Facile preparation of highly active zirconia-supported gold nanoparticle catalyst

Abstract

Efficient production of hydrogen peroxide (H₂O₂) from oxygen (O₂) is an important research topic in chemistry. Au nanoparticles loaded on n-type semiconducting metal oxides by the deposition precipitation (DP) method (Au/n-MOs) are known to possess catalytic activity for two electron-oxygen reduction reaction. In this study, zirconia-supported Au nanoparticles (NPs) were prepared by the usual DP method (Au/ZrO₂-DP) and a modified DP method with a pre-step of long-term stirring of ZrO₂ NPs in HAuCl₄ solution (Au/ZrO₂-MDP). Regardless of the insulating character of ZrO₂, Au/ZrO₂-MDP stably yields H₂O₂ from O₂ and HCOOH at 25 °C with a turnover frequency (TOF) of 1.0 × 10² h⁻¹ and selectivity of 97%. The TOF is much larger than the values of Au/ZrO₂-DP and various Au/n-MOs. Spectroscopic measurements indicated that the unexpectedly high catalytic activity of Au/ZrO₂-MDP can be induced by effective introduction of O-vacancies to the ZrO₂ surface.

---

Introduction

Metal oxide-supported Au nanoparticles (NPs) smaller than 10 nm are known to exhibit high catalytic activity for various important reactions including CO oxidation, alcohol oxidation, hydrogen evolution, hydrogenation, and C–C coupling, while bulk Au is chemically inactive.^1–4 The catalytic activity strongly depends on the kind of the metal oxide supports as well as the Au particle size. Among them, n-type semiconducting metal oxides such as TiO₂ and CeO₂ are most widely used as the support for Au NPs (Au/n-MOs).^3,4 A positive correlation was previously recognized between the catalytic activity of Au/n-MOs for alcohol oxidation and the conduction band (CB) minimum of n-MOs.⁵ Also, Au/n-MTiO₃ (M = Ca, Sr, Ba) with high CB minimum has recently been reported to show high catalytic activity for H₂O₂ production from O₂ with formic acid (HCOOH) as a proton and electron donor at ambient temperature and pressure (eqn (1)).⁶ HCOOH with hydrogen storage capacity of 4.3 wt% and normal boiling point of 374 K can be a highly promising hydrogen storage and carrier material.^7,8 HCOOH is one of major products from non-edible biomass,⁹ while it has recently been synthesized from H₂ and CO₂ using an iridium catalyst under mild conditions.¹⁰ Thus, the process (eqn (1)) can be regarded as a green and sustainable process for H₂O₂ production.

O₂ + HCOOH → H₂O₂ + CO₂ (1)

On the other hand, insulating ZrO₂-supported Au NPs (Au/ZrO₂) were reported to show catalytic activity for CO oxidation,¹¹ although the activity is much lower than Au/TiO₂.¹² The significant catalytic activity of Au/ZrO₂ can originate from the oxygen vacancy at the Au–ZrO₂ interface.¹³ In usual, catalytically active Au/MOs are conveniently prepared by the deposition precipitation (DP) process consisting of neutralization of HAuCl₄ solution (step 1), adsorption of the resulting [Au(OH)₃Cl]⁻ complex on MO (step 2), and Au NP formation by heating in the air (step 3).¹⁴

While repeating the preparation of Au/ZrO₂ catalysts by the DP method, we have found that a highly active Au/ZrO₂ catalyst can be obtained by stirring the ZrO₂ particle-dispersed HAuCl₄ solution for a long period of time. Here we present the simple modified DP method for the preparation of Au/ZrO₂ (Au/ZrO₂-MDP). As a test reaction, Au/MO-catalyzed production of H₂O₂ from O₂ and HCOOH (eqn (1)) was carried out. Au/ZrO₂-MDP prepared at heating temperature = 400 °C has been shown to exhibit an extraordinarily high catalytic activity for H₂O₂ production.

Results and discussion

Catalyst preparation and characterization

Commercial non-faceted monoclinic ZrO₂ particles (mean particle size = 20 nm, specific surface area = 100 m² g⁻¹ UEP-100, Daiichi Kigenso Kagaku Kogyo Co.) were used as the support of Au NPs. In the normal DP method, just after adding ZrO₂ particles into HAuCl₄ aqueous solution, the suspension was neutralized to pH 7.0, and the suspension was stirred at 80 °C for 2 h. The resulting particles were washed and dried, and calcined 1 h. In the present modified DP method, ZrO₂ particles were added to the HAuCl₄ solution, and then, the suspension was stirred at 25 °C for 24 h. Subsequent application of the same procedures as the normal DP method yielded Au/ZrO₂. The samples prepared by the normal and modified DP methods at heating temperature t_c (°C) are designated as Au/ZrO₂-DP_tc and Au/ZrO₂-MDP_tc, respectively. The Au loading amounts of Au/ZrO₂-DP and Au/ZrO₂-MDP are quantified to be 0.69 mass% and 0.67 mass%, respectively, by inductively coupled plasma spectroscopy. In the XRD patterns of ZrO₂, Au/ZrO₂-DP₄₀₀, and Au/ZrO₂-MDP₄₀₀ (Fig. S1†), the diffraction peaks are observed at 2θ = 28.2°, 31.5°, and 50.1° due to the diffraction from the (−111), (111), and (−220) planes for the monoclinic ZrO₂ (ICDD No 00-036-0420), respectively. No diffraction of Au is observed for Au/ZrO₂-DP and Au/ZrO₂-MDP because of the small loading amount. Fig. 1a shows the high resolution-transmission electron microscopic (HR-TEM) image of Au/ZrO₂-MDP₄₀₀. The d-spacings of the deposit and support are in agreement with the values of Au(111) and monoclinic ZrO₂(011) planes, respectively. Fig. 1b shows Kubelka–Munk transformed UV-vis-NIR absorption spectra for ZrO₂, Au/ZrO₂-DP₄₀₀, and Au/ZrO₂-MDP₄₀₀. Unmodified ZrO₂ only has the interband-transition absorption at wavelength (λ) < 240 nm. In the absorption spectra of Au/ZrO₂-DP₄₀₀ and Au/ZrO₂-MDP₄₀₀, new absorption appears around 550 nm due to the localized surface plasmon resonance (LSPR) of Au NPs. Fig. 1c and the inset show the TEM image and Au particle size distribution of Au/ZrO₂-MDP₄₀₀, respectively, and the data on Au/ZrO₂-DP₄₀₀ are provided in ESI† (Fig. S2). Au NPs with a mean particle size (d_Au) = 3.6 ± 0.5 nm for Au/ZrO₂-MDP₄₀₀ and d_Au= 4.6 ± 0.7 nm for Au/ZrO₂-DP₄₀₀ are deposited on the ZrO₂ surface. Fig. 1d shows high-angle annular dark field scanning TEM (HAADF-STEM) image of Au/ZrO₂-MDP₄₀₀. It can be seen more clearly that the Au NPs shown by the bright spots are highly dispersed on the ZnO surface. Previously, the formation of noble metal (core)–metal oxides (shell) NPs (M@MOs) such as Ru@SnO₂ was previously reported during a reduction–oxidation process through the strong metal support interaction,¹⁵ but no such particle is observed in the present Au/ZrO₂ systems (Fig. S3†). The similar absorption spectra of Au/ZrO₂-DP₄₀₀ and Au/ZrO₂-MDP₄₀₀ also indicate that the loading amount, size, and shape of Au NPs are comparable each other.

Fig. 1 (a) HR-TEM image of Au/ZrO₂-MDP₄₀₀. (b) Kubelka–Munk-transformed UV-vis-NIR absorption spectra of ZrO₂, Au/ZrO₂-DP₄₀₀, and Au/ZrO₂-MDP₄₀₀. (c) TEM image of Au/ZrO₂-MDP₄₀₀. The inset shows the Au particle size distribution. (d) HAADF-STEM image of Au/ZrO₂-MDP₄₀₀.

Catalytic H₂O₂ production and stability test

The catalytic activities of Au/ZrO₂ and various Au/n-MOs for H₂O₂ production from aerated 1.06 M aqueous solution of HCOOH (eqn (1)) were evaluated at 25 °C. The concentration of H₂O₂ was quantified by the iodometric titration.¹⁶Fig. 2a shows time courses for H₂O₂ generation in the ZrO₂, Au/ZrO₂-DP₄₀₀, Au/ZrO₂-MDP₄₀₀ systems, and the Au/TiO₂ system as a typical Au/n-MO for comparison. While unmodified ZrO₂ is inactive, every Au/MO system shows catalytic activity for the generation of H₂O₂ whose amount increases almost in proportion to reaction time. Fig. 2b shows the rate of H₂O₂ generation (v) in each system. Au/ZrO₂-DP₄₀₀ provides a v value of 0.19 mM h⁻¹ comparable with that of Au/TiO₂ (0.18 mM h⁻¹). The rate of reaction in the Au/ZrO₂-MDP₄₀₀ system (0.80 mM h⁻¹) is 4.2-fold greater than that of the Au/ZrO₂-DP₄₀₀ system. Fig. 2c compares the v values and turnover frequencies (TOFs) of Au/ZrO₂-DP₄₀₀ and Au/ZrO₂-MDP₄₀₀ with those of various Au/n-MOs. The TOF was calculated by assuming that the surface Au atoms are catalytically active sites to compare the values previously reported for various Au/MOs (Table S1†). The surface Au atoms were calculated by the d_Au and the dispersion.¹⁷ Surprisingly, Au/ZrO₂-MDP₄₀₀ exhibits the highest catalytic activity among the Au/MOs, and the TOF reaches 1.0 × 10² h⁻¹. The analysis by gas chromatography confirmed the evolution of CO₂, and HCOOH utilization efficiency (η) defined by eqn (2) reaches 97% (Fig. S4†).

η (%) = (mole number of H₂O₂/mole number CO₂) × 100 (2)

To check the stability of Au/ZrO₂-MDP, the reaction was repeated 5 times using the same Au/ZrO₂-MDP₄₀₀. As shown in Fig. 2d, no decay in the catalytic activity is observed. Further, the characterization by means of XRD, TEM, and XP spectra also confirm that the structure and electronic state of Au/ZrO₂-MDP are maintained after the repeated reactions (Fig. S5†). Clearly, the introduction of the long-time stirring of ZrO₂ particles in the HAuCl₄ solution to the DP process as a pre-step remarkably enhances the catalytic activity of Au/ZrO₂ for two electron-oxygen reduction reaction (2e⁻-ORR).

Fig. 2 (a) Time courses for the H₂O₂ generation in the systems of ZrO₂, Au/ZrO₂-DP₄₀₀, Au/ZrO₂-MDP₄₀₀, and Au/TiO₂ in aerated HCOOH aqueous solution at 25 °C. (b) Rate of H₂O₂ generation (v) in the Au/ZrO₂ systems under the same conditions, and controlled conditions without O₂ or HCOOH. (c) Comparison of the catalytic activities (v and TOF) of Au/ZrO₂-DP₄₀₀ and Au/ZrO₂-MDP₄₀₀ with those of various Au/n-MOs. (d) Repeated reactions in aerated HCOOH aqueous solution at 25 °C using the same Au/ZrO₂-MDP₄₀₀ catalyst.

Activation mechanism

The catalytic activity of Au NPs is well known to strongly depend on the size and shape.^3,4 On the other hand, the LSPR-peak wavelength of Au NPs is sensitive to the size and shape of Au NP. The similar absorption spectra of Au/ZrO₂-DP and Au/ZrO₂-MDP suggest that the size and shape of Au NPs are not the main factors causing the large difference in their catalytic activities.

To gain the insight into the origin for the outstanding catalytic activity of Au/ZrO₂-MDP₄₀₀, firstly, the change in the state of the samples during step 2 to step 3 in the DP and MDP processes were traced by thermal analysis for the Au complex-adsorbed on ZrO₂ (Au³⁺/ZrO₂). Fig. 3a shows thermogravimetry (TG)-differential thermal analysis (DTA) curves for Au³⁺/ZrO₂-DP and Au³⁺/ZrO₂-MDP in the air. In each DTA curve, an endothermic peak due to the desorption of physically adsorbed H₂O is observed around 100 °C with several exothermic peaks in the range from 200 °C to 300 °C. The latter peaks are ascribable to the decomposition of the adsorbed Au³⁺-complex yielding Au NP since the absorption due to the LSPR characteristic of Au NPs appears at t_c ≳ 350 °C (Fig. S6†). Noticeably, an additional exothermic peak is present at 450 °C only in the DTA curve for Au³⁺/ZrO₂-MDP. Then, Au/ZrO₂-MDP samples were prepared at 450 °C and 500 °C, and the catalytic activity for H₂O₂ generation from aerated HCOOH solution was examined. As shown in Fig. 2b, the catalytic activity is greatly reduced at t_c ≥ 450 °C to be comparable with that of Au/ZrO₂-DP₄₀₀.

Fig. 3 (a) TG (broken lines)-DTA (solid lines) curves for Au³⁺/ZrO₂-MDP (red) and Au³⁺/ZrO₂-DP (blue). (b) PL spectra of Au/ZrO₂-MDP_400,450,500 and Au/ZrO₂-DP₄₀₀ measured at l_ex = 280 nm at 77 K. (c) O 1s-XP spectra of Au/ZrO₂-MDP₄₀₀, Au/ZrO₂-DP₄₀₀, and ZrO₂. (d) Au 4f-XPS spectra for Au/ZrO₂-MDP₄₀₀, Au/ZrO₂-DP₄₀₀, and ZrO₂.

Secondly, photoluminescence (PL) spectra of Au/ZrO₂-MDP_400,450,500 and Au/ZrO₂-DP₄₀₀ were measured at excitation wavelength (λ_ex) = 280 nm at 77 K. Every sample has a broad emission band around 480 nm due to the O-vacancies in ZrO₂.^18,19 The emission of Au/ZrO₂-MDP₄₀₀ is significantly stronger than that of Au/ZrO₂-DP₄₀₀. The emission intensity of Au/ZrO₂-MDP steeply decreases at t_c ≥ 450 °C. These results indicate that Au/ZrO₂-DP₄₀₀ has more O-vacancies than the other samples. Thirdly, X-ray photoelectron (XP) spectra of Au/ZrO₂-MDP₄₀₀, Au/ZrO₂-DP₄₀₀, and ZrO₂ were measured. No discernable difference is observed between the Zr3d-XP spectra (Fig. S7†). In Fig. 3c, each O1s-XP spectrum can be deconvoluted into two signals with the binding energies (E_B) of 530. 2 eV (S1) and 532.0 eV (S2) assignable to the lattice O and the lattice oxygen associated with in the oxygen deficient region (or surface OH groups), respectively.²⁰ The S2-signal percentage calculated from {area(S2)/area(S1) + area(S2)} × 100 increases in the order of ZrO₂ (17%) < Au/ZrO₂-DP (19%) < Au/ZrO₂-MDP (23%), which is also consistent with the conclusion drawn from the PL data. On the other hand, bulk Au possesses two signals at E_B = 84.0 eV and 87.7 eV due to the emission from the Au4f_7/2 and Au4f_5/2 orbitals, respectively.²¹ In Fig. 3d, the signals of Au/ZrO₂-DP₄₀₀ and Au/ZrO₂-MDP₄₀₀ shift towards the lower energy side, and the E_B shift in the Au/ZrO₂-MDP₄₀₀ system (ΔE_B, Au4f_7/2 = −0.9 eV) is more pronounced than the Au/ZrO₂-DP₄₀₀ system (ΔE_B, Au4f_7/2 = −0.7 eV). Evidently, O-vacancies are effectively introduced into the ZrO₂ surface by the MDP process simultaneously with the excessive electrons transferred to Au NPs.

Further, the mechanism on the introduction of O-vacancy into the ZrO₂ support of Au/ZrO₂-MDP₄₀₀ was studied by means of diffuse reflectance Fourier-transformed infrared (DRIFT) spectroscopy. Most Au complexes probably exist as [HAu(OH)₃Cl]⁻ at pH ∼7.²² [HAu(OH)₃Cl]⁻ complex is adsorbed on ZrO₂via the ligand exchange between Cl⁻ and surface hydroxy groups (Zr_s–OH).²³Fig. 4 shows the DRIFT spectra for Au³⁺/ZrO₂-MDP before and after calcination at t_c = 400 °C, 450 °C, and 500 °C, and for comparison, ZrO₂ and Au/ZrO₂-DP₄₀₀. While these spectra are similar, a closer inspection indicates the presence of the absorption at 1232 cm⁻¹ and 1160 cm⁻¹ only in the spectra of Au³⁺/ZrO₂-MDP and Au/ZrO₂-MDP₄₀₀. The former and latter signals can be assigned to the adsorbed monodentate carbonate and bidentate carbonate, respectively,^24–26 disappearing at t_c ≥ 450 °C. Thus, the exothermic peak around 450 °C in the Au³⁺/ZrO₂-MDP (Fig. 3a) results from the decomposition of the adsorbed carbonates. Recent temperature programed desorption measurements have shown that the decomposition of carbonate on Au/ZrO₂ occurs at 450 °C with the dehydration of surface hydroxy groups.^27,28

Fig. 4 DRIFT spectra for Au³⁺/ZrO₂ (MDP and DP) before calcined, Au/ZrO₂ (MDP and DP) with the calcination at t_c = 400 °C, 450 °C, and 500 °C, and ZrO₂ for comparison.

In the MDP process (Scheme 1), the long-time stirring of the suspension of ZrO₂ particles in acidic HAuCl₄ solution causes the hydrolysis of the ZrO₂ surface to form Zr_s–OH groups and O-vacancies, where the subscript s denotes the surface atom. During the neutralization, CO₂ is dissolved into the ZrO₂ suspension to be adsorbed on the surface as carbonates with the chemisorption of [Au(OH)₃Cl]⁻ complexes. Upon heating at t_c = 400 °C, the adsorbed Au³⁺ complexes are transformed to Au NPs, while the surface carbonates and oxygen vacancies remain. At t_c ≥ 450 °C, the carbonates are decomposed into CO₂ with the partial disappearance of surface O-vacancies.

Scheme 1 Proposed activation mechanism on the DP method for the preparation of Au/ZrO₂.

These results above strongly suggest that the effective introduction of O-vacancies to the ZrO₂ surface by the MDP process is deeply associated with the high catalytic activity of Au/ZrO₂-MDP for 2e⁻-ORR by HCOOH. Recently, Au/ZrO₂ prepared by a colloidal deposition method has been reported to exhibit catalytic activity for CO oxidation although it is significantly lower than that of Au/TiO₂.¹² The authors have revealed by the temporal analysis of products technique that the CO oxidation on Au/TiO₂ at 80 °C and higher proceeds via the Au-assisted Mars–van Krevelen mechanism where CO is oxidized by the lattice O on the TiO₂ surface to be reproduced by O₂.²⁹ Also, recent first principle calculations have indicated that the Au/ZrO₂-catalyzed CO oxidation can also proceed via the same mechanism.^13,30 Importantly, these studies have pointed out the significance of O-vacancies on the surface of the MO support for the activation of O₂. However, the Au-assisted Mars–van Krevelen mechanism would be excluded from the possible mechanisms in this 2e⁻-ORR because the O–O bond of O₂ is maintained and the reaction temperature is 25 °C.³¹ Meantime, the electron-richness of the MO-supported Au NPs can be one of the controlling factors of the catalytic activity for 2e⁻-ORR.^32,33 Consequently, the outstanding catalytic activity of Au/ZrO₂-MDP₄₀₀ for 2e⁻-ORR may stem from fine tuning of the electronic state of Au NPs on ZrO₂ with the effective introduction of O-vacancies to the support surface (Fig. 3d). However, there is the possibility that oxygen defect sites around the perimeter interface of Au nanoparticles with ZrO₂ work catalytically active sites, and in this case, the TOF values will be even larger (Table S1†). Further work is needed to clarify the catalytically active sites at the atomic level to elucidate the detailed action mechanism.

Conclusions

We have presented a modified DP method with a pre-step of long-time stirring of ZrO₂ particles in HAuCl₄ solution for the preparation of Au/ZrO₂ (Au/ZrO₂-MDP). The sample prepared at heating temperature of 400 °C (Au/ZrO₂-MDP₄₀₀) has been found to exhibit an extraordinarily high catalytic activity for 2e⁻-ORR by HCOOH much exceeding those of various Au/n-MOs. Spectroscopic experiments indicated that the catalytic activity of Au/ZrO₂-MDP₄₀₀ can be induced by effective introduction of O-vacancies to the ZrO₂ surface. Historically, the discovery of the catalytic activity of Au NPs and the subsequent remarkable progress in the application to various chemical reactions owe much to the development of the DP methods. Thus, we anticipate that the present modified DP method can contribute to further advance in this field.

Data availability

The data supporting the findings can be found in the article and ESI,† and are available from the authors upon reasonable request.

Author contributions

S. N. and M. T. conducted catalysts synthesis, characterization, and catalytic reaction experiments, and H. T. supervised the work and data analysis.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by JSPS KAKENHI Grant-in-Aid for Scientific Research (C) no. 20K05674 and 21K05236, the Futaba Foundation, Nippon Sheet Glass Foundation for Materials Science and Engineering, Sumitomo Foundation, Takahashi Industrial and Economic Research Foundation, and Kato Foundation for Promotion of Science.

Notes and references

1. Q.-Y. Bi, X.-L. Du, Y.-M. Liu, Y. Cao, H.-Y. He and K.-N. Fan, J. Am. Chem. Soc., 2012, 134, 8926–8933.
2. X. Liu, L. He, Y.-M. Liu and Y. Cao, Acc. Chem. Res., 2014, 47, 793–804.
3. T. Ishida, T. Maruyama, A. Taketoshi and M. Haruta, Chem. Rev., 2020, 120, 464–525.
4. M. Sankar, Q. He, R. V. Engel, M. A. Sainna, A. J. Logsdail, A. Roldan, D. J. Willock, N. Agarwal, C. J. Kiely and G. J. Hutchings, Chem. Rev., 2020, 120, 3890–3938.
5. S. Naya, M. Teranishi, R. Aoki and H. Tada, J. Phys. Chem. C, 2016, 120, 12440–12445.
6. M. Teranishi, S. Naya and H. Tada, J. Phys. Chem. C, 2019, 123, 9831–9837.
7. M. Grasemann and G. Laurenczy, Energy Environ. Sci., 2012, 5, 8171–8181.
8. X. Liu, L. He, Y.-M. Liu and Y. Cao, Acc. Chem. Res., 2014, 47, 793–804.
9. A. Boddien, B. Loges, F. Gaertner, C. Torborg, K. Fumino, H. Junge, R. Ludwig and M. Beller, J. Am. Chem. Soc., 2010, 132, 8924–8934.
10. J. F. Hull, Y. Himeda, W.-H. Wang, B. Hashiguchi, R. Periana, D. J. Szalda, J. T. Muckerman and E. Fujita, Nat. Chem., 2012, 4, 383–388.
11. X. Zhang, H. Wang and B. Xu, J. Phys. Chem. B, 2005, 109, 9678–9683.
12. D. Widmann, Y. Lin, F. Schűth and R. J. Behm, J. Catal., 2010, 276, 292–305.
13. P. Schlexer, A. Ruiz Puigdollers and G. Paccioni, Top. Catal., 2019, 62, 1192–1201.
14. S. Tsubota, M. Haruta, T. Kobayashi, A. Ueda and Y. Nakahara, Preparation of Catalysis V, Elsevier, Amsterdam, 1991.
15. T. Mitsui, K. Tsutsui, T. Matsui, R. Kikuchi and K. Eguchi, Appl. Catal., B, 2008, 81, 56–63.
16. R. Cai, Y. Kubota and A. Fujishima, J. Catal., 2003, 219, 214–218.
17. G. A. Somorjai, Introduction to Surface Chemistry and Catalysis, John Wiley & Sons, New York, 1994.
18. T. V. Perevalov, D. V. Gulyaev, V. S. Aliev, K. S. Zhuravlev, V. A. Gritsenko and A. P. Yelisseyev, J. Appl. Phys., 2014, 116, 244109.
19. C. Lin, C. Zhang and J. Lin, J. Phys. Chem. C, 2007, 111, 3300–3307.
20. S. R. Teeparthi, E. W. Awin and R. Kumar, Sci. Rep., 2018, 8, 5541.
21. K. Tanaka, S. Tanuma, K. Dohmae, Y. Nagoshi and A. A. Nisawa, X-Ray Photoelectron Spectroscopy, ed. T. Sawada, S. Tanuma and K. Tanaka, Maruzen, Tokyo, 1998.
22. F. Moreau, G. C. Bond and A. O. Taylor, J. Catal., 2005, 231, 105–114.
23. G. C. Bond, C. Loius and D. T. Thompson, Catalysis by Gold, Imperial Collage Press, London, 2006.
24. K. Pokrovski, K. T. Jung and A. T. Bell, Langmuir, 2001, 17, 4297–4303.
25. J. Li, J. Chen, W. Song, J. Liu and W. Shen, Appl. Catal., A, 2008, 334, 321–329.
26. H. Takano, Y. Kirihata, K. Izumiya, N. Kumagai, H. Habazaki and K. Hashimoto, Appl. Surf. Sci., 2016, 388, 653–663.
27. X. Zhang, H. Shi and B.-Q. Xu, J. Catal., 2011, 279, 75–87.
28. Q.-Y. Bi, J.-D. Lin, Y.-M. Liu, H.-Y. He and F.-Q. Huang, J. Power Sources, 2016, 327, 463–471.
29. D. Widmann and R. J. Behm, Angew. Chem., Int. Ed., 2011, 50, 10241–10245.
30. A. R. Puigdollers, P. Schlexer, S. Tosoni and G. Pacchioni, ACS Catal., 2017, 7, 6493–6513.
31. D. K. Widmann, A. Krautsieder, P. Walther, A. Brűckner and R. J. Behm, ACS Catal., 2016, 6, 5005–5011.
32. M. Teranishi, S. Naya and H. Tada, J. Am. Chem. Soc., 2010, 132, 7850–7851.
33. H. Kobayashi, M. Teranishi, R. Negishi, S. Naya and H. Tada, J. Phys. Chem. Lett., 2016, 7, 5002–5007.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3cy00689a

---