Rapid mass production of iron nickel oxalate nanorods for efficient oxygen evolution reaction catalysis

Abstract

The NiFe layered-double-hydroxide (NiFe-LDH) and the NiFe metal–organic framework (NiFe-MOF) demonstrate the best catalytic activity among NiFe-based materials for the oxygen evolution reaction (OER), which is important for efficient hydrogen production. However, the preparation processes of these materials are usually cumbersome and have a low yield, which eliminates their most critical advantage of being low cost. We propose a method for rapidly and efficiently preparing porous (Ni_0.5Fe_0.5)C₂O₄ nanorods with an excellent OER catalytic performance. The overpotential of (Ni_0.5Fe_0.5)C₂O₄ is 266 mV at 20 mA cm⁻² under alkaline conditions, and the Tafel slope is 54.39 mV dec⁻¹. Furthermore, analysis of the changes in the surface properties of the material before and after catalysis determined that the real active material is (Ni_0.5Fe_0.5)(OH)_x(C₂O₄)_1−x. Using a simple scaled-up experiment, (Ni_0.5Fe_0.5)C₂O₄ is mass-produced (40 g) via direct synthesis in 5 min. The composition and performance of the mass-produced sample are analysed under the same conditions, and (Ni_0.5Fe_0.5)C₂O₄ still has a good catalytic performance and its composition has not changed. The efficient synthesis of (Ni_0.5Fe_0.5)C₂O₄ nanorods with a porous structure provides a new option for the development of commercial catalysts using non-precious metals.

---

1. Introduction

Hydrogen is supposed to be a potential replacement for traditional energy sources.^1,2 However, 95% of the hydrogen industrially produced originates from fossil fuels, and the dependence on traditional energy sources has not been resolved.^3,4 Currently, the cleanest method for producing hydrogen is via the electrolysis of water.^5–7 The electrolysis of water comprises the oxygen evolution reaction (OER) and the hydrogen evolution reaction (HER).^8,9 The OER involves four electrons and has a slow reaction rate, which slows down the overall kinetics of the electrolysis of water.^10,11 The best catalysts for the OER are currently noble metal-based oxides such as RuO₂ and IrO₂.^12,13 However, the high cost of the required precious metals limits the large-scale application of these catalysts.¹⁴ Thus, the development of a low-cost and efficient OER catalyst is a crucial task.

An enormous amount of research in the field of OER catalysis has focused on NiFe-based materials owing to their low cost and excellent catalytic activity in alkaline environments.^15–19 Among them, NiFe layered-double-hydroxide (NiFe-LDH) has been intensively studied for more than 30 years, and it represents a class of promising electrocatalysts for the OER.^20–23 However, many studies on NiFe-LDH catalysts still use traditional preparation methods, such as solvothermal and electrodeposition methods.²⁴ In recent years, many studies have started considering the NiFe metal–organic framework (NiFe-MOF) for OER catalysis.^25,26 Unfortunately, most NiFe-MOF catalysts are produced under high-temperature and -pressure conditions with a low yield.^27–29 Therefore, developing a simple, efficient and scalable preparation process for NiFe-based OER catalysts is crucial.

We developed a rapid and simple method for synthesising a new type of mixed NiFe oxalate nanorods (NRs). Experiments were performed to demonstrate the OER catalytic activity of the synthesised nanorods and the effectiveness of the proposed method even when increased to the mass-production scale This study presents a theoretical and experimental basis for effectively solving the bottleneck problem of the low synthesis efficiency for NiFe-based OER catalysts.

2. Experiment

2.1 Reagents

The chemical reagents used in the experiment were not further purified, and included N,N-dimethylacetamide (DMAC) (analytical reagent), oxalic acid (analytical reagent), ferrous chloride tetrahydrate (FeCl₂·4H₂O) (analytical reagent), nickel chloride hexahydrate (NiCl₂·6H₂O) (analytical reagent) and ruthenium oxide (RuO₂) (Macklin). The water used in the experiment was deionized water.

2.2 Synthesis of FeC₂O₄ (NiC₂O₄) NRs

Oxalic acid (10 mmol) was dissolved in DMAC as a solution A. FeCl₂·4H₂O (NiCl₂·6H₂O) (2 mmol) was dissolved in 25 mL deionized water as solution B. Solution B was slowly added to solution A with stirring for 5 min. The above solution was centrifuged with ethanol three times. The precipitate was dried in an oven at 60 °C and ground to obtain FeC₂O₄ (NiC₂O₄) nanorods (NRs).

2.3 Synthesis of (Fe_0.5Ni_0.5)C₂O₄ NRs

Oxalic acid (10 mmol) was dissolved in DMAC as solution A. FeCl₂·4H₂O (1 mmol) and NiCl₂·6H₂O (1 mmol) were dissolved in 25 mL deionized water as solution B. Solution B was slowly added to solution A with stirring for 5 min. The above solution was centrifuged with ethanol three times. The precipitate was dried in an oven at 60 °C and ground to obtain (Fe_0.5Ni_0.5)C₂O₄ NRs.

2.4 Synthesis of G-(Ni_0.5Fe_0.5)C₂O₄

Take 1 g of the prepared NiC₂O₄ and 1 g of the prepared FeC₂O₄ into a clean agate mortar, grind for 10 minutes, and name the sample G-(Ni_0.5Fe_0.5)C₂O₄.

2.5 Synthesis of 40-(Ni_0.5Fe_0.5)C₂O₄

Oxalic acid (0.2 mol) was dissolved in DMAC as solution A. FeCl₂·4H₂O (20 mmol) and NiCl₂·6H₂O (20 mmol) were dissolved in 0.5 L deionized water as solution B. Solution B was slowly added to solution A with stirring for 5 min. The above solution was centrifuged with ethanol three times. The precipitate was dried in an oven at 60 °C and ground to obtain 40-(Fe_0.5Ni_0.5)C₂O₄ NRs.

2.6 Electrochemical test

The catalytic activity of oxalate was studied using a CH Instruments Electrochemical Analyzer (CHI 760E). The classic three-electrode system was adopted in which a silver chloride electrode, a platinum wire and a foamed nickel (NF 2 mm × 2 mm see supporting documents for processing) were used as the reference electrode, the auxiliary electrode and the working electrode, respectively. The overpotential calculation formula was as follows: E_RHE = E_Ag/AgCl + 0.197 + 0.0592pH. Linear sweep voltammograms (LSVs) were measured in 1 mol L⁻¹ KOH saturated oxygen solution at a sweep speed of 5 mV s⁻¹. The LSV measurements were iR compensated at 95%. Cyclic voltammetry (CV) was used to study the electrochemically active surface area (ECSA), and was carried out at different sweep speeds (20, 40, 60, 80, 100 and 120 mV s⁻¹) when the voltage was 1.223–1.323 V (vs. RHE) in 1 mol L⁻¹ KOH saturated oxygen solution. Electrochemical impedance spectroscopy (EIS) was recorded under 1. 6 V (vs. RHE) from 0.1 Hz to 100 kHz at the amplitude of the sinusoidal voltage of 5 mV. See supporting documents for all other detection methods.

3. Results and discussion

X-Ray diffraction (XRD) was used to analyse the composition of the oxalate. As shown in Fig. 1a, the diffraction peaks at 18.4°, 23.0° and 34.2° of the prepared FeC₂O₄ NRs corresponded well to the crystal plane facets (110), (−202) and (−121) of FeC₂O₄·H₂O (JCPDS Card, No. 23-0293). Compared with the standard spectrum of FeC₂O₄·H₂O, the diffraction peaks of the prepared NiC₂O₄ NRs were shifted to the right. This was attributed to the differences in the radii of Fe and Ni. Furthermore, the diffraction peaks of the (Fe_0.5Ni_0.5)C₂O₄ NRs at 18.5°, 29.5° and 34.8° were clearly located between the diffraction peaks of the FeC₂O₄ and NiC₂O₄ NRs. This result confirms that Fe and Ni existed in a certain proportion in the prepared binary metal oxalate NRs.

Fig. 1 (a) XRD patterns of NiC₂O₄, FeC₂O₄ and (Fe_0.5Ni_0.5)C₂O₄. (b) SEM image of (Fe_0.5Ni_0.5)C₂O₄. (c) TEM image of (Fe_0.5Ni_0.5)C₂O₄. (d–g) Dynamic map and EELS mapping of Fe, Ni, C and O, respectively, in (Fe_0.5Ni_0.5)C₂O₄.

Scanning electron microscopy (SEM) was used to analyse the morphology of the prepared samples. As shown in Fig. S2a, b (ESI†) and Fig. 1b, the prepared NiC₂O₄, FeC₂O₄ and (Fe_0.5Ni_0.5)C₂O₄ samples all exhibited a rod-like morphology. This was attributed to the small ionisation energy of metal ions in DMAC, which made the oxalate grow laterally.³⁰ However, the samples had fairly different sizes. The length of the prepared NiC₂O₄ NRs was 100 nm, which is relatively short. Conversely, the length of the prepared FeC₂O₄ NRs was 5 μm. The prepared (Fe_0.5Ni_0.5)C₂O₄ NRs had a length between 5 and 100 μm. The microstructure and composition of the (Fe_0.5Ni_0.5)C₂O₄ NRs were studied via high-magnification transmission electron microscopy (TEM). Fig. S2d–f (ESI†) clearly show many pores on the surface of the (Fe_0.5Ni_0.5)C₂O₄ NRs. In addition, lattice fringes are clearly visible in Fig. 1c. The measured lattice spacings were 0.286 and 0.312 nm, which are similar to the lattice spacings of the FeC₂O₄ (202) and (112) crystal planes. The elemental distribution of the (Fe_0.5Ni_0.5)C₂O₄ NRs was confirmed through elemental mapping from energy-dispersive X-ray spectroscopy. As shown in Fig. 1d–g, Fe, Ni, C and O were uniformly distributed in the NRs. FeC₂O₄, NiC₂O₄ and (Fe_0.5Ni_0.5)C₂O₄ were analysed via their nitrogen adsorption method–desorption isotherm plots. From Fig. 2a–c, it can be clearly seen that the Brunauer–Emmett–Teller surface area and the pore volume of FeC₂O₄, NiC₂O₄ and (Fe_0.5Ni_0.5)C₂O₄ were 6.6581 m² g⁻¹, 63.0562 m² g⁻¹ and 15.2492 m² g⁻¹, and 0.015696 cm³ g⁻¹, 0.193547 cm³ g⁻¹ and 0.067713 cm³ g⁻¹, respectively. These results reveal that NiC₂O₄ has the largest specific surface area, which is consistent with the conclusion obtained by SEM. This, in addition with the above XRD and TEM results, confirms that the (Fe_0.5Ni_0.5)C₂O₄ NRs exist as a single pure phase rather than two separate phases.

Fig. 2 (a–c) Adsorption–desorption isotherms of the FeC₂O₄, NiC₂O₄ and (Fe_0.5Ni_0.5)C₂O₄ samples, respectively.

The catalytic activity of the sample was explored using the classic three-electrode system in 1 M KOH as the oxygen-saturated electrolyte. The linear sweep voltammetry (LSV) curves in Fig. 3a and Fig. S3 (ESI†) show that, at 50 mA cm⁻², the overpotentials of (Fe_0.75Ni_0.25)C₂O₄, (Fe_0.25Ni_0.75)C₂O₄ (see ESI† for the preparation process), (Fe_0.5Ni_0.5)C₂O₄, NiC₂O₄, FeC₂O₄, and the blank nickel foam (NF) were 329, 315, 284, 365, 312, and 382 mV, respectively. Results when the ratio of Fe : Ni was 1 : 1 showed that the prepared samples had the best OER catalytic activity. (Fe_0.5Ni_0.5)C₂O₄, NiC₂O₄, FeC₂O₄ and NF had Tafel slopes of 54.39, 71.61, 61.42, and 71.36 mV dec⁻¹, as presented in Fig. 3b. The (Fe_0.5Ni_0.5)C₂O₄ NRs had the smallest Tafel slope, which indicates that they can obtain a higher catalytic current density at the same overpotential. The electrochemically active surface area (ECSA) was determined via cyclic voltammetry (CV), as shown in Fig. 3c and Fig. S6a, b (ESI†). Fig. 3d shows that the double-layer capacitance values of (Fe_0.5Ni_0.5)C₂O₄, NiC₂O₄ and FeC₂O₄ were 13.88, 23.3 and 13.69 mF cm⁻², respectively. The NiC₂O₄ NRs had the largest ECSA, because they had the smallest size. However, this result is inconsistent with the LSV curve, which indicates that the ECSA is not the primary cause for the good OER catalytic activity of (Fe_0.5Ni_0.5)C₂O₄. Electrochemical impedance spectroscopy (EIS) was used to measure the charge-transfer process, and the fitted data are shown in Fig. 3e. (Fe_0.5Ni_0.5)C₂O₄ had a much smaller radius than NiC₂O₄, which indicates that the (Fe_0.5Ni_0.5)C₂O₄ NRs had the least resistance to charge transfer. Therefore, promotion of the charge transfer is the main factor for the improved reaction kinetics of the OER. Stability is essential for catalyst materials. Thus, CV measurements were repeated to compare the catalytic activity after different cycles in 1 M KOH. As shown in Fig. 3f, minor attenuation occurred for (Ni_0.5Fe_0.5)C₂O₄ after 1000 cycles. The potential changes after 2000 and 3000 cycles were negligible. These results illustrate that the catalyst exhibited good stability after a long period of time without any significant change in current.

Fig. 3 (a) LSV polarisation curves and (b) Tafel plots for the OER of NiC₂O₄, FeC₂O₄, (Fe_0.5Ni_0.5)C₂O₄ and blank NF with 95% iR compensation. (c) CV curves of (Fe_0.5Ni_0.5)C₂O₄ in 1 M KOH solution with different scan rates. (d) Relationship between the difference between the anode and cathode current densities and the scanning rate for all catalysts. (e) Nyquist diagram obtained via EIS at 1.6 V vs. RHE for NiC₂O₄, FeC₂O₄ and (Fe_0.5Ni_0.5)C₂O₄ electrocatalysts (inset: equivalent resistance circuit model). (f) LSV curves of (Fe_0.5Ni_0.5)C₂O₄ before and after 1000, 2000 and 3000 cycles in 0.1 M KOH solution.

To clarify the catalytic mechanism and specific active substances, we used XPS to analyse the chemical composition and state of the (Ni_0.5Fe_0.5)C₂O₄ sample before and after OER catalytic process. The XPS spectrum of C1s (Fig. 4a) indicated that the peak intensity of C₂O₄²⁻ (288.7 eV) in the sample greatly weakened after catalysis but did not disappear completely.^32,33 The XPS spectrum of O1s (Fig. 4b) showed only one O1s peak belonging to C₂O₄²⁻ (532.1 eV) before the OER measurements. However, after the OER measurements, the C₂O₄²⁻ peak was obviously weakened and the strongest peak belonged to O–H (531.1 eV).^31–33 The Fe2p spectrum (Fig. 4c) showed that the main peak of Fe2p_3/2 shifted by 1.2 eV after the catalytic process. Before the OER measurements, the peaks at 710.4 and 714.8 eV corresponded to Fe2p_3/2. After the OER measurements, the peak at 711.6 eV and 716.6 eV corresponded to Fe2p_3/2 of Fe(OH)₂ and the peak  724.9 eV and 733.07 eV corresponded to Fe2p_1/2 of Fe(OH)₂^34,35 The spectrum of Ni2p (Fig. 4d) showed a prepeak (853.7 eV) and a main peak (856.7 eV) of Ni²⁺ in Ni2p_3/2 before the OER measurements. After the OER measurements, the main peak of Ni2p shifted to the right by 1.1 eV. The peaks at 855.6 eV and 861.6 eV corresponded to Ni2p_3/2 of Ni(OH)₂. The peaks at 873.3 eV and 880.2 eV corresponded to Ni2p_1/2 of Ni(OH)₂.^36–38 The prepeak disappeared because Ni changed from oxalate to hydroxide. The above results indicate that most of the (Ni_0.5Fe_0.5)C₂O₄ NRs immersed in alkaline solution were transformed into the corresponding hydroxide. To further study the phase transformation process, the G-(Ni_0.5Fe_0.5)C₂O₄ catalytic performance and phase transition before and after catalysis were studied. The LSV curve of G-(Ni_0.5Fe_0.5)C₂O₄ is shown in Fig. S7a (ESI†). The overpotential was 291 mV at 50 mA cm⁻², and the Tafel slope was 59.62 mV dec⁻¹. This is comparable to the catalytic performance of the prepared (Ni_0.5Fe_0.5)C₂O₄. Significantly, the XPS spectrum of G-(Ni_0.5Fe_0.5)C₂O₄ after the OER measurements (Fig. S4, ESI†) was completely consistent with that of the prepared (Ni_0.5Fe_0.5)C₂O₄. Furthermore, the results indicated that the real active substance was (Ni_0.5Fe_0.5)(OH)_x(C₂O₄)_1−x.

Fig. 4 (a) C_1s, (b) O_1s, (c) Fe_2p and (d) Ni_2p XPS spectra of (Ni_0.5Fe_0.5)C₂O₄ before and after the OER measurements. (e) Schematic diagrams of the structure of (Ni_0.5Fe_0.5)C₂O₄ changing to (Ni_0.5Fe_0.5)(OH)_x(C₂O₄)_1−x.

Therefore, we believe that the surface of the (Ni_0.5Fe_0.5)C₂O₄ NRs first dissolves in the KOH solution and is then deposited to form the active substance of (Ni_0.5Fe_0.5)(OH)_x(C₂O₄)_1−x. The structural transformation is shown in Fig. 4e. The reaction formulae are as follows:

½NiC₂O₄ + ½FeC₂O₄ + xOH + 2xe⁻ → (Ni_0.5Fe_0.5)(OH)_x(C₂O₄)_1−x + xC₂O₄²⁻ (Ni_0.5Fe_0.5)C₂O₄ + xOH + 2xe⁻ → (Ni_0.5Fe_0.5)(OH)_x(C₂O₄)_1−x + xC₂O₄²⁻

As shown in Fig. S8 (ESI†), we conducted a mass production experiment and obtained 40 g of (Ni_0.5Fe_0.5)C₂O₄. The composition, morphology and catalytic activity were analysed, and the XRD, SEM and XPS results are shown in Fig. S1a, S2c and S5 (ESI†), respectively. The mass production experiment had no effect on the morphology, composition and surface properties of the catalyst. Fig. S6c–e and S7a–c (ESI†) show the LSV curve and EIS measurement results. The overpotential of the sample was 278 mV at 50 mA cm⁻², and the Tafel slope was 51.92 mV dec⁻¹. As shown in Fig. S5d and e (ESI†), the double-layer capacitance was 12.55 mF cm⁻². The catalytic performance also did not change significantly. Therefore, the proposed method is suitable for the rapid mass production of (Ni_0.5Fe_0.5)C₂O₄ as an efficient OER catalyst. In addition, as shown in Fig. S8 (ESI†), we compared the OER catalytic activity of (Ni_0.5Fe_0.5)C₂O₄ with the recently reported FeNi-based catalyst. It can be concluded that the preparation method of (Ni_0.5Fe_0.5)C₂O₄ is not only simple, but also has the same performance as other FeNi-based catalysts.^39–43

4. Conclusion

We developed a simple and rapid co-precipitation process to prepare (Ni_0.5Fe_0.5)(C₂O₄) with a uniform morphology. Electrochemical tests confirmed that the (Ni_0.5Fe_0.5)(C₂O₄) NRs exhibited an excellent OER catalytic performance and stability.^44,45 We analysed the phase transition before and after the catalytic process and confirmed that the real active material is (Ni_0.5Fe_0.5)(C₂O₄)_1−x(OH)_x. We conducted mass production experiments and analysed the morphology, composition and catalytic activity of the products. Results showed that the mass production experiment does not affect the performance of the catalyst. The developed method and catalyst should be helpful for realising the large-scale production and application of NiFe-based OER catalysts.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Educational Commission of Jiangxi Province of China (GJJ180470 and GJJ190431) and PhD Research Startup Funding of Jiangxi University of Science and Technology (jxxjbs18062). Jiangxi Provincial Key Laboratory of Flash Green Development and Recycling (20193BCD40019).

References

1. M. F. Lagadec and A. Grimaud, Nat. Mater., 2020, 19, 1140–1150.
2. K. P. Brooks, S. J. Sprik, D. A. Tamburello and M. J. Thornton, Int. J. Hydrogen Energy, 2020, 45, 24917–24927.
3. M. Kroschel, A. Bonakdarpour, J. T. H. Kwan, P. Strasser and D. P. Wilkinson, Electrochim. Acta, 2019, 317, 722–736.
4. S. E. Hosseini and M. A. Wahid, Renewable Sustainable Energy Rev., 2016, 57, 850–866.
5. F. Dionigi, Z. Zeng, I. Sinev, T. Merzdorf, S. Deshpande, M. B. Lopez, S. Kunze, I. Zegkinoglou, H. Sarodnik and D. Fan, Nat. Commun., 2020, 11, 1–10.
6. M. Zha, C. Pei, Q. Wang, G. Hu and L. Feng, J. Energy Chem., 2020, 47, 166–171.
7. Y. Sun, H. Liao, J. Wang, B. Chen, S. Sun, S. J. H. Xi, S. Ong, C. Diao, Y. Du and J. Wang, Nat. Catal., 2020, 3, 554–563.
8. H. Liu, Z. Yan, X. Chen, J. Li, L. Zhang, F. Liu, G. Fan and F. Cheng, Research, 2020, 2020, 1–11.
9. S. Li, W. Chen, H. Pan, Y. Cao, Z. Jiang, X. Tian, X. Hao, T. Maiyalagan and Z. Jiang, ACS Sustainable Chem. Eng., 2019, 7, 8530–8541.
10. L. Wang, D. Kong, T. Xin, X. Shu, K. Zheng, L. Xiao, X. Sha, Y. Lu, Z. Zhang and X. Han, Appl. Phys. Lett., 2016, 108, 151903.
11. X. Zhu, T. Jin, C. Tian, C. Lu, X. Liu, M. Zeng, X. Zhuang, S. Yang, L. He and H. Liu, Adv. Mater., 2017, 29, 1704091.
12. X. Liu, F. Liu, J. Yu, G. Xiong, L. Zhao, Y. Sang, S. Zuo, J. Zhang, H. Liu and W. Zhou, Adv. Sci., 2020, 7, 2001526.
13. Y. Qu, B. Chen, Z. Li, X. Duan, L. Wang, Y. Lin, T. Yuan, F. Zhou, Y. Hu and Z. Yang, J. Am. Chem. Soc., 2019, 141, 4505–4509.
14. S. Ghosh and R. N. Basu, Nanoscale, 2018, 10, 11241–11280.
15. H. Liu, J. Guan, S. Yang, Y. Yu, R. Shao, Z. Zhang, M. Dou, F. Wang and Q. Xu, Adv. Mater., 2020, 32, 2003649.
16. M. L. Lindstrom, R. Gakhar, K. Raja and D. Chidambaram, J. Electrochem. Soc., 2020, 167, 46507.
17. L. Wu, G. Wan, N. Hu, Z. He, S. Shi, Y. Suo, K. Wang, X. Xu, Y. Tang and G. Wang, Nanomaterials, 2018, 8, 451.
18. S. Zhao, Y. Wang, J. Dong, C. He, H. Yin, P. An, K. Zhao, X. Zhang, C. Gao and L. Zhang, Nat. Energy, 2016, 1, 1–10.
19. M. Luo, C. Zhao, C. Wang, Y. Bi, Q. Li, Y. Hao, L. Li, Y. Kuang, Y. Li and X. Lei, Nano Res., 2017, 10, 1732–1739.
20. Z. Cai, X. Bu, P. Wang, W. Su, R. Wei, J. C. Ho, J. Yang and X. Wang, J. Mater. Chem. A, 2019, 7, 21722–21729.
21. W. Zhang, J. Qi, K. Liu and R. Cao, Adv. Energy Mater., 2016, 6, 1502489.
22. S. Dutta, A. Indra, Y. Feng, T. Song and U. Paik, ACS Appl. Mater. Interfaces, 2017, 9, 33766–33774.
23. Y. Wang, S. Tao, H. Lin, G. Wang, K. Zhao, R. Cai, K. Tao, C. Zhang, M. Sun and J. Hu, Nano Energy, 2021, 81, 105606.
24. H. H. Sun, A. Dolocan, J. A. Weeks, A. Heller and C. B. Mullins, ACS Nano, 2020, 14, 17142–17150.
25. S. Xu, J. Du, J. Li, L. Sun and F. Li, J. Mater. Chem. A, 2020, 8, 16908–16912.
26. J. Du, S. Xu, L. Sun and F. Li, Chem. Commun., 2019, 55, 14773–14776.
27. H. Jia, Y. Yao, J. Zhao, Y. Gao, Z. Luo and P. Du, J. Mater. Chem. A, 2018, 6, 1188–1195.
28. L. Gan, J. Fang, K. Zhang, Y. Lai and J. Li, Int. J. Hydrog. Energy, 2019, 44, 2841–2847.
29. H. Sun, Y. Lian, C. Yang, L. Xiong, P. Qi, Q. Mu, X. Zhao, J. Guo, Z. Deng and Y. Peng, Energy Environ. Sci., 2018, 11, 2363–2371.
30. W. A. Ang, N. Gupta, R. Prasanth and S. Madhavi, ACS Appl. Mater. Interfaces, 2012, 4, 7011–7019.
31. X. Gao, D. Chen, J. Qi, F. Li, Y. Song, W. Zhang and R. Cao, Small, 2019, 15, 1904579.
32. A. M. Salvi, F. Langerame, A. E. Pace, M. E. E. Carbone and R. Ciriello, Surf. Sci. Spectra, 2015, 22, 21–31.
33. V. Hasija, P. Raizada, A. Hosseini-Bandegharaei, P. Singh and V. Nguyen, Top. Catal., 2020, 63, 1030–1045.
34. B. M. Hunter, W. Hieringer, J. R. Winkler, H. B. Gray and A. M. Müller, Energy Environ. Sci., 2016, 9, 1734–1743.
35. Z. Zhang, B. He, L. Chen, H. Wang, R. Wang, L. Zhao and Y. Gong, ACS Appl. Mater. Interfaces, 2018, 10, 38032–38041.
36. W. Zhang, J. Qi, K. Liu and R. Cao, Adv. Energy Mater., 2016, 6, 1502489.
37. Y. Zhou and N. López, ACS Catal., 2020, 10, 6254–6261.
38. M. C. Biesinger, B. P. Payne, L. W. M. Lau, A. Gerson and R. S. C. Smart, Surf. Interface Anal., 2009, 41, 324–332.
39. J. W. Zhang, H. Zhang, T. Z. Ren, Z. Y. Yuan and T. J. Bandosz, Front. Chem. Sci. Eng., 2020, 15, 3.
40. X. Xie, C. Cao, W. Wei, S. Zhou and Q. L. Zhu, Nanoscale, 2020, 12, 10.
41. L. Cao, Z. Li, K. Su, M. Zhang and B. Cheng, J. Energy Chem., 2020, 54, 143.
42. X. Gu, Z. Liu, M. Li, J. Tian and L. Feng, Appl. Catal., B, 2021, 297, 120462.
43. Y. Liu, G. Jia, Q. Wu, D. Zhang and X. Cui, Mater. Lett., 2021, 291, 129564.
44. X. Jia, Y. Zhao, G. Chen, L. Shang, R. Shi, X. Kang, G. I. N. Waterhouse, L. Z. Wu, C. H. Tung and T. Zhang, Adv. Energy Mater., 2016, 6, 1502585.
45. Q. Wang, L. Shang, R. Shi, X. Zhang, Y. Zhao, G. I. N. Waterhouse, L. Z. Wu, C. H. Tung and T. Zhang, Adv. Energy Mater., 2017, 7, 1700461.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1nj04668c

‡ These two authors are contributed equally to this work.

---