Speaker
Description
Numerical simulations of low-temperature plasma discharges are an essential tool for optimizing plasma-based devices and for improving the understanding of the physical mechanisms governing discharge behavior. These simulations, however, pose major numerical challenges due to the strong nonlinear coupling among the governing equations, arising from plasma chemistry source terms and the electrostatic interaction described by Poisson’s equation. In addition, the system is highly stiff, as the characteristic time scales of the relevant processes span several orders of magnitude, from nanoseconds for electron transport and inelastic collisions to milliseconds for ion motion and metastable species dynamics.
Historically, simulation codes have relied on fully explicit or semi-explicit time integration schemes. While robust, these methods impose stringent limitations on the time-step size, dictated by stability constraints such as the Courant–Friedrichs–Lewy (CFL) condition or the Maxwell relaxation time. This often results in excessive computational costs, especially when fine spatial discretizations are employed. Conversely, fully implicit time integration schemes remove these stability constraints but have traditionally been avoided because of their unfavorable wall-clock-time scaling with the number of degrees of freedom and their high memory requirements.
In this work, we present Merlino2D, an open-source fluid plasma solver developed for the simulation of low-temperature gas discharges and plasma-based devices. The solver operates on unstructured triangular meshes and enables robust two-dimensional simulations in both steady-state and time-dependent regimes. Merlino2D is provided with a native interface to the open-source LisbOn KInetics Boltzmann solver (LoKI-B), for eedf and swarm parameters computation. The governing equations are formulated as a system of differential-algebraic equations and integrated in time using a fully implicit scheme with adaptive time stepping. This approach ensures effective handling of numerical stiffness while maintaining favorable computational performance as the mesh resolution increases.
The numerical methodology underlying Merlino2D is described, and the capabilities of the code are demonstrated through two-dimensional simulations in Cartesian and axisymmetric geometries. Both DC and AC discharges are considered, including corona discharges and surface dielectric barrier discharges (DBDs). The results show good agreement with experimental trends and data reported in the literature, confirming the accuracy and versatility of the proposed solver.