Thermodynamically consistent dynamic boundary conditions of phase field models

TL;DR

This paper introduces a hierarchical method based on the generalized Onsager principle to derive thermodynamically consistent phase field models with dynamic boundary conditions, ensuring entropy production and energy dissipation.

Contribution

The paper presents a novel two-step method to systematically derive thermodynamically consistent phase field models and boundary conditions using the generalized Onsager principle.

Findings

Method ensures non-negative entropy production in bulk and boundary.

Two types of boundary conditions are formulated for binary phase field models.

Numerical simulations demonstrate the impact of boundary conditions on crystal growth.

Abstract

We present a general, constructive method to derive thermodynamically consistent models and consistent dynamic boundary conditions hierarchically following the generalized Onsager principle. The method consists of two steps in tandem: the dynamical equation is determined by the generalized Onsager principle in the bulk firstly, and then the surface chemical potential and the thermodynamically consistent boundary conditions are formulated subsequently by applying the generalized Onsager principle at the boundary. The application strategy of the generalized Onsager principle in two-step yields thermodynamically consistent models together with the consistent boundary conditions that warrant a non-negative entropy production rate (or equivalently non-positive energy dissipation rate in isothermal cases) in the bulk as well as at the boundary. We illustrate the method using phase field models of binary materials elaborate on two sets of thermodynamically consistent dynamic boundary conditions. These two types of boundary conditions differ in how the across boundary mass flux participates in boundary surface dynamics. We then show that many existing thermodynamically consistent, binary phase field models together with their dynamic or static boundary conditions are derivable from this method. As an illustration, we show numerically how dynamic boundary conditions affect crystal growth in the bulk using a binary phase field model.