Zhenhua Zeng

The performance of water electrolysis is largely limited by the sluggish oxygen evolution reaction (OER). Ni- and Ir-based materials are the catalysts of choices for alkaline and acidic OER, respectively. To design and engineer catalysts with further improved activity, it is desirable to identify catalytic active sites, and establish structure-activity relationship. Despite decades of intensive research, the precise atomic-scale configuration of active sites under operating conditions—and the resulting structure–activity relationships—have remained unresolved. This long-standing uncertainty has fundamentally limited the rational design of next-generation OER catalysts.

Although it is known that the oxygen reaction reaction (ORR) happens on catalyst surfaces, only specific surface sites have high catalytic activity. Thus, identifying active sites and maximizing their presence lies at the heart of ORR research. For ORR on Pt group metal surfaces, the classic model categorizes active sites based on surface motifs, such as terraces and steps. However, this simplistic approach often leads to orders of magnitude errors in catalyst activity predictions and qualitative uncertainties of active sites, thus limiting opportunities for catalyst design. Using stepped Pt(111) surfaces as examples, I will illustrate that the root cause of larger errors and uncertainties is such a simplified categorization overlooks atomic site-specific reactivity driven by surface stress release. Specifically, I will show how surface stress release at steps introduces inhomogeneous strain fields, resulting in distinct electronic structures and reactivity for terrace atoms with identical local coordination. This phenomenon leads to a cluster of active sites flanking both sides of the step edge. I will demonstrate strategies to enhance ORR activity by leveraging this effect, such as varying terrace widths, adjusting the thickness of 2D nanosheets or controlling external stress. If time permits, I will briefly introduce our recent work regarding how to tune the surface chemistry of inverse catalysts oxide/metal without site-blocking effects.

The performance of water electrolysis is largely limited by the sluggish oxygen evolution reaction (OER). To design and engineer electrocatalysts with higher and higher activity, the first step is to identify catalytic active phases and reaction mechanisms. In this presentation, I will delve into the active phases and reaction mechanics of OER on Ni-based oxyhydroxides in alkaline media. I will show that under extremely oxidative conditions, the active phases are typically not as-prepared precursor phases, as an oxidative phase transition may occur. I identified the atomic-scale structure and electronic structures of active phases through ab initio molecular dynamics simulation and XANES simulation at the O k edge. The predicted structure and electronic structures were confirmed by operando wide-angle X-ray scatting measurement and XNAES measurement at the O k edge. We further found that, under OER conditions, surfaces are not pristine but are in equilibrium with the electrolyte and electrode potential, i.e., they are covered by reaction intermediates. Thus, OER goes through Mars van Krevelen mechanics, rather than traditionally assumed Langmuir-hinshelwood mechanics. Also, active sites can be described by a synergistic dual-metal reaction center. This center is characterized by switching active sites during OER, and through simple surface Fe doping and the formation of dual Fe-Fe centers, they increase OER activity by over two orders of magnitude. This fundamental understanding not only elucidates the high OER activity of Ni (oxy)hydroxides with Fe-doping but also provides principles for developing OER catalysts with further improved performance. If time permits, I will also introduce our recent work in acidic water electrolysis.

Alkaline anion exchange membrane water electrolysis can utilize Ir-free catalysts and fluorine-free membranes and thus solve significant sustainability challenges posed by acidic IrO₂-catalyzed proton exchange membrane water electrolysis. In all alkaline water electrolysis technologies, Ni (oxy)hydroxides serve as the hosts of non-precious electrocatalysts for the anodic oxygen evolution reaction (OER). Despite over five decades of reseach, however, the atomic-scale strutures of the active phases remain elusive, hindering the identification of active sites and established rigorous structure-activity reaction. I will first provide direct atomic-scale insights into their crystal structures and structural transformations, shedding light on unique active sites: synergistic dual-metal reaction centers. There center are characterized by switching active sites during OER, and through simple surface Fe doping and the formation of dual Fe-Fe center, they increases OER activity by over two orders of magnitude. This fundamental understanding not only elucidates the high OER activity of Ni (oxy)hydroxides with Fe-doping but also provides principles for developing OER catalysts with further improved performance.^1-3 References Dionigi, F.; Zeng, Z.; Sinev, I., et al. Nature Communications 2020, 11, (1), 2522. Dionigi, F.; Zhu, J.; Zeng, Z.; Merzdorf, T.; Sarodnik, H.; Gliech, M.; Pan, L.; Li, W.-X.; Greeley, J.; Strasser, P. Angew. Chem., Int. Ed. 2021, 60, (26), 14446-14457. M. Klingenhof, H. T., S. Koch, J. Zhu, Z. Zeng, L. Metzler, A. Klinger, M. Elshamy, F. Lehmann, P. W. Buchheister, A. Weisser, G. Schmid, S. Vierrath, F. Dionigi and P. Strasser. Nature Catalysis 2024.

Advances in the theoretical understanding of electrochemical systems have, over the past decade, led to growing use of periodic Density Functional Theory (DFT) studies to treat a surprisingly large ensemble of electrocatalytic reactions, ranging from carbon dioxide electroreduction to oxygen evolution. Many such studies have employed simplified models of the electrochemical environment to determine reactivity trends across a broad space of catalytic materials, while other efforts have focused on developing detailed descriptions of electrochemical phenomena, such as the structure of electrochemical double layers, on model catalyst structures. An emerging challenge is to combine these approaches to ultimately enable theoretical design of electrocatalysts for reactions of significantly expanded chemical and materials complexity. In this talk, we begin by discussing how we develop and apply DFT-based methods to study the electrooxidation of ethanol and related oxygenated species on Pt surfaces. We show how explicit double layer models and ab initio molecular dynamics provide exciting insights beyond the traditional computational hydrogen electrode treatments, leading to improved descriptions of elementary reaction steps involving adsorption/desorption of reaction intermediates and proton-coupled electron transfer processes. By combining the aggregate results with detailed microkinetic models and explicit descriptions of adsorbate-adsorbate interactions, we further demonstrate how these effects can influence both the predicted overpotentials and the selectivities to acetic acid, acetaldehyde, and carbon dioxide. We conclude with some perspectives on how these insights may be used to enhance the search for improved oxygenate electrooxidation catalysts.

3d transitional metal oxyhydroxides, such as Ni and Co oxyhydroxides with Fe doping, are among the most active and most studied OER catalysts in alkaline electrolytes. They are also the catalysts of choice for industrial alkaline water electrolysis. Despite decades of research, the atomic-scale structures of these materials, particularly key Ni (oxy)hydroxide hosts, remain elusive. This presentation aims to fill this critical knowledge gap by combining rigorous Density Functional Theory calculations with in situ Wide Angle X-ray Scattering techniques under quasi-steady state electrochemical cycling conditions. This unique combination has enabled us to unveil the in-situ structures and (ir)reversible transformations of pure undoped Ni (oxy)hydroxides. Such an advance consequently allows us to transcend the limitations of previous modeling based on approximated structures. Specifically, based on the newly identified actual in-situ structures, we discover the existence of a previously unknown synergy at the reaction centers, coupled with the polaron effects, in promoting the oxygen evolution reaction. Such a finding sheds light on the catalytic active sites and oxygen evolution reaction pathways, particularly the exceptional activity of multiple Fe reaction centers on Ni oxyhydroxides. In addition, we found that such a phenomenon is general and also exists on NiM and CoM layered double hydroxides,^{1, 2} A2B2O7 pyrochlores,³ layered Co oxyhydroxides intercalated with phenanthroline through non-covalent ligand-oxide interaction.⁴ References: Dionigi, F.; Zeng, Z.; Sinev, I., et al. Nature Communications 2020, 11, (1), 2522. Dionigi, F.; Zhu, J.; Zeng, Z.; Merzdorf, T.; Sarodnik, H.; Gliech, M.; Pan, L.; Li, W.-X.; Greeley, J.; Strasser, P. Angew. Chem., Int. Ed. 2021, 60, (26), 14446-14457. Yan, J.; Zhu, J.; Chen, D.; Liu, S.; Zhang, X.; Yu, S.; Zeng, Z.; Jiang, L.; Du, F. Journal of Materials Chemistry A 2022, 10, (17), 9419-9426. Wu, Q.; Liang, J.; Xiao, M.; Long, C.; Li, L.; Zeng, Z.; Mavrič, A.; Zheng, X.; Zhu, J.; Liang, H.-W.; Liu, H.; Valant, M.; Wang, W.; Lv, Z.; Li, J.; Cui, C. Nature Communications 2023, 14, (1), 997.

For ORR on Pt, the classic model categorizes active sites based on surface motifs, such as terraces and steps. However, this simplistic approach often leads to orders of magnitude errors in catalyst activity predictions and qualitative uncertainties of active sites, thus limiting opportunities for catalyst design. Using stepped Pt(111) surfaces and ORR as examples, we will illustrate that the root cause of larger errors and uncertainties is such a simplified categorization overlooks atomic site-specific reactivity driven by surface stress release. Specifically, we will show how surface stress release at steps introduces inhomogeneous strain fields, resulting in distinct electronic structures and reactivity for terrace atoms with identical local coordination. This phenomenon leads to a cluster of active sites flanking both sides of the step edge. We will demonstrate strategies to enhance ORR activity in low temperature (100°C) PEM fuel cells, we found that, in contrast to ORR under hydrous conditions, (111) terrace sites are not active for ORR under anhydrous conditions, because of weakened binding of ORR intermediates induced by O* accumulation on the surface. On the other hand, step edges, which are generally not active for ORR under hydrous conditions, are predicted to be the active sites for ORR under anhydrous conditions. Among them, (110) type step edge with a unique configuration of accumulated O stabilizes O2 adsorption and facilitates O2 dissociation, which lead an overpotential [4] [1] G. Liu, A. J. Shih, H. Deng, K. Ojha, X. Chen, M. Luo, I. T. McCrum, M. T. M. Koper, J. Greeley, Z. Zeng, Nature 2024, 626, 1005-1010. [2] L. Wang, Z. Zeng, W. Gao, T. Maxson, D. Raciti, M. Giroux, X. Pan, C. Wang, J. Greeley, Science 2019, 363, 870-874. [3] K. J. Sawant, Z. Zeng, J. P. Greeley, Angew. Chem., Int. Ed. 2024, 63, e202312747. [4] G. Liu, H. Deng, J. Greeley, Z. Zeng, Chinese Journal of Catalysis 2022, 43, 3126-3133.