Electron transportation path build for superior photoelectrochemical performance of Ag₃PO₄/TiO₂

Abstract

TiO₂ is an attractive photoanode material with its large band gap, whilst its performance largely suffers from low efficiency on both charge separation and solar conversion. Herein, a self-organized TiO₂ nanotube arrays (TNT) is prepared by anodized Ti foil in ethylene glycol electrolyte to ameliorate charge transmission ability. Ag₃PO₄ is further synthesized on TNT substrate by dipping method. HRTEM images results indicate Ag₃PO₄ nanoparticles are successfully deposited on the surfaces of TNT. Photoelectrochemical tests show the Ag₃PO₄/TiO₂ heterojunction has a higher photocurrent density of 2.34 mA cm⁻² at 0 V than that of pure TNT (0.38 mA cm⁻²). This is attributed to an Ag “pump” reduced on the interface of Ag₃PO₄/TiO₂, therefore electron transportation path is built between Ag₃PO₄ and TiO₂ leading to photogenerated electrons and holes effective separation. This high photocurrent density array films facilitates it a desirable photoelectrochemical material for water splitting.

---

1 Introduction

Photoelectrochemica (PEC) water splitting is recognized as one of the most promising strategies in solar energy conversion. Titanium dioxide (TiO₂) has been specifically selected out among various PEC materials since Honda–Fujishima first found the ability of TiO₂ for photo-splitting water under ultraviolet light in 1972.¹ In 1976, Carey found that PCBs in TiO₂ suspension solution successfully dechlorinated under UV irradiation.² In 1991, Graztel successfully developed a dye-sensitized solar cell by using TiO₂ nanoparticles.³ But TiO₂ suffers from its poor charge separation and high charge transmission resistance.^4,5 One dimensional (1-D) nanostructure allows a short diffusion length for holes in radial direction, whereas the long axial direction of the structure becomes the preferred electron channel that provides enough length of light attenuation as well.^6,7 Besides these above, 1-D nanostructures also provide excellent photovoltaic, photocatalytic, and PEC properties relative to random-shaped particles.^8–12 Therefore, the 1-D self-organized TiO₂ nanotube array (TNT) is expected to design. In 2001, Grimes and co-workers successfully prepared TNT by anodization in hydrofluoric acid electrolyte.¹³ Further, Macak synthesized TNT by anodizing in glycerol electrolyte containing 0.5% NH₄F, the morphology more smooth than that obtained in aqueous electrolyte, and the length of nanotube reached above 10 μm.¹⁴ Also, TNT exhibits more attractive PEC performance and larger specific surface area than TiO₂ nanoparticles or nanosheets.^15,16 However, band gap of TiO₂ at 3.0–3.2 eV can only absorb ultraviolet light corresponding for 5% of sunlight, which means the solar utilization efficiency is less than satisfactory.

Ag₃PO₄ a n-type semiconductor can absorb visible light wavelengths less than 520 nm (indirect band gap of 2.4 eV), and the quantum yield reaches more than 90%, which aroused attention of many researchers;^17–19 but its conduction band (CB) is at 0.45 V vs. NHE, so Ag⁺ of Ag₃PO₄ will be easily reduced to elemental Ag under visible light irradiation.^10,20–26 Wang et al. studied Ag₃PO₄/AgBr/Ag degradation of organic matter, due to better stability of AgBr and plasma effect of Ag nanoparticles, which improved Ag₃PO₄ stability and photocatalytic capacity.²⁷ Teng found that Ag/Ag₃PO₄/TiO₂ can photodegrade chlorophenol efficiently, and Ag nanoparticles can effectively prevent the occurrence of Ag₃PO₄ light corrosion, and photocurrent density at 0.28 mA cm⁻².²⁸ Xu et al. reported Ag₃PO₄/TiO₂/Fe₃O₄ photodegraded benzene sulfonate and had a great bactericidal effect.²⁹ Photodegradation properties on Ag₃PO₄ or its composites can be found in many research, but seldom on PEC properties, which is closely relative to photo-water-splitting. Moreover, all of works is not enough to achieve high performance because electrons randomly flow after the electron–hole separation resulting in fast electron–hole recombination.

In this study, Ag₃PO₄ nanoparticles were chemically deposited on 1-D TNT. A heterojunction of Ag₃PO₄/TiO₂ was formed and linear sweep voltammetry test showed photocurrent was about 6 times bigger than pure TNT and the band flat potentials of Ag₃PO₄/TiO₂ positively shifted, which meant solar absorption range of TiO₂ was significantly broadened. Ag was also synthesized on the interface between Ag₃PO₄ and TiO₂ due to light illumination, which formed an efficient system for separation of photo-generated charges and improvement of PEC properties.

2 Experimental

2.1 Sample Ag₃PO₄/TiO₂ preparation

Ti foil of 1 cm² was provided with anodization in ethylene glycol electrolyte containing 4 wt% H₂O and 0.25 wt% NH₄F for 1 h under constant potential (50 V) at room temperature.^14,30 The samples were washed with distilled water and dried at 60 °C. The as-prepared TiO₂ was thermally treated at 450 °C for 2 h under atmospheric conditions.

Ag₃PO₄ nanoparticles were deposited into the crystallized TiO₂ nanotubes by sequential chemical bath deposition method. Typically, the sample was successively immersed in four different beakers for 3 minutes in each beaker. One beaker contained 5 mM AgNO₃ aqueous solution, another contained 5 mM NaH₂PO₄, and the other two contained distilled water to rinse the samples from the excess of each precursor solution. Such an immersion cycle was repeated several times, typically between 2 and 8 cycles.

2.2 Characterization

The surface morphology was observed through a field-emission scanning electron microscope (Hitachi S-4800, Japan). Transmission electron microscopy (TEM) images were taken with JEOL JEM-2100 transmission electron microscope at 200 kV. The absorbance was measured with UV-vis spectrophotometer (Hitachi U-3900, Japan). The elemental chemical status was measured by X-ray photoelectron spectroscopy (Thermo ESCALAB 250Xi), equipped with Al Kα radiation. All energies were calibrated to spurious carbon at 284.8 eV.

The PEC properties of samples were investigated by a three-electrode configuration electrochemical workstation (CHI 760E), employing the samples, Ag/AgCl and Pt mesh as working, reference and counter electrode, respectively (Fig. S4†). The supporting electrolyte used was 1 M KOH (pH = 14) aqueous solution. Mott–Schottky plots were measured at 1000 Hz. Electrochemical impedance spectroscopic (EIS) measurements were performed between 10⁵ Hz and 10⁻² Hz dark. The working electrode was illuminated with a 300 W xenon lamp. The photocurrent was recorded concurrently with the light switching on and off without applied voltage.

3 Results and discussion

3.1 Electron transportation analysis

After deposited Ag₃PO₄, a heterojunction of Ag₃PO₄/TiO₂ first formed and promoted separation of the photogenerated carriers.^25,31,32 Additionally, Ag⁺ of Ag₃PO₄ was proved photoreduced and decomposed to weakly active Ag under photo-illumination,^31–35 which could obviously enhance visible light absorbance above 700 nm (Fig. 1c) and accelerate photogenerated carrier separation due to its surface plasma resonance (SPR).^28,36 As reported by many researchers, SPR in Ag can be directly excited under visible light illumination to generate and inject hot carriers into the CB of semiconductors.^37,38 This role is just like a “pump” (Fig. 1a and b), electrons in Ag₃PO₄ are absorbed by Ag and then “pumped” into higher energy level, the CB of TiO₂, due to lower conductivity of Ag₃PO₄ and SPR of Ag which assisted electrons to get over the barrier of Ag₃PO₄/TiO₂ junction. Then electrons transported from the CB of TiO₂ to Ti base along 1-D electron channel of TNT. Equally, holes were injected from the valence band (VB) of TiO₂ into the VB of Ag₃PO₄, and then participated in oxidation reaction. The transfer route of electron can be E_CB(Ag₃PO₄) → E_f(Ag) → E_CB(TiO₂). If without the SPR of Ag, charges transportation route are shown in Fig. S4,† the light induced electrons transported from higher conductor band (−0.612 V vs. NHE) of TiO₂ to lower conductor band (+0.45 V vs. NHE) of Ag₃PO₄, and the holes were accumulated in the valence band of TiO₂, which is inconsistent with experimental results of electron transportation from TiO₂ to Ti foil.

Fig. 1 Schematic diagram of bidirectional photo-induced electron transportation (a), photogenerated electron–hole pairs separation process (b) and UV-vis spectra (c).

3.2 XPS analysis

Ag₃PO₄/TiO₂ sample was examined by X-ray photoelectron spectroscopy (XPS), and the results are shown in Fig. 2. The binding energies of the XPS spectra were calibrated by C 1s (284.8 eV). The carbon peak is due to the adventitious hydrocarbon from the XPS instrument itself (Fig. S2†). In Fig. 2a, the characteristic peak at 458.5 eV and 464.2 eV assigned to Ti⁴⁺ in TiO₂ are depicted.³⁹ Two binds at 367.7 and 373.7 eV, are ascribed to Ag 3d_5/2 and Ag 3d_3/2 bonding energies in Fig. 2b. These bands could be further deconvoluted into two peaks, respectively, at 366.4, 367.3 eV and 372.8, 373.7 eV, where the bands at 366.4 and 372.8 eV are ascribed to the Ag⁺ of Ag₃PO₄, and those at 367.3 and 373.7 eV are attributed to the metallic Ag⁰. As some papers have reported,^40,41 these results verify the existence of metallic Ag⁰ on Ag₃PO₄/TiO₂ photocatalysts after the reaction.

Fig. 2 High resolution XPS spectra of (a) Ti 2p and (b) Ag 3d of Ag₃PO₄/TiO₂.

3.3 Morphology characterization

Diameter of the self-organized pure TNT prepared by anodizing method is about 60 to 90 nm (Fig. 3a and b). Fig. 3b showed SEM images of the sample for 4 cycles deposited Ag₃PO₄ on TNT, and the nanoparticles of Ag₃PO₄ covered the surface with a diameter of about 10 nm. Also, some particles with a diameter of 40 to 50 nm were formed because of severe agglomeration for 8 cycles, shown in Fig. S3.† Illustration of EDS indicated that the atomic proportion of Ag and P is 2.44 to 1, close to stoichiometric ratio of 3 to 1 (Fig. 3c).

Fig. 3 SEM images of pure TiO₂ (a), Ag₃PO₄/TiO₂ 4 cycles (b), insert is partial enlarged view of (a) and (b), EDS of 4 cycles (c).

TEM images of the sample for 4 cycles deposited Ag₃PO₄ on TNT were shown in Fig. 4a and b. Ag₃PO₄ nanoparticles located at both surface TNT and inside of nanotubes at a diameter of 10–20 nm. To further confirm the existence of Ag₃PO₄, HRTEM (Fig. 4c) showed the lattice fringe of 0.35 nm was consistent with (101) facet of anatase TiO₂, while the lattice fringe of 0.42 nm was assigned to the (110) facet of Ag₃PO₄.

Fig. 4 TEM (a, b) and HRTEM (c) images of Ag₃PO₄/TiO₂, 4 cycles.

3.4 Photoelectrochemical properties

Under scanning potential between −0.9 V to 0.3 V vs. Ag/AgCl, the current density of TNT deposited Ag₃PO₄ were bigger than that of the pure TNT (Fig. 5a), only 0.387 mA cm⁻² at 0 V vs. Ag/AgCl, whereas the current density of 4 cycles deposition was almost 5 times higher, reached to 2.340 mA cm⁻². It reported that maximum photocurrent of Ag₃PO₄/TiO₂ nanowire array heterostructure photoelectrodes was about 0.7 mA cm⁻² at an applied bias 0.23 V vs. Ag/AgCl.⁴² The improvement of photocurrent could be ascribed to Ag “pump”, preferred electron transportation direction along TNT and large absorption spectrum of Ag₃PO₄.^29,43 UV-vis spectra (Fig. 1c) proved the band gap of pure TNT and Ag₃PO₄ at 3.1 eV and 2.38 eV, respectively. And the recombination of photogenerated charge carriers was significantly reduced in the heterojunction (Fig. S1b†). Under intermittent light irradiation, time-dependent photocurrent generation is presented in Fig. 5b. The photocurrents respond exactly to the presence of solar light was interrupted every 50 s (light on/off) under 0 V vs. Ag/AgCl, and the steady state behavior of each photoanode follows the same trend as that of the I–V curves (Fig. 5a). Furthermore, the loading amounts of Ag₃PO₄ on TNT increase with the increasing of cycle time (Fig. 3c and S3c†). Whatever I–V curves or I–t curves showed photocurrent increased initially and then decreased with increase loading amount of Ag₃PO₄. Therefore appropriate loading amounts of Ag₃PO₄ is fatal to enhance PEC performance of TNT. In addition, there are no appreciable dark currents, demonstrating the photochemical stability of all of photoanodes in the alkaline solution. The steady-state photocurrents do not show any significant degradation with time.⁴⁴

Fig. 5 Ag₃PO₄/TiO₂ composites linear sweeps voltammetry (a), i–t curve (b), UV-vis spectra.

Fig. 6a represented Nyquist diagrams from electrochemical impedance spectroscopy (EIS) for samples tested in dark. The equivalent circuit for this cell system was depicted in inset of Fig. 6a. R_b was the bulk resistance of the electrolyte. C_sc and R_sc were the capacitance and the resistance of the solid-state interfacial layer which was formed at the highly charged state due to the passivation reaction between the electrolyte and the surface of the electrode. C_dl and R_ct were the double layer capacitance and the charge transfer resistance. W was the Warburg resistance result from the diffusion resistance of redox couple.

Fig. 6 Ag₃PO₄/TiO₂ Nyquist diagrams (a), insert is equivalent circuit fitting the EIS and Mott–Schottky plots (b).

The Mott–Schottky plots of Ag₃PO₄/TiO₂ are used to analyze the flat band potential and carrier concentrations of semiconductor (Fig. 6b). Under the applied voltage, the Fermi level can be changed and bended the energy band, and flat band potential V_fb refers the applied potential which makes the inner electric field intensity of semiconductor approach zero. The carrier concentrations of semiconductor N_D and the slope of linear part of Mott–Schottky plots have the relation as follow:

where, e is the elementary charge (1.602 × 10⁻¹⁹ C), ε is the dielectric constant of sample and the dielectric constant of anatase is 48, ε₀ is the vacuum dielectric constant (8.854 × 10⁻¹⁴ F cm⁻¹), m is the slope of the linear part of Mott–Schottky plots. The flat band potential V_fb, C_SC and carrier concentrations N_D have a relationship, below:where, C_SC is the space charge capacitance, ε₀, ε and e are constant, N_D is carrier concentrations, V and V_fb is respectively the applied potential and the flat band potential (vs. the potential of the reference electrode), (k_BT)/e is 25.8 mV at room temperature.

The parameters obtained from EIS and Mott–Schottky plots are shown in Table 1. The charge transfer resistances R_ct of 4 cycles deposition Ag₃PO₄ minimal only 404.7 ohm cm⁻², since the carrier concentration N_D of it was 2.6 times higher than that of the pure TiO₂ about 4.85 × 10²² cm⁻³. Moreover, the carrier concentration of Ag₃PO₄/TiO₂ samples was obviously higher than that of the pure TiO₂. And valence bands of two semiconductors were very close,^{29,33,43,45,46} which inferred the heterojunction formed by depositing Ag₃PO₄ nanoparticles on TNT (Fig. 1b). Potentials of both conduction band (−0.612 V vs. NHE) and valence band (+2.588 V vs. NHE) of TiO₂ are more negative than those of Ag₃PO₄ (conduction band potential: +0.45 V vs. NHE, valence band potential: +2.9 V vs. NHE).³³ It benefits for transfer of photogenerated carriers and separation of photogenerated electron–hole pairs, also increases the carrier concentrations. The flat band potential (V_fb) is close to the conduction band which makes the curving band of space charge layer straight. The V_fb of TNT was −0.612 V, after deposition Ag₃PO₄ of 4 cycles V_fb lowed to −0.410 V. The reason for lower V_fb was the photogenerated electrons were “pumped” into CB of TiO₂ by SPR of Ag. It meant the range of absorption spectrum broaden and photocatalytic properties enhancement. As shown in Fig. 1c, the absorbance of Ag₃PO₄/TiO₂ was significantly increased than pure TiO₂, implying Ag₃PO₄ is a very promising material for building fast electron transportation path to improve PEC performance of TiO₂.

Table 1 Physics parameters of Ag₃PO₄/TiO₂

Sample	Rct (ohm cm^−2)	ND (cm^−3)	Vfb (V)
2 cycles	667.5	4.40 × 10^22	−0.478
4 cycles	404.7	4.85 × 10^22	−0.410
6 cycles	698.8	2.42 × 10^22	−0.433
8 cycles	621.4	3.16 × 10^22	−0.468
Blank	594.8	1.35 × 10^22	−0.612

---

4 Conclusion

A preferred electron transportation path was designed to improve the photoelectrocatalytic activity through deposition Ag₃PO₄ nanoparticles to the self-organized TiO₂ nanotube arrays. Ag₃PO₄/TiO₂ effectively reduced the recombination of photogenerated electron–hole pairs due to unique 1-D nanostructure, as well as SPR of photoreduced Ag from Ag₃PO₄. The charge carrier concentration of Ag₃PO₄/TiO₂ increased 2.6 times than that of pure TNT. The photocurrents of Ag₃PO₄/TiO₂ were significantly improved by 5 times. Therefore, fabrication of Ag₃PO₄/TiO₂ nanotube arrays is a highly efficient method to build a noteworthy photoelectrode for PEC water splitting.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The financial support for this study by National Natural Science Foundation of China (No. 21476262). The Technology Development Plan of Qingdao (No. 14-2-4-108-jch). Research Funds for the Central Universities (No. 15CX05032A) are gratefully acknowledged.

References

1. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
2. J. H. Carey, J. Lawrence and H. M. Tosine, Bull. Environ. Contam. Toxicol., 1976, 16, 697–701.
3. B. O'Regan and M. Gratzel, Nature, 1991, 353, 737–740.
4. K. Park, Q. Zhang, D. Myers and G. Cao, ACS Appl. Mater. Interfaces, 2013, 5, 1044–1052.
5. J. Low, B. Cheng and J. Yu, Appl. Surf. Sci., 2017, 392, 658–686.
6. W.-T. Sun, Y. Yu, H.-Y. Pan, X.-F. Gao, Q. Chen and L.-M. Peng, J. Am. Chem. Soc., 2008, 130, 1124–1125.
7. X. Feng, K. Shankar, O. K. Varghese, M. Paulose, T. J. Latempa and C. A. Grimes, Nano Lett., 2008, 8, 3781–3786.
8. H. Xiong, M. D. Slater, M. Balasubramanian, C. S. Johnson and T. Rajh, J. Phys. Chem. Lett., 2011, 2, 2560–2565.
9. B. Liu and E. S. Aydil, J. Am. Chem. Soc., 2009, 131, 3985–3990.
10. Q. Liang, Y. Shi, W. Ma, Z. Li and X. Yang, Phys. Chem. Chem. Phys., 2012, 14, 15657–15665.
11. H. Cheng-Xing, Y. Lian-Qing, Z. Ya-Ping, D. Kai-Tuo and H. Lan-Zhong, J. Inorg. Mater., 2016, 31, 1237–1241.
12. L. Yu, Y. Zhang, Q. Zhi, Q. Wang, F. Gittleson, J. Li and A. D. Taylor, Sens. Actuators, B, 2015, 211, 111–115.
13. D. Gong, C. A. Grimes and O. K. Varghese, et al., J. Mater. Res., 2001, 16, 3331–3334.
14. J. M. Macak, H. Tsuchiya, L. Taveira, A. Ghicov and P. Schmuki, J. Biomed. Mater. Res., Part A, 2005, 75, 928–933.
15. Z. Zhang and P. Wang, Energy Environ. Sci., 2012, 5, 6506–6512.
16. K. Lee, A. Mazare and P. Schmuki, Chem. Rev., 2014, 114, 9385–9454.
17. C. Feng, Y. Pang, Y. Wang, M. Sun, C. Zhang, L. Zhang, Y. Zhou and D. Li, Appl. Surf. Sci., 2016, 376, 188–198.
18. L. Ma, H. Han, L. Pan, M. Tahir, L. Wang, X. Zhang and J.-J. Zou, RSC Adv., 2016, 6, 63984–63990.
19. X. Yang, H. Cui, Y. Li, J. Qin, R. Zhang and H. Tang, ACS Catal., 2013, 3, 363–369.
20. N. Umezawa, O. Y. Shuxin and J. H. Ye, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 8.
21. Y. Bi, H. Hu, S. Ouyang, Z. Jiao, G. Lu and J. Ye, J. Mater. Chem., 2012, 22, 14847–14850.
22. Y. Bi, H. Hu, S. Ouyang, G. Lu, J. Cao and J. Ye, Chem. Commun., 2012, 48, 3748–3750.
23. H. Hu, Z. Jiao, T. Wang, J. Ye, G. Lu and Y. Bi, J. Mater. Chem. A, 2013, 1, 10612–10616.
24. H. Hu, Z. Jiao, H. Yu, G. Lu, J. Ye and Y. Bi, J. Mater. Chem. A, 2013, 1, 2387–2390.
25. C. Cui, Y. Qiu, H. Hu, N. Ma, S. Li, L. Xu, L. Chaorong, J. Xu and T. Weihua, RSC Adv., 2016, 6, 43697–43706.
26. J. Li, X. Ji, X. Li, X. Hu, Y. Sun, J. Ma and G. Qiao, Appl. Surf. Sci., 2016, 372, 30–35.
27. W.-S. Wang, H. Du, R.-X. Wang, T. Wen and A.-W. Xu, Nanoscale, 2013, 5, 3315–3321.
28. W. Teng, X. Li, Q. Zhao and G. Chen, J. Mater. Chem. A, 2013, 1, 9060–9068.
29. J.-W. Xu, Z.-D. Gao, K. Han, Y. Liu and Y.-Y. Song, ACS Appl. Mater. Interfaces, 2014, 6, 15122–15131.
30. F.-Q. Xiong, X. Wei, X. Zheng, D. Zhong, W.-H. Zhang and C. Li, J. Mater. Chem. A, 2014, 2, 4510–4513.
31. X. Yang, J. Qin, Y. Jiang, K. Chen, X. Yan, D. Zhang, R. Li and H. Tang, Appl. Catal., B, 2015, 166–167, 231–240.
32. X. Yang, J. Qin, Y. Jiang, R. Li, Y. Li and H. Tang, RSC Adv., 2014, 4, 18627–18636.
33. W. Yao, B. Zhang, C. Huang, C. Ma, X. Song and Q. Xu, J. Mater. Chem., 2012, 22, 4050–4055.
34. H. Tang, Y. Fu, S. Chang, S. Xie and G. Tang, Chin. J. Catal., 2017, 38, 337–347.
35. X. Yang, Z. Chen, J. Xu, H. Tang, K. Chen and Y. Jiang, ACS Appl. Mater. Interfaces, 2015, 7, 15285–15293.
36. D. Wang, L. Li, Q. Luo, J. An, X. Li, R. Yin and M. Zhao, Appl. Surf. Sci., 2014, 321, 439–446.
37. S. Bai, X. Li, Q. Kong, R. Long, C. Wang, J. Jiang and Y. Xiong, Adv. Mater., 2015, 27, 3444–3452.
38. X. Yang, H. Tang, J. Xu, M. Antonietti and M. Shalom, ChemSusChem, 2015, 8, 1350–1358.
39. C. Wagner, W. Riggs, L. Davis, J. Moulder and G. Muilenberg, Handbook of X-ray Photoelectron Spectroscopy, Perkin-Elmer Corp, 1979.
40. P. Wang, B. Huang, Z. Lou, X. Zhang, X. Qin, Y. Dai, Z. Zheng and X. Wang, Chem.–Eur. J., 2010, 16, 538–544.
41. P. Wang, B. Huang, Q. Zhang, X. Zhang, X. Qin, Y. Dai, J. Zhan, J. Yu, H. Liu and Z. Lou, Chem.–Eur. J., 2010, 16, 10042–10047.
42. B. Jin, X. Zhou, J. Luo, X. Xu, L. Ma, D. Huang, Z. Shao and Z. Luo, RSC Adv., 2015, 5, 48118–48123.
43. B. Lu, N. Ma, Y. Wang, Y. Qiu, H. Hu, J. Zhao, D. Liang, S. Xu, X. Li, Z. Zhu and C. Cui, J. Alloys Compd., 2015, 630, 163–171.
44. J. S. Jang, C. W. Ahn, S. S. Won, J. H. Kim, W. Choi, B.-S. Lee, J.-H. Yoon, H. G. Kim and J. S. Lee, J. Phys. Chem. C, 2017, 121, 15063–15070.
45. S. B. Rawal, S. D. Sung and W. I. Lee, Catal. Commun., 2012, 17, 131–135.
46. L. Zhang, H. Zhang, H. Huang, Y. Liu and Z. Kang, New J. Chem., 2012, 36, 1541–1544.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7ra11283a

---