Christodoulos Chatzichristodoulou

The discovery of efficient, durable, and platinum group metal‐free electrocatalysts for alkaline water electrolysis is essential to reduce the cost of green hydrogen. Efficient exploration of the vast possible synthetic space is nevertheless hindered by the slow pace of manual experimentation. Here, a closed‐loop optimization is presented that combines an automated electrocatalyst synthesis and testing system, CatBot, with Bayesian optimization and accelerated stress testing at industrial conditions for the rapid discovery of active and durable NiMo coatings. Distinct regions with exceptionally high activity and stability are identified within limited experimental budgets on a nickel wire substrate, using different search strategies to efficiently navigate the six‐dimensional synthesis space exceeding possible combinations. The most promising coatings are upscaled onto technological nickel foam electrodes showing platinum‐level catalytic activity with overpotentials of 173 mV at –1 in 30 wt.% KOH at for the hydrogen evolution reaction. This work establishes a generalizable strategy for the autonomous optimization of functional electrocatalyst coatings under realistic operating conditions.

Operando X-ray photoelectron spectroscopy has emerged as a powerful technique for probing the surface composition and chemical state of electrocatalysts under working conditions. In this study, we report a strategy...

Proton exchange membrane (PEM) water electrolysis offers significant potential for sustainable production of green hydrogen due to its high current density, superior hydrogen purity, and excellent conversion efficiency. However, widespread industrial implementation remains challenging because state-of-the-art oxygen evolution reaction (OER) catalysts heavily rely on iridium, a precious metal with limited global supply and high costs. ^1–4 Therefore, considerable research efforts have been dedicated to developing new catalysts requiring lower iridium loadings. Although PEM electrolysis typically operates at elevated temperatures (50-80°C), ¹ most studies assessing catalyst activity and stability have been performed at room temperature, and the limited available temperature-dependent studies primarily focus on alkaline environments. ^5,6 In this work, we have systematically investigated the temperature-dependent activity and stability of two representative Ir-based catalysts: a commercial IrO₂ catalyst and a supported core-shell Ir@IrOₓ catalyst. Using a high-temperature rotating-disk electrode (RDE) setup, electrochemical assessments of the activity and stability of the catalysts were performed across a range of operating temperatures from 20°C up to 80°C. We investigated how the increased operating temperature affects the OER activities of these two catalysts, observing not only enhanced kinetics with the increased working temperatures but also altered reaction mechanisms, as indicated by the Tafel analysis of the OER polarization curves. Furthermore, by employing an accelerated stress test protocol, we evaluated the stability of the two catalysts under varied temperatures. Comprehensive in-situ and post-mortem characterization techniques, including in-situ XRD, XPS (Figure 1), TEM, and ICP, were employed to investigate the structural changes of the catalysts during the stability tests, which revealed the accelerated degradation at higher working temperatures, but also the surprisingly high OER stability of the core-shell Ir@IrOx structure under high working temperatures. Figure 1: Comparison the Ir 4f XPS spectra of the sample before and after the stability test at 60 °C. References Carmo, M., Fritz, D. L., Mergel, J. & Stolten, D. A comprehensive review on PEM water electrolysis. Int J Hydrogen Energy 38 , 4901–4934 (2013). Wang, C. & Feng, L. Recent advances and perspectives of Ir-based anode catalysts in PEM water electrolysis. Energy Advances 3 , 14–29 (2024). Kibsgaard, J. & Chorkendorff, I. Considerations for the scaling-up of water splitting catalysts. Nature Energy 2019 4:6 4 , 430–433 (2019). Reier, T., Oezaslan, M. & Strasser, P. Electrocatalytic oxygen evolution reaction (OER) on Ru, Ir, and pt catalysts: A comparative study of nanoparticles and bulk materials. ACS Catal 2 , 1765–1772 (2012). Shi, G. et al. Temperature Dependence of Oxygen Evolution Reaction Activity in Alkaline Solution at Ni-Co Oxide Catalysts with Amorphous/Crystalline Surfaces. ACS Catal 12 , 14209–14219 (2022). Lyu, X., Li, J., Yang, J. & Serov, A. Significance of slight ambient temperature variation on the electrocatalyst performance toward oxygen evolution reaction. J Environ Chem Eng 11 , 111492 (2023). Figure 1

Introduction The transition to sustainable energy systems is a global priority. With the Paris Agreement ¹ setting a goal of achieving net-zero emissions by 2050, there is an imperative need to develop efficient and greener energy conversion & storage technologies. One of the promising technologies addressing this challenge is Solid Oxide Electrolysis Cells (SOECs), which play a crucial role in Power-to-X applications and hard-to-abate sectors. SOECs operate at high temperatures (700–900°C) and are currently being tested at scales of several hundred kilowatts, with commercial units available in Europe and globally. SOEC technology has higher efficiency than low-temperature electrolysis, can operate in reverse mode as fuel cells, and can be integrated into other high-temperature processes or synthetic fuel production. State-of-the-art (SoA) SOECs typically use a nickel/yttria-stabilized zirconia (Ni/YSZ) fuel electrode, a YSZ electrolyte, and an oxygen electrode made of lanthanum strontium cobalt (LSC) or lanthanum strontium cobalt ferrite (LSCF). Although cobalt (Co)-based oxygen electrodes exhibit high catalytic activity, they present several challenges, including high thermal expansion coefficient (TEC) mismatch with other cell components, which can lead to delamination. Additionally, they are highly expensive and have limited chemical and redox stability. Moreover, cobalt is classified as a critical raw material (CRM) by the European Union (EU) under the EU Critical Raw Materials Act ² , due to its supply risk. Thus, there is a need for an alternative Cobalt-free electrode. This study focuses on characterizing La _0.6 Sr _0.4 FeO _{3- δ} (LSF) as a cobalt-free oxygen electrode. LSF has shown good performance and stability as reported in published literatures ^3–7 . Also, it is comparatively cheap and lean in CRMs. However, as of today, according to the authors’ knowledge, there is a lack of detailed studies on LSF oxygen electrodes integrated into SoA SOEC cells based on Ni-YSZ fuel electrodes and YSZ electrolytes and the long-term performance of such cells. Thus, durability testing with the underlying degradation mechanism for LSF as an oxygen electrode remains unclear on the full cell level. In this work, we fabricate Ni-YSZ fuel electrode supported SOEC cells, with PVD (Physical Vapor Deposition) CGO (Gadolinium doped Ceria) barrier layer and LSF oxygen electrode. The performance and durability of such cells are investigated via detailed electrochemical characterization combined with post-mortem SEM (Scanning electron microscopy) analyses. Methodology The single cell was fabricated using tape casting for the Ni-YSZ fuel electrode and YSZ electrolyte ⁸ , followed by the CGO barrier layer deposition through the PVD technique. Further, the LSF oxygen electrode was screen printed on the half-cell. The cells were sintered at high temperatures, and an LSM contact layer was applied to the oxygen electrode. The cell was tested using in-house DTU Energy rigs consisting of an alumina cell test house, a furnace, relevant gas controllers and sensors, and an electrochemical measurement system. The tested cells had a footprint of 5.3*5.3 cm ² , with an active area of 4*4 cm ² . A gold mesh was used as the current collector on the oxygen electrode, while a nickel mesh was used on the fuel electrode. During start-up, the cell was heated up to 850°C at 1°C/min in argon. For reduction, firstly the cell was reduced in safety gas (5% H ₂ in N ₂ ) for 2 hours, then in pure hydrogen with 20 L/h for 1 hour. The fingerprint test was performed at open circuit voltage (OCV) at various temperatures of 850°C,800°C,750°C,700°C and under different gas compositions. The fuel electrode was supplied with varying hydrogen/steam ratios, i.e. 96/4 (H ₂ /H ₂ O), 80/20 (H2/H2O), 50/50 (H2/H2O), and 10/90 (H ₂ /H ₂ O). While the oxygen electrode was supplied with either pure oxygen or unfiltered compressed air. After the fingerprint testing, a durability test was performed in electrolysis mode at -0.5 A/cm ² , 750°C, and 10/90 (H2/H2O) on the fuel side and unfiltered compressed air on the oxygen electrode with a fuel/steam utilization of 28%. The Electrochemical Impedance Spectroscopy (EIS) measurements were measured with a frequency response analyzer Solartron 1260, at a frequency range of 96850-0.1 Hz, with 12 points/decade and a total impedance time of 67 min. Results & Discussion The electrochemical performance analysis of the LSF oxygen electrode-based SOEC from the fingerprint testing is presented in Figures 1a and 1b. The EIS at various gas compositions and temperatures shows an expected increase in ohmic ( R _s ) and polarization ( R _p ) resistance with a decrease in temperature. The Bode plot reveals thermally activated characteristic frequency peaks, indicating processes influenced by temperature variations. In Figure 1a, at a temperature of 750°C and varying hydrogen/steam ratios, processes associated with the fuel electrode can be observed at a characteristic frequency range of 10³–10⁵ Hz, typically corresponding to bulk ionic transport and charge transfer reactions ⁹ . Figure 1b shows that when changing from air to pure oxygen at the oxygen electrode, at a given temperature and gas composition, the middle characteristic frequency range (10–300 Hz) clearly highlights processes occurring at the oxygen electrode, likely related to oxygen surface exchange kinetics and oxygen ion bulk transport ¹⁰ . The lower characteristic frequency peak (0.1–4 Hz) does not exhibit strong thermal activation but varies with changes in the gas atmosphere at the fuel electrode, as seen in Figures 1a and 1b. This suggests that the low-frequency region corresponds to physical processes such as mass transfer, gas conversion, and diffusion at the electrodes ¹¹ . Following fingerprint analysis, durability testing was conducted for 1050 hours under galvanostatic conditions. Figure 2a illustrates the voltage progression over time, showing significant degradation within the first 500 hours, with a degradation rate of 17.5%/kh (194.4 mV/kh). However, the cell voltage stabilized in the last 550 hours, with a reduced degradation rate of 5.2%/kh (62.9 mV/kh). The overall degradation rate for the 1050-hour test is calculated to be 11.3%/kh (125.5 mV/kh). During the durability test, impedance measurements were recorded every 12 hours under current conditions. Figure 2c shows Nyquist and Bode plots showing impedance evolution every 50 hours. The characteristic frequencies identified at OCV suggest that degradation in the fuel electrode is more pronounced during the first 500 hours but stabilizes in the later stages. Moreover, in the first 350 hours, the middle-frequency peak shifts to a higher frequency region, which might suggest the activation of the middle-frequency process, predominantly associated with processes at the oxygen electrode. From 350 to 1050 h, the middle-frequency peak shifts significantly to the lower-frequency region, which suggests oxygen electrode degradation. As the oxygen electrode is fed with compressed unfiltered air, there might be possibilities of surface poisoning through impurities from the air, such as sulfates, which can be a probable degradation mechanism for the oxygen electrode. This degradation due to impurities needs further investigation with microstructural characterization and surface analysis. Figure 2b shows the evolution of R _s and R _p with respect to time. R _s tends to be quite stable during the operation, indicating robust cells and reliable cell contacting. Meanwhile, there is an increase in R _p over time, as expected, with an R _p degradation rate of 0.26 Ω∙cm ² / kh. Conclusions This study presents the electrochemical performance and durability of cobalt-free LSF oxygen electrode-based solid oxide electrolysis cells. The cells were fabricated using tape casting and screen printing. The cells were further electrochemically tested at different temperatures and gas compositions. The fingerprint analysis confirms an increase in R _s and R _p with decreasing temperature. Characteristic frequency peaks in bode plots indicate different electrode processes, with high frequency (10³–10⁵ Hz) related to the fuel electrode, middle frequency (10-300 Hz) related to the oxygen electrode, and low frequency (0.1-5 Hz) related to mass transfer processes. Durability testing for over 1000 hours under galvanostatic conditions shows an overall degradation rate of 11.3%/kh (125.5 mV/kh). During the durability test, the R _s remains stable while R _p degrades at the rate of 0.26 Ω∙cm ² /kh. Overall, LSF shows potential as a cobalt-free oxygen electrode for SOEC, with promising performance and stability. Further investigation into microstructural characterization for degradation mechanisms and modifications to LSF electrodes can lead to improved performance. Thus, LSF can emerge as a potential CRM lean, cheap, and stable alternative for solid oxide electrolysis cells. References UNFCCC. THE PARIS AGREEMENT . https://unfccc.int/documents/184656 (2018). Commission, E., Directorate-General for Internal Market Entrepreneurship and SMEs, I., Grohol, M. & Veeh, C. Study on the Critical Raw Materials for the EU 2023 – Final Report . (Publications Office of the European Union, 2023). doi:doi/10.2873/725585. Khoshkalam, M. et al. Improving oxygen incorporation rate on (La0. 6Sr0. 4) 0.98 FeO3-δ via Pr2Ni1-xCuxO4+ δ surface decoration. J Power Sources 457 , 228035 (2020). Wang, W., Huang, Y., Jung, S., Vohs, J. M. & Gorte, R. J. A Comparison of LSM, LSF, and LSCo for Solid Oxide Electrolyzer Anodes. J Electrochem Soc 153 , A2066 (2006). Kammer Hansen, K. & Mogensen, M. Evaluation of LSF based SOFC Cathodes using Cone-shaped Electrodes. ECS Trans 13 , 153 (2008). Gibelli, M., Cordaro, G. & Donazzi, A. Preparation, Characterization, and Kinetic Testing of Infiltrated LSF-YSZ Electrodes for Symmetric Solid Oxide Cells. Ind Eng Chem Res 60 , 6639–6652 (2021). Zuo, X. et al. Molten Salt Synthesis of High-Performance, Nanostructured La0.6Sr0.4FeO3−δ Oxygen Electrode of a Reversible Solid Oxide Cell. Materials 13 , (2020). Hauch, A., Brodersen, K., Chen, M. & Mogensen, M. B. Ni/YSZ electrodes structures optimized for increased electrolysis performance and durability. Solid State Ion 293 , 27–36 (2016). Padinjarethil, A. K., Bianchi, F. R., Bosio, B. & Hagen, A. Electrochemical Characterization and Modelling of Anode and Electrolyte Supported Solid Oxide Fuel Cells. Front Energy Res 9 , 668964 (2021). Leonide, A., Sonn, V., Weber, A. & Ivers-Tiffee, E. Evaluation and Modelling of the Cell Resistance in Anode Supported Solid Oxide Fuel Cells. ECS Trans 7 , 521 (2007). Taubmann, J., Sun, X., Rizvandi, O. B. & Frandsen, H. L. Advanced insights into gas conversion and diffusion impedance of solid oxide cells by 2D multi-physics modelling. J Power Sources 588 , 233739 (2023). Disclaimer The authors acknowledge financial support from the SUSTAINCELL project which is supported by the Clean Hydrogen Partnership and its members under Grant Agreement No 101101479.Co-funded by the European Union. Views and opinions expressed are, however, those of the author(s) only and do not necessarily reflect those of the European Union or Clean Hydrogen JU. Neither the European Union nor the granting authority can be held responsible for them. Figure 1 Figure 2

Solid oxide electrolyzers and fuel cells (SOEC/SOFC) hold significant potential for reducing global carbon emissions. However, achieving greater durability and cost-effectiveness requires a deeper understanding of their degradation mechanisms during operation. Traditional post-test analyses offer limited insights into the dynamic degradation processes occurring in real-time. To address this challenge, we have developed an innovative method that integrates TEM characterization with simultaneous electrochemical testing, allowing a direct correlation between structural and compositional changes and electrochemical performance. This methodology employs a specialized TEM holder equipped with a heating-biasing MEMS chip and an environmental TEM capable of introducing reactive gases such as O ₂ , H ₂ , and H ₂ O, effectively mimicking real-world SOEC/SOFC operating conditions. A dedicated sample preparation technique enables the simultaneous application of heat, reactive gases, and electrical bias [1-4]. In this study, perovskite-based electrodes, including lanthanum-strontium-cobalt oxide (LSC) and lanthanum-strontium-ferrite (LSF), paired with yttrium-stabilized zirconia (YSZ) electrolytes and gadolinium-doped ceria (CGO) buffer layer, were investigated. To study degradation phenomena, the cells were polarized under 2.7 mbar O ₂ at 750 °C. (S)TEM imaging revealed phase transformations in negatively polarized perovskite electrodes and crack formation at the LSC/YSZ interface in positively polarized regions. Combined STEM-EELS and STEM-EDS analyses tracked valence state changes and the migration of iron, cobalt and strontium, while EIS measurements captured variations in surface reaction resistance simultaneously. These findings demonstrate a method for real-time observation of dynamic structural evolution, directly correlated with electrochemical performance testing. Acknowledgements This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 850850)” Reference [1] Ma, Zhongtao, et al. “Experimental Requirements for High‐Temperature Solid‐State Electrochemical TEM Experiments.” Small Methods (2024): 2301356. [2] Ma, Zhongtao, et al. “Electrochemical impedance spectroscopy integrated with environmental transmission electron microscopy. ” Small Methods (2023): 2201713. [3] Ma, Zhongtao, et al. “Operando Electron Microscopy and Impedance Analysis of Solid Oxide Electrolysis and Fuel Cells” ACS Energy Letters (2024): 9, 2007 [4] Ma, Zhongtao, et al. “Visualizing interface degradation of solid oxide cell”, in submission.

The durability of a solid oxide electrolysis cell (SOEC) is an ever-present concern for the technology to achieve long-term operation. This is a particular challenge in the widely pursued and researched composite fuel electrode of nickel (Ni) and stabilised zirconia (commonly Y ₂ O ₃ stabilised ZrO ₂ (YSZ)) [1,2]. The observed changes in microstructure are namely 1) the contact loss between particles in the composite (both between Ni-Ni or Ni-YSZ), and 2) the migration of Ni away from regions with highest overpotentials [2,3]. While clear correlations between operating conditions and degradation were established [2,3], corresponding changes in material properties measured under realistic SOEC operating conditions remained elusive. However, such measurements are crucial to establish the causality behind Ni migration and to advance the development of cells with improved durability. What further complicates establishing any conclusive degradation mechanism at the atomistic level is the irregular and porous nature of the Ni/YSZ composite. The presented study aims at advancing the understanding of Ni migration by measuring distinct material changes in SOEC mode. Realistic Ni/YSZ model electrode cells are prepared in a cleanroom by photolithography and magnetron sputtering of Ni. Heat treatment of the cells under SOEC fabrication conditions establishes an interface of Ni particles with YSZ alike the porous composite. Operando micro-Raman spectroscopy is performed due to 1) the high operation temperatures possible with the in-situ Raman setup (up to 1000 °C) and 2) the opportunities in analysing micrometre-sized volumes on the sample surface. Furthermore, the interactions of light with colour centre electrons (F-centres) allow us to probe the evolution of the defect chemistry of the oxygen ion conducting oxide (YSZ) under applied electrolysis potentials. Changes in oxygen vacancy concentration under cathodic potentials were previously reported by Hendriks et al. based on differential capacitance measurements on an Au/YSZ blocking electrode sample around 550 °C [4,5]. In this work, the changes in defect concentrations are measured at 750 °C for Ni/YSZ electrodes resembling the composites used on the device-scale. A sketch of the test setup is shown in Figure 1 a). The partially reduced YSZ regions around the triple phase boundary (TPB) are further correlated with Ni transport from the TPB onto the YSZ surface, into the bulk and the coarsening of Ni particles. The morphology of the particles and Ni transport are analysed by atomic force microscopy (AFM) (as shown in Figure 1 b)), secondary ion mass spectrometry (SIMS) and low energy ion scattering (LEIS). These approaches assure nanometre resolution in investigating the microstructure and highest surface sensitivity in the analysis of chemical composition. The characteristic decrease in Raman intensity around the cathodically polarised electrode (SOEC fuel electrode) is repeatedly observed. The lower intensity originates from the higher defect concentration (oxygen vacancies and F-centre electrons) at the cathode. On the same samples, the Ni morphology changes due to the reducing cathodic conditions. These changes are exemplified in Figure 1 b) and c) as the nanoparticles of Ni spreading onto the free YSZ surface (marked by arrows in Figure 1 c)) and the coarser Ni particles around the cathode TPB. The interaction between Ni and YSZ defects are further explored by density functional theory (DFT). A favourable interaction of Ni with reduced YSZ surfaces is established through different DFT calculations on ZrO ₂ and YSZ surfaces as well as the oxide bulk. It is found that despite the strong interaction with an oxygen vacancy site, Ni can easily be transported between vacancy sites on the ZrO ₂ surface. The DFT calculations, presenting an atomistic picture of the interactions between YSZ defect chemistry and Ni, can help elucidate the experimental observations. A multi-physics model of the defect chemistry inside a SOEC is used to connect the findings to the device-scale. In the presented work, the operando spectroscopic approach and findings are introduced. The Ni transport due to the defect chemical driving force are shown through results of AFM, SIMS and LEIS. Characteristic properties in Ni/YSZ interactions are explained on a DFT basis. Ultimately, a holistic mechanism is proposed connecting these observations to the common observations of Ni migration in SOECs. This is showcased by a multi-physics degradation model based on the defect chemical changes of YSZ under SOEC polarisation. Overall, a fundamental atomistic mechanism of the underlying driving force behind Ni migration is outlined based on the presented findings from experiments and modelling. References [1] Hauch, Anne, et al. Science 370.6513 (2020): eaba6118. [2] Mogensen, Mogens B., et al. Fuel Cells 21.5 (2021): 415-429. [3] Bilalis, Vasileios, et al. Journal of Power Sources 606 (2024): 234490. [4] Hendriks, Mark G. H. M., et al. Solid State Ionics 154 (2002): 467-472. [5] Hendriks, Mark G. H. M., et al. Solid State Ionics 146.3-4 (2002): 211-217. Figure 1

Fast and accurate synthesis and testing of electrocatalysts is essential to accelerate development of next generation catalysts for sustainable energy technologies. In this paper, we introduce CatBot, a fully automated platform for reliable synthesis and testing of electrocatalysts capable of operating at temperatures of up to 100 °C from highly acidic to highly alkaline conditions. The platform leverages roll-to-roll transfer, integrating customizable stages for substrate cleaning, catalyst loading, and electrochemical testing, with a custom made liquid distribution system enabling multi-element electrocatalyst synthesis via electrodeposition. CatBot enables fabrication and testing of up to 100 electrocatalysts per day, significantly accelerating catalyst discovery and optimization. We demonstrate the platform's reproducibility, through synthesis and testing of various catalytic coatings for the hydrogen evolution reaction (HER) in alkaline conditions, achieving overpotential uncertainties in the range of 4–13 mV at −100 mA cm⁻². Additionally, we benchmark the platform by comparing anodic and cathodic redox peaks for nickel in alkaline solutions confirming consistency with previous studies. Thus, CatBot comprises an automated, fast, reproducible, accurate and scalable synthesis and testing system for the accelerated development of next generation electrocatalysts.