Pore scale modelling of a cathode catalyst layer in fuel cell environment: agglomerate reconstruction and variables optimization

Abstract

In this work, a new strategy is adopted to optimize the cathode catalyst layer of a PEM fuel cell and to analyse the impact of three pertinent design variables. A pore scale, three-dimensional model is developed to predict the performance of a single agglomerate of the catalyst layer. The pore scale modelling is implemented as a two-step procedure. First, the microstructure of the catalyst layer is computationally reconstructed and then the governing and constitutive equations can be discretized and numerically solved. The agglomerate reconstruction follows a stochastic approach of the catalyst layer structure, and random routines are added to capture the variability and imperfections of the fabrication methods. From the obtained results, it is concluded that the design variables with the highest impact on performance were the carbon particle diameter and the Nafion volume fraction. In contrast, the agglomerate diameter did not show any impact on performance for the tested operational conditions. Another important finding was related with the influence of the operational pressure on the design variables performance impact. As pressure increased, the design variables contribution to the potential variability vanished. This work also demonstrated that performance was markedly affected by the random nature of the agglomerate structure.