Engineering intense Ru–TiO₂ interaction for robust hydrogen oxidation reaction

Abstract

Developing high-performance ruthenium (Ru)-based electrocatalysts for alkaline hydrogen oxidation reaction (HOR) is crucial for the practical application of anion exchange membrane fuel cells. However, the inherent high oxophilicity of Ru leads to severe performance degradation at elevated anodic potentials. Herein, we construct a Ru/TiO₂ heterostructure electrocatalyst via a reverse two-step approach that enhances the interaction between Ru and TiO₂. The optimal Ru/TiO₂-400 electrocatalyst exhibits remarkable HOR performance with a mass activity of 0.559 A mg_Ru⁻¹ at 50 mV (vs. RHE) and a specific exchange current density of 0.484 mA cm⁻², which are 2.0 and 4.8 times higher than those of Pt/C, respectively. Notably, the Ru/TiO₂-400 electrocatalyst exhibits remarkable HOR performance with minimal current degradation even at anodic potentials up to 0.6 V (vs. RHE). Experimental results demonstrate that the robust Ru–Ti and Ru–O bonds derived from the intense Ru–TiO₂ interaction effectively prevent Ru from combining with O from the adsorbed OH, thereby enhancing Ru inoxidizability at high potentials. Moreover, the metallic Ru becomes electron rich due to the electron transfer from TiO₂ to Ru, which weakens the adsorption of H reaction intermediates. The Ru and TiO₂ domains at Ru–TiO₂ interfaces are the optimal H and OH adsorption sites, respectively. Therefore, the enhanced electrocatalytic performance of the Ru/TiO₂ electrocatalyst is attributed to the robust and multifunctional Ru–TiO₂ interfaces with intense Ru–TiO₂ interaction.

---

Xiaoyu Zhang

Xiaoyu Zhang has been a Research Fellow at the Zhejiang University of Technology in China since 2022. She received her B.E. degree from the University of Jinan in 2013 and her PhD from Shandong University in 2019. From 2019 to 2022, she worked as a Postdoctoral Fellow at Zhejiang University. Her research focuses on electrocatalysis, fuel cells, and lithium/potassium-ion batteries, with particular emphasis on designing novel electrocatalytic materials and understanding the fundamental electrochemical processes for CO2 reduction, water splitting, and hydrogen oxidation reactions. She has published more than 20 papers as the first or corresponding author and has received multiple research grants.

Introduction

Anion exchange membrane fuel cells (AEMFCs) have been considered as a potential energy-conversion technology, since they offer the advantages of applying cost-effective membranes and cathodic non-precious metal electrocatalysts, when compared with proton exchange membrane fuel cells (PEMFCs).^1,2 However, a key challenge hindering the commercialization of AEMFCs lies in the sluggish reaction kinetics of the anode hydrogen oxidation reaction (HOR).³ Platinum (Pt)-based electrocatalysts demonstrate excellent HOR performance in acidic environments, but their activity in alkaline solutions is significantly reduced, showing a decrease in the reaction rate by 2–3 orders of magnitude.^4,5 This degradation derived from strong hydrogen binding energy and weak hydroxyl adsorption on Pt surfaces under alkaline conditions.⁶ Furthermore, the extreme scarcity and high cost of Pt result in additional difficulty for the commercialization of AEMFCs.^7,8 Therefore, developing cost-effective Pt-free alkaline HOR electrocatalysts is critical to enhance the performance and commercial viability of AEMFCs.

Ruthenium (Ru) is an attractive alternative to Pt due to its comparable hydrogen binding energy and significantly lower price.^9,10 Ru-based electrocatalysts thus hold great promise as efficient HOR electrocatalysts in alkaline media. However, the high oxophilicity of Ru leads to the surface being easily occupied by oxygen species, resulting in performance degradation at elevated anodic potentials (>0.1 V vs. RHE).¹¹ To alleviate the adsorption of oxygen species on Ru sites, various oxophilic supports, such as TiO₂, WO₃, and MoO_x, have been explored to construct Ru-based heterostructures.^12–21 For instance, the reported Ru/WO_2.9 and Ru/TiO₂ heterostructure electrocatalysts typically demonstrate improved HOR kinetics and intrinsic activity, which can be attributed to the abundant hydroxyl (OH) adsorption sites provided by supports and the reduced surface oxophilicity of Ru.^16,20 Besides kinetics and intrinsic activity, the activity decay of Ru-based electrocatalysts at high anodic potentials is also a vital parameter that affects the cycling stability of the fuel cells, which has been overlooked by most existing research.^3,9 A novel strategy is urgently needed to enhance passivation resistance and the overall stability of Ru-based electrocatalysts at high potentials. The metal–support interaction plays a critical role in modulating the electrocatalytic performance of heterogeneous catalysts.²² Specifically, electron transfer between the metal and support can significantly alter the electronic structure of the active metal, thereby enhancing its activity and stability.^23,24 Previous reports have investigated the utilization of metal–support interaction to ameliorate the high-potential passivation resistance of Ru-based electrocatalysts. For example, Wei's group confined Ru clusters within the TiO₂ lattice.¹⁷ Electron transfer from TiO₂ to Ru weakens the hydrogen binding energy (HBE) and oxygen species binding energy on the surface of Ru clusters. As a result, this electrocatalyst achieves efficient acidic/alkaline HOR performance up to 0.9 V (vs. RHE). However, most of the reported Ru-based electrocatalysts, such as Ru NPs/def-TiO₂(A) and Ru–MoO_x, generally exhibit unsatisfactory passivation resistance at high potentials.^15,21 Thus, the metal–support interaction that affects the high-potential performance of Ru-based HOR electrocatalysts still remains to be elucidated. Weak metal–support interaction hinders metal anchoring, while strong metal–support interaction (SMSI) generally results in the overcoverage of oxides on metal surface, which leads to the loss of active sites.²⁴ Therefore, constructing electrocatalysts with an optimal metal–support interaction is considered as an effective strategy to improve the performance of Ru-based catalysts.

Herein, we propose an innovative two-step synthesis approach to construct Ru/TiO₂ heterostructures from RuO₂/TiO₂, which leads to intense Ru–TiO₂ interaction at Ru/TiO₂ heterointerfaces. The interaction promotes electron transfer from TiO₂ to Ru, which optimizes the Ru electronic structure and further tunes the adsorption of H intermediates. In addition, the robust Ru–Ti and Ru–O bonds derived from the Ru–TiO₂ interaction effectively prevent Ru from combining with O from OH intermediates or other oxygen species, endowing the Ru/TiO₂ heterostructure with robust performance even at anodic potentials as high as 0.6 V (vs. RHE). This work provides valuable insights into the role of metal–support interaction in enhancing the activity of Ru-based electrocatalysts for alkaline HOR.

Results and discussion

The synthesis process of Ru/TiO₂ electrocatalysts is illustrated in Fig. 1a, and the detailed procedures are provided in the ESI.† Firstly, defective TiO₂ nanosheets (D-TiO₂) were synthesized via a solvothermal process followed by an annealing treatment.^25,26 A mixture of D-TiO₂, RuCl₃, and NaNO₃ was then calcined to obtain the RuO₂/TiO₂ precursor. Finally, Ru/TiO₂ electrocatalysts were obtained through a reduction annealing treatment of the RuO₂/TiO₂ precursor. Fig. 1b shows the changes in the crystal structure of different samples, as detected by X-ray diffraction (XRD). The RuO₂/TiO₂ precursor exhibits two broad diffraction peaks at 28° and 35° (gray zones), which are ascribed to the (110) and (101) planes of RuO₂ (PDF#40-1290), respectively. After annealing in a H₂/Ar atmosphere at 200 °C, these RuO₂ diffraction peaks disappear, while new diffraction peaks corresponding to the (100), (002), and (101) planes of metallic Ru (PDF#06-0663) appear at 38°, 42°, and 44° (red zones), respectively. The intensities of Ru diffraction peaks increase with increasing annealing temperature, indicating an improvement in the crystallinity of Ru nanoparticles (NPs). Notably, the D-TiO₂ support remains unchanged during the whole synthesis process, as evidenced by the stable diffraction peaks corresponding to TiO₂ (PDF#21-1272). The morphologies of RuO₂/TiO₂ and different Ru/TiO₂ samples were further characterized using the transmission electron microscope (TEM) and high-resolution transmission electron microscope (HRTEM). The lattice spacings of 0.319 and 0.257 nm are identified as the (110) and (101) planes of RuO₂, respectively, while a lattice spacing of 0.351 nm corresponds to the TiO₂ (101) plane (Fig. S1†). After annealing treatment, the morphologies of Ru/TiO₂ electrocatalysts remain similar to that of the RuO₂/TiO₂ precursor (Fig. 1c, d and S2–S4†). The clear lattice fringes observed in the HRTEM images confirm the successful conversion from RuO₂ to metallic Ru. Fig. 1e further exhibits the heterointerface structure of Ru/TiO₂-400. It can be observed that Ru NPs with a lattice spacing of 0.206 nm are well-dispersed on the TiO₂ (101) plane, exposing a distinct Ru–TiO₂ interface (Fig. 1f–h). In addition, the average size and crystallinity of Ru NPs increase with higher annealing temperature (Fig. S5†), which is consistent with the XRD results.

Fig. 1 (a) Schematic diagram of the synthesis of Ru/TiO₂. (b) XRD patterns of RuO₂/TiO₂ and different Ru/TiO₂ samples. (c–e) TEM and HRTEM images of Ru/TiO₂-400. (f–h) Enlarged HRTEM images of the yellow, blue, and green dashed regions in (e), respectively. (i–k) The corresponding FFT patterns of yellow, blue, and green solid-line regions in (e), respectively.

The valence states of Ru in Ru/TiO₂ samples were probed via X-ray photoelectron spectroscopy (XPS). The Ru 3d spectrum of the RuO₂/TiO₂ precursor exhibits typical peaks of RuO₂, which can be deconvoluted into two spin–orbit doublets (Fig. 2a). One group of peaks with binding energies of ≈280.79 and 284.99 eV corresponds to Ru 3d_5/2 and Ru 3d_3/2, while another group of peaks with binding energies of ≈282.69 and 286.89 eV is identified as satellite peaks.^27,28 These satellite peaks of RuO₂ disappear after the annealing treatment. Additionally, as the annealing temperature increases, the Ru 3d_5/2 peaks of Ru/TiO₂ samples shift to lower binding energies, gradually approaching the value of metallic Ru (≈280.2 eV) (Fig. 2a and S6a†).²⁹ Because the overlap between Ru 3p and Ti 2p orbitals is relatively minimal, we further quantitatively analyzed the ratios of Ru⁰ and Ru⁴⁺ in Ru/TiO₂ samples based on Ru 3p spectra (Fig. 2b and S6b†).^29,30 The peaks at ≈461.9 and 484.3 eV are ascribed to the Ru 3p_3/2 and Ru 3p_1/2 peaks of Ru⁰, respectively.^29,31–33 The Ru 3p_3/2 and Ru 3p_1/2 of Ru⁴⁺ can be observed at binding energies of ≈464.1 and 486.5 eV.^34,35 Notably, the RuO₂/TiO₂ precursor already contains a significant amount of Ru⁰ (Ru⁰/(Ru⁰ + Ru⁴⁺) = 0.42), suggesting the presence of oxygen vacancies that could facilitate metal–support interaction.^19,36 The ratio of Ru⁰/(Ru⁰ + Ru⁴⁺) in Ru/TiO₂-200 (0.85) and Ru/TiO₂-300 (0.88) is similar (Fig. S6b†), while the ratio is considered as 1.0 for Ru/TiO₂-400 and Ru/TiO₂-450, since Ru⁴⁺ is almost undetectable in these two samples. However, the Ru 3p_3/2 peak at 461.90 eV exhibits a higher Ru valence state in Ru/TiO₂-400 (Fig. 2b). Overall, the content of Ru⁰ in Ru/TiO₂ increases with the increase of annealing temperature, and Ru⁴⁺ is no longer present on the surfaces of Ru/TiO₂-400 and Ru/TiO₂-450. The binding energy of Ti 2p_3/2 shows a significant positive shift of 0.52–0.65 eV in all samples compared to standard TiO₂ (458.5 eV).³⁷ This shift reveals an intense interaction between Ru and TiO₂, which enhances electron transfer from TiO₂ to Ru NPs. Consequently, an abundant electron supply helps to regulate the electron filling level of the Ru d-band and optimize the adsorption of intermediates on its surface.³⁸ Meanwhile, the decrease in electron cloud density around the Ti atoms facilitates the attraction of OH species.^13,14 Furthermore, the Ti 2p_3/2 peak of Ru/TiO₂-450 (459.02 eV) demonstrates the smallest shift, indicating a relatively weaker Ru–TiO₂ interaction compared to other Ru/TiO₂ samples.

Fig. 2 (a) C 1s and Ru 3d XPS spectra of RuO₂/TiO₂, Ru/TiO₂-400, and Ru/TiO₂-450. (b) Ti 2p and Ru 3p XPS spectra of RuO₂/TiO₂, Ru/TiO₂-400, and Ru/TiO₂-450. (c) Ru K-edge XANES spectra of the samples and reference materials (Ru foil and RuO₂). (d) Ru valence states for samples obtained by linear fitting of the Ru K-edge absorption energy (E₀). (e) Fourier-transformed k³-weighted χ(k)-function of the EXAFS spectra for the Ru K-edge. (f) Relationship between the fitting coordination number and annealing temperature. (g–i) Wavelet transforms of the k³-weighted EXAFS spectra for RuO₂, Ru/TiO₂-400, and Ru foil.

The electronic and coordination structures of RuO₂/TiO₂ and various Ru/TiO₂ samples were further investigated using X-ray absorption spectroscopy (XAS). The X-ray absorption near-edge structure (XANES) spectra of the Ru K-edge are shown in Fig. 2c. The spectral profile of RuO₂/TiO₂ is similar to that of standard RuO₂. After annealing treatment, the absorption edge energy decreases in the order of Ru/TiO₂-200 > Ru/TiO₂-300 > Ru/TiO₂-400 ≈ Ru/TiO₂-450, indicating an inverse correlation with the annealing temperature. Linear fitting of Ru K-edge absorption energy (E₀) quantifies the average valence state of Ru as 3.39, 0.31, 0.25, 0.14, and 0.13 for RuO₂/TiO₂, Ru/TiO₂-200, Ru/TiO₂-300, Ru/TiO₂-400, and Ru/TiO₂-450, respectively (Fig. 2d).³⁹ Fourier-transformed extended X-ray absorption fine-structure (FT-EXAFS) spectra were collected to investigate the local coordination environment (Fig. 2e), and the fitting results are listed in Fig. S7–S11 and Table S1.† As shown in Fig. 2e, RuO₂/TiO₂ possesses a Ru–O bond peak near 1.96 Å, while the Ru–O–Ru coordination at ≈3.5 Å presents a much weaker intensity compared to the reference RuO₂.^40,41 This reduced intensity may be attributed to the RuO₂–TiO₂ interaction through Ru–O–Ti bonds. A well-defined peak at 2.67 Å corresponding to the Ru–Ru bond is observed in all Ru/TiO₂ samples.⁴¹Fig. 2f reveals that the coordination number of Ru–Ru increases with increasing temperature, while the coordination number of Ru–O decreases. Wavelet transform (WT) analysis further reveals changes in the Ru coordination environment (Fig. 2g–i and S12†).^42,43 RuO₂/TiO₂ exhibits reduced intensity of the Ru–O–Ru center compared to RuO₂, which is consistent with the observed decrease in the peak near 3.5 Å in R space EXAFS spectra. Notably, the Ru–Ru center of Ru/TiO₂-400 shifts slightly towards lower k numbers, indicating that some Ru atoms are bonded with lighter atoms, specifically Ti.^44,45 This provides direct evidence for the Ru–Ti connection. Besides, all Ru/TiO₂ samples retain a small amount of Ru–O bonds. Since XPS results indicate that Ru–O bonds are not present on the surface, these bonds are probably located at the Ru/TiO₂ heterointerfaces. Therefore, Ru NPs are connected to D-TiO₂ through Ru–Ti and Ru–O bonds, which facilitates electron transfer and regulates the electronic structure of Ru.

The HOR performance of Ru/TiO₂ electrocatalysts was examined using a rotating disk electrode (RDE) at a rotation speed (ω) of 1600 rpm and a scan rate of 5 mV s⁻¹ in a H₂-saturated 0.1 M KOH solution. As shown in Fig. 3a, Ru/TiO₂ electrocatalysts exhibit significantly enhanced HOR performance at high potentials compared to most reported Ru-based electrocatalysts. The optimal Ru/TiO₂-400 displays negligible current decay up to 0.6 V (vs. RHE). This excellent catalytic current originates from the HOR process, since the electrocatalysts possess currents of 0 in an Ar-saturated 0.1 M KOH solution (Fig. S13†). The valence state of Ru affects the HOR performance. For instance, Ru/TiO₂-450 with the lowest Ru valence state presents the highest extreme diffusion current density (j_d). However, the current decay reoccurs at high potentials in Ru/TiO₂-450 due to the weakened Ru–TiO₂ interaction compared to Ru/TiO₂-400. Besides, the difference between the Ru–O coordination number of Ru/TiO₂-400 and that of Ru/TiO₂-450 demonstrates that an optimal number of Ru–O bonds could prevent the oxidation of metal Ru at high potentials. The relationship between the inverse of current density (j⁻¹) and ω^−1/2 of Ru/TiO₂-400 follows a linear relationship with a slope of 4.96 cm² mA⁻¹ s^−1/2, which is close to the theoretical value (4.87 cm² mA⁻¹ s^−1/2) for a two-electron reaction (Fig. 3b).⁵ The kinetic current densities (j_k) are calculated through the Koutecky–Levich equation and further fitted with the Butler–Volmer equation to obtain the exchange current densities (j₀) (Fig. 3c). Both j_k and j₀ increase with higher annealing temperature, indicating improved HOR kinetics. Geometric j₀ values of Ru/TiO₂-400 and Ru/TiO₂-450 are calculated to be 4.01 and 5.92 mA cm⁻², respectively (Table S2†). To ensure quality normalization, inductively coupled plasma-atomic emission spectroscopy (ICP-OES) was conducted to assess the Ru content of Ru/TiO₂ (11.42 wt%, Table S3†). The mass-normalized exchange current densities (j_0,m) and mass activity (j_k,m) of electrocatalysts clearly demonstrate that the activity and kinetics of Ru/TiO₂ improve with increasing annealing temperature (Fig. 3d). Ru/TiO₂-400 possesses a j_k,m of 0.559 A mg_Ru⁻¹, which is 2.0 times higher than that of commercial Pt/C (0.272 A mg_Ru⁻¹, Table S3†). Meanwhile, Ru/TiO₂-450 has an even higher j_k,m of 1.529 A mg_Ru⁻¹ than that of Ru/TiO₂-400. The electrochemical active surface areas (ECSAs) of the as-prepared samples were evaluated via a Cu underpotential deposition (Cu_UPD) stripping experiment.⁴⁶ Ru/TiO₂-300 exhibits the highest ECSA (38.31 m² g_Ru⁻¹, Fig. S14†) among Ru/TiO₂ electrocatalysts, since the reduction of RuO₂ generates more active metallic Ru sites. However, the sintering of Ru nanoparticles occurs when the annealing temperature exceeds 300 °C, which is consistent with TEM images, leading to a decrease in the ECSA. The ECSA normalized exchange current densities (j_0,s) of Ru/TiO₂-400 (0.484 mA cm⁻²) and Ru/TiO₂-450 (0.921 mA cm⁻²) are 4.8 and 9.2 times higher than that of Pt/C (0.10 mA cm⁻²), respectively. This indicates that the kinetics of individual active sites has been improved, possibly owing to Ru–TiO₂ interaction. Durability tests performed on the RDE at 0.1 V (vs. RHE) are illustrated in Fig. S15.† Based on the above discussion, Ru/TiO₂-400 with reasonably fast kinetics, high activity, and acceptable durability can be identified as the preferred electrocatalyst, especially due to its performance at high potentials. This outstanding performance derives from the optimized Ru valence state and the intense Ru–TiO₂ interaction. Electron-rich Ru provides generous H adsorption sites for the HOR, making the Ru valence state positively correlated with HOR kinetics and activity. TiO₂ at the Ru–TiO₂ heterointerface is the optimal OH adsorption sites.¹⁴ Furthermore, the intense Ru–TiO₂ interaction leads to robust Ru–Ti and Ru–O bonds, helping to prevent Ru surfaces from being occupied by oxygen species at elevated anodic potentials.

Fig. 3 (a) LSV curves of RuO₂/TiO₂, Ru/TiO₂ samples, and commercial Pt/C. (b) LSV curves of Ru/TiO₂-400 at different rotation rates (ω) of 400, 900, 1600, and 2500 rpm. Inset shows the Koutecky–Levich plot at an overpotential of 20 mV. (c) Kinetic current density at 1600 rpm and fitting based on the Butler–Volmer equation. (d) Comparison of the mass-normalized exchange current densities (j_0,m) and mass activity (j_k,m, η = 50 mV) of different samples. (e) CV curves of Ru/TiO₂ samples and commercial Pt/C. Inset shows the relationship between hydrogen desorption potentials and annealing temperature. (f) CO stripping voltammograms of Ru/TiO₂-400, Ru/TiO₂-H, and Ru/TiO₂-P.

The catalytic activity of electrocatalysts in alkaline media is closely related to the binding energy of adsorbed hydrogen (H_ad) and adsorbed hydroxyl (OH_ad) species on the electrocatalyst surface.⁴ The binding energy of H_ad species was evaluated by an underpotential deposited hydrogen (H_UPD) experiment. The H_UPD desorption peaks of Ru/TiO₂ electrocatalysts shift to lower potential compared to Pt/C (0.298 V vs. RHE), indicating weakened hydrogen binding energies (Fig. 3e).¹¹ Therefore, the enriched electrons on Ru weaken the adsorption of H_ad species on its surface.³⁸ CO stripping tests were performed to evaluate the hydroxyl binding energies (OHBEs) of electrocatalysts, as OH_ad species participate in the oxidation process of CO.¹⁶ However, no CO oxidation peak was observed during CO stripping tests for Ru/TiO₂ electrocatalysts, which is probably attributed to the enhanced electron transfer caused by intense Ru–TiO₂ interaction (Fig. 3f and S16†). A typical CO adsorption process includes donation and back-donation of CO electrons, according to the classical model of CO–metal bonding.¹⁷ Electrons from TiO₂ reduce the number of empty orbitals of Ru and limit the donation of CO electrons to Ru, and thus hinder CO adsorption.¹⁷ To further highlight the role of the intense Ru–TiO₂ interaction in Ru/TiO₂ electrocatalysts, comparison samples were synthesized by preparing the same amount of Ru on D-TiO₂via one-step hydrothermal and pyrolysis approaches, denoted as Ru/TiO₂-H and Ru/TiO₂-P, respectively. These samples possess similar morphology to that of Ru/TiO₂, but show obvious CO oxidation peaks and high-potential passivation (Fig. 3f, S17 and S18†). The effect of the Ru valence state has been excluded using Ru 3p XPS spectra (Fig. S19†), since the Ru 3p_3/2 peaks of Ru/TiO₂-H (461.85 eV) and Ru/TiO₂-P (461.90 eV) are close to that of Ru/TiO₂-400 (461.90 eV). However, the positive shifts of Ti 2p_3/2 peaks for Ru/TiO₂-H (458.85 eV) and Ru/TiO₂-P (458.89 eV) compared to standard TiO₂ (458.5 eV) are not as remarkable as that of Ru/TiO₂-400 (459.07 eV), indicating a stronger electron transfer in Ru/TiO₂-400. This enhanced electronic effect provides evidence for the intense Ru–TiO₂ interaction in Ru/TiO₂ electrocatalysts. In addition, EXAFS spectra reveal similar Ru–Ru coordination in both Ru/TiO₂-H and Ru/TiO₂-P samples to that of Ru/TiO₂-400, and the coordination structure of Ru is not the primary factor affecting HOR performance (Fig. S20†). Therefore, it can be inferred that the excellent anti-passivation ability of Ru/TiO₂ at high potentials derives from the intense Ru–TiO₂ interaction generated during the reverse two-step process. Ru/TiO₂-H with the weakest Ru–TiO₂ interaction demonstrates the most significant current decay, which also confirms the above viewpoint. Notably, Ru/TiO₂-H with a lower Ti 2p_3/2 binding energy possesses a CO oxidation peak at a lower potential (0.670 V vs. RHE) compared to Ru/TiO₂-P (0.789 V vs. RHE), revealing a stronger OHBE on Ru sites (Fig. 3f).^47,48 Thus, the OHBE of Ru/TiO₂-400 may be the weakest due to its highest Ti 2p_3/2 binding energy. The robust Ru–Ti and Ru–O bonds tightly anchor Ru NPs, which increases the energy barrier for oxygen species adsorption, thereby weakening the OHBE on the Ru surface and enhancing its oxidation resistance. The multifunctional Ru/TiO₂ heterostructure with intense Ru–TiO₂ interaction optimizes the adsorption of intermediates. Electron-rich Ru serves as the preferred adsorption site for H_ad species, while TiO₂ at the heterointerface is more favorable to adsorb OH_ad species. Moreover, Ru/TiO₂ exhibits excellent alkaline HER performance (Fig. S21†). The optimal Ru/TiO₂ electrocatalyst possesses a lower overpotential compared to that of commercial Pt/C and faster HER kinetics, which can also be attributed to the appropriate adsorption of H_ad and OH_ad species on Ru/TiO₂ heterointerfaces.

Conclusions

In summary, we have successfully developed a Ru/TiO₂ heterostructure electrocatalyst with enhanced alkaline HOR electrocatalytic performance at high potentials. The strategic transformation from RuO₂/TiO₂ to Ru/TiO₂ facilitates the formation of intense interaction between Ru and TiO₂, which prevents Ru from being oxidized and optimizes the electronic structure of Ru. As a result, the Ru/TiO₂ electrocatalysts demonstrate a high mass activity and maintain improved performance even at anodic potentials up to 0.6 V (vs. RHE). The high performance is attributed to the intense Ru–TiO₂ interaction that enables appropriate adsorption of reaction intermediates on Ru/TiO₂ heterointerfaces. This work reveals the significance of optimizing metal–support interaction in overcoming the performance degradation at elevated anodic potentials in Ru-based electrocatalysts.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

W. S. conceived the idea for the experiments. X. J. designed and executed experiments. X. J. and X. Z. wrote and revised the manuscript. B. Y. and X. Z. assisted in data analysis. W. S., H. P. and M. G. provide financial support and equipment for the experiments.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (52171224 and 92261119), National Natural Science Foundation Joint Fund Project (U24A2042), National Science Foundation of Zhejiang Province (LZ22B030006), and Basic Research Foundation of Zhejiang Provincial Universities (G23224161033). The authors thank beamline BL14W1 in the Shanghai Synchrotron Radiation Facility (SSRF) for XAFS characterization.

Notes and references

1. J. Hyun and H.-T. Kim, Energy Environ. Sci., 2023, 16, 5633–5662.
2. Y. X. Yang, P. Li, X. B. Zheng, W. P. Sun, S. X. Dou, T. Y. Ma and H. G. Pan, Chem. Soc. Rev., 2022, 51, 9620–9693.
3. J. T. Ren, L. Chen, H. Y. Wang, Y. Feng and Z. Y. Yuan, Energy Environ. Sci., 2024, 17, 3960–4009.
4. L. X. Su, H. Wu, S. K. Zhang, C. X. Cui, S. N. Zhou and H. Pang, Adv. Mater., 2024, 37, 2414628.
5. X. Y. Tian, P. C. Zhao and W. C. Sheng, Adv. Mater., 2019, 31, 1808066.
6. X. Q. Mu, S. L. Liu, L. Chen and S. C. Mu, Small Struct., 2023, 4, 2200281.
7. X. Zhang, Y. Xie and L. Wang, Nano Res., 2023, 17, 960–981.
8. Y. W. Cao, Y. H. Mou, J. L. Zhang, R. Zhang and Z. Z. Liang, Mater. Today Catal., 2024, 4, 100044.
9. X. Y. Zhang, X. Z. Xiao, J. Chen, Y. Y. Liu, H. G. Pan, W. P. Sun and M. X. Gao, Energy Environ. Sci., 2022, 15, 4511–4526.
10. J. Kim and J. Lee, Energy Fuels, 2023, 37, 17765–17781.
11. X. Y. Zhang, L. X. Xia, G. Q. Zhao, B. X. Zhang, Y. P. Chen, J. Chen, M. X. Gao, Y. Z. Jiang, Y. F. Liu, H. G. Pan and W. P. Sun, Adv. Mater., 2023, 35, 2208821.
12. L. Jing, G. Jie, W. Q. Yu, H. W. Ren, X. J. Cui, X. Chen and L. H. Jiang, Chem. Eng. J., 2023, 472, 145009.
13. Y. Y. Cong, M. L. Liu, L. Y. Chen, Y. R. Qi, L. M. Zhang, Q. P. Zhao, C. L. Li and H. B. Wang, Chem. Eng. J., 2024, 501, 157669.
14. L. Y. Chen, C. L. Li, M. L. Liu, Z. R. Dai, H. B. Wang, X. Zhou, Q. P. Zhao and Y. Y. Cong, Adv. Sci., 2024, 11, 2410881.
15. X. T. Fu, X. X. Huang, Y. P. Cen, X. Y. Ren, L. Yan, S. Jin, Z. B. Zhuang, W. L. Li and S. B. Tian, Small, 2024, 20, 2406387.
16. J. X. Jiang, S. C. Tao, Q. He, J. Wang, Y. Y. Zhou, Z. Y. Xie, W. Ding and Z. D. Wei, J. Mater. Chem. A, 2020, 8, 10168–10174.
17. Y. Y. Zhou, Z. Y. Xie, J. X. Jiang, J. Wang, X. Y. Song, Q. He, W. Ding and Z. D. Wei, Nat. Catal., 2020, 3, 454–462.
18. Y. C. Pi, Z. M. Qiu, Y. Sun, H. Ishii, Y. F. Liao, X. Y. Zhang, H. Y. Chen and H. Pang, Adv. Sci., 2023, 10, 2206096.
19. H. Liu, Z. Zhang, M. X. Li, Y. P. Li, Y. Kuang and X. M. Sun, Nanoscale, 2023, 15, 12064–12070.
20. Z. Y. Cheng, Y. Yang, P. C. Wang, P. C. Wang, J. H. Yang, D. D. Wang and Q. W. Chen, Small, 2024, 20, 2307780.
21. L. G. Li, C. Liu, S. H. Liu, J. Wang, J. J. Han, T. S. Chan, Y. Y. Li, Z. W. Hu, Q. Shao, Q. B. Zhang and X. Q. Huang, ACS Nano, 2022, 16, 14885–14894.
22. T. W. v. Deelen, C. H. Mejía and K. P. d. Jong, Nat. Catal., 2019, 2, 955–970.
23. K. F. Xie, K. Xu, M. Q. Liu, X. H. Song, S. Y. Xu and H. Y. Si, Mater. Today Catal., 2023, 3, 100029.
24. Z. X. Luo, G. Q. Zhao, H. G. Pan and W. P. Sun, Adv. Energy Mater., 2022, 12, 2201395.
25. G. Xiang, T. Li, J. Zhuang and X. Wang, Chem. Commun., 2010, 46, 6801–6803.
26. X. J. Wang, X. H. Wan, X. X. Qin, C. Chen, X. S. Qian, Y. Z. Guo, Q. J. Xu, W. B. Cai, H. Yang and K. Jiang, ACS Catal., 2022, 12, 9437–9445.
27. Y. Wang, Y. X. Hao, L. Q. Wang, C. S. Li, J. W. Ren, Y. Sun, F. Hu, L. L. Li and S. J. Peng, Inorg. Chem. Front., 2023, 10, 6015–6022.
28. J. X. Zhang, R. Lin, Y. C. Zhao, H. Wang, S. C. Liu and X. Cai, ACS Sustain. Chem. Eng., 2023, 11, 9489–9497.
29. Z. M. Wang, Q. L. Hong, X. H. Wang, H. Huang, Y. Chen and S. N. Li, Acta Phys.–Chim. Sin., 2023, 39, 2303028.
30. X. Mu, X. Zhang, Z. Chen, Y. Gao, M. Yu, D. Chen, H. Pan, S. Liu, D. Wang and S. Mu, Nano Lett., 2024, 24, 1015–1023.
31. Z. H. Liang, H. B. Liu, S. Q. Huang, M. H. Xing, Z. Y. Li, S. T. Wang, L. Yang and D. P. Cao, J. Mater. Chem. A, 2023, 11, 849–857.
32. H. Zhang, T. Y. Gao, Q. Z. Zhang, B. Gu, Q. H. Tang, Q. Cao and W. H. Fang, Mater. Today Catal., 2023, 3, 100013.
33. H. B. Yang, D. W. Ma, Y. Li, Q. H. Zhao, F. Pan, S. S. Zheng and Z. R. Lou, Chin. J. Struct. Chem., 2023, 42, 100031.
34. Q. K. Wu, W. J. Yang, X. D. Wang, W. Zhu, S. S. Lv, Y. Zhou, T. Y. Chen, S. H. Liu, W. Y. Li and Z. Chen, Appl. Catal., B, 2023, 335, 122896.
35. Y. Liu, L. R. Cheng, S. Q. Zhou, Y. T. Yang, C. G. Niu, T. T. Isimjan, B. Wang and X. l. Yang, J. Energy Chem., 2024, 94, 332–339.
36. Z. Z. Wei, Z. J. Zhao, J. Wang, Q. Zhou, C. X. Zhao, Z. H. Yao and J. G. Wang, J. Mater. Chem. A, 2021, 9, 10160–10168.
37. M. Kuang, Y. Wang, W. Fang, H. Tan, M. Chen, J. Yao, C. Liu, J. Xu, K. Zhou and Q. Yan, Adv. Mater., 2020, 32, 2002189.
38. Y. N. Gao, B. Ouyang, Y. Shen, W. Wen, J. X. Wang, M. Z. Wang, Y. Q. Sun and K. Xu, Adv. Energy Mater., 2025, 2406114.
39. W. D. He, X. H. Tan, Y. Y. Guo, Y. H. Xiao, H. Cui and C. X. Wang, Angew. Chem., Int. Ed., 2024, 63, e202405798.
40. Y. Zuo, S. Bellani, G. Saleh, M. Ferri, D. V. Shinde, M. I. Zappia, J. Buha, R. Brescia, M. Prato, R. Pascazio, A. Annamalai, D. O. de Souza, L. De Trizio, I. Infante, F. Bonaccorso and L. Manna, J. Am. Chem. Soc., 2023, 145, 21419–21431.
41. Y. Li, W. Wang, M. Cheng, Y. Feng, X. Han, Q. Qian, Y. Zhu and G. Zhang, Adv. Mater., 2023, 35, 2206351.
42. H. Funke, M. Chukalina and A. Rossberg, Phys. Scr., 2005, T115, 232–234.
43. H. Funke, M. Chukalina and A. C. Scheinost, J. Synchrotron Radiat., 2007, 14, 426–432.
44. Z. X. Luo, X. Han, Z. T. Ma, B. X. Zhang, X. S. Zheng, Y. F. Liu, M. X. Gao, G. Q. Zhao, Y. Lin, H. G. Pan and W. P. Sun, Angew. Chem., Int. Ed., 2024, 63, e202406728.
45. J. Timoshenko and A. Kuzmin, Comput. Phys. Commun., 2009, 180, 920–925.
46. B. X. Zhang, B. H. Zhang, G. Q. Zhao, J. M. Wang, D. Q. Liu, Y. P. Chen, L. X. Xia, M. X. Gao, Y. F. Liu, W. P. Sun and H. G. Pan, Nat. Commun., 2022, 13, 5894.
47. B. X. Zhang, J. M. Wang, G. M. Liu, C. M. Weiss, D. Q. Liu, Y. P. Chen, L. X. Xia, P. Zhou, M. X. Gao, Y. F. Liu, J. Chen, Y. S. Yan, M. H. Shao, H. G. Pan and W. P. Sun, Nat. Catal., 2024, 7, 441–451.
48. L. P. Zhang, S. J. Chen, T. H. Du, X. Z. Zhao, A. Q. Dong, L. F. Zhang, T. F. Li, L. B. Li, C. L. Yan and T. Qian, ACS Nano, 2024, 18, 34195–34206.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta02938d

---