Investigation of the property hull for solid oxide fuel cell microstructures

Abstract

Designing optimal microstructures for solid oxide fuel cell (SOFC) electrodes is a subtle task owing primarily to the multitude of the electro-chemo-physical phenomena taking place simultaneously that directly affect working conditions of a SOFC electrode and its performance. In this study, a novel design paradigm is presented to obtain desired triple phase boundary length (TPBL), ionic/electronic phase conductivity and gas diffusion properties for a SOFC cell. The method builds on top of a previously developed methodology for digital realization of generic microstructures. The obtained realizations are then used to predict TPBL, ion conductivity and gas diffusion of their representing SOFC electrodes. Study follows by building a database obtained from these realizations to train a neural network that relates input geometrical parameters to these three properties. It is shown that the presented methodology allows one to obtain property and microstructure hull for SOFC electrodes and finally achieve optimal microstructure. The results indicate that the gas diffusion for SOFC electrodes is strongly depended on the geometry of the microstructure, while TPBL and ionic conductivity, to less extents, are related to the geometry. The analysis also shows that there exists a trade-off between ionic conductivity, TPBL and diffusion factor.