Synthesis and characterization of B–N co-doped mesoporous TiO₂ with enhanced photocatalytic activity

Abstract

Boron doped, nitrogen doped, and boron–nitrogen co-doped mesoporous TiO₂ photocatalysts were successfully synthesized via a fast sol–gel method. The prepared photocatalysts were characterized by X-ray diffraction (XRD), thermalgravimetric analysis (TGA), UV-vis diffuse reflectance spectroscopy (DRS), transmission electron microscopy (TEM), N₂ adsorption–desorption, and X-ray photoelectron spectroscopy (XPS). It was found that the photocatalysts have a typical mesostructure and a large specific surface area. Compared with that of the undoped TiO₂, the absorption band edges of the doped samples exhibit an evident red-shift and the absorption intensity of the visible region increases obviously. The XPS analysis shows that Ti–N–B–O structures formed on the surface of the boron–nitrogen co-doped TiO₂, indicating a synergistic effect of the two dopants that enhances the photocatalytic activity. This was evaluated by the degradation of methyl blue (MB), weak acid red (WAR), and rhodamine B (RB) dyes under visible and UV light irradiation.

---

1. Introduction

Titanium dioxide (TiO₂) is widely investigated because of its high photocatalytic activity, non-toxicity, and chemical stability, which make it functional in a wide range of areas such as dye-sensitized solar cells,^1,2 degradation of pollutants in water and air,^3–6 and hydrogen evolution.^7,8 Nevertheless, the application of conventional TiO₂ is very limited, mainly due to two problems: (1) the band gap of anatase TiO₂ is 3.2 eV, which indicates that it is only active in the UV region, accounting for less than 5% of the solar energy; (2) the low surface area of the conventional TiO₂, resulting in a low efficiency when absorbing and reacting. It is therefore urgent to develop some modifications of TiO₂ to produce a novel photocatalyst, which has a high surface area and can be activated by visible light.

Since the report by Asahi et al.⁹ revealed that nitrogen doping can significantly improve the photocatalytic activity of TiO₂ under visible light, doping with non-metal ions has been an effective means to narrow the band gap of TiO₂, and thus expand the light response range to the visible light region. Extensive efforts have been made in TiO₂ doping with non-metal ions, such as B, C, N, S and F.^10–15 In recent years, the TiO₂ materials co-doped with non-metals such as S–N, C–N, F–N, B–F and B–N,^16–20 which are expected to produce a synergistic effect, have drawn more and more attention. To our knowledge, the energies of both the B2p and N2p orbitals are higher than that of the O2p orbital. Therefore, the valence band of TiO₂ will rise when the B2p orbital or N2p orbital mixes with the O2p orbital, resulting in the narrowing of the band gap. Owing to the synergistic effect of B and N, B–N co-doped TiO₂ is considered as a much more effective photocatalyst under visible light. In a previous report, Xing et al.²¹ prepared a B and N co-doped TiO₂ by a novel double hydrothermal method. They revealed that when nitrogen was introduced into the material before boron, the synergistic effect may be caused by the formation of Ti–B–N–Ti and Ti–N–B–O compounds on the surface of the catalyst, while only Ti–N–B–O formed when boron was introduced into the material before nitrogen. Some studies have been performed to investigate the synergistic effect of boron and nitrogen in the TiO₂ catalysts,^22–24 but few of these studies have focused on modifying the crystalline structure and improving the specific surface area.

Mesoporous materials have shown unparalleled advantages in photocatalytic applications, especially due to their high specific surface area, which may produce numerous surface reaction sites for the adsorption of the reactant molecule. In addition, for mesoporous TiO₂, the small grain size results in a shorter distance for the electrons and holes to transfer to the reaction sites.²⁵ Since Antonelli and Ying²⁶ first synthesized mesoporous TiO₂ in 1995, many attempts have been made to modify TiO₂ by giving it a mesoporous structure.^27–30 However, few studies reported the mesoporous TiO₂ doped with other elements.^31,32 In our previous work, we developed a fast sol–gel method using polyacrylamide (PAM) and polyethylene (PEG) as templates to synthesize N-doped mesoporous TiO₂ with a high specific surface area.³³ In the present work, B-doped, N-doped, and B–N co-doped mesoporous TiO₂ photocatalysts were prepared by the fast sol–gel method using tetrabutyl titanate, urea, and boric acid as the Ti precursor, N source, and B source, respectively. The photocatalysts with high specific surface areas were characterized by X-ray diffraction (XRD), thermalgravimetric analysis (TGA), transmission electron microscopy (TEM), X-ray photoelectron spectroscopy (XPS), UV-vis spectroscopy, and N₂ adsorption–desorption. The photocatalytic activity of the photocatalysts is evaluated by the photocatalytic degradation of MB, WAR, and RB dyes in water under UV and visible light irradiation.

2. Experimental

2.1 Preparation of photocatalysts

A fast sol–gel preparation of the B-doped, N-doped, and B–N co-doped mesoporous TiO₂ was performed as follows. 10 mL tetrabutyl titanate was dissolved into 20 mL anhydrous ethanol with stirring to obtain solution A. 10 mL nitric acid (5 wt%), H₃BO₃ (0.055, 0.091, 0.182 g corresponding to 0.03, 0.05, 0.10 of B/Ti molar ratios), and urea (0, 1.76 g corresponding to 0, 1 of N/Ti molar ratios) were added to a solution of 10 mL deionized water and 280 mL anhydrous ethanol with stirring to obtain solution B. 0.05 g PAM and 8.8 g PEG were added to 20 mL anhydrous ethanol with stirring to obtain solution C. Solution C was added to solution B while stirring to get a mixture. Then solution A was added dropwise to the mixture above under stirring. The resultant mixture was stirred at room temperature for 1 h until a white gel was obtained. The gel was dried at 30 °C for 16 h to produce the xerogel. The resultant xerogel was crushed to obtain fine powder and further calcined in air at 400 °C for 4 h to obtain the catalyst. The samples were labeled as (mB,nN)-TiO₂, where m and n corresponded to the initial molar ratios of B to Ti and N to Ti, respectively. For comparison, undoped TiO₂ was prepared according to the above procedure in the absence of solution B.

2.2 Characterizations

XRD patterns of the photocatalysts were collected at room temperature in the range 10–75° (2θ) using a Rigaku D/MAX 2550 diffractometer (Cu Kα radiation, λ = 1.5406 Å), operated at 40 kV and 100 mA. The crystalline size was estimated by applying the Scherrer equation. Thermalgravimetric analysis (TGA) was performed using a SDT Q600 V8.0 Build 95 instrument. The dry xerogel powders were heated 800 °C in oxygen at a scan rate of 5 °C min⁻¹, and the observed mass loss was attributed to the quantitative pyrolysis of the polymer template component. N₂ adsorption isotherms were collected on an AUTOSORB-1 nitrogen adsorption apparatus at −196 °C and all samples were degassed at 120 °C for 2 h. A PHI 5300 ESCA with an Mg K X-ray source at a power of 250 W was employed for XPS analysis. The shift of binding energy due to relative surface charging was corrected using the C1s lever at 284.6 eV as an internal standard. The UV-vis absorbance spectra were obtained using a Shimadzu-2501 spectrophotometer. BaSO₄ was the reference sample and the spectra were recorded at room temperature in air within the range 200–900 nm. TEM images were recorded on a JEM-2100F made in Japan.

2.3 Evaluation of photocatalytic activity

The photocatalytic activity of the as-prepared samples was evaluated by degradation of MB, WAR, and RB dyes in water (10 mg L⁻¹) under UV and visible light irradiation. The catalyst (0.05 g) was added into a quartz reactor (100 mL), which contained 50 mL dye solution. Prior to irradiation, the suspension was magnetically stirred for 40 min in the dark to reach an adsorption–desorption equilibrium. A 300 W high-pressure mercury lamp was used as the UV radiation, with a peak of 365 nm, and a 300 W xenon lamp through a UV-cutoff filter (≤400 nm) was used as the visible light source. The lamp was cooled by flowing water in the quartz jacket around the lamp during the photocatalytic reaction, thus the ambient temperature stayed constant. The analytic suspension (5 mL) was taken out of the reactor at regular intervals, and centrifuged immediately, before being filtered to separate the TiO₂ from the solution. The residual concentration of the dyes in the remaining clear liquid was analyzed by a spectrophotometer (UV-7220, Beifenruili, China). The degradation ratio of the dyes can be calculated by (A₀ − A)/A₀ × 100%.

3. Results and discussion

3.1 TG-DSC analysis

The TG-DSC curve of the as-prepared xerogel was used out to investigate whether the volatiles had escaped or not following calcination. As shown in Fig. 1(a), the process of thermal gravity loss can be divided into two stages. It was observed that weight loss of greater than 30% occurred between 0–220 °C, which was ascribed to the volatilizing of residual ethanol, solvents, and organic templates absorbed on the surface of the xerogel. The weight loss between 220 and 400 °C was evaluated to be 28%, mainly due to the decomposition of organic templates composed of PEG and PAM.^34,35 It is noteworthy that there was a small weight loss (2%) in the range of 500–600 °C. This may be caused by the decomposition of a minute quantity of residual PAM which has higher thermostability than PEG. The result is quite consistent with the DSC curve (Fig. 1(b)). We can observe that several endothermic peaks are presented. The two peaks in the range between 0–250 °C, which are not marked in the figure, are due to the volatilizing of solvents and other volatiles. The strong peak at 338 °C was ascribed to the decomposition of organic templates and the phase transformation from amorphous to anatase. The weak peak at 468 °C was caused by the same phase transformation process. The peak at 575 °C was probably caused by the phase transformation from anatase to rutile.³⁶

Fig. 1 TG-DSC curve of the xerogel of (0.03B,1.0N)-TiO₂.

3.2 XRD and BET analysis

The XRD patterns of the as-synthesized TiO₂, B-TiO₂, N-TiO₂ and B–N-TiO₂ samples are presented in Fig. 2. All the diffraction peaks can be ascribed to pure anatase phase TiO₂ after calcination. The indexes at the top of each peak indicate the existence of different lattice planes, suggesting a multicrystal structure of TiO₂ which will be further confirmed by the SAED pattern. We can also conclude that there has been virtually no phase transition in TiO₂ in the process of B-doping or B–N co-doping. The diffraction peak at 25.4°, corresponding to the characteristic peak of the crystal plane (101) of the anatase phase TiO₂, became broader and the relative intensity decreased after boron and nitrogen were introduced into TiO₂, suggesting the decrease of the crystal size. According to the Scherrer equation, the average crystallite sizes of TiO₂, B-TiO₂, N-TiO₂, and B–N-TiO₂ samples are estimated to be about 10.9, 9.5, 9.9, and 9.1 nm, respectively, corresponding to (a), (b), (c), and (d) in Fig. 2. It confirms that the boron and nitrogen doping can efficiently decrease the crystal size mainly due to the deformation of the lattice and oxygen vacancies left by the substitution of O atoms by B or N atoms.¹⁰ The size decreases from (b) and (c) to (d) may be caused by the new structure Ti–N–B–O, formed at the internal surface of TiO₂.²¹

Fig. 2 XRD patterns of samples: (a) TiO₂; (b) (0.03B)-TiO₂; (c) (1.0N)-TiO₂; (d) (0.03B,1.0N)–TiO₂.

Fig. 3 shows the nitrogen adsorption–desorption isotherms and pore size distribution curves of the TiO₂, B-TiO₂, N-TiO₂ and B–N-TiO₂ samples. As shown in Fig. 3(a), the N₂ sorption isotherms of all samples are consistent with type IV of hysteresis loops, which are the typical mesoporous structures of the materials according to the IUPAC classification. For (0.03B,1.0N)–TiO₂, there is a hysteresis loop appearing at a relatively low pressure (0.4 < P/P₀ < 0.8), indicating a narrow range of pore size distribution, which is centered at 16.8 nm, estimated by the Barrett–Joyner–Halenda (BJH) approach (Fig. 3(b)). The mesopores allow light to scatter inside their pore channel, thus enhancing the harvesting of light.³⁷ According to the linear part of the sorption isotherm, the specific surface area and pore volume of (0.03B,1.0N)–TiO₂ were determined to be 125.4 m² g⁻¹ and 0.1776 cm³ g⁻¹, respectively. The other samples have similar nitrogen adsorption–desorption isotherms and BJH pore size distributions to the (0.03B,1.0N)–TiO₂ sample. Their textural properties are listed in Table 1, from which we can conclude that the change in the structure of TiO₂ caused by dopants is negligible. Considering the crystallite size of the sample, the mesopores are believed to be formed by the agglomeration and connection of adjacent nanoparticles in the sample.³³ This network nanostructure offers more efficient transportation for the reactant molecules to the active sites, which are expected to enhance the photocatalytic activity.^26,38

Fig. 3 N₂ adsorption–desorption isotherms (a) and the BJH corresponding pore size distribution curves (b) of pure TiO₂, B-TiO₂, N-TiO₂, and B–N-TiO₂.

Table 1 The characterization results of different samples

Sample	Crystallite size^a(nm)	SBET^b (m^2 g^−1)	Pore size^c(nm)	Pore volume^d (cm^3 g^−1)
a Evaluated by the Scherrer equation.b Measured by the BET method.c Estimated from the Barrett–Joyner–Halenda (BJH) formula.d Estimated by the N2 adsorption volume at a relative pressure of 0.99.
TiO2	10.9	121.6	16.9	0.1754
(0.03B)-TiO2	9.5	130.6	15.6	0.1804
(1.0N)-TiO2	9.9	127.8	16.8	0.1695
(0.03B,1.0N)-TiO2	9.1	125.4	16.8	0.1776

---

3.3 TEM analysis

The microstructure of the (0.03B,1.0N)–TiO₂ sample was investigated with transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM). TEM images of the sample are shown in Fig. 4(a) and (b). As estimated from the TEM images, the grains are all round-shaped with a uniform size of about 10 nm, which is in agreement with the XRD analysis. In addition, it is clearly seen that the sample has a typical mesopore structure with a wormhole-like framework. As observed in Fig. 4(b), the pore size is about 15–20 nm, which is consistent with the N₂ sorption analysis. The Debye rings shown in Fig. 4(d) suggest a sequence of diffraction rings consistent with what is expected for anatase TiO₂ as the nanocrystal of the mesoporous sample. The average interplanar space is determined to be 0.35 nm (Fig. 4(c)), which is assigned to the anatase TiO₂ facet (101) from the corresponding wide-angle XRD pattern.^39,40

Fig. 4 TEM characterization of (0.03B,1.0N)–TiO₂: (a and b) TEM images, (c) HRTEM images, (d) SAED pattern.

3.4 UV-vis DRS of catalysts

The UV-vis diffraction reflectance spectra of the (mB)-TiO₂ samples are shown in Fig. 5. It is found that the amount of boron doped into TiO₂ affects the optical absorption of the sample in both the UV and visible regions. The (0.1B)-TiO₂ sample exhibits the strongest absorption in the visible region, and (0.03B)-TiO₂ exhibits the strongest response to light in wavelengths > 550 nm. Furthermore, it can be clearly observed that the absorption edge, which is decided by the intercept on the wavelength axis for a tangent drawn on the absorption spectra, of all the samples doped with boron extend into visible light from about 420 to 480 nm. Taking into account the absorption intensities in the UV and visible regions, the amount of boron doped in the (0.03B)-TiO₂ sample is appropriate. The relative atomic concentration of boron in the sample is about 1.23 atom% determined by the XPS data. This relative atomic concentration of boron is approximately equal to the optimal relative atomic concentration of doped boron (about 1.13 atom%), which was demonstrated to result in very significant activity for B doped TiO₂.²³

Fig. 5 UV-vis absorption spectra of (mB)-TiO₂ samples.

In our previous work, we demonstrated that the (1.0N)-TiO₂ sample exhibited higher photocatalytic activity than other N doped samples with different molar ratios of N : Ti.³³ Therefore, the molar ratio of N : Ti was limited to 1 : 1 in this paper. Fig. 6(a) shows the UV-vis diffraction reflectance spectra of pure TiO₂, (0.03B)-TiO₂, (1.0N)-TiO₂ and (0.03B,1.0N)–TiO₂. Compared with the pure TiO₂, the absorption of the doped samples extends somewhat into the visible region. Further, it is clear that the absorption edge of the N doped and B–N co-doped samples exhibits an evident red-shift. The plots of transformed Kubelka–Munk function versus the energy of absorbed light give the band gap of the samples.²⁵ As shown in Fig. 6(b), the band gap energies are estimated to be 3.18, 3.02, 2.80 and 2.78 eV for pure TiO₂, (0.03B)-TiO₂, (1.0N)-TiO₂, and (0.03B,1.0N)–TiO₂, respectively, indicating a possible synergistic effect of boron and nitrogen. The narrowed band gap of the B–N co-doped TiO₂ arises from the contribution of the B–N co-doping atoms, which formed the Ti–N–B–O structure demonstrated by the XPS tests. The results confirm that both dopants will make the band gap of TiO₂ narrower and therefore improve the visible light photocatalytic activity.

Fig. 6 UV-vis absorption spectra of pure TiO₂, B-TiO₂, N-TiO₂, and B–N–TiO₂.

3.5 XPS analysis

The XPS spectra of (0.03B,1.0N)–TiO₂ is shown in Fig. 7. The obvious peaks of titanium, oxygen, nitrogen, boron, and carbon can be detected in Fig. 7(a), and the binding energies of Ti2p, O1s, N1s, B1s, and C1s are 459.0, 531, 400.5, 192.1, and 284.6 eV, respectively, which are approximately in agreement with the previous data by other researchers.^20–23 The C1s peak (284.6 eV) is usually associated with the residual carbon from the precursor solution and the adventitious hydrocarbon from the XPS instrument itself. The relative atomic concentrations of boron and nitrogen in the sample, estimated from the XPS data, were determined to be about 1.23 and 1.78 atom%, respectively.

Fig. 7 (a) XPS spectra of (0.03B,1.0N)–TiO₂, (b) high-resolution XPS spectra of B1s, (c) high-resolution XPS spectra of N1s, and (d) high-resolution XPS spectra of Ti2p.

Fig. 7(b) shows the high-resolution XPS spectra for B1s. Four XPS peaks at 187.3, 189.2, 192.1, and 193.7 eV were observed. According to the previous literature, the peak at 193.7 eV is attributed to the B–O bond.⁴¹ The peaks at 187.3 and 192.1 eV are ascribed to the structures of Ti–B and Ti–O–B,^10,42 respectively. The peak at 189.2 eV is attributed to the B–N bond, which is supposed to exist in Ti–B–N or Ti–N–B forms.³¹ To make sure of the existence of the structural forms of B and N in the sample, the N1s XPS spectra of the sample were measured and the high-resolution XPS spectra of N1s are shown in Fig. 7(c). As seen from Fig. 7(c), three peaks at 399.1, 400.5, and 401.5 eV appeared on the spectra. The peak at 400.5 eV is attributed to the structure of Ti–O–N.¹⁹ The peak at 401.5 eV may be attributed to the Ti–N–O bond.^25,43 The peak at 399.1 eV indicates a structure of Ti–N–B (ref. 21) rather than Ti–B–N, in which the binding energies of B1s and N1s are 190.5 and 398.1 eV, respectively.³¹ Taking the spectra of B1s and N1s into account in this work, we may conclude that the dopants of boron and nitrogen form a Ti–N–B–O structure in the sample.

The high-resolution XPS spectrum of Ti2p, shown as Fig. 7(d), can further verify the existence of this Ti–N–B–O structure. As seen from Fig. 7(d), the characterization peak shifts from 458.2 eV, which is the normal binding energy for pure TiO₂, to 459 eV, the binding energy for TiO₂ co-doped with boron and nitrogen.^21,31 This change in the binding energy should be attributed to the electronegativity of O in the Ti–N–B–O structure, which is in good agreement with the results of other literature.²¹

3.6 Photocatalytic activity

The UV light and visible light photocatalytic activities of pure TiO₂, (0.03B)-TiO₂, (1.0N)-TiO₂ and (0.03B,1.0N)–TiO₂ samples were evaluated by photocatalytic degradation of a MB aqueous solution (shown in Fig. 8). Fig. 8(a) shows a comparison of the photocatalytic activity of the samples under UV light irradiation. As seen from the figure, although all the samples exhibit high photocatalytic activity under UV irradiation, it is somewhat different for each sample. It can be seen that all the doped samples, except (1.0N)-TiO₂, exhibit a higher photocatalytic activity, which is attributed to the distinction of absorption intensity for all samples in the UV region, as shown in Fig. 6(a).

Fig. 8 Photocatalytic degradation kinetics of MB solutions under irradiation of (a) UV light, (b) visible light.

The photocatalytic activity of the samples under visible light is shown in Fig. 8(b). Compared with pure TiO₂, the doped TiO₂ samples showed an obviously enhanced effect on the photocatalytic activity, especially for the B–N co-doped TiO_2. The degradation rate of the B–N co-doped TiO₂ photocatalyst is more than 70% for 240 min of irradiation time, which is roughly three times that of pure TiO₂ for the same irradiation time. This is largely due to the outstanding synergistic effect of boron and nitrogen co-doping by forming the Ti–N–B–O structure.²¹ The result is in great agreement with the UV-vis DRS analysis.

Comparing Fig. 8(a) with (b), it is evident that the photocatalytic activities of all the samples in the UV region are higher than in the visible region. This is because the absorption in the UV region is stronger than in the visible region, which has been shown in Fig. 6. It is also found that the enhancement of photocatalytic activity for the samples doped by boron and nitrogen in the UV region seems to be negligible compared with that of the samples in the visible region. This may be ascribed to the Ti–N–B–O structure formed on the surface of the sample. To the best of our knowledge, the UV light irradiation will produce a lot of electrons and holes, which will be recombined by the excess amount of B–N species on the surface of the sample^21,44 and therefore affect the photocatalytic activity.

To confirm the universality of the photocatalytic degradation by the B–N co-doped mesoporous TiO₂, we also tested the photocatalyst using other dyes such as weak acid red (WAR) and rhodamine B (RB). For the purpose of eliminating the interference caused by the photocatalytic degradation environment, all conditions were kept the same as in the case of MB. As illustrated in Fig. 9, the degradation curves of both WAR and RB under illumination of both UV light and visible light showed similar characteristics with those of MB. The degradation rates of all dyes under irradiation of UV light for 90 minutes (Fig. 9(a)) were over 90%, owing to the strong absorption in the UV region by the B–N co-doped TiO₂. Degradation rates under illumination of visible light for 4 h (Fig. 9(b)) were over 60% for all dyes, suggesting a high photocatalytic activity of B–N co-doped mesoporous TiO₂ in the visible region. Thus, it should be a reasonable conclusion that the B–N co-doped mesoporous TiO₂ obtained has a potential application in the degradation of various organic pollutants.

Fig. 9 Photocatalytic degradation kinetics of dyes under irradiation of (a) UV and (b) visible light.

4. Conclusions

In summary, mesoporous TiO₂ photocatalysts co-doped with boron and nitrogen were successfully prepared by using the fast sol–gel method. The co-doped catalyst has a specific surface area of 125.4 m² g⁻¹ and the average pore size is decided to be 16.8 nm, which implies a typical mesopore structure. The XPS analysis shows that Ti–N–B–O structures formed on the surface of the co-doped sample. The special structure greatly improves the photocatalytic activity under visible light, suggesting a synergistic effect of boron and nitrogen. The absorption band edge of the B–N co-doped mesoporous TiO₂ sample exhibits an evident red-shift and its absorption intensity of the visible region is higher than that of the undoped sample, which makes the energy gap of TiO₂ narrower. Compared with the un-doped sample, the B–N co-doped mesoporous TiO₂ appears to have higher photocatalytic activity under irradiation of visible light.

Acknowledgements

This work was supported by the Science and Technology Planning Project of Shaanxi Province (2013K09-04) and the Scientific Research Project of the Provincial College Key Laboratory of Shaanxi Province (2010JS007).

References

1. P. Docampo, S. Guldin, U. Steiner and H. J. Snaith, J. Phys. Chem. Lett., 2013, 4, 698–703.
2. H. Nishikiori, Y. Uesugi, R. A. Setiawan, T. Fujii, W. Qian and M. A. El-Sayed, J. Phys. Chem. C, 2012, 116, 4848–4854.
3. D. Wang, X. Li, J. Chen and X. Tao, Ind. Eng. Chem. Res., 2012, 51, 218–224.
4. J. He, G. K. Reddy, S. W. Thiel, P. G. Smirniotis and N. G. Pinto, J. Phys. Chem. C, 2011, 115, 24300–24309.
5. G. K. Prasad, P. V. R. K. Ramacharyulu, J. Praveen Kumar, A. R. Srivastava and B. Singh, Thin Solid Films, 2012, 520, 5597–5601.
6. P. Wang, Y. Ao, C. Wang, J. Hou and J. Qian, J. Hazard. Mater., 2012, 223, 79–83.
7. Y. Chen, Y. Chang, J. Huang, I. Chen and C. Kuo, J. Phys. Chem. C, 2012, 116, 3857–3865.
8. C. G. Silva, R. Juarez, T. Marino, R. Molinari and H. Garcia, J. Am. Chem. Soc., 2011, 133, 595–602.
9. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
10. N. Lu, H. M. Zhao, J. Y. Li, X. Quan and S. Chen, Sep. Purif. Technol., 2008, 62, 668–673.
11. P. Xu, J. Lu, T. Xu, S. Gao, B. Huang and Y. Dai, J. Phys. Chem. C, 2010, 114, 9510–9517.
12. T. M. Triantis, T. Fotiou, T. Kaloudis, A. G. Kontos, P. Falaras, D. D. Dionysiou, M. Pelaez and A. Hiski, J. Hazard. Mater., 2012, 211, 196–202.
13. Q. Xiang, J. Yu, W. Wang and M. Jaroniec, Chem. Commun., 2011, 47, 6906–6908.
14. M. R. Bayati, A. Z. Moshfegh and F. Golestani-Fard, Appl. Catal., A, 2010, 389, 60–67.
15. S. Tosoni, O. Lamiel-Garcia, D. Fernandez Hevia, J. M. Dona and F. Illas, J. Phys. Chem. C, 2012, 116, 12738–12746.
16. Q. Xiang, J. Yu and M. Jaroniec, Phys. Chem. Chem. Phys., 2011, 13, 4853–4861.
17. J. Xu, B. Yang, M. Wu, Z. Fu, Y. Lv and Y. Zhao, J. Phys. Chem. C, 2010, 114, 15251–15259.
18. T. Huang, S. Mao, J. Yu, Z. Wen, G. Luc and J. Chen, RSC Adv., 2013, 3, 16657–16664.
19. E. A. Reyes-Garcia, Y. Sun and D. Raftery, J. Phys. Chem. C, 2007, 111, 17146–17154.
20. X. Zhou, F. Peng, H. Wang and H. Yu, J. Solid State Chem., 2011, 184, 3002–3007.
21. M. Xing, W. Li, Y. Wu, J. Zhang and X. Gong, J. Phys. Chem. C, 2011, 115, 7858–7865.
22. N. O. Gopal, H. H. Lo and S. C. Ke, J. Am. Chem. Soc., 2008, 130, 2670–2671.
23. S. In, A. Orlov, R. Berg, F. Garcia, S. Pedrosa-Jimenez, M. S. Tikhov, D. S. Wright and R. M. Lambert, J. Am. Chem. Soc., 2007, 129, 13790–13791.
24. X. Ding, X. Song, P. Li, Z. Ai and L. Zhang, J. Hazard. Mater., 2011, 190, 604–612.
25. D. Wang, L. Jia, X. Wu, L. Lu and A. Xu, Nanoscale, 2012, 4, 576–584.
26. D. M. Antonelli and J. Y. Ying, Angew. Chem., Int. Ed. Engl., 1995, 34, 2014–2017.
27. L. Chen, B. Yao, Y. Cao and K. Fan, J. Phys. Chem. C, 2007, 111, 11849–11853.
28. W. Shao, F. Gu, C. Li and M. Lu, Inorg. Chem., 2010, 49, 5453–5459.
29. T. Coquil, C. Reitz, T. Brezesinski, E. J. Nemanick, S. H. Tolbert and L. Pilon, J. Phys. Chem. C, 2010, 114, 12451–12458.
30. D. Lee, S. Park, S. Ihm and K. Lee, Chem. Mater., 2007, 19, 937–941.
31. G. Liu, Y. Zhao, C. Sun, F. Li, G. Lu and H. Cheng, Angew. Chem., Int. Ed., 2008, 47, 4516.
32. Y. Wu, M. Xing and J. Zhang, J. Hazard. Mater., 2011, 192, 368–373.
33. E. Liu, X. Guo, L. Qin, G. Shen and X. Wang, Chin. J. Catal., 2012, 33, 1665–1671.
34. D. Huang, S. Liao, J. Liu, Z. Dang and L. Petrik, J. Photochem. Photobiol., A, 2006, 184, 282–288.
35. A. Mukhopadhyay, N. Joshi, K. Chattopadhyay and G. De, ACS Appl. Mater. Interfaces, 2012, 4, 142–149.
36. X. Wang, D. Hu and J. Yang, Chem. Mater., 2007, 19, 2610–2621.
37. Y. Li, T. White and S. Lim, J. Solid State Chem., 2004, 177, 1372–1381.
38. G. Tian, H. Fu, L. Jing, B. Xin and K. Pan, J. Phys. Chem. C, 2008, 112, 3083–3089.
39. G. Liu, H. Yang, X. Wang, L. Cheng, J. Pan, G. Lu and H. Cheng, J. Am. Chem. Soc., 2009, 131, 12868–12869.
40. X. Xue, Y. Wang and H. Yang, Appl. Surf. Sci., 2013, 264, 94–99.
41. D. M. Chen, D. Yang, Q. Wang and Z. Y. Jiang, Ind. Eng. Chem. Res., 2006, 45, 4110–4116.
42. Q. Ling, J. Sun and Q. Zhou, Appl. Surf. Sci., 2008, 254, 3236–3241.
43. H. Irie, Y. Watanabe and K. Hashimoto, J. Phys. Chem. B, 2003, 107, 5483–5486.
44. N. Murakami, T. Chiyoya, T. Tsubota and T. Ohno, Appl. Catal., A, 2008, 348, 148–152.

---