Boosting Hydrogen Evolution via Phase Engineering-Modulated Crystallinity of Ruthenium–Zinc Bimetallic MOFs

Abstract

The systematic design of ruthenium-based electrocatalysts for the hydrogen evolution reaction (HER) is crucial for sustainable hydrogen production via electrocatalytic water splitting in an alkaline medium. However, the mismatch between water dissociation and hydrogen adsorption kinetics limits its HER activity. Herein, we present a phase engineering-modulated strategy to develop an ultrasmall ZnRu bimetallic metal–organic framework electrocatalyst (ZnRu₃₀-ZIF) for catalyzing alkaline HER. Experimental results and density functional theory calculations indicate that the incorporation of Ru atoms modifies the crystal structure of the ZIF-8 phase, resulting in enlarged facet spacing and smaller nanocrystals (45 ± 3 nm). This optimization of the crystal structure regulates the electronic properties of the ZnRu₃₀-ZIF, forming a higher d-band center (−5.91 eV), which reduces the water dissociation energy (0.19 eV) and facilitates hydrogen desorption (ΔG_H* = 1.09 eV). The prepared ZnRu₃₀-ZIF exhibits a low overpotential of 48 mV at 10 mA cm⁻² and an excellent mass activity of 2.9 A mg_Ru⁻¹ at 0.1 V (vs. RHE). This work establishes a phase-engineering strategy for the preparation of high-performance Ru-based MOF electrocatalysts for HER.

1. Introduction

Hydrogen, recognized as one kind of clean energy carrier, is highly valued for its carbon-free nature and high energy density (~39 kWh kg⁻¹), making it a promising alternative to fossil fuels [1,2,3]. Electrocatalytic water splitting powered by renewable energy is the most effective method for large-scale green hydrogen production. However, the sluggish kinetics of the cathodic hydrogen evolution reaction (HER), which generally follow the Volmer–Tafel and Volmer–Heyrovsky mechanisms, result in low water-splitting efficiency and high energy consumption [4,5]. According to Sabatier’s principle, an intermediate hydrogen binding energy on the catalyst surface is indicative of an ideal HER catalyst [6]. Additionally, the kinetic energy barrier of HER is limited by the water dissociation process in an alkaline medium. As exhibited in the volcano plots of HER activity, platinum (Pt)-based catalysts exhibit the best HER performance, but their high cost and scarcity hinder large-scale applications [6,7]. To address this limitation, significant efforts have been devoted to developing non-Pt HER electrocatalysts over recent decades. Nevertheless, most of the non-Pt catalysts developed, including transition metal phosphides [8,9,10] and selenides [11,12,13], still face challenges in the mismatch between water dissociation and hydrogen adsorption capacities [4,14].

Recently, ruthenium has garnered significant attention for its application in alkaline hydrogen evolution reactions (HERs) due to its comparable electron configuration and water-binding energy to platinum (Pt) [15,16]. However, the unsatisfactory hydrogen binding capacity and water dissociation of Ru atoms hinder hydrogen desorption and mass transfer, limiting its activity [17,18]. Additionally, as a member of the noble metals, the high cost of Ru restricts its widespread application [19,20]. Dispersing Ru at the atomic level on suitable support has emerged as an effective strategy to improve the utilization of Ru and reduce the cost of Ru-based catalysts [21,22,23,24]. Furthermore, this method can optimize the HER kinetics via the metal–support interaction, which modulates the electronic structure of Ru atoms [25,26,27]. Therefore, designing novel supports with controllable composition and electronic structure for the incorporated ruthenium atoms is a crucial and challenging endeavor. Among various reported carrier materials, including carbon nanotubes [28,29], graphene oxide [30,31], and metal foams [32,33], metal–organic frameworks (MOFs) stand out as promising carriers due to their tunable structures, high porosity, and large specific surface area [6,15,34]. For instance, Ru@Ni-MOF [35], Ru-doped Ni-MOF [36], and Ru@ZIF-L(Co) [21] have demonstrated high hydrogen evolution reaction (HER) activity in alkaline electrolytes. However, these Ru-based MOFs often incorporate additional active components like nickel (Ni) and cobalt (Co) sites, which complicates the exploration of the HER mechanism and the understanding of metal–support interactions. Furthermore, several pristine MOFs cannot be directly used for catalytic HER due to their poor conductivity and limited catalytic activity. To overcome these limitations, innovative strategies to modulate the structure of MOFs are necessary [37]. Typically, the carbonization of ruthenium-based metal–organic frameworks (Ru-based MOFs) at high temperatures has been widely recognized and employed to enhance conductivity and catalytic performance [38,39]. Several Ru-based MOFs exhibit rapid kinetic performance for HER, such as Ru@NC [15], RuO₂/CuCoN@NC [40], and Ru-MoO₂@PC [41]. However, the carbonization process usually leads to structural collapse and Ru aggregation, which reduces atomic utilization and obstructs active site exposure [42]. Therefore, optimizing catalytic activity while simultaneously increasing ruthenium atomic utilization presents a significant challenge in the development of Ru-based MOF catalysts for HER.

Recently, the phase-engineering strategy has garnered significant attention in the design of Ru-based electrocatalysts owing to its advantages in precisely regulated shapes of nanocrystals and a profound comprehension of catalytic mechanisms [43,44,45]. Moreover, incorporating Ru atoms into the MOF lattice via a phase-engineering method can improve the stability of atomically distributed Ru sites while maintaining the metal–support interaction [6,10]. Hence, it is a promising strategy to prepare high-performance Ru-based MOFs using a phase-engineering strategy. Considering the inertness of zinc (Zn) in the HER and the disparities in electronic states and atomic sizes between Zn and Ru atoms, the design and synthesis of ZnRu bimetallic MOFs provide a distinctive approach to optimize Ru site dispersion and investigate HER mechanisms.

In this study, we report a one-pot solvothermal method to synthesize a ZnRu bimetallic MOF (ZnRu₃₀-ZIF) for HER in alkaline media. The synthesized ZnRu₃₀-ZIF exhibits an ultrasmall dodecahedral structure with a size of 45 ± 3 nm, which is beneficial to enhancing the exposure of Ru atoms and facilitating mass transfer. X-ray photoelectron spectroscopy (XPS) and theoretical analyses indicate that the coexistence of Zn and Ru within the ZIF-8 lattice leads to a lower d-band center and electron rearrangement, optimizing hydrogen adsorption and water dissociation. Consequently, the ZnRu₃₀-ZIF demonstrates superior alkaline HER activity (48 mV at 10 mA cm^−²), rapid HER kinetics (55.1 mV dec⁻¹), and excellent stability (50 h at 10 mA cm⁻²). Furthermore, the ZnRu₃₀-ZIF exhibits a significantly higher Ru mass-normalized activity of 2.4 A mg_Ru⁻² at −0.1 V (vs. RHE). This work proposes a simple strategy for synthesizing a novel dodecahedral ZnRu bimetallic metal–organic framework (MOF) that simultaneously optimizes water dissociation and hydrogen adsorption energy at Ru active sites.

2. Results and Discussion

2.1. Morphological and Structural Characterizations

The ZnRu bimetallic MOF was synthesized using a simple one-pot solvothermal method. Ruthenium(III) chloride and zinc(II) nitrate hexahydrate were separately dissolved in methanol and subsequently added to a 2-methylimidazole solution sequentially. The resulting mixture was then heated at 140 °C for 20 h (details are provided in the Experimental Section 3.3). The prepared catalysts are denoted as ZnRu_x-ZIF, where x (=10, 20, and 30, in this work) represents the feeding ratios of ruthenium chloride. Powder XRD patterns were employed to investigate the crystal structures of the synthesized catalysts. As shown in Figure 1a, all prepared catalysts exhibit strong diffraction peaks at 7.3°, 10.4°, 12.8°, 14.8°, 16.5°, and 18.1°, corresponding to the (011), (002), (112), (022), (013), and (222) planes of the standard ZIF-8, respectively. Notably, no diffraction peaks related to metallic Ru and RuO₂ were detected in any samples, indicating that Ru exists as a coordinated metal in the catalyst and that Ru agglomeration is absent [46]. Compared to ZIF-8, the diffraction peaks of ZnRu_x-ZIF shift to lower angles with the increased concentration of RuCl₃, suggesting enlarged interplanar spacing. Moreover, XRD patterns demonstrate that the diffraction peak broadens with increasing Ru content, indicating reduced crystallinity of ZnRu_x-ZIF. This could be ascribed to the greater atomic size of Ru compared to Zn, which is conducive to modulated interaction and an electronic structure [47]. The XRD results confirm the successful synthesis of the ZnRu bimetallic MOF via a phase-engineering method.

Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) measurements were conducted to reveal the morphologies of the prepared catalysts. As expected, in Figure 1b,f, the ZnRu₃₀-ZIF exhibits an ultrasmall dodecahedral shape of 45 ± 3 nm (Figure 1d), which is advantageous for increasing the specific surface area and exposing the active sites. The SEM-EDX analysis reveals a strong signal of Ru and Zn atoms, indicating the successful synthesis of the ZnRu bimetallic MOF (Figures S1 and S2). Moreover, the forms of Ru in the ZnRu₃₀-ZIF were further analyzed by high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM). The HAADF-STEM and the corresponding EDX element mapping demonstrate the uniform distribution of Ru, Zn, and N on the surface of ZnRu₃₀-ZIF (Figure 1g), confirming the presence of atomically dispersed Ru sites, consistent with the XRD results. For comparison, pristine ZIF-8 displays a larger standard dodecahedral shape of 612 ± 60 nm (Figure 1c,e and Figure S4). Based on theory and XRD results, the decrease in size of ZnRu₃₀-ZIF can be attributed to two factors: (1) the addition of RuCl₃ solution, with a lower pH of 3.2 compared to Zn(NO)₃ (6.5), which facilitates the linker deprotonation process and reduces the nucleation time, resulting in smaller particles [48]; (2) the incorporation of Ru, which disrupts the periodic arrangement of the ZIF-8 secondary building units, inhibiting particle growth [49]. The difference in size between ZIF-8 and ZnRu₃₀-ZIF confirms that phase engineering has a well-controlled effect on the morphology of nanocrystals. The substantial bulk nanostructure of ZIF-8 hinders the mass transfer within the catalyst and limits the utilization of active sites [1,11]. These characterization results confirm the successful synthesis of the ultrasmall dodecahedral ZnRu bimetallic MOF and atomically dispersed Ru sites.

The composition and electronic structure of ZnRu₃₀-ZIF were first elucidated using X-ray photoelectron spectroscopy (XPS). The binding energy of all elements was calibrated using the C-C band at 284.8 eV. XPS survey spectra of ZnRu₃₀-ZIF showed obvious peaks for Ru (Figure 2a), confirming the successful introduction of Ru atoms. In the C 1s spectra of ZnRu₃₀-ZIF (Figure 2b), three primary peaks can be deconvoluted at 284.8 eV, 286.5 eV, and 287.3 eV, corresponding to C-C, C-O-C, and O-C=O bonds, respectively [21]. Furthermore, the spin–orbit doublets located at 282.6 eV and 287.0 eV could be assigned to Ru⁴⁺ with an energy difference of 0.4 eV [4,22]. Moreover, the Ru 3p XPS spectra of ZnRu₃₀-ZIF show three peaks located at 463.9 eV, 485.9 eV, and 475.4 eV, corresponding to Ru 3p_5/2, Ru 3p_3/2, and Zn LM1, respectively. The higher binding energy of Ru 3p further confirms the Ru valence located at +4 [4]. The Ru⁴⁺-dominated high-resolution XPS spectrum of ZnRu₃₀-ZIF indicates that the surface Ru composition is in a high-oxidation state rather than metallic Ru. The N 1s XPS spectra of ZnRu₃₀-ZIF reveal the presence of the Ru-N bond, Zn-N bond, N-C bond, and O-N bond (Figure 2e and Figure S6), in which the O-N might be attributed to surface oxidation. In the high-resolution Zn 2p XPS spectra (Figure 2d), a pair of peaks at 1021.6 eV and 1044.6 eV are assigned to the Zn²⁺ in ZnRu₃₀-ZIF. In contrast to pure ZIF-8, the Zn²⁺ peak of ZnRu₃₀-ZIF exhibits a negative shift, implying that the introduction of Ru causes electron rearrangement and electron transfer from the Ru atom to the ligands. The asymmetrical distribution of the electron structure in ZnRu₃₀-ZIF is conducive to improving the poor conductivity characteristics of ZIF-8.

To further evaluate the structural information and properties of ZnRu₃₀-ZIF, the XPS valence band spectra (XPS-VBS) were analyzed. As shown in Figure 2f,g, the ZnRu₃₀-ZIF shows a higher d-band center (ε_d) of −5.91 eV than ZIF-8 (−9.35 eV). A higher ε_d prevents the antibonding orbital from being occupied by electrons, thereby enhancing adsorption energy [1,6,47]. Considering the hydrophobic properties of MOFs, the high ε_d for ZnRu₃₀-ZIF will facilitate mass transfer and enhance hydrogen adsorption. Furthermore, the ZnRu₃₀-ZIF shows an increased electron density near the Fermi level (0 eV), which is beneficial to facilitate electron storage and enhance HER kinetics [5,47]. Relying on the ultrasmall morphology and higher value of ε_d brought about by the effect of phase-engineering modulation, the ZnRu₃₀-ZIF can facilitate HER kinetics. In addition, inductively coupled plasma atomic emission spectrometry (ICP-OES) measurements were used to confirm the Ru content in ZnRu₃₀-ZIF. As displayed in Figure 2h, the loading of Ru in ZnRu_x-ZIF gradually increases with the addition of raw RuCl₃. Excitingly, ZnRu₃₀-ZIF exhibits a high Ru content of 7.8%, confirming the successful synthesis of Ru-based MOFs with high Ru content. These results exhibit promising prospects of ZnRu₃₀-ZIF for HER.

2.2. Electrocatalytic HER Performance

The HER performance of the synthesized samples was evaluated using a standard three-electrode system in a 1 M KOH electrolyte at room temperature. Linear sweep voltammetry (LSV) curves were recorded at a scan rate of 5 mV s⁻¹ with 100% iR compensation. Notably, as the incorporation of Ru increases, the HER activity is significantly enhanced, as illustrated in the LSV curves (Figure 3a). The ZnRu₃₀-ZIF displays superior HER activity, with an overpotential of 48 mV at the benchmark current density of 10 mA cm⁻², which is significantly lower than that of ZnRu₂₀-ZIF (98 mV) and ZnRu₁₀-ZIF (370 mV). For comparison, the current of the LSV curves for ZIF-8 and carbon black (CB) at the applied potential is negligible, indicating their intrinsic inertness for HER. The results confirm that Ru atoms in ZnRu_x-ZIF are the active sites. The ZnRu₃₀-ZIF also shows a low overpotential of 130 mV at 50 mA cm⁻² (Figure 3b), suggesting excellent structural stability at high current density. Furthermore, the Tafel slopes were derived from the iR-compensation LSV curves to reveal the HER kinetics. As shown in Figure 3c, the Tafel slope of ZnRu₃₀-ZIF is 55.1 mV dec⁻¹, significantly lower than that of ZnRu₂₀-ZIF (199.2 mV dec⁻¹), ZnRu₁₀-ZIF (390.5 mV dec⁻¹), and ZIF-8 (515.6 mV dec⁻¹). The low Tafel slope of ZnRu₃₀-ZIF indicates the rapid HER kinetics and implies a typical Volmer–Heyrovsky mechanism [32,35].

To further investigate the intrinsic activity of ZnRu₃₀-ZIF, the mass activity and the electrochemical surface area (ECSA) of ZnRu₃₀-ZIF were determined. Also, Ru mass-normalized activity was calculated based on the ICP-OES results. ZnRu₃₀-ZIF exhibits a mass-normalized activity of 2.4 A mg_Ru⁻¹ at −0.1 V (vs. RHE) (Figure S7), which is higher than that of ZnRu₂₀-ZIF (1.5 A mg_Ru⁻¹) and ZnRu₁₀-ZIF (1.0 A mg_Ru⁻¹). The ECSA of the sample was determined by double-layer capacitance (C_dI), calculated from the CV curves in the non-Faradic potential range at a scan rate of 5–25 mV s⁻¹. All CV curves of samples are displayed in Figure S8. As shown in Figure 3d, ZnRu₃₀-ZIF demonstrates a larger C_dI of 18.9 mF cm⁻² compared to ZIF-8 (0.23 mF cm⁻²), indicating the existence of more active sites in ZnRu₃₀-ZIF than ZIF-8. This could be attributed to the incorporation of Ru and the effect of the smallness of ZnRu₃₀-ZIF [50]. Additionally, electrochemical impedance spectroscopy (EIS) curves of the catalysts were recorded at −0.1 V (vs. RHE) across a frequency range of 100 kHz to 0.01 Hz to gain insight into HER kinetics. The Nyquist plots of the samples were fitted using an electrical equivalent circuit, as shown in Figure S9a. As expected, the ZnRu₃₀-ZIF has the lowest charge transfer resistance (R_ct) of 165.4 Ω compared with the ZIF-8 (33,573 Ω), ZnRu₁₀-ZIF (2065 Ω), ZnRu₂₀-ZIF (589.1 Ω), and reported Ru-based HER catalysts. The Bode plots (Figure S9b,c) illustrate the reduced frequency required to achieve the largest phase angle with increasing Ru content, with the highest frequency observed in ZnRu₃₀-ZIF (approximately 10^2.3). This result demonstrates the rapid filling of the double layer on the ZnRu₃₀-ZIF surface and the rapid depletion of the surface charge, indicating fast charge transfer and HER kinetics. This result is consistent with the Tafel slope analysis. According to the XPS results, the excellent HER kinetics for ZnRu₃₀-ZIF than ZIF-8 are attributed to the higher d-band center and electronic rearrangement by the incorporation of Ru. Importantly, the long-term stability of the HER catalyst is another key factor for its practical application. To investigate the durability of ZnRu₃₀-ZIF, a chronopotentiometry test was carried out at a benchmark current density of 10 mA cm⁻² (Figure 3e). As shown in Figure 3e, there is a negligible change in potential during the 50 h test at 10 mA cm⁻², verifying the robust stability of ZnRu₃₀-ZIF. According to HAADF-STEM and XRD results, Ru atoms were successfully implanted into the crystal structure of ZIF-8. The high thermodynamic stability of the crystal and the robust coordination configuration effectively anchor Ru atoms, preventing their dissolution and leaching. Therefore, ZnRu₃₀-ZIF has excellent stability over long operating periods. Ex situ characterization was performed using ICP, SEM, XRD, and XPS to analyze the structural changes in ZnRu₃₀-ZIF after stability testing (A-ZnRu₃₀-ZIF). Compared to pristine electrolytes, the Ru concentration remains almost unchanged (Figure S10), proving the stability of Ru sites in ZnRu₃₀-ZIF. The XRD pattern for A-ZnRu₃₀-ZIF is similar to that of the fresh ZnRu₃₀-ZIF (Figure S11), indicating the retention of the ZIF-8 structure during the stability test. SEM and corresponding EDS results reveal that the A-ZnRu₃₀-ZIF retained the ultrasmall dodecahedral morphology and the uniform distribution of Ru (Figure S12). The XPS survey spectrum of A-ZnRu₃₀-ZIF exhibits a new peak belonging to the F element, resulting from the Nafion-117 ionomer (Figure S13a). After HER, the peaks of Ru 3p for A-ZnRu₃₀-ZIF shift to lower binding energy (Figure S13c), indicating that the Ru sites trap electrons during the HER process. The similar binding energy shift is observed in Zn 2p spectra, which confirm that Ru is the active site in ZnRu₃₀-ZIF for HER. In particular, the ZnRu₃₀-ZIF exhibits excellent HER performance in alkaline electrolytes, which is superior to most of the reported Ru-based HER catalysts (Figure 3f and Table S1).

2.3. DFT Calculations

The differential charge density, water dissociation energy barrier, and Gibbs free energy of the intermediate adsorption were investigated via spin-polarized density functional theory (DFT) calculations. Models of ZnRu₃₀-ZIF and metallic Ru were constructed and optimized to reveal the modulation of the ZIF-8 phase for the electronic structure of Ru sites. As shown in Figure 4b, the differential charge density of ZnRu₃₀-ZIF demonstrates significant electron transfer between Ru and the ligands, as well as increased electron density at the Zn sites. The results reveal a substantial interaction between the Ru sites and the ZnRu₃₀-ZIF carrier, corresponding to the XPS study. This indicates the unique electronic structure of the phase engineering-modulated Ru sites. The alkaline HER process on the catalysts involves three steps: water dissociation, hydrogen adsorption, and desorption (Figure 4a). As shown in Figure 4c, the adsorption energy of H₂O is similar for ZnRu₃₀-ZIF (−0.8 eV) and metallic Ru (−0.88 eV), suggesting excellent mass transfer performance of ZnRu₃₀-ZIF. Notably, ZnRu₃₀-ZIF exhibits a lower energy barrier of water dissociation (0.19 eV) than metallic Ru (0.52 eV), indicating that the Ru sites in ZnRu₃₀-ZIF are more effective at cleaving the HO-H bond. The enhanced water dissociation comes from the unique coordination structure and the metal–support interaction, which regulate the d-band structure of the Ru sites in ZnRu₃₀-ZIF. Subsequently, the Gibbs free energy for hydrogen adsorption (ΔG_H*) of ZnRu₃₀-ZIF is −1.09 eV (Figure 4d), which is much lower than that of metallic Ru (−1.43 eV). The lower ΔG_H* is more favorable for hydrogen molecule desorption [4,34,47]. In conclusion, DFT calculations demonstrate the excellent alkaline HER kinetics of ZnRu₃₀-ZIF with the balance of water dissociation and hydrogen adsorption, confirming that the Ru sites in ZnRu₃₀-ZIF are favorable for HER.

3. Materials and Methods

3.1. Chemicals

CoZinc(II) nitrate hexahydrate (Zn(NO₃)₂, AR 98%), methanol (AR 99.9%), isopropyl alcohol (AR 99.7%) and 2-methylimidazole (AR) were purchased from Aladdin, Shanghai, China. Ruthenium(III) chloride (RuCl₃, AR 99.5%), Nafion solution (5%), carbon black (CB, 30 nm), ethanol absolute (99.8%) were obtained from InnoChem, Beijing, China_.

3.2. Synthesis of ZIF-8

Quantities of 0.7 g zinc nitrate hexahydrate and 1.5 g 2-methylimidazole were dissolved in 40 mL and 20 mL methanol, respectively, under ultrasound. Subsequently, the Zn(NO₃)₂ solution was added to the 2-methylimidazole solution with strong stirring. After continuous stirring for 0.5 h, the mixture was transferred to a 100 mL Teflon-lined steel vessel and heated at 140 °C for 20 h. After cooling down to room temperature, the resulting solid was separated from the mixture solution via centrifugation at 10,000 rpm. The solid was dispersed into 50 mL methanol, followed by ultrasound for 0.5 h and centrifugation to remove the unreacted salts and ligands. After three washes, the ZIF-8 power was dried at 80 °C in a vacuum.

3.3. Synthesis of ZnRu₃₀-ZIF

Next, 0.3 g RuCl₃ was dissolved in 5 mL methanol. The Zn(NO₃)₂ solution and 2-methylimidazole solution were prepared using the same recipe as ZIF-8. The RuCl₃ solution was first added to the 2-methylimidazole solution drip by drip under ultrasound. After continuous ultrasound for 30 min, the Zn(NO₃)₂ solution was added to the above solution with strong stirring. The subsequent hydrothermal process was carried out using the same method as for the preparation of ZIF-8. The resulting catalyst was named ZnRu₃₀-ZIF. For comparison, the ZnRu₂₀-ZIF and ZnRu₁₀-ZIF were synthesized with 0.2 g RuCl₃ and RuCl₃, respectively.

3.4. Material Characterizations

A Malvern Panalytical Empyrean powder X-ray diffraction (XRD) diffractometer with Cu Kα radiation was used to analyze the sample lattice structure. The sample morphology was characterized using a Zeiss Gemini 300 scanning electron microscope and a JEM-F200 transmission electron microscope. X-ray photoelectron spectroscopy (XPS) characterizations were carried out using a Thermo Scientific K-Alpha electron spectrometer. The X-ray source was monochromatic Al Kα (hv = 1486.6 eV) and the degree of vacuum of the analysis chamber was 5.0 × 10⁻⁷ mPa. Inductively coupled plasma optical emission spectroscopy (ICP) was carried out to analyze the metal content using a Prodigy 7 spectrometer.

3.5. Electrochemical Measurement

For the electrochemical test, 2 mg catalyst powder and 0.5 mg carbon black were first dispersed in a 1.0 mL mixture solution (0.8 mL ethanol absolute, 0.18 mL deionized water, and 0.02 mL 5% Nafion solution) to form a homogeneous catalyst ink. To prepare the working electrode, 0.006 mL of the catalyst ink was dropped onto a 3 mm-diameter glass carbon electrode (GCE) with IR lamp heat. The catalyst loading of the working electrode was 0.17 mg cm⁻².

In this work, all electrochemical measurements were performed using a Corrtest electrochemical analyzer (CS310H) at room temperature. For the hydrogen evolution reaction test in 1 M KOH, the GCE with a catalyst loading of 0.17 mg cm⁻² was used as the working electrode. A saturated Ag/AgCl and graphite rod were carried out in a three-electrode system and employed as the reference and counter electrode, respectively. The potential was converted into the reversible hydrogen electrode using the formula:

$${\mathrm{E}}_{\mathrm{RHE}}={\mathrm{E}}_{\mathrm{Ag}/\mathrm{AgCl}}+0.196+0.0591\times \mathrm{pH}.$$

Before the HER testing, the dissolved air in the electrolyte was exhausted using a vacuum pump. Linear sweep voltammetry (LSV) measurements were performed throughout the potential range of 0.1–−0.5 V (vs. RHE) with a scan rate of 5 mV s⁻¹. The Tafel slope was calculated from the LSV result via the following equation:

$$E=\mathrm{a}+\mathrm{b}\times \log \mathrm{j}$$

where E is the potential, b is the Tafel slope, and j is the current density.

The electrochemical surface area (ECSA) was estimated by calculating the double-layer capacitance, which was determined using CV measurement in a potential range of 0.1–0.2 V (vs. RHE). The electrochemical impedance spectra (EIS) were conducted at a potential of 0.1 V (vs. RHE) with an amplitude voltage of 10 mV. The high-frequency resistance was taken as the solution resistance for LSV curve correction.

3.6. Theoretical Calculations

The Perdew–Burke–Ernzerhof (PBE) exchange-correlation function in the generalized gradient approximation (GGA) method was used to set the cutoff energy to 400 eV. The energy and force convergence criteria were 10⁻⁵ eV and 0.02 eV/Å. The vacuum spacing was set to 15 Å for all periodic slab calculations. A 2 × 2 × 1 Monkhorst-pack k-point was selected. The Gibbs free energy (∆G) of the reaction intermediates was calculated by the following equation:

$$∆G=∆E+∆E_{\mathrm{ZPE}}-T\times ∆S$$

where ∆E is the adsorption energy of the reaction intermediate (H₂O*, OH* and H*) on the metallic sites, ∆E_ZPE is the zero-point energy, and ∆S is the entropy change. The values of ∆E_ZPE and ∆S were derived from frequency calculations.

4. Conclusions

In summary, we successfully synthesized an ultrasmall ZnRu bimetallic MOF (ZnRu_x-ZIF) with atomically dispersed Ru sites to catalyze alkaline HER. The optimized ZnRu₃₀-ZIF demonstrates excellent HER activity, characterized by a low overpotential of 48 mV to reach 10 mA cm⁻², along with remarkable stability for 50 h at 10 mA cm⁻². The ultrasmall dimensions of ZnRu₃₀-ZIF (45 ± 3 nm) and the atomically dispersed Ru sites synergistically enhance the utilization of Ru atoms. DFT calculations and experimental analyses indicate that the incorporation of Ru facilitates electron transfer within the ZnRu₃₀-ZIF dodecahedron and improves the d-band center of the catalyst. In addition, the Ru sites in ZnRu₃₀-ZIF demonstrate rapid HER kinetics with lower water dissociation energy (0.19 eV) and hydrogen desorption (1.09 eV). Therefore, the ZnRu₃₀-ZIF achieves an outstanding mass activity of 2.4 A cm⁻² at −0.1 V (vs. RHE), which is 1.6 and 2.4 times higher than that of ZnRu₂₀-ZIF and ZnRu₁₀-ZIF, respectively. This work provides a novel method for the design and development of high-performance Ru-based MOFs for alkaline HER.