Proceedings of MATSUS Spring 2025 Conference (MATSUSSpring25)
DOI: https://doi.org/10.29363/nanoge.matsusspring.2025.308
Publication date: 16th December 2024
Ammonia is a key fertilizer ingredient and, if synthesized in a sustainable way, a carbon-free alternative to liquid fuel. An alluring prospect is the production of ammonia via the electrochemical reduction of atmospheric nitrogen and water, driven by renewable electricity. However, a catalyst that facilitates the formation of ammonia while simultaneously hindering the parasitic side reaction of hydrogen evolution in aqueous electrolytes, has yet to be identified. Moreover, the persistent disagreement between theoretical predictions and experimental results hinders the further development of the field. Herein, we scrutinize the methodology and assumptions used in the bulk of theoretical studies on nitrogen reduction, suggesting improvements where applicable.
Over the last decade we have been searching, using density functional theory (DFT) calculations, for alternative materials that can catalyze nitrogen reduction reaction (NRR) while suppressing hydrogen evolution reaction (HER). The class of materials we have investigated are transition metal ceramics of e.g. nitrides, oxides, sulfides, carbides, oxynitrides and carbonitrides. Several promising candidates are predicted within each class of materials and we have tested several of them experimentally. There, we grow the catalysts in thin-films using magnetron sputtering, which are then tested in a micro reactor for electrocatalytic performance. The electrochemical micro reactor is connected in-line with the ammonia detection unit, preventing any possible contamination which makes the results reliable and robust. Experiments are done both in N2 saturated electrolyte and in Ar saturated electrolyte and isotope labelled 15N2 is used to proof catalysis. In this presentation, we discuss both the theoretical predictions and the experimental performance of several candidates for NRR. Finally, we explore the extent to which deep neural networks can be used to aid in the search for a novel catalyst, highlighting the pitfalls that must be avoided.