A structure-preserving finite element discretization for the time-dependent Nernst-Planck equation

It is still a challenging task to get a satisfying numerical solution to the time-dependent Nernst-Planck (NP) equation, which satisfies the following three physical properties: solution nonnegativity, total mass conservation, and energy dissipation. In this work, we propose a structure-preserving finite element discretization for the time-dependent NP equation combining a reformulated Jordan-Kinderlehrer-Otto (JKO) scheme and Scharfetter-Gummel (SG) approximation. The JKO scheme transforms a partial differential equation solution problem into an optimization problem. Our finite element discretization strategy with the SG stabilization technique and the Fisher information regularization term in the reformulated JKO scheme can guarantee the convexity of the discrete objective function in the optimization problem. In this paper, we prove that our scheme can preserve discrete solution nonnegativity, maintain total mass conservation, and preserve the decay property of energy. These properties are all validated with our numerical experiments. Moreover, the later numerical results show that our scheme performs better than the traditional Galerkin method with linear Lagrangian basis functions in keeping the above physical properties even when the convection term is dominant and the grid is coarse.