Fundamental analysis of the influence of the geometrical parameters on the transport properties of Gas Diffusion Electrodes

This paper deals with a fundamental analysis of the structure–effective property relationship in gas diffusion electrodes (GDE) that are gaining importance in electrocatalytic devices for energy and materials. First, a virtual representation of the carbon-paper based gas diffusion layer (GDL) is reconstructed combining computer-aided design with advanced image processing techniques. Next, a rigorous validation of the stochastic geometry in terms of its morphological and transport properties is carried out using benchmark experimental data for commercially available GDL materials. Once validated, three-dimensional (3D) transport based diffusion and electronic/thermal conduction simulations are carried out to study the dependence of the GDL effective properties on the geometrical meso-scale parameters such as fiber diameter, binder volume fraction, fiber density per layer. The presented analysis shows that the effect of the variation of these individual parameters are equivalent, as long as the bulk porosity of the geometry is kept constant, thus validating the usability of two-parameter porosity–tortuosity based correlations for the GDL. As the final step, the simulation data are used to develop accurate structure–property correlations for the GDL. The developed correlations provide more realistic estimates of the GDL effective properties, compared to their semi-empirical counterparts such as the Bruggeman correlation. This improvement is attributed to the retention of the information regarding the underlying true tortuosity and non-isotropic/non-spherical pore space, in the functional forms of the developed correlations. The refined correlations in this work can be applied to improve the prediction accuracy of computationally lean lumped-parameter GDE-based electrolyzer models.