Suppressing migration of Ru in a high-entropy alloy for durable acidic oxygen evolution

Abstract

Ru-based catalysts are recognized as some of the most advanced materials for acidic oxygen evolution reaction (OER). Nonetheless, their vulnerability to over-oxidation to soluble RuO₄ during catalytic processes presents considerable challenges for their long-term stability. To address the issue of Ru atom leaching within the lattice, a Ru-based high entropy alloy electrocatalyst (HEA/PANI-CP) has been synthesized on polyaniline-modified carbon paper utilizing an innovative “thermal shock” technique. This method employs compositional engineering to amplify the high entropy effects, thereby effectively suppressing element diffusion and controlling the migration energies of active Ru sites. Consequently, the catalyst demonstrates a significant decrease in elemental dissolution, which contributes to enhanced stability during the OER. Importantly, the HEA catalyst exhibits remarkable performance, achieving an overpotential of 258 mV at a current density of 10 mA cm⁻² while exhibiting extraordinary stability with minimal degradation over 300 hours of operation. These results position HEA/PANI-CP as a leading candidate for acid-stable OER catalysts, offering a promising avenue for the advancement of durable Ru-based electrocatalysts.

---

Hao Wu

Hao Wu earned his PhD from Fudan University in 2016, followed by postdoctoral training at the National University of Singapore (2016–2018) and King Abdullah University of Science & Technology (2018–2020). He is a Professor in the School of Chemistry and Chemical Engineering at Shandong University, where he leads a research group focused on electrocatalysis and electrosynthesis, with an emphasis on designing novel metal–organic framework (MOF) catalysts for applications in renewable energy conversion and green chemical production. Dr Wu has served as a Session Presider at the ACS Fall Meetings (2023 and 2024).

1. Introduction

Proton-exchange-membrane water electrolysis (PEMWE) has emerged as a promising technology for producing green hydrogen gas, owing to its ampere-level current density at low overpotentials, minimal ohmic losses, and low gas crossover rates.^1–3 However, the implementation of PEMWE is hindered by the need for efficient and stable electrocatalysts for the oxygen evolution reaction (OER), particularly under harsh corrosive and oxidative conditions.^4–6 The requirement for enhanced catalyst activity and stability in PEM systems restricts viable electrocatalysts to noble metals such as iridium (Ir) and ruthenium (Ru). While Ru-based catalysts present a significant economic advantage over Ir-based counterparts, RuO₂ is prone to excessive oxidation to soluble RuO₄ derivatives under highly acidic conditions, which compromises long-term stability and operational efficiency.⁷ Hence, it is imperative to develop strategies aimed at preventing the leaching of Ru from electrocatalysts.

High-entropy alloy (HEA) materials, characterized by their composition of five or more elements randomly arranged in a single solid solution phase, have garnered increasing attention in catalysis due to their unique properties.^8–13 The random mixing of elements, along with complex multi-element synergies, allows for tunable electronic structures and adjustable adsorption energies, thereby enhancing catalytic performance across various applications.^14–19 In particular, the inherent lattice distortion in HEAs, arising from discrepancies in atomic sizes, may elevate the energy barriers for element diffusion, thereby mitigating the leaching of components into corrosive environments. The multicomponent interactions present in HEAs may also provide synergistic effects, such as ligand and strain effects, which can modify the d-band structure and electronic characteristics of active sites, further facilitating catalytic reactions.^20,21 Thus, leveraging the advantages of HEAs in the design of Ru-based electrocatalysts may enhance both catalytic durability and activity, while simultaneously reducing the reliance on Ru.

Herein, we synthesized an HEA catalyst comprising five metals (FeCoNiMnRu) in a single solid solution phase, which is supported on polyaniline-modified carbon paper (HEA/PANI-CP) using thermal shock technology for efficient and stable OER performance. Following the thermal shock treatment, the PANI coating layer undergoes conversion into N-doped carbon species, which can enhance the robustness of the carbon support through mitigating gasification of the carbon and the formation of passivated oxide layers in oxidative environments.^22–24 Moreover, the self-supporting HEA/PANI-CP electrode would allow for uniform catalyst distribution, improving electrolyte–electrode interactions, and therefore accelerating the reaction kinetics. Density functional theory (DFT) calculations indicate that Ru atoms within the HEA exhibit greater migration energy compared to those in binary alloys, contributing to their enhanced durability under acidic conditions. The HEA/PANI-CP catalyst demonstrated remarkable OER activity with an overpotential of 258 mV at 10 mA cm⁻², with excellent stability over 300 hours.

2. Experimental section

2.1 Materials and chemicals

All chemicals were of analytical grade and directly used without further purification in the experiments. Iron chloride hexahydrate (FeCl₃·6H₂O), cobalt chloride hexahydrate (CoCl₂·6H₂O), nickel chloride hexahydrate (NiCl₂·6H₂O), ruthenium chloride hydrate (RuCl₃·3H₂O) and manganese chloride tetrahydrate (MnCl₂·4H₂O) were purchased from Sinopharm Chemical Reagent Co., Ltd. Ruthenium oxide (RuO₂) and Nafion (5 wt%) were purchased from Aladdin Ltd.

2.2 Preparation of polyaniline-deposited CP (PANI-CP)

A 2 × 3 cm² piece of carbon paper (CP) was typically subjected to heat treatment at 580 °C for 10 minutes in air using a muffle furnace, followed by a 30-minute wash with concentrated nitric acid under ultrasonic agitation at room temperature. Subsequently, the processed CP, along with a carbon rod and Ag/AgCl, served as the working, counter, and reference electrodes in a three-electrode electrochemical setup. The electrolyte consisted of 3.5 mL concentrated nitric acid, 44 mL deionized water, and 2.5 mL aniline. Polyaniline (PANI) was electrochemically deposited onto the CP at a constant voltage of 0.7 V versus Ag/AgCl for 600 seconds. After deposition, the CP was rinsed three times with deionized water and then dried at 60 °C for 2 hours to yield the PANI-decorated CP sample.

2.3 Preparation of HEA/PANI-CP

The precursor solution (0.05 mol L⁻¹) was formulated by dissolving equal molar concentrations of various metal salts CoCl₂·6H₂O, NiCl₂·6H₂O, FeCl₃·6H₂O, MnCl₂·4H₂O, and RuCl₃·3H₂O in ethanol. This mixed non-noble metal solution was subsequently introduced into PANI-CP at a predetermined loading concentration. The resulting mixture was then dried in a vacuum oven to produce precursor-loaded PANI-CP. Finally, a thermal shock method was employed to synthesize the catalysts, involving a high-temperature shock created by applying a large current pulse (20 A for 1 s) in a vacuum environment. Additionally, a binary alloy was synthesized through the same procedure, utilizing only two of the metal salts, including Ru.

3. Results and discussion

The synthesis process of the HEA on carbon paper (CP) is illustrated in Fig. 1a and consists of two primary steps. Initially, polyaniline (PANI) nanofibers were electrodeposited onto pristine CP, referred to as PANI-CP, utilizing a three-electrode setup. Subsequently, PANI-CP was immersed in metal precursors and subjected to thermal shock treatment, yielding HEA/PANI-CP. When a current pulse (20 A, 1 s) was applied to PANI-CP, it was instantly heated to 1783 K in 40 ms, which then rapidly cooled to room temperature within 650 ms (Fig. S1†). The elemental compositions of Fe, Co, Ni, Mn, and Ru in HEA/PANI-CP, determined via inductively coupled plasma mass spectrometry, were found to be 2.64, 3.38, 3.21, 3.27, and 4.65 wt%, respectively. This outcome confirms that the elements are approximately mixed in equal ratios. According to the Hume-Rothery rule, the formation of solid solutions requires three conditions: (1) minimal atomic radius discrepancies (δ ≤ 6.6%); (2) comparable crystal structures between the solute and solvent; (3) nearly equivalent electronegativities that facilitate the creation of solid solution alloys. The calculated δ for the HEA is 2.87%, fulfilling the geometric prerequisites for the establishment of a single-phase disordered solid solution. The Gibbs free energy (ΔG_mix) thermodynamic equation indicates that increasing entropy can alleviate the constraints on mutual solubility among metal elements. Generally, HEAs are defined as alloys comprising at least five elements with a mixing entropy (ΔS) exceeding 1.5R (where R = 8.314 J mol⁻¹ K⁻¹). Based on the ICP data presented in Table S1,† the mixing entropy (ΔS_mix = 13.299 J mol⁻¹ K⁻¹) for the HEA is determined to be greater than 1.5R. This analysis suggests that the HEA possesses a sufficiently high ΔS_mix and a low δ, meeting the fundamental conditions for the development of a single-phase disordered solid solution, thus making the synthesis of high-entropy alloys from these five elements theoretically viable.

Fig. 1 (a) Schematic illustration of the preparation process of HEA/PANI-CP; (b) XRD patterns of HEA/PANI-CP; (c) high magnification SEM images of HEA/PANI-CP; (d) HRTEM images of HEA/PANI-CP; inset panel in (d) shows the FFT pattern; (e) EDS elemental mapping of HEA/PANI-CP.

As noted by Guo et al., the valence electron concentration (VEC) of a solid solution influences the crystalline phase of HEAs.²⁵ Typically, a VEC below 6.87 results in a body-centered cubic (bcc) phase, while a VEC ranging from 6.87 to 8.0 leads to a mixed face-centered cubic (fcc) and bcc phase. A VEC exceeding 8.0 indicates the formation of the fcc phase. The VEC for the HEA is calculated to be 8.3, favoring the fcc structure. The X-ray diffraction (XRD) pattern of HEA/PANI-CP, shown in Fig. 1b, confirms successful synthesis, with diffraction peaks at 43.3°, 53.0°, and 77.9° corresponding to the (111), (200), and (220) planes, respectively, indicative of a distinct fcc structure. Fig. S2a† displays the XRD of PANI-CP, revealing two prominent peaks at 26.3° and 54.3°, which can be attributed to the (002) and (004) planes of graphite. Scanning electron microscopy (SEM) images of PANI-CP in Fig. S2b† exhibit a loose porous structure, facilitating effective impregnation of metal salts and uniform catalyst loading. Fig. 1c illustrates the presence of nanoparticles adhering to the carbon fiber surface, along with numerous nanoneedles uniformly distributed across it, enhancing electron and water molecule transfer and adsorption while providing additional active sites for catalytic reactions. This unique morphology also promotes the formation of superhydrophobic interfaces, enabling the rapid release of bubbles and preventing significant energy loss due to blockage of active centers and ionic conduction pathways. The high-resolution transmission electron microscopy (HR-TEM) image in Fig. 1d shows clear lattice fringes with an interplanar distance of 0.228 nm, corresponding to the (111) plane of HEA/PANI-CP. The inset displays the associated fast Fourier transform pattern, where the hexagonal arrangement of spots indicates the [111] direction of the projected face. Elemental mapping in Fig. 1e further confirms the uniform distribution of these elements.

X-ray photoelectron spectroscopy (XPS) tests were performed to further investigate the surface chemical state of these five elements (Fig. 2a). The high-resolution XPS spectrum for Fe 2p displays two prominent peaks at 707.8 eV and 720.4 eV (Fig. 2b), which indicate the presence of metallic Fe in the HEA. Moreover, two peaks at binding energies of 712.4 eV and 725.1 eV correspond to Fe²⁺. Satellite peaks are detected at binding energies of 715.3 eV and 730.8 eV.^26,27Fig. 2c shows the deconvoluted core-level Co 2p spectra, where the primary peaks at 781.1 eV and 795.7 eV are attributed to Co³⁺, while those at 783.4 eV and 797.7 eV are linked to Co²⁺. Furthermore, the peaks at 778.7 eV and 793.7 eV represent Co⁰, with satellite peaks found at 786.5 eV and 802.8 eV.²⁸ The Ni 2p spectrum exhibits two satellite peaks at 859.2 eV and 876.9 eV alongside two spin–orbit peaks (Fig. 2d). Specifically, the spin–orbit double peaks correspond to Ni 2p_3/2 and Ni 2p_1/2, while peaks at 853.2 eV and 870.5 eV are associated with Ni⁰, and those at 855.7 eV and 873.4 eV correspond to Ni²⁺.²⁹ The Mn 2p XPS spectrum (Fig. 2e) reveals peaks at 638.5 eV and 645.3 eV for Mn⁰ and peaks at 640.9 eV and 648.7 eV for Mn²⁺. In the Ru 3p core-level XPS spectra (Fig. 2f), peaks at 462.1 eV and 484.9 eV are attributed to Ru 3p_3/2 and Ru 3p_1/2, respectively.³⁰

Fig. 2 (a) XPS spectra of the HEA/PANI-CP composite; high-resolution XPS spectra of (b) Fe 2p; (c) Co 2p; (d) Ni 2p; (e) Mn 2p; (f) Ru 3p.

The OER performance of CP, PANI-CP, HEA/CP and HEA/PANI-CP was first assessed under acidic conditions using a three-electrode electrochemical system with a 0.5 M H₂SO₄ electrolyte. An Ag/AgCl electrode served as the reference, while a carbon rod functioned as the counter electrode. Prior to conducting electrochemical tests, each electrode underwent 50 cycles of cyclic voltammetry (CV) to ensure stabilization. The linear sweep voltammetry (LSV) curves, with 100% iR compensation, for CP, PANI-CP, HEA/CP, and HEA/PANI-CP are illustrated in Fig. 3a. Notably, HEA/PANI-CP exhibited an overpotential of 258 mV at 10 mA cm⁻², significantly lower than those observed for CP (664 mV), PANI-CP (643 mV), and HEA/CP (286 mV). To further highlight the advantages of high-entropy alloys over conventional binary alloys, the OER performance of Ru-based binary alloy catalysts containing various elements was examined (Fig. S4a†). In this comparison, the HEA demonstrated superior catalytic activity. Notably, the Ru-free quaternary alloy FeCoNiMn/PANI-CP exhibited an overpotential of 417 mV at 10 mA cm⁻², dramatically surpassing 258 mV for HEA/PANI-CP, excluding the contributions of Fe, Co, Ni, and Mn for the OER activity (Fig. S5†). The mass-normalized LSV profiles of HEA/PANI-CP (Fig. S6†), combined with a remarkable turnover frequency (TOF) of 0.38O₂ s⁻¹ at an overpotential of 258 mV, provide compelling evidence for the catalyst's superior intrinsic activity in the OER. To delve deeper into the catalytic characteristics of HEA/PANI-CP, the Tafel relationship was employed to assess the kinetics of the catalyst. The corresponding Tafel curves were derived from the LSV data, as depicted in Fig. 3b. The Tafel slope for HEA/PANI-CP was determined to be 105 mV dec⁻¹, markedly lower than those for CP (290 mV dec⁻¹), PANI-CP (359 mV dec⁻¹), and HEA/CP (109 mV dec⁻¹). Fig. 3c shows that all samples present typical semicircles of varying diameters, with R_ct serving as a key indicator of charge transfer kinetics; a lower R_ct value signifies a faster reaction rate at the interface of the electrode and electrolyte. The R_ct for HEA/PANI-CP was approximately 0.93 Ω, indicating relatively rapid reaction kinetics for this catalyst. It is well established that the electrochemically active surface area (ECSA) is directly proportional to the electrochemical double-layer capacitance (C_dl), which indicates the number of exposed active sites. The C_dl values were assessed through CV measurements performed at varying scan rates ranging from 20 mV s⁻¹ to 140 mV s⁻¹ (Fig. S7†). As shown in Fig. 3d, the C_dl value for HEA/PANI-CP was 28.7 mF cm⁻², surpassing those of the other catalysts. The durability of the HEA/PANI-CP electrode was further evaluated via chronopotentiometry tests conducted at a constant current density of 10 mA cm⁻² for 300 hours and 100 mA cm⁻² for 180 hours (Fig. S8†), confirming its exceptional long-term OER stability. Although the catalytic activities of HEA/PANI-CP and HEA/CP are nearly identical, HEA/CP exhibits significantly inferior operational stability compared to HEA/PANI-CP at the same current density, as illustrated in Fig. S9.† This phenomenon can be attributed to the role of graphitic nitrogen in PANI-modified carbon paper, which enhances the electric conductivity of the carbon matrix while redistributing charge density to mitigate the oxidation propensity of carbon atoms. TEM imaging after continuous OER electrolysis reveals the formation of an amorphous surface layer with more lattice disorder compared to the bulk crystalline structure (Fig. S10†). This morphological evolution suggests that surface reconstruction occurred during the OER. The comparative analysis of XPS spectra before and after OER (Fig. S11†) reveals a significant increase in oxygen/hydroxide content on the catalyst surface following stability testing. This observation indicates progressive oxidation of the metallic components, leading to the formation of M–OH and M–OOH species, which are widely recognized as the true active phases governing the OER. Potentiodynamic polarization (PDP) curves serve as an effective means to assess the corrosion resistance of the catalysts. The PDP curves (Fig. 3f) indicate that the HEA and other Ru-based binary alloys exhibit comparable electrochemical corrosion behaviors; however, the HEA's corrosion potential (0.91 V vs. RHE) is greater than that of CoRu (0.87 V vs. RHE), FeRu (0.84 V vs. RHE), MnRu (0.78 V vs. RHE), and NiRu (0.71 V vs. RHE), suggesting that the HEA possesses superior anticorrosive properties, making it less susceptible to corrosion and degradation in the acidic electrolyte, thereby enhancing its catalytic stability.

Fig. 3 (a) OER polarization curves with iR-compensation of CP, PANI-CP, HEA/CP and HEA/PANI-CP; (b) the corresponding Tafel plots of the OER; (c) the Nyquist plots at an overpotential of 50 mV for the OER; the inset shows equivalent circuit models for the OER; (d) C_dl of the corresponding sample; (e) the long-term durability measurement of HEA/PANI-CP for 300 h; (f) corrosion potential analysis of the catalysts.

We employ density functional theory (DFT) calculations to explore the benefits of a high-entropy environment in enhancing the stability of Ru atoms. During the oxygen evolution reaction (OER), Ru catalysts can dissolve under acidic conditions, which involves the migration of surface Ru atoms. Therefore, we assess the atomic migration of Ru within binary and high-entropy alloys to evaluate their stability. Utilizing a face-centered cubic structure for the HEA model, we investigate the migration process of Ru atoms positioned in hexahedral sites (Fig. 4a). As illustrated in Fig. 4b, one Ru atom experiences a transient energy barrier of 1.51 eV on the (111) surface. In comparison to other binary Ru-based catalysts such as RuFe (1.09 eV), RuCo (1.30 eV), RuNi (1.15 eV), and RuMn (0.99 eV), the energy barrier for Ru in the HEA is the highest, indicating that Ru atoms in this environment are more resistant to migration from the surface. The stability of the high-entropy material was further corroborated through an ICP test, which assessed the dissolution of elements from various electrodes after a 24-hour stability test conducted at a current density of 100 mA cm⁻² (Table S2†). The dissolved amount of Ru was related to the evolved oxygen and quantified using a metric known as the stability number (S-number), which allows for a quantitative assessment of catalyst stability. The calculated S-number for the HEA is 18 250, demonstrating a significant enhancement in stability compared to other binary alloys during prolonged operation.

Fig. 4 (a) Migration path of Ru atoms in the HEA; (b) calculated migration energies of Ru atoms on the HEA, FeRu, CoRu, NiRu and MnRu; (c) the evolution of Ru dissolution in the electrolyte during a 24 h electrolysis period at 100 mA cm⁻² and the corresponding evolution of the S-number values of the catalysts.

4. Conclusions

In summary, we employed electrochemical deposition combined with thermal shock to prepare HEA/PANI-CP. This catalyst significantly enhanced OER activity and demonstrated superior long-term durability compared to binary alloy catalysts. Density functional theory (DFT) calculations further substantiate that the high entropy effect is instrumental in providing excellent stability during the OER, as it greatly improves the surface migration energy of Ru atoms by establishing a high entropy environment. This study offers valuable insights into the design of stable high entropy electrocatalysts.

Data availability

The data that support the findings of the present study are available in the ESI† of this article.

Author contributions

Le Su: data curation and wrote the manuscript. Xiaokang Chen: synthesized the sample. Yi Tan: investigation and methodology. Wei-Qiao Deng: conceptualization and project administration. Hao Wu: resources, supervision, and revised the manuscript. Xiangyang Miao: formal analysis.

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

This work was supported by the National Key R&D Program of China (no. 2022YFA1503104), the Natural Science Foundation of Shandong Province (no. 2022HWYQ-009), the Taishan Scholars Project (no. tspd20230601), and the Qilu Young Scholars Program of Shandong University. This work was also funded by the Basic Research Program of Jiangsu (no. BK20230243).

References

1. J. Kwon, S. Sun, S. Choi, K. Lee, S. Jo, K. Park, Y. Kim, H. Park, H. Park, J. Jang, H. Han, U. Paik and T. Song, Adv. Mater., 2023, 35, 2300091.
2. Y. Wang, X. Lei, B. Zhang, B. Bai, P. Das, T. Azam, J. Xiao and Z. Wu, Angew. Chem., Int. Ed., 2024, 63, e202316903.
3. N. Thao, K. Kim, J. Ryu, B. An, A. Nayak, J. Jang, K. Na, W. Choi, G. Ali, K. Chae, M. Akbar, K. Chung, H. Cho, J. Park, B. Kim and H. Han, Adv. Sci., 2023, 10, 2207695.
4. D. Chen, H. Zhao, R. Yu, K. Yu, J. Zhu, J. Jiao, X. Mu, J. Yu, J. Wu and S. Mu, Energy Environ. Sci., 2024, 17, 1885–1893.
5. X. Liu, S. Xi, H. Kim, A. Kumar, J. Lee, J. Wang, N. Tran, T. Yang, X. Shao, M. Liang, M. Kim and H. Lee, Nat. Commun., 2021, 12, 5676.
6. W. Zhao, F. Xu, Z. Wang, Z. Pan, Y. Ye, S. Hu, B. Weng and R. Zhu, Small, 2022, 18, 2205495.
7. X. Wang, H. Jang, S. Liu, Z. Li, X. Zhao, Y. Chen, M. Kim, Q. Qin and X. Liu, Adv. Energy Mater., 2023, 13, 2301673.
8. X. Xiang, B. Cheng, B. Zhu, C. Jiang and G. Liang, Chin. J. Catal., 2025, 68, 326–335.
9. H. Cai, H. Yang, S. He, D. Wan, Y. Kong, D. Li, X. Jiang, X. Zhang, Q. Hu and C. He, Angew. Chem., Int. Ed., 2025, e202423765.
10. Y. Wang, H. Meng, R. Yu, J. Hong, Y. Zhang, Z. Xia and Y. Wang, Angew. Chem., Int. Ed., 2025, 64, e202420752.
11. T. Chen, J. Cai, H. Wang, C. Gao, C. Yuan, K. Zhang, Y. Yu, W. Xiao, T. Luo and D. Xia, Energy Environ. Sci., 2025, 18, 853–861.
12. X. Zou, X. Zhao, B. Pang, H. Ma, K. Zeng, S. Zhi and H. Guo, Adv. Mater., 2024, 36, 2412954.
13. M. Xie, Y. Lu, X. Xiao, D. Wu, B. Shao, H. Nian, C. Wu, W. Wang, J. Gu, S. Han, M. Gu and Q. Xu, Adv. Funct. Mater., 2025, 35, 2414537.
14. K. Zeng, J. Zhang, W. Gao, L. Wu, H. Liu, J. Gao, Z. Li, J. Zhou, T. Li, Z. Liang, B. Xu and Y. Yao, Adv. Funct. Mater., 2022, 32, 2204643.
15. N. Jannath and K. Gilliard-AbdulAziz, Matter, 2024, 7, 2750–2752.
16. Y. Yao, Q. Dong, A. Brozena, J. Luo, J. Miao, M. Chi, C. Wang, I. Kevrekidis, Z. Jason Ren, J. Greeley, G. Wang, A. Anapolsky and L. Hu, Science, 2022, 376, 151.
17. W. Shi, H. Liu, Z. Li, C. Li, J. Zhou, Y. Yuan, F. Jiang, K. Fu and Y. Yao, SusMat, 2022, 2, 186–196.
18. R. Guo, L. Yu, Z. Liu, J. Pan, Y. Yao and L. Liu, Nano Res., 2022, 15, 4893–4901.
19. L. Yao, F. Zhang, S. Yang, H. Zhang, Y. Li, C. Yang, H. Yang and Q. Cheng, Adv. Mater., 2024, 36, 2314049.
20. Y. Mei, J. Chen, Q. Wang, Y. Guo, H. Liu, W. Shi, C. Lin, Y. Yuan, Y. Wang, B. Xia and Y. Yao, Sci. Adv., 2024, 10, eadq6758.
21. J. Hao, Z. Zhuang, K. Cao, G. Gao, C. Wang, F. Lai, S. Lu, P. Ma, W. Dong, T. Liu, M. Du and H. Zhu, Nat. Commun., 2022, 13, 2662.
22. S. Yu, Z. Levell, Z. Jiang, X. Zhao and Y. Liu, J. Am. Chem. Soc., 2023, 145, 25352–25356.
23. W. Zhang, B. Dijk, L. Wu, C. Maheu, V. Tudor, J. Hofmann, L. Jiang, D. Hetterscheid and G. Schneider, ACS Catal., 2024, 14, 11065–11075.
24. L. Zhang, N. Zhang, H. Shang, Z. Sun, Z. Wei, J. Wang, Y. Lei, X. Wang, D. Wang, Y. Zhao, Z. Sun, F. Zhang, X. Xiang, B. Zhang and W. Chen, Nat. Commun., 2024, 15, 9440.
25. S. Guo and C. Liu, Prog. Nat. Sci.: Mater. Int., 2011, 21, 433–446.
26. Y. Wang, Y. Zhuang, Y. Hu, F. Kong, G. Yang, O. Rojas and M. He, Nanoscale, 2023, 15, 18033.
27. H. Wang, H. Jiang, C. Gao, Y. Qiu, H. Guo, J. Zhang, M. Amin and J. Lin, Chem. Eng. J., 2024, 501, 157738.
28. H. Ding, D. Liu, X. Liu, L. Zhang and G. Xu, J. Colloid Interface Sci., 2025, 682, 392–402.
29. Y. Long, G. Zhao, L. Yang, Y. Xu and C. Xu, J. Mater. Chem. A, 2024, 12, 33301.
30. H. Li, Q. Quan, H. Dong, Y. Zhang, P. Xie, D. Chen, D. Yin, C. Wong and J. Ho, Adv. Funct. Mater., 2025, 2420810.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta01949d

---