Recent achievements in noble metal-based oxide electrocatalysts for water splitting

Abstract

The search for sustainable energy sources has accelerated the exploration of water decomposition as a clean H₂ production method. Among the methods proposed, H₂ production via water electrolysis has garnered considerable attention. However, the process of H₂ production from water electrolysis is severely limited by the slow kinetics of the anodic oxygen evolution reaction and large intrinsic overpotentials at the anode; therefore, suitable catalysts need to be found to accelerate the reaction rate. Noble metal-based oxide electrocatalysts retain the advantages of abundant active sites, high electrical conductivity of noble metals, and low cost, which make them promising electrocatalysts; however, they suffer from the challenge of an imbalance between catalytic activity and stability. This review presents recent research progress in noble metals and their oxides as electrocatalysts. In this review, two half-reactions (the hydrogen evolution reaction and the oxygen evolution reaction) of water electrolysis are described. Recently reported methods for the synthesis of noble metal-based oxide electrocatalysts, improvement strategies, and sources of enhanced activity and stability for these types of catalysts are presented. Finally, the challenges and future perspectives in the field are summarised. This review is expected to help improve the understanding of noble metal-based oxide electrocatalysts.

---

Feng Wang

Feng Wang is a master's student jointly trained by Kashi University and Yunnan University. Her current research areas of interest include the preparation of noble metal composites and related hydrolysis applications. She received her bachelor's degree in applied chemistry from Liaoning University in 2023.

Linfeng Xiao

Linfeng Xiao is currently a PhD candidate in Resources and Environment at Yunnan University, China. His research centers on the development of advanced functional materials for energy and environmental catalysis. With a strong focus on sustainability, he has made significant contributions, authoring three SCI-indexed papers as the lead author.

Xijun Liu

Xijun Liu received his PhD degree from College of Science, Beijing University of Chemical Technology in 2014. Then, he joined the School of Materials Science and Engineering of Tianjin University of Technology. Currently, he is a full time professor at the School of Resources, Environment and Materials of Guangxi University. His current scientific interests focus on sub-nano materials, heterogeneous catalysis, and materials design for catalysts and energy conversion/storage.

Xue Zhao

Xue Zhao completed his PhD in 2022 at Wuhan University, in China. Dr Zhao is currently working in the Faculty of Chemistry and Chemical Engineering, Yunnan Normal University in China. His research interests include the development of functional materials for energy catalysis, and environmental catalysis.

Abdukader Abdukayum

Abdukader Abdukayum received his PhD degree in Chemistry from the College of Chemistry, Nankai University, China in 2014. Currently, he is a full professor at the college of chemistry and environmental sciences, Kashi University, China. His research interests include the design and synthesis of advanced functional materials and their application to health, energy and environment fields.

Guangzhi Hu

Guangzhi Hu completed his PhD in 2010 under the supervision of Prof. Shijun Shao at the Lanzhou Institute of Chemical Physics (LICP), Chinese Academy of Sciences (CAS), in China. He then embarked on a four-year postdoctoral research fellowship at the Department of Physics, Umeå University, in Sweden. Following this, he worked as a full-time professor at the Xinjiang Technical Institute of Physics and Chemistry (XTIPC), CAS. Currently, he leads a research group at Yunnan University in China, focusing on the electrochemical applications of nanostructured materials in both energy storage and environmental protection.

---

Wider impact

With the continuous depletion of fossil energy sources, there is an urgent need to replace traditional fossil-energy hydrogen production technology with clean and non-polluting electrolytic water H₂ production technology. However, it is limited by slow kinetics in the water decomposition process and requires the use of appropriate catalysts for efficient hydrogen production. Noble metal electrocatalysts, which are rich in active sites, are effective electrocatalysts; however, the cost of pure noble metals is high, so there is an urgent need to develop lower-cost electrocatalysts. Noble metal-based oxides not only reduce the use of noble metals, to lower the cost, but also retain the active sites of noble metals, making them a perfect alternative. In this review, from the perspective of developing better advanced electrocatalysts with noble metal-based oxides, we comprehensively reviewed the existing research progress, elucidate the mechanism of water cracking for the HER and the OER, propose potential improvement pathways to facilitate the green conversion of water to high value-added small molecules, and finally suggest that future research needs to be directed toward catalyst recycling and regeneration of activity.

1. Introduction

Over the past few decades, traditional resources have been used to promote industrial advancement. However, the problem of energy depletion is rooted in the excessive utilisation of conventional fossil fuels, which also produces a series of pollutants that affect our daily lives.¹ Therefore, many environmentally friendly and renewable energy resources are being developed to overcome the current energy and environmental crises.² H₂, one of the most promising clean energy sources, has gradually become a new international focus and achieved rapid development.³ H₂ has been used in many fields, include power generation,^4–6 transportation industry,^7–9 medical domain,^10,11 energy storage,¹² high energy density fuel (122 kJ g⁻¹),¹³ raw materials for industry, and product cleaning.^14,15 Currently, H₂ is produced from coal,^16–19 fossil oil,^20,21 natural gas,^22,23 and water splitting^24–28 by many researchers worldwide. Unfortunately, obtaining H₂ energy generally involves the extensive use of fossil energy sources, which hinders humans from achieving the goal of carbon neutrality.^29,30 In recent years, electrocatalytic hydrolysis has garnered considerable attention as a non-polluting and highly efficient H₂ production method,³¹ which can be directly powered by the splitting of water to produce H₂, and theoretically reduces pollution and greenhouse gas emissions.^32,33 The earliest commercial application of this approach dates back to the 1890s.³⁴ Depending on the electrolyser, H₂ production methods are generally classified as alkaline solution electrolysis, proton exchange membrane electrolysis, and solid oxide electrolysis, among which however, alkaline solution electrolysis is the most widely used.^35–38

The kinetics of water decomposition is slow, enabling water molecules (H₂O) to exist in a stable form in nature. Therefore, to achieve efficient electrolysis of water for H₂ production, the reaction rates of the two half-reactions of water decomposition (the hydrogen evolution reaction (HER) and the oxygen evolution reaction (OER)) must be increased.^39–42 Numerous studies have found that the application of appropriate electrocatalysts to the H₂ production process can significantly improve efficiency by reducing the specific energy requirements.^32,43 These catalysts are classified as noble metal catalysts,^44,45 transition metal catalysts,^46–51 and non-metallic catalysts. Among these, noble metal catalysts are the most active catalysts and have numerous active sites. However, noble metals are very expensive and rare, causing the production cost of noble metal catalysts to be very high and hampering the large-scale application of electrolytic H₂ production to a certain extent.^26,52–58 To reduce the production cost, the amounts of noble metals used in such catalysts must be minimised, and scientists have conducted extensive research on this topic. However, as the use of noble metals decreases, the number of active sites with catalytic ability is inevitably reduced, and the catalytic activity decreases accordingly, thus, methods that can reduce the amounts of noble metals used while maintaining good catalytic activity are necessary. The ultimate goal of these methods is to increase the atomic utilisation of noble metals so that more active sites are exposed and employed. These methods can be classified into two categories: those using noble metal oxides and those loading very small amounts of noble metals onto other non-precious-metal oxides. However, these noble metal oxides have difficulty overcoming the ‘see-saw effect’ between high activity and high stability, and various improvement strategies have been proposed to optimise these catalysts, including the creation of heterostructures, the fabrication of crystal defects, and doping with other atoms.^59–63

With the continuous development of water electrolysis technology, scholars have found that the best catalytic performance for the HER is achieved using Pt, whereas the best catalytic performance for the OER is realised using RuO₂ and IrO₂.^64,65 The evolution of noble metal oxides is illustrated in Fig. 1a. As early as in 1971, Trasatti et al.⁶⁶ discovered the potential of RuO₂ as an electrode material that could be used in supercapacitors. In 1976, Iwakura et al.⁶⁷ prepared Pt oxide electrodes by thermal decomposition, studied the anodic evolution of O and found that the decisive step of the OER varied with changes in the overpotential. In 1979, Beni et al.⁶⁸ prepared Ir oxide films (SIROFs) by reactive sputtering Ir deposition under humidified O discharge conditions and found that SIROFs not only had higher catalytic activity than pure Ir, but also introduced a high degree of stability. Denton et al.⁶⁹ concluded that the catalytic preparation of chlorine with RuO₂–TiO₂ was accompanied by a parallel oxygenation reaction and found that RuO₂–TiO₂ accelerated the kinetics of this reaction under acidic conditions in 1981. By studying the electroreduction of aqueous Pd oxide layers formed at high positive potentials in different electrolyte solutions, Bolzán et al.⁷⁰ found that the composition of the electrolyte solution affected the electroreduction reaction. Fifteen years later (i.e. in 2007), Rh oxide thin-film electrodes Ti/Rh_xTi_(1−x)O_y (0.1 ≤ x ≤ 1.0) were prepared by Gouveia et al.,⁷¹ which exhibited Rh-dominated surface electrochemistry, contributing to the development of Rh-based catalysts. In 2015, Vargas-Uscategui et al.⁷² synthesised Re oxide by pulsed-current electrodeposition and found that the catalytic behaviour of the catalyst varied with the morphology and relative abundance of the Re oxide species. After a defect-rich RuO₂ catalyst was engineered in 2019, Ge et al.⁷³ discovered that the introduction of defects not only increased the number of active sites but also improved the intrinsic OER activity. In 2021, Zhang et al.⁷⁴ synthesised a PtNi@Pt dendritic nanowire (WO_x–PtNi@PtDNW) catalyst modified with WO_x, which improved the HER efficiency. In 2023, Lee⁷⁵ introduced an excess electron reservoir in an Sb-doped SnO₂ carrier of IrO_x and found that the mass activity of the prepared catalyst was 75 times higher than that of commercial nanoparticle-based catalysts, suggesting that the introduction of electron reservoir energy improves the catalytic performance.

Fig. 1 (a) Timeline of the use of noble metals and metal oxides. (b) Numbers of publications annually and cumulatively over the last 10 years. (c) Worldwide research status.

The trend in annual publication volume can indicate whether a particular research topic has value for continued study. Fig. 1b shows the number of annual and cumulative publications on the application of noble metal-based oxides in water electrolysis over the past 10 years. The increasing trends in these quantities demonstrate that the amount of research on noble metal-based oxide electrocatalysts has been increasing every year for the last 10 years, which also indicates that research on this type of catalyst remains a popular topic in contemporary times. In the line graph of the annual number of publications, several gentle upward phases are observed, which may be due to some researchers shifting their research toward photoelectrocatalytic water decomposition^76,77 or searching for transition metal catalysts^78–80 to replace noble metals. In addition, understanding the current status of worldwide developments in a given field will help researchers better choose their research directions. In this review, more than 800 papers on the catalytic electrolysis of water with noble metals and metal oxides published in the last decade were analysed using scientometric methods,⁸¹ and a grid map of cooperation among different countries is shown in Fig. 1c. This grid map includes 53 nodes (dots) and 201 links (connecting lines), indicating that 53 countries have published relevant papers in the last 10 years. The closeness and interlocking complexity of the lines indicate that individual countries are closely connected. Among them, the top five countries in terms of number of publications in the last decade are: China, the United States, South Korea, Germany, and Japan. Most articles are from China (459 articles), followed by the United States (110 articles), although the number of published articles from the United States is only a quarter of the number from China. Thus, China has contributed significantly to the development of this field.

In the past few years, increasing numbers of reports have been published on synthesis and modification strategies for noble metal-based oxide electrocatalysts. To the best of our knowledge, most studies on RuO₂ have focused on improving its stability, whereas most research on IrO₂ focuses on improving its reactivity, and studies on other noble metal oxides are surfacing. Relatively few reviews exist that discuss noble metal oxides in an integrated manner, which leads to the lack of a more macroscopic understanding of noble metal-based oxide electrocatalysts. Considering the rapid development of noble metal-based oxide electrocatalysts, an urgent need exists to understand the latest advances, obstacles, and prospects in this cutting-edge field promptly. In this paper, we present a thorough review of the recent research advances related to noble metal-based oxide electrocatalysts, starting with descriptions of the two half-reaction mechanisms of water electrolysis. Next, the most common synthetic methods for noble metal oxides and strategies for their improvement are discussed. The reasons for the increased activity and stability of the optimised catalysts are also described in detail. Finally, challenges and future research directions for noble metal-based oxide electrocatalysts are outlined. This review is hoped to guide the design and optimisation of noble metal-based oxide electrocatalysts with excellent catalytic activity, remarkable stability, and competitive cost-effectiveness.

2. Water splitting mechanism

The two half-cell reactions of water splitting exhibit diverse forms when they occur in different media,⁴¹ as summarised in Table 1. The HER occurs at the cathode and produces H₂, and the OER occurs at the anode and produces O₂. A strong link exists between the HER and the OER, where an increase in efficiency on one side is followed by an increase in efficiency on the other side.

Table 1 Two half reactions of water splitting

	In acidic media	In alkaline media
Cathode:	2H^+ + 2e^− → H2	2H2O + 2e^− → H2 + 2OH^−
Anode:	2H2O → O2 + 4e^− + 4H^+

---

2.1. HER mechanism

The commonly acknowledged mechanism of the HER in alkaline or acidic media involves three elementary reactions: Volmer, Heyrovsky, and Tafel steps.⁸² Because the Volmer reaction is always present in the HER, the HER involves only two mechanisms: the Volmer–Heyrovsky and Volmer–Tafel mechanisms, as shown in Fig. 2. The Volmer step is the first step in water dissociation. In this step, adsorbed proton (H⁺) intermediates (M–H), where M donates the active site of the catalyst, are generated via the reaction of M with H⁺ or H₂O through the following reactions:

H⁺ + M + e⁻ → M–H (in acidic media) (1)

H₂O + e⁻ + M → M–H + OH⁻ (in alkaline media). (2)

Fig. 2 Schematic of the HER.

Subsequently, H₂ is processed and desorbed from the surface of the catalysts through the Heyrovsky step or the Tafel step, depending on the characteristics of the catalysts. Generally, M–H bonds react with H⁺ or H₂O during the Heyrovsky step, and the adjacent coupled M–Hs interact with each other during the Tafel step. This process is described by the following formulas:

Heyrovsky step:

M–H + H⁺ + e⁻ → H₂ + M (in acidic media) (3)

M–H + H₂O + e⁻ → H₂ + OH⁻ + M (in acidic media) (4)

Tafel step:

M–H + M–H → H₂ + 2M (in acidic and alkaline media) (5)

The rate of the HER is largely dependent on the adsorption-free energy (ΔG) of H⁺. If the binding of H⁺ to the surface is too weak, the adsorption Volmer step will become a rate-determining step due to the small number of M–Hs. If the binding is too strong, impeding the desorption of H₂, the Heyrovsky/Tafel step will limit the total reaction rate. Therefore, the necessary but not sufficient condition for an active HER catalyst is ΔG ≈ 0.⁸³ However, many scholars believe that the progress of desorption is the rate-determining step because if H₂ does not diffuse from the catalyst surface into the solution in time, the active sites on the catalyst will be blocked, decreasing the effective catalytic reaction surface area and impeding the progress of the reaction. Thus, the catalytic efficiency is related not only to the active site, but also to the degree of gas aggregation on the surface of the catalyst, showing a volcano-like trend.^84,85 Numerous self-supported materials have been used as catalysts to increase the effective surface area and expedite the diffusion rate of H₂.

The electrolyte pH determines the rate-determining step. In acidic electrolytes, the concentration of H⁺ is high; therefore, the adsorption step is easy to perform. As the Volmer step is completed in a short time, the desorption of H₂ limits the reaction rate. The concentration of H⁺ in alkaline electrolytes is very low, which slows adsorption. The Volmer step takes a long time to complete, making it difficult for the subsequent desorption to proceed smoothly. Therefore, the dissociation of water to produce H₂ is the rate-determining step in alkaline electrolytes.^86,87

2.2. OER mechanism

The OER involves a four-electron transfer system with a high energy barrier. Therefore, the reaction requires a high overpotential to overcome this energy barrier. Improving the efficiency of the OER, a half-cell water decomposition reaction, improves the H₂ production efficiency of electrolysed water.⁸⁸ The four-electron transfer mechanism for the OER is commonly known as the adsorption evolution mechanism (AEM); however, it is accompanied by a lattice oxygen participation mechanism (LOM) when the catalyst is an oxide. These two pathways are shown in Fig. 3.

Fig. 3 Schematic of the HER. (a) Acidic and alkaline LOM. (b) Acidic and alkaline AEM.

When the electronic state of the catalyst is close to the Fermi energy level exhibiting metallicity, the metal acts as a redox centre, which leads to a conventional AEM mechanism in the OER process, as shown in Fig. 3a.⁵³ Under acidic conditions, the first step is the dissociation of H₂O and adsorption of the deprotonated hydroxide radical (OH⁻) on the active site to form M–OH. In the second step, further dissociation of H⁺ and formation of M–O occurs. In the third step, M–O combines with the deprotonated OH⁻ of another H₂O to form adsorbed M–OOH on the surface of the metal cation and ultimately dissociates to O₂. These steps can be expressed according to eqn (6)–(9). The alkaline medium is dominated by OH⁻, which can bind directly to the active site, enabling the OER process to proceed without the hydrolytic dissociation of OH⁻, as expressed in eqn (10)–(13).

In acidic media:

M + H₂O → M–OH + H⁺ + e⁻ (6)

M–OH → M–O + H⁺ + e⁻ (7)

M–O + H₂O → M–OOH + H⁺ + e⁻ (8)

M–OOH + H₂O → M + O₂ + H⁺ + e⁻ (9)

In alkaline media:

M + OH⁻ → M–OH + e⁻ (10)

M–OH + OH⁻ → M–O + H₂O + e⁻ (11)

M–O + OH⁻ → M–OOH + e⁻ (12)

M–OOH + OH⁻ → M + O₂ + H₂O + e⁻ (13)

When O is present in the catalyst, isotopic labelling and differential electrochemical mass spectrometry can probe the origin of the O₂ produced at the active site, confirming that some of the lattice O are involved in the OER process. This mechanism by which the lattice O is oxidised and thus participates in the OER process has been termed the LOM mechanism.⁸⁹ The first and second steps of the LOM mechanism, forming M–O, are similar to the AEM mechanism, following the coupling of M–O with the lattice O on the catalyst, generating O₂ and O. Finally, the O vacancies further react with H₂O or OH⁻, returning to the original catalyst form. This loop is shown in Fig. 3b.

The existence of the LOM mechanism has both advantages and disadvantages. On the one hand, the LOM process does not require the step of forming M–OOH, which accelerates the reaction rate. On the other hand, the catalyst of the LOM pathway is dynamically changing: the creation of O vacancies (O_v) accelerates the metal solubilisation, leading to the collapse of the catalyst structure and the rapid deactivation of the catalyst, thus decreasing the catalyst stability. To improve catalyst stability, the catalyst structure must sometimes be modulated to inhibit the LOM pathway and enhance the AEM pathway.⁹⁰

The conversion of electrical energy into H₂ energy through electrochemical workstations involves the dissociation of H₂O and the formation of H₂ and O₂ release processes. However, the OER and HER have competing reactions. The HER involves two competing reactions: the Volmer–Heyrovsky step and the Volmer–Tafel step. The OER, which is the rate-controlling step of water electrolysis, involves two competing reactions: the AEM and LOM reactions. Different reaction pathways lead to different performances; however, they also introduce corresponding shortcomings. Therefore, different catalysts must be selected and manufactured accordingly to achieve a balance between catalyst performance and stability.

3. Synthetic methods and modification strategies of noble metals and metal oxides

3.1. Synthesis methods

The catalytic properties of catalysts vary with their structures, and different synthesis methods can produce different structures. Thus, when synthesising the target catalysts, full consideration must be given to the nature of the raw materials and the selection of suitable catalysts. Common synthetic catalyst synthesis methods include hydrothermal,⁹¹ solvent-thermal,⁹² pyrolysis,⁹³ templating,⁹⁴ precipitation,⁹⁵ sol–gel,⁹⁶ high-pressure microwave-assisted,⁹⁷ and electrodeposition⁹⁸ methods. This review focuses on the template, electrodeposition and pyrolysis methods for synthesising noble metal-based oxides and a simple flowchart summarising these three synthesis methods is shown in Fig. 4.

Fig. 4 Three simple flowcharts of the (a) template, (b) electrodeposition and (c) thermolysis methods.

3.1.1. Template method. The template method is an effective method of designing and constructing porous nanostructures. It first converts the space occupied by the template into nanopores and then fills these nanopores with precursors, thus replicating the nanostructure with the initial template morphology in reverse. Therefore, the pore structure parameters of multistage porous materials can be easily adjusted by changing the template structure and controlling the introduction of the templates. Template methods can be classified as soft and hard templates.⁹⁹ Hard templates are materials with fixed shapes that are relatively stable, such as molecular sieves¹⁰⁰ and mesoporous silica (SiO₂),^101,102 whereas soft templates are surfactants and amphiphilic block copolymers capable of self-assembly, such as liquid crystals¹⁰³ and gels.¹⁰⁴ Template synthesis methods usually include the following steps. First, templates with well-defined structures, such as nanoparticles and nanorods, are introduced into the precursors. Then, the precursor is introduced inside the pores of the template, or the precursor encapsulates the template and is cured. Finally, the template is removed by acid/base etching, dissolution, or heat treatment.¹⁰⁵

To decrease the amount of noble metals used and retain high catalytic activity, many researchers have focused on the preparation of novel noble metal oxides and have attempted to use these catalysts as substitutes for conventional noble metal catalysts. However, as the noble metal content in the catalyst decreases, the total number of active sites contained in the catalyst decreases accordingly, and to overcome this challenge, the atom utilisation rate must be maximised so that the active sites can be exposed to the maximum extent to participate in the reaction. By choosing a template with a porous structure to replicate the target material with a similar structure, an increase in the catalyst-specific surface area can be achieved. Lin et al.¹⁰⁶ used a hard template method to synthesise a Cr–Ru oxide solid solution electrocatalyst (Cr_0.6Ru_0.4O₂) using a metal–organic framework as the template, which effectively improved the OER efficiency in an acidic medium.

The traditional hard template method has the disadvantage of requiring the template to be separated from the target product, which is difficult, or sacrificing the template to complete the synthesis of the catalyst. Therefore, to overcome these shortcomings, a salt template method using soluble salt as the template was developed. Commonly used salts in the salt template method include KCl,¹⁰⁷ NaCl,^108,109 ZnCl₂,¹¹⁰ CaCl₂,¹¹¹ and other chlorinated salts. However, some special reactions exist in which achieving the desired effect by using only one type of salt is difficult; therefore, multiple salts are utilised simultaneously in what can be referred to as the multi-salt templating method.^112,113 Sun et al.¹¹⁴ synthesised RuO₂–WC composite nanoparticles as catalysts for the bifunctional electrolysis of water using the salt template method. The synthesis process is shown in Fig. 5a, where NaCl is employed as the internal template. First, (NH₄)₁₀W₁₂O₄₀·xH₂O is mixed with RuCl₃ to obtain the precursor, which is evenly covered on the NaCl surface. This precursor is carbonised into WC cores via carbothermal reduction and subsequently stretched along the NaCl template, resulting in the formation of WC nanoparticles (WCNPs). Under a reducing atmosphere, Ru³⁺ is converted into Ru particles and attached to the surface of the WCNPs (step 1). Subsequently, Ru–WCNPs@NaCl are oxidised to form RuO₂-WCNPs@NaCl (step 2). The RuO₂–WCNPs are then washed in deionised water to solubilise and eliminate the NaCl template and obtain RuO₂–WCNPs (step 3).

Fig. 5 (a) Flowchart of RuO₂–WCNP synthesis. (b) and (c) SEM, (d) TEM, (e) SAED, and (f) HRTEM images of RuO₂–WCNPs. (g) HRTEM image of the selected region in (f). (h) TEM-EDS elemental mapping in RuO₂–WCNPs. Reproduced from ref. 114 with permission from Wiley-VCH GmbH, copyright 2022.

The scanning electron microscopy (SEM) images of the synthesised RuO₂–WCNPs clearly reveal that the RuO₂–WCNPs are homogeneously distributed on the C flakes with a size of roughly 20 nm (Fig. 5b and c), forming a 0D/2D nanostructure. The low-magnification transmission electron microscopy (TEM) images (Fig. 5d) clearly show uniformly sized RuO₂–WCNPs modified on the C sheets. The accompanying selected-area electron diffraction (SAED) map (Fig. 5e) shows two sets of derivative rings, confirming the coincidence of RuO₂ and WC in the multicrystalline structure. The interfacial structure of WC–RuO₂ was observed using high-resolution transmission electron microscopy (HRTEM) (Fig. 5f). RuO₂NPs with a diameter of about 2 nm are tightly adhered to the WC surface, forming RuO₂–WCNPs. Fig. 5g clearly shows the interface between the RuO₂ and WC carriers. The (001) and (101) crystal faces of hcp-WC are observed in the yellow region, whereas the (200) crystal face of RuO₂ is visible in the red field. The results of TEM with energy-dispersive X-ray spectroscopy (EDS) mapping prove that W and C are evenly dispersed among the heterogeneous particles, whereas Ru and O show an extra-fine-grained morphology (Fig. 5h). Performance characterisation revealed that the catalyst significantly promoted the OER activity with η₁₀ of 347 mV and mass activity of 1430 A g_Ru⁻¹, which is eight times greater than that of the commercial Ru (176 A g_Ru⁻¹).

Because the traditional template method synthesises at low temperatures, the prepared catalysts can obtain nanostructures with high-quality specific activity, but their stability has some defects; increasing the temperature during synthesis is a method of enhancing the stability. Malinovic et al.¹¹⁵ synthesised structurally ordered Ir oxide catalysts for the first time at high temperatures. The presence of hydroxyl surface groups enhanced the catalytic activity on the nanoscale, as well as the stability of the catalyst through high-temperature heat treatment. The synthesis starts with the hydrolysis of an Ir precursor in a water-in-oil microemulsion, followed by sol–gel encapsulation with silica and heat treatment (800 °C) after drying, and finally, the removal of the protected silica shell using hydrofluoric acid–ethanol solution to obtain Ir oxide nanoparticles that are unsupported and surfactant-free. To determine the effect of the heat treatment temperature on the electrochemical stability of the synthesised catalysts, the authors measured the synthesised materials at 400 °C, 600 °C, and 800 °C online using a scanning flow cell with inductively coupled plasma (ICP) mass spectrometry (MS), and the test results show that IrO₂-800 °C has the highest intrinsic stability, with an initial S-number of 1, which is almost an order of magnitude higher than that of IrO_x-400 °C. Thus, the high-temperature heat treatment enhances the stability of the catalyst, and this high-temperature heat treatment strategy is expected to be utilised to enhance the stability of other noble metal oxides.

The template method is a commonly used method for the preparation of multilevel pore structure materials, but some drawbacks exist, such as the fact that the purity and crystal structure of the prepared materials are not easy to control and that the thermal stability and corrosion resistance of the prepared materials still need to be improved. In addition, the preparation process of this method is complex and sometimes even requires the use of relatively expensive sacrificial templates, limiting the large-scale commercialisation of the template method. Fortunately, electrodeposition is simpler than template deposition.

3.1.2. Electrodeposition. The electrodeposition method was used to synthesise the target materials by controlling the redox capacity of the metal ions in the electrolyte solution by varying the voltage and current, thus precisely controlling the nucleation and growth rate of the nanocrystals. The method involves both physical and chemical processes. The chemical reaction for depositing metal ions is induced by an electric current and can therefore be classified as a chemical method. However, the growth and morphology of the film precipitated on the electrode are influenced by physical effects; therefore, the method also involves physical processes. This method has the dual benefit of being highly efficient and environmentally friendly and has the advantages of low reagent consumption, cleanliness, efficiency, recyclability, and freedom from carrier limitations. It is widely recognised as an economical, fast, and efficient technique.¹¹⁶ Depending on the applied signal (voltage or current), electrodeposition methods can be classified into constant potential deposition,¹¹⁷ constant current deposition,¹¹⁸ pulsed deposition,¹¹⁹ and cyclic voltammetric electrodeposition.^116,120 Electrochemical deposition is generally categorised into anodic and cathodic deposition depending on the location of the product. Electrochemical deposition is usually performed at low temperatures (less than 100 °C), and the modulation of the morphology, thickness, and conductive behaviour of the target catalysts is achieved by adjusting the electrolyte temperature, pH, applied potential, and current.

Sub-exchange membrane water electrolysis (PEMWE) promises sustainable and scaled-up green H₂ production,⁵⁷ and PEMWE with core–shell nanostructures has significantly lower ohmic resistance and mass transfer resistance, which can greatly improve the reaction kinetics. Electrocatalytic deposition facilitates the synthesis of materials with core–shell structures. Jeon et al.¹²¹ synthesised functional electrocatalyst Si, W co-doped RuO_x nanomaterials (RuSiW) (Fig. 6Aa) by electrodeposition of the precursors onto the electrode surfaces in an acidic medium using [Ru₄(μ-O)₄(OH)₂(H2O)₄(μ-SiW₁₀O₃₆)₂](RuPOM) as the precursor. During electrodeposition, the flux was negligible for the first 2 h, followed by an exponential increase and gradual saturation by approximately 4 h (Fig. 6Ab). A similar trend was observed when the pattern of the C fibre paper electrode was analysed (Fig. 6Ac). HRTEM showed that the individual Ru–Si–W nanoparticles consisted of nanocrystalline cores and shells with thicknesses of 5–10 nm (Fig. 6Ad). To reveal the chemical composition of the shell, elemental mapping was performed using energy dispersive X-ray spectroscopy (EDX) (Fig. 6Ae) and elementary depth section analysis using time-of-flight secondary ion mass spectrometry (ToF-SIMS) (Fig. 6Af). Ru, Si, W, and O, which are all elements in the RuPOM precursor, were examined in intact nanoparticles. Notably, the high contents of Si and W near the surface (≈10 nm deep) of RuSiW nanoparticles indicate that the shell layer is composed of Si, W co-doped RuO_x and possibly silicotungstates. Ultimately, RuSiW was found to be highly stable and exhibited HER and OER activities similar to those of Pt/C and RuO₂. Moreover, theoretical calculations showed that the co-doping of RuO_x with W and Si triggered a shift in the d-band centre and the optimisation of a favourable atomic configuration for H adsorption, which enhanced the HER activity of the originally poor RuO₂ synergistically.

Fig. 6 (A) (a) Molecular structure of RuPOM and schematic diagram of RuSiW synthesised by electrodeposition. (b) Temporal maps measured during the RuSiW deposition process. (c) SEM images of RuSiW cascaded electrodes at different time intervals. (d) HRTEM image of RuSiW. (e) EDX elemental dispersion patterns of RuSiW. (f) ToF-SIMS elemental depth section of RuSiW. Reproduced from ref. 121 with permission from Wiley-VCH GmbH, copyright 2023. (B) (a) Schematic representation of the synthetic route of SACs. (b) Mass loadings of Ir and Pt SAs on various metal oxide supports. (c) HAADF-STEM image of Ir₁–Co₃O₄. (d) EDX elemental mapping images of Ir₁–Co₃O₄. Reproduced from ref. 122 with permission from Wiley-VCH GmbH, copyright 2024.

Jeong et al.¹²³ also synthesised a catalyst with a core–shell structure for stable and efficient OERs via electrodeposition. They used electrodeposited Ir oxide thin films (EIROFs) as a support enriched with metal Fe₂N nanostructures to construct a unique Fe₂N@EIROF core–shell structure. Because the application of transition metal nitrides (TMNs) in PEMWE in corrosive acidic environments is limited by their chemical instability, careful passivation of the TMN surface with a chemically stable Ir catalyst layer can overcome this instability and prolong the lifetime of the catalyst, resulting in the excellent stability of Fe₂N@EIROF. In addition, the porous Fe₂N nanostructures modified on the Ti porous transport layer (PTL) increased the surface area of the catalyst, which led to an increase in the mass activity of Ir (103 A mg⁻¹), and thus the catalytic activity of the catalyst. Thus, TMNs can be used as functional protective layers and carriers to improve the performance of catalysts by replacing conventional Pt electrodes in cost-effective PEMWEs. Jiang et al.¹²⁴ prepared an IrRu@WO₃ electrode consisting of WO₃ nanoarrays with heterogeneous IrRu bimetallic coatings by electrodeposition to improve the OER stability and activity in PEM water electrolysis. The IrRu@WO₃ electrode required only a 245 mV overpotential, and the electrodes were able to operate stably for 500 h at a voltage of 2.16 V. Due to the good O*-binding energy of the electrophilic interfacial Ru, the OER activity could be improved, and the contribution of W and Ir enhanced the electronic environment of Ru, which prevented the formation of soluble peroxide and enhanced stability.

Catalysts prepared by electrodeposition either have a core–shell structure or produce an encapsulated state, where the interaction between the different layers improves both catalyst activity and stability. Thus, the electrodeposition method has great potential for application in the synthesis of noble metals and metal oxides. However, the electrodeposition method is technically difficult because of the high requirements for devices and equipment and often requires anti-corrosion treatment to avoid corrosion of the loading materials.

3.1.3. Thermolysis. The pyrolysis method, which uses high-temperature decomposition of precursors to generate new solid-phase materials, has great advantages in dealing with metal precursors that have low reduction potentials and are not easy to reduce chemically and has been used by an increasing number of researchers to synthesise catalysts in recent years. In addition, this method has the advantages of low cost and ease and speed of operation. The larger the specific surface area of the prepared catalyst, the better the catalytic performance, and the catalyst must usually be prepared as a thin film to maximise the catalytic activity. Fortunately, spray pyrolysis meets this requirement well.¹²⁵ Spray pyrolysis is a method of synthesising a target material from solid or liquid particles suspended in a gas stream produced by the thermal decomposition of an atomiser driven by a gas stream. Commonly used atomisers are pneumatic sprayers,¹²⁶ ultrasonic sprayers,¹²⁷ and electrostatic sprayers.¹²⁸ This method enables the chemical composition of the resulting film to be controlled by changing the chemical composition of the precursor solution and has the advantage of being environmentally friendly and having easily achievable operating conditions.

In the preparation of catalysts, pyrolysis is often used in conjunction with other synthetic methods to obtain the precursor, whereas pyrolysis is utilised to obtain the target product. Zhu¹²⁹ reported a catalyst based on the doping of Pt singlet atoms (SAs) into heterogeneous interfacial Ru/RuO₂ carriers (Pt–Ru/RuO₂) and used it to promote the alkaline HER. Pt–Ru/RuO₂ was first synthesised by impregnating the RuCl₃ precursor in NaNO₃ molten salt to obtain RuO₂-rich grain boundaries, followed by the adsorption of Pt species on RuO₂, and finally by calcination in an Ar atmosphere and thermal decomposition. Furthermore, Pt–Ru/RuO₂ requires only 1.90 V overpotential to reach a current density of 1 A cm⁻² and shows excellent price activity (up to 247.1 $⁻¹ at 2.1 V) in anion-exchange membrane-electrolysed water integrated with NiFeLDH, which is almost three times as high as that of its Pt/C-based counterpart. This good performance was mainly due to the acceleration of slow water decomposition by RuO₂, whereas the Pt SAs and Ru promoted the binding of H* on the catalyst surface. This study provides an effective method of searching for catalysts that balance high activity with the economy. However, the balance between the activity and stability of noble metal-based oxide electrocatalysts requires further in-depth study. Chen et al.¹³⁰ reported a low-crystallinity iridium molybdenum oxide (IrMoO_x) nanofibre catalyst prepared by electrostatic spinning-calcination and found that IrMoO_x exhibited excellent OER activity in an acidic electrolyte, with an overpotential of only 267 mV at a current density of 10 mA cm⁻², which is lower than that of the benchmark commercial IrO₂ catalyst (330 mV). The increased catalytic activity of IrMoO_x stemmed from the introduction of Mo inducing a change in the electronic structure around the active site of Ir oxide. Although iron oxides are known for their excellent stability in acidic media, the IrMoO_x synthesised in this study was stable for only 30 h. It can be seen that the introduction of Mo improved the catalytic activity at the expense of stability.

Molten salts have low melting points (<100 °C),¹³¹ and their use in the pyrolysis process can reduce the operating temperature to a certain extent and save energy. Kaushik et al.¹²² conducted a study on the preparation of single-atom catalysts (SACs) for the promotion of acidic water decomposition by the direct pyrolysis of molten salts. The synthesis process of SACs is shown in Fig. 6Ba. First, a freshly prepared aqueous solution containing noble metal chloride (ACl_x) and non-precious-metal nitrate [B(NO₃)_y] is mixed with an aqueous solution, during which the metal nitrate in the molten state provides a suitable environment for the exchange between the noble and non-precious-metal cations due to its high mobility, forming a molten salt [ACl_x–B(NO₃)_y]. This salt is drop-coated onto C paper and then directly annealed in air at 220–300 °C. The nitrate precursor then decomposes to yield highly dispersed noble metal monoatoms embedded in the non-precious-metal oxide lattice (A1–BO_z). Finally, the undecomposed chloride precursor is washed away with ionised water, resulting in a high-purity product. The authors synthesised seven non-precious-metal oxide single-atom SACs and used ICP optical emission spectroscopy (OES) to measure the mass loading of noble metals in each catalyst (Fig. 6Bb). The mass loading of Ir SAs in Co₃O₄ is as high as 10.97 wt%, and the mass loading of Pt SAs in Co₃O₄ can reach 4.60 wt%, which is higher than that in other oxides. The aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image of Ir–Co₃O₄ (Fig. 6Bc) shows highly dispersed Ir loading on the surface of Co₃O₄. The EDX mapping image (Fig. 6Bd) reveals a uniform distribution of elements in Ir–Co₃O₄. The obtained Ir₁–Co₃O₄ and Pt₁–Co₃O₄ electrocatalysts require overpotentials as low as 227 mV and 22 mV to achieve OER and HER current densities of 10 mA cm⁻², respectively, with good electrocatalytic performance.

The preparation of catalysts by the thermal decomposition of molten salts not only improves the performance of the catalysts, but also has the universality of simple operation. Hence, this approach is expected to be used for large-scale production and applications in the future and can even be utilised to explore the performance of catalysts loaded with diatomic or triatomic noble metals on metal oxides. However, the products prepared using the thermal decomposition method have a uniform particle size distribution and good dispersion, and the particle size and morphology of the products can be regulated by changing the reaction parameters and calcination temperature of the precursor. In addition, the reaction conditions of this method are mild, and the cost of the products is low; therefore, an increasing number of researchers are synthesising noble metal and metal oxide catalysts using the thermal decomposition method.

In conclusion, catalysts with different morphologies can be prepared by employing different methods. The balance between the catalytic activity and stability of catalysts can be maximised by choosing an appropriate synthesis method to synthesise noble metal oxides based on the inherent properties of the raw materials. To achieve this balance, in addition to selecting appropriate synthetic methods, improvement strategies can be used to compensate for the shortcomings of the original catalysts.

3.2. Improvement strategies

The difficulty in balancing the cost, activity, and stability of catalysts has long been a major challenge in the development of electrolytic water-to-H₂ technology. Several improvement strategies have been proposed to enhance the inherent defects of different catalysts, many of which have positive impacts. This review focuses on four strategies for improving noble metal oxides: doping engineering, heterojunctions, strain engineering, and interfacial engineering.

3.2.1. Doping. Doping is the process of introducing other elements into an original substance to change its physical or chemical properties; therefore, it is commonly used and considered to be an effective modulation strategy to improve the electrical conductivity and catalytic properties of catalytic materials.¹³² By varying the rarity of the dopant atoms, doping methods can be classified into doping with metallic atoms^133–137 and non-metallic atoms^138–140 or co-doping with multiple atoms.^141–144 Metal doping is a relatively common method in the present day, and noble metals can be doped into transition metal oxides to prepare noble metal oxides. Zhao¹⁴⁵ doped Pt into MnO₂ nanosheets to prepare catalysts that can efficiently catalyse water decomposition (sl-Mn_0.98Ir_0.02O₂). The strain-induced elongation of the Mn–O bond lengths due to Ir doping modulated the electronic architecture and improved the catalytic behaviour of the OER. The catalyst exhibited an excellent mass activity of 5681 A g⁻¹ at an overpotential of 300 mV in 0.5 M H₂SO₄ and achieved current densities of 50 and 100 mA cm⁻² at overpotentials of 240 mV and 277 mV, respectively.

Doping transition metals into noble metal oxides is also possible. For this class of doping methods, theoretical calculations can be used to obtain the optimal doping transition metal for a particular noble metal oxide, which can minimise the experimental exploration time through theoretical calculations, and these approaches are increasingly becoming the methods preferred by researchers. Esterhuizen et al.¹⁴⁶ developed a workflow that uses machine-learning-assisted Bayesian optimisation in conjunction with density functional theory (DFT) to screen for the most superior dopants of acid-resistant Ir-based oxides for the formation of the OER. The authors also performed experimental validation and determined that Ir_0.5Mo_0.5O_x can be used as a potentially stable catalyst for the OER. The potential of Ir_0.5Mo_0.5O_x is 30 mV lower than that of pure Ir, whereas the dissolution rate of Ir is 24% lower than that of pure Ir. However, improving the durability of RuO₂ in acidic environments remains challenging. In the same year, Sun et al.¹⁴⁷ determined the solvation energy of Ru in 3d transition metal-doped MRuO_x (M = Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn) by performing DFT calculations to evaluate the stability of these catalysts in the acidic OER. The calculations revealed that ZnRuO_x had the highest stability among the above six catalysts. Experimental results were also obtained that verified this finding and demonstrated that ZnRuO_x experienced negligible decay of η₁₀ after 15 000 CV cycles. Importantly, using ZnRuO_x as the anode, PEMWE can be performed stably at a current density of 200 mA cm⁻² for 120 h. This study also shows that the use of theoretical calculations to guide the experiments is indeed a feasible approach, which can effectively shorten the experimental cycle and will have a wide range of applications in the future.

Metal doping involves regulation of the catalyst structure by replacing the position of the original metal with metal atoms. For catalysts doped with transition metals into noble metal oxides, with the introduction of dopant atoms, some of the active sites of the noble metal are replaced, which reduces the catalytic activity and even stimulates the LOM of the OER and reduces the stability of the catalyst.⁹⁰ To address this issue, Ping⁹⁰ proposed a strategy of interstitial doping with non-metals to inhibit lattice O oxidation and thus improve stability. Since the effective atomic radius of Si⁴⁺ (0.26 Å) is smaller than that of Ru⁴⁺ (0.62 Å), they doped Si into its interstitium of RuO₂ to construct stable RuO₂ electrocatalysts for the acidic OER. Fig. 7a shows the catalytic decomposition of water by conventional RuO₂ under normal conditions, where the LOM occurs, and RuO²⁻ is solubilised with the participation of lattice O in RuO₂ as the reaction proceeds. However, as shown in Fig. 7b, by doping with interstitial Si, the lattice O is immobilised and does not participate in the catalytic reaction, and the AEM mechanism occurs, enhancing the stability of the catalyst. The HAADF-STEM plots in Fig. 7c–e reveal that Si is present in the interstitials, and the same conclusion can be reached by analysing the atomic line profiles in Fig. 7f–i. In addition, because the bond dissociation energy of Si–O (798 kJ mol⁻¹) is larger than that of Ru–O (481 kJ mol⁻¹), the doping of Si can weaken the Ru–O bond covalency, thus changing the pathway through which the OER occurs and enhancing the stability, as evidenced by the longer Ru–O bond length in the Fourier-transformed expanded X-ray absorption fine structure spectra (FT-EXAFS) (Fig. 7j). The catalyst was subjected to electron paramagnetic resonance (EPR) tests, as shown in the spectra in Fig. 7k, in which the O_vs before and after the reaction are near 3513 G (g = 2.001) without any significant change, indicating that almost no lattice O is involved in the reaction and that the AEM pathway dominates the OER. In the performance test, it was found that the Tafel slope of 33.0 mV dec⁻¹ for Si–RuO₂-0.1 was lower than that of the commercial RuO₂ catalyst, indicating that the catalyst was effective in improving the slow kinetics of the OER, and the catalyst had a very small charge transfer resistance between the catalyst and the reactants, which provided a fast charge transfer capability and high electrical conductivity. A representative Si–RuO₂-0.1 catalyst showed high activity and stability in acid, with an overpotential of 226 mV at 10 mA cm⁻² and a negligible degradation rate of ∼52 μV h⁻¹ over 800 h of testing.

Fig. 7 (a) and (b) Schematic representations of the enhanced stability of interstitial doped Si. (c) HAADF-STEM image of Si–RuO₂-0.1. (d) and (e) Magnified HAADF-STEM images derived from the region boxed out in (c). (f)–(i) Line-scanning intensity sections from the regions boxed out in (d) and (e). (j) FT-EXAFS spectra of Si–RuO₂-0.1 and Com-RuO₂. (k) EPR spectra of Si–RuO₂-0.1 before and after the stability test. Reproduced from ref. 90 with permission from Springer Nature, copyright 2024.

In metal atom doping, although the optimal metal dopant for noble metal oxides can be derived from theoretical calculations and doping with the optimal metal dopant can enhance the stability of the catalyst to a certain extent, the degree of increase in the stability of noble metal oxides is still low compared to that in non-precious-metal interstitial doping. In the future, theoretical calculations could be combined with non-precious-metal interstitial doping to obtain an optimal non-precious-metal dopant.

3.2.2. Heterogeneous junctions. Heterojunction catalysts, which consist of two different functional materials coupled together, are non-homogeneous phase catalysts that have been widely used in water electrolysis in recent years.¹⁴⁸ Based on the properties of each phase component, heterojunctions can be classified into semiconductor–semiconductor heterojunctions (including type I, type II, type III, PN, and Z-scheme junctions) and semiconductor–metal heterojunctions (including Schottky and ohmic junctions).^149,150 In heterojunction catalysts, spontaneous electron transfer at the interface produces a charged repartition that favours electron transfer reactions over the electrode surface, significantly promotes the reaction kinetics of the entire multistep elementary reaction, and lowers the thermodynamic energy barrier, thus improving the performance of the electrocatalysts. In addition, the intermediates that are readily accessible in the heterojunction can be combined to complete the multielectron reaction synergistically. Thus, heterojunction catalysts are more efficient than single-component catalysts and have better catalytic activity.^151,152

Huang et al.¹⁵³ synthesised a homogeneous PtIr/IrO_x heterojunction catalyst on ultrafine one-dimensional nanowires. By simply balancing the Pt–IrO_x ratio, the HER and OER can be significantly accelerated. Performance tests of the synthesised catalysts revealed that the HER activity of Pt-rich PtIr/IrO_x-30NWs in 1.0 M KOH was approximately 11 times stronger than that of Pt/C, whereas the OER activity of IrO_x-rich PtIr/IrO_x-50NWs was 5 times higher than that of Ir/C. These findings indicate that the formation of heterojunctions in the catalysts can improve the catalytic performance. In addition, heterostructured catalysts exhibit morphological diversity, the specific surface area of the catalyst can be increased, and the utilisation of atoms can be improved by designing porous materials. Wu et al.¹⁵⁴ designed a defect-rich hexagonal MnO_x/RuO₂ heterojunction catalyst. The two-dimensional (2D) hexagonal nanosheet introduced a porous structure to expose the active sites fully and enhance the utilisation of noble metal atoms. Importantly, DFT calculations demonstrated that the O_v and heterointerfaces generated by the defective structure could modulate the electronic structure of Ru, optimise the adsorption energy of *O and *OOH intermediates (* represents the adsorbed species), and inhibit the involvement of lattice O to achieve simultaneous increases in catalytic activity and stability.

The metal–semiconductor Mott–Schottky (M–S) heterojunction has been widely investigated for the preparation of noble metal and metal oxide materials. Wang et al.¹⁵⁵ constructed an M–S heterojunction with Ru–RuO₂ material and loaded it onto N, P co-doped carbon (NPC) to obtain Ru–RuO₂@NPC nanocomposites. Fig. 8a shows a graphical model of the Ru(100)/RuO₂(101) interface, where the heterojunction can be seen, and Fig. 8b presents the atomic configurations of the dissociation steps of water on the surface of the Ru–RuO₂ heterojunction, including H₂O adhesion, H₂O degradation, and H adsorption. The authors found that the HER performance of Ru–RuO₂ was superior to that of Pt/C catalysts and the OER performance was superior to that of commercial RuO₂ catalysts over a wide range of solution pH, so it could be used as a high-performance bifunctional catalyst for the total decomposition of water. Therefore, the source of catalytic activity was determined using DFT calculations. The calculated density of states (DOS) (Fig. 8c) shows that the heterojunction has metallographic properties with zero bandgaps, and the adsorbate bonding pattern on the catalyst surface (Fig. 8d) reveals an upward shift in the d-band energy of the Ru–RuO₂ heterostructure compared to that of pure RuO₂, which suggests an increase in the anti-bonding energy state and an enhancement of the interactions between the adsorbate and heterointerfaces. In addition, the energy maps corresponding to each reaction step in the OER on Ru, RuO₂, and Ru–RuO₂ heterojunctions were calculated and plotted (Fig. 8e), showing that the energy barrier for the formation of *OOH on Ru–RuO₂ heterostructures was relatively low in the decisive step of the OER, which agrees with the experimental results of the improved performance of the OER. The energy pathway of the HER (Fig. 8f) demonstrates that ΔG on the Ru–RuO₂ heterojunction decreases drastically to only 0.05 eV, much lower than its values on the pure Ru (0.33 eV) and RuO₂ (0.12 eV) lattices, which also suggests that preparation of Ru–RuO₂ M–S heterojunctions facilitates the adsorption/desorption of H. The formation of Ru RuO₂M–S heterojunctions is a good example of an improvement in OER performance.

Fig. 8 (a) Schematic structures of a Ru-based catalyst. (b) Evolution of H₂O over catalysts. (c) DOS of a Ru–RuO₂ heterojunction. (d) Schematic illustration of adsorbent bond generation over the catalyst surface. (e) Free energy diagrams for the OER. (f) Energetic pathway of the HER. Reproduced from ref. 155 with permission from Elsevier, copyright 2022.

These findings demonstrate that the construction of heterojunctions can not only improve the specific surface area through the change of morphology, but also can help regulate the electronic structure of the active site, shift the d-band centre, and even reduce the decisive step reaction energy barrier or lower the ΔG, thereby improving the catalytic performance of the material. Heterojunctions play an important role in the synthesis of noble metal oxides.

3.2.3. Defects. Defects are distortions of the perfect state, and in the process of synthesising materials, defects are inevitably generated, which can change the surface properties of the material as well as the structure of the electrons, thus altering the properties of the material. Therefore, defect engineering can be used as an effective improvement strategy. Common defects include vacancy defects,¹⁵⁶ functional group defects,¹⁵⁷ doping defects,¹⁵⁸ edge defects,¹⁵⁹ amorphous defects,¹⁶⁰ boundary defects,¹⁶¹ and strain defects.¹⁶² This review classifies the defects into point defects (Fig. 9(a)), surface defects (Fig. 9(b)) and edge defects (Fig. 9(c)) based on the number of defects as well as their location.

Fig. 9 Three different types of defects. (a) Point defects, (b) surface defects and (c) edge defects.

In the preparation of noble metal and metal oxide catalysts for water electrolysis, scholars generally believe that defects decrease the activation energy required for water adsorption and that they are the source of the high activity of the amorphous oxide structure. Zhang et al.¹⁶³ developed Ta-doped amorphous hydrated IrO_x with abundant O_v (Ta@IrO_x) as a highly active and stable electrocatalyst for acidic OERs. Ta acts as an electron acceptor dopant, optimising the structural morphology of IrO_x and generating surface O_v. Among synthesised catalysts, the best-performing 350-Ta@IrO_x has an overpotential of 223 mV at 10 mA cm⁻² and a mass activity of 1207.4 A g_Ir⁻¹ at 1.55 V, which is 147.7 times higher than that of commercial IrO₂. The generation of this excellent catalytic performance is mainly due to the combined effect of Ta doping and O-deficiency engineering, effectively promoting the nucleophilic attack of H₂O, which accelerates the decisive step in the high catalytic OER activity of Ta-doped IrO_x. Similarly, Zhang¹⁶⁴ synthesised MnRuO_x containing microcrystals that could be used to enhance the activity and stability of O₂ precipitation. They found that the amorphous phase in MnRuO_x introduced a number of defects that created favourable conditions for OH⁻ adsorption and provided a basis for structural reconfiguration of the catalytic material, contributing to its stability. However, the enhancement of the catalytic performance originates not only from the introduction of defects, but also from the formation of crystalline heterojunctions, which interact with each other to optimise the catalyst performance.

In many cases, defect engineering is used in conjunction with other improvement strategies, whereas in other cases, scholars have been able to accomplish catalyst optimisation using defect engineering directly. Wang et al.¹⁶⁵ synthesised the layered Ir salt Na₂IrO₃ (Na-213), which was then protonated and stripped to prepare layered Ir oxide nanosheets (e-H–Na-213). These nanosheets were used to optimise the marginal active sites using in-plane Ir defects to improve the catalytic activity. Fig. 10a shows the N protonation process, in which Na⁺ from the interlayer is replaced by H⁺. In the Na₂IrO₃ layer, the IrO₆ octahedra form a hexagonal honeycomb structure through a co-edge mode, and inside the honeycomb skeleton, the intrinsic Ir vacancies are partially filled with Na atoms (interfacial Na). The HAADF-STEM results for e-H–Na-213 in Fig. 10b clearly reveal the presence of intrinsic Ir vacancies (yellow circles) within the catalyst, and a portion of the Ir vacancies occupied by Na atoms (blue circles) can also be seen. The stripped e-H–Na-213 is still a highly crystalline material, and three sets of crystalline surfaces can be observed in Fig. 10c. The subsequent comparison between e-H–Na-213 and IrO₂ in Fig. 10d shows that the overpotentials required for both OERs decrease with increasing Ir content, and the overpotentials required for e-H–Na-213 are lower than those for IrO₂. The catalytic performance of e-H–Na-213 is superior to that of IrO₂ at Ir loadings of 130 μg cm⁻² (Fig. 10e). The source of the superior catalytic performance is that e-H–Na-213 can form densely stacked catalyst deposits, whereas the IrO₂ nanoparticles converge to accumulate, and only part of the electrode surface is covered under the same Ir loading conditions (Fig. 10f). Second, e-H–Na-213 exhibited an Ir mass activity of 355 Ag⁻¹, which is 16.5 times higher than that of IrO₂ (21.6 Ag⁻¹) and exceeds those of a series of well-designed Ir oxides (Fig. 10g) and the catalyst was able to achieve a stability of 1300 h. These excellent properties enable e-H–Na-213 to solve the problems of low Ir catalyst activity and large electrolysis unit size in the current proton exchange membrane water electrolysis and are expected to be commercially applied. This finding indicates that the presence of in-plane defects can optimise the catalyst performance. Finally, the authors analysed the results of DFT calculations and concluded that the Ir defects within the intrinsic plane are the key to building a distinctive local context of edge active sites with the best surface O adsorption properties and thus superior catalytic activity.

Fig. 10 (a) Schematic diagram of the formation of H–Na-213. Special aberration-corrected HAADF-STEM image (b) and SAED pattern (c) of e-H–Na-213. (d) Relationship between Ir loading and overpotential for e-H–Na-213 and IrO₂. (e) Linear sweep voltammetry (LSV) curves of e-H–Na213 and IrO₂ at a loading of 130 μg cm⁻². (f) Schematic and SEM images of e-H–Na-213 and IrO₂ loaded on Ti foil. (g) Comparison of the Ir mass activity (j_Ir) of e-H–Na-213 with that reported for Ir-based catalysts. Reproduced from ref. 165 with permission from Wiley-VCH GmbH, copyright 2024.

Defect construction is a widely used method in catalyst design. Whether defect engineering is combined with other types of engineering to improve catalytic activity through synergistic effects or defect engineering is used alone to optimise the performance of catalysts, the physicochemistry of the raw material needs to be fully analysed before selecting the appropriate type of defect. Defect engineering is expected to be used more frequently in future studies.

3.2.4. Strain engineering. Strain engineering can change the internal electronic structure of electrocatalysts and is an effective method of optimising catalyst performance.¹⁶⁶ In general, the basic rule of strain engineering is that compression or stretching of the atomic arrangement occurs, which induces a shift in the density centre of the d-orbitals concerning the Fermi energy level and modulates the electronic configuration of the surface, thus, strain engineering plays a key role in modulating the interaction between the catalytic surface and adsorbate molecules.^166–168 The common types of strain engineering are tensile and compressive strain engineering, both of which are widely used in the preparation of noble metal oxides with excellent properties. Su¹⁶⁹ proposed a low-Ir electrocatalyst by localising tensile-strained Ir atoms at Mn oxide surface cationic sites (TS-Ir/MnO₂) to activate the LOM mechanism and achieve extraordinary OER activity. Meanwhile, due to the presence of stretch-strained Ir sites, H₂O adsorption and transformation occurred on the O_v around the Ir sites, which inhibited the involvement of lattice O in MnO₂, enhancing the activity and stability of the catalyst. Synthesis methods and materials are typically used to prepare catalysts with specific strain effects. Huang¹⁷⁰ utilised a burst fire to introduce a specific tensile strain to prepare RuO₂ nanorods loaded on antimony tin oxide (ATO) (s-RuO₂/ATO), which optimised the catalytic performance of RuO₂ by changing its electronic structure.

In particular, the strain present in a catalyst is not constant; it can be affected by the external environment, and real-time strain modulation can be achieved by regularly changing the physical conditions. Du et al.¹⁷¹ found that anisotropic thermal expansion can generate a compressive strain of IrO₆ octahedra in Sr₂IrO₄ catalysts and enable real-time modulation of the compressive strain by changing the temperature. With increasing temperature, the tilt of the IrO₆ octahedra in Sr₂IrO₄ decreases, inducing pronounced lattice distortions, expansions along the a and b axes directions, and contractions along the c direction (Fig. 11a). In addition, the number of unoccupied 5d states of Ir increases, suggesting a higher valence state of Ir (Fig. 11b), and the Ir–O_ab-plane and Ir–O_c-axis bond lengths are both shortened (Fig. 11c). These findings confirm the compressive strain in the IrO₆ octahedron upon heating of Sr₂IrO₄ and the positive shift in the Ir⁴⁺ binding energy (Fig. 11d), suggesting that the compressive strain in the IrO₆ unit favours the stabilisation of the higher valence states of the Ir species. The binding energy of the Ir–O π band increases in Sr₂IrO₄ for the J_eff = 1/2 and J_eff = 3/2 subbands (Fig. 11e), indicating that the thermal strain in the IrO₆ unit shifts the d band downward away from E_f. In addition, as seen from the linear sweep voltammetry (LSV) curves at different temperatures (Fig. 11f), the overpotential of the OER required for this catalyst decreases as the temperature increases, and the catalytic activity increases. The Tafel slope decreased with increasing temperature (Fig. 11g), demonstrating that thermal strain greatly accelerates the OER kinetics. The catalyst breaks the linear Arrhenius relationship (Fig. 11h), illustrating that the instantaneous thermal strain from temperature change can change the electronic state of Sr₂IrO₄, thus causing the OER rate to vary with temperature. The results of free-energy calculations on the thermally strained Sr₂IrO₄ (001) crystalline surface showed that the free energy required for the decisive rate step decreases with increasing temperature (Fig. 11i and j) and that the catalytic activity increases. These findings suggest that heating may be a way to modulate the electronic structure and that more than just the traditional thermal diffusion effect occurs during heating, however, a positive response may also occur that accelerates the slow kinetics of the OER.

Fig. 11 (a) IrO₆ octahedra inside Sr₂IrO₄, which expand along the a and b axes and contract along the c axis. (b) XANES plots of Ir L₃-edges of Sr₂IrO₄ at different temperatures. (c) k³-weighted FT curves corresponding to the data in (c). (d) Ir_4f XPS spectra. (e) Valence band spectra of Sr₂IrO₄ under different temperatures. (f) LSV curves were normalised by electrochemically active surface area (ECSA) for Sr₂IrO₄ in 1.0 M KOH with increasing temperature. (g) Temperature-dependent Tafel slopes of Sr₂IrO₄. (h) Calculated ln j₀–T⁻¹ plot for the OER on Sr₂IrO₄. (i) and (j) ΔG of adsorbed intermediates at 25 °C (i) and 90 °C (j). Reproduced from ref. 171 with permission from Springer Nature, copyright 2024.

Strain engineering is effective for changing electronic structures and has long been a popular strategy for improving catalyst performance. This approach has been increasingly applied in the preparation of noble metal oxides. However, the use of constant strain fails to meet the needs of people because of differences in their environments, which may affect the catalyst performance. Therefore, catalysts with good activity in all environments must be found, and strategies such as thermal strain application, which can regulate instantaneous strain, may become popular research topics in the future.

Pure noble metal catalysts have excellent catalytic activity and long-term stability in the two half-reactions of water electrolysis. However, the cost of preparing the catalysts has largely limited the development of water electrolysis technology. To overcome this problem, noble metal oxides have been proposed, which can significantly reduce the amounts of noble metals used in catalysts. However, achieving a balance between catalytic activity and stability using noble metal oxides alone is difficult. Fortunately, various strategies to optimise the catalyst performance have been discovered by an increasing number of scholars, increasing the possibility of the wide application of noble metal oxides in the future. Among them, doping engineering, heterojunction, defect engineering, and strain engineering are the most frequently used for the preparation of these catalysts. Of course, not all catalysts can be improved through these improvement strategies, and appropriate improvement strategies must be selected according to the inherent properties of the catalysts. Only in this way can the stability of such catalysts be maintained for a long time of stability after improving their catalytic activity. In the future, the performance of catalysts can be optimized instantaneously, depending on the environment, rather than performing a single optimisation.

4. Catalytic properties of noble metal oxides

To date, researchers have conducted a considerable amount of research to explore catalysts that can replace pure noble metals and have found that noble metal oxides are excellent alternatives that not only reduce the amounts of noble metals used, but also provide excellent catalytic activity. Based on previous studies, they can be classified into three categories: modified noble metal oxides, noble metal–transition metal oxides, and noble metal oxides–transition metal oxides. Next, this review introduces the OER and HER performances of these three types of catalysts.

4.1. Modified noble metal oxides

The emergence of noble metal oxides has contributed significantly to the reduction of the cost of noble metal-based catalysts, the most common being RuO₂ and IrO₂, which are good catalysts for the OER. However, some of their inherent defects have limited large-scale applications; therefore, in recent years, research has been devoted to the search for more stable Ru- and Ir-based catalysts with better activity. Other noble metal oxides with good catalytic properties are also being explored.

4.1.1. Ru oxides. Ru oxides are often used as catalysts for the OER because of their good catalytic activity in this reaction, and only a few studies on H₂ precipitation have been conducted. One of the more common research strategies is to change the charge distribution around the Ru active sites by elemental doping to improve the catalyst performance. Deng et al.¹⁷² proposed preparing robust W–Ru oxide heterostructures [(Ru–W)O_x] by performing successive rapid oxidation and metal thermal migration processes and introducing valence oscillations of the Ru site induced by high-valence W species during the OER process. These oscillations promoted the cyclic transition of the active metal oxidation state and maintained the continuous operation of the active site. In addition, the preferential oxidation of the W species and electron gain of the Ru site significantly enhanced the stability of RuO_x, which exhibited excellent stability for 300 h in an acidic electrolyte. Metal doping may stimulate the lattice O oxidation mechanism while increasing the catalytic activity, thereby reducing the stability of the catalyst to some extent. Wu et al.¹⁷³ doped Ni into RuO₂ and improved its stability by an order of magnitude. Metal ions can even be doped to change the magnetic properties and thus optimise the catalyst performance. Li et al.¹⁷⁴ generated a net ferromagnetic moment in antiferromagnetic RuO₂ by doping dilute Mn (Mn²⁺) (S = 5/2) to enhance the OER activity in acid electrolytes, which is known as a spin-polarisation-mediated strategy. Fig. 12a shows the electrochemical test setup with an applied magnetic field (330.9 mT), which is required to synthesise Mn–RuO₂NFs. The LSV plot in Fig. 12b shows that Mn²⁺ doping reduces the overpotential required for the catalysts, suggesting enhanced catalytic activity. A stability test of the catalysts with and without the magnetic field was conducted, and the results are presented in Fig. 12c. Mn–RuO₂NFs/M with a magnetic field applied shows almost no change in overpotential after 1000 cyclic voltammetry tests, indicating that its stability is better than that in the absence of a magnetic field. The catalyst was finally measured to be able to maintain its stability for 480 h in the presence of a magnetic field, which further illustrates that a magnetic field can improve the stability of the catalyst.

Fig. 12 (a) Schematic illustration of a three-electrode system with an applied magnetic field. (b) LSV curves of Mn–RuO₂NFs/M, Mn–RuO₂NFs, RuO₂NFs, and IrO₂ in 0.5 M H₂SO₄. (c) Overpotentials at the current density of 10 mA cm⁻² of Mn–RuO₂NFs/M and Mn–RuO₂NFs before and after 1000 cycles. Reproduced from ref. 174 with permission from Wiley-VCH GmbH, copyright 2023. (d) and (e) Schematic diagrams of dynamic electron transfer in Re_0.06Ru_0.94O₂. (f) LSV curves and (g) Tafel slope plots for Re_0.06Ru_0.94O₂, RuO₂, and commercial RuO₂ in O₂-saturated 0.1 M HClO₄. (h) Constant-current chronopotentiometric stability measurements at an anodic current density of 10 mA cm⁻² for Re_0.06Ru_0.94O₂ and RuO₂. (i) Dissolved Ru and Re ion concentrations in an electrolyte for Re_0.06Ru_0.94O₂ and RuO₂ determined via ICP-MS. Reproduced from ref. 175 with permission from Springer Nature, copyright 2023.

Jin¹⁷⁵ proposed the synthesis of Re_0.06Ru_0.94O₂ electrocatalysts using the noble metal Re as the dopant. Unlike the static doping of conventional dopants, the doping of Re introduces a dynamic process that accepts electrons at the in situ potential to activate the Ru sites, returns electrons, and prevents Ru solvation at large overpotentials (Fig. 12d and e), which adaptively improves the activity and stability. The LSV curves and corresponding Tafel slopes of Re_0.06Ru_0.94O₂, RuO₂, and commercial RuO₂ are given in Fig. 12f and g, showing that Re_0.06Ru_0.94O₂ has an overpotential of 190 mV at a current density of 10 mA cm⁻² and a Tafel slope of 45.5 mV dec⁻¹, which is superior to those of RuO₂ and C–RuO₂. The stability test results demonstrate that the potential of Re_0.06Ru_0.94O₂ remains stable for 200 h of continuous testing, whereas that of RuO₂ exhibits rapid activity decay within 19 h. Thus, the introduction of Re improves the stability of the catalyst. In addition, the concentrations of Ru and Re in the electrolyte were tested over time (Fig. 12h). RuO₂ shows a 2.7% loss of Ru over 20 h, whereas Re_0.06Ru_0.94O₂ exhibits only a 0.34% loss of Ru and a 0.62% loss of Re after 200 h of stability testing. Thus, not all metal doping causes a loss of catalyst stability. However, noble metal doping inevitably increases the cost of catalysts, which is contrary to the original purpose of reducing the production cost of catalysts and may have certain limitations in future large-scale applications.

The doping strategy for Ru oxides can be implemented by doping not only metallic elements, but also non-metallic elements.⁹⁰ Non-metallic elements with small atomic radii can even be doped into the interstitials of RuO₂, which can improve stability without reducing the number of active sites. In addition to doping, an electron reservoir can be introduced into Ru oxides to provide a constant supply of electrons to the active sites to avoid overoxidation and catalyst spillage. Hao et al.¹⁷⁶ chose boron nitride (BN) as an electron reservoir capable of providing and receiving electrons instantaneously and loaded RuO₂ onto ultrathin BN nanosheets, which successfully improved the stability of RuO₂. However, not all carriers are used to provide electrons to the active site; some carriers are designed to hold the active site in place. Yao¹⁷⁷ developed a robust metal–organic framework anchoring strategy by stabilising atomically isolated Ru oxides on UiO-67-bpydc (bpydc = 2,2′-bipyridine-5,5′-dicarboxylic acid) using strongly coordinating pyridine ligands and found that the Ru–N bond between Ru oxides and UiO-67 bpydc not only accelerates the involvement of lattice O in the OER process, but also stabilises the soluble *V_O–RuO₄²⁻ intermediate, resulting in improved OER performance and stability for up to 115 h. The Ru oxide can be used as an intermediate in the OER. In addition, suitable defects and interfaces can be designed to improve the stability of Ru oxides. Chen et al.¹⁷⁸ proposed a bicontinuous nanoreactor consisting of multiscale defective RuO₂ nanomonomers (MD–RuO₂–BN), which provide abundant active sites and fast mass transfer capability through cavity-limited effects. The presence of vacancies and grain boundaries makes MD–RuO₂–BN rich in low-coordinated Ru atoms, which attenuate the Ru–O interaction and inhibit the oxidation of lattice O and dissolution of high-valent Ru, thereby improving the stability of the catalyst.

Ru oxides have been the star catalysts for OER rate enhancement of electrocatalytic water decomposition by virtue of the abundance of active sites on the noble metal ruthenium, and researchers have been enthusiastically working on this type of catalyst, which is expected to lead to breakthroughs in the future. However, although Ru oxide catalysts have excellent catalytic activity, the LOM mechanism inevitably occurs during realistic water resolution of oxygen, triggering lattice oxygen to participate in the reaction. The continuous transition of lattice oxygen between the dissociated and replenished forms will cause the catalyst to collapse and cause the catalyst to reconfigure, thus blocking the mutual contact between the active sites of the catalyst and the electrolyte, and even allowing a large number of ruthenium ions, which are rich in active sites, to leach out of the electrolyte, resulting in the drastic attenuation of the catalytic activity. Therefore, the excellent catalytic activity of ruthenium oxide electrocatalysts at the start is not enough to meet the demand, but it is also necessary to have very good stability to maintain the excellent catalytic activity. Even though a lot of research has been done to improve the stability of these catalysts, most of the published results are in the range of 100–300 h, and very few studies have been done beyond 500 h or even 1000 h. Although it is time-consuming and costly to test the stability of catalysts over long periods, future research in the area of catalytic activity will be able to maintain the catalytic activity of ruthenium oxide electrocatalysts for a much longer period. The development of ruthenium oxide electrocatalysts with ultra-long catalyst activity will lead to further developments in water electrolysis.

4.1.2. Ir oxides. Ir oxide is a relatively stable electrocatalyst that has great advantages in commercial applications. The current more popular research on iridium oxide electrocatalysts tends to improve the catalytic activity. It is well known that among the two mechanisms (the AEM and the LOM) in the OER, the catalytic activity of the catalyst will be better when the reaction tends to occur using the LOM mechanism, so the probability of the LOM mechanism can be increased by rational design to improve the catalytic activity. Wang et al.¹⁷⁹ induced the occurrence of the LOM mechanism by doping the large-ionic-radius Nd into IrO₂ crystals and introducing a large amount of O_v, which exhibited an overpotential of 263 mV at a current density of 10 mA cm⁻².

Increasing the catalytic activity of the catalyst can also be achieved by lowering the decisive rate step energy barrier for the OER of electrolytic water. Kuang et al.¹⁸⁰ synthesised solid-solution structured ultrafine IrMnO₂ nanoparticles doped with Sr on C nanotubes (Sr–IrMnO₂/CNTs), which can be driven by an overpotential of only 236.0 mV to 10.0 mA cm⁻². Sr–IrMnO₂/CNTs had excellent stability (>400.0 h) and 39.6 times more Ir mass activity than IrO₂ at 1.53 V. DFT calculations show that the solid-solution structure promotes strong electronic coupling between Ir and Mn and lowers the energy barrier of the decisive OER step, thus providing high catalytic activity. Liu et al.,¹⁸¹ on the other hand, proposed a twisting strategy and successfully prepared a new class of CdRu₂IrO_x nanomaterials with twisted structures. These materials exhibited excellent stability of the OER at 10 mA cm⁻² in 0.5 M H₂SO₄, as well as an ultra-low overpotential of 189 mV and ultra-long stability of 1500 h for the OER. These characteristics were obtained mainly because the cooperative interaction between meso-Ru and Ir leads to the distortion of the Ru–O, Ir–O, and Ru–M (M = Ru, Ir) bonds, and this distorted structure can reduce the energy barrier of the decisive velocity step in the OER process.

The slow kinetics of the OER can be accelerated by increasing the flow rate of H⁺ inside the catalyst. Fan et al.¹⁸² synthesised a substable 3R-phase IrO₂ (3R-IrO₂) electrocatalyst with Ir vacancies by using a microwave-assisted mechanothermal method. In this catalyst, the co-rimmed (IrO₆) octahedra in 3R-IrO₂ formed a 2D layer (Fig. 13Aa), and three such layers were stacked to form the final material (Fig. 13Ab), which enabled the introduction of new active sites. In addition, due to the presence of Ir vacancies within the layers, H⁺ can diffuse rapidly between (Fig. 13Ac) and within (Fig. 13Ad) the layers. Furthermore, 3R-IrO₂ exhibits an overpotential of 188 mV at a current density of 10 mA cm⁻², which is much lower than that required for rutile-type IrO₂. After 511 h of testing at a current density of 10 mA cm⁻², 3R-IrO₂ maintained good stability (Fig. 13Ae), and even at current densities of 20 and 40 mA cm⁻², the stability was better than that of C–IrO₂ at 10 mA cm⁻². The mass activity of the catalysts did not deteriorate significantly after the stability test and remained high, and this phenomenon demonstrated the value of 3R-IrO₂ for long term use. Thus, by cleverly designing the structure of the catalyst, not only can new active sites be introduced, but also the H⁺ flow rate can be increased, which can effectively improve the performance of the catalyst. Chong et al.¹⁸³ rationalised the morphology and local geometrical ligand environments of Ir and Ru catalysts by using defect-rich, La- and Li-doped Co₃O₄ nanofibres (LLCF) as substrates to promote the electrocatalytic OER. They found that mass activities of LLCF were 26 and 50 times higher than those of commercial IrO₂ and RuO₂, respectively. In recent years, many studies have been devoted to the development of crystalline Ir oxides,^184–186 and because of their ordered structures, the stability of such catalysts has improved.¹¹⁵ In contrast to an ordered structure, which can lead to stability, an amorphous disordered structure results in better catalytic activity. Lee et al.¹⁸⁷ suggested that substable nanoporous and amorphous Ir oxides may indeed provide lower OER overpotentials, which reconciles their superior OER catalytic performance. Liang et al.¹⁸⁸ revealed the effects of the active site density and energetics on the Ir oxide water oxidation activity and proposed that in amorphous IrO_x, O* exhibits repulsive adsorbate–adsorbate interactions and that as their coverage increases, the O–O bond formation becomes easier, thus lowering the energy barrier required for the decisive speed step. Similarly, to reduce the adsorbate binding energy, Zhao et al.¹⁸⁹ instead doped Co into SrIrO₃ and used Co to optimise the adsorbate binding energy of IrO_x and to obtain a high electrocatalytic activity for the OER.

Fig. 13 (A) (a) Top view of the crystal structures of 3R-IrO₂. (b) Crystal structure of 3R-IrO₂. (c) and (d) H⁺ transportation pathway along interlayers in 3R-IrO₂. (e) Stability of 3R-IrO₂ by the chronopotentiometry technique at the constant current densities of 10, 20, 40, and 100 mA cm⁻² and C–IrO₂ at 10 mA cm⁻². Reproduced from ref. 182 with permission from Elsevier, copyright 2021. (B) (a) LSV curves of KIr₄O₈ and IrO₂ in 0.5 M H₂SO₄ solution toward the OER. (b) Tafel slope plots of KIr₄O₈ and IrO₂. (c) Proposed mechanism for KIr₄O₈ toward the OER. (d) and (e) Gibbs free energy diagrams at Ir sites of KIr₄O₈ and IrO₂ for the OER. Reproduced from ref. 190 with permission from Wiley-VCH GmbH, copyright 2024.

The energy barrier of the decisive step can be reduced not only by changing the energy barrier of the original decisive step, but also by reconstructing a decisive step with a lower energy barrier. Li et al.¹⁹⁰ designed a KIr₄O₈ nanowire electrocatalyst with more exposed active sites and enriched hydroxyl groups, and the LSV curves in Fig. 13Ba demonstrate that KIr₄O₈ possesses excellent OER activity with an overpotential of 266 mV at 10 mA cm⁻², which is 48 mV lower than that of IrO₂. In addition, the Tafel slope of KIr₄O₈ is 48.2 mV dec⁻¹ (Fig. 13Bb), which is significantly smaller than that of IrO₂, suggesting that KIr₄O₈ has fast catalytic kinetics in the OER process. This promising catalytic activity is mainly because KIr₄O₈ is a hydroxyl-rich substance, and the direct involvement of hydroxyl groups in the reaction during water electrolysis lowers the energy barrier of the decisive speed step. In the presence of this catalyst, the initial step of water electrolysis is deprotonation of the structural hydroxyl groups (Fig. 13Bc). Examination of the Gibbs free energy diagrams of IrO₂ (Fig. 13Bd) and KIr₄O₈ (Fig. 13Be) reveals that the rate-controlling step of IrO₂ is the conversion of O* into OOH*, with a required energy barrier of 1.82 eV. In addition, the rate-controlling step of KIr₄O₈ is the conversion of OH* into O*, which is intrinsic to the catalyst, with a required energy barrier of 1.62 eV. Furthermore, the energy barrier for KIr₄O₈ is less than that of IrO₂, suggesting that KIr₄O₈ has a smaller energy barrier than IrO₂ and indicating that KIr₄O₈ has high OER catalytic activity.

Among the noble metal-based oxide electrocatalysts for the OER of electrolytic water, Ir oxide and Ru oxide are the two major research hotspots. Compared with Ru oxide electrocatalysts, Ir oxide electrocatalysts have much higher stability and poorer catalytic activity. Therefore, to achieve the expected catalytic effect, the content of the noble metal Ir needs to increase, which will increase the production cost to a large extent due to the scarcity of the noble metal, which is very unfavourable to the development of Ir oxide electrocatalysts. In addition, the larger volume will also be inconvenient for industrial applications as well as storage. Therefore, improving the mass activity of the catalyst is crucial for the development of Ir oxide electrocatalysts. There are many contemporary studies to reconstruct the structure of Ir-based oxide electrocatalysts by doping metal ions, which induces such catalysts to be more inclined to undergo the LOM mechanism during the electrolysis of water, thus increasing the catalytic activity of the catalysts. With the increasing chance of the LOM mechanism, the risk of catalyst reconfiguration and collapse in the process of catalyst use increases, and therefore a sacrifice in terms of stability is required to improve the catalytic activity, which does not fully demonstrate the stability advantages of iridium-based oxides. Of course, not all doping methods are used for the OER mechanism, there are also many studies under the condition of not changing the original reaction mechanism, through some distortions, defects and amorphous structure design, to reduce the reaction energy barrier of the original reaction mechanism of the deceleration step, or even reconstruct a deceleration step with a lower required energy barrier, which improves the catalytic activity without sacrificing the stability. In the future, more researchers will be devoted to the study of iridium oxide noble metal oxides that maintain or even improve the original stability and increase the catalytic activity.

4.1.3. Other noble metal oxides. Ru and Ir oxides are primarily used in the OER, whereas other noble metal oxides are also employed in the HER. In particular, Pt is widely utilised in the HER due to its abundant active sites and good electrical conductivity, and the corresponding Pt oxides also exhibit good H₂ precipitation activity. Yu et al.¹⁹¹ investigated whether Pt still has the ability to catalyse water decomposition through the HER after oxidation, and found that the Pt–O bond produced after Pt oxidation is a new active site for the HER, which is still able to achieve the effect of water electrolysis. Researchers have even found that the HER activity of oxidised Pt is superior to that of metallic Pt⁰.¹⁹² Yang et al.¹⁹² synthesised a new 2D substable Pt oxide (1T-PtO₂), which, in the acidic HER, required a low overpotential of only 12 mV at a current density of 10 mA cm⁻². DFT calculations showed that the (Pt–O) active site is easily attacked by H⁺ to form the intermediate state (Pt–H), which readily combines with any H atoms on the neighbouring O to form H₂, thereby increasing the catalytic activity. In addition, the ability to maintain stability for 100 h at a current density of 10 mA cm⁻² breaks the conventional perception that Pt oxides are very unstable during water electrolysis. It can be seen that when using Pt to improve the catalytic performance of the HER with other catalysts, it is not necessary to reduce Pt to the metallic state Pt⁰, and the presence of suitable Pt oxides may lead to similar or even better catalytic effects. This would provide an alternative approach to the development of Pt-based catalysis. This finding explains the relationship between the Pt–O structure and catalytic activity.

The noble metal Pd, as a Pt-group noble metal material, is also rich in active sites. In published studies, the introduction of a small amount of Pd into transition metal-based materials can lead to excellent catalytic performance. Choi et al.¹⁹³ have used Pd to construct active sites on nickel oxide (NiO) nanoplates to obtain excellent HER performance. However, the application of Pd oxides for water electrolysis is minimal. Li et al.¹⁹⁴ combined metallic Pd with Pd oxides for the preparation of H₂ sensing materials, and Ebenezer et al.¹⁹⁵ used Pd–PdO as an electrocatalyst for nitrate reduction. Lee et al.¹⁹⁶ synthesised Pd/PdO–NiPh catalysts using Pd/PdO modification of phosphate groups in nickel phosphate through metal-carrier interactions and obtained good hydrogen precipitation performance under hydrazine assistance. It can be seen that Pd oxides have some electrocatalytic properties of their own, and further study of them is likely to lead to discoveries similar to those for Pt oxides.

The frequency of noble metal Rh in electrocatalytic water decomposition has been increasing in recent times, and Rh oxides have received more and more attention from researchers due to their good catalytic activity. Manna et al.¹⁹⁷ synthesised a prepared Pt/Rh oxide–N-doped carbon composite (Pt/Rh₂O₃–CN_x), where during H₂ precipitation, H₂O dissociates at Pt to form Pt-adsorbed H* species (Pt–H). Some of the H moves to Rh₂O₃ and reacts with neighbouring H⁺ or adsorbed OH⁻ to form H₂ or H₂O, respectively, and the catalytic performance of Pt/RhO–CN was about 3.5/6 times higher than that of the Pt/C catalyst. The fact that the catalytic performance of the HER can still be increased with the reduction of the noble metal Pt suggests that the existence of active sites in Rh oxides is expected to be more widely used in future water electrolysis projects. The original OER process in conventional water electrolysis requires a high applied voltage for total water decomposition because of the slow kinetics of the four-electron process that needs to occur. If the conventional OER can be replaced by organic molecule oxidation, it will make total water decomposition much easier. Shi et al.¹⁹⁸ prepared suitable tripartite Rh oxide (Tri-RhO) electrocatalysts with adjustable lattice strain, which accelerated the slow kinetics of water decomposition through hydrazine oxidation, exhibited an overpotential of 103 mV at a current density of 10 mA cm⁻², and were able to remain stable for 72 h. Gao et al.¹⁹⁹ synthesised Rh–RhO nanoclusters and obtained trifunctional electrocatalysts with excellent performance in the HER, the hydroxidation reaction (HOR), and the hydrazine oxidation reaction (HzOR) through the modulation of surface oxygen vacancy content. It can be seen that combining polymer oxidation with water electrolysis can reduce the cost of electricity consumption of water electrolysis to a great extent. Rh oxides have great catalytic potential for anodic oxidation, and will become a promising class of electrocatalysts after optimisation to improve stability.

The noble metal Ag is a bit cheaper compared to other noble metals, and if it can be used to replace other noble metal materials, it can greatly reduce the production cost of catalysts, which is of great potential value. Raghav et al.²⁰⁰ prepared bifunctional electrocatalysts by sputtering Ag₂O/AgO on flower-like Mn–Co–Cu ternary metal oxides using a DC magnetron sputtering method, and the catalyst showed excellent OER catalytic performance, with a fully hydrolytic current density of 10 mA cm⁻² achieved by supplying only 1.37 V. In some recent studies, few electrocatalysts existed with Ag oxide alone, and most of them used Ag oxide in combination with other transition metal oxides. It is very likely that fewer active sites on Ag can catalyse water decomposition, and synergistic interactions with the active sites of other metals are needed to increase the catalytic activity. Compared with other noble metal oxides, the research on Ag oxide electrocatalysts in the field of water electrolysis is still relatively limited, and there is much room for development.

Optimisation of noble metal oxides, among which the most common choices are Ru and Ir oxides, as substrates in recent years has yielded good results. Other noble metal oxides are also being explored by researchers, but particularly significant results have not yet been achieved; hence, research is still mainly focused on Ru-based and Ir-based catalysts. Table 2 lists some noble metal oxide catalysts discovered in recent years. Fortunately, the further development of emerging Pt oxide, Pd oxide, Rh oxide and Ag oxide electrocatalysts is expected to bring new breakthroughs in the development of electrocatalysts that take into account the stability and catalytic activity of the catalysts. However, a large portion of the noble metal-based oxide electrocatalysts have poor stability problems, especially the Ru oxide electrocatalysts, which had serious metal ion leaching during the process of use, and if no follow-up treatment is done, a lot of noble metals will be wasted, which will aggravate the environmental problem of resource shortage. Therefore, whether it is to improve the original noble metal-based oxide-based electrocatalysts or to develop new catalysts, the recycling of catalysts, especially the recovery of noble metals, is an issue that needs to be focused on in future research.

Table 2 Noble metal oxide catalysts

Catalysts	Reaction	η 10 (mV)	Tafel slope (mV dec^−1)	Stability (h)	Ref.
Ca2Y0.2Ir0.8O4	OER	213	56.6	168η10	201
Si–RuOx@C	OER	220	53	100η10	202
La–RuO2	OER	208	57.4	28η10	203
	HER	71	49.3	28η10
PbO2 + Ag	OER	454	310	5η20	204
Bi0.15Ru0.85O2	OER	200	59.6	100η10	205
Er–RuOx	OER	208	45	100η500	206
Ru/HfO2	OER	39	29	28η10	207
py-RuO2:Zn	OER	173	41.2	1000η10	208
Ru0.5Ir0.5O2	OER	151	45	618.3η10	209

---

4.2. Noble metal–transition metal oxides

The combination of noble metals and transition metal oxides to form an active catalyst is effective. First, it can reduce the use of noble metals to lower the production cost. In addition, some transition metal oxides also have certain active sites, but due to the very unstable nature of these oxides, they have not been widely promoted. The introduction of noble metals can enhance the stability of these catalysts.

The most common of these catalysts are SACs, among which the SA Pt catalyst is the most widely used. Wei et al.²¹⁰ synthesised a PtSAC (PtSA–Mn₃O₄) by precisely anchoring Pt atoms in situ on spinel Mn₃O₄, which modulated the electronic structure through the strong interactions generated between Pt and Mn₃O₄, thus optimising the binding strength of the d-band centre and the intermediate and yielding excellent alkaline H₂ precipitation activity. In addition, using noble metal atoms to optimise the properties of non-noble metal oxides, Yan et al.²¹¹ used O_v to populate Pt atoms uniformly in Ni vacancy (Ni_v) sites with double-deficient NiO (D-NiO–Pt) (Fig. 14a), which ultimately resulted in the free energies of H adsorption on both Ni and Pt sites being closer to 0. Fig. 14b compares the X-ray photoelectron spectra (XPS) of Pt Pt_4f and NiOPt, showing that the valence of Pt²⁺ in D-NiO–Pt is reduced, suggesting that the vacancy induces a strong interaction between the Pt site and the NiO substrate. Analysis of the distribution function indicated that the atomic minimum spacing of D-NiO–Pt was smaller than those of pure NiO and D-NiO, increasing the appearance of tiny peaks near the main peak and confirming the filling of the lattice with Pt cations (Fig. 14c). The synchrotron X-ray absorption spectroscopy (XANES) analysis results in Fig. 14d demonstrate that the valence state of Pt in D-NiO–Pt is between 0 and +4. D-NiO–Pt exhibits excellent HER activity (Fig. 14e) with an overpotential of only 20 mV at a current density of 10 mA cm⁻², which is superior to those of the other catalysts, suggesting a synergistic effect of the O_v and Pt populations on the reaction kinetics. D-NiO–Pt exhibits the smallest Tafel slope of 31.1 mV dec⁻¹ (Fig. 14f), demonstrating that the Volmer–Tafel mechanism is greatly improved with a shift from a rate-determining step to a Tafel step. D-NiO–Pt has the minimum R_ct (0.73 Ω), which confirms that the Pt filling in Ni_v with additional O_v binding favours the kinetic behaviour of the HER (Fig. 14g). In the stability test results, D-NiO–Pt shows almost no decay during the 100 h chronopotentiometry study (Fig. 14h), indicating good stability. The doped Pt atom reduces the formation energy of Ni_v to increase the crystal stability and also binds to O_v to modulate the electronic structure of the surrounding Ni sites. Notably, O_v plays a very important role in such catalysts for the HER.

Fig. 14 (a) Schematic synthesis of D-NiO–Pt. (b) XPS spectra of Pt_4f. (c) Analysis of the sample as a function of the distribution of pairs. (d) XANES spectra of different samples on the Pt L₃-edge. (e) LSV curves. (f) Tafel slopes. (g) Nyquist plots. (h) Continuous 100 h long-term stability test. Reproduced from ref. 211 with permission from Wiley-VCH GmbH, copyright 2022.

In addition to Pt, the noble metal Pd is also often used to prepare SAC electrocatalysts. Li et al.²¹² prepared PdSA–Co₃O₄ electrocatalysts by synergistically coordinating Pd single atoms with Co₃O₄ nanosheets to generate octahedral Pd–O–Co active units, and they exhibited good catalytic activity and maintained their stability for 80 h in the electrolysis of neutral seawater. Wang et al.²¹³ deposited ultrafine nanoparticles of Pd onto TiO₂ and also obtained good alkaline hydrogen precipitation properties. Furthermore, Li²¹⁴ synthesised Ru/MoO_2–x, in which Ru was doped into MoO₃, inducing a transition from MoO₃ to MoO₂ and generating O_v and Ru–O–Mo sites. The electrocatalytic activity of the resulting catalysts for the HER was enhanced by the synergistic effect of the Ru nanoclusters, Ru–O–Mo sites, and O_v-enriched MoO₂. For the OER, the combination of noble and non-precious-metal oxides can not only enable full use of the active sites on the metal, but also activate the active sites on the lattice O to increase the catalytic performance. Wang et al.²¹⁵ loaded Ir species on Ni-doped Co₃O₄ (Ir/Ni–Co₃O₄), in which the Co sites were the most favourable for promoting the OER through the adsorbate evolution mechanism, and Ni and Ir were the most favourable for promoting the OER through the adsorbate evolution mechanism. Ni sites synergistically interact with Ir atoms to shift the energy position of the O_p band centre upward, resulting in the activation of lattice O connecting the Ni and Ir atoms, which participate in the OER by coupling with the adsorbed O on the Ir site. Ir/Ni–Co₃O₄ only needs to provide an OER overpotential of 177 mV at the corresponding current density of 10 mA cm⁻². However, the poor stability of the catalysts after the involvement of lattice O in the reaction requires further investigation. O not only appears as a companion of non-precious metals in this type of catalyst, but also can play a certain role itself. Therefore, the full study and utilisation of O_v and lattice O may continue to be a research hotspot in future studies on water decomposition electrocatalysts.

In addition, combining noble metals with transition metal oxides can introduce heterojunctions that modulate the distribution of electrons, thereby improving catalyst performance. Peng et al.²¹⁶ modified Ru species onto W oxides to prepare Ru–WO_2.72 electrocatalysts and constructed Mott–Schottky heterojunctions, which induced electron enrichment and increased the H₂ yield. The Ru–WO_2.72 exhibited 161.6 times the mass activity of commercial Ru/C with an overpotential of 40 mV at a current density of 10 mA cm⁻². Chen et al.²¹⁷ designed a through-hole Ru/MoO₂ domain-limited heterostructure for the HER at a universal pH, where the Ru–O–Mo bridge formed could regulate the electron transport at the interface, and the strengthened metal Ru–Ru bond could induce H⁺ adsorption and transfer in the kinetics. Ultimately, the accelerated HER kinetics was attributed to strong mass transfer due to the H₂ overflow effect at the interface of Ru and MoO₂. Also accelerating the slow HER kinetics through enhanced mass transfer, Chen et al.²¹⁸ integrated RuNPs with O-deficient WO_3–x, which resulted in a 24.0-fold increase in HER activity compared to that of commercial Ru/C in a neutral electrolyte. This excellent catalytic activity is mainly due to the large capacity of WO_3–x to store H⁺, which can be transferred to RuNPs at the cathodic potential and increase the H₂ coverage on the surface of the RuNPs, thus changing the decisive step of the HER on Ru to H₂ complexation. Hence, increasing the transfer capacity of H⁺ on the catalyst can effectively improve the HER activity of the catalyst.

Although catalyst loading is an effective way to combine noble metals with transition metal oxides for interfacial engineering, cluster heterojunctions can deliver better catalytic performance than loaded catalysts. Zhang et al.²¹⁹ synthesised a cluster heterostructured catalyst (Ru–CrO_x@CN) consisting of crystalline Ru and amorphous Cr oxide clusters with excellent properties of alkaline H₂ electrocatalysis. In loaded catalysts (Fig. 15a), the transfer of active metal species from SAs to clusters to NPs is achieved through carrier–metal interactions in large catalysts. In contrast, in cluster heterojunction catalysts, strong cluster–cluster interactions occur (Fig. 15b), where Ru atoms from the Ru clusters can cross the boundary into CrO_x, improving the interfacial charge redistribution and thus optimising the adsorption energy of the reactive intermediates on each cluster. The HAADF-STEM images in Fig. 15c–e illustrate that most nanoclusters in Ru–CrO_x@CN exhibit distinct cluster–cluster interfaces. In the HER, the overpotential of Ru–CrO_x@CN is 7.0 mV at 10 mA cm⁻², significantly smaller than those of Ru@CN and Pt/C, and the mass activities of Ru–CrO_x@CN are 2.1 and 23.0 times higher than those of the other catalysts, respectively (Fig. 15f–h). The Tafel slopes of Ru–CrO_x@CN of 30.1 mV dec⁻¹ are smaller than those of the other two catalysts (Fig. 15i), suggesting that the deceleration step of Ru–CrO_x@CN may be transformed from a slow Volmer step into a Tafel step. The HER activity of Ru–CrO_x@CN is higher than that of Pt-group metal-based catalysts (Fig. 15j). The native HER activity was evaluated by calculating the transition frequency (TOF) value, and the TOF of Ru–CrO_x@CN was found to be higher than those of most reported HER catalysts (Fig. 15k). Thus, Ru–CrO_x@CN has good HER catalytic activity and the cluster heterojunction is indeed effective in promoting the HER rate; however, this catalyst remains stable for only 20 h, which requires further improvement.

Fig. 15 (a) Commonly supported catalysts. (b) Cluster heterostructured catalyst. Representative (c) and enlarged (d) HAADF-STEM images of Ru–CrO_x@CN. (e) Inverse fast Fourier transform pattern of (d). (f) LSV curves of different catalysts in a N-saturated 1 M KOH electrolyte. (g) LSV curves for various mass standardised catalysts. (h) Comparison of overpotential at 10 mA cm⁻² and mass activity at 100 mV of different catalysts. (i) Tafel plots. (j) Overpotential and Tafel slope of various Ru-based catalysts. (k) TOF values of different catalysts. Reproduced from ref. 219 with permission from Springer Nature, copyright 2024.

By doping transition metal oxides with noble metals, different effects can be obtained, depending on the nature of the material chosen. The bond lengths of oxides can be altered by doping with noble metals to tune the electronic structure further, or the noble metals can be partially oxidised to higher valence states with other non-precious-metal oxides, thus effectively increasing the electronic conductivity and optimising the energy barriers.²²⁰ In addition, tuning the macrostructure of active-site-rich composites, such as by designing hollow structures,²²¹ can sufficiently improve the utilisation of active sites. Even the formation of short-range metal arrays by tightly controlling the spacing between noble metal atoms attached to the oxides can alter the OER pathway, thus improving the catalytic efficiency.²²² It is exciting to note that the introduction of noble metals into the catalyst will change the performance of the catalyst so much that it can even be used to catalyse the electrolysis of seawater and the degradation of plastics at the same time, solving the environmental problems in the process of hydrogen production, which will bring a huge industrial value.²²³Table 3 compares some noble metal–transition metal oxide catalysts synthesised using other improved strategies.

Table 3 Noble metal-transition metal oxide catalysts

Catalysts	Reaction	η 10 (mV)	Tafel slope (mV dec^−1)	Stability (h)	Ref.
Ru/Co3O4−x	OER	280	86.9	150η10	224
Ir–MnO2(160)–CC	OER	181	74.8	180η10	225
Pt/TiO2/Ti	HER	23	28.7	25η180	226
Pt/CuO@C	HER	39	41.7	72η10	227
Ru/P–TiO2	HER	27	28.3	24η10	228
Pt@CoOx	HER	82	51.5	24η100	229
MoOx–Rh/C	HER	15	16	12η50	230
PtSA–M–CeO2−x/rGO	HER	33	57.9	90η100	231
c/a-Ru/VOx-500	HER	33	59	50η50	232
Ru–VO2	OER	228	48.27	125η10	233
	HER	46	39.1

---

Thus, it is also a good choice to improve the catalytic activity by modifying (defecting/doping) the noble metal-based oxide electrocatalysts, because the modified electrocatalysts can activate the active sites on other non-noble metal materials as far as possible, while making full use of the active sites of the noble metal, to achieve the purpose of consuming a very small amount of noble metal materials when preparing electrocatalysts with excellent performance. The aim of preparing electrocatalysts with excellent performance is achieved by consuming a very small amount of noble metal materials. However, the catalytic performance of some noble metal-based oxide electrocatalysts deteriorates with time of use, which may be caused by the loss of active sites in the catalysts, or the catalysts may be poisoned, leading to the deterioration of performance. This calls for consideration of noble metal recycling to improve the utilisation of scarce noble metal materials. In this case, regeneration of the catalyst can be considered, and the original catalytic activity of the poisoned catalyst can be restored through appropriate physical or chemical methods, which will largely improve the service life of the catalyst, and also achieve the goal of reducing the amount of noble metals used, and minimise the adverse impact of industrial development on the environment.

4.3. Noble metal oxides–transition metal oxides

To achieve an effective balance between the catalytic activity and stability of noble metal oxides, in addition to combining transition metals with noble metal oxides and combining noble metals with transition metal oxides, noble metal oxides can be combined with transition metals to improve the catalytic activity and stability of the catalysts through the synergistic effect of the two oxides. Ge et al.²³⁴ prepared a Cr–Ir oxide electrocatalyst (CrO₂–0.16IrO₂), which doped IrO₂ into the pores of CrO₂ (Fig. 16a and b), forming a strongly coupled interface through which the performance of the catalyst was optimised. X-ray absorption spectroscopy (XAS) measurements were conducted on the Ir L₃-side (Fig. 16c), and CrO₂–0.16IrO₂ exhibits higher energies than the reference IrO₂ sample, CrO₂–0.16IrO₂, suggesting that Cr⁴⁺ has a higher valence state. The peak at 1.5 Å that appears in the FT-EXAFS spectrum of CrO₂–0.16IrO₂ (Fig. 16d) is the peak of the Ir–O bond in CrO₂–0.16IrO₂, suggesting that combining the two oxides only alters the electronic structure and does not disrupt the original chemical bond. The overpotential of the CrO₂–0.16IrO₂ electrocatalyst in the LSV curve (Fig. 16e) at a current density of 2000 mA cm⁻² is only 425 mV, which is much smaller than those of the other analogous catalysts. In addition, CrO₂–0.16IrO₂ has a Tafel slope much smaller than those of the other catalysts (41.5 mV dec⁻¹; Fig. 16f), and a mass activity higher than those of the other catalysts (762 A g_Ir⁻¹; Fig. 16g). These findings indicate that the OER activity of the catalyst is increased after the combination of the two oxides.

Fig. 16 (a) Design of the CrO₂–0.16IrO₂ electrocatalyst. (b) XRD pattern. (c) and (d) XANES and EXAFS spectra of CrO₂–0.16IrO₂ and an IrO₂ reference for the Ir L₃-edge. (e) LSV curves of different samples in a 0.5 M H₂SO₄ electrolyte. (f) Δη/Δlog|j| values for the CrO₂–0.16IrO₂ and commercial IrO₂ electrocatalysts. (g) Mass activity comparison. (h) Stability test results of the CrO₂–0.16IrO₂ electrocatalyst at 1000 mA cm⁻². (i) Comparison of Ir element dissolution in different electrocatalysts. Reproduced from ref. 234 with permission from Royal Society of Chemistry, copyright 2023.

Furthermore, in the stability tests at a current density of 1000 mA cm⁻², CrO₂–0.16IrO₂ maintained high stability over 100 h (Fig. 16h). The dissolution of Ir in CrO₂–0.16IrO₂ in tests with Ir (Fig. 16i) yields an almost constant value that is much lower than those of commercial IrO₂ electrocatalysts over 100 h, suggesting that CrO₂–0.16IrO₂ has better stability and that the part of dissolution can be attributed to isolated IrO₂NPs without a strong coupling interface. Hence, interfacial engineering can be fully utilised to improve the activity and stability of the catalysts after combining noble metal oxides and transition metal oxides. Heterojunction engineering is a commonly used improvement strategy. In particular, Karthikeyan et al.²³⁵ reported a porous heterostructured nanofoam catalyst (NiO/IrO₂) in which the electronic structure modulation at the heterojunction and the multiphase coupling strategy work together to improve the catalyst performance, resulting in excellent HER (η₁₀ = 42 mV), OER (η₁₀ = 240 mV), and ORR (E_1/2 = 0.80 V) performance. Although the formation of heterojunctions can modulate the electronic structure, if the material forming the heterojunction changes, the heterojunction may disappear, rendering the originally designed improvement strategy ineffective. Long et al.²³⁶ loaded Ru–RuO₂ heterojunctions with Mn₃O₄ as the substrate and prevented the overoxidation of Ru by sacrificing Mn₃O₄ during the OER process, thereby maintaining the original heterojunction, enhancing the stability while accelerating the involvement of AEM through the heterojunction to improve the catalytic activity.

The OER is the rate-limiting step in the decomposition of electrolytic water because the complex band of four successive H⁺-coupled electron-transfer steps leads to slow kinetics, and the desire to accelerate such slow kinetics requires an increase in the reaction rate of the decisive step of deprotonation, as well as the rate of electron transfer. Wang et al.²³⁷ reported a Ru_0.6W_17.4–δ electrocatalyst with a 3D sea-urchin-like morphology O_49–δ electrocatalyst, which introduced a Ru–O_bri–W acid site into Ru_0.6W_17.4O_49–δ. This approach promotes H⁺ migration from the O intermediate to the neighbouring Obri site (Fig. 17a), accelerating the deprotonation step and thus facilitating the acidic OER kinetics. In addition, in Ru_0.6W_17.4O_49–δ, the overoxidation of the surface Ru site is prevented by the increased electron density around the Ru site due to O-bridging from the neighbouring W, thus prolonging the long-term stability. The overpotential of Ru_0.6W_17.4O_49–δ is 252 mV at a baseline current density of 10 mA cm⁻², which is higher than that of commercial RuO₂ (Fig. 17b). In addition, the Tafel slope of Ru_0.6W_17.4O_49–δ is 50 mV dec⁻¹, which is smaller than that of commercial RuO₂ (Fig. 17c), suggesting that this catalyst has faster reaction kinetics. This characteristic which is visually demonstrated in the histogram of the overpotential and Tafel slope at 10 mA cm⁻² in Fig. 17d, which shows that Ru_0.6W_17.4O_49–δ has a lower overpotential and faster reaction kinetics. The Ru normalised mass activity of this catalyst at 1.53 V and 1.6 V is 10.8 and 10.9 times higher than that of commercial RuO₂, respectively (Fig. 17e). Ru_0.6W_17.4O_49–δ has a C_dl 2.3 times higher than that of commercial RuO₂ (Fig. 17f), indicating that the former has a higher active area. The catalyst exhibits an ECSA-normalised current density of 0.237 mA cm⁻² at 1.575 V, which is 1.5 times higher than that of commercial RuO₂ (Fig. 17g). However, the stability of Ru_0.6W_17.4O_49–δ is only 48 h, and the stability needs to be improved further in subsequent studies. Li et al.,²³⁸ on the other hand, prepared three-dimensional bicontinuous nanoporous Co@CoO/RuO₂ electrocatalytic materials with tunable dimensions and chemical compositions by introducing gas-phase dealloying of Co-based alloys, which enhanced the interfacial CoORu bonding by improving the charge transfer rate, nanopore connectivity, and stability of the CoO/RuO₂ interface, while also enhancing the activity and stability of nanoporous catalysts.

Fig. 17 (a) Schematic representation of the H⁺ transfer process (orange balls-W, grey balls-Ru, white balls-H, red balls-O). (b) LSV curves of Ru_0.6W_17.4O_49–δ, W₁₈O₄₉, and commercial RuO₂. (c) Tafel slopes. (d) Histograms of η₁₀ and Tafel slopes. (e) and (g) Ru mass-normalised activities and the specific activities of Ru_0.6W_17.4O_49–δ and commercial RuO₂. (f) C_dl of Ru_0.6W_17.4O_49–δ, W₁₈O₄₉, and commercial RuO₂. Reproduced from ref. 237 with permission from Wiley-VCH GmbH, copyright 2023.

For IrO₂ catalysts with good stability but low catalytic activity, increasing the valence of Ir is an effective way to improve their catalytic performance. Li et al.²³⁹ reported a method of oxidising Ir atoms using MnO₂ to obtain an atomically dispersed hexavalent Ir oxide (Ir^VI-ado), which exhibited good catalytic activity for the OER. First, MnO₂ was deposited on a corrosion-resistant Pt-coated Ti mesh (MnO₂/PTL) by electrodeposition, followed by immersion in 0.01 M H₂SO₄ incorporating K₂IrCl₆ at 95 °C to obtain the precatalyst (Ir-pre). Then, the Ir-pre was annealed in air at 450 °C to obtain Ir^VI-ado (Fig. 18a). The position of the strongest peak in the XANES spectrum of Ir shifted to a higher energy by 0.9 eV after H₂SO₄ treatment and then further shifted to a higher energy by 0.5 eV after annealing, indicating that Ir was oxidised from the 4-oxidation state to a higher oxidation state (Fig. 18b and c). During Ir oxidation, MnO₂ acted as an oxidant for Ir^VI formation. High-energy resolution fluorescence detected XANES (HERFD-XANES) was used to verify the valence state of Ir atoms, and then the oxidation state of Ir^VI-ado was determined to be +5.8 ± 0.1 by using the integral white line (WL) intensity proportional to the number of 5d holes (Fig. 18d and e). The X-ray absorption spectra (Fig. 18f) show that the current density achieved in the in situ proton exchange membrane (PEM) electrolysis channel is 0.02, 0.42, and 2.30 A cm⁻² when the channel voltage is 1.5, 1.8, and 2.5 V, respectively. The position of the Ir L₃-edge of the WL changes by less than 0.1 eV with increasing slot voltage (Fig. 18g), suggesting that the hexavalent state of Ir^VI is maintained with increasing voltage. The stabilisation of the catalysts was assessed by PEM electrolysis (Fig. 18h), showing that Ir^VI-ado is the most stable at a mass-specific activity of 1.25 × 10⁴ Ag_Ir⁻¹. At a mass-specific activity of 5 × 10⁴ Ag_Ir⁻¹, the catalyst is relatively unstable but maintains a durability of 1710 h and a turnover number of 1.5 × 10⁸. The turnover was calculated by dividing the total amount of O₂ released by the Ir charge and surpassing those assessed for the Ir-based catalysts found in the three-pole electrodes and the PEM apparatus (Fig. 18i). Finally, the Pt-encapsulated Ti mesh was replaced with Pt surface-modified sintered Ti powder (p-PTL), and the thickness of the MnO₂ layer was halved to improve the PEM performance of Ir^VI-ado (Fig. 18j). Thus, oxidising Ir to a higher valence state improves the catalytic activity of IrO₂ because doing so also improves its stability. In future research, oxidation to a higher valence state to improve catalytic activity will be a promising research topic for materials that have good stability on their own but lack catalytic activity.

Fig. 18 (a) Synthetic preparation of Ir^VI-ado. (b) and (c) XANES results of the Ir L₃-edge as a function of time while synthesising Ir^VI-ado. (d) HERFD-XANES results at the Ir L₃-border of Ir^VI-ado and control samples. (e) Intensity of the WL peak as a function of the cavity number in the d-band for Ir^VI-ado (green circles) and control samples. (f) Current density reached at a given voltage in the PEM electrolysis chamber for in situ XAS determination. (g) XANES results of in situ Ir L₃-side of Ir^VI-ado at various battery voltages. (h) Battery voltage of Ir^VI-ado with various quality ratios of activity across a commercial PEM electrolytic cell. (i) Number of turnovers based on catalyst lifetime. (j) LSV of Ir^VI-ado packed on MnO₂/p-PTL. Reproduced from ref. 239 with permission from American Association for the Advancement of Science, copyright 2024.

Among the emerging noble metal oxides, Ag oxides combined with transition metal oxides are the choice of most researchers. Shaghaghi et al.²⁴⁰ synthesized AgO/NiO hybrid nanostructured electrocatalysts by the co-precipitation method, in which the synergistic effect between Ag₂O and NiO increased the number of active catalytic sites. This solved the problem of poor intrinsic conductivity of Ag to improve its OER performance (430 mV) and HER performance (140 mV) in alkaline solution.

Transition metal oxides can be used as carriers of noble metal oxides, and by changing the structure of transition metal oxides, the catalyst surface area can be enlarged, thus increasing the availability of the active sites of the noble metal oxides. In addition, the transition metal oxide carrier can act as an electron reservoir, continuously providing electrons to the noble metal sites, preventing the noble metal from being over-oxidised and dissolved in the electrolyte, avoiding collapse, and enhancing the stability of the catalyst. Transition metal oxides can oxidise noble metals to more catalytically active valence states, thereby increasing their catalytic activity. Compared with previous electrocatalysts doped with transition metals in noble metal oxides and electrocatalysts doped with noble metals in transition metal oxides, the combination of the two oxides has rarely been studied. Nonetheless, the existing reports indicate that the combination of the two oxides can have many surprising effects, and this combination will be widely applied in the future development of electrolytic water technology. Some noble metal oxide and transition metal oxide electrocatalytic materials are listed in Table 4.

Table 4 Noble metal oxide–transition metal oxide catalysts

Catalysts	Reaction	η 10 (mV)	Tafel slope (mV dec^−1)	Stability (h)	Ref.
IrOx/CeO2-0.6	OER	220	63.0	300η10	241
IrO2@TaB2	OER	288	42.6	120η10	242
c-IrOx–MoO3/Ti	OER	200	58.6	130η100	243
B0.6–IrO2–Ta2O5	OER	210	34.2	400η100	244
RuO2/(Co,Mn)3O4	OER	270	77.0	24η10	245
NCO@RuO2–NCs	OER	188	74.3	15η10	246
	HER	90	54.9	38η10

---

However, when combining noble metal oxides with transition metal oxides to prepare electrocatalysts, it is inevitable that a part of the noble metal will be in the inner part of the catalyst, which will result wasting some of the active sites on the noble metal. To make full use of the noble metal resources, it is a good choice to develop nanomaterial electrocatalysts with a small size as much as possible. Because, as the size decreases, the surface area of the catalyst will correspondingly increase, and more noble metals will be exposed, which not only provides more active sites but also improves the mass activity of the catalyst, which brings better catalytic activity and reduces the cost of the catalyst.

In the process of presenting the source of activity of the optimised noble metal-based oxide electrocatalysts, this review focuses on illustrating the excellent catalytic performance of the listed catalysts by analysing the catalysts' required overpotentials for the OER or HER, Tafel slopes, active area, impedance and catalyst stability at a current density of 10 mA cm⁻² and comparing these values with the performance of commercial catalysts that have excellent catalytic performance. However, in addition to the comparison with commercial catalysts, comparison between the optimised catalysts is still relevant. Therefore, in this review, the bifunctional water electrolysis performance of noble metal-based oxide electrocatalysts is discussed. After a literature search, we found that the bifunctional noble metal-based oxide electrocatalysts used for total water electrolysis are mainly concentrated on Ru-based oxides, Ir-based oxides and Pt-based oxides, and other emerging noble metal oxides accounted for a relatively small proportion of the total water electrolysis performance (Fig. 19a). This phenomenon suggested that the emerging noble metal-based oxides have the potential to catalyse total hydrolysis, which will be remarkably rewarding upon further development. In addition, we also compared the cell voltages of the most frequently occurring Ru-based oxides for the total hydrolysis process (Fig. 19b), and found that to achieve a current density of 10 mA cm⁻², the external voltage needed to be applied was in the range of 1.4–1.6 V. It was found that the smallest cell voltages were required after doping metal ions in RuO₂, and that the smallest cell voltages were required after combining RuO₂ with other metal oxides. For oxides, the required battery voltage is a little higher. Thus, it can be seen that obtaining energy-saving bifunctional electrocatalysts, is a simple and effective way to find suitable doping metals.

Fig. 19 (a) Percentage of six types of noble metal-based oxides as bifunctional electrocatalysts. (b) Comparison of cell voltages for Ru-based oxide electrocatalysts.

5. Commercial electrolytic water technology

At the end of the 19th century, water electrolysis entered into industrial applications, and electrolysis plants (using alkaline electrolyte) began to operate.²⁴⁷ The global water electrolyzer market has been expanding with the continuous development of hydrogen production from water electrolysis. Currently, the existing electrolyzers can be classified into two categories based on the operating temperature: low-temperature electrolyzers and high-temperature electrolysers,²⁴⁸ based on the electrolyte technology, representative water electrolysis technologies can be classified into: alkaline water electrolysis (AWE), proton exchange membrane water electrolysis (PEMWE), anion-exchange membrane water electrolysis (AEMWE), and solid oxide water electrolysis (SOEWE).²⁴⁹

The AWE is a very mature technology with electrolyzers operating in the megawatt range and is commercially available worldwide.²⁵⁰ The AWE is also an optimally cost-effective technology because the electrode material used is a nickel-based non-precious metal. The AWE typically uses KOH or NaOH lye as the electrolyte into which the electrodes are immersed and separated by a diaphragm to separate the hydrogen and oxygen produced. OH⁻ in the lye on the anode side loses electrons and undergoes oxidation to produce O₂, while H₂O molecules on the cathode side gain electrons and undergo reduction to produce H₂, with a total water consumption of about 11.5 times the amount of hydrogen produced.²⁵¹ However, it is necessary to balance the pressure on both sides during the operation, otherwise, the high-purity hydrogen will explode when it comes into contact with oxygen, which will lead to safety accidents. In addition, the use of alkaline electrolytes brings some disadvantages, which include a corrosive nature, high maintenance, limited dynamic operation of the system, and difficulty in coupling with fluctuating energy sources such as wind or solar energy. These disadvantages have largely limited the development of the AWE, and therefore researchers are committed to developing other water electrolysis technologies with better performance.

The PEMWE, also known as the AWE due to the strong acidity of the electrolyte after the anodic reaction, is also a relatively mature technology for hydrogen production. Fortunately, the PEMWE is safer than the AWE because the electrolyte used in its electrolysis tank is pure water, avoiding the corrosion problems caused by alkaline electrolyte, and therefore has good chemical stability. Moreover, the proton exchange membrane separating the anode and cathode has good proton conductivity and gas separation, which can increase the efficiency of electrolysis and ensure the safety of the environment at the same time. In addition, the PEMWE is also the most commercially used electrolytic water hydrogen production technology with noble metal-based oxide electrocatalysts, with relatively more Ir-based oxides being used. In recent studies, researchers have worked on improving the catalytic activity with reduced loading of Ir and minimising the size of the PEMWE system. Zhang²⁵²et al. prepared Sr–IrO_x electrocatalysts by doping strontium element into IrO_x, which increased the mass activity of Ir to 150 Ag_Ir⁻¹. Moreover, the catalyst was stable at a current density of 1 A cm⁻² and a PEMWE performance test was conducted for 500 h, which is realistic for industrial applications. In addition, the Ir^VI-ado electrocatalyst prepared by Li et al.²³⁹ by combining MnO₂ with atomic-level Ir^VI was used in a PEW cell for a long-time stability test, and it was found that it could achieve a durability of 1710 h, and the loading of Ir was reduced by more than 96% compared with that of the original Ir-based catalyst to reach a mass activity of 5 × 104 Ag_Ir⁻¹. This good stability and mass activity make Ir^VI-ado an excellent electrocatalyst for PEMWE. Luckily, a self-preserved nanocrystalline solid solution catalyst of Ir_ySn_0.9(1−y)Sb_0.1(1−y)O_x was developed by Venkatesan et al.²⁵³ to enhance the OER activity and stability by having highly dispersed Ir nanoparticles and abundant oxygen vacancies. The degradation rate was only 18 μV h⁻¹ at a current density of 2 A cm⁻² and a low Ir loading of 0.2 mg cm⁻², exceeding the ultra-high stability of 2000 h, as tested in PEMWE. The publication of these research results meant that Ir-based oxide electrocatalysts would make a breakthrough in the development of PEMWE.

Although some Ir-based oxide electrocatalysts have already achieved some results in PEMWE, the catalytic activity of most Ir oxides is at a low level, and a large amount of catalysts is required to achieve the desired catalytic effect, which makes most of the PEMWE systems applying Ir oxides as the catalysts still have a large volume. Therefore, new materials have been sought to replace the traditional Ir-based oxide catalysts. Among them, Ru oxide is a hot topic of research. Optimised Ru oxide electrocatalysts are highly likely to be applied to PEMWE when their stability is improved without sacrificing the original catalytic activity. Zhou and his team²⁵⁴ introduced acid-resistant Pd into RuO₂, which increased the stability of the catalysts up to 1100 h. This discovery made the PEMWE with Pd–RuO₂ only require 1.688 V to reach 1000 mA cm⁻² and operate at 500 mA cm⁻² for more than 250 h. Also optimising RuO₂, Shen et al.²⁵⁵ doped Cr into RuO₂ and synthesised the electrocatalyst of Cr_0.2Ru_0.8O_2−x, which was applied to a practical PEMWE device with great industrial value. The development of large amounts of stable RuO₂ will cause a change in PEMWE. Among the noble metals of the Pt group, Pd and Pt can be modified with other oxides to obtain stable electrocatalysts for PEMWE devices. The Pt and Ru co-doped oxide electrocatalyst (PtRu–Co₃O₄) can also exhibit excellent performance in PEM electrolysis baths as reported by Li and his team.²⁵⁶ Most of the other noble metal oxides are still in the laboratory exploration stage, and further exploration is needed to reach commercial applications. In the case of commercial PEMWE, large-scale use requires consuming a large amount of catalyst. It is important to minimise the cost of noble metal-based oxide electrocatalysts, and the recovery and regeneration of the catalysts take on practical significance.

In addition, AEMWE and SOEWE are two emerging hydrolysis technologies. AEMWE, developed from PEMWE and AWE, has the advantages of both, using non-precious-metal catalysts, and can also be used under alkaline conditions, but lacks anion-exchange membranes that are both sufficiently conductive and stable, so AEMWE is still in the beginning stages of laboratory research. The electrolysis tank of SOEWE was operated at high temperatures of about 500–1000 °C, and part of the energy required for water decomposition was provided in the form of heat. Some of the energy required was provided in the form of heat, which can store electrical and thermal energy from renewable sources such as chemical energy (H₂), is environmentally friendly, and uses non-precious-metal catalysts, which makes it less expensive. However, temperature gradients of several hundred degrees can cause mechanical stresses and lead to material failure, so it is still not commercially available. After further research, AEMWE and SOEWE will show similar or even better industrial value than PEMWE.

6. Conclusions and perspectives

The production of “grey H₂” and “blue H₂” causes certain environmental pollution, whereas the “green H₂” produced by the electrolysis of water is environmentally friendly and thus has gained attention. The process of water electrolysis needs to be catalysed by suitable catalysts, among which the noble metal-based oxide electrocatalysts developed from noble metals can not only retain the active sites of noble metals, but also reduce the production costs, which is crucial for the evolution of H₂ production technology by water electrolysis. This review summarised the recent progress in research on noble metal oxides in the field of water electrolysis. First, the development process of noble metal oxides, the number of publications in the past 10 years, and the worldwide research status were introduced. Then, the mechanisms of the two half reactions for the electrolysis of water were introduced in detail. In addition, three typical synthesis methods (template, electrodeposition, and pyrolysis methods) and four improvement strategies (doping, heterogeneous junction, defect and strain engineering strategies) were introduced, as well as detailed analyses of the sources of the improved catalytic performance of three types of noble-metal-based electrocatalysts, namely, noble metal oxides, noble metal–transition metal oxides, and noble metal oxides–transition metal oxides. Finally, some commercial applications of noble metal-based oxide electrocatalysts were also summarised.

Compared with pure noble metal-based oxide electrocatalysts, the use of noble metals is indeed reduced, which largely reduces the production costs, and they are promising electrocatalysts. However, this type of catalyst still has difficulty in achieving a balance between stability and catalytic activity. In addition, with the further development of noble metal-based oxide electrocatalysts and their large-scale commercial application in the future, more and more noble metals will be consumed, and the environmental problem of resource scarcity will be aggravated. Therefore, there is still much room for improvement of the noble metal-based oxide electrocatalysts.

(1) Conducting computational simulations and developing advanced characterisation techniques. In the process of catalyst modification, the full use of computational simulation can greatly reduce the exploration cost and shorten the experimental cycle. In addition, advanced characterisation techniques can allow researchers to fully grasp the catalyst information for finer tuning when modifying the catalyst.

(2) Development of small-sized nanomaterial electrocatalysts. As the size decreases, the surface area of the catalyst will correspondingly increase, and more noble metals will be exposed to it, which will not only provide more active sites but also increase the mass activity of the catalyst, thus improving the catalytic activity and achieving the purpose of reducing the cost of the catalyst.

(3) Recovery and regeneration of catalysts. The recovery of catalysts, especially the noble metals contained therein, can improve the utilisation of noble metals and alleviate the shortage of noble metal resources. In addition, the active regeneration of catalysts is a feasible way to extend the life of catalysts, which can also reduce the consumption of noble metal resources. Therefore, the recovery and regeneration of catalysts need to be taken into full consideration in future studies of noble metal-based oxide electrocatalysis.

(4) Electrocatalysts for universal pH. Electrocatalysts can be used under both acidic and alkaline conditions and will greatly facilitate the real-time application of electrolytic water technology. Thus far, researchers have mainly focused on different acidic or alkaline conditions and the preparation of specific composite materials. However, these materials can only be used under a single catalyst condition, preventing them from meeting the needs of all people. Hence, a wider range of conditions for the electrocatalyst must be considered to achieve large-scale production and application of noble metal oxides.

(5) Multifunctional applications of noble metal-based oxide electrocatalysts. Noble metal oxides with abundant active sites have the potential for use in multifunctional applications beyond catalysis, energy conversion, and environmental remediation. In addition to catalysing water decomposition, catalysts in this class can also be used to catalyse the degradation of several pollutants and can even be applied in the development of batteries. Further studies are required to explore the new properties and applications in a wider range of fields.

In conclusion, although some challenges are faced in research on noble metal oxide catalysts, opportunities also exist to achieve a balance between catalytic activity, stability, and cost, which will promote the development of electrolytic H₂ production technology and the completion of the energy transition. Such work will enable the application of these materials in a wider range of fields. As an important and promising research topic in the 21st century, noble metal oxides will make significant contributions to the progress of humankind.

Data availability

No primary research results, software or codes have been included and no new data were generated or analysed as part of this review.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (U2002213), Tianshan Innovation Team Plan of Xinjiang Uygur Autonomous Region (2023D14002), Open Foundation of Guangxi Key Laboratory of Processing for Non-Ferrous Metals and Featured Materials (2022GXYSOF10), Science and Technology Talent and Platform Program of Yunnan Provincial Science and Technology Department (202305AM070001), Scientific and Technological Project of Yunnan Precious Metals Laboratory (YPML-20240502065), The Yunnan Fundamental Research Projects (202401CF070026), The Xingdian Talent Program of Yunnan Province, and Double-First Class University Plan (C176220100042).

References

1. Y. Y. Liu, B. Tian, F. D. Ning, Y. L. Li, C. Y. Zhao, C. He, Q. L. Wen, X. Dan, Z. Chai, W. Li, M. Shen, L. He, W. X. Li and X. C. Zhou, Adv. Energy Mater., 2023, 14, 2303353.
2. T. Z. Ang, M. Salem, M. Kamarol, H. S. Das, M. A. Nazari and N. Prabaharan, Energy Strategy Rev., 2022, 43, 100939.
3. W. Liu, Y. M. Wan, Y. L. Xiong and P. B. Gao, Int. J. Hydrogen Energy, 2022, 47, 24584–24591.
4. A. Boretti, Int. J. Hydrogen Energy, 2024, 49, 166–172.
5. A. H. Schrotenboer, A. A. T. Veenstra, M. A. Juithet Broek and E. Ursavas, Renewable Sustainable Energy Rev., 2022, 168, 112744.
6. B. Nastasi and S. Mazzoni, Energy Convers. Manage., 2023, 291, 117293.
7. R. Förster, M. Kaiser and S. Wenninger, Appl. Energy, 2023, 334, 120653.
8. A. Di Martino, M. Longo and F. Foiadelli, Energy Convers. Manage., 2024, 304, 118211.
9. W. Y. Zhang, X. M. Fang and C. W. Sun, J. Environ. Manage., 2023, 341, 118019.
10. L. B. Qin, Z. Xu, B. Zhao, C. Zou, W. S. Chen and J. Han, J. Energy Inst., 2022, 102, 14–21.
11. K. X. Yin, R. Q. Zhang, M. Yan, L. Sun, Y. X. Ma, P. Z. Cui, Z. Y. Zhu and Y. L. Wang, Process Saf. Environ. Prot., 2023, 178, 8–17.
12. D. M. Polyukhov, N. A. Kudriavykh, S. A. Gromilov, A. S. Kiryutin, A. S. Poryvaev and M. V. Fedin, ACS Energy Lett., 2022, 7, 4336–4341.
13. M. Wang, H. R. Zhou and F. Wang, Joule, 2024, 8, 1–18.
14. W. Liu, Y. X. Li, Y. X. Dou, N. Xu, J. J. Wang, J. T. Xu, C. M. Li and J. Q. Liu, J. Colloid Interface Sci., 2024, 655, 800–808.
15. Y. Z. Wang, Y. M. Ding, C. H. Zhang, B. W. Xue, N. W. Li and L. Yu, Rare Met., 2021, 40, 2785–2792.
16. S. Xu, M. Li, Y. J. Wang and Z. L. Jin, Int. J. Hydrogen Energy, 2024, 51, 16–30.
17. X. S. Yan, H. L. Liu, M. Luo, J. J. Cai and R. C. Shen, Int. J. Hydrogen Energy, 2023, 48, 11180–11190.
18. L. Zhang, Y. Zhang, J. P. Tang, L. X. Kang and Y. Z. Liu, Processes, 2022, 10, 2600.
19. S. X. Zhu, H. Jin, X. J. Meng, W. W. Wei, Z. S. Ou, M. M. Song and L. J. Guo, Fuel, 2023, 333, 126319.
20. T. Anand and S. Debbarma, J. Energy Resour. Technol., 2024, 146, 012302.
21. H. F. He, Q. Li, J. S. Tang, P. C. Liu, H. R. Zheng, F. Zhao, W. L. Guan, E. Guo and C. F. Xi, Int. J. Hydrogen Energy, 2023, 48, 2161–2170.
22. F. Angikath, F. Abdulrahman, A. Yousry, R. Das, S. Saxena, O. Behar, H. Alhamed, T. Altmann, B. Dally and S. M. Sarathy, Int. J. Hydrogen Energy, 2024, 49, 246–262.
23. I. J. Okeke, B. A. Saville and H. L. MacLean, Int. J. Hydrogen Energy, 2023, 48, 12581–12599.
24. V. H. Do, P. Prabhu, V. Jose, T. Yoshida, Y. T. Zhou, H. Miwa, T. Kaneko, T. Uruga, Y. Iwasawa and J. M. Lee, Adv. Mater., 2023, 35, 2208860.
25. L. M. Wang, L. L. Zhang, W. Ma, H. Wan, X. J. Zhang, X. Zhang, S. Y. Jiang, J. Y. Zheng and Z. Zhou, Adv. Funct. Mater., 2022, 32, 2203342.
26. M. Z. You, X. Du, X. H. Hou, Z. Y. Wang, Y. Zhou, H. P. Ji, L. Y. Zhang, Z. T. Zhang, S. S. Yi and D. L. Chen, Appl. Catal., B, 2022, 317, 121729.
27. W. L. Yu, H. R. Liu, Y. Zhao, Y. L. Fu, W. P. Xiao, B. Dong, Z. X. Wu, Y. M. Chai and L. Wang, Nano Res., 2023, 16, 6517–6530.
28. M. X. Zhong, M. J. Xu, S. Y. Ren, W. M. Li, C. Wang, M. B. Gao and X. F. Lu, Energy Environ. Sci., 2024, 17, 1984–1996.
29. Y. X. Du, Q. C. Li, L. P. Han, P. F. Yang, L. T. Xin, W. Jin, W. P. Xiao, Z. J. Li, J. S. Wang, Z. X. Wu and L. Wang, Appl. Catal., B, 2024, 344, 123617.
30. V. D. Dang, T. Annadurai, A. P. Khedulkar, J. Y. Lin, J. Adorna, W. J. Yu, B. Pandit, T. V. Huynh and R. A. Doong, Appl. Catal., B, 2023, 320, 121928.
31. H. L. Guang, S. L. Zhu, Y. Q. Liang, S. L. Wu, Z. Y. Li, S. Y. Luo, Z. D. Cui and A. Inoue, Rare Met., 2021, 40, 1031–1039.
32. W. Z. Zhang, M. H. Liu, X. Gu, Y. X. Shi, Z. F. Deng and N. S. Cai, Chem. Rev., 2023, 123, 7119–7192.
33. J. P. Sun, Z. Zhao, J. Li, Z. Z. Li and X. C. Meng, Rare Met., 2022, 42, 751–768.
34. A. Konieczny, K. Mondal, T. Wiltowski and P. Dydo, Int. J. Hydrogen Energy, 2008, 33, 264–272.
35. D. Yang, J. Lee, N. C. Song, S. Lee, S. Kim, S. Lee and S. Choi, Int. J. Hydrogen Energy, 2023, 48, 32241–32260.
36. H. R. Cheng, Y. H. Xia, Z. Y. Hu and W. Wei, Appl. Energy, 2024, 358, 122510.
37. W. A. Hindson and S. James, Int. J. Hydrogen Energy, 2024, 49, 1040–1047.
38. S. Kim, S. W. Lee, S. Lee, J. H. Kim and T. H. Shin, Electrochim. Acta, 2023, 463, 142751.
39. J. Wang, H. Yang, F. Li, L. G. Li, J. B. Wu, S. H. Liu, T. Cheng, Y. Xu, Q. Shao and X. Q. Huang, Sci. Adv., 2022, 8, eabl9271.
40. W. Gao, Z. M. Xia, F. X. Cao, J. C. Ho, Z. Jiang and Y. Q. Qu, Adv. Funct. Mater., 2018, 28, 1706056.
41. J. Wang, J. Hu, S. Q. Niu, S. W. Li, Y. C. Du and P. Xu, Small, 2022, 18, 2105972.
42. X. Z. Ren, X. H. Li, Y. J. Peng, G. Z. Wang, J. Yin, X. C. Zhao, W. Wang and X. B. Wang, Rare Met., 2022, 41, 4127–4137.
43. F. Zhou, Y. Zhou, G. G. Liu, C. T. Wang and J. Wang, Rare Met., 2021, 40, 3375–3405.
44. H. Cui, H. X. Liao, Z. L. Wang, J. P. Xie, P. F. Tan, D. W. Chu and P. Jun, Rare Met., 2022, 41, 2606–2615.
45. Y. H. Wang, R. Q. Li, H. B. Li, H. L. Huang, Z. J. Guo, H. Y. Chen, Y. Zheng and K. G. Qu, Rare Met., 2021, 40, 1040–1047.
46. H. X. Chen, H. Xu, Z. R. Song, Y. Liu, H. Cui and J. K. Gao, Rare Met., 2022, 42, 155–164.
47. R. Hao, Q. L. Feng, X. J. Wang, Y. C. Zhang and K. S. Li, Rare Met., 2021, 41, 1314–1322.
48. F. H. Yuan, M. R. Mohammadi, L. L. Ma, Z. D. Cui, S. L. Zhu, Z. Y. Li, S. L. Wu, H. Jiang and Y. Q. Liang, Rare Met., 2022, 41, 2624–2632.
49. T. Wu, M. Z. Sun and B. L. Huang, Rare Met., 2022, 41, 2169–2183.
50. L. H. Liu, N. Li, M. Han, J. R. Han and H. Y. Liang, Rare Met., 2021, 41, 125–131.
51. S. Sajjad, C. Wang, C. W. Deng, F. Ji, T. Ali, B. Shezad, H. Q. Ji and C. L. Yan, Rare Met., 2022, 41, 1851–1858.
52. H. Chen, R. Ding, B. W. Liu, F. R. Zeng and H. B. Zhao, Small, 2024, 20, 2306274.
53. G. L. Gao, Z. X. Sun, X. L. Chen, G. Zhu, B. W. Sun, Y. Yamauchi and S. D. Liu, Appl. Catal., B, 2024, 343, 123584.
54. S. Iqbal, B. Safdar, I. Hussain, K. Zhang and C. Chatzichristodoulou, Adv. Energy Mater., 2023, 13, 2203913.
55. X. Y. Liu, X. T. Song, G. M. Jiang, L. J. Tao, Z. Y. Jin, F. K. Li, Y. Z. He and F. Dong, Chem. Eng. J., 2024, 481, 148430.
56. J. Y. Zhang, J. Dang, X. H. Zhu, J. G. Ma, M. G. Ouyang and F. Y. Yang, Appl. Catal., B, 2023, 325, 122296.
57. L. Tao, F. Lv, D. W. Wang, H. Luo, F. X. Lin, H. Y. Gong, H. T. Mi, S. G. Wang, Q. H. Zhang, L. Gu, M. C. Luo and S. J. Guo, Joule, 2024, 8, 450–460.
58. C. Liu, Z. N. Wei, M. N. Cao and R. Cao, Nano Res., 2024, 17, 4844–4849.
59. Y. Jiang, H. Liu, Y. M. Jiang, Y. N. Mao, W. Shen, M. Li and R. X. He, Appl. Catal., B, 2023, 324, 122294.
60. V. H. Do, P. Prabhu, V. Jose, T. Yoshida, Y. Zhou, H. Miwa, T. Kaneko, T. Uruga, Y. Iwasawa and J. M. Lee, Adv. Mater., 2023, 35, 2208860.
61. J. Y. Wu, W. W. Zou, J. X. Zhang, L. H. Zhang, H. Y. Song, Z. M. Cui and L. Du, Small, 2023, 20, 2308419.
62. H. Q. Song, X. Yong, G. I. N. Waterhouse, J. K. Yu, H. Wang, J. M. Cai, Z. Y. Tang, B. Yang, J. W. Chang and S. Y. Lu, ACS Catal., 2024, 14, 3298–3307.
63. D. Chen, H. Y. Zhao, R. H. Yu, K. S. Yu, J. W. Zhu, J. X. Jiao, X. Q. Mu, J. Yu, J. S. Wu and S. C. Mu, Energy Environ. Sci., 2024, 17, 1885–1893.
64. R. Z. Jiang, D. T. Tran, J. T. Li and D. Chu, Energy Environ. Mater., 2019, 2, 201–208.
65. S. J. Li, W. F. Xie, Y. K. Song, Y. Li, Y. J. Song, J. Z. Li and M. F. Shao, Chem. Eng. J., 2022, 437, 135473.
66. S. Trasatti and G. Buzzanca, J. Electroanal. Chem. Interfacial Electrochem., 1971, 29, A1–A5.
67. C. Iwakura, K. Fukuda and H. Tamura, Electrochim. Acta, 1976, 21, 501–508.
68. G. Beni, L. M. Schiavone, J. L. Shay, W. C. Dautremont-Smith and B. S. Schneider, Nature, 1979, 282, 281–283.
69. D. A. Denton, J. A. Harrison and R. I. Knowles, Electrochim. Acta, 1981, 26, 1197–1207.
70. A. E. Bolzán and A. J. Arvia, J. Electroanal. Chem., 1992, 322, 247–265.
71. H. Gouveia, R. Landers and J. F. C. Boodts, Electrochim. Acta, 2007, 52, 2359–2369.
72. A. Vargas-Uscategui, E. Mosquera, B. Chornik and L. Cifuentes, Electrochim. Acta, 2015, 178, 739–747.
73. R. X. Ge, L. Li, J. W. Su, Y. C. Lin, Z. Q. Tian and L. Chen, Adv. Energy Mater., 2019, 9, 1901313.
74. W. Y. Zhang, B. L. Huang, K. Wang, W. X. Yang, F. Lv, N. Li, Y. G. Chao, P. Zhou, Y. Yang, Y. J. Li, J. H. Zhou, W. S. Zhang, Y. P. Du, D. Su and S. J. Guo, Adv. Energy Mater., 2021, 11, 2003192.
75. G. R. Lee, J. Kim, D. Hong, Y. J. Kim, H. Jang, H. J. Han, C. K. Hwang, D. H. Kim, J. Y. Kim and Y. S. Jung, Nat. Commun., 2023, 14, 5402.
76. W. T. Bao, Y. Tang, J. Yu, W. X. Yan, C. X. Wang, Y. Y. Li, Z. M. Wang, J. F. Yang, L. Zhang and F. Yu, Appl. Catal., B, 2024, 346, 123706.
77. X. B. Li, T. Han, Y. T. Zhou, M. Wang, Z. L. Tian, F. Deng, Y. D. Luo, Y. Xie, J. T. Huang, L. Han, Z. Chen, Z. J. Feng and W. Chen, Appl. Catal., B, 2024, 350, 123913.
78. L. Dai, Z. N. Chen, L. X. Li, P. Q. Yin, Z. Q. Liu and H. Zhang, Adv. Mater., 2020, 32, 1906915.
79. R. X. Deng, H. Q. Yao, Y. X. Wang, C. N. Wang, S. Q. Zhang, S. S. Guo, Y. L. Li and S. L. Ma, Chem. Eng. J., 2024, 488, 150996.
80. D. I. Jeong, D. Kang, B. K. Kang, U. Y. Lee, I. Y. Suh, Y. Kim, B. M. Weon, S. W. Kim and D. H. Yoon, Small, 2024, 2400374,  DOI:10.1002/smll.202400374.
81. J. Ma, R. H. Zhou and F. Yu, Desalination, 2024, 571, 117107.
82. Y. Jiang, S. Gao, X. Liu, Y. Wang, S. Zhou, Q. Liu, A. Abdukayum and G. Hu, Nano Res., 2024, 17, 5763–5785.
83. Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K. Nørskov and T. F. Jaramillo, Science, 2017, 355, eaad4998.
84. Z. Y. Chen, L. T. Li, Y. Q. Chu, F. M. Zhao, Y. H. Zhu, S. P. Tong and H. J. Zheng, Small, 2024, 20, 2309675.
85. T. Liu, Y. H. Zhao and T. Y. Zhai, J. Am. Chem. Soc., 2024, 146, 6461–6465.
86. Q. Fu, X. J. Wang, J. C. Han, J. Zhong, T. R. Zhang, T. Yao, C. Y. Xu, T. L. Gao, S. B. Xi, C. Liang, L. L. Xu, P. Xu and B. Song, Angew. Chem., Int. Ed., 2021, 60, 259–267.
87. Z. Liu, H. W. He, Y. X. Liu, Y. Zhang, J. W. Shi, J. Xiong, S. F. Zhou, J. Li, L. Y. Fan and W. W. Cai, J. Colloid Interface Sci., 2023, 635, 23–31.
88. S. Anantharaj and S. Noda, Energy Environ. Sci., 2022, 15, 1461–1478.
89. Q. L. Ma and S. C. Mu, Interdiscip. Mater., 2023, 2, 53–90.
90. X. Y. Ping, Y. D. Liu, L. X. Zheng, Y. Song, L. Guo, S. G. Chen and Z. D. Wei, Nat. Commun., 2024, 15, 2501.
91. M. K. Jiang, Y. X. Liu, S. S. Kan, S. X. Deng, Z. K. Ren, X. M. Jiang and C. B. Yao, Chem. Eng. J., 2024, 487, 150724.
92. B. Salhi, N. Baig and I. Abdulazeez, npj Clean Water, 2024, 7, 9.
93. B. F. Lin, C. L. Tang, Z. J. Zheng, M. F. Zhang, Y. P. Huang, X. Yuan, X. H. Liu and Y. H. Wu, Chem. Eng. J., 2024, 488, 150858.
94. N. N. He, Y. D. Zou, C. Chen, M. H. Tan, Y. D. Zhang, X. F. Li, Z. M. Jia, J. Zhang, H. H. Long, H. Y. Peng, K. F. Yu, B. Jiang, Z. Q. Han, N. Liu, Y. Li and L. J. Ma, Nat. Commun., 2024, 15, 3896.
95. N. Gorouhi, M. Haghighi, M. Shabani and C. Almasi Jaf, J. Cleaner Prod., 2024, 450, 141850.
96. Y. Luo, K. Li, Y. J. Hu, T. Chen, Q. C. Wang, J. Q. Hu, J. Feng and J. Z. Feng, Small, 2024, 20, 2309822.
97. X. Ma, X. Y. Zhang, M. Yang, J. Y. Xie, R. Q. Lv, Y. M. Chai and B. Dong, Rare Met., 2021, 40, 1048–1055.
98. A. Q. Li, X. N. Bai, Y. T. Xie, H. B. Bao, Y. Yang, M. He, W. J. Yang, Q. W. Zhang, Y. H. Zhang and X. M. Li, Chem. Eng. J., 2024, 490, 151887.
99. X. Zhang, C. C. Shan, S. M. Ma, S. S. Zhao and J. Yang, Inorg. Chem. Commun., 2022, 135, 109089.
100. Y. M. Park, B. G. Kim, X. H. Gao, X. J. Zhang, H. S. Roh, C. H. Chung and J. W. Bae, Fuel, 2023, 343, 127943.
101. T. Nardin, J. Cambedouzou, J. Ravaux, C. Rey, D. Meyera and O. Diat, RSC Adv., 2015, 5, 86156–86162.
102. T. Guo, G. Y. Luo, C. Y. Shi, H. L. Shi, Z. X. Shi, B. F. He and J. B. Chen, ACS Sustainable Chem. Eng., 2022, 10, 4031–4039.
103. S. M. Youssry, I. S. El-Hallag, R. Kumar, G. Kawamura, A. Matsuda and M. N. El-Nahass, J. Electroanal. Chem., 2020, 857, 113728.
104. R. Q. Chen, Y. K. Yan, G. D. Wen, D. W. Zhang, Y. T. Zhou, C. J. Xiao, C. L. Yang, D. Na and J. S. Zhang, J. Mater. Sci., 2022, 57, 18770–18786.
105. J. T. Ren, Y. J. Song and Z. Y. Yuan, J. Energy Chem., 2018, 32, 78–84.
106. Y. C. Lin, Z. Q. Tian, L. J. Zhang, J. Y. Ma, Z. Jiang, B. J. Deibert, R. X. Ge and L. Chen, Nat. Commun., 2019, 10, 162.
107. X. W. Hou, W. J. Wang, X. H. Gao, K. Ran, Y. L. Huang, Z. D. Zhang, Y. Fang, S. Wang, D. X. He, W. P. Ye, R. Zhao and W. D. Xue, Carbon, 2022, 199, 268–278.
108. Y. M. Cheng, Y. X. Han, W. Zhang, L. Zeng, Y. Y. Long, S. Y. Wang and Q. H. Weng, Chem. Eng. J., 2022, 437, 135304.
109. K. Sun, C. Wang, M. Tebyetekerwa and X. S. Zhao, Chem. Eng. J., 2022, 446, 137211.
110. M. Y. Chen, W. X. Zhu, H. L. Guo, Z. H. Tian, L. Q. Zhang, J. T. Wang, T. X. Liu, F. L. Lai and J. J. Huang, Energy Storage Mater., 2023, 59, 102760.
111. C. Li, E. Lepre, M. Bi, M. Antonietti, J. W. Zhu, Y. S. Fu and N. López-Salas, Adv. Sci., 2023, 10, 2300526.
112. T. B. Nguyen, B. Yoon, T. D. Nguyen, E. Oh, Y. Ma, M. Wang and J. Suhr, Carbon, 2023, 206, 383–391.
113. W. X. Zhong, B. Li, Z. Ma, C. L. Zhu, F. Yan, X. T. Zhang and Y. J. Chen, Carbon, 2023, 202, 235–243.
114. S. C. Sun, H. Jiang, Z. Y. Chen, Q. Chen, M. Y. Ma, L. Zhen, B. Song and C. Y. Xu, Angew. Chem., Int. Ed., 2022, 61, e202202519.
115. M. Malinovic, P. Paciok, E. S. Koh, M. Geuß, J. Choi, P. Pfeifer, J. P. Hofmann, D. Göhl, M. Heggen, S. Cherevko and M. Ledendecker, Adv. Energy Mater., 2023, 13, 2301450.
116. Q. Zhang, P. Y. Pan, X. W. Xia, S. J. Wang, Z. Y. Pang, G. S. Li, L. Ji, X. Yu, X. G. Lu and X. L. Zou, J. Mater. Chem. A, 2024, 12, 15386–15397.
117. Y. H. Sun, J. J. Zhang, L. Y. Liu, Y. K. Qin, Y. M. Sun, W. Xu and D. B. Zhu, Sci. China:Chem., 2016, 59, 1323–1329.
118. M. C. Lu, J. Sun, B. Y. Cui, J. Zhang, J. L. Ren and R. Y. Li, Sep. Purif. Technol., 2024, 340, 126736.
119. M. Surendran, S. Singh, H. D. Chen, C. Wu, A. Avishai, Y. T. Shao and J. Ravichandran, Adv. Mater., 2024, 36, 2312620.
120. A. Kaboli, N. Esfandiari, G. B. Darband, R. Sharifi, M. Aliofkhazraei and A. S. Rouhaghdam, J. Electroanal. Chem., 2024, 958, 118151.
121. D. Jeon, D. Y. Kim, H. Kim, N. Kim, C. Lee, D. H. Seo and J. Ryu, Adv. Mater., 2024, 36, 2304468.
122. S. Kaushik, D. J. Wu, Z. Zhang, X. Xiao, C. Zhen, W. J. Wang, N. Y. Huang, M. Gu and Q. Xu, Adv. Mater., 2024, 36, 2401163.
123. H. Y. Jeong, J. Oh, G. S. Yi, H. Y. Park, S. K. Cho, J. H. Jang, S. J. Yoo and H. S. Park, Appl. Catal., B, 2023, 330, 122596.
124. G. Jiang, H. M. Yu, D. W. Yao, Y. H. Li, J. Chi, H. J. Zhang and Z. G. Shao, J. Mater. Chem. A, 2022, 10, 11893–11903.
125. I. Suzuki, S. Suzuki, T. Watanabe, M. Kita and T. Omata, J. Asian Ceram. Soc., 2022, 10, 520–529.
126. M. Espindola-Rodriguez, Y. Sanchez, S. López-Marino, D. Sylla, M. Placidi, M. Neuschitzer, H. Xie, V. Izquierdo-Roca, O. Vigil-Galán and E. Saucedo, J. Anal. Appl. Pyrolysis, 2016, 120, 45–51.
127. J. C. Li, T. Li, G. G. Xue, J. Liu, P. Y. Ma and L. Wang, Sep. Purif. Technol., 2024, 346, 127524.
128. L. Y. Niu, B. C. Zhang, J. Sun, J. J. Wang, C. X. Qin and L. X. Dai, Composites, Part B, 2023, 266, 111046.
129. Y. M. Zhu, M. Klingenhof, C. L. Gao, T. Koketsu, G. Weiser, Y. C. Pi, S. H. Liu, L. J. Sui, J. R. Hou, J. Y. Li, H. M. Jiang, L. M. Xu, W. H. Huang, C. W. Pao, M. H. Yang, Z. W. Hu, P. Strasser and J. W. Ma, Nat. Commun., 2024, 15, 1447.
130. X. J. Chen, W. M. Li, N. Song, M. X. Zhong, S. Yan, J. Q. Xu, W. D. Zhu, C. Wang and X. F. Lu, Chem. Eng. J., 2022, 440, 135851.
131. B. Małecka, A. Łącz, E. Drożdż and A. Małecki, J. Therm. Anal. Calorim., 2014, 119, 1053–1061.
132. Y. F. Zhang, T. Y. Wang, L. Mei, R. J. Yang, W. W. Guo, H. Li and Z. Y. Zeng, Nano-Micro Lett., 2024, 16, 180.
133. J. F. Jiang, L. Xu, L. J. Du, L. Li, G. Y. Zhang, C. G. Qiu and L. M. Peng, Nat. Electron., 2024, 7, 545–556.
134. W. W. Shi, B. X. Ge, P. Y. Jiang, Q. Y. Wang, L. F. He and C. J. Huang, Appl. Catal., B, 2024, 354, 124121.
135. J. Wang, M. Zhang, Z. H. Chen, L. N. Li, G. C. Jiang and Z. Q. Li, ACS Energy Lett., 2024, 9, 653–661.
136. X. J. Fang, L. P. Ren, F. Li, Z. X. Jiang and Z. G. Wang, Rare Met., 2021, 41, 901–910.
137. W. Y. Fu, Y. X. Lin, M. S. Wang, S. Si, L. Wei, X. S. Zhao and Y. S. Wei, Rare Met., 2022, 41, 3069–3077.
138. H. H. Lyu, X. Wang, P. Li, P. Yan and J. C. Tang, Appl. Catal., B, 2024, 354, 124123.
139. Y. R. Wang, X. X. Li, W. J. Xu, D. Y. Chen, N. J. Li, Q. F. Xu, H. Li and J. M. Lu, Appl. Catal., B, 2024, 348, 123839.
140. D. Wang, Y. X. Chang, Y. R. Li, S. L. Zhang and S. L. Xu, Rare Met., 2021, 40, 3156–3165.
141. L. F. Yang, C. L. Du, J. C. Tian, X. Y. Yao, Q. W. Zhang, X. L. Ma, Y. Q. Zhu, M. S. Zou and C. B. Cao, Appl. Catal., B, 2024, 355, 124190.
142. M. Wang, X. L. Liu, W. Zhang, H. W. Niu, L. M. Xu, J. T. Han and Y. H. Lu, Energy Storage Mater., 2024, 70, 103459.
143. C. H. An, W. Kang, Q. B. Deng and N. Hu, Rare Met., 2021, 41, 378–384.
144. B. L. Deng, L. P. Guo, Y. Lu, H. B. Rong and D. C. Cheng, Rare Met., 2021, 41, 911–920.
145. W. L. Zhao, F. H. Xu, L. Q. Liu, M. Liu and B. C. Weng, Adv. Mater., 2023, 35, 2308060.
146. J. A. Esterhuizen, A. Mathur, B. R. Goldsmith and S. Linic, J. Am. Chem. Soc., 2024, 146, 5511–5522.
147. P. P. Sun, Z. L. Qiao, X. B. Dong, R. Jiang, Z. T. Hu, J. Yun and D. P. Cao, J. Am. Chem. Soc., 2024, 146, 15515–15524.
148. J. Liang, X. T. Gao, K. Xu, J. Lu, D. L. Liu, Z. W. Zhao, E. C. M. Tse, Z. Q. Peng, W. B. Zhang and J. X. Liu, Small, 2023, 19, 2304889.
149. C. Pan, Z. Mao, X. Yuan, H. J. Zhang, L. Mei and X. Y. Ji, Adv. Sci., 2022, 9, 2105747.
150. F. Han, W. Xu, C. X. Jia, X. T. Chen, Y. P. Xie, C. Zhen and G. Liu, Rare Met., 2022, 42, 853–861.
151. H. B. Liu, J. Gao, X. C. Xu, Q. H. Jia, L. Yang, S. T. Wang and D. P. Cao, Chem. Eng. J., 2022, 448, 137706.
152. G. Ren, B. Liu, X. Xu, P. Jing, J. Wu and J. Zhang, Nano Today, 2024, 56, 102315.
153. H. P. Huang, L. H. Fu, W. Q. Kong, H. R. Ma, X. Zhang, J. L. Cai, S. P. Wang, Z. X. Xie and S. F. Xie, Small, 2022, 18, 2201333.
154. Z. X. Wu, Y. L. Wang, D. Z. Liu, B. W. Zhou, P. F. Yang, R. Z. Liu, W. P. Xiao, T. Y. Ma, J. S. Wang and L. Wang, Adv. Funct. Mater., 2023, 33, 2307010.
155. N. Wang, S. L. Ning, X. L. Yu, D. Chen, Z. L. Li, J. C. Xu, H. Meng, D. K. Zhao, L. G. Li, Q. M. Liu, B. Z. Lu and S. W. Chen, Appl. Catal., B, 2022, 302, 120838.
156. J. J. Zhang, X. Yang, G. F. Xu, B. B. Kumar and R. Balasubramanian, Adv. Mater., 2024, 36, 2309205.
157. X. W. Sun, W. W. Meng, K. H. Ngai, Z. G. Nie, C. H. Luan, W. J. Zhang, S. A. Li, X. H. Lu, B. Wu, G. F. Zhou, M. Z. Long and J. B. Xu, Adv. Mater., 2024, 36, 2400347.
158. L. J. Yang, G. Sun, H. L. Fu and L. Zhang, Chem. Eng. J., 2023, 472, 145052.
159. L. Lv, R. H. Lu, J. X. Zhu, R. H. Yu, W. Zhang, E. H. Cui, X. B. Chen, Y. H. Dai, L. M. Cui, J. Li, L. Zhou, W. Chen, Z. Y. Wang and L. Q. Mai, Angew. Chem., Int. Ed., 2023, 62, e202303117.
160. M. Fathabadi, M. Qorbani, A. Sabbah, S. Quadir, C. Y. Huang, K. H. Chen, L. C. Chen and N. Naseri, J. Mater. Chem. A, 2022, 10, 16655–16665.
161. L. B. Wu, F. H. Zhang, S. W. Song, M. H. Ning, Q. Zhu, J. Q. Zhou, G. H. Gao, Z. Y. Chen, Q. C. Zhou, X. X. Xing, T. Tong, Y. Yao, J. M. Bao, L. Yu, S. Chen and Z. F. Ren, Adv. Mater., 2022, 34, 2201774.
162. S. L. Liu, W. X. Miao, K. Ma, H. F. Teng, X. Zhang, J. L. Li, W. Z. Li, X. J. Cui and L. H. Jiang, Appl. Catal., B, 2024, 350, 123919.
163. N. Zhang, J. Y. Du, N. Zhou, D. P. Wang, D. Bao, H. X. Zhong and X. B. Zhang, Chin. J. Catal., 2023, 53, 134–142.
164. J. J. Zhang, L. L. Xu, X. X. Yang, S. Guo, Y. F. Zhang, Y. Zhao, G. Wu and G. Li, Angew. Chem., Int. Ed., 2024, e202405641,  DOI:10.1002/anie.202405641.
165. L. Wang, R. F. Du, X. Liang, Y. C. Zou, X. Zhao, H. Chen and X. X. Zou, Adv. Mater., 2024, 36, 2312608.
166. Z. Q. Hou, C. H. Cui, Y. N. Li, Y. J. Gao, D. M. Zhu, Y. F. Gu, G. Y. Pan, Y. Q. Zhu and T. Zhang, Adv. Mater., 2023, 35, 2209876.
167. A. Bao, Y. H. Gu, Y. X. Zhang, B. W. Zhang, J. C. Wu, B. Ni, X. Y. Zhang, H. J. Pan and X. W. Qi, Carbon Energy, 2024, 6, e519.
168. Y. T. Yan, J. H. Lin, K. K. Huang, X. H. Zheng, L. Qiao, S. D. Liu, J. Cao, S. C. Jun, Y. Yamauchi and J. Qi, J. Am. Chem. Soc., 2023, 145, 24218–24229.
169. H. Su, C. Y. Yang, M. H. Liu, X. Zhang, W. L. Zhou, Y. H. Zhang, K. Zheng, S. X. Lian and Q. H. Liu, Nat. Commun., 2024, 15, 95.
170. B. Huang, H. Y. Xu, N. N. Jiang, M. H. Wang, J. R. Huang and L. H. Guan, Adv. Sci., 2022, 9, 2201654.
171. Y. Du, F. K. Xie, M. F. Lu, R. X. Lv, W. X. Liu, Y. D. Yan, S. C. Yan and Z. G. Zou, Nat. Commun., 2024, 15, 1780.
172. L. M. Deng, S. F. Hung, Z. Y. Lin, Y. Zhang, C. C. Zhang, Y. X. Hao, S. Y. Liu, C. H. Kuo, H. Y. Chen, J. Peng, J. Z. Wang and S. J. Peng, Adv. Mater., 2023, 35, 2305939.
173. Z. Y. Wu, F. Y. Chen, B. Y. Li, S. W. Yu, Y. Z. Finfrock, D. M. Meira, Q. Q. Yan, P. Zhu, M. X. Chen, T. W. Song, Z. Y. Yin, H. W. Liang, S. Zhang, G. F. Wang and H. T. Wang, Nat. Mater., 2022, 22, 100–108.
174. L. Li, J. Zhou, X. Wang, J. Gracia, M. Valvidares, J. Ke, M. M. Fang, C. Q. Shen, J. M. Chen, Y. C. Chang, C. W. Pao, S. Y. Hsu, J. F. Lee, A. Ruotolo, Y. Y. Chin, Z. W. Hu, X. Q. Huang and Q. Shao, Adv. Mater., 2023, 35, 2302966.
175. H. Y. Jin, X. Y. Liu, P. F. An, C. Tang, H. M. Yu, Q. H. Zhang, H. J. Peng, L. Gu, Y. Zheng, T. Song, K. Davey, U. Paik, J. C. Dong and S. Z. Qiao, Nat. Commun., 2023, 14, 354.
176. Y. X. Hao, S. F. Hung, C. Tian, L. Q. Wang, Y. Y. Chen, S. Zhao, K. S. Peng, C. C. Zhang, Y. Zhang, C. H. Kuo, H. Y. Chen and S. J. Peng, Angew. Chem., Int. Ed., 2024, e202402018,  DOI:10.1002/anie.202402018.
177. N. Yao, H. N. Jia, J. Zhu, Z. P. Shi, H. J. Cong, J. J. Ge and W. Luo, Chem, 2023, 9, 1882–1896.
178. D. Chen, R. H. Yu, K. S. Yu, R. H. Lu, H. Y. Zhao, J. X. Jiao, Y. T. Yao, J. W. Zhu, J. S. Wu and S. C. Mu, Nat. Commun., 2024, 15, 3928.
179. Y. Wang, Y. Qin, R. Wen, L. Wang, M. Dou and F. Wang, Small, 2024, 20, e2401964.
180. J. R. Kuang, B. L. Deng, Z. Q. Jiang, Y. J. Wang and Z. J. Jiang, Adv. Mater., 2024, 36, 2306934.
181. S. H. Liu, H. Tan, Y. C. Huang, Q. B. Zhang, H. P. Lin, L. Li, Z. W. Hu, W. H. Huang, C. W. Pao, J. F. Lee, Q. Y. Kong, Q. Shao, Y. Xu and X. Q. Huang, Adv. Mater., 2023, 35, 2305659.
182. Z. L. Fan, Y. J. Ji, Q. Shao, S. Z. Geng, W. X. Zhu, Y. Liu, F. Liao, Z. W. Hu, Y. C. Chang, C. W. Pao, Y. Y. Li, Z. H. Kang and M. W. Shao, Joule, 2021, 5, 3221–3234.
183. L. N. Chong, J. G. Wen, E. Song, Z. Z. Yang, I. D. Bloom and W. J. Ding, Adv. Energy Mater., 2023, 13, 2302306.
184. K. Y. Qin, H. Yu, W. X. Zhu, Y. J. Zhou, Z. Y. Guo, Q. Shao, Y. B. Wu, X. P. Wang, Y. Y. Li, Y. J. Ji, F. Liao, Y. Liu, Z. H. Kang and M. W. Shao, Adv. Funct. Mater., 2024, 2402226,  DOI:10.1002/adfm.202402226.
185. J. Qi, H. Y. Zeng, L. Gu, Z. F. Liu, Y. Q. Zeng, E. N. Hong, Y. C. Lai, T. H. Liu and C. Z. Yang, ACS Appl. Mater. Interfaces, 2023, 15, 15269–15278.
186. G. Q. Zhao, Z. X. Luo, B. H. Zhang, Y. P. Chen, X. Z. Cui, J. Chen, Y. F. Liu, M. X. Gao, H. G. Pan and W. P. Sun, Nano Res., 2023, 16, 4767–4774.
187. S. Lee, Y. J. Lee, G. Lee and A. Soon, Nat. Commun., 2022, 13, 3171.
188. C. W. Liang, R. R. Rao, K. L. Svane, J. H. L. Hadden, B. Moss, S. B. Scott, M. Sachs, J. Murawski, A. M. Frandsen, D. J. Riley, M. P. Ryan, J. Rossmeisl, J. R. Durrant and I. E. L. Stephens, Nat. Catal., 2024, 7, 763–775.
189. J. W. Zhao, K. H. Yue, H. Zhang, S. Y. Wei, J. W. Zhu, D. D. Wang, J. Z. Chen, V. Y. Fominski and G. R. Li, Nat. Commun., 2024, 15, 2928.
190. Z. Y. Li, X. Li, M. N. Wang, Q. Wang, P. F. Wei, S. Jana, Z. Q. Liao, J. C. Yu, F. Lu, T. F. Liu and G. X. Wang, Adv. Mater., 2024, 36, 2402643.
191. F. Y. Yu, Z. L. Lang, L. Y. Yin, K. Feng, Y. J. Xia, H. Q. Tan, H. T. Zhu, J. Zhong, Z. H. Kang and Y. G. Li, Nat. Commun., 2020, 11, 490.
192. H. W. Yang, Y. J. Ji, Q. Shao, W. X. Zhu, M. M. Fang, M. J. Ma, F. Liao, H. Huang, Y. Zhang, J. J. Yang, Z. L. Fan, Y. Y. Li, Y. Liu, M. W. Shao and Z. H. Kang, Energy Environ. Sci., 2023, 16, 574–583.
193. J. Choi, Q. T. Nguyen, S. Park, B. G. Ghule, J. H. Park, J. R. Park, U. T. Nakate, J.-H. Jang, D.-W. Kim and S. Park, Chem. Eng. J., 2024, 497, 154407.
194. Z. Li, X. Xing, D. Feng, L. Du, Y. Tian, X. Chen and D. Yang, Ceram. Int., 2024, 50, 21519–21525.
195. J. Ebenezer, A. Lal, P. Velayudham, A. Borenstein and A. Schechter, ACS Appl. Mater. Interfaces, 2024, 16, 36433–36443.
196. H. Lee, J. Theerthagiri, M. L. Aruna Kumari, A. Min, C. J. Moon, V. Anbazhagan, R. L. Brutchey and M. Y. Choi, Int. J. Hydrogen Energy, 2024, 57, 176–186.
197. B. K. Manna, R. Samanta, R. K. Trivedi, B. Chakraborty and S. Barman, J. Colloid Interface Sci., 2024, 670, 258–271.
198. J. Shi, Q. Sun, W. Zhu, T. Cheng, F. Liao, M. Ma, J. Yang, H. Yang, Z. Fan and M. Shao, Chem. Eng. J., 2023, 463, 142385.
199. J. Gao, W. Yu, J. Liu, L. Qin, H. Cheng, X. Cui and L. Jiang, J. Colloid Interface Sci., 2024, 664, 766–778.
200. J. Raghav, D. Deepak, S. Sinha Roy and S. Roy, ACS Appl. Energy Mater., 2023, 6, 2286–2295.
201. Y. Y. Liu, L. Cai, Q. Q. Ji, C. Wang, Z. Y. Liu, L. Y. Lv, B. Tang, H. L. Duan, F. C. Hu, H. J. Wang, N. Li, Z. H. Sun and W. S. Yan, ACS Energy Lett., 2022, 7, 3798–3806.
202. C. X. Liu, Y. B. Jiang, T. Wang, Q. S. Li and Y. Z. Liu, Adv. Sci., 2023, 10, 2207429.
203. Y. Wu, R. Yao, Q. Zhao, J. P. Li and G. Liu, Chem. Eng. J., 2022, 439, 135699.
204. S. M. Li, M. Shi, C. X. Wu, K. Q. Nie, Z. Wei, X. P. Jiang, X. B. Liu, H. L. Chen, X. L. Tian, D. X. Wu and Y. W. Li, Chem. Eng. J., 2024, 485, 150043.
205. L. Q. Wu, Q. Liang, J. Y. Zhao, J. Zhu, H. G. Jia, W. Zhang, P. Cai and W. Luo, Chin. J. Catal., 2023, 55, 182–190.
206. L. Li, G. W. Zhang, C. H. Zhou, F. Lv, Y. J. Tan, Y. Han, H. Luo, D. W. Wang, Y. X. Liu, C. S. Shang, L. Y. Zeng, Q. Z. Huang, R. J. Zeng, N. Ye, M. C. Luo and S. J. Guo, Nat. Commun., 2024, 15, 4974.
207. G. K. Li, H. Jang, S. G. Liu, Z. J. Li, M. G. Kim, Q. Qin, X. Liu and J. Cho, Nat. Commun., 2022, 13, 1270.
208. D. F. Zhang, M. N. Li, X. Yong, H. Q. Song, G. I. N. Waterhouse, Y. F. Yi, B. J. Xue, D. L. Zhang, B. Z. Liu and S. Y. Lu, Nat. Commun., 2023, 14, 2517.
209. W. X. Zhu, X. C. Song, F. Liao, H. Huang, Q. Shao, K. Feng, Y. J. Zhou, M. J. Ma, J. Wu, H. Yang, H. W. Yang, M. Wang, J. Shi, J. Zhong, T. Cheng, M. W. Shao, Y. Liu and Z. H. Kang, Nat. Commun., 2023, 14, 5365.
210. J. X. Wei, K. Xiao, Y. X. Chen, X. P. Guo, B. L. Huang and Z. Q. Liu, Energy Environ. Sci., 2022, 15, 4592–4600.
211. Y. T. Yan, J. H. Lin, T. X. Xu, B. S. Liu, K. K. Huang, L. Qiao, S. D. Liu, J. Cao, S. C. Jun, Y. Yamauchi and J. Qi, Adv. Energy Mater., 2022, 12, 2200434.
212. J. Li, J. Zhao, Z. Lai, Y. Zhang, Y. Wang, R. Jia, H. Wang, Z. Yin, J. Zhang, X. Zheng, Y. Zhang and Y. Deng, Chem. Eng. J., 2024, 492, 152226.
213. R. Wang, D. Zhang, S. Luo, L. Jiang, Q. Wang, L. Chen, G.-F. Wei and X. Wang, Electrochim. Acta, 2024, 507, 145198.
214. C. Li, H. Jang, M. G. Kim, L. Hou, X. Liu and J. Cho, Appl. Catal., B, 2022, 307, 121204.
215. A. S. Wang, W. Y. Wang, J. C. Xu, A. Zhu, C. N. Zhao, M. Yu, G. L. Shi, J. G. Yan, S. H. Sun and W. C. Wang, Adv. Energy Mater., 2023, 13, 2302537.
216. L. X. Peng, L. Su, X. Yu, R. Y. Wang, X. Z. Cui, H. Tian, S. W. Cao, B. Y. Xia and J. L. Shi, Appl. Catal., B, 2022, 308, 9.
217. Y. Chen, Y. D. Liu, L. Li, T. Sakthivel, Z. X. Guo and Z. F. Dai, Adv. Funct. Mater., 2024, 34, 2401452.
218. J. D. Chen, C. H. Chen, M. K. Qin, B. Li, B. B. Lin, Q. Mao, H. B. Yang, B. Liu and Y. Wang, Nat. Commun., 2022, 13, 5382.
219. B. X. Zhang, J. M. Wang, G. M. Liu, C. M. Weiss, D. Q. Liu, Y. P. Chen, L. X. Xia, P. Zhou, M. X. Gao, Y. F. Liu, J. Chen, Y. S. Yan, M. H. Shao, H. G. Pan and W. P. Sun, Nat. Catal., 2024, 7, 441–451.
220. 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.
221. L. Liu, Y. Wang, Y. Z. Zhao, Y. Wang, Z. L. Zhang, T. Wu, W. J. Qin, S. J. Liu, B. R. Jia, H. Y. Wu, D. Y. Zhang, X. H. Qu, M. Chhowalla and M. L. Qin, Adv. Funct. Mater., 2022, 32, 2112207.
222. J. W. Chang, Y. Y. Shi, H. Wu, J. K. Yu, W. Jing, S. Y. Wang, G. I. N. Waterhouse, Z. Y. Tang and S. Y. Lu, J. Am. Chem. Soc., 2024, 146, 12958–12968.
223. K. Liu, X. Gao, C. X. Liu, R. Shi, E. C. M. Tse, F. Liu and Y. Chen, Adv. Energy Mater., 2024, 14, 2304065.
224. C. Z. Yuan, S. Wang, K. S. Hui, K. X. Wang, J. F. Li, H. X. Gao, C. Y. Zha, X. Zhang, D. A. Dinh, X. L. Wu, Z. K. Tang, J. W. Wan, Z. P. Shao and K. N. Hui, ACS Catal., 2023, 13, 2462–2471.
225. Y. X. Weng, K. Y. Wang, S. Y. Li, Y. X. Wang, L. F. Lei and L. Z. Zhuang, Adv. Sci., 2023, 10, 2205920.
226. H. F. Yuan, J. W. Li, Z. F. Tang, Y. J. Wang, T. Wu, M. Huang, L. L. Zhao, Z. H. Zhao, H. Liu, C. X. Xu, X. Y. Liu and W. J. Zhou, Chem. Eng. J., 2024, 492, 152339.
227. P. J. Wang, Y. T. Yan, P. C. Wang, Z. Y. Ye, X. H. Zheng and W. Cai, Chem. Eng. J., 2023, 455, 140856.
228. S. Z. Zhou, H. Jang, Q. Qin, L. Q. Hou, M. G. Kim, S. G. Liu, X. Liu and J. Cho, Angew. Chem., Int. Ed., 2022, 61, e202212196.
229. L. L. Zhai, X. J. She, L. C. Zhuang, Y. Y. Li, R. Ding, X. Y. Guo, Y. Q. Zhang, Y. Zhu, K. Xu, H. J. Fan and S. P. Lau, Angew. Chem., Int. Ed., 2022, 61, e202116057.
230. J. D. Wu, J. C. Fan, X. Zhao, Y. Wang, D. W. Wang, H. T. Liu, L. Gu, Q. H. Zhang, L. R. Zheng, D. J. Singh, X. Q. Cui and W. T. Zheng, Angew. Chem., Int. Ed., 2022, 61, e202207512.
231. X. Y. Liu, Y. X. Zhou, J. K. Lin, X. Xiao, Z. J. Wang, L. Y. Jia, M. Y. Li, K. Yang, J. C. Fan, W. W. Yang and G. S. Li, Angew. Chem., Int. Ed., 2024, 19, e202406650.
232. Z. H. Tao, H. Y. Zhao, N. Lv, X. Luo, J. Yu, X. Tan and S. C. Mu, Adv. Funct. Mater., 2024, 34, 2312987.
233. Z. Q. Niu, Z. K. Lu, Z. L. Qiao, S. T. Wang, X. H. Cao, X. D. Chen, J. Yun, L. R. Zheng and D. P. Cao, Adv. Mater., 2024, 36, 2310690.
234. S. Y. Ge, R. K. Xie, B. Huang, Z. Y. Zhang, H. M. Liu, X. Kang, S. Q. Hu, S. H. Li, Y. T. Luo, Q. M. Yu, J. W. Wang, G. L. Chai, L. H. Guan, H. M. Cheng and B. L. Liu, Energy Environ. Sci., 2023, 16, 3734–3742.
235. S. C. Karthikeyan, S. Sidra, S. Ramakrishnan, D. H. Kim, P. J. J. Sagayaraj, K. Sekar and D. J. Yoo, Appl. Catal., B, 2024, 355, 124196.
236. X. Long, B. Zhao, Q. Q. Zhao, X. X. Wu, M. N. Zhu, R. F. Feng, M. Shakouri, Y. Zhang, X. X. Xiao, J. J. Zhang, X. Z. Fu and J. L. Luo, Appl. Catal., B, 2024, 343, 123559.
237. X. F. Wang, H. Jang, S. G. Liu, Z. J. Li, X. H. Zhao, Y. F. Chen, M. G. Kim, Q. Qin and X. Liu, Adv. Energy Mater., 2023, 13, 2301673.
238. Y. Y. Li, Q. W. Zhang, X. R. Zhao, H. F. Wu, X. Y. Wang, Y. Q. Zeng, Q. Chen, M. W. Chen and P. Liu, Adv. Funct. Mater., 2023, 33, 2214124.
239. A. L. Li, S. Kong, K. Adachi, H. Ooka, K. Fushimi, Q. K. Jiang, H. Ofuchi, S. Hamamoto, M. Oura, K. Higashi, T. Kaneko, T. Uruga, N. Kawamura, D. Hashizume and R. Nakamura, Science, 2024, 384, 666–670.
240. Z. Shaghaghi and S. Akbari, Int. J. Hydrogen Energy, 2024, 51, 936–949.
241. W. Y. Gou, Z. M. Xia, X. H. Tan, Q. Y. Xue, F. Ye, S. Dai, M. K. Zhang, R. Si, Y. Zou, Y. Y. Ma, J. C. Ho and Y. Q. Qu, Nano Energy, 2022, 104, 107960.
242. Y. N. Wang, M. C. Zhang, Z. Y. Kang, L. Shi, Y. C. Shen, B. Y. Tian, Y. C. Zou, H. Chen and X. X. Zou, Nat. Commun., 2023, 14, 5119.
243. Y. R. Chen, D. N. Liu, Q. Q. Zhao, X. Long, J. Wang, J. J. Zhang, X. Z. Fu and J. L. Luo, Chem. Eng. J., 2023, 475, 146255.
244. Q. B. Huang, S. J. Zhuang, Y. X. Zheng, X. Y. Peng, Z. G. Ye and D. S. Li, Chem. Eng. J., 2024, 491, 152040.
245. S. Q. Niu, X. P. Kong, S. W. Li, Y. Y. Zhang, J. Wu, W. W. Zhao and P. Xu, Appl. Catal., B, 2021, 297, 120442.
246. Z. H. Zhang, X. H. Liu, D. Wang, H. Wan, Y. Zhang, G. Chen, N. Zhang and R. Z. Ma, Chem. Eng. J., 2022, 446, 137037.
247. M. Paidar, V. Fateev and K. Bouzek, Electrochim. Acta, 2016, 209, 737–756.
248. S. Trasatti, J. Electroanal. Chem., 1999, 476, 90–91.
249. N. Dubouis and A. Grimaud, Chem. Sci., 2019, 10, 9165–9181.
250. J. E. Chae, S. Y. Lee, S. Y. Baek, K. H. Song, C. H. Park, H.-J. Kim and K.-S. Lee, J. Membr. Sci., 2021, 638, 119694.
251. F. Z. Aouali, M. Becherif, H. S. Ramadan, M. Emziane, A. Khellaf and K. Mohammedi, Int. J. Hydrogen Energy, 2017, 42, 1366–1374.
252. W. Zhang, C. Zhu, Y. Wen, M. Wang, Z. Lu and Y. Wang, Angew. Chem., Int. Ed., 2024, e202418456,  DOI:10.1002/anie.202418456.
253. S. Venkatesan, J. Mitzel, S. S. Ambu, T. Morawietz, I. Biswas, O. Recalde, E. Adabifiroozjaei, L. Molina-Luna, D. Estes, K. Wegner, P. Gazdzicki, A. Gago and K. Friedrich, Adv. Energy Mater., 2024, 2401659.
254. C. Zhou, L. Li, Z. Dong, F. Lv, H. Guo, K. Wang, M. Li, Z. Qian, N. Ye, Z. Lin, M. Luo and S. Guo, Nat. Commun., 2024, 15, 9774.
255. Y. Shen, X. L. Zhang, M. R. Qu, J. Ma, S. Zhu, Y. L. Min, M. R. Gao and S. H. Yu, Nat. Commun., 2024, 15, 7861.
256. D. Li, D. Xu, Y. Pei, Q. Zhang, Y. Lu and B. Zhang, J. Am. Chem. Soc., 2024, 146, 28728–28738.

---

Footnote

† Feng Wang and Linfeng Xiao contributed equally to this article.

---