A Thermodynamically Consistent Free Boundary Model for Two-Phase Flows in an Evolving Domain with Bulk-Surface Interaction

TL;DR

This paper develops a thermodynamically consistent mathematical model for two-phase flows in evolving domains with bulk-surface interactions, incorporating phase-fields, contact line dynamics, and boundary slip conditions.

Contribution

It introduces a novel, comprehensive model derived via Lagrange multipliers and energetic variational methods, generalizing previous models in the literature.

Findings

Model captures bulk-surface material transfer and variable contact angles.

Inclusion of generalized Navier slip boundary condition improves contact line description.

Derivation from fundamental principles ensures thermodynamic consistency.

Abstract

We derive a thermodynamically consistent model, which describes the time evolution of a two-phase flow in an evolving domain. The movement of the free boundary of the domain is driven by the velocity field of the mixture in the bulk, which is determined by a Navier--Stokes equation. In order to take interactions between bulk and boundary into account, we further consider two materials on the boundary, which may be the same or different materials as those in the bulk. The bulk and the surface materials are represented by respective phase-fields, whose time evolution is described by a bulk-surface convective Cahn--Hilliard equation. This approach allows for a transfer of material between bulk and surface as well as variable contact angles between the diffuse interface in the bulk and the boundary of the domain. To provide a more accurate description of the corresponding contact line motion, we include a generalized Navier slip boundary condition on the velocity field. Based on local mass balance laws, we derive our model from scratch in two different ways: by the Lagrange Multiplier Approach and (in the case of matched densities and no mass flux between bulk and surface) by the Energetic Variational Approach. We further show that our model generalizes previous models from the literature, which can be recovered from our system by either dropping the dynamic boundary conditions or assuming a static boundary of the domain.