Artificial Correlation Heating in PIC simulations
POSTER
Abstract
The Particle-in-Cell (PIC) method, a cornerstone in plasma modeling, is widely employed for its ability to simulate kinetic phenomena in device-scale domains. Here, we introduce a new constraint necessary to avoid a novel numerical heating mechanism, Artificial Correlation Heating (ACH). This requires that the macroparticle coupling strength be smaller than one 𝚪w<1, where 𝚪w ≡ 𝚪 w2/3, 𝚪 is the physical coupling strength and w is the macroparticle weight. If this condition is violated, the finite macroparticle weight artificially enhances the coupling strength and causes the plasma to heat until the macroparticle coupling strength is near unity, depending on the grid resolution. A comprehensive model of ACH is developed that incorporates electron density, temperature, macroparticle weight, and grid resolution. It is then tested using PIC simulations, delineating the boundaries of the method's applicability and offering a predictive framework for ACH. Moreover, the research explores a runaway heating process induced by ACH in the presence of ionization, which can lead to numerical instability. A conclusion of this study is that the onset of ACH imposes more stringent constraints than those typically employed in standard PIC simulations.
Publication: M D Acciarri et al 2024 Plasma Sources Sci. Technol. 33 035009
M D Acciarri et al 2024. Artificial Correlation Heating in PIC simulations. PSST. Submitted
Presenters
-
Marco D Acciarri
University of Michigan
Authors
-
Marco D Acciarri
University of Michigan
-
Christopher H Moore
Sandia National Laboratories
-
Scott D Baalrud
University of Michigan