greenwald (MIT), W. Hoffman (Kitware), J. Myra (Lodestar Research), Tynan (U. California San Diego), P. Worley (PHWorley Consulting)