Process based reconstruction and simulation of a three-dimensional fuel cell catalyst layer

Abstract

Fundamental understanding of catalyst layer nanostructure of hydrogen polymer electrolyte membrane (PEM) fuel cells is critical for improvement in performance and durability. A process based 3D mathematical model has been developed to elucidate the effect of electrode composition, porosity and ionomer weight fraction in catalyst layers on electrochemical and nano-scale transport phenomena. Numerical reconstruction of catalyst layer random structure has been performed through a controlled random algorithm, mimicking the experimental fabrication process. Nano-scale species transport properties, e.g., Knudsen diffusion of oxygen in nano-pores and proton transport in thin-film electrolyte, have been included in the model, allowing for more rigorous study of the catalyst layer. It was found that there is a threshold in both porosity and ionomer weight fractions, below which species percolation through the random structure becomes difficult due to reduced connectivity and increased isolation. The degree of mixing or size of agglomerates has been studied and it was discovered that increasing or decreasing the agglomerate number from the optimum value reduces the electrochemically active area (ECA) and deteriorates species transport, suggesting an optimum level of stirring of the catalyst ink during catalyst layer preparation is critical.