Stochastic microstructural modeling of fuel cell gas diffusion layers and numerical determination of transport properties in different liquid water saturation levels

•A stochastic, microstructural GDL modeling framework is developed and validated.•Anisotropic GDL properties are determined in dry and partially saturated states.•PTFE and liquid water are found to have significant impact on GDL properties.•The modeling framework can be employed as a GDL design and prototyping tool.

Gas diffusion layer (GDL) materials in polymer electrolyte membrane fuel cells (PEMFCs) are commonly made hydrophobic to enhance water management by avoiding liquid water blockage of the pores and facilitating reactant gas transport to the adjacent catalyst layer. In this work, a stochastic microstructural modeling approach is developed to simulate the transport properties of a commercial carbon paper based GDL under a range of PTFE loadings and liquid water saturation levels. The proposed novel stochastic method mimics the GDL manufacturing process steps and resolves all relevant phases including fiber, binder, PTFE, liquid water, and gas. After thorough validation of the general microstructure with literature and in-house data, a comprehensive set of anisotropic transport properties is simulated for the reconstructed GDL in different PTFE loadings and liquid water saturation levels and validated through a comparison with in-house ex situ experimental data and empirical formulations. In general, the results show good agreement between simulated and measured data. Decreasing trends in porosity, gas diffusivity, and permeability is obtained by increasing the PTFE loading and liquid water content, while the thermal conductivity is found to increase with liquid water saturation. Using the validated model, new correlations for saturation dependent GDL properties are proposed.