Porous materials can be characterized by well-trained neural networks. In this study, fibrous paper-type gas diffusion layers were trained with artificial data created by a stochastic geometry model. The features of the data were calculated by means of transport simulations using the Lattice–Boltzmann method based on stochastic micro-structures. A convolutional neural network was developed that can predict the permeability and tortuosity of the material, through-plane and in-plane. The characteristics of real data, both uncompressed and compressed, were predicted. The data was represented by reconstructed images of different sizes and image resolutions. Image artifacts are also a source of potential errors in the prediction. The Kozeny–Carman trend was used to evaluate the prediction of permeability and tortuosity of compressed real data. Using this method, it was possible to decide if the predictions on compressed data were appropriate.

Mixtures of ethanol and methanol being synthesized from CO2 and green H2 can serve as sustainable base chemicals for a number of chemical processes. Amongst these processes, the catalytically-supported synthesis of CO2-neutral C4 to C10 alcohols is of increasing importance as, e.g., iso-butanol can be used as a drop-in fuel or after dehydration to produce iso-butene as a feedstock for the synthesis of plastics. 2-ethyl-hexanol can be further refined into solvents, tensides, or monomers. In this respect, NiPt alloys on an activated carbon support were found to be active and stable catalysts for the synthesis of iso-butanol following the Guerbet reaction scheme. In this study, two different routes are applied to the synthesis of these NiPt catalysts: a more conventional one based on the impregnation of Ni and Pt salts and an advanced path with a surface redox reaction between elemental Ni on the support and Pt ions in a polar solution. The experimental evaluation showed that the Pt particles from the surface redox reaction being exposed on the Ni particles are more active than those on the impregnated catalysts due to their high surface energy. Their specific space-time yields were 10–20 times higher.