Z-scheme photocatalytic hydrogen production over WO₃/g-C₃N₄ composite photocatalysts

Abstract

WO₃/g-C₃N₄ composite photocatalysts were prepared by a simple calcination method and H₂ production activity of these composites was evaluated. The photocatalytic activity of the composites highly depended on WO₃ content. The enhanced photocatalytic activity could be ascribed to the Z-scheme mechanism, which results in the efficient charge separation.

---

As a typical metal free inorganic semiconductor, graphitic C₃N₄ (g-C₃N₄) has attracted intensive attention for H₂ generation,¹ pollutant degradation² and CO₂ reduction.³ It is well-known that the band gap of g-C₃N₄ is about 2.7 eV, which can absorb visible light up to 460 nm.⁴ Furthermore, the CB minimum (−1.12 eV vs. NHE) of g-C₃N₄ is extremely negative, so photo-generated electrons should have high reduction ability. However, the photocatalytic efficiency of the pure g-C₃N₄ is limited by the high recombination rate of its photo-generated electron–hole pairs.⁵ One of the techniques for increasing the separation efficiency of photo-generated electron–hole pairs is to form a composite photocatalyst using two kinds of semiconductors. Suitable matching of the band levels of the conduction and valence bands in the two semiconductors offers appropriate driving forces to separate and transfer photo-generated electron–hole pairs.⁶ To improve g-C₃N₄ photocatalytic activity, various semiconductor/g-C₃N₄ composite photocatalysts have been reported, such as ZnO,⁷ TiO₂,⁸ Ag₃PO₄,⁹ AgBr,¹⁰ Bi₂WO₆,¹¹ MoS₂,¹² etc., and used for the photodegradation of organic dyes in solution. However, the photocatalytic H₂ production over semiconductor/g-C₃N₄ composite photocatalysts has been proposed in limited reports.¹³

On the other hand, studies have shown that WO₃ is a visible-light responsive photocatalyst with a relatively narrow band-gap energy (2.4–2.8 eV) and a VB potential similar to that of TiO₂.¹⁴ Therefore, the oxidizing power of holes in the VB of WO₃ and TiO₂ are considered to be almost the same. However, pure WO₃ is not an efficient photocatalyst because of its low CB level, which limits the photocatalyst's ability to react with electron acceptors such as oxygen.¹⁵ The low CB level also increases the recombination of photo-generated electron–hole pairs leading to lower photocatalytic activity. Many attempts have been made to improve the photocatalytic activity of WO₃, such as noble metal loading¹⁶ and coupling with other semiconductors.¹⁷ In our previous study, the activity of WO₃ particles could be significantly improved by the binary loading of Ag and CuO as cocatalysts.¹⁸ Among them, studies have confirmed that WO₃ is a good candidate for synthesizing semiconductor heterojunctions with higher photocatalytic activity. As WO₃ and g-C₃N₄ are both visible-light-driven photocatalysts, after the polymeric g-C₃N₄ photocatalyst being combined with WO₃ the obtained WO₃/g-C₃N₄ composite may be a promising candidate for efficient photocatalytic activity under solar light irradiation. However, there are a few reports on the photocatalytic activity evaluation of WO₃/g-C₃N₄ composite.^19,20 Furthermore, to the best of our knowledge, there are no reports on the application of WO₃/g-C₃N₄ composite photocatalysts for H₂ production from aqueous solution, and no attention has been paid to the photocatalytic mechanism of the composite-catalysed reaction, which has remained unclear to date.

In this paper, different ratios of WO₃/g-C₃N₄ composite photocatalysts were synthesized via the calcination process. The photocatalysts were characterized by various techniques such as powder X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), UV-visible diffuse reflectance spectra (DRS), and so on. The photocatalytic activity was evaluated by H₂ production from triethanolamine (TEA) aqueous solution under artificial solar light. The separation mechanisms of photo-excited carriers for the composite photocatalysts were also proposed on the basis of the results for the photoluminescence (PL) analysis.

The XRD patterns of WO₃/g-C₃N₄ composites, g-C₃N₄ and WO₃ are shown in Fig. 1. For pure g-C₃N₄, two broad diffraction peaks around 27.5 and 13.0° were observed, corresponding to the (002) and (100) diffraction planes, respectively. The former, which corresponds to the interlayer distance of 0.325 nm, is attributed to the long-range interplanar stacking of aromatic units; the latter with a much weaker intensity, which corresponds to a distance d = 0.681 nm, is associated with interlayer stacking.¹ For WO₃/g-C₃N₄ composites, the XRD patterns reveal a coexistence of WO₃ and g-C₃N₄. The peak intensities of g-C₃N₄ rapidly decreased with increasing the WO₃ contents. It is difficult to confirm the diffraction peaks of g-C₃N₄ in the XRD pattern of the 30 wt% WO₃/g-C₃N₄ sample. The XRD pattern of WO₃ could be indexed as the monoclinic structure. These results along with FTIR spectra (Fig. S1†) and XPS spectra (Fig. S2†) showed clearly that WO₃/g-C₃N₄ composite photocatalysts could be synthesized by the calcination method.

Fig. 1 XRD patterns of g-C₃N₄, WO₃ and WO₃/g-C₃N₄ composite photocatalysts.

Fig. 2 shows SEM and TEM images of g-C₃N₄, WO₃ and 10 wt% WO₃/g-C₃N₄ composite samples. It is clearly seen in Fig. 2a and d that the morphology of g-C₃N₄ was smooth, thin and flat sheets. In addition, a typical porous morphology of g-C₃N₄ powders was exhibited.²¹ From Fig. 2b and e, WO₃ showed aggregated particles with the particle size of 20–150 nm. In the composite sample, WO₃ particles were sparsely observed onto the g-C₃N₄ surface. WO₃ particles did not agglomerate and were directly attached to the surface of g-C₃N₄. With increasing WO₃ content, a large number of WO₃ particles were observed on the g-C₃N₄ surface (Fig. S3a–c†). Further, HRTEM observation was conducted to investigate the interfacial structure of the composite sample (Fig. S3d†). The (020) lattice fringe of monoclinic WO₃ (0.375 nm) was clearly observed in the HRTEM image. The gray area can be ascribed to g-C₃N₄. From the SEM, TEM and HRTEM analyses, it can be concluded that the heterojunction structure was formed in the composite.

Fig. 2 SEM and TEM images of (a and d) g-C₃N₄; (b and e) WO₃; (c and f) 10 wt% WO₃/g-C₃N₄.

The UV-vis DRS of g-C₃N₄, WO₃ and all composite photocatalysts are shown in Fig. 3. For all samples, the optical absorption edge was estimated to be at around 450 nm. The composite samples displayed better photon absorption than both WO₃ and g-C₃N₄ because the composites would have high crystallinity due to the calcination at 450 °C during the composite photocatalysts preparation. The band gap can be estimated from the following equation:

αhν = A(hν − E_g)^n/2

where α, h, ν, and E_g are the absorption coefficient, Plank's constant, light frequency, and band gap, respectively, and A is a constant. The factor n depends on the characteristics of the optical transition of a semiconductor (n = 1 for direct transition and n = 4 for indirect transition). The band gaps of g-C₃N₄ and WO₃ were estimated to be 2.78 and 2.60 eV, respectively (Fig. S4†).^1,14 After coupling these two semiconductors, the band gaps of WO₃/g-C₃N₄ composite photocatalysts kept at around 2.8 eV, implying the composite photocatalysts are also responsible for the visible light region.

Fig. 3 UV-vis DRS of g-C₃N₄, WO₃ and WO₃/g-C₃N₄ composite photocatalysts.

The photocatalytic H₂ production over the all samples is shown in Fig. 4a. It revealed that the loading of WO₃ greatly influenced the photocatalytic performance of g-C₃N₄. The H₂ production rate over g-C₃N₄ was 54 μmol h⁻¹ g⁻¹. As a comparison, the composite photocatalysts showed that the H₂ production rates were lower than that on g-C₃N₄ except for 10 wt% WO₃/g-C₃N₄ photocatalyst in which the H₂ production rate was 110 μmol h⁻¹ g⁻¹, i.e., the photocatalytic activity of this composite was about 2 times higher than that of g-C₃N₄. However, the photocatalytic activity of the mechanical mixture of WO₃ and g-C₃N₄ sample with 10 wt% WO₃ content (60 μmol h⁻¹ g⁻¹) was almost the same as that of pure g-C₃N₄. The photocatalytic activity of 10 wt% WO₃/g-C₃N₄ composite was also evaluated under visible light irradiation (>420 nm). As a result, the H₂ production rate over 10 wt% WO₃/g-C₃N₄ (66 μmol h⁻¹ g⁻¹) was higher than that of g-C₃N₄ (27 μmol h⁻¹ g⁻¹) under visible light irradiation. The apparent quantum efficiencies at 405 nm were 0.34 and 0.90% for g-C₃N₄ and 10 wt% WO₃/g-C₃N₄, respectively.

Fig. 4 (a) Photocatalytic H₂ production over g-C₃N₄ and WO₃/g-C₃N₄ composite photocatalysts under artificial solar light irradiation; (b) time courses of H₂ production over g-C₃N₄ and 10 wt% WO₃/g-C₃N₄ composite photocatalyst artificial solar light irradiation.

Fig. 4b shows the time courses of H₂ production obtained over g-C₃N₄ and 10 wt% WO₃/g-C₃N₄ samples under light irradiation. For both samples, the production of H₂ steadily increased with prolonged time of light irradiation. However, the photocatalytic activity of g-C₃N₄ for H₂ production gradually decreased during the photocatalytic reaction and a total H₂ production was about 1.6 mmol g⁻¹ (36 mL g⁻¹) after 30 h. On the other hand, a total of 3.3 mmol g⁻¹ H₂ gas (74 mL g⁻¹) was produced over 10 wt% WO₃/g-C₃N₄, and no obvious deactivation of the composite photocatalyst was found, suggesting the good stability of WO₃/g-C₃N₄ as an organic–inorganic composite photocatalyst for solar H₂ production. Furthermore, no changes of the composite sample after photocatalytic reaction were observed in XRD pattern and TEM image (Fig. S5†).

To understand the higher photocatalytic activity of 10 wt% WO₃/g-C₃N₄ relative to other composites, g-C₃N₄ and WO₃, the PL spectra of the samples at 270 nm were recorded (Fig. 5). It is clear that the PL spectra of the composite and g-C₃N₄ photocatalysts have a strong emission peak at around 450 nm, which could be related to the recombination of the photo-excited electron–hole of g-C₃N₄^7,11,22 while a weak emission peak at around 460 nm of WO₃ was assigned to localized state in the band gap due to oxygen vacancies or defects.²³ From Fig. 5, it can be seen that the PL intensities of the 5, 15 and 30 wt% WO₃/g-C₃N₄ photocatalysts exhibited the stronger emission than that of the pure g-C₃N₄, suggesting that the recombination of the photo-excited electron–hole on the g-C₃N₄ photocatalyst surface is higher. On the contrary, the PL intensity of 10 wt% WO₃/g-C₃N₄ composite photocatalyst was lower than that of pure g-C₃N₄, which means that the recombination of the photo-excited electron–hole of the composite photocatalyst was lower than pure g-C₃N₄. It indicates that when the amount of WO₃ is suitable (10 wt%), the recombination of the photo-excited electron–hole on the g-C₃N₄ surface is suppressed. However, 10 wt% WO₃/g-C₃N₄ showed the highest PL intensity at around 360 nm among the all samples examined (inset of Fig. 5). The peak around at 360 nm was assigned to the electron–hole recombination on the WO₃ surface.²³ Therefore, the recombination of the charge carriers on the interface of 10 wt% WO₃/g-C₃N₄ in the composite photocatalyst would be higher than that of pure WO₃. The PL results are agreement with the results of photocatalytic H₂ production onto WO₃/g-C₃N₄ composite photocatalysts. It means that higher and lower PL intensities at 360 and 450 nm indicate a higher photocatalytic activity in the experimental conditions. The higher PL intensities at 360 nm of the samples would be attributed to the higher recombination rate between photo-excited electrons in the CB of WO₃ and photo-excited holes in the VB of g-C₃N₄ on the interface of the composite, suggesting that rich electrons in the CB of g-C₃N₄ and holes in the VB of WO₃ participate in the reduction reaction of H⁺ and the oxidation of TEA, respectively. As a result, the charge separation could be promoted on the 10 wt% WO₃/g-C₃N₄ composite, leading to the higher photocatalytic activity of the composite photocatalyst. Based on these results, it could be concluded that the WO₃/g-C₃N₄ system is a typical Z-scheme photocatalyst.

Fig. 5 PL spectra of g-C₃N₄, WO₃ and WO₃/g-C₃N₄ composite photocatalysts.

On the basis of the above results, the photocatalytic mechanism for the WO₃/g-C₃N₄ composite sample is tentatively proposed and schematically illustrated in Fig. 6. For pure g-C₃N₄, the photo-generated electrons and holes in g-C₃N₄ tend to recombine and only a fraction of them participates in the photocatalytic reaction, resulting in a relative low activity. For WO₃/g-C₃N₄ composite with an optimal WO₃ content, i.e. 10 wt%, part of the surface of g-C₃N₄ is covered by WO₃ particles, leading to the formation of Z-scheme photocatalytic system. According to previous studies, the CB and VB positions of WO₃ are about +0.74 and +3.4 V, respectively,¹⁵ while those of g-C₃N₄ are about −1.13 and +1.57 V, respectively.²² Further, the band structure of WO₃/g-C₃N₄ used in this study can be also estimated according to the empirical equations as shown below:

E_VB = χ − E^e + 0.5E_g

E_CB = E_VB − E_g

where E_VB and E_CB are the valence and conduction band edge potentials, respectively; χ is the electronegativity of the semiconductor, which is the geometric mean of the electronegativity of the constituent atoms; E^e is the energy of free electrons on the hydrogen scale (about 4.5 eV vs. NHE). The estimated CB and VB positions for WO₃ and g-C₃N₄ are shown in Fig. 6 and were similar values to those of previous studies.^15,22 Thus, under artificial solar light irradiation, the photo-induced holes tend to keep in the VB of WO₃, while the electrons in CB of WO₃ combine with the holes in VB of g-C₃N₄ at the interface of the composite. The electrons in the VB of g-C₃N₄ are further excited to its CB. This results in an efficient space separation of the photo-induced charge carriers. Then, the electrons stored in the CB of g-C₃N₄ react with H⁺ in water near the surface of the photocatalyst to produce H₂ gas while the holes in the VB of WO₃ oxidize TEA molecules. The above hydrogen production experiments and PL analysis would support the Z-scheme photocatalytic mechanism. If the charge carriers of WO₃/g-C₃N₄ photocatalyst transfer according to the conventional electron–hole separation process for a great number of composite photocatalysts; the electrons in the CB of g-C₃N₄ would migrate to the CB of WO₃, and holes in the VB of WO₃ would transfer to the VB of g-C₃N₄. This can result from the efficient charge separation of the photo-induced charge carriers. However, the electrons in the CB of WO₃ cannot produce H₂ gas from water because the CB potential of WO₃ is positive than that of the H₂/H⁺ couple. In addition, the oxidation power of the composite photocatalyst decreases because the VB potential of g-C₃N₄ is relatively low. This would lead that the WO₃/g-C₃N₄ composite has a lower reduction/oxidation ability and photocatalytic activity than pure g-C₃N₄ and WO₃. Therefore, it can be concluded that the photocatalytic mechanism of WO₃/g-C₃N₄ composite is not accordance with the traditional charge separation process. Namely, a typical Z-scheme photocatalyst is favourable for the production of H₂ gas from water. The Z-scheme mechanism of WO₃/g-C₃N₄ composite photocatalyst has been also reported by the other researchers.²⁰

Fig. 6 Schematic diagram of Z-scheme photocatalytic mechanism of WO₃/g-C₃N₄ composite photocatalyst.

In this study, the unsuitable contents of WO₃ in WO₃/g-C₃N₄ photocatalyst showed low photocatalytic activity for H₂ production. Unfortunately, the reasons are not fully understood currently. We need the further studies to be clear the more detail photocatalytic mechanism of WO₃/g-C₃N₄ composite.

Conclusions

In summary, the composite photocatalyst WO₃/g-C₃N₄ was fabricated via a simple calcination method. The highest photocatalytic activity was achieved for 10 wt% WO₃/g-C₃N₄ composite at H₂ production rate of 110 μmol h⁻¹ g⁻¹, which was about 2 times higher compared to that of pure g-C₃N₄. Furthermore, the composite showed a good stability for the repeated H₂ production reaction. The enhanced activity was due to the formation of WO₃/g-C₃N₄ Z-scheme photocatalytic system and an efficient charge separation of the photo-generated electron–hole pairs. This present study would provide new insights on enhancing the photocatalytic H₂ production activity of g-C₃N₄ by the formation of Z-scheme WO₃/g-C₃N₄ composite photocatalyst.

Acknowledgements

This work was partly supported by Grant-in-Aid for Scientific Research (C) no. 24510095 from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

Notes and references

1. X. C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76; Q. J. Xiang, J. G. Yu and M. Jaroniec, J. Phys. Chem. C, 2011, 115, 7355; J. G. Yu, S. H. Wang, B. Cheng, Z. Lin and F. Huang, Catal. Sci. Technol., 2013, 3, 1782.
2. Y. J. Cui, Z. X. Ding and X. Z. Fu, Angew. Chem., Int. Ed., 2012, 51, 11814.
3. G. H. Dong and L. Z. Zhang, J. Mater. Chem., 2012, 22, 1160; J. Mao, T. Peng, X. Zhang, K. Li, L. Ye and L. Zan, Catal. Sci. Technol., 2013, 3, 1253.
4. X. Wang, K. Maeda, X. Chen, K. Takanabe, K. Domen, Y. Hou, X. Fu and M. Antonietti, J. Am. Chem. Soc., 2009, 131, 1680; Y. Wang, X. Wang and M. Antonietti, Angew. Chem., Int. Ed., 2012, 51, 68.
5. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2009, 25, 10397.
6. J. Zhang, M. Zhang, R. Q. Sun and X. Wang, Angew. Chem., Int. Ed., 2012, 51, 10145.
7. Y. J. Wang, R. Shi, J. Lin and Y. F. Zhu, Energy Environ. Sci., 2011, 4, 2922; J. X. Sun, Y. P. Yuan, L. G. Qiu, X. Jiang, A. J. Xie, Y. H. Shen and J. F. Zhu, Dalton Trans., 2012, 41, 6756.
8. X. Lu, Q. Wang and D. Cui, J. Mater. Sci. Technol., 2010, 26, 925.
9. S. Kumar, T. Surendar, A. Baruah and V. Shanker, J. Mater. Chem. A, 2013, 1, 5333.
10. H. Xu, J. Yan, Y. G. Xu, Y. H. Song, H. M. Li, J. X. Xia, C. J. Huang and H. L. Wan, Appl. Catal., B, 2013, 129, 182; S. Yang, W. Zhou, C. Ge, X. Liu, Y. Fang and Z. Li, RSC Adv., 2013, 3, 5631.
11. L. Ge, C. Han and J. Liu, Appl. Catal., B, 2011, 108/109, 100.
12. W. C. Peng and X. Y. Li, Catal. Commun., 2014, 49, 63.
13. B. Chai, T. Peng, J. Mao, K. Li and L. Zan, Phys. Chem. Chem. Phys., 2012, 14, 16745; Y. D. Hou, A. B. Laursen, J. S. Zhang, G. G. Zhang, Y. S. Zhu, X. C. Wang, S. Dahl and I. Chorkendorff, Angew. Chem., Int. Ed., 2013, 52, 3621; J. Chen, S. Shen, P. Guo, P. Wu and L. Guo, J. Mater. Chem. A, 2014, 2, 4605.
14. M. Miyauchi, M. Shibuya, Z. G. Zhao and Z. Liu, J. Phys. Chem. C, 2009, 113, 10642.
15. M. Miyauchi, Phys. Chem. Chem. Phys., 2008, 10, 6258.
16. T. Arai, M. Horiguchi, M. Yanagida, T. Gunji, H. Sugihara and K. Sayama, Chem. Commun., 2008, 5565; Q. Xiang, G. F. Meng, H. B. Zhao, Y. Zhang, H. Li, W. J. Ma and J. Q. Xu, J. Phys. Chem. C, 2010, 114, 2049.
17. D. Bi and Y. Xu, Langmuir, 2011, 27, 9359; H. Widiyandari, A. Purwanto, R. Balgis, T. Ogi and K. Okuyama, Chem. Eng. J., 2012, 180, 323.
18. H. Katsumata, Y. Oda, S. Kaneco and T. Suzuki, RSC Adv., 2013, 3, 5028.
19. L. Huang, H. Xu, Y. Li, H. Li, X. Cheng, J. Xia, Y. Xu and G. Cai, Dalton Trans., 2013, 42, 8606; Y. Zang, L. Li, Y. Zuo, H. Lin, G. Li and X. Guan, RSC Adv., 2013, 3, 13646; K. Katsumata, R. Motoyoshi, N. Matsushita and K. Okada, J. Hazard. Mater., 2013, 260, 475.
20. Z. Jin, N. Murakami, T. Tsubota and T. Ohno, Appl. Catal., B, 2014, 150/151, 479; S. Chen, Y. Hu, S. Meng and X. Fu, Appl. Catal., B, 2014, 150/151, 564.
21. Y. Zhang, J. Liu, G. Wu and W. Chen, Nanoscale, 2012, 4, 5300.
22. S. C. Yan, S. B. Lv, Z. S. Li and Z. G. Zou, Dalton Trans., 2010, 39, 1488.
23. K. Lee, W. S. Seo and J. T. Park, J. Am. Chem. Soc., 2003, 125, 3408.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details and Fig. S1–S5. See DOI: 10.1039/c4ra02511c

---