a geometric modeling approach to find the best microstructure for infiltrated sofc electrodes

In this study, a novel design paradigm is presented to obtain some geometryrelated electrochemical and physical properties of an infiltrated sofc electrode. a range of digitally realized microstructures with different backbone geometric properties and virtual electrocatalyst particle loadings under various deposition conditions are generated. triple phase boundary (tpb), the active surface density of particles and gas transport factor are evaluated in those realized models based on selected infiltration strategy. based on this database, a neural network is trained to relate the desired range of input geometric parameters to a property hull. the effect of porosity and geometric anisotropy in backbones in addition to the loading, distribution and aggregation behavior of particles is systematically investigated on those performance indicators. the results indicated that microstructures with very high amount of tpb and contact surface density of particle have a relatively low gas diffusion factor, meanwhile increasing these parameters does not involve  a sensible contradiction. also, by adding particles, the tpb density variation is changed as a function of backbone porosity and the average shape of aggregated particles. a direct search into the microstructure and property hull is applied to find the best parameters in modeling approach aiming the maximum effective geometric properties. finally, a genetic algorithm is employed to detect appropriate geometric factors getting the maximum acquirable performance in infiltrated sofc electrodes.