Development of a multi-component diffusion fluid solver for two-temperature binary plasma sheath

Studying the plasma sheath is fundamental to a number of applications, ranging from arcing in high
vacuum electronics to charging of space platforms due to plume contamination of electric thrusters:
these fields differ for the variety of the species involved in the simulation of the fluid and for the
different nature of the wall confining it.
In this work we investigate two different approaches to plasma fluid simulations: the multi-fluid,
commonly used in the plasma physics community, and the multi-component approach (a general-
ization of the simpler drift-diffusion model), commonly used in the combustion and re-entry flows
community. The former considers the species within the mixture as interpenetrating fluids, solving
(under the isothermal assumption) for each fluid one equation for mass conservation and one for
momentum conservation; the latter (in this work in its binary diffusion approximation) assumes the
plasma as a single fluid with diffusing species, hence we solve (in isothermal condition) for one mass conservation equation for each species and only one equation for momentum conservation of the bulk velocity of the plasma. In both models solving for the electrostatic Poisson equation ensures charge conservation inside the domain. The multi-component model offers advantages when simulating mixtures with large number of species as the number of equations (and the model complexity) decreases, with a reduction in computational cost.
In this work we compare the results of the two methods by simulating a one-dimensional discharge
with a binary mixture of argon plasma (electrons and positive ions in a uniform bath of argon neu-
trals). We adapt the boundary conditions from the classic multi-fluid approach and implement a
semi-implicit treatment of the electric potential in order to improve performance in terms of stability
of the time integration scheme. Different simulations have been performed with different degree of
collisionality in the mixture by changing the neutrals pressure.
The proposed approach shows good agreement above gas pressure of 10 Pa with a noticeable de-
crease in the computational complexity; such results pave the way to the development of a full
multi-component model that will allow to simulate more complex mixtures.