Ammonia is a vital fertilizer component and offers a carbon-neutral alternative to liquid fuels when synthesized sustainably. A promising method for ammonia production is the electrochemical reduction of atmospheric nitrogen and water, powered by renewable electricity. However, a significant challenge lies in identifying a catalyst that effectively promotes ammonia formation while suppressing the hydrogen evolution reaction in aqueous electrolytes. This issue is exacerbated by discrepancies between theoretical predictions and experimental outcomes, which hinder the field’s progress.
In this presentation, we critically assess the methodologies and assumptions in theoretical studies of nitrogen reduction, suggesting key improvements. Over the past decade, we have employed density functional theory (DFT) calculations to explore transition metal ceramics, including nitrides, oxides, sulfides, carbides, oxynitrides, and carbonitrides, in search of materials that can catalyze the nitrogen reduction reaction (NRR) while inhibiting the hydrogen evolution reaction (HER). This has led to the identification of several promising candidates.
We synthesized catalysts as thin films using magnetron sputtering and tested their electrocatalytic performance in a microreactor system connected to an ammonia detection unit, ensuring reliable, contamination-free results. Experiments were conducted in both N2-saturated and Ar-saturated electrolytes, using isotope-labeled 15N2 to validate catalysis. We will discuss the correlation between theoretical predictions and experimental performance for these NRR candidates, as well as the application of deep neural networks in the discovery of novel catalysts, highlighting common pitfalls to avoid.