Temperature Dependence of SEI Formation and Faradaic Efficiency in Electrochemical Lithium Mediated Nitrogen Reduction to Ammonia

Lithium-mediated nitrogen reduction (Li-N2R) is a promising electrochemical ammonia production alternative to the Haber Bosch process, a process responsible for 1.3% of CO2 emissions yearly. 1 While Haber Bosch requires high temperatures and pressures in large centralized facilities, Li-N2R works at near-ambient temperatures and pressures allowing for decentralized production and can be powered by renewable electricity sources. In Li-N2R, Li ions are plated as Li metal on the working electrode, which then reacts with N 2 gas to form lithium nitride (Li 3 N). Lithium nitride reacts with a proton source to form ammonia (NH 3 ) (Figure left). Past research indicates that the solid electrolyte interface (SEI) formed in the electrochemical system has a large impact on the NH 3 faradaic efficiency (FE) of the cell. 2 Variations in the temperature during cell cycling can influence both the thickness and composition of the SEI. This, in turn, alters the kinetics of species transport to and from the electrode. Consequently, SEIs formed at different temperatures lead to varying proportions of products in the reaction, and are a helpful tool to tune the system towards NH 3 selectivity. We elucidate temperature effects on SEI formation and FE for electrolytes with 1M LiBF 4 in two common solvents, Tetrahydrofuran (THF) and diglyme (DG), with a 1v% ethanol (EtOH) proton source (Figure right). We also confirm the validity of our results with an Ar control. By dissolving SEI species in a D 2 O rinsate, we can quantify the composition of the SEI formed at different temperatures using a combination of nuclear magnetic resonance (NMR), inductively coupled plasma mass spectrometry (ICP-MS), and ion chromatography (IC). With our strategy, we correlate SEI composition trends with changes in FE, and provide pathways towards engineering the SEI for enhanced Li-N2R FE to NH 3 and improved electrolyzer lifetime. 1 J. W. Erisman et al., Nat. Geosci 2008, 1, 636–639 2 K. Steinberg et al., Nat. Energy 2022, 8 (2), 138 Figure 1