A Representative Volume Element Approach for Pore-Scale Modeling of Fuel Cell Materials

Abstract

The objective of this work is to utilize the recent advances in microstructure quantification to select small volumes (referred to as representative volume elements, or "RVEs") for use in pore scale modeling, which accurately reflect the overall microstructure and transport properties of large fuel cell material datasets. The micro-porous layer (MPL) of polymer electrolyte fuel cells is chosen for demonstration. Focused ion beam scanning electron microscopy is utilized to obtain a 3-D structural dataset for the selected MPL sample. Using n-point statistics, RVEs are selected from the full dataset to reflect the ensemble averaged statistics of the full dataset to within acceptable tolerance. Metric comparisons between RVEs and the full dataset indicate that the selected RVEs offer a very good representation of the full dataset, albeit in a volume that is significantly smaller in spatial extent, therefore providing a computationally efficient and reliable model domain for pore-scale modeling efforts.