Determining the Composition and Structure of a High-Performing SEI Layer for Lithium-Mediated Nitrogen Reduction to Ammonia

Valerie Anne Niemann,Mathieu Doucet,Hanyu Wang,Niklas Henrik Deissler,Peter Benedek,Jon Bjarke Valbaek Mygind,Sang-Won Lee,William Abraham Tarpeh,Ib Chorkendorff,Adam C. Nielander,Thomas F. Jaramillo

DOI: https://doi.org/10.1149/ma2023-02602902mtgabs

2023-12-22

ECS Meeting Abstracts

Abstract:The industrial synthesis of ammonia (NH 3 ) via the Haber-Bosch process has enabled mass production of fertilizer and the human population growth to 8 billion. 1 NH 3 also shows promise as a long-term, energy-dense fuel amenable to widespread distribution. 2 However, the Haber-Bosch process operates at high temperatures and pressures in centralized facilities and accounts for 1.3% of global carbon emissions. 1,2 To ensure a food-secure future and realize NH 3 as a truly carbon-free fuel, we must develop an NH 3 synthesis alternative that operates at ambient conditions and can be driven by renewables. One of the most promising NH 3 synthesis method under mild conditions is the electrochemical, Li-mediated process (Li-NRR). In this process, Li is electrodeposited from solution and reacted with N 2 to form Li 3 N. A proton shuttle, e.g., ethanol, protonates Li 3 N to form NH 3 . 3 The electrodeposited Li forms a solid-electrolyte interphase (SEI) layer that controls N 2 , Li + , and H + transport and is key to system performance. The use of fluorinated Li salts, e.g., LiBF 4 , have recently advanced Li-NRR system performance. Improved selectivity has been attributed to changes in the SEI layer, but evaluating SEI structure and composition remains a significant challenge. 3 Neutron-based characterization methods, complementary to X-ray methods, offer unique benefits to study SEI layers containing light elements such as Li- and H; for example, an element’s neutron scattering cross section does not depend on its atomic number. 4 In this work, we studied the formation of the SEI layer in-situ before and after applying -0.3 mA/cm 2 in galvanostatic experiments (chronopotentiometry, CP) on a copper (Cu) working electrode. We utilized a combination of deuterated and non-deuterated solvents and proton donors to distinguish their contributions to the SEI. We found that the SEI forms two layers: a compact, lithium rich layer, and a diffuse layer with carbonaceous species (Figure 1). 4 We also identified the presence of boron in the layers, indicated by neutron absorption. Finally, we observed more proton-rich species in experiments containing the non-deuterated H + donor, corroborating previous studies that indicate that Li-ethoxide may be a major component of the diffuse SEI (Figure 1b-c). 5 Compared to LiClO 4 -based electrolytes, the structure of these LiBF 4 -derived layers is consistent even after repeated current cycling, indicating a more stable SEI structure and suggesting a relationship between SEI stability and increased Li-NRR performance. References Nørskov, J. et al. DOE Roundtable Report (2016). Chang, F. et al. Advanced Materials 33 , 2005721 (2021). Li, S. et al. Joule 6 , 2083–2101 (2022). Blair, S.J., Doucet, M. et al. ACS Energy Lett. , 7 , 6, 1939-1946 (2022). Steinburg, K. et al . Nature Energy , 8 , 138-148 (2023). Figure 1 : (a) reflectivity curve before and after 4 mins constant applied current at -0.3 mA/cm 2 (chronopoteniometry, CP) to plate Li using 1 M LiBF 4 with 1 vol% ethanol, (b) resulting scattering length density (SLD) profile, and (c) schematic of layers indicated in SLD profile Figure 1