Novel g-C₃N₄/BiIO₄ heterojunction photocatalysts: synthesis, characterization and enhanced visible-light-responsive photocatalytic activity

Abstract

The g-C₃N₄/BiIO₄ composite photocatalysts were successfully synthesized using a simple-mixed-calcinations method, and their photocatalytic activities for degradation of rhodamine B (RhB) under visible-light (λ > 420 nm) were investigated for the first time. The crystal structure and optical property of the as-synthesized samples were characterized by XRD, FTIR, SEM, TEM, HRTEM and DRS spectroscopy. The photodegradation experiments indicated that the g-C₃N₄/BiIO₄ composite photocatalyst displays a higher photocatalytic activity than the two individuals, which was also confirmed by the PL spectra and photoelectrochemical experiments. This remarkably improved photocatalytic performance can be attributed to the heterojunction structure of g-C₃N₄/BiIO₄ composites, which possess stronger oxidation and reduction capability, thus resulting in the efficient separation of photoinduced charge carriers as demonstrated in the active species experiments. The present study will be beneficial for the design of high performance photocatalysts.

---

1. Introduction

Visible-light photocatalysis has attracted worldwide interest, as it provides a green and potential route for wide applications in hydrogen evolution and environmental purification.^1–6 Studies on the development of efficient and visible-light-driven semiconductor photocatalysts have received increasing attention and have been a worldwide endeavor. To date, the major strategies for improving the photocatalytic activity of the semiconductor photocatalysts are the modification of TiO₂,^7–9 combination of semiconductor photocatalysts^10–12 and exploitation of novel photocatalytic materials.^13–15 Among the strategies, the composite semiconductor photocatalysts have higher photocatalytic activity because they can largely reduce the recombination and enhance the separation rate of photogenerated charge carriers. Therefore, many researches focus on the preparation of composite photocatalysts.

Graphitic carbon nitride (g-C₃N₄) is a metal-free polymeric photocatalyst, with a graphitic stacking structure of the g-C₃N₄ layers consisting of tristriazine units connected through amino groups.^16,17 This material can be successfully activated by visible light irradiation and exhibits high photocatalytic activity for H₂ evolution from water¹⁸ and degradation of organic pollutants.¹⁹ However, the photocatalytic performance of g-C₃N₄ has been restricted because of its low efficiency, mainly because of the fast charge recombination. Nowadays, various methods have been developed to improve the photocatalytic activity, such as designing novel nanostructures,^20,21 doping with non-metal or metal elements,^22,23 building heterostructures.^24–26 Among these, the construction of heterostructures, combining g-C₃N₄ with other appropriate semiconductors, is an effective and sustainable method to improve the photocatalytic activity.

Nowadays, Bi-based compounds have attracted considerable attention for their potential application as novel photocatalysts. Moreover, all the Bi-based photocatalysts exhibit high efficiency in the degradation of organic pollutants and interesting structure–property relationships because of the existence of an active (Bi₂O₂)²⁺ layer.¹⁵ BiIO₄ is a new bismuth iodate, as a nonlinear optical material, recently synthesized by Nguyen et al.²⁷ It also exhibits a layered structural topology. Instead of a perovskite-like anion block separating the (Bi₂O₂)²⁺ layers in Aurivillius phases, the locally polar iodate (IO₃)⁻ anions are observed between (Bi₂O₂)²⁺ layers in the structure of BiIO₄.⁴ The layered configuration will also benefit the charge transfer. After investigating the energy levels of g-C₃N₄ and BiIO₄, we found that the E_CB and E_VB for g-C₃N₄ are estimated to be −1.13 eV and 1.57 eV, while those of BiIO₄ are calculated to be 0.85 eV and 3.84 eV, respectively. Such that, the energy levels of g-C₃N₄ and BiIO₄ are efficiently matched overlapping band-structures. Therefore, it is possible to construct a heterojunction between g-C₃N₄ and BiIO₄ with a high visible-light photocatalytic activity.

In the present work, we successfully synthesized the heterostructured g-C₃N₄/BiIO₄ composites using a simple mixed calcinations method and their photocatalytic performance on rhodamine B degradation under visible light irradiation was studied. The g-C₃N₄/BiIO₄ heterojunction composites result in improved separation efficiency of the charge carriers, which leads to enhanced photocatalytic activity for RhB degradation and photocurrent generation. The mechanism for the enhancement of photocatalytic activity over the g-C₃N₄/BiIO₄ composites were investigated in detail.

2. Experimental section

2.1 Preparation of the photocatalyst

All the chemicals were of analytical grade, and used as received without further purification. The g-C₃N₄ was synthesized according to the literature method.²⁸ In a typical procedure, melamine (10 g) was placed in a crucible without a cover at ambient pressure under nitrogen. Then, the precursor was heated to 520 °C at a heating rate of 3 °Cmin⁻¹ in a tube furnace for 4 h under nitrogen. The yellow product obtained was collected and ground into powder after the muffle furnace was naturally cooled to room temperature.

BiIO₄ was obtained by a hydrothermal method. In a typical procedure, 0.003 mol Bi(NO₃)₃·5H₂O was added to 15 mL deionized water and the breaker was placed in an ultrasonic bath for 10 min to dissolve the raw materials. Moreover, a certain amount of I₂O₅ was dissolved in 15 mL deionized water to obtain a clear solution. Then, the solution was added to the suspension and subsequently stirred for another 30 min at room temperature. The resulting suspension was subsequently transferred into a 50 mL Teflon-lined stainless autoclave and heated at 180 °C for 24 h. After cooling, the products were collected by filtration and washed repeatedly with deionized water and ethanol and then dried at 60° for 12 h.

The g-C₃N₄/BiIO₄ composite was prepared according to the following procedure: the synthesis of g-C₃N₄/BiIO₄ with molar ratio 8 : 1, 0.008 mol of g-C₃N₄ and 0.001 mol of BiIO₄ were mixed and ground in an agate mortar for 10 min. Then, the mixture was calcined at 400 °C for 4 h to obtain the 8 : 1 g-C₃N₄/BiIO₄ photocatalyst. According to this method, different molar ratios of g-C₃N₄/BiIO₄ at 6 : 1 and 10 : 1 were prepared, respectively.

2.2 Characterization

The structural properties of the samples were characterized at room temperature using X-ray powder diffraction (XRD) on a Bruker D8 focus using Cu Kα radiation (40 kV/40 mA). The morphologies of the products were observed by field emission scanning electron microscopy (FE-SEM) (Hitachi S-4800, Japanese), transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) (JEM-2100, Japanese). Fourier-transform infrared (FTIR) spectra were obtained using a Bruker spectrometer in the frequency range of 2000–450 cm⁻¹. The UV-vis diffuse reflectance spectra (DRS) of catalysts were obtained using a Cary 5000 UV-vis spectrometer (America Varian) equipped with an integrating sphere. The Brunauer–Emmett–Teller (BET) specific surface areas of the samples were analyzed by nitrogen adsorption–desorption (Micromeritics ASAP 2460, USA). The photoluminescence (PL) spectra were recorded on a Hitachi F-4600 fluorescence spectrophotometer, using a Xe lamp (excitation at 365 nm) as light source.

2.3 Evaluation of photocatalytic activity under visible light

The visible-light photocatalytic activities of g-C₃N₄/BiIO₄ composites were evaluated by the oxidation of RhB in an aqueous solution at room temperature. The visible-light source was a 500 W Xenon lamp. For the photocatalytic test, a glass optical filter was inserted to cut off the short wavelength components (λ < 420 nm). The photocatalytic activities of the g-C₃N₄/BiIO₄ composites can be quantitatively evaluated by comparing the apparent reaction rate constants. During photocatalytic degradations of dyes in the liquid phase, 50 mg photocatalyst was suspended in 50 mL RhB solution (1 × 10⁻⁵ mol L⁻¹) in a quartz tube under visible-light irradiation. Before irradiation, the suspensions were stirred for 60 min in the dark to ensure the establishment of the adsorption–desorption equilibrium. 3.0 mL liquid was taken at a certain time interval during the experiment and centrifuged to remove the catalyst. The filtrates were analyzed on a Varian UV-vis spectrophotometer (Shimadzu UV-5500PC).

2.4 Active species trapping experiment

In order to detect the active species in the photocatalytic process, a quantity of different appropriate species quenchers were introduced into the photocatalytic degradation process of RhB in a manner similar to the photodegradation experiment. The dosages of these species quenchers were referred to the previous studies.²⁸

2.5 Photoelectrochemical measurement

The photocurrent measurement were conducted on a Chi 660B electrochemical system (Shanghai, China) using a standard three-electrode cell with a working electrode, a platinum wire as a counter electrode, and a standard calomel electrode (SCE) as a reference electrode. The working electrodes are g-C₃N₄, BiIO₄ and 8 : 1 g-C₃N₄/BiIO₄ film electrodes. Na₂SO₄ (0.1 M) was used as the electrolyte solution.

3. Results and discussion

3.1. Characterization

3.1.1 XRD analysis. The detailed information on the purity and crystallinity of the as-prepared samples was obtained in the XRD measurement. Fig. 1 shows the XRD patterns of the g-C₃N₄, BiIO₄ and g-C₃N₄/BiIO₄ composites with different molar ratios. It can be seen that all the samples were well crystallized and all the diffraction peaks of g-C₃N₄ could be well indexed to the g-C₃N₄ peaks (JCPDS 87-1526). Two distinct diffraction peaks at 27.40° and 13.04° can be assigned to the corresponding (002) and (100) diffraction planes of the g-C₃N₄.^23,24,28 All the diffraction peaks of the BiIO₄ can be indexed into the pure orthorhombic BiIO₄ (ICSD # 262019) and the strongest peak of BiIO₄ is assigned to (121) plane,²⁹ which is in good agreement with the following HRTEM analyses: no impurity peaks were observed, which implies that the final products of g-C₃N₄ and BiIO₄ were of pure phases. Because of the low intensity of the g-C₃N₄, there are no obvious peaks that can be detected in g-C₃N₄/BiIO₄ composites. However, the narrow sharp peaks suggest that the g-C₃N₄/BiIO₄ products are highly crystalline.

Fig. 1 XRD pattern of g-C₃N₄, BiIO₄ and g-C₃N₄/BiIO₄ composites.

3.1.2 SEM and TEM analysis. The morphology and microstructure of the samples were investigated by SEM, TEM and HRTEM, as shown in Fig. 2 and 3. For pure g-C₃N₄ (Fig. 2a), the obtained samples were composed of stacking layers. Fig. 2b presents the SEM micrograph of BiIO₄ that shows a different behavior with anomalous smooth nanorods structure, and the length of these nanorods ranges from 300 nm to 4 μm, while the width was in the range of 50 nm to 1 μm. After chemical reaction, parts of the BiIO₄ nanorods were assembled on the surface of the layered g-C₃N₄ (Fig. 2c), and the other parts of BiIO₄ nanorods were covered by the g-C₃N₄, shown in Fig. 2d. These results show that the g-C₃N₄/BiIO₄ heterostructure composites were obtained.

Fig. 2 SEM images of the samples: (a) g-C₃N₄, (b) BiIO₄, (c) and (d) 8 : 1 g-C₃N₄/BiIO₄.

Fig. 3 TEM images of the samples: (a) g-C₃N₄, (b) BiIO₄, (c) 8 : 1 g-C₃N₄/BiIO₄, and (d) HRTEM images of the 8 : 1 g-C₃N₄/BiIO₄.

To obtain a better understanding of the interfacial interaction between g-C₃N₄ and BiIO₄, the g-C₃N₄/BiIO₄ heterostructure and pure samples were further characterized by TEM and HRTEM. As displayed in Fig. 3a and b, the layered structure of g-C₃N₄ and the nanorod structure of BiIO₄ were confirmed, respectively. Fig. 3c presents the TEM image of the 8 : 1 g-C₃N₄/BiIO₄ composites, in which the firmly constructed g-C₃N₄/BiIO₄ heterojunction is clearly observed. Furthermore, clear fringe with an interval of 0.326 nm form Fig. 3d can be indexed into the (121) lattice plane of orthorhombic BiIO₄. Thus, it can be inferred that the interfacial interaction may mainly happen on the (121) and (002) facets of BiIO₄ and g-C₃N₄, respectively.

3.1.3 FTIR analysis. Fig. 4 displays the FTIR spectra of g-C₃N₄, BiIO₄ and g-C₃N₄/BiIO₄ to investigate the chemical bonding of the as-prepared samples. The characteristic vibration peak of the pure g-C₃N₄ appears at wavenumbers of 808, 1245, 1322, 1574, and 1633 cm⁻¹. The peak at 808 cm⁻¹ was assigned to the triazine units, and the stretching vibration of C–N heterocyclics was reflected at 1245 to 1574 cm⁻¹. The peak at 1322 cm⁻¹ and 1633 cm⁻¹ corresponds to the C–N and C=N stretching mode, respectively.³⁰ As for the BiIO₄, the intensive peaks at about 419 and 517 cm⁻¹ were attributed to the stretching vibrations of the Bi–O in BiIO₄, while the peaks at about 681 and 764 cm⁻¹ were assigned to the stretching vibrations of I–O,²⁹ respectively. All the FTIR spectra exhibit the characteristic peaks of both g-C₃N₄ and BiIO₄, and no impurity or solvent residue absorption peaks were detected. The above results demonstrated the successful combination of the two components.

Fig. 4 FTIR spectra of the g-C₃N₄, BiIO₄ and g-C₃N₄/BiIO₄ composites.

3.1.4 UV-vis DRS analysis. The optical absorption of the as-prepared g-C₃N₄/BiIO₄ samples is shown in Fig. 5. The pure g-C₃N₄ sample can absorb both UV and visible light with an absorption edge at 460 nm, which can be assigned to the intrinsic band gap of g-C₃N₄ (2.7 eV). The pure BiIO₄ also has slight light absorption in the visible region with a band edge 415 nm. Compared to BiIO₄, all the g-C₃N₄/BiIO₄ composites exhibit a red shift and a shoulder on the adsorption edge that reaches further out in the visible region. These observations are attributed to the interaction between g-C₃N₄ and BiIO₄ in the composite samples. The covered spectral range increases with an increase in the g-C₃N₄ content. The total absorption of the composite samples increases the production of electron–hole pairs. As a result, this may lead to a higher photocatalytic activity.

Fig. 5 UV-vis diffuse reflectance spectra of the g-C₃N₄, BiIO₄ and g-C₃N₄/BiIO₄ composites.

The band gap is determined by optical absorption near the band edge by the following equation:

αhv = A (hv − E_g)^n/2 (1)

where α, h, v, E_g, and A are the optical absorption coefficient, Plank constant, light frequency, band gap, and a constant, respectively.⁶ In the equation, n is 1 for direct transitions and 4 for indirect transitions. Both, g-C₃N₄ and BiIO₄ possess indirect transition band gaps^4,25 and the value of n equals 4. By extrapolating the straight line to the x-axis in this plot, the E_g of g-C₃N₄ and BiIO₄ was estimated to 2.70 eV and 2.99 eV (as shown in the figure insert in the Fig. 5), respectively. Furthermore, we can calculate their conduction and valance band positions through the following eqn (2) and (3):⁶

E_VB = X − E_e + 0.5E_g (2)

E_CB = E_VB − E_g (3)

where X is the electronegativity of the semiconductor, E_e is the energy of free electrons on the hydrogen scale (E_e = 4.5 eV), and E_g is the band gap energy of the semiconductor. For g-C₃N₄, the X is calculated to be 4.67 eV. The E_CB and E_VB are estimated to be −1.13 eV and 1.57 eV, respectively. The X of BiIO₄ is calculated to be 6.84 eV and the E_CB and E_VB are estimated to be 0.85 eV and 3.84 eV, respectively.

3.1.5 BET surface area. The BET surface area of the 8 : 1 g-C₃N₄/BiIO₄ photocatalyst has been measured by nitrogen adsorption–desorption. The specific surface area of 8 : 1 g-C₃N₄/BiIO₄ composite is 4.24 m² g⁻¹, which is a little higher than bulk g-C₃N₄ (4.0 m² g⁻¹), suggesting that there may be more adsorption sites for O₂ molecule in the 8 : 1 g-C₃N₄/BiIO₄ composite. It may also contribute to the enhancement of photocatalytic activity.

3.2. Photocatalytic activities of g-C₃N₄/BiIO₄ heterostructures

The photocatalytic activities of g-C₃N₄/BiIO₄ composites were measured by the degradation of RhB in water under visible light irradiation as shown in Fig. 6a. For the purpose of comparison, the direct photolysis of RhB in the absence of the photocatalyst was tested under the same conditions. The direct photolysis of RhB from the blank experiment could almost be neglected. As shown in Fig. 6a, the 8 : 1 g-C₃N₄/BiIO₄ composite exhibits the highest photocatalytic activity. The degradation efficiency reaches 54% after 5 h irradiation compared to 29%, 3%, 31%, 33% and 25% for g-C₃N₄, BiIO₄, 10 : 1 g-C₃N₄/BiIO₄, 6 : 1 g-C₃N₄/BiIO₄, 8 : 1 MM g-C₃N₄/BiIO₄ composites, respectively. The photocatalytic degradation of organic pollutants generally follows the pseudo-first-order kinetics. As shown in Fig. 6b, the 8 : 1 g-C₃N₄/BiIO₄ heterojunction possesses the optimum activity among all the samples. Its apparent rate constant is 0.140 min⁻¹, which is about 2.5, 23.3 and 2.8 times as high as those of g-C₃N₄ (k = 0.055 min⁻¹), BiIO₄ (k = 0.006 min⁻¹) and 8 : 1 MM (k = 0.050 min⁻¹), respectively. It has been further proved that the enhancement of the visible light photocatalytic activity in 8 : 1 g-C₃N₄/BiIO₄ composite should be attributed to the more reliable interfacial interaction of g-C₃N₄ and BiIO₄. Fig. 6c shows the time-dependent absorption spectra of RhB solution in the presence of g-C₃N₄/BiIO₄ heterostructure. The shift of maximum absorption of RhB solution from 554 to 533 nm corresponds to the N-demethylation processes occurred during the photocatalytic reaction.

Fig. 6 (a) Photocatalytic degradation curves of RhB over the various samples; (b) apparent rate constants for the photodegradation of RhB under the irradiation of visible-light (λ > 420 nm); (c) UV-visible spectra of RhB at different visible irradiation times in the presence of 8 : 1 g-C₃N₄/BiIO₄.

It is generally accepted that the main reactive species including h⁺, ˙OH and ˙O₂⁻ are involved in the photocatalytic oxidation process. Therefore, the effects of some scavengers on the degradation of RhB were examined in an attempt to elucidate the reaction mechanism. As an ˙O₂⁻ scavenger, benzoquinone (BQ) was added to the reaction system, isopropanol (IPA) was introduced as the scavenger of ˙OH, and ethylenediamine tetra acetic acid (EDTA) was adopted to quench h⁺.^31–34 The effects of a series of scavengers on the degradation efficiency of RhB are shown in Fig. 7. The degradation efficiency of RhB for the 8 : 1 g-C₃N₄/BiIO₄ photocatalyst was greatly inhibited after adding EDTA, indicating h⁺ is the main active species in the photocatalytic process. When BQ and IPA are added, the photocatalytic degradation efficiencies of dye also decrease to 24.9% and 26.7%, respectively. It is suggested that ˙O₂⁻and ˙OH play an equally important role in the photocatalytic reaction. In summary, the main reactive species involved in the photocatalytic degradation of RhB are h⁺, ˙O₂⁻and ˙OH.

Fig. 7 Photocatalytic degradation of RhB over the 8 : 1 g-C₃N₄/BiIO₄ photocatalyst alone and with the addition of BQ, IPA or EDTA.

3.3. Mechanism of the improved photocatalytic activity of the g-C₃N₄/BiIO₄ heterostructure

The separation efficiency of the photogenerated electrons and holes for the g-C₃N₄/BiIO₄ composite was investigated by photoluminescence at an excitation wavelength of 365 nm. It is well known that the higher PL emission intensity indicates the higher recombination efficiency of the photogenerated carriers, and thus the lower photocatalytic activity.³⁵ Fig. 8 shows the PL spectra of the g-C₃N₄, 8 : 1 g-C₃N₄/BiIO₄ composites and the mechanically-mixed 8 : 1 g-C₃N₄/BiIO₄ samples. Pure g-C₃N₄ exhibits the highest emission intensity, centered at about 459 nm, which is approximately equal to its optical band gap. It is important to note that the emission intensity of the 8 : 1 g-C₃N₄/BiIO₄ heterostructure is lower than that of the mechanically-mixed g-C₃N₄/BiIO₄ sample, indicating that the recombination rate of the photogenerated charge carriers is truly lower in the g-C₃N₄/BiIO₄ composite. The PL results confirm the importance of the heterojunctions in hindering the recombination of electrons and holes.

Fig. 8 Photoluminescence spectra of the g-C₃N₄, 8 : 1 g-C₃N₄/BiIO₄ and mechanically mixed 8 : 1 g-C₃N₄/BiIO₄ samples excited by wavelength of 365 nm visible-light.

It was also widely known that the separation efficiency of electrons and holes played vital roles in the photocatalytic reaction.³⁶ The photocurrent was generated from the photogenerated electrons in the conduction bands of semiconductor photocatalysts, leaving holes in their valence bands. Therefore, the higher the photocurrent, the better the electron and hole separation efficiency would be, and thus the higher the photocatalytic activity would be.^36,37 The transient photocurrent responses of the g-C₃N₄, BiIO₄ and 8 : 1 g-C₃N₄/BiIO₄ heterostructure electrodes were recorded for several on-off cycles of visible light irradiation, as shown in Fig. 9. The 8 : 1 g-C₃N₄/BiIO₄ sample shows increased current as compared to g-C₃N₄ and BiIO₄. It demonstrated that g-C₃N₄/BiIO₄ heterostructure can effectively reduce the recombination of photogenerated electrons and holes and produce longer living photogenerated carriers. Therefore, the result of photocurrent-time measurement suggests that 8 : 1 g-C₃N₄/BiIO₄ heterogeneous photocatalyst has a stronger ability to separate electron–hole pairs than pure g-C₃N₄ and BiIO₄.

Fig. 9 Comparison of transient photocurrent response of the g-C₃N₄, BiIO₄ and 8 : 1 g-C₃N₄/BiIO₄.

According to the above results, the detailed photocatalytic mechanism of the enhanced photocatalytic activity was proposed (shown in Fig. 10) based on the energy band structures of g-C₃N₄/BiIO₄. From Fig. 7, holes are the main active species in the process of RhB degradation because the photoexcited holes in the VB of BiIO₄ were easily transferred to the VB of g-C₃N₄ under visible-light irradiation. However, the gap (1.98 eV) between CB of g-C₃N₄ and CB of BiIO₄ is considerably larger than that (0.72 eV) between CB of BiIO₄ and VB of g-C₃N₄, which may delay the electrons transference leaving some electrons in the CB of g-C₃N₄. The electrons left in the CB of g-C₃N₄ have more negative potential to generate active ˙O₂⁻ radicals with powerful reduction. Moreover, the gap (2.27 eV) between VB of g-C₃N₄ and VB of BiIO₄ is also larger than that (0.72 eV) between CB of BiIO₄ and VB of g-C₃N₄, which can similarly delay the holes transference leaving parts of holes in the VB of BiIO₄, resulting in more positive potential holes to oxidize H₂O to yield ˙OH radicals. Therefore, RhB is photodegraded mainly through direct h⁺ oxidation, and secondarily through a ˙O₂⁻ and ˙OH pathway corresponding to the results of the active species experiments. As a result, the photocatalytic activity of g-C₃N₄/BiIO₄ composites is much higher than that of the single g-C₃N₄ and BiIO₄.

Fig. 10 Schematic diagram of electron–hole pairs separation and the possible reaction mechanism over the g-C₃N₄/BiIO₄ photocatalyst under visible light irradiation.

4. Conclusions

The visible-light-driven composite photocatalyst g-C₃N₄/BiIO₄ was synthesized by a simple mixed-calcinations method for the first time. All the as-synthesized g-C₃N₄/BiIO₄ composites exhibited improved photocatalytic performance than that of pure samples. Among which, the 8 : 1 g-C₃N₄/BiIO₄ composite possesses the highest photocatalytic activity for RhB degradation under visible-light irradiation. The detailed active species trapping experiments of h⁺, ˙O₂⁻ and ˙OH production over g-C₃N₄/BiIO₄ composite photocatalyst revealed that the highly improved photocatalytic activity should be ascribed to the fabrication of g-C₃N₄/BiIO₄ heterostructure, thus resulting in the more efficient separation and transfer of charge and the largely reduced probability of recombination. The present study will provide new insights into the design of high-performance photocatalysts.

Acknowledgements

This work was supported by the National Natural Science Foundations of China (Grant no. 51302251), the Fundamental Research Funds for the Central Universities (2652013052), and the National High Technology Research and Development Program (863 Program 2012AA06A109) of China.

Notes and references

1. S. W. Liu, J. G. Yu and M. Jaroniec, Chem. Mater., 2011, 23, 4085.
2. Q. J. Xiang, J. G. Yu and M. Jaroniec, Chem. Soc. Rev., 2012, 41, 782.
3. H. W. Huang, L. J. Liu, S. F. Jin, W. J. Yao, Y. H. Zhang and C. T. Chen, J. Am. Chem. Soc., 2013, 135, 18319.
4. H. W. Huang, S. B. Wang, N. Tian and Y. H. Zhang, RSC Adv., 2014, 4, 5561.
5. H. W. Huang, H. J. Qi, Y. He, N. Tian and Y. H. Zhang, J. Mater. Res., 2013, 28, 2977.
6. N. Tian, H. W. Huang, Y. H. Zhang and Y. He, J. Mater. Res., 2014, 29, 641.
7. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269.
8. D. V. Bavykin, A. A. Lapkin, P. K. Plucinski, L. Torrente-Murciano, J. M. Friedrich and F. C. Walsh, Top. Catal., 2006, 39, 151.
9. J. M. Feng, J. J. Han and X. J. Zhao, Prog. Org. Coat., 2009, 64, 268.
10. L. Spanhel, H. Weller and A. Henglein, J. Am. Chem. Soc., 1987, 109, 6632.
11. S. Eibl, B. C. Gates and H. Knozinger, Langmuir, 2001, 17, 107.
12. M. C. Long, W. M. Cai, J. Cai, B. X. Zhou, X. Y. Chai and Y. H. Wu, J. Phys. Chem. B, 2006, 110, 20211.
13. X. Xiao, C. L. Xing, G. P. He, X. X. Zuo, J. M. Nan and L. S. Wang, Appl. Catal., B, 2014, 148–149, 154.
14. H. W. Huang, J. Y. Yao, Z. S. Lin, X. Y. Wang, R. He, W. J. Yao, N. X. Zhai and C. T. Chen, Angew. Chem., Int. Ed., 2011, 50, 9141.
15. H. W. Huang, Y. He, Z. S. Lin, L. Kang and Y. H. Zhang, J. Phys. Chem. C, 2013, 117, 22986.
16. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76.
17. A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J. O. Müller, R. Schlögl and J. M. Carlsson, J. Mater. Chem., 2008, 18, 4893.
18. K. Schwinghammer, B. Tuffy, M. B. Mesch, E. Wirnhier, C. Martineau, F. Taulelle, W. Schnick, J. Senker and B. V. Lotsch, Angew. Chem., Int. Ed., 2013, 52, 2435.
19. Y. Cui, Z. Ding, P. Liu, M. Antonietti, X. Fu and X. Wang, Phys. Chem. Chem. Phys., 2012, 14, 1455.
20. L. Chen, D. Huang, S. Ren, T. Dong, Y. Chi and G. Chen, Nanoscale, 2013, 5, 225.
21. P. Niu, L. Zhang, G. Liu and H. Cheng, Adv. Funct. Mater., 2012, 22, 4763.
22. G. Liu, P. Niu, C. Sun, S. C. Smith, Z. Chen, G. Lu and H. Cheng, J. Am. Chem. Soc., 2010, 132, 11642.
23. X. F. Song, H. Tao, L. X. Chen and Y. Sun, Mater. Lett., 2014, 116, 265.
24. D. M. Chen, K. W. Wang, D. G. Xiang, R. L. Zong, W. Q. Yao and Y. F. Zhu, Appl. Catal., B, 2014, 147, 554.
25. Y. X. Yang, W. Guo, Y. N. Guo, Y. H. Zhao, X. Yuan and Y. H. Guo, J. Hazard. Mater., 2014, 271, 150.
26. X. J. Wang, Q. Wang, F. T. Li, W. Y. Yang, Y. Zhao, Y. J. Hao and S. J. Liu, Chem. Eng. J., 2013, 234, 361.
27. S. D. Nguyen, J. Yeon, S. H. Kim and P. S. Halasyamani, J. Am. Chem. Soc., 2011, 133, 12422–12425.
28. X. W. Wang, G. Liu, Z. G. Chen, F. Li, L. Z. Wang, G. Q. Lu and H. M. Cheng, Chem. Commun., 2009, 3452.
29. Z. B. Cao, Y. C. Yue and Z. G. Hu, J. Synth. Cryst., 2011, 4, 858.
30. Y. P. Zang, L. P. Li, Y. Zuo, H. F. Lin, G. S. Li and X. F. Guan, RSC Adv., 2013, 3, 13646.
31. M. Sun, D. Z. Li, W. J. Zhang, Z. X. Chen, H. J. Huang, W. J. Li, Y. H. He and X. Z. Fu, J. Solid State Chem., 2012, 190, 135.
32. W. J. Li, D. Z. Li, Y. M. Lin, P. X. Wang, W. Chen, X. Z. Fu and Y. Shao, J. Phys. Chem. C, 2012, 116, 3552.
33. Y. M. Lin, D. Z. Li, J. H. Hu, G. C. Xiao, J. X. Wang, W. J. Li and X. Z. Fu, J. Phys. Chem. C, 2012, 116, 5764.
34. W. J. Wang, L. Z. Zhang, T. C. An, G. Y. Li, H. Y. Yip and P. K. Wong, Appl. Catal., B, 2011, 108–109, 108.
35. L. L. Chen, W. X. Zhang, C. Feng, Z. H. Yang and Y. M. Yang, Ind. Eng. Chem. Res., 2012, 51, 4208.
36. J. Jiang, X. Zhang, P. B. Sun and L. Z. Zhang, J. Phys. Chem. C, 2011, 115, 20555.
37. Q. J. Xiang, J. G. Yu and M. Jaroniec, J. Phys. Chem. C, 2011, 115, 7355.

---