A new hybrid scheme for simulations of highly collisional RF-driven plasmas

Abstract

This work describes a new 1D hybrid approach for modeling atmospheric pressure discharges featuring complex chemistry. In this approach electrons are described fully kinetically using ParticleIn-Cell/Monte-Carlo (PIC/MCC) scheme, whereas the heavy species are modeled within a fluid description. Validity of the popular drift-diffusion approximation is verified against a "full" fluid model accounting for the ion inertia and a fully kinetic PIC/MCC code for ions as well as electrons.The fluid models require knowledge of the momentum exchange frequency and dependence of the ion mobilities on the electric field when the ions are in equilibrium with the latter. To this end an auxiliary Monte-Carlo scheme is constructed. It is demonstrated that the drift-diffusion approximation can overestimate ion transport in simulations of RF-driven discharges with heavy ion species operated in the γ mode at the atmospheric pressure or in all discharge simulations for lower pressures. This can lead to exaggerated plasma densities and incorrect profiles provided by the drift-diffusion models. Therefore, the hybrid code version featuring the full ion fluid model should be favored against the more popular drfit-diffusion model, noting that the suggested numerical scheme for the former model implies only a small additional computational cost.

Hybrid Simulation of Low Temperature Plasmas: A Brief Tutorial

“…Eremin et al proposed a PIC‐Fluid hybrid for highly collisional RF helium discharges between parallel plate electrodes. Electrons were treated by PIC‐MCC while heavy species were treated as fluid.…”

Observations of Surface Mode Influence on Plasma Uniformity in PIC/MCC Simulations of Large Capacitive Discharges

“…We employed a model 2d (xz) rectangular discharge geometry with L z = 2.5 cm and L x = 50 cm under the uniformity assumption in the infinite y direction. A voltage source was used to drive the discharge with the voltage amplitude adjusted during the simulation to meet prescribed absorbed power value per unit length . The boundary conditions for the potential set it to zero at the top electrode and the sidewalls, to the driving voltage at the bottom electrode and linear fall from the driving voltage to zero was assumed in the transitional region corresponding to the dielectric spacer, which was located at the bottom of the discharge and span over 2.5 cm from each grounded sidewall.…”

“…The boundary conditions for the potential set it to zero at the top electrode and the sidewalls, to the driving voltage at the bottom electrode and linear fall from the driving voltage to zero was assumed in the transitional region corresponding to the dielectric spacer, which was located at the bottom of the discharge and span over 2.5 cm from each grounded sidewall. A voltage source was used, and in order to match a prescribed power per unit length we have used the amplitude adjustment procedure such as the one described in ref . To model the neutral background gas (assumed to be argon at 3 Pa) chemistry, a set of three reactions (the elastic collision, the ionization, and the excitation from the ground level) for the electron‐neutral collisions and a set of two reactions (the elastic collision and the charge exchange) for the ion‐neutral collisions were used with the cross‐sections for the sets taken from ref .…”

“…A voltage source was used to drive the discharge with the voltage amplitude adjusted during the simulation to meet prescribed absorbed power value per unit length. [28] The boundary conditions for the potential set it to zero at the top electrode and the sidewalls, to the driving voltage at the bottom electrode and linear fall from the driving voltage to zero was assumed in the transitional region corresponding to the dielectric spacer, which was located at the bottom of the discharge and span over 2.5 cm from each grounded sidewall. A voltage source was used, and in order to match a prescribed power per unit length we have used the amplitude adjustment procedure such as the one described in ref.…”

“…A voltage source was used, and in order to match a prescribed power per unit length we have used the amplitude adjustment procedure such as the one described in ref. [28] To model the neutral background gas (assumed to be argon at 3 Pa) chemistry, a set of three reactions (the elastic collision, the ionization, and the excitation from the ground level) for the electron-neutral collisions and a set of two reactions (the elastic collision and the charge exchange) for the ion-neutral collisions were used with the cross-sections for the sets taken from ref. [29] to [30] , respectively.…”