PicFoam: An OpenFOAM based electrostatic Particle-in-Cell solver

Abstract

picFoam is a fully kinetic electrostatic Particle-in-Cell (PIC) solver, including Monte Carlo Collisions (MCC), for non-equilibrium plasma research in the open-source framework of OpenFOAM. The solver’s modular design, based on the same principles used in OpenFOAM, makes it highly flexible, by allowing the user to choose different methods at run time, and extendable, by building upon templated modular classes. The implementation of the PIC method employing the finite volume method, allows it to simulate on arbitrary geometries in one to three dimensions. OpenFOAM’s barycentric particle tracking is used effectively to perform charge and field weighting from the Lagrangian particle based description to the Eulerian field description and backwards without computational expensive particle searching algorithm. picFoam also includes open and general circuit boundary models for the description of real plasma devices. Program summaryProgram Title: picFoamCPC Library link to program files:http://dx.doi.org/10.17632/bbsm8tjgjy.1Developer’s repository link:https://github.com/TFDzarm/picFoamLicensing provisions: GPLv3Programming language: C++Nature of problem: The description of equilibrium and non-equilibrium plasma with its complex collective behavior between charged species. As well as their interaction with circuits connected to the boundaries of the plasma. The applications are versatile and lie e.g. in the description of plasma phenomena like discharges in arcjet electric propulsion systems, as well as a general study of plasma phenomena.Solution method: Implementation of the Particle-in-Cell method with Monte Carlo Collisions [1] in the open-source numerical toolbox OpenFOAM [2]. By this the newly implemented solver gains access to OpenFOAM’s powerful tool sets and an easy to set up simulation case structure. Fields are solved by employing OpenFOAM’s cell-centered finite volume solvers, this allows for the simulation on arbitrary structured meshes. The leapfrog scheme [1] is employed for the integration of the particle’s equation of motion combined with relativistic and non-relativistic integration schemes for Newton’s second law of motion [3]. Monte Carlo Collision methods include models for the collision of electrons and neutral species, considered collision events are elastic collision events and inelastic excitation and ionization events. Additional models include Coulomb collisions and isotropic scattering of neutral species [4]. The solver is able to bind circuit models, include open circuit, ideal voltage and current sources, as well as general purpose circuits with an ideal voltage source and a resistance, an impedance and a capacity in series, to the plasma domain [5].Additional comments including restrictions and unusual features: The finite volume method is a relatively rarely used method for implementations of the PIC method, here by employing this method we gain the ability to use arbitrary structured meshes in one to three dimensions. OpenFOAM’s barycentric particle tracking is a novel approach for moving the particles through the mesh. With this method we are able to interpolate (weight) the charges efficiently to the mesh without using computational expensive algorithms to locate the particle’s positions beforehand.