Nonaqueous Li-Mediated Nitrogen Reduction: Taking Control of Potentials

Abstract

The performance of the Li-mediated ammonia synthesis has progressed dramatically since its recent reintroduction. However, fundamental understanding of this reaction is slower paced, due to the many uncontrolled variables influencing it. To address this, we developed a true nonaqueous LiFePO4 reference electrode, providing both a redox anchor from which to measure potentials against and estimates of sources of energy efficiency loss. We demonstrate its stable electrochemical potential in operation using different N2- and H2-saturated electrolytes. Using this reference, we uncover the relation between partial current density and potentials. While the counter electrode potential increases linearly with current, the working electrode remains stable at lithium plating, suggesting it to be the only electrochemical step involved in this process. We also use the LiFePO4/Li+ equilibrium as a tool to probe Li-ion activity changes in situ. We hope to drive the field toward more defined systems to allow a holistic understanding of this reaction.

Water Increases the Faradaic Selectivity of Li-Mediated Nitrogen Reduction

“…They suggested that Li 2 O species formed in the SEI reduced Li + diffusivity in the SEI, thus kinetically suppressing excessive lithium plating, a significant parasitic side-reaction, as well as other electrolyte-degrading side reactions . Quantitative electrochemical measurements of suppression of Li plating will be possible given recent developments of more accurate reference electrodes for the Li-mediated system. , Our own XPS results are remarkably similar to those of Li et al, as shown in Figure a-d: increasing water generally shifts the peaks away from those corresponding to LiClO n under dry conditions (i.e., 533 eV for O 1s and 56.9 eV for Li 1 s) to those corresponding to toward Li 2 O under moist conditions (i.e., 531.5 eV for O 1s and 55.3 eV for Li 1 s. Our results suggest that the replacement of LiClO n , LiCl, and related species with nonchlorinated species such as Li 2 O, consistent with earlier reports from the battery literature on the effect of trace H 2 O on the SEI . Increasing the water concentration beyond the optimum appears to increase surface Li atomic concentration (46% to 54%) and decrease O concentration slightly (41% to 34%), which could indicate increased Li 2 O formation, as shown in Figure d.…”

“… 24 Quantitative electrochemical measurements of suppression of Li plating will be possible given recent developments of more accurate reference electrodes for the Li-mediated system. 25 , 26 Our own XPS results are remarkably similar to those of Li et al, as shown in Figure 3 a-d: increasing water generally shifts the peaks away from those corresponding to LiClO n under dry conditions (i.e., 533 eV for O 1s and 56.9 eV for Li 1 s) to those corresponding to toward Li 2 O under moist conditions (i.e., 531.5 eV for O 1s and 55.3 eV for Li 1 s. Our results suggest that the replacement of LiClO n , LiCl, and related species with nonchlorinated species such as Li 2 O, consistent with earlier reports from the battery literature on the effect of trace H 2 O on the SEI. 4 Increasing the water concentration beyond the optimum appears to increase surface Li atomic concentration (46% to 54%) and decrease O concentration slightly (41% to 34%), which could indicate increased Li 2 O formation, as shown in Figure 3 d. The decrease in Faradaic efficiency at higher water concentrations may be due to excessive proton activity in moist conditions, which favors hydrogen evolution as a side reaction, as measured by Tsuneto et al at higher ethanol concentrations, 1 and proposed by Nørskov and co-workers in theoretical work, 15 , 27 and may still modify SEI bulk characteristics, such as porosity.…”

The origin of overpotential in lithium-mediated nitrogen reduction

“…Since acceptance of this manuscript, we have developed a stable reference electrode based on a half-lithiated battery cathode material, Lithium Iron Phosphate (LFP). [32][33][34] This has been shown to have a stable open circuit even under oxygen and hydrogen atmosphere. [32][33][34] Previous work evaluating hydrogen redox has used an Ag wire pseudo-reference calibrated against the ferrocene redox couple.…”

“…[32][33][34] This has been shown to have a stable open circuit even under oxygen and hydrogen atmosphere. [32][33][34] Previous work evaluating hydrogen redox has used an Ag wire pseudo-reference calibrated against the ferrocene redox couple. 19 However, our initial experiments show that an Ag wire pseudo-reference was not stable in a 0.2 M LiBF 4 1% vol ethanol in THF electrolyte under nitrogen reduction conditions (see ESI Fig.…”

Quantifying Influence of the Solid-Electrolyte Interphase in Ammonia Electrosynthesis

“…Methods . LiMEAS experiments were conducted using a one-compartment cell with a Cu cathode, Pt mesh anode, and Li 0.5 FePO 4 reference electrode, , all submerged in 16 mL of electrolyte solution. 5% H 2 in N 2 gas was bubbled through the electrolyte at 5 mL·min –1 for 20 min prior to current passage and continued to flow while current was applied.…”