An IrRuO_x catalyst supported by oxygen-vacant Ta oxide for the oxygen evolution reaction and proton exchange membrane water electrolysis

Abstract

The sustainable development of proton exchange membrane water electrolysis (PEMWE) requires a dramatic reduction in Ir while maintaining good catalytic activity and stability for the oxygen evolution reaction (OER). Herein, high-surface-area Ta₂O₅ with abundant oxygen vacancies is synthesized via a facile process, followed by anchoring IrRuO_x onto a Ta₂O₅ support (IrRuO_x/Ta₂O₅). IrRuO_x and Ta₂O₅ work synergistically to afford excellent catalytic performance for the acidic OER. At 0.3 mg_Ir cm⁻², IrRuO_x/Ta₂O₅ only needed an overpotential of 235 mV to deliver 10 mA cm⁻² in an acidic half cell and needed a cell potential of 1.91 V to deliver 2 A cm⁻² in a PEM water electrolyzer. The characterization results show that doping Ir into RuO_x significantly improves the stability and the electrochemically active surface area of RuO_x. In IrRuO_x/Ta₂O₅, IrRuO_x interacts with Ta₂O₅ through more electron-rich Ir, indicating strong synergy between the catalyst and the support. The use of a metal oxide support improves the catalyst dispersion, optimizes electronic structures, facilitates mass transport, and stabilizes active sites. This work demonstrates that compositing Ir with less expensive Ru and anchoring catalyst nanoparticles on platinum-group metal (PGM)-free metal oxide supports represents one of the most promising strategies to reduce Ir loading and achieve an activity–stability trade-off. Such a strategy can benefit future catalyst design for other energy storage and conventional processes.

---

Hao Wang

Dr Hao Wang received his Ph.D. degree from the Beijing University of Chemical Technology in 2019. He was a guest graduate student at Argonne National Laboratory and a postdoctoral researcher at National Renewable Energy Laboratory. He is now a staff scientist in the Institute of Process Engineering, Chinese Academy of Sciences. His expertise includes rational design of electrocatalysts and membrane electrode assemblies in hydrogen fuel cells and water electrolyzers. He has published more than 30 papers in international journals such as Adv. Mater., ACS Energy Lett., Appl. Catal. B, Adv. Funct. Mater., etc.

---

1. Introduction

The development of hydrogen energy is critical for becoming a sustainable society.¹ As a raw material and an energy carrier, hydrogen can be used to decarbonize many sectors, such as industry, transportation, and power.² In hydrogen-driven scenarios, hydrogen production technology lays the foundation for determining whether hydrogen energy can be implemented at large scales.³ Water electrolysis driven by renewable electricity is promising due to its high efficiency and emission-free nature.⁴ In particular, proton exchange membrane water electrolysis (PEMWE) can produce highly pure hydrogen at a high current density and can work easily with intermittent renewable electricity.⁵ Unfortunately, the anodic oxygen evolution reaction (OER) of PEMWEs is kinetically sluggish and requires a highly oxidative and acidic environment.⁶ The current catalyst of choice uses a high loading of Ir, which has a lower crustal reserve than Pt and causes an extremely high cost hurdle for commercialization.^7,8 Therefore, reducing the Ir loading while maintaining the OER activity and stability is necessary for the sustainable development of PEMWEs.^9–11

One popular strategy for developing low-Ir catalysts is alloying with Ru. Although Ru is also a platinum-group metal (PGM), it is much more affordable than Ir. In the acidic OER, Ru is more active but less stable than Ir;¹² thus, precisely controlling the IrRu composition can achieve an activity–stability trade-off.^13–15 However, carefully designed alloy nanoparticles always agglomerate into large clusters whose surface is unable to fully participate in the OER. To overcome this problem, catalyst supports are often introduced to better disperse nanoparticles and generate catalyst–support interactions to stabilize the active sites.^16,17 In particular, the PGM-free support in the anode will significantly reduce the PGM loading, thus improving the competitiveness of the PEMWE cost. Unfortunately, the harsh conditions of the acidic OER make it impossible to use conventional PGM-free supports with good electronic conductivity, such as carbonaceous materials and transition metals.¹⁸ These supports will be oxidized or dissolved under working conditions. The dissolved transition metal ions penetrate the membrane under an electric field and block the proton transfer path. Therefore, finding stable and conductive PGM-free supports is quite meaningful but challenging.^19,20

A few insightful studies have used Ti-, W-, and Sn-based oxides as PGM-free supports.^21–24 These oxides are relatively stable in acids, and the main idea behind using these supports is to construct metal–metal oxide interactions to modify the electronic structures and optimize binding energies with the oxygen intermediates.^25,26 Ta oxides are also attractive candidates for exploration due to their tunable electronic structures and excellent stability against corrosion and dissolution.²⁷ Recent studies demonstrated that oxygen-induced Ta₂O₅ ²⁸ and Fe-doped Ta₂O₅ ²⁹ have good OER activity in alkaline media, suggesting that they are active toward water oxidation. In terms of catalyst supports for the acidic OER, Xu and co-workers³⁰ designed an IrO₂–Ta₂O₅ anode and found that different calcination temperatures had a great impact on the exposed crystal plane of IrO₂. Amano and co-workers³¹ coated Ti felt with amorphous IrO₂–Ta₂O₅ layers and found that the acidic OER activity was dependent on the calcination temperature. In a more recent study, Pak and co-workers³² supported Ir with mesoporous Ta₂O₅ to improve Ir utilization, electronic conductivity, and the electrochemically active surface area. Despite these insightful studies, Ir supported by Ta oxides for the acidic OER still requires an overpotential of 300 mV at 10 mA cm⁻² in half cells, and the loading of Ir is still higher than 0.5 mg_Ir cm⁻² to achieve satisfactory single cell performance. Therefore, the potential of the Ta₂O₅ support to reduce Ir loading and achieve high catalytic activity has not yet been fully demonstrated.

Herein, we composited IrRuO_x with Ta₂O₅ to reduce Ir loading and further improved the OER and PEMWE performance. The Ta₂O₅ support developed in this work was confirmed to be amorphous with a high surface area and a high concentration of oxygen vacancies, and such advantageous structures resulted in uniformly distributed IrRuO_x with intimate contact. In the active IrRuO_x phase, Ir atoms were doped into RuO_x and coupled with Ta to form strong interactions. As tested in half cells, at 0.03 mg_Ir cm⁻², the catalyst needed only a 284 mV overpotential to deliver 10 mA cm⁻². By improving to 0.3 mg_Ir cm⁻², IrRuO_x only needed an overpotential of 235 mV to deliver 10 mA cm⁻² in a half cell and needed a potential of 1.91 V to deliver 2 A cm⁻² in a PEM water electrolyzer. The good dispersion of the catalyst with high utilization, the close contact between IrRuO_x and Ta₂O₅, and the modified electronic structures of the active sites are responsible for the enhanced electrochemical performance.

2. Experimental

To synthesize the samples studied in this work, dihydrogen hexachloroiridate(IV) hydrate (H₂IrCl₆·6H₂O, Ir ≥39%) was purchased from Sigma-Aldrich. Tantalum ethoxide (C₁₀H₂₅O₅Ta, 99.99%), ruthenium(III) chloride (RuCl₃, 99.5%, Ru: 47%), sodium borohydride (NaBH₄, ≥95%), ethylene glycol (C₂H₆O₂, >99.5), and tantalum oxide (Ta₂O₅, 99.99%) were purchased from Adamas-Beta. Ammonium fluoride (NH₄F, ≥98.0%) was purchased from Greagent. All chemicals were used as received, without further purification.

2.1 Synthesis of Ta₂O₅

To prepare Ta₂O₅, 0.3 g of tantalum ethoxide was dissolved in 6 mL of 0.1 M NH₄F in ethylene glycol. The resulting solution was mixed with 6 mL of water in an inner Teflon tube under rapid stirring, and the inner Teflon tube was sealed in a stainless-steel autoclave and placed in an oven. The oven was heated to 220 °C, maintained for 20 h, and naturally cooled to room temperature. The precipitate was collected by centrifugation and washed with water 3 times and then with ethanol 3 times to remove the unreacted residues. Ta₂O₅ powder was finally obtained after vacuum drying at 60 °C for 10 h.

2.2 Synthesis of IrRuO_x/Ta₂O₅

To composite IrRuO_x, Ta₂O₅, RuCl₃ hydrate, and H₃IrCl₆ with a mass ratio of 1/2.07/0.44 were dissolved in 10 mL of water to form solution A. In a subsequent step, 20 mg of NaBH₄ was dissolved in 6 mL of ice-cold water to form solution B. Solution B was then added to solution A, and the resulting mixture was magnetically stirred for 6 h at room temperature. After the reaction, the solid sample was collected by centrifugation and washed with water 3 times and then with ethanol 3 times, followed by vacuum drying at 60 °C for 10 h. The IrRuO_x/Ta₂O₅ catalyst was finally obtained after heat treatment in a tube furnace, which was heated to 350 °C and maintained for 2 h under an air atmosphere. The overall process of catalyst synthesis is schematically shown in Scheme 1. For comparison, IrO_x/Ta₂O₅ and RuO_x/Ta₂O₅ were also prepared following the same synthesis procedure except for the addition of only Ru and Ir precursors, respectively. To study the PGM loadings and their impact on the acidic OER, 0.5IrRuO_x/Ta₂O₅ (half Ir and Ru precursors added) and 1.5IrRuO_x/Ta₂O₅ (1.5-fold Ir and Ru precursors added) were also prepared following the same synthesis procedure. The sample-C used in this work refers to a commercially available sample.

Scheme 1 Synthesis procedure for Ta₂O₅ and IrRuO_x/Ta₂O₅.

2.3 Materials characterization

The crystalline structures of the studied samples were examined using an X-ray diffractometer (XRD, Rigaku Co. Ltd, Japan) employing Cu Kα radiation (λ = 1.54056 Å) operated at 40 kV. A JEM-2100F high-resolution field-emission transmission electron microscope (FETEM, JEOL, Japan) and a SIGMA 500/VP scanning electron microscope (Zeiss, Germany) were used to observe the morphology of the samples. X-ray photoelectron spectroscopy (XPS) measurements were conducted on an EscaLab 250Xi spectrometer (Thermo Fisher Scientific, USA), and the spectra were processed by calibrating C 1s to 284.8 eV. The specific surface area was measured by N₂ adsorption–desorption at 77 K with an ASAP 2460 gas sorption analyzer (Micromeritics, USA). Bulk oxygen vacancy signals were detected by low temperature electron paramagnetic resonance (EPR, Bruker, EMXplus-6/1, Germany) under liquid nitrogen conditions, and the sample dosage was approximately 10 mg.

The experimental details for synchrotron and electrochemical characterization are available in the ESI.†

3. Results and discussion

3.1 Physical characterization of the catalysts

The crystalline structures of the prepared samples were determined by XRD. According to Fig. 1A, the diffraction peaks of the Ta sample developed in this work are indexed to Ta₂O₅ (JCPDS, PDF no. 08-0225). However, compared with commercial Ta₂O₅ powder (Ta₂O₅-C), Ta₂O₅ has broader diffraction peaks. After integration with the IrRuO_x active species (Fig. 1B), all the Ta₂O₅ peaks were retained, with several new broad peaks emerging at (2θ=) 28.6°, 35.5°, and 54.5°. While IrO₂ (JCPDS, PDF no. 43-1019) and RuO₂ (JCPDS, PDF no. 43-1027) show similar XRD peak positions, the new peaks on Ta₂O₅ are closer to those of RuO₂, indicating that Ir is doped into the RuO_x phase.

Fig. 1 (A) Comparison of XRD patterns between Ta₂O₅ and Ta₂O₅-C. (B) Comparison of XRD patterns between Ta₂O₅, IrO_x/Ta₂O₅, RuO_x/Ta₂O₅, and IrRuO_x/Ta₂O₅. (C) TEM image and (c) HRTEM image of Ta₂O₅. (D) TEM image and (d) HRTEM image of IrRuO_x/Ta₂O₅.

The morphologies of the studied samples were observed by TEM. The Ta₂O₅ sample consists of sphere-like particles agglomerated into large clusters (Fig. 1C and Fig. S1, see the ESI†). There are no other notable features on its smooth surface. HRTEM further revealed the (210) lattice plane of Ta₂O₅ with a lattice spacing of 0.233 nm (Fig. 1c). After introducing the IrRuO_x active species, IrRuO_x/Ta₂O₅ showed a bumpy surface, which was caused by IrRuO_x growth on the Ta₂O₅ surface (Fig. 1D). However, no individual well-defined IrRuO_x nanoparticles were observed, suggesting a uniform distribution of IrRuO_x on the Ta₂O₅ substrate. HRTEM revealed a lattice spacing of 0.325 nm, which is assigned to the (110) lattice plane of RuO₂ but is larger than the standard value of RuO₂ (Fig. 1d). This was caused by Ir doping, which slightly expanded the lattice. To determine how the elements were distributed within the IrRuO_x/Ta₂O₅ sample, EDS was performed, and the resulting elemental mapping images are shown in Fig. 2. According to the images, Ta, Ir, Ru, and O are uniformly distributed, and IrRuO_x and Ta₂O₅ are undistinguishable, which is evidence of intimate contact and synergy. Uniform element distributions were also observed for RuO_x/Ta₂O₅ (Fig. S2, see the ESI†) and IrO_x/Ta₂O₅ (Fig. S3, see the ESI†). The atomic ratio of O/Ru/Ta/Ir in IrRuO_x/Ta₂O₅ was determined to be 69.68/16.45/11.75/2.12 by EDS. This ratio was used to determine the catalyst loading on the electrodes.

Fig. 2 (A) TEM image of the selected area to perform EDS. (B) Element mapping image with all elements overlapped. Individual element mapping of (a) Ta, (b) Ir, (c) Ru, and (d) O.

To assess the surface areas and pore structures, the studied samples were tested at 77 K for N₂ adsorption–desorption behavior. The resulting isotherms are displayed in Fig. 3A, and the parameters are summarized in Table S1 (see the ESI†). Notably, Ta₂O₅-C has little N₂ adsorption, and its BET surface area was determined to be extremely low at 2.92 m² g⁻¹. By using the method developed in this work, the as-prepared Ta₂O₅ showed a much greater N₂ uptake, confirming its porous nature, and its BET surface area significantly improved to 231.56 m² g⁻¹. Such a high surface area is favorable for Ta₂O₅ to serve as a catalyst support to confine the catalyst size and improve the catalyst dispersion. After integration with IrRuO_x, the IrRuO_x/Ta₂O₅ sample shows a decreased BET surface area of 98.18 m² g⁻¹, suggesting that the IrRuO_x nanoparticles are nonporous. Overall, owing to the highly porous Ta₂O₅ support, IrRuO_x/Ta₂O₅ still possesses a high surface area as an anode catalyst. To determine the pore structures in the studied samples, pore size distributions were plotted as shown in Fig. 3B. Unsurprisingly, Ta₂O₅-C has no pores. Ta₂O₅ and IrRuO_x/Ta₂O₅ have pores primarily distributed in widths ranging from 2 to 20 nm and 2 to 7 nm, respectively, confirming that they are both mesoporous. The pore volumes of Ta₂O₅ and IrRuO_x/Ta₂O₅ were determined to be 0.097 cm³ g⁻¹ and 0.043 cm³ g⁻¹, respectively (Table S1, see the ESI†). IrRuO_x/Ta₂O₅ also shows a small number of macropores with pore widths greater than 50 nm, which results from the stacking of the agglomerates. The abundant hierarchical pores of IrRuO_x/Ta₂O₅ will facilitate mass transport during the PEMWE.³³

Fig. 3 (A) N₂ adsorption–desorption isotherms and (B) the resulting pore size distributions for the studied samples. (C) EPR spectra of the studied samples. (D) XPS full spectrum, (E) deconvoluted Ta 4f, (F) deconvoluted Ir 4f, (G) deconvoluted Ru 3p, and (H) deconvoluted O 1s spectra for the studied samples.

To gain further insight into the as-prepared Ta₂O₅ structure, EPR was conducted to examine the Ta–O coordination environment and the unpaired electrons. Ta₂O₅-C was also tested as a reference. As shown in Fig. 3C, Ta₂O₅-C shows a flight curve without any EPR response. However, Ta₂O₅ synthesized in this work has a high signal density, suggesting the existence of a high concentration of oxygen vacancies in Ta₂O₅. Abundant oxygen vacancies are usually favorable for catalyst growth and often result in interactions between the catalyst and support. Additionally, Ta oxides with oxygen vacancies are more electronically conductive than regular oxygen-saturated Ta₂O₅.³⁴ After compositing with IrRuO_x, the signal intensity decreased, but vacancies were still detectable in IrRuO_x/Ta₂O₅. Since Ta₂O₅ was fully covered by IrRuO_x and underwent air treatment, the remaining oxygen vacancies also originated from IrRuO_x clusters due to their high surface-to-volume ratios.³⁵

The surface compositions of the studied samples were examined by XPS, and Ta, Ir, Ru, and O were all detected in the full spectrum of IrRuO_x/Ta₂O₅ (Fig. 3D). The atomic composition obtained by XPS is in good agreement with that obtained by EDS (Table S2, see the ESI†). Fig. 3E shows the deconvoluted Ta 4f spectra and the fitted results for Ta₂O₅-C, Ta₂O₅, and IrRuO_x/Ta₂O₅. The Ta 4f spectra for all three samples can be fitted into two peaks, with the one at a lower binding energy assigned to Ta 4f_7/2 and the other to Ta 4f_5/2.³⁶ Compared to Ta₂O₅-C, Ta₂O₅ developed in this work showed lower Ta 4f binding energies, indicating a lower Ta valence in Ta₂O₅, and the abundant oxygen vacancies in Ta₂O₅ were responsible for these results, as confirmed by the EPR experiments. After integration with IrRuO_x, the Ta 4f spectra of IrRuO_x/Ta₂O₅ retained similar features but showed slightly higher binding energies than those of Ta₂O₅. This is an indication of interactions between the PGM catalyst and the Ta₂O₅ support. Fig. 3F shows the deconvoluted Ir 4f spectra and the fitted results for IrRuO_x/Ta₂O₅ and commercial IrO₂ (IrO₂-C). Even though IrRuO_x/Ta₂O₅ was fully oxidized in air during its preparation, the Ir 4f spectra of IrRuO_x/Ta₂O₅ showed slightly lower binding energies compared to those of Ta₂O₅, confirming the strong interactions between the PGM catalyst and the Ta₂O₅ support. It can be concluded that Ir active sites were more electron-rich compared to IrO₂-C.^32,35 Ir with a lower valence state usually shows better sustainability with higher stability.³⁷Fig. 3G shows the deconvoluted Ru 3p spectra and the fitted results for IrRuO_x/Ta₂O₅ and commercial RuO₂ (RuO₂-C). These two samples show similar peaks, indicating similar Ru valence states. No binding energy shifts were observed, which was not expected from the Ir results, indicating that interactions mainly occur between Ta and Ir. Fig. S4 (see the ESI†) shows a comparison of the Ru 3d spectra and the fitting results for IrRuO_x/Ta₂O₅ and RuO_x/Ta₂O₅. In the figure, Ru 3d_5/2, Ru 3d_3/2, and their satellite peaks were observed. The other three peaks are related to carbon species. However, the Ru 3d_5/2 and Ru 3d_3/2 binding energies of IrRuO_x/Ta₂O₅ were positively shifted (∼0.1 eV) compared to those of RuO_x/Ta₂O₅, suggesting that the Ru in IrRuO_x/Ta₂O₅ had a higher oxidation state. Previous studies have indicated that a higher valence state of Ru is more favorable for the OER.^38,39

Since the oxygen electronic structures are critical in this study, we further fitted the deconvoluted O 1s spectra as shown in Fig. 3H, and the resulting parameters are summarized in Table S3 (see the ESI†). To understand the differences between the individual components, Ta₂O₅-C, IrO₂-C, and RuO₂-C were also tested as references. The O 1s spectra of Ta₂O₅-C can be fitted by four peaks centered at 530.0 eV, 531.1 eV, 532.1 eV, and 533.2 eV, which are assigned to lattice O, vacant O, water, and adventitious species, respectively. For the Ta₂O₅ sample developed in this work, the four peaks still exist with similar positions but different ratios compared to Ta₂O₅-C. One noticeable feature is that the peak at 531 eV becomes dominant in Ta₂O₅, which indicates a high concentration of oxygen vacancies on its surface and reconfirms the EPR results. After integration with IrRuO_x, the O 1s spectra of IrRuO_x/Ta₂O₅-C can be fitted into five peaks, centered at 529.1 eV, 530.2 eV, 531.0 eV, 532.0 eV, and 533.3 eV. By comparison with the peaks for IrO₂-C and RuO₂-C, it can be concluded that the peak at 529.1 eV in IrRuO_x/Ta₂O₅-C is assigned to the lattice O of RuO₂; the peak at 530.2 eV contains the lattice O of IrO₂, Ta₂O₅, and the hydroxyl species on Ru; and the peak at 531.0 eV still refers to vacant O, as confirmed by EPR. The aforementioned results indicate strong interactions between Ta and Ir, which could be due to the similar binding energies of lattice O in IrO₂ and Ta₂O₅.

To better understand how the loaded active species impacted the Ta₂O₅ support, we conducted synchrotron X-ray absorption characterization of IrRuO_x/Ta₂O₅. The standard Ta₂O₅ power sample and the Ta foil were also tested as references. Fig. 4A displays the X-ray absorption near edge structure (XANES) results to show the Ta-L3 edge of the samples. The peak intensity of the white line in the IrRuO_x/Ta₂O₅ catalyst is between Ta (Ta⁰) and Ta₂O₅ (Ta^V), indicating that the average Ta oxidation state of the catalyst is between Ta⁰ and Ta^V but closer to Ta^V. This observation agrees well with the XPS results. The Fourier transform extended X-ray absorption fine structure (EXAFS) spectra are shown in Fig. 4B–D, from which the local coordination environments of Ta can be interpreted. The bond length of Ta–O is 1.53 Å in the standard Ta₂O₅ sample. However, that length in IrRuO_x/Ta₂O₅ was determined to be 1.46 Å. Similar results were also observed from the wavelet transform EXAFS as shown in Fig. 4E–G. While the coordination number of Ta–O in IrRuO_x/Ta₂O₅ was close to 6 and similar to that in the Ta₂O₅ standard sample, the difference in the numbers of Ta–O in the first shell and the second shell still suggested different coordination configurations (Table S4, see the ESI†). Therefore, the decrease in the Ta–O bond length in IrRuO_x/Ta₂O₅ was mainly due to the interactions between Ta₂O₅ and IrRuO_x.⁴⁰

Fig. 4 (A) Ta L3-edge XANES spectra. FT-EXAFS spectra of (B) IrRuO_x/Ta₂O₅, (C) standard Ta₂O₅, and (D) Ta foil. WT-EXAFS spectra of (E) IrRuO_x/Ta₂O₅, (F) standard Ta₂O₅, and (G) Ta foil.

3.2 Electrochemical evaluation of the catalysts

To evaluate the OER catalytic responses of the studied samples, a mixture of the catalyst and ionomer was deposited onto the electrode surface with the catalyst loading precisely controlled. Linear sweep voltammetry (LSV) was first conducted for the electrodes with a total catalyst loading of 0.4 mg cm⁻² in 0.5 M H₂SO₄ (Fig. 5A). Ta₂O₅ showed negligible OER activity but displayed a slight current improvement after integration with IrO_x. However, after integrating with RuO_x, the RuO_x/Ta₂O₅ sample had an excellent OER catalytic response and only needed an overpotential of 326 mV to deliver 10 mA cm⁻². After integration with both Ir and Ru, the OER catalytic activity of IrRuO_x/Ta₂O₅ significantly increased, and the overpotential decreased to 284 mV at 10 mA cm⁻². To understand whether the catalyst loading had a great impact on the catalytic performance, all catalysts with a total catalyst loading of 4 mg cm⁻² were also evaluated under the same test conditions. Even at a high total catalyst loading of 4 mg cm⁻², the Ir loading of IrRuO_x/Ta₂O₅ was still fairly low (ca. 0.3 mg_Ir cm⁻²). According to Fig. 5B, increasing catalyst loading did not help the Ta₂O₅ and IrO_x/Ta₂O₅ samples but further improved the OER catalytic activity of RuO_x/Ta₂O₅ and IrRuO_x/Ta₂O₅ with overpotentials of 240 mV and 235 mV at 10 mA cm⁻², respectively. We also adjusted the catalyst-to-support ratios to understand their impact on the catalytic performance of the OER. The results showed that IrRuO_x/Ta₂O₅ was optimal, and further increasing or decreasing the ratio suppressed the OER activity (Fig. 5C). To understand the OER kinetics, Tafel curves were plotted as shown in Fig. 5D. The Tafel slopes of IrO_x/Ta₂O₅, RuO_x/Ta₂O₅, and IrRuO_x/Ta₂O₅ were determined to be 94.2, 47.8, and 32.3 mV dec⁻¹, respectively, following the same trend as the OER voltammograms. With increasing current, IrRuO_x/Ta₂O₅, which had the smallest Tafel slope, had the smallest overpotential, confirming its excellent OER catalytic activity.

Fig. 5 OER LSV curves for the samples tested at (A) 0.4 mg cm⁻² and (B) 4 mg cm⁻², and (C) different PGM/Ta ratios at 4 mg cm⁻² in 0.5 M H₂SO₄. (D) Tafel plots derived from OER voltammograms recorded at 4 mg cm⁻². (E) EIS curves of the studied samples at 4 mg cm⁻². (F) Chronopotentiometric curves of 4 mg cm⁻² IrRuO_x/Ta₂O₅ RuO_x/Ta₂O₅ at 10 A cm⁻² for 90 h.

The electronic conductivity of the catalyst is critical for maximizing the participation of active sites in the OER. Therefore, the four studied samples were tested by electrochemical impedance spectroscopy (EIS), and the resulting curves are displayed in Fig. 5E. All four curves showed a similar starting point, which refers to the solution and connection resistance, indicating the reliability of the electrochemical testing system. The diameter of the semicircle in the high-frequency zone refers to the charge-transfer resistance of the catalyst. The diameter was the largest for both Ta₂O₅ and IrO_x/Ta₂O₅, decreased for RuO_x/Ta₂O₅, and the smallest for IrRuO_x/Ta₂O₅. The results suggested that IrRuO_x/Ta₂O₅ possessed the best electronic conductivity, which explained its excellent OER activity and fast kinetics. Adding RuO_x or IrRuO_x actually decreased the electron transfer resistance of Ta₂O₅, suggesting that the catalyst was more charge conductive than the support. This also explains why the catalyst and support required an optimal ratio to afford the best catalytic activity (Fig. 5C). To understand how the loaded catalyst impacted the Ta₂O₅ double layer capacitance (C_DL), CV curves were recorded at different scanning speeds, and the results are shown in Fig. S5 (see the ESI†). The C_DL of Ta₂O₅ was determined to be 4.8 mF cm⁻², which improved to 56.3 mF cm⁻² after integration with IrRuO_x, which explains the enhanced OER catalytic performance of IrRuO_x/Ta₂O₅. To understand the OER stability of the samples with and without Ir, chronopotentiometry was conducted at 10 mA cm⁻² for RuO_x/Ta₂O₅ and IrRuO_x/Ta₂O₅. According to Fig. 5F, IrRuO_x/Ta₂O₅ maintained stable performance for up to 90 h, and RuO_x/Ta₂O₅ showed a slightly greater performance decay. Both catalysts have reasonable stabilities, benefiting from catalyst–support interactions, in half cells. The IrRuO_x/Ta₂O₅ catalyst developed in this work is among the best-performing Ta-supported catalysts reported in the literature as shown in Table S5 (see the ESI†).

The top two samples of IrRuO_x/Ta₂O₅ and RuO_x/Ta₂O₅ in half-cells were further evaluated in a PEM water electrolyzer. CCM was prepared and assembled with a cathodic gas diffusion layer and an anodic Ti felt porous transport layer, as schematically shown in Fig. 6A, and a photograph of CCM with gaskets is shown in Fig. S6 (see the ESI†). The catalyst loading for IrRuO_x/Ta₂O₅ was 4 mg cm⁻² (ca. 0.3 mg_Ir cm⁻²); and for 60 wt% Pt/C, the loading was 0.2 mg_Pt cm⁻². The V–I polarization curves were first recorded as shown in Fig. 6B. An electrolyzer with the IrRuO_x/Ta₂O₅ catalyst only needed 1.71 V to deliver 1 A cm⁻² and 1.91 V to deliver 2 A cm⁻². However, the electrolyzer with the RuO_x/Ta₂O₅ catalyst needed higher cell potentials to deliver 1 A cm⁻² (1.72 V) and 2 A cm⁻² (1.94 V). Therefore, RuO_x/Ta₂O₅ showed less activity than IrRuO_x/Ta₂O₅, following the trend observed in half-cell tests. By comparing the EIS curves (Fig. 6C), it was found that both samples have a similar starting point and diameter of the semicircle, indicating a similar ohmic resistance in MEAs. The double layer capacitance between the catalyst and ionomer in the catalyst layer is important for enabling the participation of active sites in the OER. To evaluate this, cyclic voltammetry (CV) was conducted for both MEAs, and the resulting voltammograms in Fig. 6D indicate that the doping of Ir in RuO_x improved the electrochemically active surface area. The Ir dopants also improved the stability of the catalyst in MEA. Although RuO_x/Ta₂O₅ showed decent stability in a half cell at 10 mA cm⁻², it showed a fast performance decay when tested in MEA at 500 mA cm⁻² (Fig. 6E).

Fig. 6 (A) Scheme of a PEM water electrolyzer studied in this work. (B) Polarization curves, (C) EIS curves, (D) CV curves, and (E) chronopotentiometric curves at 500 mA cm⁻² for IrRuO_x/Ta₂O₅ and RuO_x/Ta₂O_{5m^{4 mg cm−2}}.

4. Conclusion

In summary, a Ta₂O₅ support with an amorphous structure, a high surface area, and a high concentration of surface oxygen vacancies was synthesized by a facile method. After integration with the IrRuO_x catalyst, strong interactions were found and they mainly occurred between Ir and Ta, and Ir became more electron-rich. In the IrRuO_x active species, the doping of Ir in RuO_x improved the electrochemically active surface area and stability of RuO_x; therefore, IrRuO_x/Ta₂O₅ outperformed IrO_x/Ta₂O₅ and RuO_x/Ta₂O₅ in catalytic activity and stability. All the components in IrRuO_x/Ta₂O₅ worked synergistically to catalyze the OER so that a high loading of Ir was no longer needed. At 0.3 mg_Ir cm⁻², IrRuO_x/Ta₂O₅ only needed a 235 mV overpotential at 10 mA cm⁻² in a half cell and a 1.91 V cell potential at 2 A cm⁻² in a PEM water electrolyzer. This work opens up a new avenue for designing PEMWE catalysts supported by PGM-free metal oxides to reduce Ir loading while offering excellent activity and stability.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 22308356), the Key R&D Program of Henan Province (no. 231111241800), Henan Provincial Science and Technology R&D Program Joint Fund (Advantageous Discipline Cultivation, no. 222301420045), Zhongke Technology Achievement Transfer and Transformation Center of Henan Province (no. 2024147), the Young Elite Scientists Sponsorship Program by CAST (no. 2021QNRC001), and the State Key Laboratory of Catalytic Materials and Reaction Engineering (RIPP, SINOPEC). The authors sincerely thank Prof. Suojiang Zhang (IPE, CAS) for his careful academic guidance and great support.

References

1. D. Guan, B. Wang, J. Zhang, R. Shi, K. Jiao, L. Li, Y. Wang, B. Xie, Q. Zhang, J. Yu, Y. Zhu, Z. Shao and M. Ni, Energy Environ. Sci., 2023, 16, 4926–4943.
2. A. M. Oliveira, R. R. Beswick and Y. Yan, Curr. Opin. Chem. Eng., 2021, 33, 100701.
3. B. Pivovar, N. Rustagi and S. Satyapal, Electrochem. Soc. Interface, 2018, 27, 47.
4. D. Yao, L. Gu, B. Zuo, S. Weng, S. Deng and W. Hao, Nanoscale, 2021, 13, 10624–10648.
5. M. Carmo, D. L. Fritz, J. Mergel and D. Stolten, Int. J. Hydrogen Energy, 2013, 38, 4901–4934.
6. E. Fabbri and T. J. Schmidt, ACS Catal., 2018, 8, 9765–9774.
7. C. Minke, M. Suermann, B. Bensmann and R. Hanke-Rauschenbach, Int. J. Hydrogen Energy, 2021, 46, 23581–23590.
8. Z. Lin, T. Wang and Q. Li, Ind. Chem. Mater., 2023, 1, 299–311.
9. K. Ayers, Curr. Opin. Chem. Eng., 2021, 33, 100719.
10. M. A. Hubert, L. A. King and T. F. Jaramillo, ACS Energy Lett., 2022, 7, 17–23.
11. Y. Lin, Y. Dong, X. Wang and L. Chen, Adv. Mater., 2023, 35, 2210565.
12. D. Escalera-López, S. Czioska, J. Geppert, A. Boubnov, P. Röse, E. Saraçi, U. Krewer, J.-D. Grunwaldt and S. Cherevko, ACS Catal., 2021, 11, 9300–9316.
13. J. Xu, J. Li, Z. Lian, A. Araujo, Y. Li, B. Wei, Z. Yu, O. Bondarchuk, I. Amorim, V. Tileli, B. Li and L. Liu, ACS Catal., 2021, 11, 3402–3413.
14. J. Zhang, X. Cao, Y.-F. Jiang, S.-F. Hung, W. Liu, H. B. Yang, C.-Q. Xu, D.-S. Li, T. Zhang, Y. Li, J. Li and B. Liu, Chem. Sci., 2022, 13, 12114–12121.
15. Y. Zheng, F. Zhang, G. Wang, D. Lai, L. Zou, Q. Cheng, J. Li, Z. Zou and H. Yang, J. Power Sources, 2022, 528, 231189.
16. M. Bernt, A. Hartig-Weiß, M. F. Tovini, H. A. El-Sayed, C. Schramm, J. Schröter, C. Gebauer and H. A. Gasteiger, Chem. Ing. Tech., 2020, 92, 31–39.
17. A. S. Pushkarev, I. V. Pushkareva and D. G. Bessarabov, Energy Fuels, 2022, 36, 6613–6625.
18. H. Chang, Z. Liang, L. Wang and C. Wang, Nanoscale, 2022, 14, 5639–5656.
19. Z. Shi, X. Wang, J. Ge, C. Liu and W. Xing, Nanoscale, 2020, 12, 13249–13275.
20. Y. Xu, K. Fan, Y. Zou, H. Fu, M. Dong, Y. Dou, Y. Wang, S. Chen, H. Yin, M. Al-Mamun, P. Liu and H. Zhao, Nanoscale, 2021, 13, 20324–20353.
21. S. Zhao, A. Stocks, B. Rasimick, K. More and H. Xu, J. Electrochem. Soc., 2018, 165, F82.
22. G. Jiang, H. Yu, Y. Li, D. Yao, J. Chi, S. Sun and Z. Shao, ACS Appl. Mater. Interfaces, 2021, 13, 15073–15082.
23. G. Jiang, H. Yu, D. Yao, Y. Li, J. Chi, H. Zhang and Z. Shao, J. Mater. Chem. A, 2022, 10, 11893–11903.
24. T. L. Doan, H. E. Lee, M. Kim, W. C. Cho, H. S. Cho and T. Kim, J. Power Sources, 2022, 533, 231370.
25. J. Cheng, J. Yang, S. Kitano, G. Juhasz, M. Higashi, M. Sadakiyo, K. Kato, S. Yoshioka, T. Sugiyama, M. Yamauchi and N. Nakashima, ACS Catal., 2019, 9, 6974–6986.
26. J. Zhang, Q. Zhang and X. Feng, Adv. Mater., 2019, 31, 1808167.
27. C. Han and T. Wang, Energy Fuels, 2023, 37, 13624–13644.
28. W. Xiao, X. Huang, W. Song, Y. Yang, T. S. Herng, J. M. Xue, Y. P. Feng and J. Ding, Nano Energy, 2016, 25, 60–67.
29. A. Liu, Z. Chen, X. Wei, W. Xiao and J. Ding, MRS Commun., 2017, 7, 563–569.
30. W. Xu, G. M. Haarberg, S. Sunde, F. Seland, A. P. Ratvik, E. Zimmerman, T. Shimamune, J. Gustavsson and T. Åkre, J. Electrochem. Soc., 2017, 164, F895.
31. F. Amano, Y. Furusho and Y.-M. Hwang, ACS Appl. Energy Mater., 2020, 3, 4531–4538.
32. C. Baik, J. Cho, J. I. Cha, Y. Cho, S. S. Jang and C. Pak, J. Power Sources, 2023, 575, 233174.
33. J. Islam, B. S. Yoon, P. T. Thien, C. H. Ko and S.-K. Kim, Catal. Today, 2024, 425, 114349.
34. Y. Liu, L. Xu, Y. Xin, F. Liu, J. Xuan, M. Guo and T. Duan, J. Electrochem. Soc., 2022, 169, 046516.
35. W. Li, J. Lv, D. Liu, W. Cai, X. Chen, Q. Huang, L. Wang and B. Wang, Chem. Mater., 2023, 35, 3892–3901.
36. H. Li, Y. Pan, L. Wu, R. He, Z. Qin, S. Luo, L. Yang and J. Zeng, Int. J. Hydrogen Energy, 2023, 48, 26021–26031.
37. H.-S. Oh, H. N. Nong, T. Reier, A. Bergmann, M. Gliech, J. Ferreira de Araújo, E. Willinger, R. Schlögl, D. Teschner and P. Strasser, J. Am. Chem. Soc., 2016, 138, 12552–12563.
38. Z. L. Zhao, Q. Wang, X. Huang, Q. Feng, S. Gu, Z. Zhang, H. Xu, L. Zeng, M. Gu and H. Li, Energy Environ. Sci., 2020, 13, 5143–5151.
39. 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.
40. Y. Wang, H. Meng, M. He and H. Yu, J. Alloys Compd., 2020, 822, 153711.

---

Footnote

† Electronic supplementary information (ESI) available: Synchrotron and electrochemical experimental details, TEM images, element mapping images, BET surface area and pore volume summary, XPS and EDS elemental content summary, XPS O 1s peak summary, XPS Ru 3d fitting results, double layer capacitance, a photograph of CCM with gaskets, and comparison of OER catalytic performance with literature reports. See DOI: https://doi.org/10.1039/d3nr06211b

---