Modeling the dynamic water transport in the porous layers of PEM fuel cells based on numerical upscaling

A one-dimensional two-phase model is developed to study the dynamic water transport in the porous layers of polymer electrolyte membrane (PEM) fuel cells. Liquid water transport and diffusion of oxygen are described by a coupled differential equation system. Capillary pressure, relative permeability, and oxygen diffusivity are determined based on the results of microstructural simulations. The model is compared quantitatively to two different chronoamperometric experiments, potential step voltammetry, and sine wave testing, whereby the optimization algorithm of Nelder and Mead is used. The model predicts the measured dynamic current behavior for both experiments. As a result of the optimization, important two-phase and material parameters are extracted.