Proton Donors Boost Interfacial Water Activation on RuO₂‐Embedded RuWO_x for Durable Acidic Oxygen Evolution

ABSTRACT

The activity‐stability dilemma in acidic oxygen evolution reaction (OER) stems from a fundamental trade‐off in the interface hydrogen‐bond network: while promoting interfacial water enrichment, it often impedes efficient proton transfer. Achieving simultaneous enhancement of interfacial water accumulation and deprotonation kinetics remains a critical challenge. Herein, we design a tungsten‐stabilized ruthenium oxide heterostructure (RuWO_x‐300) that overcomes this limitation in acidic OER. The high‐valence W⁶⁺ species induce a localized electrostatic field, which enhances interfacial water enrichment. Concurrently, the W‐O_bri‐Ru moieties serve as active Brønsted acid sites with high proton‐donor capability, accelerating surface deprotonation. As a result, RuWO_x‐300 exhibits exceptional OER performance, requiring overpotentials of only 179 and 231 mV to achieve 10 and 100 mA cm⁻², respectively, and demonstrating remarkable stability for over 4820 h at 50 mA cm⁻², surpassing commercial RuO₂ and most reported Ru‐based catalysts. Through ab initio molecular dynamics (AIMD) simulations and in situ Raman spectroscopy, we elucidate a dual‐functional mechanism: increased interfacial water coverage reduces the activation barrier for O─O bond formation, while proton‐donor‐enhanced transfer kinetics suppresses Ru dissolution. This work establishes a new design principle for highly active and stable acidic OER electrocatalysts via proton‐donor‐mediated interfacial water activation.

A tungsten‐stabilized amorphous RuWO_x/RuO₂ heterostructure breaks the activity‐stability trade‐off in acidic OER by enriching and activating interfacial water molecules, while its Brønsted‐acidic W‐O_bri‐Ru sites with high proton‐donor capability facilitate proton transfer to suppress catalyst degradation.

1. Introduction

The conversion of intermittent renewable energy (e.g., solar and wind) into storable hydrogen fuels relies critically on efficient and durable water electrolysis technologies [1, 2]. Among these, proton exchange membrane water electrolysis (PEMWE) has gained prominence due to its ability to overcome limitations inherent in alkaline systems, including gas crossover, low operating pressure, and limited current density, while offering superior energy efficiency [3, 4]. However, the widespread adoption of PEMWE is hindered by the sluggish kinetics of the anodic oxygen evolution reaction (OER) under acidic media, necessitating electrocatalysts with exceptional activity and corrosion resistance [5, 6]. Currently, noble metal oxides, particularly iridium‐ and ruthenium oxides (IrO₂ and RuO₂), serve as the benchmark anodic catalysts in PEMWE [7]. Among them, RuO₂ exhibits superior intrinsic OER activity and cost‐effectiveness compared to IrO₂, highlighting its considerable potential for practical application [8, 9]. Nevertheless, Ru sites are prone to over‐oxidation under anodic potentials, leading to the formation of soluble RuO₄ species and rapid catalyst degradation [8, 9]. Therefore, designing Ru‐based OER catalysts that simultaneously achieve high activity and long‐term stability remains a pivotal challenge.

To address this issue, strategies such as doping and hybridization have been extensively employed to fine‐tune the adsorption energetics of reaction intermediates on Ru‐based catalysts [10, 11]. Despite these efforts, the overall OER performance remains constrained by the thermodynamics of water activation and kinetics of proton‐coupled electron transfer [12, 13, 14, 15]. Recent studies emphasize the pivotal role of the interfacial water structure within the electrical double layer (EDL), which governs both reactant activation and deprotonation efficiency [8, 16]. In acidic media, water molecules form an extended hydrogen‐bond network composed primarily of 4‐coordinated (4‐HB‐H₂O) and 2‐coordinated (2‐HB‐H₂O) structures, which is more organized than that in alkaline environments [16, 17]. This network imposes a high kinetic barrier for water molecules to access active sites within the inner Helmholtz plane and impedes the rate‐limiting water dissociation step. Under operating potentials, the resulting scarcity of interfacial free water can induce lattice oxygen participation in the OER, triggering irreversible structural degradation and catalyst deactivation [18].

Although a weakening hydrogen‐bond network could enhance local water accessibility and reduce the activation barrier for water dissociation, such an approach inevitably slows down proton transfer kinetics. The resulting disruption of the proton‐transfer pathway leads to proton accumulation at the catalyst interface, which promotes electron depletion from Ru active sites and accelerates dissolution via over‐oxidation [19, 20]. Therefore, rational engineering the interfacial microenvironment to simultaneously facilitate free water accessibility and proton transfer kinetics offers a promising strategy to break the intrinsic activity‐stability trade‐off in Ru‐based OER electrocatalysts.

Herein, we propose an interfacial engineering strategy that synergistically enhances both interfacial water coordination and proton transfer kinetics to resolve the long‐standing activity‐stability dilemma in acidic OER. Through atomic‐level control of crystallization kinetics, we precisely constructed a hetero‐phase architecture consisting of an amorphous RuWO_x (0< x < 3) matrix with embedded strain‐stabilized microcrystalline RuO₂ domains. This unique structure generates oxygen vacancies and W‐O_bri‐Ru Brønsted acid sites with high proton‐donor capability. The synergistic interaction between W⁶⁺ cations and oxygen vacancies creates a strong local electrostatic field, which promotes the adsorption of free water at the electrode‐electrolyte interface while suppressing lattice oxygen oxidation. Moreover, the W‐O_bri‐Ru motifs, serving as localized proton donors, facilitate proton transfer from the oxo‐intermediates to adjacent O_bri, alleviating the excessive proton adsorption typically observed at conventional Ru‐O_bri‐Ru configurations. The resulting RuWO_x‐300 catalyst exhibits outstanding OER performance, requiring an overpotential of only 179 mV at 10 mA cm⁻² and demonstrating exceptional stability for over 4820 h at 50 mA cm⁻² in acidic electrolyte. When integrated into a PEMWE device, RuWO_x‐300 enables stable operation for 200 h at 0.5 A cm⁻², underscoring its practical applications. This work provides a rational design strategy for high‐performance acidic OER electrocatalysts through proton‐donor‐mediated interfacial microenvironment engineering, opening a pathway toward scalable and cost‐effective green hydrogen production.

2. Results and Discussion

2.1. Theoretical Investigations

Conventional Ru‐based catalysts undergo rapid dissolution, a process primarily driven by electron loss at Ru sites due to insufficient water replenishment and interfacial proton accumulation (Figure 1a). To address this, we designed a W‐doped amorphous RuWO_x catalyst with embedded RuO₂ microcrystals. The incorporation of high‐valence W⁶⁺ promotes the formation of oxygen vacancies and W‐O_bri‐Ru Brønsted acid sites. These sites generate localized positive charges, which establish a strong interfacial electric field that preferentially attracts the oxygen atom (δ⁻) of polar water molecules, thereby enriching free water at the interface [21]. Simultaneously, the engineered W‐O_bri‐Ru motif serves as efficient Brønsted acid sites, which facilitate the deprotonation process and significantly enhance water dissociation kinetics (Figure 1b) [20, 22]. This dual functionality creates preferential proton transfer pathways while maintaining optimal binding energies for reaction intermediates, thereby overcoming the kinetic limitations inherent to conventional Ru‐based catalysts.

FIGURE 1.

Modulating the interfacial microenvironment via Brønsted Acid Sites for stable acidic OER. (a, b) Schematic illustrations contrasting the conventional RuO₂ catalyst, which suffers from Ru dissolution due to insufficient interfacial water and sluggish proton transfer (a), with the proposed RuWO_x‐300 catalyst, where Brønsted acid sites optimize the microenvironment to enhance water enrichment and facilitate proton transfer, thereby stabilizing Ru sites (b). (c, d) AIMD simulations depicting the orientation and hydrogen‐bonding network of interfacial water molecules on (c) com. RuO₂ and (d) RuWO_x‐300. N H‐bonds: number of hydrogen bonds; IHP: inner Helmholtz plane; OHP: outer Helmholtz plane. (e) Calculated hydrogen adsorption energy (E _ads‐H) on the O_bri sites. (f) Schematic of the proton transfer pathway facilitated by the W‐O_bri‐Ru motif. (g) Schematic band structure diagram illustrating the electronic interaction between W and Ru.

To unravel the role of W‐doping in modulating interfacial water orientation within the EDL, we carried out ab initio molecular dynamics (AIMD) simulations. Representative snapshots and quantitative hydrogen‐bond characterization demonstrate a 18.6% reduction in coordinated water molecules at the inner Helmholtz plane (IHP) on RuWO_x‐300 compared to commercial RuO₂ (Figure 1c,d, and Figure S1). This indicates a significant disruption of the rigid hydrogen‐bond network, which facilitates the enrichment of reactive free water. Although such network disruption typically impedes proton conductivity, the engineered W‐O_bri‐Ru Brønsted acid sites exhibit an optimal proton affinity, with a calculated hydrogen adsorption (E _ads‐H) of 0.17 eV, markedly lower than the ‐0.70 eV for conventional Ru‐O_bri‐Ru configurations (Figure 1e and Figure S2). This suitable energy enables spontaneous proton transfer, effectively overcoming the limitations of classical conduction mechanisms.

The detailed deprotonation pathway on RuWO_x‐300 is illustrated in Figure 1f. The *OOH intermediate first adsorbs on the Ru site, followed by adjacent Brønsted acid‐assisted deprotonation, where the hydrogen migrates to the bridging oxygen (O_bri). This is followed by the desorption of the proton from the O_bri site. Furthermore, density‐of‐states analysis (Figure 1g) reveals strengthened hybridization among Ru 4d, W 5d, and O 2p orbitals. W incorporation leads to a widening of the energy gap (Δɛ) between the Ru d‐band and O p‐band centers. Such electronic‐structure modulation may influence charge‐transfer behavior and help suppress Ru over‐oxidation after oxygen‐vacancy formation [23]. The constructed model was designed to capture the representative local coordination environment at the amorphous/crystalline interface, particularly the W‐O_bri‐Ru motifs that are directly relevant to proton transfer and electronic regulation, thereby providing an appropriate basis for mechanistic investigation of their role in OER catalysis. These findings suggest that engineering proton‐transfer‐active Brønsted acid sites may be an effective strategy for simultaneously regulating interfacial water structure and proton‐transfer kinetics, offering a pathway to overcome the activity‐stability trade‐off in acidic OER catalysts.

2.2. Material Synthesis and Characterization

Guided by theoretical predictions, we synthesized hybrid RuWO_x catalysts via a hydrothermal‐annealing approach (Figure 2a). Briefly, an aqueous solution containing RuCl₃ and Na₂WO₄ was hydrothermally treated at 180°C for 12 h, yielding an amorphous precursor (Figure S3a). The precursor was subsequently calcined in air at different temperatures to regulate crystallization. In situ XRD analysis tracked the phase evolution from 25°C to 500°C, revealing a systematic shift of the RuO₂ (110) diffraction peak (PDF#00‐043‐1027) toward higher 2θ angles, indicative of compressive lattice strain in the microcrystalline domains (Figure 2b) [10]. Complementary ex situ XRD of isothermally annealed samples further elucidated the crystallization pathway (Figure S4a–d). The material remained amorphous at 200°C, began crystallizing into microcrystalline RuO₂ at 300°C, and showed progressive grain growth alongside WO₃ phase emergence (PDF#04‐007‐1276) at 400°C. Complete phase segregation into distinct RuO₂ and WO₃ occurred at 500°C. For comparison, a tungsten‐free home‐made sample (denoted home. RuO₂‐300) was synthesized under identical conditions (300°C), exhibiting similar amorphous/microcrystalline characteristics to RuWO_x‐300 (Figure S5a), in contrast to the well‐crystallized commercial RuO₂ (com. RuO₂) (Figure S5b). Systematic optimization of the Ru:W molar feeding ratio (1:0.25–1:4) identified the 1:1 composition as the most active catalyst for OER. Based on preliminary screening, RuWO_x‐300 (annealed at 300°C, Ru:W = 1:1) was selected as the optimal catalyst for detailed evaluation. Inductively coupled plasma‐mass spectrometry (ICP‐MS) confirmed its elemental composition as Ru: 10.5 at%, W: 8.2 at%, and O: 81.3 at%.

FIGURE 2.

Morphology and structure characterizations of the RuWO_x‐300 Catalyst. (a) Schematic illustration of the synthetic procedure for the catalyst. (b) In situ XRD patterns of the RuWO_x precursor treated at different temperatures. (c) TEM image (left) and corresponding elemental mappings (right) and (d) HRTEM image of RuWO_x‐300. (e) Detailed structural analysis from aberration‐corrected HRTEM image: (i) Atomic‐resolution image, (ii) intensity profile across the circled area in (i), (iii) fast Fourier transform patterns, and (iv) GPA map visualizing the in‐plane axial strain (ε _xx). (f) Schematic diagram depicting the heterogeneous strain distribution within the RuWO_x‐300 architecture, arising from the embedded crystalline domains within the amorphous matrix.

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) were employed to investigate the thermally induced microstructure evolution. The precursor exhibited a granular morphology with homogeneous elemental distribution (Figure S3b). Increasing the annealing temperature from 200°C to 500°C progressively transformed the morphology from granular to short rod‐like, eventually forming mixed sea urchin‐like and granular aggregates (Figure S4e–h). Elemental mapping confirmed uniform Ru and W distribution below 300°C, with clear phase segregation emerging at 400°C, corroborating the formation of separate RuO₂ and WO₃ phases (Figure S4g). A schematic illustration of this thermal evolution is provided in Figure S4i–l. TEM and elemental mapping further corroborate this phase evolution (Figure 2c and Figures S6–S9). High‐resolution TEM (HRTEM) of RuWO_x‐300 revealed interplanar distances of 0.261 and 0.325 nm, corresponding to the (101) and (110) planes of rutile RuO₂, interspersed within an amorphous matrix (Figure 2d and Figure S10). Additionally, the HRTEM image also showed an amorphous phase occupying the spaces between microcrystalline regions. Aberration‐corrected scanning transmission electron microscope (AC‐STEM) image showed uniform atomic‐scale contrast without isolated bright dots, confirming homogeneous elemental distribution and the absence of dopant clustering (Figure 2e‐i, ii). Fast Fourier transform (FFT) analysis identified the microcrystalline domains as RuO₂, with diffraction angles consistent with the (1$\overline{2}$1) and ($\overline{1}\overline{1}$0) planes along the [$\overline{1}$13] direction (Figure 2e‐iii). Geometric phase analysis (GPA) further verified the presence of compressive strain within the RuO₂ microcrystalline (Figure 2e‐iv), likely induced by the thermal expansion mismatch between the amorphous RuWO_x matrix and the RuO₂ domains (thermal expansion coefficient: WO₃ = 4.5 × 10⁻⁶°C⁻¹ versus RuO₂ = 6.4 × 10⁻⁶°C⁻¹). This strain modulates the electronic structure of RuO₂, facilitating electron transfer and reducing the OER activation barrier.

X‐ray photoelectron spectroscopy (XPS) was employed to probe the surface electronic states and oxidation chemistry of the catalyst. Deconvolution of the O 1s XPS spectra (Figure 3a) reveal three peaks at 530.8, 532.6, and 533.9 eV, corresponding to the lattice oxygen (O_L), adsorbed hydroxyl groups (OH_ads), and molecular water (H₂O_ads), respectively [24]. The peaks of Ru 3p_3/2 spectra (Figure 3b) at 462.2 and 464.8 eV correspond to Ru⁴⁺ and Ru³⁺, respectively [25]. Notably, RuWO_x‐300 exhibits a higher proportion of Ru³⁺ compared to home. RuO₂‐300 and com. RuO₂, suggesting that W doping effectively lowers the average Ru oxidation state. In the W 4f XPS region (Figure 3c), two spin‐orbit doublets are identified at 34.6 eV/36.8 eV and 35.5 eV/37.7 eV, corresponding to W⁵⁺ and W⁶⁺ species, respectively [26]. Complementarily, electron paramagnetic resonance (EPR) spectroscopy demonstrates a pronounced signal at g = 2.003 for RuWO_x‐300 (Figure 3d; Table S1), characteristic of oxygen vacancy (O_v), with an intensity substantially exceeding those of the references. The coexistence of W⁶⁺ and O_v establishes localized synergistic pairs that are expected to generate interfacial electric fields. To further verify this effect, Kelvin probe force microscopy (KPFM) was performed. As shown in Figure S11, RuWO_x‐300 exhibits a substantially higher surface contact potential difference (CPD, 612 mV) than com. RuO₂ (127 mV), indicating a stronger local positive surface potential. Such a positively polarized surface is expected to facilitate water accumulation at the catalyst/electrolyte interface, promote water activation, and reinforce the surface‐adsorbed oxygen layer, thereby effectively suppressing lattice oxygen evolution.

FIGURE 3.

Electronic structure and atomic coordination environment. (a) O 1s and (b) Ru 3p_3/2 XPS spectra of RuWO_x‐300, home. RuO₂‐300, and com. RuO₂. (c) W 4f XPS spectra of RuWO_x‐300. (d) EPR spectra of RuWO_x‐300, home. RuO₂‐300, and com. RuO₂ at g = 2.003. (e) Normalized Ru K‐edge and W L₃‐edge XANES spectra for RuWO_x‐300 and references, (f) Fourier‐transformed EXAFS spectra, and (g) WT images for RuWO_x‐300.

To further elucidate the electronic and coordination structures, x‐ray absorption spectroscopy (XAS) measurements were conducted. The Ru K‐edge XANES (x‐ray absorption near‐edge structure) spectrum of RuWO_x‐300 lies between those of Ru₂O₃ and RuO₂ standards (Figure 3e), reflecting an intermediate oxidation state of Ru [27]. Linear fitting of the first‐derivative edge position gives an average Ru valence of +3.68 (Figure S12b) [25], consistent with the XPS results. This oxygen vacancy‐mediated reduction of Ru valence mitigates lattice oxygen participation and stabilizes Ru centers against over‐oxidation, thereby enhancing OER durability. The W L₃‐edge XANES spectrum exhibits a positive shift relative to metallic W, corresponding to an average W oxidation state of +5.23 as determined by linear combination fitting (Figure S12d). Fourier‐transformed (FT) EXAFS (extended x‐ray absorption fine structure) spectra at the Ru K‐edge (Figure 3f) exhibit two prominent peaks at ∼1.48 and 3.22 Å, attributed to the Ru‐O and Ru‐O‐Ru/W scattering paths, respectively [20]. Similarly, the W L₃‐edge FT‐EXAFS spectrum shows contributions from W‐O (∼1.17 Å) and W‐O‐W/Ru (∼3.47 Å) coordination [26]. Wavelet transform (WT) analysis of the Ru K‐edge spectra (Figure 3g and Figure S13) clearly reveals the Ru‐O (1.48 Å) and Ru‐O‐Ru/W (2.46 Å) coordination shells [20], while the W L₃‐edge WT spectrum exhibits a pronounced maximum at R ≈ 3.20 Å and k ≈ 12 Å⁻¹, attributing to W‐O‐W/Ru contributions [26]. Quantitative EXAFS analysis (Figure S14 and Table S2) further reveals a reduced Ru‐O coordination number of 5.6 in RuWO_x‐300, compared to 6.4 in RuO₂, confirming that W incorporation induces partial Ru substitution and stabilizes oxygen vacancies. These oxygen vacancies create localized electron‐rich regions that activate adjacent Ru sites, thereby reducing the energy barrier for *OH/*OOH adsorption. Concurrently, the electrophilic W⁶⁺ centers withdraw excess electrons, effectively suppressing unfavorable Ru valence fluctuations [28]. The synergistic W⁶⁺‐O_v coupling thus optimizes intermediate adsorption energetics for improved activity, while inhibiting lattice oxygen oxidation to enhance stability, collectively breaking the activity‐stability trade‐off in acid OER.

2.3. Electrocatalytic Oxygen Evolution Performance

The acidic OER performance of catalyst was evaluated in a standard three‐electrode system with 0.5 M H₂SO₄ electrolyte. As shown in Figure 4a, RuWO_x‐300 demonstrates exceptional activity, requiring overpotentials (η) of only 179 and 231 mV to achieve current densities of 10 and 100 mA cm⁻², respectively, outperforming both home. RuO₂‐300 (η ₁₀ = 187 mV, η ₁₀₀ = 268 mV) and com. RuO₂ (η ₁₀ = 293, η ₁₀₀ = 387 mV) (Table S3). Notably, the polarization curves collected without iR correction show the same trend, with RuWO_x‐300 still exhibiting the lowest overpotential (η ₁₀ = 190 mV) among all samples (Figure S15). The corresponding Tafel slope of RuWO_x‐300 (48.9 mV dec⁻¹) is likewise lower than those of home. RuO₂‐300 (56.8 mV dec⁻¹) and com. RuO₂ (72.9 mV dec⁻¹), demonstrating accelerated OER kinetics after W incorporation (Figure 4b) [29]. This enhancement is further supported by electrochemical impedance spectroscopy (EIS), where RuWO_x‐300 shows a smaller charge‐transfer resistance relative to the reference samples (Figure S16) [30].

FIGURE 4.

Electrocatalytic OER performance and PEMWE single‐cell evaluation. (a) Polarization curves, (b) Tafel plots, (c) double‐layer capacitance C _dl, and (d) chronopotentiometry curves of samples in 0.5 M H₂SO₄ (three‐electrode system). (e) Comparison of the operational stability of RuWO_x‐300 with other reported Ru‐based acidic OER catalysts. The diagonal lines represent the total charge transferred before catalyst deactivation. (f) Schematic diagram of the single‐cell PEMWE. (g) Polarization curves of PEMWE devices employing RuWO_x‐300, home. RuO₂‐300, or com. RuO₂ as anode catalyst and commercial Pt/C as the cathode. (h) Chronopotentiometry stability tests of the PEMWE cells of RuWO_x‐300||Pt/C, home. RuO₂‐300||Pt/C, and com. RuO₂||Pt/C at 500 mA cm⁻² and 60°C.

The electrochemical active surface area (ECSA) was estimated from the double‐layer capacitance (C _dl), derived from cyclic voltammetry measurements at varying scan rates (Figure S17) [31]. As summarized in Figure 4c, RuWO_x‐300 exhibits a C_dl of 138.2 mF cm⁻², corresponding to an ECSA of 3948.6 ${\mathrm{cm}}_{\mathrm{ECSA}}^2$, substantially larger than those of home. RuO₂‐300 (117.4 mF cm⁻², 3354.3 ${\mathrm{cm}}_{\mathrm{ECSA}}^2$) and com. RuO₂ (16.7 mF cm⁻², 477.1 ${\mathrm{cm}}_{\mathrm{ECSA}}^2$). The Ru:W molar feeding ratio was also systematically varied from 1:0.25 to 1:4. Among the resulting catalysts, RuWOx‐300 (Ru:W = 1:1) exhibited the lowest overpotential, the smallest charge‐transfer resistance, and the largest ECSA, establishing it as the optimal composition for OER (Figures S18 and S19).

Notably, ECSA‐normalized LSV curves (Figure S20a) confirm that RuWO_x‐300 possesses superior intrinsic activity. Moreover, the Ru‐mass activity (MA) of RuWO_x‐300 reaches 2252.9 A ${\mathrm{g}}_{\mathrm{Ru}}^{-1}$ at 1.50 V versus RHE, which is 6.8 and 146.3 times higher than those of home. RuO₂‐300 (333.6 A ${\mathrm{g}}_{\mathrm{Ru}}^{-1}$) and commercial RuO₂ (15.4 A ${\mathrm{g}}_{\mathrm{Ru}}^{-1}$), respectively, demonstrating its exceptional intrinsic catalytic performance (Figure S20b). As shown in Figure S21, the turnover frequency (TOF) values of RuWO_x‐300 reach 0.437 s⁻¹ based on Ru sites and 0.246 s⁻¹ based on total metal sites (Ru + W), both of which are significantly higher than those of home. RuO₂‐300 (0.061 s⁻¹) and com. RuO₂ (0.002 s⁻¹). These results highlight the synergistic contribution of both enlarged active surface area and optimized electronic structure induced by W incorporation.

To investigate the influence of thermal treatment on catalytic behavior, a series of RuWO_x samples annealed at different temperatures (200°C–500°C) were systematically evaluated. The OER performance displays a strong dependence on annealing temperature. Samples treated at low temperature (200°C–250°C) retained a largely amorphous structure but suffered from poor electrical conductivity, leading to inferior performance. Conversely, annealing at high temperatures (≥400°C) induced excessive crystallization and phase segregation into RuO₂ and WO₃, which reduced active sites' accessibility and accelerated Ru dissolution, thereby degrading both activity and stability. Among all samples, RuWO_x‐300 achieves an optimal balance between amorphous and microcrystalline domains, delivering the lowest overpotential and the smallest Tafel slope (Figures S22 and S23). Consistently, it exhibits the lowest charge‐transfer resistance in EIS Nyquist plots (Figure S24) and the highest C _dl and ECSA values (Figure S25), indicating favorable interfacial charge transfer and abundant active sites. The ECSA‐normalized LSV curves further confirm that RuWO_x‐300 possesses the highest intrinsic activity in the series (Figure S26). These findings underscore that the heterophase architecture, consisting of an amorphous RuWO_x matrix with embedded microcrystalline RuO₂ domains, is essential for achieving high conductivity, high active‐site density, and structural robustness.

The operational stability of RuWO_x‐300 was evaluated employing titanium felt as the current collector. After 3000 cyclic voltammetry (CV) cycles in acidic OER test, XRD and TEM analyses confirm that the phase composition and microstructure of RuWO_x‐300 remain unchanged, preserving its mixed amorphous‐microcrystalline structure and uniform elemental distribution (Figures S27 and S28). Post‐OER XPS analysis reveals a slight decrease in the lattice oxygen (O_L) content (Figure 3a and Figure S29a), consistent with limited lattice oxygen participation without structure collapse. The Ru⁴⁺ content remains stable, while the proportion of W⁶⁺ increases from 50.88% to 67.71% (Figure 3b,c and Figure S29b,c), suggesting that W acts as an oxidant buffer, preferentially undergoing oxidation over Ru and thereby inhibiting Ru over‐oxidation and dissolution [32].

Chronopotentiometry testing at 50 mA cm⁻² demonstrates that RuWO_x‐300 operates stably for over 4820 h (200 days), whereas both home. RuO₂‐300 and commercial RuO₂ degrade rapidly (Figure 4d). Figure 4e compares the stability of RuWO_x‐300 with other state‐of‐the‐art Ru‐based acidic OER catalysts (Table S4), where the diagonal lines in Figure 4e represent the total charge transferred before deactivation. While most reported catalysts achieve OER stability in the range of 10^2–10⁴ C cm⁻², RuWO_x‐300 exceeds 10⁵ C cm⁻². ICP‐MS analysis further confirms significantly suppressed Ru dissolution in RuWO_x‐300 (< 20 ppb) compared to home. RuO₂‐300 (>120 ppb) and com. RuO₂ (>200 ppb) (Figure S30). In addition, RuWO_x‐300 maintains stable operation for over 1000 h at 50 mA cm⁻² on a 1 cm² electrode (Figure S31), further demonstrating its outstanding durability. These results collectively suggest that W incorporation not only preserves high catalytic activity but also substantially enhances durability, effectively breaking the activity‐stability trade‐off in acidic OER.

Building on the exceptional OER activity and stability of RuWO_x‐300, we further assembled a membrane electrode assembly (MEA) employing Nafion 115 as the proton exchange membrane (Figure S32), with RuWO_x‐300 serving as the anode catalyst and commercial Pt/C (40 wt.%) as the cathode. The full‐cell configuration was systematically evaluated under operational conditions (60°C, deionized water), emulating industrially relevant environments to assess practical viability (Figure 4f). As shown in Figure 4g, the PEMWE cell with RuWO_x‐300 as the anode (RuWO_x‐300||Pt/C) achieves an industry‐relevant current density of 1.0 A cm⁻² at an ultralow cell voltage of 1.67 V (without iR compensation). This represents a reduction of 90 and 180 mV compared to cells using home. RuO₂‐300 (1.76 V) and com. RuO₂ (1.85 V) anodes, respectively. Beyond high activity, operational stability under industrial applications is crucial for practical deployment. Impressively, the RuWO_x‐300‐based cell exhibits exceptional durability, operating continuously for 200 h at 500 mA cm⁻² with negligible voltage decay (Figure 4h). The initial voltage increase is attributed to the break‐in process of the membrane‐electrode assembly, during which membrane hydration, catalyst‐layer wetting, and interfacial equilibration gradually reach steady‐state conditions [33, 34]. This performance corresponds to a 50‐fold stability improvement over both home. RuO₂‐300 and com. RuO₂ references, which suffered rapid degradation within 4 h under identical operating conditions. The results above demonstrate the great potential of RuWO_x‐300 as a highly efficient and stable anode catalyst for industrial PEMWE.

2.4. Mechanism Insights Into the Enhanced OER Kinetics

To elucidate the origin of the enhanced OER kinetics, a series of electrochemical measurements probing proton transfer behavior were performed. As shown in Figure 5a, the CV profiles of home. RuO₂‐300 and com. RuO₂ display two distinct redox couples at ∼0.65 and ∼1.24 V versus RHE, corresponding to the Ru^III/Ru^IV and Ru^IV/Ru^VI transitions, respectively [23, 25]. In contrast, RuWO_x‐300 exhibits a broad capacitive plateau in the low‐potential region (0–0.4 V versus RHE), while the higher‐potential redox features are significantly suppressed [20]. This low‐potential feature resembles the hydrogen desorption behavior typically observed on WO₃ or Pt electrodes [35, 36], indicating that surface deprotonation on RuWO_x‐300 occurs more readily than on the RuO₂ counterparts.

FIGURE 5.

Probing deprotonation kinetics and interfacial water structure. (a) CV curves of RuWO_x‐300, home. RuO₂‐300, and com. RuO₂. in 0.5 M H₂SO₄. (b) Logarithmic plots of OER current density (at 1.50 V vs. RHE) as a function of electrolyte pH. (c) Logarithmic plots of MOR current density versus methanol concentration on different catalysts. (d) In situ Raman spectra of RuWO_x‐300 acquired from the OCP to 1.8 V versus RHE. (e) Potential‐dependent population distribution of different interfacial water structures derived from spectral deconvolution. (f) Potential‐dependent frequency shifts of the Raman bands corresponding to different interfacial water. (g) Schematic models comparing the electrochemical interface of com. RuO₂ (left) and RuWO_x‐300 (right) during the OER.

To further verify this facilitated deprotonation, pH‐dependent OER measurements were performed in acidic electrolytes (Figure S33). As illustrated in Figure 5b, RuWO_x‐300 demonstrates the most pronounced pH dependence among the investigated catalysts, with a reaction order (ρ) of ‐0.69, which is significantly more negative than those of home. RuO₂‐300 (ρ = ‐0.41) and com. RuO₂ (ρ = ‐0.25). This result confirms a higher propensity for proton dissociation at the RuWO_x‐300 surface [8, 37]. The enhanced pH sensitivity is attributed to the stronger Brønsted acidity of the Ru‐O_bri‐W bridging oxygen motif, where the proton dissociation constant (pKa) of O_bri is comparable to the electrolyte pH, thereby lowering the deprotonation barrier. In contrast, the O_bri sites in home. RuO₂‐300 and com. RuO₂ are proton‐saturated (pKa ≫ pH), leading to weak pH sensitivity. Solid‐state ¹H nuclear magnetic resonant (¹H‐NMR) spectroscopy further corroborates the Brønsted acidity of W‐OH_bri‐Ru linkages, showing a split peak indicative of distinct OH_bri environments (Figure S34) [38, 39].

The influence of surface deprotonation was further investigated by using the methanol oxidation reaction (MOR) as a probe. The com. RuO₂ exhibits the highest MOR activity, followed by home. RuO₂‐300, whereas RuWO_x‐300 is nearly inert (Figure 5c and Figure S35). Given that MOR proceeds via nucleophilic attack of methanol on electrophilic *OH species, the suppressed activity of RuWO_x‐300 indicates a much lower surface *OH coverage. This finding demonstrates that W incorporation reduces *OH accumulation on Ru sites, thereby facilitating its faster deprotonation to form subsequent *O/*OOH intermediates and ultimately enhancing the OER kinetics [8, 20]. Moreover, as shown in Figure S36, RuWO_x‐300 exhibits a substantially larger kinetic isotope effect (KIE) than com. RuO₂ over the investigated potential range, reaching a maximum value of 9.09 compared with 7.17 for com. RuO₂. The stronger isotope effect indicates a more pronounced involvement of proton‐transfer processes in the kinetically relevant step of RuWO_x‐300.

In situ Raman spectroscopy was employed to investigate the potential‐dependent evolution of the interfacial water structure within the EDL (Figure 5d; Figures S37a and S38a). The broad O–H stretching band was deconvoluted into three components, corresponding to 4‐HB‐H₂O, 2‐HB‐H₂O, and weakly hydrogen‐bonded free‐H₂O [8]. Quantitative analysis (Figure 5e; Figures S37b and S38b) reveals that RuWO_x‐300 possesses a substantially higher proportion of free H₂O compared to the reference catalysts. This unique hydration structure arises from the incorporation of high‐valence W⁶⁺ and the resulted oxygen vacancies, which collectively generate strong local electrostatic fields that disrupt the interfacial hydrogen‐bond network. The fields promote water adsorption while simultaneously increasing the interfacial free‐H₂O population, thereby ensuring a sustained reactant supply for efficient OER.

The vibrational Stark effect further supports this mechanism [40]. RuWO_x‐300 exhibits a markedly enhanced sensitivity of interfacial free‐H₂O band to an applied potential (Stark slope = 47.45 cm⁻¹ V⁻¹), whereas home. RuO₂‐300 (54.09 cm⁻¹ V⁻¹) and com. RuO₂ (‐61.03 cm⁻¹ V⁻¹) exhibit characteristic responses dominated by 4‐HB‐H₂O (Figure 5f; Figures S37c and S38c). This difference provides direct quantitative evidence for the weakening of the hydrogen‐bond network in the W‐doped system, which facilitates water reorientation and dissociation under the applied electric field. The resulting enhancement in interfacial water dynamics promotes efficient water activation while suppressing detrimental lattice oxygen participation during prolonged OER. A schematic comparison of the electrochemical interfaces during water oxidation is presented in Figure 5g. In contrast to conventional RuO₂, the localized electrostatic field in RuWO_x‐300 enables dynamic replenishment of free‐H₂O at the interface. This synergistic mechanism effect ensures a sustained reactant supply, simultaneously enhancing structural durability and accelerating acidic OER kinetics.

2.5. OER Pathway Analysis

To gain deeper insight into the reaction kinetics and the evolution of oxygen‐containing intermediates, we conducted in situ electrochemical impedance spectroscopy (EIS) under varying applied potentials (Figure S39). In the Bode phase plots, RuWO_x‐300 consistently exhibits smaller phase peak angles than home. RuO₂‐300 and com. RuO₂, indicating reduced interfacial resistance and enhanced electron transfer efficiency for OER (Figure S39a–c) [30]. Notably, the maximum phase angle reduction for RuWO_x‐300 occurs at the lowest potential, reflecting the accelerated reaction kinetics (Figure S39d–f). With increasing potential, the phase peak of RuWO_x‐300 rapidly shifts toward higher frequencies, signifying facilitated adsorption‐desorption of OER intermediates and lower Faradaic resistance (Figure S39a–c) [41]. These results collectively confirm that RuWO_x‐300 promotes the interfacial charge transfer more efficiently, thereby accelerating acidic OER kinetics.

To elucidate the OER pathway, ¹⁸O isotope‐labeling differential electrochemical mass spectrometry (DEMS) was employed (Figure 6a). For RuWO_x‐300, the DEMS data exhibited an intense ³²O₂ signal and a very weak ³⁴O₂ signal, with no detectable ³⁶O₂. The minimal ³⁴O₂ signal is attributable to the natural isotopic abundance of ¹⁸O in the electrolyte, confirming that the O–O coupling does not involve two lattice oxygen atoms (which would yield ³⁶O₂) nor the combination of a lattice oxygen with an adsorbate (which would yield significant ³⁴O₂). This isotopic signature effectively excludes the participation of the lattice oxygen mechanism (LOM), and the suppression of the LOM pathway is a key factor underpinning the high stability.

FIGURE 6.

Mechanistic analysis of the OER Pathway. (a) DEMS signals of different O₂ products for ¹⁸O‐labeled RuWO_x‐300 in 0.5 M H₂SO₄ using H₂ ¹⁶O as electrolyte. (b) In situ Raman spectra of RuWO_x‐300 recorded from the OCP to 1.8 V vs. RHE. PDOS plots for (c) com. RuO₂ and (d) RuWO_x‐300. Gibbs free energy diagrams (ΔG) for the (e) AEM pathway and (f) BOAD pathway on com. RuO₂ and RuWO_x‐300. (g) Schematic illustration of the AEM and BOAD pathway on RuWO_x‐300.

In situ Raman spectroscopy was further employed to monitor the evolution of oxo‐intermediates during OER (Figure 6b and Figure S40). Characteristic vibrational modes of rutile RuO₂ were observed at 518.0 cm⁻¹ (E_g), 614.9 cm⁻¹(A_1g), and 715.2 cm⁻¹ (B_2g) [23, 42]. The weaker peak intensities for RuWO_x‐300 and home. RuO₂‐300 compared to com. RuO₂ indicate the formation of an amorphous‐crystalline heterostructures with reduced long‐range order, consistent with the incorporation of W and controlled crystallization. A band at 984.96 cm⁻¹ is attributed to ${\mathrm{SO}}_4^{2-}$ species adsorbed from the H₂SO₄ electrolyte [43], while bands at 952.3 and 1078.6 cm⁻¹ correspond to adsorbed *OOH and *OO intermediates [42, 44], respectively (Figure 6b and Figures S41–S43). Remarkably, these intermediates emerge on RuWO_x‐300 at ∼ 1.4 V versus RHE, which is 100 mV lower than the onset potential for home. RuO₂‐300 (1.5 V vs. RHE) and 300 mV below that of commercial RuO₂ (≥1.7 V vs. RHE). This earlier appearance confirms that W incorporation accelerates the formation of key oxygen intermediates.

In situ Fourier‐transform infrared (FT‐IR) spectra were further employed to probe the evolution of surface hydroxyl species and oxygenated intermediates during OER (Figure S44). In the O–H stretching region, RuWO_x‐300 exhibits a hydroxyl‐related band at a higher wavenumber and with a more reversible intensity change than com. RuO₂, indicating a distinct proton‐binding environment and more dynamic proton exchange behavior [45]. More importantly, the bands assigned to hydrogen‐bonded *OOH species with adjacent sites (∼1062 cm⁻¹) [22, 30], *OOH (∼1227 cm⁻¹) [46], and *O–O (∼1126 cm⁻¹) [47] on RuWO_x‐300 increase upon anodic polarization and decrease reversibly when the potential is stepped back, whereas the corresponding oxygenated intermediates on com. RuO₂ remains largely accumulated after potential reversal. The fully reversible evolution of these oxygenated species suggests that proton‐coupled intermediate formation and consumption proceed more dynamically on RuWOx‐300. In particular, the potential‐dependent and reversible evolution of the hydrogen‐bonded *OOH band (∼1062 cm⁻¹) suggests the involvement of adjacent bridging oxygen sites in proton‐coupled transformations of oxygenated intermediates during OER [22, 30].

Our mechanistic investigations reveal fundamental differences in the OER pathways between conventional and W‐modified RuO₂ catalysts. The co‐existence of *OOH and *OO intermediates on both home. RuO₂‐300 and com. RuO₂ suggests a mixed reaction pathway involving both the adsorbate evolution mechanism (AEM) and the lattice oxygen mechanism (LOM). In contrast, the Brønsted acidic W‐O_bri‐Ru sites in RuWO_x‐300 facilitate surface deprotonation, as evidenced by the pH‐dependent measurements, ¹H‐NMR, KIE, DEMS, and in situ FT‐IR spectra, enabling a shift toward a bridging‐oxygen‐assisted deprotonation (BOAD) pathway. In this mechanism, the W‐O_bri‐Ru motifs critically assist the deprotonation of oxygen‐containing intermediates, circumventing the linear scaling relations typical of single‐site Ru catalysts and thereby enhancing both OER kinetics and stability.

Density functional theory (DFT) calculations provided atomic‐level insights into the OER mechanism. Projected density of states (PDOS) analysis shows that in com. RuO₂, the Ru d‐band center (ɛ_d) and O p‐band center (ɛ_p) lie at ‐1.179 and ‐2.878 eV, respectively. In RuWO_x‐300, the Ru d‐band shifts upward to ‐0.931 eV, enhancing orbital hybridization with oxygen intermediates and facilitating O–O coupling [48]. Meanwhile, the O p‐band shifts downward to ‐2.954 eV, which reduces the redox activity of lattice oxygen and suppresses its participation, improving structural stability in OER [49]. Furthermore, the widened gap between the Ru d‐band and O p‐band centers also indicates tuned Ru‐O covalency, which optimizes the adsorption‐desorption balance of intermediates during OER [48].

Gibbs free energy diagrams were computed for the AEM, BOAD, and LOM pathways (Figures S45–S50). For com. RuO₂, both AEM and BOAD share the same rate‐determining step (RDS, *O → *OOH) with a theoretical overpotential of 0.68 V (Figure 6e,f), while LOM exhibits an RDS of *O → *OO with η = 0.69 V (Figure S51), suggesting a preference for the AEM route. In contrast, RuWO_x‐300 shows significantly lower theoretical overpotential: along the AEM pathway, the RDS shifts to *OOH → O₂ with η = 0.50 V (Figure 6e). More importantly, in the BOAD pathway, the Brønsted acidic W‐O_bri‐Ru sites facilitates facile deprotonation of *OOH, making *OOH → *OO + *H the RDS with an exceptionally low η = 0.36 V (Figure 6f). For LOM, the RDS is *OO → O₂ with η = 0.41 V. These results confirm that RuWO_x‐300 preferentially follows the BOAD pathway, where enhanced deprotonation at W‐O_bri‐Ru sites leads to superior OER activity and stability.

3. Conclusion

In summary, we have demonstrated a rational interfacial microenvironment engineering strategy to simultaneously address the longstanding challenges of limited interfacial water supply and sluggish proton transfer in acidic OER. The tungsten‐stabilized RuWO_x‐300 catalyst breaks the conventional activity‐stability trade‐off through synergistic electronic and structural modulation. Comprehensive electrochemical, in situ/ex‐situ spectroscopic, and theoretical analyses reveal a dual‐function mechanism: high‐valent W⁶⁺ species induce a localized electrostatic field that enriches free water at the interface, while Brønsted acidic W‐O_bri‐Ru sites, acting as efficient interfacial proton donors, facilitate efficient proton transfer via a BOAD pathway. These cooperative effects enhance the OER kinetics while efficiently suppressing Ru dissolution under harsh acidic operation. The resulting RuWO_x‐300 catalyst exhibits outstanding performance, with ultralow overpotentials (179 mV@10 mA cm⁻², 231 mV@100 mA cm⁻²) and exceptional long‐term stability (> 4820 h at 50 mA cm⁻²), surpassing the most reported Ru‐based acidic OER electrocatalysts. This work pioneers an atomic‐level microenvironment engineering approach for designing highly efficient and stable Ru‐based electrocatalysts, offering a viable pathway to overcome the activity‐stability dilemma in acidic water oxidation.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adma74093‐sup‐0001‐SuppMat.pdf.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.