Rapid Joule-Heating Synthesis of Efficient Low-Crystallinity Ru-Mo Oxide Catalysts for Alkaline Hydrogen Evolution Reaction

Abstract

Electrocatalytic water splitting has been demonstrated to be a highly efficient and promising technology for green hydrogen production. However, the inefficiency and instability of the cathode hinder its wide application in water electrolysis. Herein, we report a rapid Joule heating method for synthesizing the Ru-Mo oxide catalyst. Comprehensive characterization results confirmed that the as-prepared catalyst featured an internal porous structure with low crystallinity, which weakened the strength of Ru-H bonds through structural and electronic modulation. The enhanced HER performance was attributed to the incorporation of Mo⁴⁺ species, which strengthened Ru-O-Mo interactions. As tested, the optimized catalyst exhibited ultralow overpotentials (25.08 mV and 120.52 mV @ 10 and 100 mA cm⁻², respectively) and excellent stability (100 h @ 100 mA cm⁻²) in a 1 M KOH solution. Meanwhile, the as-prepared catalyst was equipped in an anion exchange membrane (AEM) alkaline water electrolyzer, which could deliver 185 mA cm⁻² at only 2.16 V with 100% Faradaic efficiency. This study provides a feasible strategy for constructing highly efficient low-crystallinity electrocatalysts.

1. Introduction

Hydrogen (H₂), characterized by zero emissions and high energy density (140.4 MJ/kg), has emerged as a sustainable alternative to fossil fuels and an efficient energy carrier [1]. However, hydrogen production primarily relies on fossil fuel reforming and industrial by-products, which result in substantial energy consumption and environmental degradation due to carbon emissions. In contrast, electrocatalytic water splitting is regarded as a promising pathway for industrial-scale green hydrogen generation, featuring both pure H₂ production and environmental friendliness [2]. Nevertheless, the high energy consumption (4.2 kWh Nm⁻³ H₂) restricts its widespread applications [3]. Therefore, the development of efficient and robust electrocatalysts is essential for energy conservation and advancing large-scale utilization. Platinum-based catalysts (e.g., commercial Pt/C) exhibit exceptional hydrogen evolution reaction (HER) activity due to nearly optimal hydrogen adsorption energy (ΔG_*H ≈ 0 eV). However, the low reserves and high cost of platinum severely impede its large-scale application in water electrolysis [4]. As a result, it is critical to develop cost-effective electrocatalysts that deliver high efficiency and stability, thereby advancing electrocatalytic water splitting toward an industrially viable process.

Among various alternatives, ruthenium (Ru)-based materials, such as RuO₂ and unsaturated RuO_x clusters, have received much attention in water electrolysis due to their higher reserves and lower cost (approximately half that of Pt) [5,6,7,8]. Specifically, RuO₂ facilitates the adsorption and dissociation of water molecules, which makes it an ideal catalyst for the water dissociation reaction under alkaline conditions [9,10]. However, the high adsorption of hydrogen intermediates on Ru (ΔG_*H ≪ 0 eV) results in inferior HER performance [11]. Introducing suitable secondary metals (e.g., Mo, Ni, and Co) into Ru-based materials has proven to be an effective strategy to overcome the dynamic constraint [12]. In detail, the interaction among these elements and Ru-based materials has been reported to optimize the electronic structure by lowering the d-band center, thereby weakening Ru-H interactions and tuning the hydrogen adsorption free energy (ΔG_*H) closer to the ideal value (0 eV) [13,14]. Accordingly, molybdenum (Mo) has been comprehensively explored due to its abundant reserves and tunable electronic structures [15,16] and exhibits potential as a rational secondary metal. Recent studies have demonstrated that the synergistic effect between Ru and Mo species can accelerate the efficiency of water splitting, in accordance with the Brewer–Engel theory [17]. For instance, Fan et al. synthesized Ru-, RuO₂-, and MoO₃-embedded carbon nanorods (Ru-RuO₂/MoO₃ CNRs-350) [18], wherein the strong interaction between Ru and Mo species enabled the catalyst to exhibit ultralow overpotential (9.2 and 65.4 mV, respectively) at a current density of 10 and 100 mA cm⁻² in 1.0 M KOH, which is comparable to that of Pt/C (20 wt%).

Despite the advances described above, it is difficult to overcome the inherent activity limitations of Ru-based materials. It is critical to develop effective strategies, such as phase modulation [19], morphology engineering [20], heteroatom doping [21], heterostructure construction [22], etc. Recently, low-crystallinity materials have gained attention as another promising approach for enhancing intrinsic activity [23]. Specifically, the low crystallinity not only provides abundant defects and active sites to facilitate the adsorption of intermediates but also enhances the conductivity of active materials and accelerates electron transfer rates [24,25]. Furthermore, the relatively disordered atomic arrangement of low-crystallinity materials can realize rapid charge transfer, thereby creating an ideal electronic structure for active sites [8,26]. Guo et al. created a rough layer with extremely low crystallinity on the surface of Co-Ni-Mo-O nanosheets using an electrochemical activation strategy [27], which increased the active sites and improved internal conductivity compared to fully polycrystalline or amorphous counterparts. Such optimized Co-Ni-Mo-O/NF-ER catalysts exhibited exceptional HER activity (42 mV @ 10 mA cm⁻² and 400 h @ 250 mA cm⁻²). However, the conventional synthesis method (e.g., chemical vapor deposition) can destroy the low-crystalline matrix during long-term operation with high temperatures (>700 K) [28]. Therefore, the controllable synthesis of low-crystallinity materials remains a bottleneck for the development of scalable applications.

Based on the considerations described above, we report a rapid Joule heating method for the synthesis of low-crystallinity Ru-Mo oxide catalysts (Ru₁Mo₁-JH). Compared to the catalyst synthesized via conventional tube furnace heating methods (Ru₁Mo₁-LT), Ru₁Mo₁-JH features significantly reduced crystallinity and increased Mo⁴⁺ content, which strengthens metal-oxygen bonds and weakens Ru-H interactions. Such synergistic electronic modulation endows the catalyst with exceptional HER performance. The catalyst Ru₁Mo₁-JH exhibits ultralow overpotentials (25.08 mV and 120.52 mV at current densities of 10 and 100 mA cm⁻², respectively) and excellent stability (100 h @ 100 mA cm⁻²) in 1 M KOH, which surpassed commercial Pt/C (42.27 mV and 149.27 mV) under identical conditions. Furthermore, an anion exchange membrane electrolyzer (AEM) assembled with an Ru₁Mo₁-JH electrode achieves 185 mA cm⁻² at only 2.16 V with 100% Faradaic efficiency. These results highlight the potential of Ru₁Mo₁-JH as a cost-effective alternative to Pt/C for industrial applications.

2. Experimental

2.1. Synthesis of Ru₁Mo₁-JH Catalyst

This work successfully synthesized low-crystallinity Ru-Mo oxide catalysts through a two-step process that combines hydrothermal treatment and rapid Joule heating. To optimize the strength of the Ru-H bond by tailoring the electronic structure, we first investigated the impact of Mo species on the water splitting performance of Ru-based catalysts. Homogeneous solutions of the precursors, including RuCl₃ (Aladdin Ltd., Shanghai, China) and MoCl₅ (Aladdin Ltd., Shanghai, China) with different Ru/Mo ratios of 0:4, 1:3, 2:2, 3:1, and 4:0, were transferred to a reactor, in which hydrothermal treatment took place at 160 °C for 5 h. Subsequently, these catalysts were heated to 470 °C for 60 s using a rapid Joule heating device, denoted as Mo-JH, Ru₁Mo₃-JH, Ru₁Mo₁-JH, Ru₃Mo₁-JH, and Ru-JH, respectively. To further elucidate the role of low crystallinity, a parallel catalyst (Ru₁Mo₁-LT) was synthesized using the homogeneous solution with the RuCl₃/MoCl₅ ratio (1:1) and treated using a conventional thermal method at the same temperature (470 °C) for 1 h to ensure full crystallization. The following text will focus on the performance comparison of these catalysts and explore the improvement mechanism of the Ru₁Mo₁-JH.

2.2. Characterizations

X-ray diffraction (XRD) was employed on the Bruker D8 focus (Bruker AXS GmbH, Berlin, Germany) using Cu Kα radiation with a scan rate of 10°min⁻¹. Transmission electron microscopy (TEM) was performed on an FEI Tecnai F20 microscope (FEI, Hillsboro, Oregon, OR, USA) at an acceleration voltage of 200 kV, yielding high-resolution transmission electron microscopy (HRTEM) and energy-dispersive X-ray spectroscopy (EDX) data via a JEOL JEM 2100 microscope (JEOL, Tokyo, Japan) at the same voltage. Scanning electron microscopy (SEM) images were obtained using a Hitachi S-4800 microscope (Hitachi, Nagano, Japan) at an acceleration voltage of 5–10 kV. The X-ray photoelectron spectroscopy (XPS) spectra were obtained using an ESCALAB 250Xi (Thermo Fisher Scientific, Waltham, MA, USA) with a monochromatic Al Kα source, and the spectra were analyzed using Avantage 6.9 software. Brunauer–Emmett–Teller (BET) measurements were performed on an ASAP2460 (Micromeritics Instrument Corporation, Norcross, Georgia, GA, USA) using nitrogen as the adsorption gas, which was degassed at 120 degrees Celsius for 8 h. Fourier-transform infrared spectroscopy (FTIR) analysis was employed on a Nicolet iS 10 (Thermo Fisher Scientific, Waltham, MA, USA) using a silicon crystal to obtain the absorption infrared spectrum of the samples.

2.3. Electrochemical Measurements

The electrochemical HER of catalysts was performed on a CHI 660E workstation (CH Instruments, Shanghai, China) using a typical three-electrode system in 1 M KOH (Sinopharm Chemical Reagent Co. Ltd., Shanghai, China) electrolyte. A graphite rod, Hg/HgO, and a catalyst-loaded glassy carbon electrode (GCE) (Shanghai Jingchong Electronic Technology Development Co. Ltd., Shanghai, China) served as the counter, reference, and working electrodes, respectively. The catalyst ink consisted of 5 mg of catalyst, 245 μL of ethanol (Sinopharm Chemical Reagent Co. Ltd., Shanghai, China), 245 μL of water, and 10 μL of 5 wt% Nafion solution (Shanghai Jingchong Electronic Technology Development Co. Ltd., Shanghai, China). Afterward, the GCE was dropped into 24 μL of as-prepared ink and dried before undergoing electrochemical tests. Linear sweep voltammetry (LSV) measurements were performed at a scan rate of 5 mV s⁻¹ with 95% iR-compensation. The Nernst equation (E_RHE = E_Hg/HgO + 0.0592pH + 0.098 V) was referenced for the conversion of the applied potentials. The Tafel slope was obtained via linear fit to LSV curves using the Tafel equation (η = a + b log(j)). Cyclic voltammetry (CV) tests with different scan rates (ranging from 20 to 100 mV s⁻¹) were conducted to calculate the double-layer capacitance (C_dl) values of the catalysts. Electrochemical impedance spectroscopy (EIS) was performed at frequencies ranging from 10⁵ Hz to 10⁻² Hz. The Faradaic efficiency of Ru₁Mo₁-JH was calculated by measuring the amount of evolved H₂ through a water drainage method at 185 mA cm⁻², expressed as follows:

Faradaic efficiency = N_G/N_T

(1)

N_G = V/V_m

(2)

N_T = Q/(2 × F)

(3)

where N_G and N_T are the practical and theoretical number of hydrogen molecules, respectively; V is the measured volume of hydrogen; V_m is the molar volume of gas; and Q indicates the number of electrons transferred to generate hydrogen.

3. Results and Discussion

3.1. Structural Characterization of Ru₁Mo₁-JH

The Ru₁Mo₁ precursor was synthesized using the hydrothermal method to form powdered nanoparticles. As shown in its SEM image (Figure 1a), the nanoparticles exhibited a 3D-connected structure. After rapid Joule heating treatment at 470 °C for 60 s, the Ru₁Mo₁-JH catalyst exhibited significantly reduced agglomeration and uniform dispersion of nanoparticles (Figure 1b), facilitating the exposure of abundant active sites and thus enhancing catalytic activity, which benefited from the transient heating and rapid quenching characteristics of Joule heating technology [29]. Joule heating is capable of reaching high temperatures up to 3000 K within hundreds of milliseconds and cooling below 323 K within seconds. Based on rapid energy input and instantaneous quenching, ultrafast Joule heating technology can effectively circumvent the aggregation of nanoparticles and maintain the stability of catalytic particle dispersion [30]. In contrast, Ru₁Mo₁ was heated for a longer time to obtain Ru₁Mo₁-LT, which induced sintering, resulting in irregular blocks with smooth surfaces (Figure 1c). Such morphology reduced the number of active sites and diminished the catalytic performance. The TEM image (Figure 1d) revealed that the morphology of Ru₁Mo₁-JH nanoparticles was maintained after rapid Joule heating. The HRTEM images (Figure 1e,f) revealed that Ru₁Mo₁-JH had a large area of amorphous parts with interconnected channels and voids between nanoparticles, thus exposing more active sites. Such a structure was also conducive to electrolyte ion transport, intermediate activation, and the escape of gases produced by the reaction [31]. Region I in Figure 1f was selected for fast Fourier transform (FFT), which indicated that Ru₁Mo₁-JH was composed of randomly oriented nanoparticles (Figure 1g). Furthermore, the elemental mapping of Ru₁Mo₁-JH showed that the Mo, Ru, and O elements were uniformly distributed over the entire structure of Ru₁Mo₁-JH (Figure 1h). However, the TEM of Ru₁Mo₁-LT showed larger particle dimensions and increased agglomeration (Figure 1i), which limited the exposure of active sites. As shown in Figure 1j, Ru₁Mo₁-LT was distributed with obvious lattice fringes. It was initially determined that the crystallinity of the catalyst was relatively lower after rapid Joule heating, while a crystal material was formed after conventional thermal treatment. It should be noted that low-crystallinity materials have a high specific surface area, which can improve the efficiency of catalysts by providing fully exposed active sites. This finding was also confirmed via SEM in Figure 1b. According to the line intensity distribution map of area III in Figure 1k, the spacing of the fringes was 0.34 nm, which confirmed the complete crystallization of Ru₁Mo₁-LT. These results further provided strong evidence for the low crystallinity of Ru₁Mo₁-JH and the complete crystallization of Ru₁Mo₁-LT.

The phase composition and crystal structure of the samples were further analyzed via XRD. As shown in Figure 2a, XRD of as-prepared catalysts exhibited phase structures of RuO₂ (PDF#21-1172), MoO₂ (PDF#50-0739), and MoO₃ (PDF#85-2405). The gradual enhancement of XRD diffraction peak intensities with prolonged heating time indicated an increase in crystallinity. The half-peak full widths (FWHM) of the strongest diffraction peaks for Ru₁Mo₁, Ru₁Mo₁-JH, and Ru₁Mo₁-LT were calculated as 0.60 Å⁻¹, 0.40 Å⁻¹, and 0.24 Å⁻¹, respectively. According to previous studies, an FWHM range of 0.3–0.5 Å⁻¹ was associated with amorphous structures, which confirmed the low crystallinity of Ru₁Mo₁-JH [32,33,34]. Quantitative phase analysis (Figure 2b) revealed that Joule heating in air resulted in Ru and Mo species existing in the form of oxides. XPS was employed to further probe the elemental composition and chemical state of Ru₁Mo₁-JH and Ru₁Mo₁-LT. The XPS full-spectrum scanning results (Figure 2c) confirmed the existence of Ru, Mo, and O in both samples. In the Mo 3d orbital spectra (Figure 2d), there was no significant difference in the binding energy of Mo⁶⁺ between the two catalysts. It should be noted that the peak of Mo⁴⁺ in Ru₁Mo₁-JH exhibited a 0.34 eV shift toward higher binding energy compared to Ru₁Mo₁-LT. Moreover, the peak integral area ratio of Mo⁶⁺/Mo⁴⁺ increased from 1.5 to 3.3, suggesting the progressive oxidation of Mo⁴⁺ to Mo⁶⁺. The Ru 3p spectra (Figure 2e) revealed a 0.20 eV positive shift in the Ru 3p_1/2 peak for Ru₁Mo₁-JH, accompanied by a 3.95% increase in the Ru 3p_3/2 peak area. According to the results described above, it could be concluded that MoO₂, as a superior electron-accepting support, formed strong Ru-MoO₂ interactions in Ru₁Mo₁-JH, allowing for significant electron transfer from Ru to MoO₂ and eventually resulting in the redistribution of electrons among Ru sites [35]. These observations suggest that Mo⁴⁺ incorporation strengthened Ru-O-Mo interactions and reduced the d-electron density of Ru, which could lower the d-band center, thereby optimizing the electronic structure [36]. For the O 1s in Figure 2f, Ru₁Mo₁-JH exhibited an approximately 0.45 eV positive shift compared to Ru₁Mo₁-LT, with a higher proportion of O²⁻ in the metal oxide, which proved that enhanced metal–oxygen bonding occurred. Notably, the contents of OH⁻ and H₂O on the surface of Ru₁Mo₁-LT were 54.11%, whereas the OH⁻ and H₂O on the surface of Ru₁Mo₁-JH accounted for only 23.89%. According to previous studies, low crystallinity formed by rapid heating could optimize the electronic structure of Ru/Mo-based catalysts and reduce the d-electron density of Ru, weakening the adsorption strength of Ru for H and ultimately affecting HER performance [37]. As shown in Figure 2g, a band at 1958 cm⁻¹ on the spectra of Ru₁Mo₁-LT was observed and attributed to Ru-H species [38,39], while a band at 1956 cm⁻¹ on the spectra of Ru₁Mo₁-JH was much weaker, which further confirmed that low crystallinity contributed to weakening Ru-H interactions. In addition, the BET surface area of Ru₁Mo₁-JH was found to be 69.01 m²/g, which exceeded that of other reported catalysts, indicating that Ru₁Mo₁-JH was characterized by abundant active sites.

To investigate the morphological and electronic structures of Ru₁Mo₁-JH after water splitting, XRD and XPS were carried out. The XRD patterns showed that the characteristic peak intensities of Ru₁Mo₁-JH had slightly decreased after the reaction, but the phase structures of RuO₂, MoO₂, and MoO₃ still existed (Figure 3a). Typically, in order to collect Ru₁Mo₁-JH after the reaction more effectively, the powder of Ru₁Mo₁-JH was dripped onto nickel foam (NF), and then an LSV scan was carried out before XRD was conducted. The three dominant peaks originated from the NF substrate. Moreover, XPS was employed to investigate the elemental composition and electronic structure of the material after the reaction. The XPS full-spectrum scanning results confirmed the existence of Ru, Mo, and O in Ru₁Mo₁-JH both before and after the reaction (Figure 3b). As shown in Figure 3c, the binding energies of the Mo⁴⁺ and Mo⁶⁺ peaks in Ru₁Mo₁-JH revealed positive shifts of 0.17 eV and 0.24 eV, respectively. The Ru 3p spectra revealed negligible binding energy shifts for Ru 3p_1/2 and Ru 3p_3/2 peaks in Ru₁Mo₁-JH after the reaction (Figure 3d). The elements Ru and Mo in Ru₁Mo₁-JH existed in the same valence state as Mo⁴⁺, Mo⁶⁺, and Ru⁴⁺, both before and after the reaction. However, the O 1s spectrum exhibited relatively significant changes after the reaction. Ru₁Mo₁-JH after the reaction showed a positive shift of approximately 0.40 eV compared to before the reaction (Figure 3e), indicating enhanced metal–oxygen bonding. Additionally, the content of OH⁻ and H₂O on the surface of Ru₁Mo₁-JH increased, which might be attributed to the influence of alkaline solutions during the water splitting process. These results confirmed the excellent structural stability of Ru₁Mo₁-JH.

3.2. HER Electrochemical Performance of Ru₁Mo₁-JH

A conventional three-electrode cell configuration was used to comprehensively evaluate the electrocatalytic activity of Ru₁Mo₁-JH in a 1 M KOH solution at 25 °C, with other as-prepared catalysts measured for comparison. To confirm the optimum content of Mo, HER activities of catalysts synthesized with homogeneous solutions of different RuCl₃/MoCl₅ ratios were explored. As shown in Figure 4a, the overpotential of Ru₁Mo₁-JH at 10 mA cm⁻² was only 25.08 mV, which was much lower than that of the contrasted samples. Typically, the Tafel slope of Ru₁Mo₁-JH (48.89 mV dec⁻¹) surpassed that of Mo-JH (99.15 mV dec⁻¹), Ru₁Mo₃-JH (65.42 mV dec⁻¹), Ru₃Mo₁-JH (242.30 mV dec⁻¹), and Ru-JH (274.97 mV dec⁻¹), implying the superior catalytic kinetics of Ru₁Mo₁-JH (Figure 4b,c). Furthermore, the C_dl value of Ru₁Mo₁-JH (83.09 mF cm⁻²) was significantly higher than that of Mo-JH (0.30 mF cm⁻²), Ru₁Mo₃-JH (11.10 mF cm⁻²), Ru₃Mo₁-JH (8.89 mF cm⁻²), and Ru-JH (2.20 mF cm⁻²), which revealed that Ru₁Mo₁-JH was characterized by a large electrochemically active area (Figure 4d). To investigate the electron transfer ability and electrical conductivity of catalysts, EIS measurements were carried out with the corresponding equivalent circuit model, wherein R_s and R_ct represented the solution resistance and charge transfer resistance, respectively. As shown in Figure 4e,f, Ru₁Mo₁-JH displayed the lowest R_ct of 14.48 Ω compared with other reference catalysts, indicating its rapid electron transport and high electrical conductivity. These results confirmed the successful optimization of the electronic structure of Ru₁Mo₁-JH.

The precursors of Ru₁Mo₁, after activation at 470 °C for different amounts of time (0 s, 60 s, and 1 h), were measured to explore the role of low crystallinity, along with the control groups, Mo-JH, Ru-JH, and commercial Pt/C. As shown in Figure 5a, Ru₁Mo₁-JH displayed overpotentials of 25.08 mV and 120.52 mV to deliver current densities of 10 mA cm⁻² and 100 mA cm⁻², respectively, outperforming commercial Pt/C (the overpotentials at 10 mA cm⁻² and 100 mA cm⁻² were 42.27 mV and 149.27 mV, respectively). Moreover, as shown in Figure 5b, the Tafel slope of Ru₁Mo₁-JH (49.89 mV dec⁻¹) was also lower than that of Mo-JH (99.15 mV dec⁻¹), Ru-JH (274.97 mV dec⁻¹), Ru₁Mo₁ (99.40 mV dec⁻¹), Ru₁Mo₁-LT (173.04 mV dec⁻¹), and Pt/C (88.81 mV dec⁻¹), indicating its rapid reaction kinetics following the Volmer–Heyrovsky mechanism (Volmer step: H₂O + e⁻ + * → *H + OH⁻; Heyrovsky step: *H +H₂O + e⁻ → H₂ + OH⁻ + *) [40,41]. Based on the comparison of the catalysts in Figure 5c, it can be seen that the Mo species significantly enhanced catalytic activity, which further optimized HER activity through the lowering of crystallinity achieved via Joule heating. Notably, Ru₁Mo₁-LT, synthesized through conventional thermal treatment, exhibited a high overpotential of 253.01 mV to deliver 10 mA cm⁻², which was attributed to the blockage of active sites. Meanwhile, a large Tafel slope of Ru₁Mo₁-LT (173.04 mV dec⁻¹) indicated its high kinetic barrier. This might be due to the excessive adsorption capacity of Ru₁Mo₁-LT for OH⁻ and H₂O intermediates, which significantly hindered the performance of HER activity at high current densities [42]. These results showcased the critical role of crystallinity modulation in enhancing HER performance.

As shown in Figure 5d, Ru₁Mo₁-JH showed a much lower R_ct value of 2.2 Ω (Figure 4d) compared with Pt/C (24.57 Ω), Ru-JH (547.3 Ω), Mo-JH (4826 Ω), Ru₁Mo₁ (528.7 Ω), and Ru₁Mo₁-LT (481.9 Ω), which was consistent with its low overpotential and favorable kinetics. In contrast, Ru₁Mo₁-LT showed a significantly higher R_ct than that of Ru₁Mo₁-JH, highlighting the improved charge transport enabled by low crystallinity [43]. Unsurprisingly, Ru₁Mo₁-JH possessed a high C_dl value of 83.09 mF cm⁻² (Figure 5e), which indicated that Ru₁Mo₁-JH had a larger electrochemically active area and could expose a greater number of active sites compared with the crystalline Ru₁Mo₁-LT. Furthermore, the durability of Ru₁Mo₁-JH was evaluated via a chronopotentiometry test (100 mA cm⁻²). In addition, the superior alkaline HER performance of as-prepared Ru₁Mo₁-JH was retained even after 100 h of operation at 100 mA cm⁻² without obvious decay (Figure 5f), indicating its high stability for potential industrial applications. According to XPS analysis, the interactions of Ru-Mo-O in Ru₁Mo₁-JH were enhanced, making it less likely for active site loss to occur when the catalyst exerts its catalytic effect, resulting in excellent stability. In order to clarify the excellent HER performance of Ru₁Mo₁-JH, the electrocatalytic performances of as-prepared Ru₁Mo₁-JH and recently reported state-of-the-art HER catalysts in alkaline solution are summarized in Figure 5g. It should be noted that the performance of the Ru-Mo oxide low-crystallinity catalyst prepared via Joule heating in this study outperformed most other Mo-based and Ru-based electrocatalysts, proving that Ru₁Mo₁-JH exhibited excellent HER performance.

3.3. Practical Water Splitting

To evaluate the practical HER activity of Ru₁Mo₁-JH, an AEM electrolyzer coupled with a quantification system for H₂ was constructed to simulate industrial hydrogen production. As illustrated in Figure 6a, the AEM electrolyzer was equipped with Ru₁Mo₁-JH loading on nickel foam (NF) and RuO₂ as the cathode and anode, respectively, which were separated by an anion exchange membrane in a 1 M KOH electrolyte for overall water splitting. First, the actual volume of H₂ produced by Ru₁Mo₁-JH//RuO₂ under a constant current density of 100 mA cm⁻² was measured (Figure 6c). A comparative analysis with the theoretical volume of H₂ in Figure 6b confirmed that the Faraday efficiency of the Ru₁Mo₁-JH catalyst could reach 100%. Remarkably, the AEM electrolyzer composed of Ru₁Mo₁-JH and RuO₂ required only 2.16 V to operate at 185 mA cm⁻² (Figure 6d), which was lower than that of Pt/C//RuO₂. These results demonstrated the exceptional HER activity and industrial viability of Ru₁Mo₁-JH in practical electrolysis systems.

4. Conclusions

In summary, this work reports a rapid Joule heating method for synthesizing low-crystallinity Ru-Mo oxide catalysts featuring an internal porous structure that maximizes active site exposure. The optimized Ru₁Mo₁-JH catalyst exhibits excellent HER performance in 1 M KOH, achieving a low overpotential of 25.08 mV at 10 mA cm⁻² and a Tafel slope of 48.89 mV dec⁻¹, which is indicative of its rapid reaction kinetics. The high value of C_dl (83.09 mF cm⁻²) and minimal R_ct (14.48 Ω) further confirm Ru₁Mo₁-JH has a large electrochemically active surface area and efficient interfacial charge transfer. Compared with the fully crystallized Ru₁Mo₁-LT, the results indicate that low crystallinity plays a critical role in weakening Ru-H bond strength and optimizing hydrogen intermediate adsorption. Furthermore, Ru₁Mo₁-JH achieved 100% Faradaic efficiency and required only 2.16 V to deliver 185 mA cm⁻² in an AEM electrolyzer. All these results position Ru₁Mo₁-JH as a highly promising and cost-effective catalyst for industrial hydrogen production.