Intrinsic local Gauss's law preserving PIC method: A self-consistent field-particle update scheme for plasma simulations

TL;DR

This paper introduces a novel local PIC method that inherently preserves Gauss's law, ensuring physically accurate plasma simulations without auxiliary corrections, and is computationally efficient for large-scale parallel applications.

Contribution

It presents a self-consistent, local update scheme for electric fields in PIC simulations that exactly enforces Gauss's law without solving Poisson or Ampère equations.

Findings

Maintains discrete Gauss's law exactly in simulations.

Demonstrates effectiveness on classical plasma benchmarks.

Offers a linear complexity local algorithm suitable for parallel computing.

Abstract

In order to perform physically faithful particle-in-cell (PIC) simulations, the Gauss's law stands as a critical requirement, since its violation often leads to catastrophic errors in long-term plasma simulations. This work proposes a novel method that intrinsically enforces the Gauss's law for the Vlasov-Amp\`ere/Vlasov-Poisson system without requiring auxiliary field corrections or specialized current deposition techniques. The electric field is managed to get updated locally and consistently with the motion of particles via splitting the motion into sub-steps along each dimension of the computational mesh. To further obtain a curl-free electric field, a local update scheme is developed to relax the electric-field free energy subject to the Gauss's law. The proposed method avoids solving the Poisson's or Amp\`ere's equation, resulting in a local algorithm of linear complexity for each time step which can be flexibly combined with various temporal discretization for particle motion in PIC simulations. Theoretical analysis verifies that the proposed method indeed maintains the discrete Gauss's law exactly. Numerical tests on classical benchmarks, including the Landau damping, two-stream instability and Diocotron instability, demonstrate the key advantages of the proposed method. It is expected that the local nature of the proposed method makes it a promising tool in parallel simulations of large-scale plasmas.