Bayesian Optimization for Improved Electrode Conditioning: Ni(-Fe) Electrodes for the Alkaline Oxygen Evolution Reaction

ECS Meeting Abstracts, vol. MA2026-01, pp. 2669

Abstract

In alkaline water electrolysis, the sluggish kinetics of the oxygen evolution reaction (OER) at the anode still cause significant overpotentials. Thus, electrode materials with enhanced activity are desirable. NiFeO x H y materials are promising OER electrocatalysts. Their activity was shown to be strongly influenced by electrochemical conditioning using potential cycling [1]. To determine a good electrode conditioning procedure, several studies have tested a set of potential cycling profiles or manually optimized the potential cycling parameters by varying one parameter at a time [1-4]. However, a systematic approach to optimize the conditioning process to achieve the best electrode activation is still missing. The use of mathematical models presents a promising strategy to develop such an approach. While mechanistic models are currently unavailable and not easily developed, data-driven models might offer a straightforward alternative. To this end, we propose the use of Bayesian Optimization with Gaussian Processes to improve the electrode conditioning process of a Ni-Fe bulk electrode [5]. We show how this approach was used to iteratively improve the electrode conditioning process for the Ni-Fe electrode and enabled a stronger activity enhancement compared to our previous manual optimization [1]. At the same time, it also required fewer experiments. We further extended the approach to transfer knowledge to new, but similar materials. To showcase this, we optimized the conditioning of a Ni electrode: The applied transfer learning approach started with the experimental data obtained for the Ni-Fe electrode and iteratively added experimental data for the Ni electrode. This required fewer experiments than applying Bayesian Optimization from scratch. All in all, Bayesian Optimization with Gaussian Processes appears promising for a systematic optimization of electrode conditioning. References: [1] C. Gohlke et al. (2024). ChemElectroChem, 11(18), e202400318. [2] M. E. G. Lyons et al. (2012). Journal of The Electrochemical Society, 159(12), H932-H944. [3] Y. J. Son et al. (2022). ACS Catalysis, 12(16), 10384–10399. [4] Y. Zuo et al. (2024). Advanced materials, 36(21), e2312071. [5] J. R. Seidenberg et al. (2025). ChemElectroChem. In press. DOI: 10.1002/celc.202500284. Acknowledgements: The authors gratefully acknowledge the financial support from the German Federal Ministry of Research, Technology and Space (BMFTR project ''PrometH2eus'', FKZ 03HY105A).

Authors 6

1. J. Raphael Seidenberg Aachen Process Systems Engineering (AVT.SVT)

   RWTH Aachen University · KU Leuven

   Affiliation as printed

   Department of Chemical Engineering, KU Leuven

   Process Systems Engineering (AVT.SVT), RWTH Aachen University

2. Clara Gohlke Aachen Electrochemical Reaction Engineering

   RWTH Aachen University

   Affiliation as printed

   Electrochemical Reaction Engineering, RWTH Aachen University

3. Raphael Diebold Aachen Electrochemical Reaction Engineering

   RWTH Aachen University

   Affiliation as printed

   Electrochemical Reaction Engineering, RWTH Aachen University

4. Anna K. Mechler Aachen Electrochemical Reaction Engineering

   Forschungszentrum Jülich · RWTH Aachen University

   Affiliation as printed

   Electrochemical Reaction Engineering, RWTH Aachen University

   Forschungszentrum Jülich GmbH, IET-4

5. Alexander Mitsos Aachen Process Systems Engineering (AVT.SVT)

   Forschungszentrum Jülich · RWTH Aachen University

   Affiliation as printed

   Energy Systems Engineering (ICE-1), Forschungszentrum Jülich

   Process Systems Engineering (AVT.SVT), RWTH Aachen University

6. Dominik Bongartz

   KU Leuven

   Affiliation as printed

   Department of Chemical Engineering, KU Leuven

   EnergyVille