Interface engineering of RuO₂/Ru–Co₃O₄ heterostructures for high-efficiency oxygen evolution in acid

Abstract

Proton exchange membrane water electrolysis (PEMWE) powered by renewable electricity is a promising technology for green hydrogen production. Ru-based catalysts are a cost-effective alternative to Ir-based counterparts for the oxygen evolution reaction (OER), yet achieving balanced activity and stability remains challenging. In this study, we optimize the performance of RuO₂ by constructing RuO₂/Ru–Co₃O₄ heterogeneous interfaces. Our results indicate that RuO₂/Ru–Co₃O₄ catalyst pyrolyzed at lower temperatures exhibits uniformly small particle sizes with abundant heterointerfaces and strong interfacial electronic interactions. Elevated pyrolysis temperatures enlarge catalyst particle sizes, simultaneously decreasing heterointerfaces and weakening the interfacial interactions. In rotating disk electrode experiments, the mass activity of the RuO₂/Ru–Co₃O₄ catalysts is significantly improved compared to the commercial RuO₂. This improvement is attributed to the heterogeneous interfaces, which offer greater utilization of active sites and enhanced charge transfer capabilities. Furthermore, the mass activity of the RuO₂/Ru–Co₃O₄ catalysts decreases with increasing pyrolysis temperatures due to the reduction in heterogeneous interfaces. Additionally, the stability of the RuO₂/Ru–Co₃O₄ catalysts is notably superior to that of commercial RuO₂, owing to the interfacial electronic interactions. However, this stability is negatively impacted by higher pyrolysis temperatures as the interfacial electronic interactions weaken. In preliminary membrane electrode assembly tests, the RuO₂/Ru–Co₃O₄ catalyst with low Ru loading shows higher activity than commercial RuO₂. Further triple-phase boundary optimization of the catalyst layer is needed. These findings contribute to advanced RuO₂-based catalyst design for PEMWE.

---

Introduction

Renewable electricity-powered proton exchange membrane water electrolysis (PEMWE) technology has garnered considerable acclaim for its superior energy efficiency, high current density, and exceptional hydrogen purity in green hydrogen production.^1–4 This technology hinges on Ir-based catalysts to facilitate the oxygen evolution reaction (OER) under challenging acidic and oxidative conditions.^5–8 However, the high cost and scarce availability of Ir significantly hinder its widespread application. In contrast, Ru-based catalysts, notably RuO₂, offer a compelling alternative.^9–11 These catalysts are approximately one-tenth the price of Ir-based catalysts and demonstrate high intrinsic activity. Further improving the OER performance of RuO₂ is crucial for advancing its application in PEMWE.^12–14

Two principal mechanisms govern the acidic OER catalyzed by RuO₂: the adsorbate evolution mechanism (AEM) and the lattice oxygen oxidation mechanism (LOM).^15–17 In the AEM, the RuO₂ catalyst demonstrates commendable stability, but the scaling relationship between the adsorption energies of intermediates constrained its activity. In contrast, its activity in the LOM improves, but its stability decreases due to structural collapse.^18,19 Various strategies have been developed to address the trade-off between catalytic activity and structural stability. These strategies include the hybridization of RuO₂ with Ir,^20,52 the introduction of stable supports to anchor RuO₂,^21,53 chemical doping with other elements to regulate the valence state of Ru,^22,54 and the formation of a solid solution.^23,55 These strategies have proven effective in improving both the catalytic performance and structural stability of RuO₂, providing valuable insights for designing advanced OER catalysts.^20–23

Among various strategies, one of the most extensively studied is the introduction of supports, such as TiO₂, CeO₂, WO₃, MnO₂, and MoO₃, to anchor Ru ions.^24–28 These supports could create synergistic interactions with the active sites, boosting activity and extending their operational lifetime. One recent work developed a RuO₂/MnO₂ heterostructure with MnO₂ acting as an electron reservoir for RuO₂. In this nanostructure, MnO₂ facilitates electron transfer from RuO₂, enhancing the OER activity, and donates electrons to RuO₂, improving its stability.²⁷ Another study designed RuO₂/MoO₃ catalysts with abundant and intimate interfaces to enable the reactive *O intermediate spillover from RuO₂ to MoO₃. The RuO₂/MoO₃ catalysts demonstrate both high activity and stability owing to the oxygen spillover process that helps suppress over-oxidation and dissolution of RuO₂.²⁸

Spinel Co₃O₄, whose theoretical activity is comparable to RuO₂, has also been explored as support for RuO₂. Extensive studies have shown that the stability and activity of RuO₂/Co₃O₄ composites are significantly improved due to electron transfer between Ru and Co.^29–32 These findings provide valuable insights for further developing RuO₂/Co₃O₄ catalysts. By optimizing interfacial interactions and structural design, it is possible to achieve more significant performance gains, addressing the key challenge of balancing activity and stability of RuO₂.

Building on this foundation, we synthesized a series of RuO₂/Ru–Co₃O₄ (RCO) catalysts by pyrolyzing Ru/ZIF-67 at controlled temperatures. Small particle sizes, abundant interfacial contacts, and strong interactions between RuO₂ and Co₃O₄ characterize the RCO catalysts. As the pyrolysis temperature increases, the particle size of the catalysts grows while the interfacial interaction between RuO₂ and Co₃O₄ weakens. In the rotating disk electrode (RDE) experiments, the RCO catalysts demonstrate significantly enhanced mass activity (MA) and stability compared to commercial RuO₂. Among these, the RCO catalyst synthesized at low temperatures exhibits the most significant increase in MA due to its abundant heterointerfaces, which provide the largest electrochemical active surface area (ECSA) and lowest charge transfer resistance (R_ct). Additionally, this low-temperature synthesized RCO catalyst demonstrates the greatest improvement in stability, attributed to its strong interfacial electronic interactions. During membrane electrode assembly (MEA) testing, the RCO catalyst, which features low Ru loading, shows superior activity to commercial RuO₂. However, the porous structure of the catalyst layer hinders the establishment of a triple-phase boundary. These findings highlight the promising potential of RCO catalysts for efficient green hydrogen production while emphasizing the need for further optimization to overcome structural constraints in PEMWE applications.

Experimental

Materials

Cobalt(II) nitrate (Co(NO₃)₂·6H₂OO, 99.7%, hexahydrate, Macklin, China), 2-methylimidazole (2-MIM, C₄H₆N₂, ≥98%, Aladdin, China), ruthenium trichloride hydrate (RuCl₃·xH₂O, Tianjin Baima Technology Co. LTD, China), ruthenium(IV) oxide (RuO₂, ≥99.9%, Aladdin, China), Pt/C (70 wt% Pt, HiCaP 70, CASI NEW Energy Technology (Ningbo) Co., Ltd, China), perchloric acid (HClO₄, for analysis, ca. 70% solution in water, Thermo Scientific Chemicals, America), isopropanol (IPA, >99.7%, analytical reagent grade, Kermel, China), methanol anhydrous (MeOH, ≥99.5%, analytical reagent grade, SCR, China), deionized water (DI water, Milli-Q, 18.2 MΩ cm at 25 °C), ultrapure nitrogen (N₂, 99.999%), ultrapure oxygen (O₂, 99.999%), and a Nafion D521 dispersion (5 wt%, EW = 1100, Alfa Aesar) were used as received.

Synthesis of the ZIF-67

To prepare ZIF-67, 2.91 g (10 mmol) of Co(NO₃)₂·6H₂O and 3.284 g (40 mmol) of 2-MIM were separately dissolved in 20 ml of methanol to form clear solutions. The 2-MIM solution was then quickly added to the Co(NO₃)₂·6H₂O solution and sonicated for 30 minutes in an ultrasonic bath. This mixture was stirred at 600 rpm for 18 h at room temperature. Finally, the precipitate was collected, rinsed 3 times with methanol, and dried under vacuum at 60 °C for about 1 h to obtain ZIF-67.

Synthesis of the Ru/ZIF-67

First, 0.33 g of ZIF-67 and 0.08 g of RuCl₃·xH₂O were separately added to 15 mL of DI water and sonicated for 5 min. The dispersion of ZIF-67 was then quickly added to the RuCl₃·xH₂O solution in an ultrasonic bath and sonicated continuously for another 45 minutes. Finally, the mixture was filtered and dried to obtain the precursor, Ru/ZIF-67, with Ru ions loaded on ZIF-67.

Synthesis of the RuO₂/Ru–Co₃O₄ catalysts

The obtained Ru/ZIF-67 was annealed separately at 300, 350, 400, and 450 °C for 3 h in air at a heating rate of 2 °C min⁻¹. The resulting samples are denoted as RuO₂/Ru-Co₃O₄_T and abbreviated as RCO_T, where T represents the annealing temperature. In addition, ZIF-67 was heat-treated at 350 °C to prepare a control sample of Co₃O₄.

Physical characterizations

A high-resolution transmission electron microscope (HR-TEM, JEM-2100F, JEOL, Japan) equipped with an energy dispersive spectrometer (EDS), a scanning electron microscope (SEM, JSM-7001F, JOEL, Japan), a thermal gravimetric (TG) and derivative thermogravimetric (DTG) analysis (TMA7100, Hitachi, Japan), a powder X-ray diffraction diffractometer (XRD, Bruker D8-Advance-A25, Britain), an X-ray photoelectron spectroscopy (XPS, Thermo Scientific, America), and inductively coupled plasma mass spectrometry (ICP-MS, NexION 5000, PerkinElmer Inc, America) were used to characterize the obtained samples. The Ru, Co, and Pt loadings in PEMWE were measured using an X-ray fluorescence spectrometer (XRF, ARL QUANT'X EDXRF Analyzer, Thermo Fisher Scientific Inc, America).

Electrochemical measurements

Electrochemical measurements were conducted using a three-electrode system controlled by a Pine Research MSR rotator (Pine Research Instrumentation Co. Ltd, America) and a bipotentiostat (CHI 760E, CH Instruments, Inc., China). The experiment utilized a five-necked electrochemical cell containing 0.10 M HClO₄ solution. A rotating disk electrode (RDE, Pine Research Instrumentation Co. Ltd, America) with a glassy carbon electrode (GCE, 5 mm diameter) was used as the working electrode. A graphite rod served as the counter electrode, and an Ag/AgCl (KCl, saturation) electrode was employed as the reference electrode. The Ag/AgCl electrode was calibrated to a reversible hydrogen electrode (RHE), which was constructed by continuously bubbling pure H₂ gas onto a Pt coil electrode placed in a glass tube filled with the electrolyte solution.

The catalyst ink was prepared by dispersing 2 mg of catalyst in a mixture of DI-water (200 μL), IPA (200 μL), and D521 Nafion dispersion (4 μL, 5 wt%) in a sonication bath for 1 hour. Then, electrodes with a catalyst loading of 0.500 mg cm⁻² for RCO_T were prepared by depositing 19.8 μL of the ink onto the GCE. For comparison, 7.28 μL of commercial RuO₂ ink and 12.44 μL of Co₃O₄ ink were deposited on GCE to obtain loadings of 0.186 mg cm⁻² and 0.314 mg cm⁻², respectively. The Ru mass (m_Ru) loadings for RCO_T and commercial RuO₂ were calculated to be 0.141 mg cm⁻². The Co mass (m_Co) loadings for RCO_T and Co₃O₄ were calculated to be 0.230 mg cm⁻².

Cyclic voltammograms (CVs) were recorded at varying scan rates (v) from 40 to 120 mV s⁻¹ in an N₂-saturated 0.10 M HClO₄ solution, with the potential window set at ± 50 mV vs. the open-circuit potential (OCP). Linear sweep voltammograms (LSVs) were recorded from 1.1 to 1.55 V vs. RHE at a scan rate of 5 mV s⁻¹ in an O₂-saturated 0.10 M HClO₄ solution, with a rotation rate set at 1600 rpm. Unless noted, all the LSV tests were corrected with 100% internal resistance (iR) compensation. The durability was assessed through chronopotentiometry testing at a constant current density of 10 mA cm⁻². Electrochemical impedance spectroscopy (EIS) measurements were performed at 1.5 V vs. RHE within a 100 kHz–0.1 Hz frequency range.

The electrochemically active surface areas (ECSAs) for RCO_T, commercial RuO₂, and Co₃O₄ were estimated from the electrochemical double-layer capacitance (C_dl) of the catalytic surfaces. The C_dl is determined from the slope of the plot of i_Cvs. ν in CV, where i_C is the charging current at the OCP corresponding to ν. ECSA was calculated using the equation:

where A is the area of GCE, C_s is the specific capacitance of an atomically smooth planar surface of the material per unit area under identical electrolyte conditions. A value of 0.040 mF cm⁻² was used for C_S, which is based on the typical reported values.⁵¹

The Tafel plots were derived from the following equation:

η = a + b log j (2)

where η is the overpotential, defined as the difference between the potential of an electrode and its equilibrium potential (E_eq, E_eq = 1.23 V vs. RHE), j is the current density, a is the intercept at the y-axis, and b is the Tafel slope.

The mass activity (MA) was calculated using the equation:

The specific activity (SA) was calculated using the equation:

PMEWE tests

Anode catalyst inks were prepared by mixing RCO_350 or commercial RuO₂ catalysts with Nafion D521 ionomer dispersion in a mixture of DI-water and IPA, followed by ultrasonication for 30 min. The catalyst concentration was 6.1 mg mL⁻¹. The volume ratio of DI-water to IPA was 0.042. The weight ratio of the ionomer to catalyst was 0.42.

A commercial Pt/C catalyst was used as the cathode catalyst. The cathode catalyst ink was prepared using a method similar to that of the anode. The catalyst concentration was 2.1 mg mL⁻¹. The weight ratio of the ionomer to carbon was 0.6.

The catalyst-coated membrane (CCM) with an active area of 5 cm² was prepared by ultrasonically spraying the anode inks onto one side of the Nafion® NR115 membranes at 60 °C, followed by spraying the cathode inks on the other side. The anode loadings were ca. 0.26 mg_Ru cm⁻² for RCO_350 and ca. 0.90 mg_Ru cm⁻² for commercial RuO₂. The cathode loading was ca. 0.40 mg_Pt cm⁻². These loadings were measured using an XRF. The obtained CCM was assembled with two porous transport layers, two polytetrafluoroethylene gaskets, a titanium felt with a 5 cm² flow field, a graphite bipolar plate also featuring a 5 cm² flow field, and two metal current collectors to complete the full PMEWE.

During the PEMWE operation, DI water was pumped into the anode and circulated between the anode and a glass flask at a 100 mL min⁻¹ flow rate. A bias voltage was applied using a high-current electrochemical workstation (VersaSTAT 3F, Princeton Applied Research, AMETEK Inc, America). The current–voltage (I–V) curves were recorded by polarizing the anode from 1.2 to 1.8 V vs. RHE at a scan rate of 10 mV s⁻¹.

Results and discussion

The morphology and composition of RCO_T catalysts

Scheme 1 illustrates the synthesis strategy for RCO_T catalysts. ZIF-67 was synthesized as a precursor for the Co₃O₄ support. The pore size of ZIF-67 is 0.34 nm, which allows the impregnated Ru³⁺ to pass through the pore channel and adsorb around CoN₄ units.³³ Subsequently, during the in situ conversion process under heating, oxygen diffuses into the micropore, forming RCO_T catalysts with contact-rich interfaces. We first examined the impregnation of Ru³⁺ within ZIF-67. Fig. S1 and S2† show that ZIF-67 and Ru/ZIF-67 possess dodecahedral structures. However, the surfaces of Ru/ZIF-67 appear rougher compared to ZIF-67. EDS elemental mappings (Fig. S3†) show distinct Ru, Co, and N signals, revealing a uniform Ru dispersion within the ZIF-67 framework. XRD patterns for both ZIF-67 and Ru/ZIF-67 (Fig. S4†) suggest no noticeable difference, indicating that the introduction of Ru³⁺ does not alter the crystal structure of ZIF-67.³⁴ Consequently, we speculated that the Ru³⁺ is effectively adsorbed within the ZIF-67. It is further supported by ICP-MS analysis, which confirms a Co : Ru atomic ratio of ca. 2.8 : 1 (Table S1†).

Scheme 1 Schematic illustration for the synthesis of RCO_T catalysts.

We then elucidated the formation process of RCO_T catalysts under different annealing temperatures. The TG-DTG curves (Fig. S5†) indicate that ZIF-67 decomposition begins at 330.3 °C (mass loss onset), with a sharp mass loss of 52.97% peaking at 375.0 °C. This peak corresponds to the collapse of the ZIF-67 framework and the formation of Co₃O₄.³⁵ Literature reports also confirm that RuCl₃ converts to RuO₂ above 300 °C.³⁶ To prevent the agglomeration of the catalyst caused by high temperature, we finally chose a pyrolysis temperature range of 300–450 °C. Fig. 1a shows the XRD patterns of RCO_T catalysts synthesized at different temperatures. At 300 °C, two broad peaks are evident at 31.3 and 36.9°, corresponding to the (220) and (311) plane of Co₃O₄ (PDF #73-1701), without signals from RuO₂. We hypothesize that Ru remains in a non-oxidized state or is amorphous at this temperature. Upon increasing the temperature to 350 °C, distinct diffraction peaks associated with RuO₂ (PDF #73-1469) emerge, indicating the formation of RuO₂. Concurrently, the diffraction peaks of Co₃O₄ become more pronounced. These diffraction peaks shift slightly towards lower angles, likely attributed to incorporating larger Ru atoms into the Co₃O₄ lattice.³⁷ As the calcination temperature is raised to 400 and 450 °C, the crystallinity of the RCO_T catalysts is further improved, as evidenced by the enhanced peak intensities. The central diffraction peaks of Co₃O₄ remain consistent with those observed at 350 °C, suggesting that temperature variations within this range do not significantly influence the doping of Ru. These results indicate that RuO₂ and Ru-doped Co₃O₄ generation begins at 350 °C, with their crystallinity enhancing as the annealing temperature increases.

Fig. 1 Phase and structural characterization: (a) XRD patterns of Co₃O₄, RuO₂, and RCO_T catalysts. (b–e) SEM image, TEM image, EDS elemental mappings, and HR-TEM image of typical RCO_350. (f) Deconvoluted high-resolution Ru 3p XPS spectra of the commercial RuO₂ and RCO_T catalysts. (g) Deconvoluted high-resolution Co 2p XPS spectra of the Co₃O₄ and RCO_T catalysts.

SEM and TEM analyses (Fig. S6 and S7† and 1b and c) reveal that the RCO_T catalysts exhibit porous dodecahedral architectures. The EDS mappings of typical RCO_350 in Fig. 1d demonstrate a homogenous distribution of Ru, O, and Co elements, suggesting significant interfacial contact between RuO₂ and Ru–Co₃O₄. Furthermore, the HR-TEM analysis of typical RCO_350 in Fig. 1e confirms the presence of heterointerfaces. Ultrasmall RuO₂ nanoparticles, ca. 5 nm in size, are observed on the surface of similarly sized Co₃O₄ nanoparticles, characterized by crystal planes (101) and (111) with d-spacing of 0.255 and 0.247 nm, respectively. Additionally, in Fig. S8,† the particle size slightly increases, and the boundaries between particles become more distinct as the calcination temperature increases from 350 to 450 °C. These findings suggest that the heating conversion process forms RCO_T catalysts with small particle sizes and contact-rich interfaces.

The elemental compositions and valence states of RCO_T catalysts, Co₃O₄, and commercial RuO₂ were examined using XPS. The XPS spectra of the RCO_T catalysts confirm the presence of Co, Ru, O, and C elements, as shown in Fig. S9.† Furthermore, the O 1s spectra of RCO_T catalysts were deconvoluted, revealing three distinct peaks corresponding to lattice O²⁻ (529.35 eV for Co–O and 528.35 eV for Ru–O), O_V (& O–H) (530.75 eV), and O_ads (532.5 eV).^38,39 Detailed fitting results were provided in Fig. S10 and Table S2.† The high-resolution Ru 3p spectrum of the RCO_350 catalyst shown in Fig. 1f is deconvoluted into four sub-peaks, corresponding to the spin orbitals of Ru 3p_3/2 (Ru⁴⁺ at 462.87 eV, Ru³⁺ at 466.0 eV) and Ru 2p_1/2 (Ru⁴⁺ at 485.12 eV, Ru³⁺ at 487.75 eV) peaks.^40,41 Compared to the commercial RuO₂, RCO_350 exhibits a shift of approximately 0.87 eV to higher binding energy in the Ru 3p_3/2 XPS peak. Similarly, RCO_400 and RCO_450 present shifts of ca. 0.62 eV in the same direction. The Ru 3p_3/2 peak of RCO_T shifts to higher binding energy, indicating the elevated oxidation state of Ru. The change is accompanied by an upward shift in the Ru d-band center position. With the increase in temperature, the degree of electron loss in RCO_T decreases, correspondingly reducing the magnitude of Ru d-band center upshift.⁴²

The high-resolution Co 2p spectrum of the RCO_350 catalyst displayed in Fig. 1g is deconvoluted into seven sub-peaks, corresponding to the spin orbitals of Co 2p_3/2 (Co³⁺ at 779.23 eV, Co²⁺ at 781.29 eV) and Co 2p_1/2 (Co³⁺ at 794.5 eV, Co²⁺ at 796.63 eV) peaks, along with three satellite peaks (784.54 eV, 788.34 eV, and 803.13 eV).^43,44 Compared to Co₃O₄, the Co³⁺ 2p_3/2 XPS of RCO_350, RCO_400, and RCO_450 present the shift of ca. 0.49, ca. 0.34, and 0.22 eV to the low binding energy. In general, the shifts in Ru 3p and Co 2p XPS spectra indicate an interface interaction between RuO₂ and Co₃O₄ in the RCO_T catalysts. As the temperature increases, the magnitude of these shifts decreases, indicating a weakening of the electronic interaction between RuO₂ and Co₃O₄. We hypothesize that the increasing temperature enhances the particles of RuO₂ and Co₃O₄, reducing the interface interaction between them.

Overall, RCO_T catalysts synthesized at different temperatures exhibit small particle sizes, contact-rich interfaces, and interaction between RuO₂ and Co₃O₄. As temperature increases, the crystallization degrees of RuO₂ and Co₃O₄ enhance, reducing their interfacial interaction.

The OER performance of RCO_T catalysts

The OER performance of the RCO_T catalysts was evaluated in 0.10 M HClO₄ and compared with commercial RuO₂ and Co₃O₄. The iR-compensated LSVs in Fig. 2a demonstrate that the RCO_T catalysts exhibit significantly higher activity than the commercial RuO₂ and Co₃O₄. Specifically, the η at 10 mA cm⁻² for RCO_350, RCO_400, and RCO_450 are 240.6 ± 1.5, 253.3 ± 3.2, and 265.7 ± 4.0 mV, respectively, all of which are lower than those of the commercial RuO₂ and Co₃O₄ (Fig. 2b). Notably, the RCO_350 requires only 275.3 ± 2.1 and 293.7 ± 2.9 mV to reach 30 and 50 mA cm⁻², respectively, outperforming RCO_400 and RCO_450. It highlights the superior OER activity of RCO_350 among the RCO_T catalysts. The Tafel slopes of the RCO_T catalysts, as depicted in Fig. 2c, increase from 59.3 mV dec⁻¹ for RCO_350 to 67.2 mV dec⁻¹ for RCO_400 and 76.7 mV dec⁻¹ for RCO_450. Despite this trend, these values remain significantly lower than the Tafel slope of commercial RuO₂ (171.1 mV dec⁻¹), further confirming the favourable OER kinetics of the RCO_T catalysts, particularly RCO_350.

Fig. 2 OER performance of catalysts in 0.1 M HClO₄: (a) LSVs, (b) η at different current density, (c) Tafel plot, (d) MA, (e) C_dl values, and (f) Nyquist plots at 1.5 V vs. RHE. (g) Correlation between the SA and XPS shift of catalysts, (h) the RCO_T catalysts activity predicted from volcano plot, where the position of RuO₂ is from ref. 47 and 48.

MA is an essential evaluation metric for assessing the performance of catalysts. The MA of the RCO_T catalysts was measured at 1.5 V vs. RHE. As shown in Fig. 2d, RCO_350 exhibits the highest MA of 137.2 ± 1.2 A g_Ru⁻¹, surpassing RCO_400 (96.7 ± 3.1 A g_Ru⁻¹) and RCO_450 (68.1 ± 3.6 A g_Ru⁻¹). These values are significantly higher than commercial RuO₂, which recorded a much lower MA of 14.6 ± 1.0 A g_Ru⁻¹, underscoring the superior activity of the RCO_T catalysts. We comprehensively analyzed the C_dl, EIS, and SA for the catalysts to gain deeper insights into the factors contributing to the enhanced MA. Fig. 2e shows that the C_dl values for the RCO_350, RCO_400, and RCO_450 are 9.1, 4.3, and 3.0 mF cm⁻², respectively, significantly higher than that of commercial RuO₂ (0.61 mF cm⁻²). It indicates that the RCO_T catalysts possess a larger ECSA than commercial RuO₂. The observed decreasing trend in C_dl from RCO_350 to RCO_450 can be attributed to the particle growth in the elevated calcination temperature. The CVs utilized to calculate C_dl at various scanning rates are shown in Fig. S11.†Fig. 2f and S12† present the obtained Nyquist plots and equivalent circuits used for curve fitting during EIS analysis. The Nyquist diagrams reveal that the RCO_T catalysts exhibit notably lower charge transfer resistance (R_ct) than commercial RuO₂ and Co₃O₄, indicating the enhanced charge transfer capability.⁴⁵ Among the RCO_T catalysts, R_ct values increase with calcination temperature, possibly due to reduced interfacial interactions. Overall, enhanced MA in RCO_T catalysts originates from their increased ECSA and reduced R_ct compared to commercial RuO₂. The superior MA of RCO_350, relative to RCO_400 and RCO_450, can be attributed to its largest ECSA and lowest R_ct.

The SA of RCO_T catalysts was analyzed, as shown in Fig. 2g, with values of 0.112 ± 0.001 for RCO_350, 0.167 ± 0.005 for RCO_400, and 0.169 ± 0.009 mA cm⁻² for RCO_450, slightly lower than the commercial RuO₂ (0.178 ± 0.013 mA cm⁻²). The decreased SA for RCO_T catalysts is attributed to their electronic structures, where the Ru⁴⁺ 3p_3/2 binding energy upshifts demonstrate an inverse relationship with SA. The most pronounced positive shift of 0.87 eV in RCO_350 corresponds to the lowest SA value, while smaller shifts of 0.62 eV in RCO_400/450 result in SA values approaching (but still below) the commercial RuO₂. Based on the volcano plot used for predicting OER activity in Fig. 2h, with the difference between ΔG_O* and ΔG_HO* (ΔG_O* − ΔG_HO*) as a descriptor, (where O*and HO* are the reaction intermediates adsorbed on the active sites of the catalyst, G_O* and G_HO* are the free energies of the active site with adsorbed O* and HO*),^46–48 and our XPS analysis of the electronic structure of the RCO_T catalyst, we propose the following interpretation. The RCO_T catalysts exhibit an upward shift of the Ru d-band, enhancing the adsorption of intermediates. For the RCO_400/450 samples, this enhanced adsorption of intermediates induces a symmetrical shift of the ΔG_O* − ΔG_HO* values from the left leg to the right leg of the volcano plot. Nevertheless, their OER activity remains comparable to that of the commercial RuO₂. In contrast, the RCO_350 catalyst exhibits a more significant enhancement in intermediate adsorption, causing a pronounced deviation of ΔG_O* − ΔG_HO* value from the thermodynamic optimum. This deviation limits its catalytic activity.

We also conducted OER tests of RCO_300. Fig. S13a† shows that the η at 10 mA cm⁻² of RCO_300 is 307 mV, much higher than that of RCO_T catalysts at other temperatures. The CVs utilized to calculate C_dl at various scanning rates are shown in Fig. S13b.† Fig. S13c† shows that the C_dl value for the RCO_300 is 10.1 mF cm⁻². Meanwhile, in Fig. S13d,† its MA is 29.6 A g_Ru⁻¹, and its SA is 0.022 mA cm⁻². Combined with Fig. 1a, it can be seen that Ru in RCO_300 is in an unoxidized state or amorphous state with a complex structure, and its MA and SA are not better than those of other samples, so we will not give more explanation on it in this work.

In addition, stability is an essential indicator in evaluating the catalyst. Chronopotentiometry was conducted to assess the stability of the catalysts. Fig. 3a presents the chronopotentiometry results for RCO_T and commercial RuO₂ at 10 mA cm⁻². RCO_350, RCO_400, and RCO_450 maintain their performance for 36 h, 26 h, and 5 h, respectively, while the commercial RuO₂ stabilizes for less than 0.2 h. The RCO_T catalysts show improved stability than commercial RuO₂, possibly due to the presence of heterogeneous interfaces. Furthermore, as shown in Fig. S14,† RCO_350 outperforms several recently reported noble metal-based electrocatalysts in the acidic solution, surpassing some Ru and Ir-based systems. We further characterized the electrochemical properties and structure changes of RCO_350 after stability testing. As shown in Fig. 3b and S15,† the CV and C_dl of RCO_350 are significantly reduced, indicating a significant loss of the active sites in the RCO_350. Fig. S16† shows the SEM image of RCO_350 after the chronopotentiometry test. Although RCO_350 maintains its dodecahedron structure, the particles have grown larger. Tables S3 and S4† show the EDS data, showing that the atomic ratio of Co : Ru shifts from 2.9 : 1 to 9.6 : 1, further confirming the loss of active sites. Fig. 3c and d show that although the apparent activity of RCO_350 decreases significantly after the test, the SA increases compared to before. This indicates that the remaining sites still maintain high activity. The abrupt voltage change observed in RCO_400/450 may be attributed to the continuous loss of active sites in the catalysts during the chronopotentiometry test. Eventually, the remaining active sites are insufficient to sustain 10 mA cm⁻², causing the voltage to surge to force the catalyst to continue supplying the required current.

Fig. 3 (a) Electrochemical stability evaluation of the RCO_T catalysts and the commercial RuO₂ (the inset) by the chronopotentiometry test at 10 mA cm⁻², (b) CV, (c) LSV, (d) SA of RCO_350 before and after chronopotentiometry test.

The RCO_T catalysts exhibit improved MA and stability compared to commercial RuO₂, as shown in Scheme 2. The enhanced apparent activity can be attributed to the rich, heterogeneous interfaces offering larger ECSA and lower R_ct. However, the interfacial interaction between RuO₂ and Co₃O₄ is too strong, leading to over-enhanced adsorption of intermediates, which limits SA further improvement. On the other hand, this strong interaction likely contributes to the enhanced stability of the Ru⁴⁺, thereby improving their OER durability.

Scheme 2 Schematic illustration of the structure–activity relationship of RCO_T catalysts for OER.

PEMWE device performance

In addition to evaluating the performance within the RDE system, it is imperative to investigate the practical implementation of PEMWE. We employed RCO_350 as the anode catalyst for the OER, commercial Pt/C as the cathode catalyst for the HER, and the PEM Nafion® NR115 to assemble a PEMWE cell for testing purposes. Fig. 4a shows the LSVs of RCO_350 and commercial RuO₂ in PEMWE. It is worth noting that when the m_Ru of RCO_350 is 0.26 mg cm⁻², the voltage to reach 100 mA cm⁻² is 1.55 V, and when the m_Ru of commercial RuO₂ is 0.90 mg cm⁻², 1.61 V is required to achieve the same current density. The performance of the RCO_350 catalyst still outperforms the commercial RuO₂. In Fig. 4b, the Tafel slopes for RCO_350 and commercial RuO₂ are 160.0 mV dec⁻¹ and 133.8 mV dec⁻¹, respectively. The higher Tafel slope of RCO_350 suggests less favourable OER kinetics, likely due to suboptimal triple-phase boundary conditions within the catalyst layer. As illustrated in Fig. 4c and d, compared to dense commercial RuO₂ MEA, the RCO_350 catalyst layer exhibits a loose and porous structure attributed to the retention of its dodecahedral structures. This morphology may hinder effective contact between the catalyst particles, compromising electrochemical performance.^49,50

Fig. 4 Electrochemical performances of RCO_350 and commercial RuO₂ on PEMWE devices: (a) LSVs. (b) Tafel plot. SEM images of (c) RCO_350 MEA and (d) commercial RuO₂ MEA.

Conclusions

In summary, we developed RCO composite catalysts through temperature-controlled pyrolysis of Ru/ZIF-67 precursors. RCO catalysts synthesized at different temperatures exhibit small particle sizes, contact-rich interfaces, and interaction between RuO₂ and Co₃O₄. As the temperature increases, the particle size of the RCO catalysts grows while the interfacial interaction between RuO₂ and Co₃O₄ weakens. In RDE experiments, the MA and stability of the RCO catalysts are significantly improved. The RCO synthesized at low temperatures exhibits the highest MA due to its smallest particle size, largest ECSA, and lowest R_ct. The heterogeneous interfaces between RuO₂ and Co₃O₄ also contributed to the superior stability of the RCO catalysts compared to commercial RuO₂, although ruthenium dissolution remains a limiting factor for long-term performance. In MEA experiments, RCO with low Ru loading demonstrates superior activity to commercial RuO₂. However, the structure of the catalyst layer hinders the construction of the triple-phase boundary. These findings highlight the potential of RCO catalysts for efficient green hydrogen production while underscoring the need for further optimization to address structural challenges in PEMWE applications.

Data availability

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

Author contributions

Yixuan Yuan: writing—original draft, methodology, investigation, visualization. Chang Liu: investigation. Shanshan Liu: investigation. Ruimin Ding: writing—review & editing, supervision, funding acquisition. Xi Yin: conceptualization, methodology, writing, review & editing, supervision, funding acquisition.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support from the State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences is greatly appreciated. This study was financially supported by the National Key Research and Development Program of China (2021YFB4001203), the Fundamental Research Program of China Postdoctoral Science Foundation (Grant No. 2022M723263), the State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences (Grant No. SCJC-HN-2023-16, SCJC-HN-2023-17).

Notes and references

1. T. Reier, H. N. Nong, D. Teschner, R. Schlögl and P. Strasser, Adv. Energy Mater., 2017, 7, 1601275.
2. Z. W. She, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K. Norskov and T. F. Jaramillo, Science, 2017, 355, eaad4998.
3. J. A. Turner, Science, 2004, 305, 972–974.
4. P. K. Katkar, S. J. Marje, S. B. Kale, A. C. Lokhande, C. D. Lokhandea and U. M. Patil, CrystEngComm, 2019, 21, 884–893.
5. C. Spöeri, J. T. H. Kwan, A. Bonakdarpour, D. P. Wilkinson and P. Strasser, Angew. Chem., Int. Ed., 2017, 56, 5994–6021.
6. F. Y. Chen, Z. Y. Wu, Z. Adler and H. T. Wang, Joule, 2021, 5, 1704–1731.
7. L. A. King, M. A. Hubert, C. Capuano, J. Manco, N. Danilovic, E. Valle, T. R. Hellstern, K. Ayers and T. F. Jaramillo, Nat. Nanotechnol., 2019, 14, 1071–1074.
8. L. C. Seitz, C. F. Dickens, K. Nishio, Y. Hikita, J. Montoya, A. Doyle, C. Kirk, A. Vojvodic, H. Y. Hwang, J. K. Nørskov and T. F. Jaramillo, Science, 2016, 353, 1011–1014.
9. Y. Lin, Z. Tian, L. Zhang, J. Ma, Z. Jiang, B. J. Deibert, R. Ge and L. Chen, Nat. Commun., 2019, 10, 162.
10. Y.-R. Zheng, J. Vernieres, Z. Wang, K. Zhang, D. Hochfilzer, K. Krempl, T.-W. Liao, F. Presel, T. Altantzis, J. Fatermans, S. B. Scott, N. M. Secher, C. Moon, P. Liu, S. Bals, S. Van Aert, A. Cao, M. Anand, J. K. Nørskov, J. Kibsgaard and I. Chorkendorf, Nat. Energy, 2022, 7, 55–64.
11. L. Deng, S. F. Hung, Z. Y. Lin, Y. Zhang, C. Zhang, Y. Hao, S. Liu, C. H. Kuo, H. Y. Chen, J. Peng, J. Wang and S. Peng, Adv. Mater., 2023, 35, 2305939.
12. Z. Y. Wu, F. Y. Chen, B. Li, S. W. Yu, Y. Z. Finfrock, D. M. Meira, Q. Q. Yan, P. Zhu, M. X. Chen, T. W. Song, Z. Yin, H. W. Liang, S. Zhang, G. Wang and H. Wang, Nat. Mater., 2023, 22, 100–108.
13. Z. S. Li, B. L. Li, M. Yu, C. L. Yu and P. K. Shen, Int. J. Hydrogen Energy, 2022, 47, 26956–26977.
14. C. Wang, L. J. Jin, H. Y. Shang, H. Xu, Y. Shiraishi and Y. Du, Chin. Chem. Lett., 2021, 32, 2108–2116.
15. X. J. Cao, H. Y. Qin, J. Y. Zhang, X. J. Chen and L. F. Jiao, J. Am. Chem. Soc., 2024, 146, 32049–32058.
16. Z. H. Yin, H. Liu, J. Hu and J. J. Wang, Natl. Sci. Rev., 2024, 11, nwae362.
17. T. Y. Gao, D. X. Jiao, L. N. Wang, X. Ge, X. Wen, L. Zhang, L. R. Zheng, X. X. Zou, W. Zhang, W. T. Zheng, J. C. Fan and X. Q. Cui, J. Am. Chem. Soc., 2024, 147, 4159–4166.
18. N. Zhang and Y. Chai, Energy Environ. Sci., 2021, 14, 4647–4671.
19. L. An, C. Wei, M. Lu, H. W. Liu, Y. B. Chen, G. G. Scherer, A. C. Fisher, P. X. Xi, Z. C. J. Xu and C. H. Yan, Adv. Mater., 2021, 33, 2006328.
20. Y. Q. Sun, C. C. Wang, Y. Wang, Y. Xu, X. D. Yang, B. Ouyang and C. C. Li, Appl. Surf. Sci., 2024, 651, 159222.
21. L. X. Zhou, Y. F. Shao, F. Yin, J. Li, F. Y. Kang and R. T. Lv, Nat. Commun., 2023, 14, 7644.
22. S. Y. Chen, H. Liu, B. C. Yuan, W. H. Xu, A. Q. Cao, M. G. Sendeku, Y. P. Li, X. M. Sun and F. M. Wang, Nanoscale, 2024, 16, 20940–20947.
23. L. Q. Hou, Z. J. Li, H. Jang, Y. Wang, X. M. Cui, X. M. Gu, M. G. Kim, L. G. Feng, S. G. Liu and X. E. Liu, Adv. Energy Mater., 2023, 13, 2300177.
24. B. R. KC, D. Kumar and B. P. Bastakoti, J. Mater. Sci., 2024, 59, 10193–10206.
25. Y. Wu, R. Yao, K. Y. Zhang, Q. Zhao, J. P. Li and G. Liu, Chem. Eng. J., 2024, 479, 147939.
26. X. Z. Zhang, F. Wu, Q. J. Zhang, Z. Y. Lu, Y. Q. Zheng, Y. A. Zhu and Y. C. Lin, Inorg. Chem., 2024, 63, 8418–8425.
27. W. Q. Jia, X. J. Cao, X. J. Chen, H. Y. Qin, L. C. Miao, Q. L. Wang and L. F. Jiao, Small, 2024, 20, 2310464.
28. W. Y. Gou, S. S. Zhang, Y. C. Wang, X. H. Tan, L. Q. Liao, Z. N. Qi, M. Xie, Y. Y. Ma, Y. Q. Su and Y. Q. Qu, Energy Environ. Sci., 2024, 17, 6755–6765.
29. C. L. Rong, S. H. Wang, X. Shen, C. Jia, Q. Sun, Q. Zhang and C. Zhao, Energy Environ. Sci., 2024, 17, 4196–4204.
30. R. Y. Fan, H. J. Liu, J. K. Ren, Y. C. Li, J. Nan, Y. L. Zhou, C. Y. Liu, Y. M. Chai and B. Dong, ACS Sustainable Chem. Eng., 2024, 12, 2313–2323.
31. R. Gong, B. W. Liu, X. L. Wang, S. C. Du, Y. Xie, W. Q. Jia, X. X. Bian, Z. M. Chen and Z. Y. Ren, Small, 2024, 19, 2204889.
32. F. F. Shang, H. J. He, Y. Lin, B. An, H. R. Cai, X. Q. Li, W. T. Wang, C. Liang, S. C. Yang and B. Wang, Inorg. Chem. Front., 2024, 11, 5265–5272.
33. L. L. Liu, M. Zhang, T. T. Ji, J. H. Yan, Y. W. Sun, G. H. He and Y. Liu, Chem. Mater., 2021, 33, 7350–7356.
34. Y. R. Lin, J. D. Bao, S. T. Zhang, Y. Zhang, J. C. Qiu, Y. F. Li, Y. P. Quan, M. W. Yuan and G. B. Sun, J. Power Sources, 2024, 623, 235394.
35. J. Li, W. Li, G. Liu, Y. Deng, J. Yang and Y. Chen, Catal. Lett., 2016, 146, 1300–1308.
36. T. Audichon, B. Guenot, S. Baranton, M. Cretin, C. Lamy and C. Coutanceau, Appl. Catal., B, 2017, 218, 385–397.
37. M. Lim, Z. Ma, G. O'Connell, J. A. Yuwono, P. Kumar, R. Jalili, R. Amal, R. Daiyan and E. C. Lovell, Small, 2024, 20, 2401333.
38. F. H. Ye, S. S. Zhang, Q. Q. Cheng, Y. D. Long, D. Liu, R. Paul, Y. M. Fang, Y. Q. Su, L. T. Qu, L. M. Dai and C. G. Hu, Nat. Commun., 2023, 14, 2040.
39. K. H. Ye, K. S. Li, Y. R. Lu, Z. J. Guo, N. Ni, H. Liu, Y. C. Huang, H. B. Ji and P. S. Wang, TrAC, Trends Anal. Chem., 2019, 116, 102–108.
40. H. Tang, M. Kong, W. W. Cao, W. G. Ma, J. J. Ding, C. X. Shi, X. D. Yang, Q. S. Shen and Y. Q. Sun, CrystEngComm, 2023, 25, 2473–2478.
41. C. J. Liu, B. B. Sheng, Q. Zhou, D. F. Cao, H. H. Ding, S. M. Chen, P. J. Zhang, Y. J. Xia, X. J. Wu and L. Song, Nano Res., 2022, 15, 7008–7015.
42. J. Yang, K. An, Z. Yu, L. Qiao, Y. Cao, Y. Zhuang, C. Liu, L. Li, L. Peng and H. Pan, ACS Catal., 2024, 14, 17739–17747.
43. M. C. Biesinger, B. P. Payne, A. P. Grosvenor, L. W. M. Lau, A. R. Gerson and R. S. C. Smart, Appl. Surf. Sci., 2011, 257, 2717–2730.
44. Y. M. Zhu, J. A. Wang, T. Koketsu, M. Kroschel, J. M. Chen, S. Y. Hsu, G. Henkelman, Z. W. Hu, P. Strasser and J. W. Ma, Nat. Commun., 2022, 13, 7754.
45. Y. X. Yao, M. F. Pei, C. Su, X. Jin, Y. P. Qu, Z. H. Song, W. Y. Jiang, X. G. Jian and F. Y. Hu, Small, 2024, 20, 2401481.
46. J. Song, C. Wei, Z. F. Huang, C. Liu, L. Zeng, X. Wang and Z. J. Xu, Chem. Soc. Rev., 2020, 49, 2196–2214.
47. I. C. Man, H.-Y. Su, F. Calle-Vallejo, H. A. Hansen, J. I. Martínez, N. G. Inoglu, J. Kitchin, T. F. Jaramillo, J. K. Nørskov and J. Rossmeisl, ChemCatChem, 2011, 3, 1159–1165.
48. H. Chen, L. Shi, X. Liang, L. Wang, T. Asefa and X. Zou, Angew. Chem., Int. Ed., 2020, 59, 19654–19658.
49. J. Kibsgaard and I. Chorkendorff, Nat. Energy, 2019, 4, 430–433.
50. T. T. Phan, S.-K. Kim, J. Islam, M.-J. Kim and J.-H. Lee, Int. J. Hydrogen Energy, 2024, 49, 875–885.
51. C. C. McCrory, S. Jung, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2013, 135, 16977–16987.
52. H. Wu, J. Chang, J. Yu, S. Wang, Z. Hu, G. I. N. Waterhouse, X. Yong, Z. Tang, J. Chang and S. Lu, Nat. Commun., 2024, 15, 10315.
53. W. X. Zheng, X. X. Cheng, P. P. Chen, L. L. Wang, Y. Duan, G. J. Feng, X. R. Wang, J. J. Li, C. Zhang, Z. Y. Yu and T. B. Lu, Nat. Commun., 2025, 16, 337.
54. X. Wang, Z. Li, H. Jang, C. Chen, S. Liu, L. Wang, M. G. Kim, J. Cho, Q. Qin and X. Liu, Adv. Energy Mater., 2025, 15, 2403388.
55. Z. L. Li, H. Y. Sheng, Y. C. Lin, H. Y. Hu, H. F. Sun, Y. Dong, X. W. Chen, L. Z. Wei, Z. Q. Tian, Q. W. Chen, J. W. Su and L. Chen, Adv. Funct. Mater., 2024, 34, 2409714.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ce00474h

---