Advanced PIC Algorithms for Modeling Plasmas

Particle-in-cell (PIC) simulation has now become a commonplace tool for simulating plasma ki- netics, but there are constraints on its effectiveness – timesteps must be small to avoid numerical instabilities, and the number of particles required to reduce statistical noise to acceptable levels may be computationally prohibitive. In plasmas with many multistate ion species, even determin- ing dominant kinetic interactions in the system may be difficult. Speed-limited PIC (SLPIC) methods which constrain the speed of the fastest simulation parti- cles (enabling larger timesteps) are developed. Kinetics-only delta-f methods, which model cold- plasma evolution as a fluid and use PIC only to capture warm-plasma effects, are explored for noise reduction. A global modeling code (GSim) is developed which can be run as a precursor to more complex PIC models as a means of determining key physics interactions in the system. Phase I primarily focused on the SLPIC method, which was shown to reduce wallclock times in simulations of electrostatic sheath formation by factors of 6 or greater, and in simulations of plasma free expansion into vacuum by factors of 2.5. Single-particle drift behaviors of SLPIC particles were shown to be correct except in cases where rapid time variation of the electromagnetic fields (e.g. in polarization drifts or particle loss rates) introduces inaccuracies. SLPIC and KODF implementations in VSim will be validated and exercised separately and in tan- dem to explore the physics of plasma-vacuum expansion and of finite-temperature effects in fusion plasmas heated by RF waves. Global modeling capabilities (e.g. for power balance or principal- path chemistry computations in low-temperature plasmas) will be developed and exercised. Developing faster and lower-noise VSim PIC modeling capabilities enables its use as a design tool for plasma manufacturing equipment, filling a need in the industrial plasma processing community for scalable computational kinetic models. Industrial plasma processing customers can use the software to more rapidly design and optimize equipment to carry out increasingly complex manu- facturing processes, using GSim as a scoping tool to determine optimal machine parameter spaces. Additionally, by studying the behavior of RF physics in fusion plasmas, we will be able to broaden the physics understanding that serves as a basis for complex fusion experiments such as ITER, a multibillion-dollar device presently under construction.