Geometrical parameters optimization to improve the effective thermal conductivity of the gas diffusion layer for PEM fuel cell

Abstract

The main concern of this article is a numerical investigation of the geometrical parameters of composite gas diffusion layers (CGDLs) and their impact on effective thermal conductivity (ETC). The geometric parameters under investigation include porosity, thicknesses of the gas diffusion layer (GDL) and micro porous layer (MPL), as well as fibers diameter and orientation. Additionally, non-geometric parameters such as saturation level and operating temperature are examined. This study utilizes realistic full microstructure simulations (GDL + MPL + PolyTetraFluoroEthylene/PTFE + binder) of a paper GDL type (SGL 25BC) to examine and enhance ETC in both in-plane (IP) and through-plane (TP) directions by optimizing the parameters under investigation. To achieve this, a MATLAB code was used to generate microstructures under various conditions, which were simultaneously imported into COMSOL multi-physics via live link technique (LLT). Subsequently, the non-dominated sorting genetic algorithm II (NSGAII) was employed as an optimization method to refine the geometric/non-geometric parameters. Finally, the ETC results for both TP and IP directions were compared with experimental data from the literature. The optimized microstructure exhibits higher ETC values compared to both the initial simulated microstructure and the experimental data. This finding indicates that fabricating CGDLs within the recommended range of optimal geometric values can lead to a substantial increase in their ETC.