In situ bubble template promoted facile preparation of porous g-C₃N₄ with excellent visible-light photocatalytic performance

Abstract

Porous graphitic carbon nitride (pg-C₃N₄) was facilely and economically prepared through in situ bubble templates such as (NH₄)₂S₂O₈, and the surface area of the resultant pg-C₃N₄ was controlled from 6.4 to 55.0 m² g⁻¹ by adjusting the mass ratio of (NH₄)₂S₂O₈/melamine. Moreover, pg-C₃N₄ was also obtained by other gas-generating porogens of ammonium salts, displaying the generality of the preparation method. The as-prepared g-C₃N₄ presented a porous structure with a higher surface area and displayed an improved separation efficiency for photogenerated electron–hole pairs. These integrative positive factors contributed to pg-C₃N₄ possessing more excellent photocatalytic activity for degrading Rhodamine B and phenol pollutants, and splitting water to H₂ than bulk g-C₃N₄ under visible-light. Moreover, the as-prepared pg-C₃N₄ exhibited unexceptionable stability and reusability even after four photocatalytic runs. The simple, economical and general fabrication strategy with bubble-generated porogen agents for porous graphitic carbon nitride with superior visible-light photocatalytic performance is attractive for environmental and energy application fields.

---

1. Introduction

Recently, semiconductor photocatalysts have drawn considerable attention owing to their outstanding performance in pollutant degradation or hydrogen production from water splitting through conversion of solar energy.^1–5 For the purpose of taking full advantage of solar energy, a large number of semiconductor materials with visible-light-responsive activity have been developed. Graphitic carbon nitride (g-C₃N₄), with a suitable band gap and unique properties, has attracted great scientific interest in visible-light-induced conversions.^6–14 Unfortunately, the photocatalytic activity of bulk g-C₃N₄ was still restricted by a high recombination rate of photogenerated electron–hole pairs and a low surface area (less than 10 m² g⁻¹). To solve the existing shortcomings, several methods such as nonmetal or metal hybridization,^15,16 coupling with semiconductors,^17–19 morphology control^20,21 and mesopore introduction,^22,23 have been exploited to enhance the photocatalytic performance of g-C₃N₄.

Photocatalysts with porous structures and large surface areas are vital for satisfactory catalytic efficiency, which could increase the mass transfer, offer more surface reactive sites and suppress the recombination of photoinduced electron–hole pairs.^22,24 Generally, porous g-C₃N₄ with a large surface area is obtained by a template-induced method. However, the synthesis and the catalytic activity of porous g-C₃N₄ are limited because soft-templating methods result in some carbon residue in the product during template calcination. Besides, the weak interaction between a soft template and the precursors of g-C₃N₄ leads to a poor porous structure, and moreover, the necessary organic mesoscaled soft template in synthesis leads to the high cost of the catalyst.²⁵ Using silica as a hard template is an alternative and successful pathway for porous g-C₃N₄ synthesis.^26–28 However, toxic ammonium bifluoride or aqueous hydrogen fluoride is necessarily involved for template removal, and the removal process is also complex and time-consuming. Hence, the facile and economic synthesis of porous g-C₃N₄ with large surface area is attractive to practical applications.

In this contribution, porous g-C₃N₄ (pg-C₃N₄) with a large surface area was facilely fabricated by in situ direct pyrolysis of (NH₄)₂S₂O₈ and melamine in air. Using (NH₄)₂S₂O₈ as a bubble-generating template resulted in the formation of a porous structure during the condensation process of the melamine precursor. Moreover, the porous structure and photocatalytic activity of the resultant pg-C₃N₄ were easily controlled by adjusting the mass ratio of (NH₄)₂S₂O₈ to melamine. Due to the low recombination rate of photogenerated electron–hole pairs and the large surface area, the resultant pg-C₃N₄ exhibited much higher efficiency than bulk g-C₃N₄ for decomposing Rhodamine B (RhB) and phenol, together with water splitting to evolve H₂. Promisingly, the preparation method for porous g-C₃N₄ was versatile for other bubble-generated porogens such as NH₄Cl and (NH₄)₂S₂O₃ besides (NH₄)₂S₂O₈. The developed strategy provided a simple and economic avenue for the fabrication of porous g-C₃N₄ with superior visible-light photocatalytic performance, which is appealing to wide application fields of g-C₃N₄.

2. Experimental methods

2.1 The preparation of photocatalysts

A certain amount of (NH₄)₂S₂O₈ was dissolved in 10 mL distilled water, then 3 g melamine was added into the above solution. After stirring for 2 h, the mixture was kept in oven at 50 °C to remove the distilled water. Then, the resultant solids were placed in a crucible with a cover and calcined at 550 °C for 2 h in air, and the final samples were obtained. According to the mass ratios of (NH₄)₂S₂O₈ to melamine, 0.15 : 1, 0.6 : 1 and 1 : 1, the as-made samples were named as pg-C₃N₄-1, pg-C₃N₄-2 and pg-C₃N₄-3, respectively. For comparison, bulk g-C₃N₄ was synthesized by direct polycondensation of melamine under the same conditions only without the addition of (NH₄)₂S₂O₈. In order to detect the universality of the preparation method, NH₄Cl and (NH₄)₂S₂O₃ were also used as bubble-generating templates. 3 g NH₄Cl or (NH₄)₂S₂O₃ was applied instead of (NH₄)₂S₂O₈, together with 3 g melamine to synthesize pg-C₃N₄ (NH₄Cl) and pg-C₃N₄((NH₄)₂S₂O₃), respectively.

2.2 Material characterizations

N₂ adsorption–desorption isotherms were collected at 77 K using a Quantachrome NOVA 2000 surface area and porosity analyzer. Samples were out gassed at 150 °C for 12 h prior to measurement. Sample morphology was examined by scanning electron microscopy (SEM, SU8010, Hitachi). The spatial elemental distributions were investigated by energy-dispersive spectrometry (EDS)-elemental mapping analysis. Moreover, the morphology was also examined by transmission electron microscopy (TEM, JEM-2010). The patterns of X-ray diffraction were carried out on a Bruker D8 Advance X-ray powder diffractometer with Cu Kα radiation (40 kV, 40 mA) for phase identification. Fourier transform infrared spectroscopy (FTIR) was recorded in a transmission mode from 4000 to 400 cm⁻¹ on a Perkin Elmer spectrum 100 FTIR spectrometer using KBr discs. X-ray photoelectron spectroscopy (XPS) measurements were recorded on a Thermo Fisher Scientific Escalab 250Xi. Elemental analysis (EA) was performed on a Vario Microcube CHN analyzer. The UV-vis diffuse reflectance spectra (DRS) were measured by a Perkin Elmer Lambda 750 UV-vis spectrometer.

2.3 Photocatalytic testing

The photocatalytic performance was evaluated through degradation of RhB dye and phenol under visible-light. A 300 W Xe lamp with a 400 nm cut-off filter was used as the visible-light source. 100 mg as-prepared photocatalyst was dispersed into 100 mL 5 mg L⁻¹ RhB or phenol solution for photocatalytic examination under magnetic stirring. Prior to the light irradiation, the dispersion was kept in the dark for 60 min under magnetic stirring to reach the adsorption–desorption equilibrium. After being irradiated by visible-light, the solutions were collected at every given time interval, centrifuged to remove the catalyst, then analyzed on UV-vis spectrometer. For comparison, the photodegradation reaction was also carried out in the absence of any catalyst. The efficiency of degradation was calculated by C/C₀, wherein C is the concentration of the remaining pollutant solution at each irradiated time, and C₀ is the initial concentration.

The visible-light-induced catalytic H₂ evolution by the as-prepared photocatalyst was carried out in a Pyrex top-irradiation reaction vessel connected with a closed glass gas-circulation system. H₂ production was performed in a solution of 50 mg photocatalyst and 10 mL triethanolamine in 90 mL distilled water. 3 wt% Pt was loaded on the surface of the photocatalyst by an in situ photodeposition method using H₂PtCl₆ as the starting material. The reactant solution was evacuated several times to remove the air prior to irradiation by a 300 W Xe lamp with a 400 nm cut-off filter and a water filter. The evolved gases were analyzed by gas chromatography (SP7800) with N₂ as the carrier gas.

3. Results and discussion

It was observed that the morphologies of bulk g-C₃N₄ and pg-C₃N₄-3 were clearly different, as shown in Fig. 1. Bulk g-C₃N₄ consisted of dense and thick layers to form a flat massive structure, while the pg-C₃N₄-3 surface consisted of a large amount of pores with different sizes. The morphology difference can be explained by the decomposition of (NH₄)₂S₂O₈ to NH₃ (g), SO₂ (g) and O₂ (g), which acted as bubble templates for the formation of the porous structure during the condensation process of the melamine precursor at 550 °C (Fig. 1C). EDS elemental mappings revealed the existence of two major elements of C and N in Fig. 1D, and the two elements were homogeneously distributed. Elemental S was not detected, which indicated that (NH₄)₂S₂O₈ only acted as the bubble template and was removed thoroughly after the high-temperature condensation of the g-C₃N₄ precursor. TEM images of bulk g-C₃N₄ and pg-C₃N₄-3 are also provided. Bulk g-C₃N₄ exhibited a flat aggregated structure and no pores were observed. For pg-C₃N₄-3, a layered structure with random pores on the surface was displayed. This result suggested that the introduced (NH₄)₂S₂O₈ acted as the porous template for the generation of pg-C₃N₄. Thus, controlling the (NH₄)₂S₂O₈ amount could adjust the porous structure and the surface area of g-C₃N₄.

Fig. 1 SEM images of (A) bulk g-C₃N₄ and (B) pg-C₃N₄-3; (C) a schematic for the formation of pg-C₃N₄; (D) EDS elemental mappings for the rectangle area; TEM images of (E) bulk g-C₃N₄ and (F) pg-C₃N₄-3.

Fig. 2A shows the N₂ adsorption–desorption isotherms of the as-prepared pg-C₃N₄ samples and bulk g-C₃N₄. Except for the bulk g-C₃N₄, all the samples exhibited type IV isotherms characteristic of mesoporous materials. And as shown in Table 1, the surface areas of pg-C₃N₄ increased progressively with increasing mass fractions of (NH₄)₂S₂O₈, due to the greater number of bubbles produced in the condensation process. Hence, by adjusting the mass ratio of (NH₄)₂S₂O₈/melamine, the surface area of pg-C₃N₄ could be successfully controlled, and the surface area of pg-C₃N₄-3 was approximately 8.5 times higher than that of bulk g-C₃N₄. The large surface area would benefit the photocatalytic activity. In addition, the wide distributions of BJH pore size in the range of 20–100 nm further confirmed the formation of different sized pores, which was consistent with the TEM and SEM results.

Fig. 2 (A) N₂ adsorption–desorption isotherms and (B) pore size distributions of (a) bulk g-C₃N₄, (b) pg-C₃N₄-1, (c) pg-C₃N₄-2 and (d) pg-C₃N₄-3.

Table 1 BET surface area and pore volume of various samples

Samples	Surface area (m^2 g^−1)	Average pore size (nm)
Bulk g-C3N4	6.5	—
pg-C3N4-1	10.4	38.7
pg-C3N4-2	34.4	33.4
pg-C3N4-3	55.0	29.5
pg-C3N4(NH4Cl)	41.8	32.7
pg-C3N4((NH4)2S2O3)	24.2	30.4

---

The typical g-C₃N₄ structure was detected by XRD in Fig. 3 and no other impurity phase was found in all the resultant samples. The main peaks at 27.6° and 13.0° were indexed to (002) and (100) planes of hexagonal g-C₃N₄ (JCPDS card no. 87-1526), corresponding to the graphite-like stacking and in-plane structural repeating motifs of the conjugated aromatic units of g-C₃N₄.²⁹ Compared with the bulk g-C₃N₄, the peak intensities of the pg-C₃N₄ samples became weaker, which were induced by the generated pores preventing the formation of extended graphitic layers during the thermal condensation process of melamine.²⁶ In order to check other ammonium salts as bubble-generating porogens for the synthesis of porous g-C₃N₄, NH₄Cl and (NH₄)₂S₂O₃ were also applied instead of (NH₄)₂S₂O₈. The XRD patterns of g-C₃N₄ obtained from the former two templates also indicated the typical g-C₃N₄ diffraction peaks (Fig. 3e and f), and the N₂ sorption isotherms further evidenced the porous structure; the surface areas were 41.8 and 24.2 m² g⁻¹, respectively (Fig. 4). These results implied that the ammonium salts were used as porogens since they decompose with gas evolution without the possibility of generating additional carbon, which might alter or interfere with the properties of the g-C₃N₄ formed from melamine.

Fig. 3 XRD patterns of (a) bulk g-C₃N₄, (b) pg-C₃N₄-1, (c) pg-C₃N₄-2, (d) pg-C₃N₄-3, (e) pg-C₃N₄ (NH₄Cl) and (f) pg-C₃N₄ ((NH₄)₂S₂O₃).

Fig. 4 N₂ adsorption–desorption isotherms of (a) pg-C₃N₄ (NH₄Cl) and (b) pg-C₃N₄ ((NH₄)₂S₂O₃), (inset is the pore size distributions).

Fig. 5 shows the FTIR spectra of the bulk g-C₃N₄ and pg-C₃N₄ samples. All the samples exhibited the typical IR patterns of g-C₃N₄, the broad bands in the 3000–3500 cm⁻¹ region were attributed to the adsorbed O–H bands and N–H components.³⁰ The absorption peak at 809 cm⁻¹ was considered as the out-of-plane skeletal bending modes of triazine.³¹ The absorption bands in the range of 1200–1700 cm⁻¹ were assigned to the typical stretching modes of C₃N₄ heterocycles.^32,33 The peak at 1642 cm⁻¹ was ascribed to the C=N stretching vibration mode, while the peaks at 1242, 1320 and 1410 cm⁻¹ were attributed to aromatic C–N stretching vibration modes.

Fig. 5 FTIR spectra of (a) bulk g-C₃N₄, (b) pg-C₃N₄-1, (c) pg-C₃N₄-2 and (d) pg-C₃N₄-3.

To further determine the surface composition and chemical state, XPS spectra of pg-C₃N₄-3 were also investigated. XPS survey spectra displayed only two peaks of C 1s and N 1s (Fig. 6A). In the C 1s spectrum, two peaks were distinguished to be centered at 284.6 and 288.1 eV. The peak at 284.6 eV was exclusively assigned to carbon atoms (C–C bonding) in a pure carbon environment, and the peak at 288.1 eV was identified as sp²-bonded carbon (C=N).³⁴ The N 1s spectrum was divided into three peaks at 398.6, 399.3 and 400.8 eV in Fig. 6C. The peak at 398.6 eV was typically attributed to an sp²-bonded N atom to two carbon atoms (C–N=C), and the two peaks at 399.3 and 400.8 eV were assigned to tertiary nitrogen (N–(C)₃) and amino functional groups with a hydrogen atom (N–H).^35,36 And as seen from the S 2p spectrum in Fig. 6D, there was no S residue in the pg-C₃N₄ sample, which further confirmed that (NH₄)₂S₂O₈ only acted as the bubble template during the g-C₃N₄ preparation process, and then (NH₄)₂S₂O₈ was removed completely after condensation, consistent with the EDS elemental mapping analysis of the sample.

Fig. 6 XPS (A) survey spectra, (B) C 1s, (C) N 1s and (D) S 2p for pg-C₃N₄-3.

The element contents in the as-prepared samples were further confirmed by elemental analyses (Table 2). The atomic ratios of carbon to nitrogen were almost 0.64, lower than 0.75 for the ideal g-C₃N₄ sample, and the hydrogen elements detected were indicative of the incomplete thermal condensation of melamine. Combined with the FTIR and XPS results, the residual hydrogen atoms were bonded to the edges of the graphene-like C–N sheet in the forms of C–NH₂ and 2C–NH bonds.^37,38

Table 2 Element content in various samples

Samples	N wt%	C wt%	H wt%	C/N atomic ratio
Bulk g-C3N4	61.32	33.90	1.60	0.64
pg-C3N4-1	60.00	33.08	1.79	0.64
pg-C3N4-2	60.22	33.28	1.75	0.64
pg-C3N4-3	59.32	33.14	1.84	0.65

---

The UV-visible DRS of the samples was investigated and shown in Fig. 7. Bulk g-C₃N₄ displayed absorption from ultraviolet to visible-light, and its absorption edge was around 463 nm, corresponding to the band gap at 2.68 eV and consistent with a previous report.¹² It was reported that if S was doped in the g-C₃N₄ sample, the corresponding band gap was narrowed.³⁹ However, the as-prepared pg-C₃N₄ samples displayed similar band gaps to bulk g-C₃N₄, proving that pg-C₃N₄ was not doped by elemental S.

Fig. 7 UV-visible DRS of (a) bulk g-C₃N₄, (b) pg-C₃N₄-1, (c) pg-C₃N₄-2 and (d) pg-C₃N₄-3.

The photocatalytic capability of the pg-C₃N₄ samples was evaluated by degrading RhB and phenol under visible-light. As shown in Fig. 8A, prior to illumination, the adsorption of RhB pollutant on bulk g-C₃N₄ was low at 9%. Due to the improved BET surface area, the adsorption amounts were enhanced to 14.6% for pg-C₃N₄-1, 24.8% for pg-C₃N₄-2 and 31.5% for pg-C₃N₄-3. For the photocatalytic process, the degradation of RhB was hardly carried out even for 60 min irradiation without any catalyst, and the degradation efficiency was also low at 50.2% over bulk g-C₃N₄. Noticeably, pg-C₃N₄ samples exhibited significantly enhanced degradation efficiencies, and the degradation rate increased with the increment of (NH₄)₂S₂O₈ addition amounts. The pg-C₃N₄-3 showed the highest activity at 96% within 40 min. Additionally, in order to exclude the effect of dye self-sensitization, the photodegradation of phenol was further conducted, as shown in Fig. 8B. The adsorption amount of phenol on the pg-C₃N₄ samples was improved compared with bulk g-C₃N₄, moreover, pg-C₃N₄-3 exhibited the best photodegradation efficiency at 55.0%, much higher than the 20.2% efficiency over bulk g-C₃N₄.

Fig. 8 (A) The adsorption and photodegradation curves of RhB; (B) the adsorption and photodegradation curves of phenol; (C) hydrogen evolution rates over various photocatalysts; (a) bulk g-C₃N₄, (b) pg-C₃N₄-1, (c) pg-C₃N₄-2, (d) pg-C₃N₄-3 and (e) without catalyst.

The photocatalytic H₂ evolution from water splitting over the as-prepared photocatalysts was also investigated (Fig. 8C). The obtained pg-C₃N₄ displayed a higher hydrogen evolution rate (HER = 35, 80, 100 μmol h⁻¹) than the bulk g-C₃N₄ (HER = 16 μmol h⁻¹), which showed a similar trend to those for RhB and phenol photodegradations. The photocatalytic H₂ evolution rate over pg-C₃N₄-3 is about 6 times higher than that of bulk g-C₃N₄. Obviously, the introduced pores caused by (NH₄)₂S₂O₈ made pg-C₃N₄ exhibit the improved photocatalytic activities for degrading pollutants and hydrogen production from water splitting. The positive effects coming from the increased surface area and the formed porosity contributed to the enhanced adsorption capacity to the pollutants, provided more photocatalytic active sites, and promoted the migration and separation of charge carriers, thus, integrally leading to the significant acceleration in photocatalytic efficiency of pg-C₃N₄.

The stability of a photocatalyst is vital to the practical applications. The photocatalytic degradation experiments of RhB and photosplitting of H₂O over pg-C₃N₄-3 were repeated up to four times under the same conditions and the results are shown in Fig. 9. After each run, the catalyst was centrifuged, washed by water and ethanol, dried then reused for the next run. In the case of four recycles, the RhB photodegradation activity did not change markedly and still remained at 90%, suggesting that pg-C₃N₄-3 was not photocorroded during the photodegradation process and kept excellent stability. Similarly, the durability of pg-C₃N₄-3 for H₂ evolution was also investigated by four consecutive operations. The produced H₂ increased steadily with irradiation time in each run without noticeable deactivation.

Fig. 9 Recycling runs of (A) photodegradation activity of RhB and (B) H₂ evolution over pg-C₃N₄-3 under visible-light irradiation.

In addition, the spent catalyst structure was further measured by XRD after four runs. Fig. 10 indicates that the peaks at 13.0° and 27.6° did not change after the repeated uses. However, the reduced diffraction intensities over the spent pg-C₃N₄-3 resulted from the narrowed interplanar distance and/or interlayer stacking distance of g-C₃N₄ after the recycles.⁴⁰

Fig. 10 The XRD patterns of the fresh and spent pg-C₃N₄-3.

4. Conclusion

Here, a facile and economic fabrication method was developed by direct pyrolysis condensation of melamine and (NH₄)₂S₂O₈. The obtained g-C₃N₄ possessed a large surface area and an abundant porous structure, which were formed by bubble templates generated from (NH₄)₂S₂O₈ decomposition during the condensation of the melamine precursor. Attractively, the preparation method for porous g-C₃N₄ was versatile for other bubble-generating porogens, such as NH₄Cl and (NH₄)₂S₂O₃ besides (NH₄)₂S₂O₈. Due to the large surface area and porous structure, pg-C₃N₄ exhibited a more efficient photocatalytic performance for degrading pollutants and higher H₂ evolution rates than the bulk g-C₃N₄ under visible-light irradiation. And the surface area and porous structure together with the photocatalytic activity pg-C₃N₄ were easily controlled by adjusting the mass fraction of porogen to melamine. The present synthesis route opened an avenue for the facile and economic preparation of porous g-C₃N₄-based materials with large surface areas, which display wide practical applications in environmental remediation and clean energy production.

Acknowledgements

We sincerely acknowledge the financial support from the National Natural Science Foundation of China (21373069), the Science Foundation of Harbin City (NJ20140037), the State Key Lab of Urban Water Resource and Environment of Harbin Institute of Technology (HIT2015DX08) and the Fundamental Research Funds for the Central Universities (HIT. IBRSEM. 201327).

Notes and references

1. Y. Tian, B. Chang, J. Lu, J. Fu, F. Xi and X. P. Dong, ACS Appl. Mater. Interfaces, 2013, 5, 7079–7085.
2. X. Chen, L. Liu, P. Y. Yu and S. S. Mao, Science, 2011, 331, 746–750.
3. J. Zhang, Q. Xu, Z. C. Feng, M. J. Li and C. Li, Angew. Chem., Int. Ed., 2008, 47, 1766–1769.
4. X. Wang, S. Blechert and M. Antonietti, ACS Catal., 2012, 2, 1596–1606.
5. Y. Wang, X. Wang and M. Antonietti, Angew. Chem., Int. Ed., 2012, 51, 68–89.
6. X. C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domenet and M. Antonietti, Nat. Mater., 2009, 8, 76–80.
7. H. Yan, Chem. Commun., 2012, 48, 3430–3432.
8. A. Suryawanshi, P. Dhanasekaran, D. Mhamane, S. Kelkar, S. Patil, N. Gupta and S. Ogale, Int. J. Hydrogen Energy, 2012, 37, 9584–9589.
9. H. W. Kang, S. N. Lim, D. Song and S. B. Park, Int. J. Hydrogen Energy, 2012, 37, 11602–11610.
10. S. C. Lee, H. O. Lintang and L. Yuliati, Chem.–Asian J., 2012, 7, 2139–2144.
11. L. Shi, L. Liang, J. Ma, F. X. Wang and J. M. Sun, Catal. Sci. Technol., 2014, 4, 758–765.
12. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2010, 26, 3894–3901.
13. P. Niu, L. Zhang, G. Liu and H. M. Cheng, Adv. Funct. Mater., 2012, 22, 4763–4770.
14. 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, 361–371.
15. G. Zhang, M. Zhang, X. Ye, X. Qiu, S. Lin and X. Wang, Adv. Mater., 2014, 26, 805–809.
16. Y. Yang, Y. Guo, F. Liu, X. Yuan, Y. H. Guo, S. Zhang, W. Guo and M. Huo, Appl. Catal., B, 2013, 142–143, 828–837.
17. L. Ge, C. Han and J. Liu, Appl. Catal., B, 2011, 108–109, 100–107.
18. Y. Yang, W. Guo, Y. Guo, Y. Zhao, X. Yuan and Y. Guo, J. Hazard. Mater., 2014, 271, 150–159.
19. J. Fu, B. Chang, Y. Tian, F. Xi and X. Dong, J. Mater. Chem. A, 2013, 1, 3083–3090.
20. M. Tahir, C. Cao, F. K. Butt, F. Idrees, N. Mahmood, Z. Ali, I. Aslam, M. Tanveer, M. Rizwan and T. Mahmood, J. Mater. Chem. A, 2013, 1, 13949–13955.
21. J. Liu, J. Huang, H. Zhou and M. Antonietti, ACS Appl. Mater. Interfaces, 2014, 6, 8434–8440.
22. X. Wang, K. Maeda, X. Chen, K. Takanabe, K. Domen, Y. Hou, X. Fu and M. Antonietti, J. Am. Chem. Soc., 2009, 131, 1680–1681.
23. J. Zhang, F. Guo and X. Wang, Adv. Funct. Mater., 2013, 23, 3008–3014.
24. L. Shi, L. Liang, F. X. Wang, M. S. Liu, S. F. Zhong and J. M. Sun, Catal. Commun., 2015, 59, 131–135.
25. W. Shen, L. Ren, H. Zhou, S. Zhang and W. Fan, J. Mater. Chem., 2011, 21, 3890–3894.
26. J. Xu, H. T. Wu, X. Wang, B. Xue, Y. X. Li and Y. Cao, Phys. Chem. Chem. Phys., 2013, 15, 4510–4517.
27. S. N. Talapaneni, S. Anandan, G. P. Mane, C. Anand, D. S. Dhawale, S. Varghese, A. Mano, T. Mori and A. Vinu, J. Mater. Chem., 2012, 22, 9831–9840.
28. S. S. Park, S. W. Chu, C. F. Xue, D. Y. Zhao and C. S. Ha, J. Mater. Chem., 2011, 21, 10801–10807.
29. D. S. Wang, H. Sun, Q. Luo, X. Yang and R. Yin, Appl. Catal., B, 2014, 156–157, 323–330.
30. F. Dong, Z. Wang, Y. Sun, W. K. Ho and H. Zhang, J. Colloid Interface Sci., 2013, 401, 70–79.
31. H. Zhao, H. Yu, X. Quan, S. Chen, Y. Zhang, H. Zhao and H. Wang, Appl. Catal., B, 2014, 152–153, 46–50.
32. L. Shi, L. Liang, F. X. Wang, J. Ma and J. M. Sun, Catal. Sci. Technol., 2014, 4, 3235–3243.
33. R. Yin, Q. Luo, D. S. Wang, H. Sun, Y. Li, X. Li and J. An, J. Mater. Sci., 2014, 49, 6067–6073.
34. H. Xu, J. Yan, Y. Xu, Y. Song, H. M. Li, J. Xia, C. Huang and H. Wan, Appl. Catal., B, 2013, 129, 182–193.
35. B. Chai, T. Y. Peng, J. Mao, K. Li and L. Zan, Phys. Chem. Chem. Phys., 2012, 14, 16745–16752.
36. L. Shi, L. Liang, F. X. Wang, M. S. Liu and J. M. Sun, J. Mater. Sci., 2015, 50, 1718–1727.
37. Y. C. Zhao, Z. Liu, W. G. Chu, L. Song, Z. X. Zhang, D. L. Yu, Y. J. Tian, S. S. Xie and L. F. Sun, Adv. Mater., 2008, 20, 1777–1781.
38. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2009, 25, 10397–10401.
39. J. Hong, X. Xia, Y. Wang and R. Xu, J. Mater. Chem., 2012, 22, 15006–15012.
40. L. Shi, L. Liang, J. Ma, F. X. Wang and J. M. Sun, Dalton Trans., 2014, 7236–7244.

---