Effects of the preparation method of Pt/g-C3N4 photocatalysts on their efficiency for visible-light hydrogen production

Xuanbo Zhou , Yunfeng Li , Yan Xing *, Junsong Li and Xin Jiang
College of Chemistry, Northeast Normal University, Changchun 130024, P.R. China. E-mail: xingy202@nenu.edu.cn

Received 18th July 2019 , Accepted 8th September 2019

First published on 9th September 2019

As a two-dimensional (2D) nanomaterial, bulk g-C3N4 (CNB) has a low specific surface area and weak electron transport ability, which limit its application in photocatalysis. In this paper, ultrathin porous g-C3N4 nanosheets (CNS) have been synthesized by thermal oxidation etching of CNB. Compared with CNB, CNS possess a larger surface area of 234.65 m2 g−1, good dispersity in water and a high electron transfer rate. As a co-catalyst, ultra-small Pt nanoparticles (NPs) with high dispersity are successfully loaded on the surface of CNS. It is found that changing the loading method of Pt NPs in the preparation step remarkably alters the efficiency for hydrogen production. The Pt/CNS-CR photocatalyst fabricated by the chemical reduction (CR) method shows a much higher efficiency for H2 evolution from water splitting, compared to the Pt/CNS-PR photocatalyst obtained by the loading of Pt NPs by the traditional photo-reduction (PR) method. When triethanolamine (TEOA) is used as a hole sacrificial agent, the hydrogen production rate of 2.0%-Pt/CNS-CR is 7862.5 μmol g−1 h−1, which is 6.92 times higher than that of 2.0%-Pt/CNS-PR (1136.8 μmol g−1 h−1). The valence states of the Pt element in the Pt/CNS-CR and Pt/CNS-PR nanocomposites have been analyzed by X-ray photoelectron spectroscopy, respectively. At the same time, the effects of the loading amount of Pt and different sacrificial reagents on the photocatalytic H2 generation activity have also been systematically investigated.

1. Introduction

Environmental pollution and energy crisis have been increasing due to the development of modern industries and the excessive use of fossil fuels. Semiconductor photocatalytic technology plays a pivotal role as a new approach for reducing environmental damage and developing green energy, such as photocatalytic degradation of pollutants,1 photocatalytic reduction of carbon dioxide,2 photocatalytic selective alcohol oxidation3 and photocatalytic hydrogen production.4,5 Ever since the photocatalytic splitting of water on TiO2 electrodes was discovered in 1972,6 TiO2 has become the most widely used photocatalyst due to its excellent redox ability and long-term stability against photocorrosion.7–9 However, the wide band gap (∼3.2 eV) limits the visible light hydrogen production of TiO2 as a photocatalyst.10 Therefore, development of efficient visible light responsive photocatalysts has become the research focus. So far, a variety of semiconductor photocatalysts including CdS,11,12 WO3,13,14 Cu2O,15,16 Bi2WO617 and BiVO418 have been reported. Among the visible light responsive photocatalysts, metal-free photocatalysts such as graphitic carbon nitride (g-C3N4) have drawn great attention. As a melon-based π-conjugated polymer, g-C3N4 exhibits a wide range of light response (∼450 nm) due to a narrow band gap of ∼2.7 eV. Moreover, g-C3N4 possesses many excellent properties for practical applications, such as non-toxicity, low cost, extraordinary stability, and easy preparation.19–22 Nevertheless, the photocatalytic performance of g-C3N4 remains unsatisfactory owing to the high recombination rate of photo-induced electron–hole pairs and the poor quantum efficiency, which greatly hinder its practical applications.23–25 Recently, it has been proved that integration of a co-catalyst with g-C3N4 is an effective strategy for enhancing its photocatalytic efficiency, ascribed to the increased separation ability of electrons and holes and the decreased activation barriers of water splitting.24,26 Platinum (Pt), with the lowest overpotential and the highest work function,27–29 has been believed to be the most highly effective co-catalyst for hydrogen evolution.30,31 Recently, uniform Pt quantum dot-decorated porous g-C3N4 nanosheets have been fabricated by Jiang et al., and the Pt/CN composite showed excellent catalytic activity in photocatalytic hydrogen production.32 Moreover, Wang and co-workers discussed the hydrogen production properties of Pt NPs anchored on the surface of g-C3N4 in both acidic and alkaline environments.33 Besides, Guan et al. reported that the Eosin Y-sensitized g-C3N4/Pt/GO nanocomposite showed excellent H2 production performance under optimized conditions.34 Most recently, the shape effect of the Pt co-catalyst on the photocatalytic activity has been investigated by Yao's group.35 Most remarkably, single-atom Pt integrated with g-C3N4 demonstrated catalytic activity nearly 50 times that of bare g-C3N4 in photocatalytic hydrogen evolution.36 However, loading of Pt NPs on g-C3N4 is normally achieved by the photo-deposition method, which is facile and easy to operate, nevertheless, the particle size and the chemical state of Pt are hard to control during the rapid photo-reduction process. A recent study revealed that the Pt0 co-catalyst served as an efficient hydrogen generation site, while Pt2+ could suppress the unfavourable hydrogen back-oxidation.37,38 Therefore, it is highly desirable to develop an effective method to load the Pt co-catalyst with a controlled proportion of different valence states of Pt on the surface of g-C3N4, leading to a high photocatalytic hydrogen production rate.

In this work, ultrathin porous g-C3N4 nanosheets (CNS) were fabricated by thermal exploitation of bulk g-C3N4 (CNB). Pt NPs were successfully anchored on the surface of CNS by either a chemical reduction (CR) or a photoreduction (PR) method at room temperature. Due to the large specific surface area of CNS, Pt NPs were uniformly dispersed on the surface of the CNS and exposed more active sites, leading to the increased utilization efficiency of Pt NPs. Furthermore, the larger proportion of Pt0 in the Pt/CNS-CR nanocomposite was more conducive to promoting the electron transfer, resulting in a higher hydrogen production rate of the Pt/CNS-CR nanocomposite than that of Pt/CNS-PR.

2. Results and discussion

2.1. Morphology and composition structures of the samples

XRD patterns of pure CNS, Pt/CNS-CR and Pt/CNS-PR nanocomposites are shown in Fig. 1. The diffraction peaks at 13.1° and 27.6° are indexed to the (1 0 0) and (0 0 2) planes of g-C3N4, respectively.4 After loading Pt NPs on the surface of CNS by either the photo-reduction or the chemical reduction method, only a new weak peak associated with the (1 1 1) plane of Pt NPs (JCPDS No. 04-0802) at 39.7° can be observed, owing to the low loading amount and the high dispersity of Pt NPs on the surface of CNS. While in the XRD pattern of the 2.0%-Pt/CNB-CR nanocomposite (Fig. S1A), peaks for the (1 1 1) and (2 0 0) planes of Pt NPs can be observed clearly, due to the poor dispersion of Pt NPs on the surface of CNB, which leads to the larger sized Pt NPs being anchored on the surface of CNB (Fig. S1C and D).
image file: c9dt02938a-f1.tif
Fig. 1 XRD patterns of the as-prepared CNS, Pt/CNS-CR and Pt/CNS-PR nanocomposites.

A SEM image of the as-synthesized CNB shows a thick block structure (Fig. S1B); after thermal oxidation exfoliation, the as-obtained CNS exhibit a 2D ultrathin sheet-like structure with a large amount of in-plane holes on the wrinkled surface, suggesting that CNS have a larger specific surface area and more active sites for the anchoring of Pt NPs (Fig. 2A and B). A TEM image (Fig. 2C) of 2.0%-Pt/CNS-CR shows that Pt NPs with a mean size of ∼2.0 nm (Fig. S2A) are uniformly dispersed on the surface or in the pores of the CNS. The HRTEM image (inset of Fig. 2C) indicates the presence of Pt NPs. The interplanar spacing of the nanoparticle is about 0.226 nm, which is consistent with the (111) plane of Pt. As the amount of Pt content increases from 1.0% to 2.5% in the Pt/CNS-CR nanocomposites (Fig. S3 and Table S1), the loading number of Pt NPs on the surface of CNS increases, however, the average size of the Pt NPs does not change obviously. As for the 2.0%-Pt/CNS-PR nanocomposite (Fig. 2D and S2B), the average size of Pt NPs is ∼1.8 nm, which is slightly smaller than that of the Pt/CNS-CR nanocomposite. However, Pt NPs are observed to be aggregated on the surface of CNB and the mean size is about 8.2 nm (Fig. S1C and D).

image file: c9dt02938a-f2.tif
Fig. 2 A SEM image of (A) CNS and TEM images of (B) CNS, (C) 2.0%-Pt/CNS-CR (inset shows the HRTEM image of Pt NPs) and (D) 2.0%-Pt/CNS-PR.

Fig. 3 shows the N2 adsorption–desorption isotherms of the as-prepared CNB, CNS and 2.0%-Pt/CNS-CR samples. The specific surface area of pure CNB is only 10.89 m2 g−1, while the CNS sample possesses a large specific surface area of 234.65 m2 g−1, which indicates that the ultrathin porous CNS might be an excellent catalyst support. After the loading of Pt NPs, the 2.0%-Pt/CNS-CR nanocomposite still displays a relatively large specific surface area of 172.01 m2 g−1. The high surface area provides more exposed active sites, thus ensuring an enhanced photocatalytic performance.

image file: c9dt02938a-f3.tif
Fig. 3 Nitrogen adsorption–desorption isotherms of CNB, CNS and 2.0%-Pt/CNS-CR samples.

XPS photoelectron spectroscopy was used to investigate the surface elemental composition and chemical states of the as-obtained nanocomposites. As shown in Fig. 4A and B, the C 1s spectra of 2.0%-Pt/CNS-PR and 2.0%-Pt/CNS-CR show two main peaks at 284.7 eV and 288.6 eV, which are attributed to the graphitic carbon and N–C[double bond, length as m-dash]N functional groups in the composites, respectively.39 The N 1s spectra of both 2.0%-Pt/CNS-PR and 2.0%-Pt/CNS-CR exhibit four peaks at 399.1, 400.2, 401.4 and 404.4 eV, which are associated with C[double bond, length as m-dash]N–C functional groups, tertiary nitrogen (N(C)3), N–H structure and π-excitations, respectively.40,41 Compared with the CNS, the peaks of C 1s and N 1s shift to the positions with higher binding energies after the deposition of Pt NPs. These shifts indicate an interaction between the Pt NPs and the CNS, owing to the polarization of electron density between the CNS support and the metal Pt.42 The strong metal–support interaction is beneficial for the electron transfer and the improved photocatalytic hydrogen production performance.

image file: c9dt02938a-f4.tif
Fig. 4 High-resolution (A) C 1s and (B) N 1s XPS spectra of CNS, 2.0%-Pt/CNS-PR and 2.0%-Pt/CNS-CR samples. Pt 4f XPS spectra of (C) 2.0%-Pt/CNS-PR and (D) 2.0%-Pt/CNS-CR.

Fig. 4C and D display the XPS spectra of Pt 4f for the Pt loaded nanocomposites via the photo-reduction and the chemical reduction method, respectively. For 2.0%-Pt/CNS-PR, the chemical states of Pt on the surface of CNS are Pt0, Pt2+ and Pt4+. Correspondingly, Pt2+ (73.58%) accounts for the largest proportion, while Pt0 (7.55%) for the smallest. The existence of Pt4+ might be due to the partly unreduced H2PtCl6·6H2O. However, there are no peaks of Pt4+ for 2.0%-Pt/CNS-CR, indicating the complete reduction of H2PtCl6·6H2O during the process of chemical reduction. The main peaks of Pt 4f at 71.0 eV and 74.1 eV correspond to Pt 4f7/2 and Pt 4f5/2 of metallic Pt0, respectively.29 The binding energies of 73.1 eV and 76.5 eV are attributed to Pt 4f7/2 and Pt 4f5/2 of Pt2+, respectively.43 It can be seen that for 2.0%-Pt/CNS-CR, the proportion of Pt0 (62.03%) is larger than that of Pt2+ (37.97%). These results suggest that the degree of reduction of metallic Pt0 through different reduction methods is different. It has been reported that the metallic Pt0 species as the active sites can trap electrons and facilitate H2 evolution.37,44 Meanwhile, oxidized Pt2+ plays an important role in inhibiting the undesirable hydrogen back-oxidation.37,38 Thus it is anticipated that the Pt/CNS-CR photocatalyst with a larger proportion of Pt0 might be more efficient than Pt/CNS-PR for hydrogen production.

2.2. Optical and electrochemical properties

UV-vis diffuse reflectance spectra of the CNS and Pt/CNS-CR samples are displayed in Fig. 5. Compared with the pure CNS, the onset absorption edges of the Pt/CNS-CR composites shift to a longer wavelength after the loading of Pt NPs on the surface of CNS. The light absorption intensities of the Pt/CNS-CR composites increase obviously, attributed to the plasma resonance effect of the Pt NPs.45 As shown in Fig. 6A, the colour of CNS is light yellow, and the corresponding absorption band edge is around 450 nm. After the deposition of Pt NPs by photo-reduction and chemical reduction methods, the colour of 2.0%-Pt/CNS-PR and 2.0%-Pt/CNS-CR becomes brown yellow and dark gray, respectively. The corresponding absorption edges of Pt/CNS-PR and Pt/CNS-CR are 460 nm and 486 nm, respectively. According to the plot of the Kubelka–Munk function versus the energy of exciting light (Fig. 6B), the bandgaps of CNS, 2.0%-Pt/CNS-PR and 2.0%-Pt/CNS-CR are estimated to be 2.75, 2.69 and 2.55 eV, respectively. According to the Mott–Schottky plots (Fig. S4), the flat band potentials of CNS, 2%-Pt/CNS-PR and 2%-Pt/CNS-CR are approximately −0.43 V (vs. NHE). As the Pt NPs are loaded on the surface of CNS, the CB potentials of the composites remain unchanged while the calculated VB potentials show an upward shift. The narrowed band gap and the increased intensity for the visible light absorption of the 2%-Pt/CNS-CR photocatalyst are beneficial for the enhancement of photocatalytic H2 production.
image file: c9dt02938a-f5.tif
Fig. 5 UV-vis diffuse reflectance spectra of CNS and Pt/CNS-CR samples.

image file: c9dt02938a-f6.tif
Fig. 6 (A) UV-vis diffuse reflectance spectra and the photographs and (B) the plots of (F(R) )2versus the energy of exciting light () for CNS, 2.0%-Pt/CNS-CR and 2.0%-Pt/CNS-PR samples.

In order to prove the positive role of metallic Pt in charge transport and separation, photoluminescence (PL) analysis and electrochemical impedance spectroscopy (EIS) have been carried out. As shown in Fig. 7A, CNS display the highest fluorescence emission peak among the as-obtained samples, indicating the highest recombination rate of the photo-induced charges. Instead, the 2.0%-Pt/CNS-PR and 2.0%-Pt/CNS-CR composites show a low fluorescence intensity, implying that the presence of the Pt co-catalyst effectively inhibits the recombination of electrons and holes. Moreover, 2.0%-Pt/CNS-CR shows a lower peak intensity than 2.0%-Pt/CNS-PR, attributed to the effect of the high content of Pt0 on the inhibition of electron–hole pair recombination. EIS results shown in Fig. 7B demonstrate that the 2.0%-Pt/CNS-CR composite shows the smallest radius of the arc, which suggests the smallest resistance of electron and hole transfer.

image file: c9dt02938a-f7.tif
Fig. 7 (A) Photoluminescence spectra and (B) electrochemical impedance spectra of CNS, 2.0%-Pt/CNS-PR and 2.0%-Pt/CNS-CR samples.

2.3. Photocatalytic activities

The photocatalytic H2 evolution activities of all the samples were investigated under visible light conditions (λ ≥ 420 nm) using TEOA as a hole sacrificial reagent. There is no H2 evolution without the photocatalyst or light irradiation, indicating that H2 is generated by the photocatalytic reactions. As displayed in Fig. 8A, only a trace amount of H2 production for the pure CNS can be observed, however, after the deposition of the Pt co-catalyst on the CNS, the H2 production performance improves significantly. The photocatalytic activities of the Pt/CNS-CR composites enhance with the increase of the loading contents of Pt, in which 2.0%-Pt/CNS-CR shows the highest amount of H2 generation of 15[thin space (1/6-em)]725.0 μmol g−1 under visible light irradiation for 120 min. Furthermore, upon increasing the loading amount of Pt to 2.5%, the amount of H2 generation decreases instead, because excessive Pt loading increases the height of Schottky barrier between the CNS and Pt interface.46 Among the Pt/CNS-PR composites with different loading amounts of Pt, the 2.0%-Pt/CNS-PR composite shows the highest hydrogen evolution efficiency (Fig. S5A and B). Nevertheless, it is still lower than those of all the Pt/CNS-CR composites. Notably, the 2.0%-Pt/CNB-CR composite shows the lowest amount of H2 production among all the samples being tested. This indicates that the larger-sized Pt NPs on the surface of the CNB and the weaker electron transport ability between the CNB and Pt NPs lead to an unfavorable effect on the photocatalytic H2 production. As shown in Fig. 8B, the H2 evolution rate of 2.0%-Pt/CNS-CR is 7862.5 μmol g−1 h−1, which is 6.92 times higher than that of the 2.0%-Pt/CNS-PR (1136.8 μmol g−1 h−1). As summarized in Table S2, the hydrogen evolution rate of the 2.0%-Pt/CNS-CR photocatalyst in our work is much higher than those of noble metal modified g-C3N4 based photocatalysts in the literature observations. Moreover, the turnover-frequency (TOF) of the 2.0%-Pt/CNS-CR and 2.0%-Pt/CNS-PR samples is calculated by the method of Yao et al.30 The TOF of 2.0%-Pt/CNS-CR (1.60 wt% of Pt loading) is 692.121 h−1, which is nine times higher than that of 2.0%-Pt/CNS-PR (1.90 wt% of Pt loading) with a TOF of 76.295 h−1. The above results indicate that the chemical reduction is an effective method to achieve a high content of Pt0, promoting the interfacial electron transfer and increasing the utilization rate of Pt NPs. Furthermore, an effective contact between the metallic Pt0 and CNS, an enhanced visible light harvesting and the synergistic effect between Pt0 and Pt2+ are all beneficial for improving the photocatalytic hydrogen production. After 10 cycling tests (Fig. 8C), the amount of hydrogen generated decreases slightly due to the agglomeration of Pt NPs during the photocatalytic process (Fig. S6). However, the amount of hydrogen generation still remains 14[thin space (1/6-em)]643.5 μmol g−1 under visible light irradiation, indicating the good stability of the photocatalyst. In addition, the effect of different hole sacrificial agents on the photocatalytic hydrogen production of the 2.0%-Pt/CNS-CR composite was also investigated (Fig. 8D). Four different sacrificial agents of TEOA, lactic acid, methanol (MeOH) and disodium ethylenediaminetetraacetic acid (EDTA4−) were chosen for comparison. It is found that when using TEOA as the hole sacrificial agent the highest hydrogen evolution performance is achieved. This is due to faster hole scavenging resulting from a higher redox potential (−3.4 eV) of TEOA, accelerating the separation rate of electrons and holes.47
image file: c9dt02938a-f8.tif
Fig. 8 (A) Amount of H2 generation of CNS, 2.0%-Pt/CNB-CR, Pt/CNS-CR and 2.0%-Pt/CNS-PR samples. (B) Hydrogen evolution rate of Pt/CNS-CR and 2.0%-Pt/CNS-PR samples under visible light irradiation. (C) The cycling test of photocatalytic H2 generation for the 2.0%-Pt/CNS-CR nanocomposite. (D) Effects of different hole scavengers for the 2.0%-Pt/CNS-CR nanocomposite under visible light irradiation.

3. Photocatalytic mechanism of H2 production

Based on the above experimental results, a possible mechanism for the visible light production of H2 on the nanocomposite is proposed in Scheme 1. Under visible light irradiation, CNS can produce photo-generated electrons (e) and holes (h+) in the conduction band (CB) and valence band (VB), respectively. The electrons from the CB of CNS transfer to the surface of the Pt NPs quickly due to the excellent electronic conductivity of metal Pt0, leading to efficient electron–hole separation. Then, H+ can accept the electrons from the Pt NPs to produce H2. At the same time, the holes in the valence band are consumed by the sacrificial agent of TEOA through the oxidation reaction.
image file: c9dt02938a-s1.tif
Scheme 1 The schematic diagram of the possible photocatalytic H2 evolution mechanism.

4. Conclusions

In conclusion, ultrathin porous g-C3N4 nanosheets have been prepared, which exhibit a large surface area and good electron transport ability. Then, as a co-catalyst, Pt NPs are successfully anchored on the surface of CNS by a chemical reduction method for efficient hydrogen production under visible light irradiation. Compared with the traditional photo-reduction method, the superior performance of 2.0%-Pt/CNS-CR is mainly attributed to the high content of metallic Pt0 facilitating the electron transport and separation, Pt2+ suppressing the undesirable hydrogen back-oxidation, and the intimate contact between the Pt NPs and the CNS. This work provides a new possibility for further improvement of the photocatalytic performances of g-C3N4-based photocatalysts, so as to exploit solar energy more efficiently.

Conflicts of interest

There are no conflicts to declare.


This work is supported by the National Natural Science Foundation of China (Grant No. 21872023 and 21473027), and Opening Fund of the State Key Laboratory of Inorganic Synthesis and Preparative Chemistry of Jilin University.

Notes and references

  1. Y. F. Li, R. X. Jin, X. Fang, Y. Yang, M. Yang, X. C. Liu, Y. Xing and S. Song, J. Hazard. Mater., 2016, 313, 219 CrossRef CAS PubMed .
  2. A. Bafaqeer, M. Tahir and N. A. S. Amin, Appl. Catal., B, 2019, 242, 312 CrossRef CAS .
  3. J. G. Wang, Z. M. Chen, G. J. Zhai and Y. Men, Appl. Surf. Sci., 2018, 462, 760 CrossRef CAS .
  4. Y. F. Li, M. Yang, Y. Xing, X. C. Liu, Y. Yang, X. Wang and S. Y. Song, Small, 2017, 13, 1701552 CrossRef .
  5. Q. Zhao, J. Sun, S. C. Li, C. P. Huang, W. F. Yao, W. Chen, T. Zeng, Q. Wu and Q. J. Xu, ACS Catal., 2018, 8, 11863 CrossRef CAS .
  6. A. Fujishima and K. Honda, Nature, 1972, 238, 37 CrossRef CAS PubMed .
  7. J. Y. Lei, B. Chen, W. J. Lv, L. Zhou, L. Z. Wang, Y. D. Liu and J. L. Zhang, Dalton Trans., 2019, 48, 3486 RSC .
  8. W. J. Foo, C. Zhang and G. W. Ho, Nanoscale, 2013, 5, 759 RSC .
  9. M. T. Uddin, Y. Nicolas, C. Olivier, W. Jaegermann, N. Rockstroh, H. Junge and T. Toupance, Phys. Chem. Chem. Phys., 2017, 19, 19279 RSC .
  10. S. B. Patil, P. S. Basavarajappa, N. Ganganagappa, M. S. Jyothi, A. V. Raghu and K. R. Reddy, Int. J. Hydrogen Energy, 2019, 44, 13022 CrossRef CAS .
  11. S. Iqbal, Z. W. Pan and K. B. Zhou, Nanoscale, 2017, 9, 6638 RSC .
  12. Y. J. Yuan, D. Q. Chen, Z. T. Yu and Z. G. Zou, J. Mater. Chem. A, 2018, 6, 11606 RSC .
  13. L. J. Wei, H. M. Zhang and J. Cao, Mater. Lett., 2019, 236, 171 CrossRef CAS .
  14. B. Song, T. T. Wang, H. G. Sun, Q. Shao, J. K. Zhao, K. K. Song, L. H. Hao, L. Wang and Z. H. Guo, Dalton Trans., 2017, 46, 15769 RSC .
  15. M. R. Abhilash, G. Akshatha and S. Srikantaswamy, RSC Adv., 2019, 9, 8557 RSC .
  16. M. M. Zhang, Z. L. Chen, Y. Wang, J. F. Zhang, X. R. Zheng, D. W. Rao, X. P. Han, C. Zhong, W. B. Hu and Y. D. Deng, Appl. Catal., B, 2019, 246, 202 CrossRef CAS .
  17. Q. Sun, X. R. Jia, X. F. Wang, H. G. Yu and J. G. Yu, Dalton Trans., 2015, 44, 14532 RSC .
  18. Z. H. Wei, Y. F. Wang, Y. Y. Li, L. Zhang, H. C. Yao and Z. J. Li, J. CO2 Util., 2018, 28, 15 CrossRef CAS .
  19. M. Xiao, B. Luo, S. C. Wang and L. Z. Wang, J. Energy Chem., 2018, 27, 1111 CrossRef .
  20. B. C. Zhu, L. Y. Zhang, B. Cheng and J. G. Yu, Appl. Catal., B, 2018, 224, 983 CrossRef CAS .
  21. B. T. Xu, M. B. Ahmed, J. L. Zhou, A. Altaee, G. Xu and M. H. Wu, Sci. Total Environ., 2018, 633, 546 CrossRef CAS PubMed .
  22. P. Wang, L. L. Zong, Z. J. Guan, Q. Y. Li and J. J. Yang, Nanoscale Res. Lett., 2018, 13, 33 CrossRef PubMed .
  23. X. J. Bai, C. P. Sun, S. L. Wu and Y. F. Zhu, J. Mater. Chem. A, 2015, 3, 2741 RSC .
  24. H. Zhao, Y. M. Dong, P. P. Jiang, H. Y. Miao, G. L. Wang and J. J. Zhang, J. Mater. Chem. A, 2015, 3, 7375 RSC .
  25. J. W. Wang, X. J. Zuo, W. Cai, J. W. Sun, X. L. Ge and H. Zhao, Dalton Trans., 2018, 47, 15382 RSC .
  26. G. G. Zhang, Z. A. Lan and X. C. Wang, Chem. Sci., 2017, 8, 5261 RSC .
  27. X. H. Li and M. Antonietti, Chem. Soc. Rev., 2013, 42, 6593 RSC .
  28. F. Fina, H. Ménard and J. T. S. Irvine, Phys. Chem. Chem. Phys., 2015, 17, 13929 RSC .
  29. S. W. Cao, J. Jiang, B. C. Zhu and J. G. Yu, Phys. Chem. Chem. Phys., 2016, 18, 19457 RSC .
  30. M. H. Luo, P. Lu, W. F. Yao, C. P. Huang, Q. J. Xu, Q. Wu, Y. Kuwahara and H. Yamashita, ACS Appl. Mater. Interfaces, 2016, 8, 20667 CrossRef CAS PubMed .
  31. J. Yao, Y. R. Zheng, X. Jia, L. X. Duan, Q. Wu, C. P. Huang, W. An, Q. J. Xu and W. F. Yao, ACS Appl. Mater. Interfaces, 2019, 11, 25844 CrossRef CAS PubMed .
  32. F. F. Sun, S. Y. Tan, H. Zhang, Z. P. Xing, R. O. Yang, B. B. Mei and Z. Jiang, J. Colloid Interface Sci., 2018, 531, 119 CrossRef CAS .
  33. Z. H. Wang, X. F. Peng, S. S. Tian and Z. Wang, Mater. Res. Bull., 2018, 104, 1 CrossRef CAS .
  34. P. Wang, Z. J. Guan, Q. Y. Li and J. J. Yang, J. Mater. Sci., 2018, 53, 774 CrossRef CAS .
  35. M. H. Luo, W. F. Yao, C. P. Huang, Q. Wu and Q. J. Xu, J. Mater. Chem. A, 2015, 3, 13884 RSC .
  36. X. G. Li, W. T. Bi, L. Zhang, S. Tao, W. S. Chu, Q. Zhang, Y. Luo, C. Z. Wu and Y. Xie, Adv. Mater., 2016, 28, 2427 CrossRef CAS PubMed .
  37. Y. H. Li, J. Xing, Z. J. Chen, Z. Li, F. Tian, L. R. Zheng, H. F. Wang, P. Hu, H. J. Zhao and H. G. Yang, Nat. Commun., 2013, 4, 2500 CrossRef .
  38. Y. H. Li, C. Z. Li and H. G. Yang, J. Mater. Chem. A, 2017, 5, 20631 RSC .
  39. Y. S. Fu, T. Huang, B. Q. Jia, J. W. Zhu and X. Wang, Appl. Catal., B, 2017, 202, 430 CrossRef CAS .
  40. L. Shi, K. Chang, H. B. Zhang, X. Hai, L. Q. Yang, T. Wang and J. H. Ye, Small, 2016, 12, 4431 CrossRef CAS .
  41. Q. L. Xu, B. Cheng, J. G. Yu and G. Liu, Carbon, 2017, 118, 241 CrossRef CAS .
  42. V. W. Lau, I. Moudrakovski, T. Botari, S. Weinberger, M. B. Mesch, V. Duppel, J. Senker, V. Blum and B. V. Lotsch, Nat. Commun., 2016, 7, 12165 CrossRef CAS PubMed .
  43. X. Q. An, W. Wang, J. P. Wang, H. Z. Duan, J. T. Shi and X. L. Yu, Phys. Chem. Chem. Phys., 2018, 20, 11405 RSC .
  44. W. N. Xing, W. G. Tu, M. Ou, S. Y. Wu, S. M. Yin, H. J. Wang, G. Chen and R. Xu, ChemSusChem, 2019, 12, 2029 CrossRef CAS PubMed .
  45. L. S. Zhang, N. Ding, L. C. Lou, K. Iwasaki, H. J. Wu, Y. H. Luo, D. M. Li, K. Nakata, A. Fujishima and Q. B. Meng, Adv. Funct. Mater., 2018, 29, 1806774 CrossRef .
  46. Y. Shiraishi, Y. Kofuji, S. Kanazawa, H. Sakamoto, S. Ichikawa, S. Tanaka and T. Hirai, Chem. Commun., 2014, 50, 15255 RSC .
  47. M. J. Berr, P. Wagner, S. Fischbach, A. Vaneski, J. Schneider, A. S. Susha, A. L. Rogach, F. Jäckel and J. Feldmann, Appl. Phys. Lett., 2012, 100, 223903 CrossRef .


Electronic supplementary information (ESI) available. See DOI: 10.1039/C9DT02938A

This journal is © The Royal Society of Chemistry 2019