Modeling and Simulating Plasma

Dr. Andrew Christlieb, Department Chair
Department of Computational Mathematics, Science and Engineering, Michigan State University

Dr. Andrew Christlieb’s research focuses on modeling and simulating materials in extreme states. The large range of temporal and spatial scales that need to be resolved in order to accurately describe these problems makes them difficult.

Dr. Christlieb is passionate about developing methods to accurately describe plasmas. Plasma is the fourth state of matter, after solid, liquid, and gas. The plasma states occurs when enough energy is added to the system such that the electrons overcome their binding energy and the systems consists of positive and negatively charged particles with a wide range of masses. The plasma state can have densities well in excess of solids; an example is the core of a star where fusion is taking place. Plasmas may have a density as low as one particle per cubic meter (e.g., interstellar material). What makes this truly difficult is that at high densities, models for accurately capturing fundamental effects are very different from models that are good at describing low-density plasmas. In many of the problems, the densities can vary by eight orders of magnitude and the temporal scales of importance can range over nine orders of magnitude.

Identifying and resolving critical scales is one of the largest challenges in accurately modeling these systems. Recently, Dr. Christlieb’s group has targeted modeling of dense correlated plasmas. These are problems where the potential energy is much higher than that of the kinetic energy. This leads to situations where the ionized gas behaves like a solid. Another area his group works on is developing new numerical methods for modeling the casting of functionalized membranes in polymer science. Such membranes are used in batteries, fuel cells and solar cells. These are systems where casting of the membrane may take as long as 24 hours, while critical chemistry happens in picoseconds.

To design algorithms to attack these problems, the Christlieb group works on methods well suited to heterogeneous computing platforms in a high performance computing setting. This includes data science tools capable of extracting essential information from the simulations.