Ammonia electrosynthesis is considered promising for the production of ammonia as an energy carrier, as the process could provide greater flexibility and compatibility to the fluctuations of renewable energy than the industrial Haber-Bosch process. One approach for enhancing the ammonia formation rate is the electrochemical promotion of catalysis (EPOC) [1], in which a non-faradaic acceleration of ammonia synthesis is observed; however, much of the mechanism remains unclear. Further mechanistic insight would aid in designing better electrode structures that are suited for efficient ammonia electrosynthesis. In this presentation, we explore the mechanism underlying the electrochemical promotion of ammonia synthesis, using electrochemical cells with iron-based cathodes on a proton-conducting BaCe_0.9Y_0.1O₃ (BCY) electrolyte.

Previously, our group reported that the ammonia formation rate of a cathodically polarized cell can be enhanced by adding H₂ to the cathode gas feed (commonly pure N₂) [2, 3]. Further kinetic and isotopic analysis using deuterium showed that the primary hydrogen source for the ammonia produced was the gaseous hydrogen feed, rather than the protons conducted via the electrolyte [4]. Cells with a porous pure Fe cathode resulted in higher ammonia formation rates than those with a porous 10 wt.% Fe-BCY cathode, suggesting that the pure Fe surfaces near the cathode-electrolyte interface, as opposed to the triple phase boundaries, are active for ammonia electrosynthesis [3]. However, the 10 wt.% Fe-BCY cathode, fabricated by the impregnation method, was thought to be limited by lack of Fe network formation needed for electron conduction.

In this study, we fabricated a cermet cathode with the composition 56 wt.% Fe-BCY, aiming for Fe and BCY to form separate networks for electron and proton conduction, respectively. Electrochemical cells of the configurations Pt|BCY|Fe and Pt|BCY|Fe-BCY [Fig. 1a], with similar Fe mass (0.5 and 0.4 mg) were compared. Ammonia synthesis was conducted in a single-chamber reactor at 500-600˚C with a 50% H₂/50% N₂ feed. The gaseous ammonia produced was captured by bubbling into a dilute H₂SO₄ solution and quantified via high-performance liquid chromatography.

Using the pure Fe electrode, an ammonia formation rate of 1.3 × 10^-8 mol s^-1 cm^-2 (26 mmol g_Fe^-1 h^-1) was obtained at 600˚C and -1 V [Fig. 1b]. This corresponds to an almost 5-fold increase in ammonia formation compared to the catalytic reaction at 0 V. In contrast, a lower ammonia formation rate of 7.0 × 10^-9 mol s^-1 cm^-2 (17 mmol g_Fe^-1 h^-1) was obtained using the Fe-BCY electrode at the same conditions. Despite the cermet electrode having a slightly higher current density, and despite the comparable iron loading on the two cells, the pure Fe cell still resulted in a faster ammonia formation rate per electrode area and per iron mass. As the cermet electrode is expected to have a significantly longer triple phase boundary length, from the comparison of these two electrochemical cells, the triple phase boundaries do not appear to be the primary active sites. We further analyze the roles of the bulk Fe surfaces and the Fe/BCY triple phase boundaries, by examining the effect of the electrode thickness and by post-experiment cross-sectional electron microscopy approaches.

Acknowledgements
This work was supported by JSPS KAKENHI Grant Number JP21H04938.

References
[1] P. Vernoux et. al., Chem. Rev. 113, 8192-8260 (2013).
[2] F. Kosaka, T. Nakamura, A. Oikawa, J. Otomo, ACS Sustainable Chem. Eng., 5, 10439-10446 (2017).
[3] C.-I. Li, H. Matsuo, J. Otomo, Sustain. Energy Fuels, 5, 188-198 (2021).
[4] C.-I. Li, H. Matsuo, J. Otomo, RSC Adv., 11, 17891-17900 (2021).