A second-order maximum bound principle-preserving exponential Runge–Kutta scheme for the convective Allen–Cahn equation

Abstract

We present and analyze a time-stepping scheme that is efficient and second-order accurate for the convective Allen–Cahn equation with a general mobility function. By employing the second-order central finite difference discretization for the diffusion term and the upwind discretization for the advection term, we develop a temporally two-stage second-order exponential Runge–Kutta scheme (ERK2) by incorporating a stabilization technique. It is demonstrated that the ERK2 unconditionally satisfies the discrete maximum bound principle (MBP) for both the polynomial Ginzburg–Landau potential and the logarithmic Flory–Huggins potential. Leveraging the uniform boundedness of numerical solutions guaranteed by the MBP, we prove the second-order convergence rate in the discrete $L^{\infty }$-norm. Furthermore, we establish the unconditionally energy dissipation property of the ERK2 scheme in the case of a constant mobility. Various numerical experiments corroborate our theoretical findings. Notably, these experiments indicate that the proposed ERK2 scheme not only achieves better computational accuracy but also significantly enhances efficiency compared to the classic second-order exponential time differencing Runge–Kutta (ETDRK2) scheme.