Cobalt oxide–supported iridium oxide nanoparticles with strong metal oxide–support interaction for efficient acidic oxygen evolution reaction

Abstract

Understanding and regulating the deprotonation process in an acidic oxygen evolution reaction (OER) is highly desirable for a proton exchange membrane water electrolyzer (PEMWE). Herein, ultrasmall IrO₂ nanoparticles were firmly anchored on an acid-resistant Co₃O₄ support (IrO₂/Co₃O₄) through galvanic replacement, with strong metal oxide–support interaction (SMOSI) induced and responsible for the accelerated deprotonation process during OER. For IrO₂/Co₃O₄, a low overpotential of 256 mV at 10 mA cm⁻² could be achieved for an acidic OER, with sustained operation exceeding 1000 h. More importantly, a PEMWE assembled with IrO₂/Co₃O₄ as the anode could survive 120 h and 40 h of operation at industrial-level current densities of 0.5 and 1 A cm⁻², with cell voltages of 1.64 and 1.77 V, respectively. Experimental results and theoretical calculations together demonstrate that the SMOSI induced by the lattice-mismatched interfaces in IrO₂/Co₃O₄ could increase the p-band center of O_bri (bridging oxygen) sites in the Ir–O_bri bonds. Such an enhanced p-band center would strengthen the proton acceptance of O_bri sites, facilitating the deprotonation process, and thus improving OER activity and stability. This work presents an alternative approach for the regulation of the deprotonation process via SMOSI and the design of an inexpensive and efficient electrocatalyst towards an industrial-level PEMWE.

---

New concepts

Accelerating the deprotonation process during an acidic oxygen evolution reaction (OER) is an effective strategy to break the scaling relationship in the conventional adsorbate evolution mechanism (AEM). In this work, the concept of strong metal oxide–support interaction (SMOSI) has been demonstrated to regulate the deprotonation process in acidic OER, to improve OER activity and reduce the mass loading of noble Ir. As evidenced by experimental results and theoretical calculations, SMOSI could be induced by anchoring ultrasmall IrO₂ nanoparticles on a Co₃O₄ support (IrO₂/Co₃O₄) through galvanic replacement. Such SMOSI strengthens the proton acceptance of bridging oxygen (O_bri) sites in the Ir–O_bri bonds, thereby accelerating the deprotonation process in OER. Consequently, a deprotonation-assisted adsorbate evolution mechanism (DAEM) could be reasonably proposed to elucidate the excellent performance for acidic OER over IrO₂/Co₃O₄, by breaking the scaling relationship in conventional AEM. This study demonstrates a promising approach to regulating the deprotonation process on O_bri sites via SMOSI for efficient and stable acidic OER, and also inspires the design of applicable and low-noble-metal electrocatalysts for industrial-level PEMWE.

1. Introduction

Hydrogen, with its high energy density, is widely recognized as a promising candidate for next-generation clean energy, when produced via water electrolysis powered by green electricity generated from renewable energy.^1–4 Among various water electrolysis technologies, the proton exchange membrane water electrolyzer (PEMWE) has attracted worldwide interest due to its high current density (>1 A cm⁻²), high hydrogen purity (>99.99%), and fast response (<5 s) for intermittent renewable electricity.^5–7 However, the industrial application of PEMWE is substantially impeded by the absence of affordable and highly efficient electrocatalysts at the anode for an acidic oxygen evolution reaction (OER).^8–10 In general, Ir and its derived oxides have been exclusively considered to be the only viable electrocatalysts to meet the requirements of high activity for OER and excellent stability under harsh acidic conditions in a PEMWE, while the scarcity of noble Ir has drastically limited its large-scale application.^11–13 Thus, significant efforts have focused on lowering the mass loading of Ir at the anode while enhancing electrocatalytic OER activity for practical PEMWE applications.^14,15

Dispersing Ir-based active species on high-surface-area supports has been considered an effective strategy towards improved OER performance with optimized utilization of noble Ir for electrocatalysis. These reported support materials, e.g., MnO₂, Nb₂O_5−x, TaO_x, TiO₂, and Co₃O₄,^16–21 are expected to exhibit high resistance to acid corrosion and oxidative decomposition to ensure stable dispersion of Ir-based active species. Among them, Co₃O₄ has been recognized as an effective support material for loading Ir-based active species,^22–26 enhancing the electrocatalytic activity and stability of Ir active sites in acidic OER. For example, by incorporating Ir single atoms on the Co₃O₄ support, the obtained Ir–Co₃O₄ electrocatalyst reached a current density of 10 mA cm⁻² for acidic OER at a small overpotential of 236 mV, with an Ir loading of only 1.05 at%.²² However, due to weak interaction with the Co₃O₄ support, the Ir single atoms in this electrocatalyst tend to dissolve during the OER process, resulting in limited stability at industrial-level current densities that requires further enhancement. Besides, the OER pathway determined by the interactions between the Ir-based active species and the Co₃O₄ support remains unclear and requires further exploration. The activity and stability for OER over Ir-based active species are believed to depend on the OER pathway. It was revealed that OER taking place at rutile IrO₂ should follow the conventional adsorbate evolution mechanism (AEM), which is limited by the scaling relationship between the adsorption energies of OH* and OOH* intermediates, with a theoretical minimum overpotential of ca. 370 ± 100 mV.^27–29 By manipulating the OER intermediates, this limitation in AEM can be overcome via the lattice-oxygen-mediated mechanism (LOM), in which activated lattice oxygen participates in the O–O coupling process to improve OER.^30–32 However, the oxygen defects generated via LOM would cause the rapid degradation and deactivation of electrocatalysts under acidic OER conditions.³³ Recently, an accelerated deprotonation process during OER by incorporating dopants into IrO₂ or RuO₂ has been proposed and confirmed to break the scaling relationship without sacrificing OER stability.^34–38 For instance, by introducing Sb dopants into the RuO₂ lattice to construct highly asymmetric Ru–O–Sb units, the deprotonation process of oxygen-containing intermediates taking place at bridging oxygen (O_bri) sites was significantly promoted for the obtained Ru_0.8Sb_0.2O₂, which required a low overpotential of 160 mV at 10 mA cm⁻² and maintained a good stability of 1100 h for acidic OER.³⁷ With such a deprotonation process regarded as an additional chemical step in AEM, the theoretical overpotential was significantly decreased to 0.35 V, compared to that of conventional AEM (0.69 V).³⁷ However, the deprotonation process regulated by incorporating dopants relies heavily on noble metals as the main components, which inevitably affords a remarkable increase in the cost of electrocatalysts. Anticipating a significant improvement in both the activity and stability of electrocatalysts for acidic OER, it would be encouraging but challenging to accelerate the deprotonation process in the OER pathway, with Ir content reduced by introducing a Co₃O₄ support to promise a high-performance and cost-effective OER. Considering the strong metal oxide–support interaction (SMOSI) between the Ir-based active species and the Co₃O₄ support,^39,40 which may determine the OER pathway and thus acidic OER activity, it is anticipated that elucidating the SMOSI-regulated deprotonation mechanism can guide the rational design of Ir-based electrocatalysts for practical industrial-scale PEMWE applications.

Motivated by the above insights, herein, ultrasmall IrO₂ nanoparticles were firmly anchored on the acid-resistant Co₃O₄ support (IrO₂/Co₃O₄) through galvanic replacement, with the induced SMOSI responsible for the accelerated deprotonation process for acidic OER. In comparison to commercial IrO₂, the obtained IrO₂/Co₃O₄ exhibits an excellent acidic OER performance, with a significantly lowered overpotential of 256 mV at a current density of 10 mA cm⁻² for sustained operation exceeding 1000 h. Impressively, a PEMWE assembled with IrO₂/Co₃O₄ as the anode and Pt/C as the cathode could be operated stably for at least 120 and 40 h at industrial-level current densities of 0.5 and 1 A cm⁻², with cell voltages of 1.64 and 1.77 V, respectively. As evidenced by experimental results and theoretical calculations, the SMOSI generated at the lattice-mismatched interfaces of the anchored IrO₂ nanoparticles and the Co₃O₄ support could increase the p-band centers of O_bri sites towards the strengthened acceptance of protons, and thus accelerate the deprotonation process during acidic OER. These findings should be able to guide the steering of proton transfer behavior and thus the deprotonation process for efficient acidic OER, while also proposing an alternative strategy for the design of inexpensive and efficient electrocatalysts via SMOSI for an industrial-level PEMWE.

2. Results and discussion

Herein, to anchor IrO₂ nanoparticles on Co₃O₄ as support (IrO₂/Co₃O₄) with strong metal oxide–support interaction (SMOSI) induced at interfaces, metallic Co was first electrodeposited on titanium felt (TF), followed by the galvanic replacement of Ir species (Fig. 1a).⁴¹ The spontaneous galvanic replacement occurring between the metallic Co and the Ir³⁺ ions in precursor solution (2Ir³⁺ + 3Co → 2Ir + 3Co²⁺) would then result in the formation of Ir nanoparticles well anchored on the surface of metallic Co (Ir/Co), due to the favorable thermodynamic process. Through the subsequent calcination and acid leaching processes, the well-optimized IrO₂/Co₃O₄ electrode with IrO₂ nanoparticles supported on Co₃O₄ could then be obtained from Ir/Co for an efficient oxygen evolution reaction (OER) (Fig. S1), with a low mass loading of Ir determined to be ca. 0.36 mg_Ir cm⁻² by inductively coupled plasma-mass spectrometry (ICP-MS, Table S1). As shown in Fig. 1b, four Raman bands are clearly positioned at ca. 480, 522, 620, and 690 cm⁻¹, corresponding to the E_g bending, F²_2g bending, F_2g stretching, and A_1g stretching vibration in the Co₃O₄ phases,⁴² respectively. Raman bands related to IrO₂ could hardly be observed, demonstrating that the IrO₂ phases exist in a small proportion or with a particle size below the Raman detection limit, with Co₃O₄ phases dominating in IrO₂/Co₃O₄. It should be carefully noted that the Raman signal peaking at ca. 690 cm⁻¹ exhibits a slight red-shift for IrO₂/Co₃O₄ compared to Co₃O₄, implying lattice strain to the Co₃O₄ phases anchored with IrO₂ nanoparticles, as also confirmed by powder X-ray diffraction (XRD) patterns (Fig. S2). Scanning electron microscopy (SEM) images reveal a particulate structure with a coarse surface for both Co₃O₄ (Fig. S3) and IrO₂/Co₃O₄ (Fig. 1c and d) deposited on TF, which should be beneficial for the exposure of abundant active sites and thus promote mass transfer. As identified by high-resolution transmission electron microscopy (HRTEM) and aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (AC HAADF-STEM) images, ultrasmall IrO₂ nanoparticles of 2–4 nm in average size (Fig. 1e) are firmly anchored on the surface of the Co₃O₄ support, given lattice fringes with d-spacings of 0.165 and 0.226 nm attributed to the (422) plane of Co₃O₄ and the (200) plane of IrO₂ (Fig. 1f), respectively. Further analysis on the top-view AC HAADF-STEM image (Fig. 1g) reveals a significant lattice mismatch at the interface between the (111) plane of Co₃O₄ and the (101) plane of IrO₂, with the lattice spacings at ordered atomic arrangements determined to be 0.465 and 0.260 nm, respectively.

Fig. 1 (a) Schematic process for the synthesis of an IrO₂/Co₃O₄ electrode with IrO₂ nanoparticles anchored on a Co₃O₄ support. (b) Raman spectra of commercial IrO₂, Co₃O₄ and IrO₂/Co₃O₄. (c) and (d) SEM images, (e) HRTEM images and (f) and (g) AC HAADF-STEM images of IrO₂/Co₃O₄.

Due to the much smaller lattice spacing of the IrO₂ (101) plane than the Co₃O₄ (111) plane, the Ir–O bonds of IrO₂ nanoparticles should be elongated by tensile strain, while the Co₃O₄ support undergoes compressive strain at the interface. Such interfacial lattice mismatch would lead to structural distortion of the IrO₂ nanoparticles and the Co₃O₄ support, referring to SMOSI, and thus inducing compressive strain in the Co₃O₄ support.^43–45 Given the homogeneous distribution of Ir, Co and O in the IrO₂/Co₃O₄ electrode (Fig. S4), abundant lattice-mismatched IrO₂/Co₃O₄ heterostructures should be created, with IrO₂ nanoparticles firmly anchored at the surface of the Co₃O₄ support through galvanic replacement.

The electronic structures and chemical compositions of the IrO₂/Co₃O₄ electrode were then investigated by X-ray photoelectron spectroscopy (XPS). It is distinctly observed that with IrO₂ nanoparticles anchored on Co₃O₄, the Co³⁺ 2p_3/2 signal located at ca. 779.6 eV is shifted to higher binding energy by 0.5 eV,⁴⁶ with the Co³⁺/Co²⁺ ratio much increased from 0.55 for Co₃O₄ to 0.94 for IrO₂/Co₃O₄ (Fig. 2a). Such an XPS peak shift and increased Co³⁺/Co²⁺ ratio suggest that the Co atoms should serve as electron donors, when IrO₂ nanoparticles interface with the Co₃O₄ support in the IrO₂/Co₃O₄ electrode. Moreover, the Ir⁴⁺ 4f_7/2 signal observed at ca. 61.3 eV is shifted to a lower binding energy by 0.2 eV for IrO₂/Co₃O₄ compared to IrO₂ (Fig. 2b).^47,48 This negative shift in binding energy indicates that the Ir atoms in the IrO₂ phases accept electrons from the Co₃O₄ support, again suggesting charge redistribution at the interfaces in IrO₂/Co₃O₄ caused by interfacial lattice mismatch (Fig. 1f and g), agreeing well with the Co 2p XPS analysis (Fig. 2a). The atomic ratio of Co:Ir was determined to be 2.8 : 1 for IrO₂/Co₃O₄ by XPS analysis (Table S2), which, together with the absence of an IrO₂ signal in Raman spectra (Fig. 1b), indicates the high dispersion of IrO₂ nanoparticles at the surface of the Co₃O₄ substrate. Density functional theory (DFT) calculations were performed to confirm the above-discovered charge transfer behavior in the lattice-mismatched structure model of IrO₂/Co₃O₄ (Fig. S5).

Fig. 2 (a) Co 2p XPS spectra of Co₃O₄ and IrO₂/Co₃O₄, (b) Ir 4f XPS spectra of IrO₂ and IrO₂/Co₃O₄. (c) Differential charge density patterns and planar-average charge density plots of IrO₂/Co₃O₄. The yellow and cyan contours represent electron accumulation and depletion regions, respectively. (d) Normalized XANES and (e) FT-EXAFS spectra of Ir foil, IrO₂ and IrO₂/Co₃O₄ collected at the Ir L₃-edge. (f) Normalized XANES and (g) FT-EXAFS spectra of Co foil, CoO, standard Co₃O₄ and IrO₂/Co₃O₄ collected at the Co K-edge. (h) Wavelet-transformed EXAFS spectra of Ir foil, IrO₂ and IrO₂/Co₃O₄ collected at the Ir L₃-edge.

It is observed that charge accumulation occurs in the IrO₂, while the charge depletion occurs in the Co₃O₄ phase (Fig. 2c), providing evidence of charge redistribution at the IrO₂/Co₃O₄ interfaces. This evolution in electronic structure in the obtained IrO₂/Co₃O₄ electrode confirms SMOSI as induced by the lattice-mismatched interfaces between the anchored IrO₂ nanoparticles and the Co₃O₄ support.

To further distinguish the atomic coordination structure of IrO₂ nanoparticles anchored on the Co₃O₄ support, X-ray absorption spectroscopy (XAS) measurements were carried out for the IrO₂/Co₃O₄ electrode. The normalized X-ray absorption near-edge structure (XANES) spectra collected at the Ir L₃-edge show that the white-line peak at 11216–11219 eV (Fig. 2d), related to the electronic transition from occupied Ir 2p to empty Ir 5d orbitals,⁴⁹ is slightly lower for IrO₂/Co₃O₄ than for IrO₂, which indicates the higher electron density at the Ir 5d orbitals in IrO₂/Co₃O₄, attributed to electrons transferred from Co₃O₄ to IrO₂ (Fig. 2b). By determining the white-line position, the average oxidation state of Ir could then be calculated as +3.60 in IrO₂/Co₃O₄ with Ir foil and IrO₂ as references (Fig. S6). The coordination structures were further investigated from the Fourier-transform (FT) extended X-ray absorption fine structure (EXAFS) spectra (Fig. 2e). The peak located at ca. 1.56 Å, assigned to the Ir–O scattering path, is slightly shifted to higher R space for IrO₂/Co₃O₄ compared to IrO₂. This slight shift suggests elongated Ir–O bonds in IrO₂/Co₃O₄, as supported by the well-fitted EXAFS spectra, with the Ir–O bond lengths determined to be 1.97 Å in IrO₂ and 2.00 Å in IrO₂/Co₃O₄ (Fig. S7 and Table S3). These elongated Ir–O bonds in IrO₂/Co₃O₄ should be ascribed to the weakened Ir–O bond strength caused by the increased electron density at Ir atoms via SMOSI (Fig. 2d). Moreover, except for the peak located at ca. 2.48 Å for the Ir–Ir scattering path, another weak shoulder peak could be noted at ca. 2.10 Å in the Ir L₃-edge EXAFS spectra of IrO₂/Co₃O₄ (Fig. 2e). Such a weak shoulder peak belongs to the Ir–Co scattering path,^50,51 resulting from the abundant interfaces created between the anchored IrO₂ nanoparticles and the Co₃O₄ support. As for normalized XANES spectra collected at the Co K-edge, the oxidation state of Co could be evaluated from the absorption edge, which is determined by the position of half height in the normalized XANES spectra.⁵² Interestingly, the absorption edge displays a slight shift to lower energy for IrO₂/Co₃O₄ compared to the standard Co₃O₄ (Fig. 2f), implying the relatively lower Co oxidation state in IrO₂/Co₃O₄. The EXAFS spectra collected at the Co K-edge were then studied to analyze the coordination structure of Co atoms in IrO₂/Co₃O₄ (Fig. 2g), with peaks clearly observed at ca. 1.5, 2.5 and 3.0 Å, corresponding to the Co–O, Co_oct–Co_oct (oct: octahedral sites) and Co_oct–Co_tet (tet: tetrahedral sites) scattering paths, respectively. By further fitting the EXAFS spectra (Fig. S8), one could note that the coordination number of Co–O is lower for IrO₂/Co₃O₄ than for standard Co₃O₄ (Table S3), which should be related to the formation of oxygen vacancies and thus, the decreased oxidation state of Co in IrO₂/Co₃O₄ (Fig. 2f). In addition, the fitted bond lengths of Co–O, Co_oct–Co_oct and Co_oct–Co_tet are smaller in IrO₂/Co₃O₄ than in Co₃O₄ (Table S3), which are again evidence for compressive strain in the Co₃O₄ support, well matching the analysis of the Raman spectra (Fig. 1b) and XRD patterns (Fig. S2b). One weak shoulder peak assigned to the Co–Ir scattering path could also be observed at ca. 2.10 Å in the Co K-edge EXAFS spectra (Fig. 2g), as previously discovered in the Ir L₃-edge EXAFS spectra (Fig. 2e). By fitting the EXAFS spectra, the Ir–Co bond lengths are determined to be 2.66 Å in IrO₂/Co₃O₄ (Table S3), smaller than the Co_oct–Co_oct bond lengths of 2.87 Å in standard Co₃O₄ and the Ir–Ir bond lengths of 3.10 Å in IrO₂ (Table S3), as also observed for the Ir–Co bonds in the structural model of IrO₂/Co₃O₄ (Fig. S5). These reduced Ir–Co bond lengths should be attributed to structural distortion at the interface between the IrO₂ nanoparticles and the Co₃O₄ support, as induced by the interfacial lattice mismatch. To distinguish the Ir–Co scattering path, wavelet-transform analysis was further performed in the Ir L₃-edge EXAFS spectra. With Ir foil and IrO₂ as references (Fig. 2h, upper and middle), the intensity maxima positioned at k ≈ 6–8.5 Å⁻¹ and 11.5–14 Å⁻¹ should be attributed to the Ir–O and Ir–Ir scattering paths, respectively. Notably, an additional intensity maximum could be detected at k ≈ 10–11 Å⁻¹ for IrO₂/Co₃O₄ (Fig. 2h, bottom), which should be ascribed to the Ir atom bonded with the Co atom, confirming the presence of the Ir–Co scattering path.²⁴ The above spectral analyses are evidence that the IrO₂ nanoparticles anchored on the Co₃O₄ support would generate abundant Ir–Co bonds at the lattice-mismatched interfaces of IrO₂/Co₃O₄, with the induced SMOSI elongating the Ir–O bonds, which is believed to regulate OER intermediate behavior for improved OER performance, as revealed in the following discussions.

The electrocatalytic activity for acidic OER over the obtained IrO₂/Co₃O₄ electrode was evaluated in 0.5 M H₂SO₄ electrolyte in a three-electrode system, with a graphite rod and a calibrated Hg/Hg₂SO₄ electrode used as the counter and reference electrodes, respectively. In comparison to Co₃O₄, the IrO₂/Co₃O₄ electrode exhibits a great increase in OER performance, with a small overpotential of only 256 mV required to achieve a current density of 10 mA cm⁻², much lower than that of commercial IrO₂ (340 mV) as the benchmark OER electrocatalyst (Fig. 3a). Depending on the increasing applied potentials, the IrO₂/Co₃O₄ electrode could reach current densities of 100 and 500 mA cm⁻² for OER at overpotentials of 308 and 373 mV, respectively. By further noting the great decrease in OER activity observed for the IrO₂ nanoparticles with Co₃O₄ support etched by acid (Fig. S9), SMOSI between the anchored IrO₂ nanoparticles and the Co₃O₄ support should contribute significantly to the enhancement in the intrinsic OER activity of IrO₂ for the IrO₂/Co₃O₄ electrode. To confirm the above deduction, the mass activity is calculated to have increased by 133 times, from 1.5 A g_Ir⁻¹ for commercial IrO₂ to 200 A g_Ir⁻¹ for the IrO₂/Co₃O₄ electrode at 1.53 V vs. RHE (Fig. 3b), which outperforms most reported Ir-based electrocatalysts for acidic OER (Table S4). Benefiting from SMOSI, the obtained IrO₂/Co₃O₄ electrode displays significantly improved OER kinetics, evidenced by a much decreased Tafel slope (54 mV dec⁻¹) compared to commercial IrO₂ (73 mV dec⁻¹) (Fig. 3c). Such kinetic improvement could be solidly supported by analysis of the electrochemical impedance spectroscopy (EIS), with charge transfer resistance (R_ct) decreasing from 3.79 Ω cm² for commercial IrO₂ to 0.42 Ω cm² for IrO₂/Co₃O₄ (Fig. 3d and Fig. S10). This reduction is attributed to the significantly increased electrochemically active surface areas (ECSAs) from 284 cm² for commercial IrO₂ to 1946 cm² for IrO₂/Co₃O₄, as determined from current–scan rate plots (Fig. S11). Notably, chronopotentiometry measurements reveal negligible change in the overpotentials for OER at current densities of 10 and 200 mA cm⁻² over 1000 (Fig. 3e) and 250 h operations (Fig. S12), respectively, indicating the excellent stability of the IrO₂/Co₃O₄ electrode for OER under acidic conditions. Scanning electron microscopy with energy-dispersive X-ray spectroscopy (SEM-EDS) and XPS analysis were then performed to assess the elemental dissolution of the IrO₂/Co₃O₄ electrode during OER. After OER for 24 h at a current density of 200 mA cm⁻², the atomic ratio of Co : Ir in the IrO₂/Co₃O₄ electrode was determined to have decreased significantly from 1.9 : 1 to 0.9 : 1 by EDS (Fig. S13) and from 2.8 : 1 to 1.2 : 1 by XPS (Table S2). Such a decrease in the Co : Ir ratio suggests that the Co₃O₄ support would be partially dissolved into the 0.5 M H₂SO₄ electrolyte under OER conditions, with IrO₂ phases dominating at the surface of the IrO₂/Co₃O₄ electrode during OER. The elemental dissolution of Co and Ir in the electrolyte during OER was further identified by ICP-MS analysis. It is revealed that Co species would be gradually dissolved from the Co₃O₄ support, especially during the initial 400 h (Fig. S14), which should be responsible for the slightly increased OER overpotentials (ΔE = 52 mV, Fig. 3e) during the chronopotentiometric measurements. Notably, the very slow dissolution of Ir species (Fig. S14) and the well-maintained XRD patterns (Fig. S15) suggest excellent OER stability. More encouragingly, the S-number (a metric for electrocatalyst stability)⁵³ reaches 5.6 × 10⁵ at 10 mA cm⁻² for the IrO₂/Co₃O₄ electrode, standing among the highest level reported for Ir-based electrocatalysts (Fig. S16 and Table S5). All these electrochemical results are evidence for the excellent activity and stability of IrO₂/Co₃O₄ for long-term OER under acidic conditions, exceeding those of most reported electrocatalysts where Ir species are dispersed on Co-based oxides (Fig. 3f and Table S4), which should be attributed to SMOSI, regulating OER intermediate behavior and thus reaction pathways.

Fig. 3 (a) Linear sweep voltammetry (LSV) plots of commercial IrO₂, Co₃O₄ and IrO₂/Co₃O₄. (b) Mass activities of Ir for commercial IrO₂ and IrO₂/Co₃O₄. (c) Tafel and (d) EIS plots of commercial IrO₂, Co₃O₄ and IrO₂/Co₃O₄. (e) Chronopotentiometric plots of IrO₂/Co₃O₄ for stability testing. (f) Activity and stability of IrO₂/Co₃O₄ and the other reported electrocatalysts of Ir species dispersed on Co-based oxides.

As inferred from the above analytical results, the OER overpotential required by IrO₂/Co₃O₄ at a current density of 10 mA cm⁻² (256 mV) is significantly lower than the theoretical minimum overpotential (370 ± 100 mV) limited by the scaling relationship, indicating that the reaction pathway for OER over IrO₂/Co₃O₄ should be different from the conventional adsorbate evolution mechanism (AEM). With the lattice-oxygen-mediated mechanism (LOM) excluded by using tetramethylammonium cation (TMA⁺) as a chemical probe (Fig. S17),⁵⁴ the pH-dependent activity was evaluated and plotted to further deepen insights into the reaction pathway taking place at the obtained IrO₂/Co₃O₄ electrode. At a potential of 1.55 V vs. RHE, the very slight change in the logarithm of OER current densities dependent on pH values with a slope of −0.14 (Fig. 4a and Fig. S18a), suggests that commercial IrO₂ should drive OER by following the conventional AEM with a concerted proton–electron transfer (CPET) process.⁵⁵ In comparison, the IrO₂/Co₃O₄ electrode exhibits more significant dependence on pH values with a slope of −0.81 for OER activity (Fig. 4a and Fig. S18b), implying a nonconcerted proton–electron transfer (nCPET) process. It has been well documented that the nCPET process should be beneficial to electrocatalytic OER activity, accompanied by a deprotonation process with decoupled electron transfer.³⁷ To investigate the deprotonation process taking place in OER over the IrO₂/Co₃O₄ electrode, kinetic isotope effect (KIE) measurements were conducted in proton (0.5 M H₂SO₄ in H₂O) and deuterium (0.5 M H₂SO₄ in D₂O) electrolytes, through Tafel analysis obtained from LSV plots (Fig. 4b and Fig. S19).⁵⁶ The KIE value was determined from the ratio of rate constants collected in proton (k_H) and deuterium (k_D) electrolytes (Fig. S19). Given the much larger mass of the D atom than of the H atom, replacing H₂O with D₂O would impede the kinetics of pathways involving protons during OER in aqueous solution, with a KIE value larger than 1 (k_H > k_D). Therefore, the KIE value can serve as a direct indicator for proton transfer being involved in the rate-determining step (RDS) for OER.^35,37,42 At low current density (less than 15 mA cm⁻²), the KIE value (k_H/k_D) based on Tafel analysis is calculated to be 1.59 for IrO₂/Co₃O₄ (Fig. S19), much smaller than for IrO₂ (1.67). This significantly decreased KIE value is good evidence for the accelerated proton transfer process for IrO₂/Co₃O₄, with SMOSI induced at interfaces benefitting the deprotonation process. Moreover, for both IrO₂ and IrO₂/Co₃O₄, the KIE values in the OER potential regions surpass the upper limit of secondary KIE (∼1.5), indicating that proton transfer should be involved in the rate-determining step (RDS) for acidic OER.⁵⁷ Such proton transfer relies on the oxygen species, as well explained by the O 1s XPS spectra. One could observe four distinct signals at ca. 530.1, 531.5, 533.0, and 534.0–535.0 eV, for IrO₂/Co₃O₄ (Fig. 4c), which should be assigned to lattice oxygen (O_lat), protonated bridging oxygen (OH_bri), adsorbed hydroxyl (OH_ads) at unsaturated Ir sites, and the gas phase of water molecules (H₂O_(g)), respectively.³⁴ Notably, the ratio of OH_bri in O 1s XPS remains almost unchanged for Co₃O₄ and IrO₂ before and after OER (Fig. S20), while IrO₂/Co₃O₄ reveals an obvious increase in the OH_bri signal intensity after OER (Fig. 4c). Given that water dissociation happened mainly at the active sites, this enhanced OH_bri signal intensity should be attributed to the highly dispersed IrO₂ nanoparticles on the IrO₂/Co₃O₄ electrode during OER, as already evidenced by SEM-EDS (Fig. S13) and XPS analyses (Table S2).

Fig. 4 (a) pH-Dependent curves plotted logarithmically for OER current densities recorded for IrO₂ and IrO₂/Co₃O₄ at 1.55 V vs. RHE at different pH values. (b) LSV curves of IrO₂ and IrO₂/Co₃O₄ measured in proton (0.5 M H₂SO₄ in H₂O) and deuterium (0.5 M H₂SO₄ in D₂O) electrolytes. (c) O 1s XPS spectra of IrO₂/Co₃O₄ before and after measuring OER at 200 mA cm⁻² for 24 hours. In situ Raman spectra of interfacial water on the (d) IrO₂ and (e) IrO₂/Co₃O₄ electrodes collected at different applied potentials, from OCP to 1.53 V vs. RHE. (f) Potential-dependent normalized signal intensity of 4-HB H₂O, 2-HB H₂O and free H₂O collected from in situ Raman spectra, with normalization based on the Raman spectra collected at OCP. (g) Schematic illustration of AEM and DAEM for OER over IrO₂/Co₃O₄. (h) Calculated Gibbs free energy diagrams for OER over IrO₂/Co₃O₄ by following AEM and DAEM.

It could thus be reasonably deduced that the bridging oxygen (O_bri, Fig. S21) sites in the IrO₂ nanoparticles anchored on the Co₃O₄ support should serve as proton acceptors to generate OH_bri intermediates, thereby accelerating the deprotonation process for OER. All the above electrochemical and spectral analyses could together claim that the deprotonation process in acidic OER could be accelerated over the IrO₂/Co₃O₄ electrode with SMOSI benefitting the O_bri sites to promote proton transfer and thus improve OER electrocatalysis.

Interfacial water as the proton source in acidic OER was monitored by in situ Raman spectra to further unravel the underlying mechanism of the accelerated deprotonation process during OER over IrO₂/Co₃O₄. Both IrO₂ and IrO₂/Co₃O₄ show a broad band at 3000–3800 cm⁻¹, related to the O–H stretching mode of interfacial water, which could be further fitted into three signals at ca. 3320, 3435 and 3620 cm⁻¹ (Fig. 4d and e), corresponding to 4-coordinated hydrogen-bonded water (4-HB H₂O), 2-coordinated hydrogen-bonded water (2-HB H₂O) and free water (Free H₂O) (Fig. S22), respectively.^58,59 With normalization based on Raman spectra collected at open circuit potential (OCP), one should note that these signals assigned to interfacial water are gradually decreased, depending on the increased applied potentials (Fig. 4f), revealing the dissociation of interfacial water along with the supply of H and O atoms for OER. Furthermore, the onset of water dissociation, defined as the potential region in which the normalized signal intensity starts to decrease, is lower for IrO₂/Co₃O₄ (at 1.23–1.33 V vs. RHE) than for IrO₂ (at 1.33–1.43 V vs. RHE), suggesting that water dissociation could be facilitated over the IrO₂/Co₃O₄ electrode, as induced by the accelerated deprotonation process. This facilitated water dissociation could be verified by the much lowered kinetic energy barrier at the O_bri sites calculated for IrO₂/Co₃O₄ relative to IrO₂ (Fig. S23). To understand water dissociation at the level of electronic structure, the projected densities of states (PDOS) of O_bri sites in IrO₂ and IrO₂/Co₃O₄ were further investigated (Fig. S24). It is clear that IrO₂/Co₃O₄ exhibits an obvious upshift in the O_bri p-band centers (E_p) of −2.41 eV compared to IrO₂ (−2.66 eV), as supported by the elongated Ir–O bonds in IrO₂/Co₃O₄ (Fig. 2e). Such an upshift in E_p indicates that SMOSI could cause less filling of antibonding states of O_bri for stronger acceptance of protons in IrO₂/Co₃O₄, facilitating the formation of OH_bri intermediates during water dissociation. These experimental results and theoretical calculations disclose that water dissociation could be accelerated for efficient acidic OER over the IrO₂/Co₃O₄ electrode via SMOSI, steering the behavior of intermediates and the reaction pathway.

Inspired by the above analysis, a deprotonation-assisted adsorbate evolution mechanism (DAEM) could be proposed for acidic OER over the IrO₂/Co₃O₄ electrode to rationalize the excellent OER activity by breaking the scaling relationship in a conventional AEM. This proposed DAEM involves water dissociation at O_bri sites, along with the formation of OH_bri intermediates through an nCPET process (Fig. 4g), as supported by the pH-dependent OER activity (Fig. 4a). With additional elementary steps involved in DAEM through the nCPET process for the generation of OH* and OOH* intermediates, the scaling relationship in a conventional AEM could be broken, contributing to a reduced energy barrier for the RDS for greatly enhanced OER activity. To confirm the above insights, the Gibbs free energy changes were then calculated for IrO₂/Co₃O₄ with OER taking place via a conventional AEM and the proposed DAEM (Fig. 4h). It is observable that the RDS transforms from the generation of OOH* intermediates (O* → OOH*) for conventional AEM to the generation of O* intermediates (OH* → O*) for the proposed DAEM on IrO₂/Co₃O₄, with the calculated OER overpotential decreasing from 0.76 to 0.21 V. For the RDS taking place on IrO₂/Co₃O₄ by following DAEM, the OH* intermediates would convert into O* intermediates via the cleavage of O–H bonds, agreeing well with proton transfer participating in the RDS, as revealed by the KIE value (Fig. S19). Compared to IrO₂ following a conventional AEM (Fig. S25), IrO₂/Co₃O₄ drives acidic OER via the DAEM with the calculated overpotential much decreased and even exceeding the top of the OER volcano plot (Fig. S26). From these theoretical analyses, it can be concluded that the SMOSI induced at the IrO₂/Co₃O₄ interface could increase the E_p of O_bri sites with the strengthened acceptance of protons for accelerating water dissociation at O_bri sites. Consequently, a DAEM could be reasonably proposed and anticipated for the excellent performance for acidic OER over IrO₂/Co₃O₄, by breaking the scaling relationship associated with the conventional AEM (370 ± 100 mV).

To evaluate the application potential of the IrO₂/Co₃O₄ electrode for water splitting at an industrial level, a proton exchange membrane water electrolyzer (PEMWE) was assembled with IrO₂/Co₃O₄ as the anode and Pt/C as the cathode for water electrolysis (Fig. 5a). The recorded polarization curves (Fig. 5b) show cell voltages of 1.64 and 1.77 V required by the IrO₂/Co₃O₄||Pt/C PEMWE to achieve current densities of 0.5 and 1 A cm⁻², respectively, much lower than those of the IrO₂||Pt/C PEMWE (1.95 V at 0.5 A cm⁻² and 2.18 V at 1 A cm⁻²). This significant enhancement in OER performance realized over the IrO₂/Co₃O₄||Pt/C PEMWE could be further proved from the fitted EIS plots (Fig. 5c and Fig. S27), with charge transfer resistance greatly decreased from 4.04 Ω cm² for the IrO₂||Pt/C PEMWE to 0.22 Ω cm² for the IrO₂/Co₃O₄||Pt/C PEMWE, indicating improved kinetics for water electrolysis. The overvoltage, which could be divided into kinetic overvoltage, ohmic overvoltage and mass-transfer overvoltage, was then analyzed to investigate the origin of the excellent water electrolysis activity for the IrO₂/Co₃O₄||Pt/C PEMWE (Fig. S28).⁶⁰ It should be noted that the kinetic overvoltage and mass-transfer overvoltage are much lower in the IrO₂/Co₃O₄||Pt/C PEMWE than in the IrO₂||Pt/C PEMWE at a current density of 1 A cm⁻², suggesting favorable reaction kinetics and fast mass transport. Such improvements can be attributed to the enhanced intrinsic OER activity and self-supported particulate structure of the IrO₂/Co₃O₄ electrode, which are key factors for a high-performance PEMWE. Chronoamperometric tests show that the IrO₂/Co₃O₄||Pt/C PEMWE could be stably operated at current densities of 0.5 and 1 A cm⁻² for at least 120 and 40 h (Fig. 5d), respectively. Encouragingly, benefiting from SMOSI-accelerated deprotonation, the PEMWE with an IrO₂/Co₃O₄ electrode of low-density Ir loading achieves remarkable performance, comparable to previously reported results (Table S6). All the above electrochemical results demonstrate that the IrO₂/Co₃O₄ electrode could serve as a promising anode in a PEMWE, with both excellent activity and stability realized for water electrolysis at an industrial level.

Fig. 5 (a) Schematic illustration of a PEMWE. (b) Polarization curves of a PEMWE assembled with IrO₂/Co₃O₄ or commercial IrO₂ as the anode and Pt/C as the cathode. (c) EIS plots recorded at a cell voltage of 1.5 V. (d) Stability test of the assembled IrO₂/Co₃O₄||Pt/C PEMWE operated at current densities of 1 and 0.5 A cm⁻².

3. Conclusion

In summary, ultrasmall IrO₂ nanoparticles were firmly anchored on an acid-resistant Co₃O₄ support through galvanic replacement, with a low overpotential of 256 mV at 10 mA cm⁻² achieved over the obtained IrO₂/Co₃O₄ electrode for acidic OER and long-term stability exceeding 1000 h. Encouragingly, a PEMWE assembled with IrO₂/Co₃O₄ as the anode and Pt/C as the cathode could survive 120 h and 40 h of operation at industrial-level current densities of 0.5 and 1 A cm⁻², with cell voltages of 1.64 and 1.77 V, respectively. It was experimentally and theoretically revealed that the SMOSI induced at the lattice-mismatched IrO₂/Co₃O₄ interfaces increases the p-band centers of O_bri sites in the Ir–O_bri bonds. Such increased p-band centers strengthen proton acceptance at O_bri sites, accelerating the deprotonation process. Consequently, a DAEM could be reasonably proposed to elucidate the excellent performance for acidic OER over IrO₂/Co₃O₄, by breaking the scaling relationship in a conventional AEM (370 ± 100 mV). This study demonstrates a promising approach for regulating the deprotonation process on O_bri sites via SMOSI for efficient and stable acidic OER, and also inspires the design of applicable and low-noble-metal electrocatalysts for an industrial-level PEMWE.

Author contributions

All authors contributed to the collection and discussion of the content. All authors helped to revise the manuscript before submission.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: Experimental section (material synthesis, characterizations, and DFT calculations), SEM, SEM-EDS, ICP-MS, XRD, EXAFS, electrocatalytic measurements, and their related discussions. See DOI: https://doi.org/10.1039/d5mh01620g.

Acknowledgements

The authors thank the financial support from the National Natural Science Foundation of China (Grant No. 52488201, 52225606), the Key R&D Program of Shaanxi Province (No. 2024CY-GJHX-28), the CNPC Innovation Fund (2022DQ02-0603), the “Fundamental Research Funds for the Central Universities”.

Notes and references

1. J. A. Turner, Science, 2004, 305, 972–974.
2. I. Staffell, D. Scamman, A. Velazquez Abad, P. Balcombe, P. E. Dodds, P. Ekins, N. Shah and K. R. Ward, Energy Environ. Sci., 2019, 12, 463–491.
3. D. Zhao, Y. Wang, C.-L. Dong, Y.-C. Huang, J. Chen, F. Xue, S. Shen and L. Guo, Nat. Energy, 2021, 6, 388–397.
4. H. Deng, C.-L. Dong, Y.-C. Huang, M. Wang, Z. Yu, Y. Wang, H. Li, J. Chen and S. Shen, ACS Mater. Lett., 2024, 6, 3272–3281.
5. L. Chong, G. Gao, J. Wen, H. Li, H. Xu, Z. Green, J. D. Sugar, A. J. Kropf, W. Xu, X.-M. Lin, H. Xu, L.-W. Wang and D.-J. Liu, Science, 2023, 380, 609–616.
6. Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K. Nørskov and T. F. Jaramillo, Science, 2017, 355, eaad4998.
7. H. Liu, Z. Zhang, J. Fang, M. Li, M. G. Sendeku, X. Wang, H. Wu, Y. Li, J. Ge, Z. Zhuang, D. Zhou, Y. Kuang and X. Sun, Joule, 2023, 7, 558–573.
8. Q. Wang, Y. Cheng, H. B. Tao, Y. Liu, X. Ma, D.-S. Li, H. B. Yang and B. Liu, Angew. Chem., Int. Ed., 2023, 62, e202216645.
9. R.-T. Liu, Z.-L. Xu, F.-M. Li, F.-Y. Chen, J.-Y. Yu, Y. Yan, Y. Chen and B. Y. Xia, Chem. Soc. Rev., 2023, 52, 5652–5683.
10. L. Li, G. Zhang, C. Zhou, F. Lv, Y. Tan, Y. Han, H. Luo, D. Wang, Y. Liu, C. Shang, L. Zeng, Q. Huang, R. Zeng, N. Ye, M. Luo and S. Guo, Nat. Commun., 2024, 15, 4974.
11. S. Ge, R. Xie, B. Huang, Z. Zhang, H. Liu, X. Kang, S. Hu, S. Li, Y. Luo, Q. Yu, J. Wang, G. Chai, L. Guan, H.-M. Cheng and B. Liu, Energy Environ. Sci., 2023, 16, 3734–3742.
12. H. Gao, Z. Xiao, S. Du, T. Liu, Y.-C. Huang, J. Shi, Y. Zhu, G. Huang, B. Zhou, Y. He, C.-L. Dong, Y. Li, R. Chen and S. Wang, Angew. Chem., Int. Ed., 2023, 62, e202313954.
13. F.-Y. Chen, Z.-Y. Wu, Z. Adler and H. Wang, Joule, 2021, 5, 1704–1731.
14. A. Li, S. Kong, K. Adachi, H. Ooka, K. Fushimi, Q. Jiang, H. Ofuchi, S. Hamamoto, M. Oura, K. Higashi, T. Kaneko, T. Uruga, N. Kawamura, D. Hashizume and R. Nakamura, Science, 2024, 384, 666–670.
15. J. Xu, H. Jin, T. Lu, J. Li, Y. Liu, K. Davey, Y. Zheng and S.-Z. Qiao, Sci. Adv., 2023, 9, eadh1718.
16. Y. Weng, K. Wang, S. Li, Y. Wang, L. Lei, L. Zhuang and Z. Xu, Adv. Sci., 2023, 10, 2205920.
17. Z. Shi, J. Li, J. Jiang, Y. Wang, X. Wang, Y. Li, L. Yang, Y. Chu, J. Bai, J. Yang, J. Ni, Y. Wang, L. Zhang, Z. Jiang, C. Liu, J. Ge and W. Xing, Angew. Chem., Int. Ed., 2022, 61, e202212341.
18. Y. Wang, M. Zhang, Z. Kang, L. Shi, Y. Shen, B. Tian, Y. Zou, H. Chen and X. Zou, Nat. Commun., 2023, 14, 5119.
19. C. Yang, W. Ling, Y. Zhu, Y. Yang, S. Dong, C. Wu, Z. Wang, S. Yang, J. Li, G. Wang, Y. Huang, B. Yang, Q. Cheng, Z. Liu and H. Yang, Appl. Catal., B, 2024, 358, 124462.
20. S. Pichaikaran, S. Kotteswaran, M. K. Francis, P. B. Bhargav, W. Bo, N. Ahmed and B. C, Mol. Catal., 2023, 547, 113383.
21. W. Q. Zaman, W. Sun, Z.-H. Zhou, Y. Wu, L. Cao and J. Yang, ACS Appl. Energy Mater., 2018, 1, 6374–6380.
22. Y. Zhu, J. Wang, T. Koketsu, M. Kroschel, J.-M. Chen, S.-Y. Hsu, G. Henkelman, Z. Hu, P. Strasser and J. Ma, Nat. Commun., 2022, 13, 7754.
23. G. Li, A. Priyadarsini, Z. Xie, S. Kang, Y. Liu, X. Chen, S. Kattel and J. G. Chen, J. Am. Chem. Soc., 2025, 147, 7008–7016.
24. J. Shan, C. Ye, S. Chen, T. Sun, Y. Jiao, L. Liu, C. Zhu, L. Song, Y. Han, M. Jaroniec, Y. Zhu, Y. Zheng and S.-Z. Qiao, J. Am. Chem. Soc., 2021, 143, 5201–5211.
25. Y. Liu, Y. Chen, X. Mu, Z. Wu, X. Jin, J. Li, Y. Xu, L. Yang, X. Xi, H. Jang, Z. Lei, Q. Liu, S. Jiao, P. Yan, X. Li and R. Cao, ACS Catal., 2023, 13, 3757–3767.
26. K. Hua, X. Li, Z. Rui, X. Duan, Y. Wu, D. Yang, J. Li and J. Liu, ACS Catal., 2024, 14, 3712–3724.
27. J. Rossmeisl, Z. W. Qu, H. Zhu, G. J. Kroes and J. K. Nørskov, J. Electroanal. Chem., 2007, 607, 83–89.
28. I. C. Man, H.-Y. Su, F. Calle-Vallejo, H. A. Hansen, J. I. Martínez, N. G. Inoglu, J. Kitchin, T. F. Jaramillo, J. K. Nørskov and J. Rossmeisl, ChemCatChem, 2011, 3, 1159–1165.
29. J. Song, C. Wei, Z.-F. Huang, C. Liu, L. Zeng, X. Wang and Z. J. Xu, Chem. Soc. Rev., 2020, 49, 2196–2214.
30. Z.-F. Huang, J. Song, S. Dou, X. Li, J. Wang and X. Wang, Matter, 2019, 1, 1494–1518.
31. Z.-F. Huang, J. Song, Y. Du, S. Xi, S. Dou, J. M. V. Nsanzimana, C. Wang, Z. J. Xu and X. Wang, Nat. Energy, 2019, 4, 329–338.
32. Z. Shi, J. Li, Y. Wang, S. Liu, J. Zhu, J. Yang, X. Wang, J. Ni, Z. Jiang, L. Zhang, Y. Wang, C. Liu, W. Xing and J. Ge, Nat. Commun., 2023, 14, 843.
33. A. Grimaud, A. Demortière, M. Saubanère, W. Dachraoui, M. Duchamp, M.-L. Doublet and J.-M. Tarascon, Nat. Energy, 2016, 2, 16189.
34. Y. Wen, C. Liu, R. Huang, H. Zhang, X. Li, F. P. García de Arquer, Z. Liu, Y. Li and B. Zhang, Nat. Commun., 2022, 13, 4871.
35. S. Chen, S. Zhang, L. Guo, L. Pan, C. Shi, X. Zhang, Z.-F. Huang, G. Yang and J.-J. Zou, Nat. Commun., 2023, 14, 4127.
36. H. Jia, Z. Liao, J. Zhu and W. Luo, Renewables, 2024, 2, 204–212.
37. L. Deng, S.-F. Hung, S. Liu, S. Zhao, Z.-Y. Lin, C. Zhang, Y. Zhang, A.-Y. Wang, H.-Y. Chen, J. Peng, R. Ma, L. Jiao, F. Hu, L. Li and S. Peng, J. Am. Chem. Soc., 2024, 146, 23146–23157.
38. H. Wu, J. Chang, J. Yu, S. Wang, Z. Hu, G. I. N. Waterhouse, X. Yong, Z. Tang, J. Chang and S. Lu, Nat. Commun., 2024, 15, 10315.
39. Y. Wu, C. Guo, R. Yao, K. Zhang, J. Li and G. Liu, Adv. Funct. Mater., 2024, 34, 2410193.
40. L.-P. Yuan, W.-J. Jiang, X.-L. Liu, Y.-H. He, C. He, T. Tang, J. Zhang and J.-S. Hu, ACS Catal., 2020, 10, 13227–13235.
41. H. Cheng, C. Wang, D. Qin and Y. Xia, Acc. Chem. Res., 2023, 56, 900–909.
42. J. Huang, H. Sheng, R. D. Ross, J. Han, X. Wang, B. Song and S. Jin, Nat. Commun., 2021, 12, 3036.
43. H. Li, W. Wang, S. Xue, J. He, C. Liu, G. Gao, S. Di, S. Wang, J. Wang, Z. Yu and L. Li, J. Am. Chem. Soc., 2024, 146, 9124–9133.
44. K. Qi, X. Cui, L. Gu, S. Yu, X. Fan, M. Luo, S. Xu, N. Li, L. Zheng, Q. Zhang, J. Ma, Y. Gong, F. Lv, K. Wang, H. Huang, W. Zhang, S. Guo, W. Zheng and P. Liu, Nat. Commun., 2019, 10, 5231.
45. Z. Li, Q. Wang, X. Bai, M. Wang, Z. Yang, Y. Du, G. E. Sterbinsky, D. Wu, Z. Yang, H. Tian, F. Pan, M. Gu, Y. Liu, Z. Feng and Y. Yang, Energy Environ. Sci., 2021, 14, 5035–5043.
46. K. Lee, J. Shim, H. Ji, J. Kim, H. S. Lee, H. Shin, M. S. Bootharaju, K.-S. Lee, W. Ko, J. Lee, K. Kim, S. Yoo, S. Heo, J. Ryu, S. Back, B.-H. Lee, Y.-E. Sung and T. Hyeon, Energy Environ. Sci., 2024, 17, 3618–3628.
47. Y. Chen, H. Li, J. Wang, Y. Du, S. Xi, Y. Sun, M. Sherburne, J. W. Ager, A. C. Fisher and Z. J. Xu, Nat. Commun., 2019, 10, 572.
48. F. Liao, K. Yin, Y. Ji, W. Zhu, Z. Fan, Y. Li, J. Zhong, M. Shao, Z. Kang and Q. Shao, Nat. Commun., 2023, 14, 1248.
49. Z. Shi, Y. Wang, J. Li, X. Wang, Y. Wang, Y. Li, W. Xu, Z. Jiang, C. Liu, W. Xing and J. Ge, Joule, 2021, 5, 2164–2176.
50. W. H. Lee, H. N. Nong, C. H. Choi, K. H. Chae, Y. J. Hwang, B. K. Min, P. Strasser and H.-S. Oh, Appl. Catal., B, 2020, 269, 118820.
51. L. Zhang, Y. Lei, Y. Yang, D. Wang, Y. Zhao, X. Xiang, H. Shang and B. Zhang, Adv. Sci., 2024, 11, 2407475.
52. J. Wang, C.-S. Hsu, T.-S. Wu, T.-S. Chan, N.-T. Suen, J.-F. Lee and H. M. Chen, Nat. Commun., 2023, 14, 6576.
53. S. Geiger, O. Kasian, M. Ledendecker, E. Pizzutilo, A. M. Mingers, W. T. Fu, O. Diaz-Morales, Z. Li, T. Oellers, L. Fruchter, A. Ludwig, K. J. J. Mayrhofer, M. T. M. Koper and S. Cherevko, Nat. Catal., 2018, 1, 508–515.
54. N. Yao, H. Jia, J. Zhu, Z. Shi, H. Cong, J. Ge and W. Luo, Chem, 2023, 9, 1882–1896.
55. L. Giordano, B. Han, M. Risch, W. T. Hong, R. R. Rao, K. A. Stoerzinger and Y. Shao-Horn, Catal. Today, 2016, 262, 2–10.
56. E. C. M. Tse, T. T. H. Hoang, J. A. Varnell and A. A. Gewirth, ACS Catal., 2016, 6, 5706–5714.
57. W. Li, F. Li, H. Yang, X. Wu, P. Zhang, Y. Shan and L. Sun, Nat. Commun., 2019, 10, 5074.
58. Y.-H. Wang, S. Zheng, W.-M. Yang, R.-Y. Zhou, Q.-F. He, P. Radjenovic, J.-C. Dong, S. Li, J. Zheng, Z.-L. Yang, G. Attard, F. Pan, Z.-Q. Tian and J.-F. Li, Nature, 2021, 600, 81–85.
59. S. Zhu, R. Yang, H. J. W. Li, S. Huang, H. Wang, Y. Liu, H. Li and T. Zhai, Angew. Chem., Int. Ed., 2024, 63, e202319462.
60. L. Tao, F. Lv, D. Wang, H. Luo, F. Lin, H. Gong, H. Mi, S. Wang, Q. Zhang, L. Gu, M. Luo and S. Guo, Joule, 2024, 8, 450–460.

---