It has been shown for Ni and Co thin-film water oxidation catalysts using surface enhanced Raman spectroscopy (SERS) that superoxo-like (“active oxygen”) species are formed during oxidizing potentials, which is visible as a broad band centered around 1050 cm-1. 2 The Raman bands are sensitive to isotope labelling, which have been used to quantify O16/O18 isotope exchange and the degree of lattice oxygen participation in OER electrocatalysis.3,4 These studies are in agreement with isotope labelling experiments using differential electrochemical mass spectrometry (DEMS), showing that only a small fraction of lattice oxygens exchange/participate during OER (proposedly only the surface sites).5,6 Herein, we probe both the redox-active oxygen species and estimate the degree of lattice oxygen participation in Fe-doped Ni, Co, and Mn electrocatalysts.
First, we evaluate soft X-ray absorption data collected at the O K-edge and metal L-edges during operando water oxidizing conditions, from where we derive the M-O covalency.7 The covalency is extracted from the O K pre-edge area (below 533 eV) normalized to the eg and t2g electron hole states. The higher the intensity under the pre-edge, the higher the M-O covalency, and the higher the probability of electrons being extracted from oxygen states close to the Fermi level (“anionic redox”). Such processes have been observed in Ni, Mn and Ir based catalysts in previous work.8–11 We further utilize SERS in attempt to correlate the superoxo-like species with the M-O covalency. We combine this with O16/O18 isotope labelling to obtain information on the lattice oxygen exchange during water oxidation.
To summarize our findings, we observe that Fe lowers the M-O covalency in all bimetallic Fe-doped catalysts (Ni-Fe, Co-Fe, Mn-Fe), and confirms a low degree of lattice oxygen participation in all catalysts. Surprisingly, largest O exchange is found in the non-catalytic state, especially in the monometallic Ni catalyst. Our data suggests that Fe suppresses lattice O exchange and lowers the M-O covalency, which correlates with a higher catalytic oxygen evolution activity. The highest population of active oxygen is found in the monometallic catalysts, which seems to scale with the population of oxidized M3+/4+ states rather than with the catalytic activity. Our study demonstrates that both anionic redox and lattice oxygen participation is minimal in catalysts with high OER activity, which shines new light onto the catalytic cycle in this type of TM oxide electrocatalysts and question the current understanding of active oxygen species in alkaline water oxidation.
References
- Nocera, D. G., Inorg. Chem., 48, 10001–10017 (2009).
- Trześniewski, B. J. et al., J. Am. Chem. Soc., 137, 15112–15121 (2015).
- Lee, S., Bai, L. & Hu, X., Angew. Chem. Int. Ed., 59, 8072–8077 (2020).
- Lee, S., Banjac, K., Lingenfelder, M. & Hu, X., Angew. Chem. Int. Ed., 58, 10295–10299 (2019).
- Roy, C. et al., Nat. Catal., 1, 820–829 (2018).
- de Araújo, J. F., Dionigi, F., Merzdorf, T., Oh, H. & Strasser, P., Angew. Chem. Int. Ed., 60(27),14981-14988 (2021).
- Suntivich, J. et al., J. Phys. Chem. C, 118, 1856–1863 (2014).
- Pfeifer, V. et al., Chem. Sci., 8, 2143–2149 (2017).
- Tesch, M. F. et al., Angew. Chem. Int. Ed., 58, 3426–3432 (2019).
- Drevon, D. et al., Sci. Rep., 9, 1532–1532 (2019).
- Ebrahimizadeh Abrishami, M. et al., Materials 9, (2016).