Porous and heterogeneous materials are core components in energy conversion and storage devices such as batteries, fuel cells and electrolyzers, or photoelectrochemical fuel generators. The heterogeneity and structural complexity of: i) the multi-functional nature of the applications requiring the presence of various functional materials in close vicinity, ii) nano- and micron-scale structuring of the material required to overcome the bulk material transport limitations, and iii) cheap and simple synthesis methods resulting in stochastic and complex morphologies. Understanding of the multi-physical transport phenomena and optimization of the component for enhanced performance, requires an accurate modelling and prediction of the transport properties, which heavily rely on the complex nano to micron-scale morphology.

In this presentation, I will show how tomography-based direct numerical simulation scan be used for the accurate numerical characterization of the heterogeneous components’ transport properties. We will use X-ray micro-tomography for the characterization of the (thermal) transport in partially saturated gas diffusion layers/electrodes or in porous thermochemical reactors, and FIB-SEM nano-tomography for the multi-physical transport characterization of photoelectordes for water splitting and catalyst layers of CO₂ reducing gas diffusion electrodes. I will show how we have built up a digital library of (photo)electrodes. Furthermore, I will show how machine learning approaches can be used to guide the design of optimized structures.