Finite size effect in the persistence probability of the Edwards-Wilkinson model of surface growth and effect of non-linearity

Abstract

The dynamical evolution of the surface height is controlled by either a linear or a nonlinear Langevin equation, depending on the underlying microscopic dynamics, and is often done theoretically using stochastic coarse-grained growth equations. The persistence probability $p(t)$ of stochastic models of surface growth that are constrained by a finite system size is examined in this work. We focus on the linear Edwards-Wilkinson model (EW) and the nonlinear Kardar-Parisi-Zhang model, two specific models of surface growth. The persistence exponents in the continuum version of these two models have been widely investigated. Krug et al.[Phys. Rev. E , 56:2702-2712, (1997)] and Kallabis et al. [EPL (Europhysics Letters) , 45(1):20, 1999] had shown that, the steady-state persistence exponents for both these models are related to the growth exponent $β$ as $θ=1-β$. It is numerically found that the values of persistence exponents for both these models are close to the analytically predicted values. While the results of the continuum equations of the surface growth are well known, we focus to study the persistence probability expressions for discrete models with a finite size effect. In this article, we have investigated the persistence probabilities for the linear Edwards-Wilkinson(EW) model and for the non-linear Kardar-Parisi-Zhang(KPZ) model of surface growth on a finite one-dimensional lattice. The interesting phenomenon which is found in this case is that the known scenario of $p(t)$ of the following algebraic decay vanishes as we introduce a finite system size.