Image analysis and modeling of spherical and channel microstructures of fuel-cell materials

The development of novel fuel-cell materials demands accurate and flexible microstructure characterization techniques. Conventional electron microscopy-based microstructural morphology analysis is carried out through the conceptual interpretation of transmission electron microscope images. With this method, only qualitative information on material morphologies can usually be obtained. This paper presents a digital image analysis system that deals with the automatic measurement and quantitative characterization of the microstructural morphologies of polymer electrolyte membrane fuel-cell materials. In this approach, two types of essential microstructural morphologies (spheral particles and interconnected graft channels) are modeled based on statistical geometry theory, and the statistical analysis schemes of the microstructural morphologies are designed and applied to the characterization of the phase-separated microstructures in fuel-cell components such as solid electrolyte ionomers, catalyst layers, and gas diffusion layers. Experimental results on real fuel-cell materials specimens demonstrate the effectiveness of the method.