β-Cyclodextrin–TiO₂: Green Nest for reduction of nitroaromatic compounds

Abstract

In this paper, highly efficient photocatalytic reduction of nitroaromatic compounds was investigated using β-cyclodextrin (β-CD) and commercial nano-TiO₂ as a host–guest system in water under sunlight irradiation. The nitroaromatic compounds solubilized in water through encapsulation in β-CD formed an inclusion complex which was attached to TiO₂ under sunlight irradiation. This ‘guest–host system’ that we call ‘Green Nest’ showed high efficiency for the reduction of nitro compounds to the corresponding amines, and more interestingly, one pot reductive N-formylation and N-acylation from nitroaromatic compounds can be carried out in the presence of triethyl orthoformate, acetic and benzoic anhydride. The β-CD–TiO₂ was characterized by transmission electron microscopy, UV-visible spectra, thermogravimetric analysis, Brunauer–Emmett–Teller measurements, UV-visible diffuse reflectance spectroscopy, and Raman spectroscopy.

---

Introduction

The concept of photochemistry as a green technology is more than a century old. Sunlight as an unending and natural source could be a key component of industry in the future. This delegates a key role to photochemistry in future industrial processes. Despite the advent of artificial light sources and other leaps in technology, it is not unexpected that mankind turns to the sun as a free energy and light source for organic photochemistry and a true ‘green’ route in organic synthesis.¹ Selective transformation of organic compounds using metal oxide semiconductors as mediators is one of the current challenging methods in organic synthesis from the viewpoints of renewable energy and environmental applications.² Owing to its biochemical compatibility, low cost, non-toxicity, optical and electronic properties and high photo-activity, the main interest among the various metal oxide semiconductors has been shifted to TiO₂.³ Also, because of its excellent photochemical properties, TiO₂ plays a crucial role in a wide range of important redox organic transformations, especially in selective reduction of nitro compounds.⁴ Up to now, a wide variety of traditional methods has been developed for the reduction of nitro compounds to corresponding amines using hazardous reagents but they show poor efficiency.⁵ To overcome these limitations, many attempts have been made at designing new sustainable and environmentally acceptable photocatalytic reductive methods for the nitro group without degrading the conjugated structure in a process that is direct and mild and could also allow transformation to other functional groups. In recent years, a great deal of attention has been paid to the photocatalytic transformation of nitro compounds into amine, imine and benzimidazole functionalities based on TiO₂.⁶

Compared to other organic solvents, water is a green, available, and more importantly cheap and safe solvent and has therefore attracted a lot of attention in organic synthesis.⁷ In contrast to the popularity of ‘TiO₂ nanoparticles’ in photocatalytic transformations in organic solvents, their use in selective organic synthesis in water is still not quite as widespread because of some limitations or disadvantages associated with the low solubility of substrates and over-oxidation of organic compounds which are a result of the oxidant species produced (such as ˙OH and O₂⁻˙/HO₂˙) in a photocatalytic process.⁸ Thus, overcoming these impediments in water is worth highlighting in photocatalytic synthesis. Recently, molecular host–guest complexes have provided one of the promising routes to adsorption of organic molecules on the surface of metal oxide semiconductors in order to improve the interaction of organic substrates with photocatalytically active sites when water is used as the reaction medium. Quite recently, Nichols et al. demonstrated that cyclodextrin–TiO₂ nanowires with an open and porous structure show self-assembly capability as a result of simulated sunlight irradiation in water.⁹ Their marine sponge-like structure, with high porosity and surface area, played a role as host for easy encapsulation of organic compounds. Recent studies indicate an increased photocatalytic performance when the surface of TiO₂ particles is modified with host molecules such as β-cyclodextrin (β-CD).¹⁰ In other words, the modification of TiO₂ with cyclodextrin through interactions between the hydroxyl groups of β-CD and TiO₂ draws the organic molecules near the surface of the photocatalyst to enhance their interaction and also results in an increase in the photocatalytic redox ability of TiO₂.¹¹ In continuation of our effort to design novel systems for photocatalytic organic transformations,¹² in this work we present a highly efficient and green photocatalytic system for selective reduction of the nitro group under sunlight irradiation based on in situ modification of TiO₂ P25 with β-CD in water. We believe that β-CD can enhance the dispersion of TiO₂ in water, and together produce a ‘host nano-reactor system’, that we call the ‘Green Nest’, in order to compensate for the weak interaction of organic nitro compounds with the surface of TiO₂ in an aqueous medium. Also, β-CD can promote the charge transfer rate from the photo-excited TiO₂ to the electron acceptor guest molecule. Thus, it is assumed that the hydrophobic inner cavity of β-CD is a nano-photo-reactor in this system. To investigate this, first nitro compounds were added to an aqueous solution of β-CD, then this inclusion complex of β-CD-nitro compound was adsorbed on the surface of titania by sunlight irradiation, and the reduction of the nitro group to amine function was carried out in the hydrophobic cavity of β-CD. Since amide derivatives such as formanilides are valuable intermediates in organic synthesis,¹³ we attempted to expand this proposed photocatalytic system to one-pot reductive N-formylation and N-acylation.

Results and discussion

β-CD is a well-known host molecule for organic compounds due to its hydrophobic cavity. The hydroxyl groups in an inclusion complex of organic compound/β-CD make it soluble in water.¹⁴ Besides, recent research in the development of high-efficiency photocatalysis processes shows that the presence of a host molecule such as β-CD is very profitable. With this in mind, nitro compounds were added to a solution of β-CD, and then the resulting inclusion complex of β-CD-nitro compound was adsorbed on the surface of titania by sunlight irradiation. Initial experiments were carried out with a water solution containing β-CD (0.1 mmol), 3-nitroacetophenone (0.1 mmol), and commercial TiO₂ (30 mg) under sunlight irradiation for 4 h (Table 1, entry 1) resulting in a 49% conversion according to gas chromatography (GC). Addition of ammonium formate or oxalic acid as a hole scavenger under the same conditions was investigated (Table 1, entries 5 and 2). No conversion of 3-nitroacetophenone was observed without TiO₂ or in the dark (Table 1, entries 6 and 7).

Table 1 Photoreduction of 3-nitroacetophenone using β-CD–TiO₂ for various conditions under sunlight irradiation

Entry	Additive	Time (h)	Yield^a (%)
a GC yield. Luximetry 10–80 × 10^3 Lux.b Without additive.c Without TiO2 P25.d In the dark.
1	—^b	4	49
2	Oxalic acid	1	100
3	Potassium iodide	3	46
4	Sodium sulfate	3	31
5	Ammonium formate	3	95
6	—^c	6	—
7	—^d	6	—

---

β-CD can play a more crucial role in the above mentioned reaction in water. First, β-CD is the host molecule for the nitro compounds and increases their solubility. This fact can be easily observed through the disappearance of nitrobenzene in water in a separate test.

Also, it was reported that the formation of host–guest complexes between β-CD and organic compounds in water affects the UV-visible properties.¹⁵ Therefore, as can be seen in Fig. 1, the intensity of UV-visible spectra of an aqueous solution of p-nitrophenol increases with the addition of a β-CD solution resulting in the formation of a host–guest complex.

Fig. 1 UV-visible absorption spectra of p-nitrophenol (1 × 10⁻⁴ M) in aqueous media and titration of this solution by β-cyclodextrin (1 × 10⁻³ M).

Also, using polyethylene glycol/water, EtOH/water and prototype polyhydroxyl carbohydrates such as glucose showed that although the solubility of nitrobenzene increased, the conversion and product selectivity decreased. An increase in the visible-light activity of TiO₂ can be achieved in combination with β-CD because of the hydroxyl group's interaction with h⁺ on the surface of photocatalyst.¹⁶ To show this fact, the above reaction was carried out in the absence of β-CD. It was observed that the reduction of nitrobenzene in water without β-CD was not a selective and clean reaction with a conversion of 30% after 3 h. On the other hand, we used a mixture of water and ethanol or polyethylene glycol instead of β-CD to increase the solubility of the nitro compounds in water. Nevertheless, a mixture of products for this condition was observed. β-CD molecule is built up from glucose units, which can also reduce the electron–hole recombination rate.¹⁷ To show the importance of the hydrophobic cavity in β-CD, glucose/TiO₂ was used in a similar manner to β-CD–TiO₂. The low yield of arylamine confirmed the crucial role of the β-CD cavity in the high performance of photocatalytic reduction of nitroaromatic compounds in water. Also, in a model experiment, changing the acidity of the mixture through attachment of β-CD to TiO₂ was measured and an increase of acidity from 7 (in the absence of β-CD) to 4 (in the presence of β-CD under sunlight irradiation) was observed. These pH changes can be due to the surface adsorption of β-CD to TiO₂ particles and formation of the β-CD–TiO₂ composite in water under sunlight irradiation¹⁸ (Scheme 1).

Scheme 1 pH enhancement of a β-CD–TiO₂ solution during sunlight irradiation results in deprotonation from a hydroxyl group of the CD molecules on TiO₂.

In the next step, studies were extended to the reduction of different types of nitroaromatic compounds including functional groups such as carbonyl, nitrile and halide using the β-CD–TiO₂ system in water under sunlight irradiation (Table 2). Several aromatic nitro compounds with electron-donating or electron-withdrawing functional groups were reduced to the corresponding amines (Table 2, entries 8–13). Consequently, a higher reduction conversion of the nitroaromatic compounds in the presence of electron-withdrawing groups was observed. High chemoselectivity of the reaction was shown with reduction of nitro functional group in the presence of other sensitive functional groups such as carbonyl and nitrile that remained intact during the photocatalytic reduction reaction (Table 2, entries 4 and 11). Also, reduction of 1,2-dinitrobenzene to the corresponding diamine was carried out (Table 2, entry 6). When the same reaction was carried out in a solution of ethanol–water (2–13 mL), benzimidazole was obtained with 88% yield (Table 2, entry 7). When the reaction was carried out on a greater scale, 2 mmol of nitrobenzene, the product (aniline) was obtained in 83% isolated yield after 12 h under solar light irradiation (Table 2, entry 1^b).

Table 2 Reduction of nitro compounds using β-CD–TiO₂ in water under sunlight irradiation

Entry	Substrate	Product	Time [h]	Yield^a [%]
a GC yield using an internal standard (biphenyl) method unless otherwise stated. Reaction conditions: nitro compounds (0.1 mmol), TiO2 (30 mg), β-CD (0.1 mmol), water (15 mL) and oxalic acid (15 mg) or ammonium formate (25 mg) was degassed by Ar gas and irradiated with solar light.b The reaction was carried out in 2 mmol of nitrobenzene with the following reaction conditions: nitro compounds (2 mmol), TiO2 (0.5 g), β-CD (0.8 mmol), water (60 mL) and ammonium formate (0.1 g), 12 h, under solar light irradiation, isolated yield.c For synthesis of benzimidazole: with the same conditions using a mixture of ethanol (2 mL) and water (13 mL). Daily sunlight (9 am–3 pm; sunlight intensity of 10–80 × 10^3 Lux).
1			3	100 (83)^b
2			1	100
3			2	100
4			1.5	92
5			3	95
6			3.5	95
7			5	88^c
8			2	100
9			3	96
10			4	89
11			1.5	100
12			1.5	100
13			3.5	95

---

The mole ratio of nitro to ammonium formate is 1 to 3.2 and nitro to oxalic acid is 1 to 1.7. The ammonium formate and oxalic acid convert to ammonia, carbon dioxide or carbon dioxide during photocatalytic reaction respectively. We found that, in an experimental control test, when the reaction was carried out in the presence of excess amount of ammonium formate, amine product was obtained in quantitative yield after separation by extraction with EtOAc. The aqueous phase including excess ammonium formate can be used for the next reduction cycle without loss of the conversion yield (see ESI, Scheme S1†).

The important factors in photocatalytic applications in aqueous medium are the stability and reusability of the nanomaterial at the end of the reaction. To study this, the photocatalytic performance of β-CD–TiO₂ was investigated after four usages (Fig. 2). At the end of each run, ethyl acetate was added to the aqueous mixture and sonicated for 5 min. The aqueous solution containing β-CD–TiO₂ nanoparticles was reused in the next run. A decrease in the photocatalytic efficiency can be due to the partial loss of photocatalyst during product separation. These results are consistent with the retention of good photocatalytic activities of β-CD–TiO₂ after four usages.

Fig. 2 Reusability and photocatalytic efficiency of the β-CD–TiO₂ aqueous solution for reduction of nitrobenzene under sunlight irradiation after 3 h.

Various characterizations of β-CD–TiO₂ confirm the presence of β-CD in the titania surface after 3 h sunlight irradiation of the solution and β-CD. Fig. S5† shows the thermogravimetric analysis (TGA) of β-CD–TiO₂ after separation and washing several times with water. From the TGA curve it was observed that there was around 4% CD in β-CD–TiO₂. Raman spectra show a band located that 2800–3000 cm⁻¹ (C–H) which is evidence of β-CD in β-CD–TiO₂ nanocomposite (Fig. S9 in ESI†). However, as presented in Scheme 2, there were no differences in the lattice structure in TEM images of β-CD–TiO₂ and TiO₂. Considering the UV-visible diffuse reflectance spectroscopy analysis of β-CD–TiO₂ (Fig. S6 in the ESI†), the latter displayed a higher absorption in the visible light region than TiO₂.¹⁹ Also, a decrease in surface area of β-CD–TiO₂ compared to TiO₂ based on the Brunauer–Emmett–Teller (BET) method can be attributed to the attachment of β-CD to the titania surface under sunlight irradiation (Table S1 in the ESI†).

Scheme 2 TEM images of β-CD–TiO₂ (a) and TiO₂ (b).

FTIR spectral analysis of β-CD–TiO₂ hybrid nanoparticles using TiO₂ as a reference was carried out (Fig. 3). In comparison with P25 (Fig. S11 in the ESI†), in the spectrum of TiO₂–β-CD the intensity increases at 3400 cm⁻¹. This result can be due to the presence of the O–H groups of β-CD. The peak centered at 1676 cm⁻¹ was assigned as the stretching of the C=C bonds. Also, the absorption at 1414 cm⁻¹ was assigned to the O–H in-plane bending, while the antisymmetric C–O–C stretch was assigned at 1156 cm⁻¹. The absorption at 1030 cm⁻¹ was assigned as the C–O stretching vibrations.²⁰

Fig. 3 FTIR spectrum of β-CD–TiO₂ hybrid nanoparticles.

In continuation, we studied a green, one-pot protocol for further transformation of photocatalytic-produced amines into valuable N-formylated and N-acylated amines in water (Scheme 3). Carboxylic anhydrides and triethyl orthoformate were used as acylation and formylation agents. In this area, Lou et al. have reported the reductive N-formylation of nitroarene compounds using Au–TiO₂ in the presence of ammonium formate at two temperatures (room temperature and reflux condition) in CH₃CN as an aprotic solvent.²¹ According to these findings, when an excess amount of ammonium formate was used in our method, no N-formylated product was detected after 5 h. Then, in the next attempt, triethyl orthoformate was used as a more active formylating agent. Interestingly, in the presence of triethyl orthoformate, high yield of the corresponding amide was achieved. Moreover, using carboxylic anhydrides as acylation agents also efficiently produced the corresponding amides. As summarized in Table 3, various types of nitroaromatics including other functional groups were selectively transformed into their corresponding amide compounds in water. The photocatalytic system also showed excellent regioselective N-formylation of p-nitrophenol with very high yield (Table 3, entry 10). As expected, when the reaction was carried out in 2 mmol of nitrobenzene, the anilide was obtained in notable decreased isolated yield after 12 h.

Scheme 3 One-pot N-formylation and N-acylation of photocatalytic produced amines through encapsulated nitrobenzene in β-CD under sunlight.

Table 3 One-pot synthesis of amides from the photocatalytic reduction of nitro compounds using triethyl orthoformate or anhydride

Entry	Substrate	Product	Time (h)	Yield^a (%)
a Isolated yields using column chromatography for N-acylation products (1–5) and GC yields for N-formylation products (6–10). Reaction conditions: nitro compounds (0.1 mmol), TiO2 P25 (30 mg), β-CD (0.1 mmol), water (15 mL), ammonium formate (25 mg) and triethyl orthoformate (0.2 mL) or anhydride (0.12 mmol) under sunlight. Daily sunlight (9 am–3 pm; sunlight intensity of 10–80 × 10^3 Lux).b The reaction was carried out in 2 mmol of nitrobenzene with the following reaction conditions: nitro compounds (2 mmol), TiO2 (0.5 g), β-CD (0.8 mmol), water (60 mL), ammonium formate (0.1 g), and triethyl orthoformate (4 mL), 12 h, under solar light irradiation.
1			2.5	100
2			1.5	91
3			3	97
4			2	94
5			3.5	100
6	CH(OEt)3		2	100 (66)^b
7	CH(OEt)3		3	94
8	CH(OEt)3		1	91
9	CH(OEt)3		2.5	90
10	CH(OEt)3		4	89

---

To study the formylation step, in control reactions, the same conditions were used for the amine as the starting material, and excellent yield of the corresponding formamide was obtained even in dark conditions, and in the absence of β-CD. This result demonstrated a high nonphotocatalytic activity of the formylation of amines. It seems that the formylation step had been carried out on the non-encapsulated amine in β-CD. Therefore, to clarify this issue, the titration of o-nitrophenol and o-aminophenol with β-CD was carried out using UV-visible data. As shown in Fig. S3 and S4 (see ESI†), with the addition of β-CD aqueous solution to o-nitrophenol and o-aminophenol, the UV-visible intensity of o-nitrophenol increases more than o-aminophenol because of the formation of the host–guest complex with β-CD in water. Thus, we suggest that the intrinsic physicochemical properties between amines (miscible) and nitro compounds (nonmiscible) caused the reduction of the nitro group in the hydrophobic cavity of β-CD and then N-formylation or N-acylation was performed in water.

Conclusions

In this work we employed β-CD and commercial nano-TiO₂ P25 self-assembled under sunlight irradiation in water for green reduction of nitro compounds. This system shows a highly efficient, eco-friendly, clear and selective reduction of the nitro group into amines through the host–guest model. From the above mentioned result we suggest the encapsulation of guest molecules in inner β-CD cavities, followed by the binding of the subsequent host–guest complexes to the surface of TiO₂ nanoparticles. Interestingly, one-pot N-acylation and N-formylation of in situ prepared amines can be carried out in an aqueous medium using anhydride or triethyl orthoformate.

Experimental section

General procedure for the reduction of nitro compounds

β-CD (0.1 mmol) was dissolved in water (15 mL) by warming to 60 °C until a clear solution was formed. Nitro compounds (0.1 mmol) were added to this solution which was sonicated for 10 min. Then, commercial TiO₂ (P25) (30 mg) and oxalic acid (15 mg) were added into a round-bottom Pyrex flask (25 mL). The reaction mixture was degassed by Ar gas (20 min) and sealed with a septum. Afterwards, the flask was irradiated under stirring with sunlight according to the data in Table 1. After the completion of the reaction, a suitable amount of NaHCO₃ was added to control the pH at around 7 and the mixture was stirred at room temperature. The organic material was extracted with ethyl acetate. The organic phase was then dried using anhydrous MgSO₄. The reaction mixture was analyzed with a gas chromatograph.

General procedure for one-pot N-formylation and N-acylation of nitro compounds

β-CD (0.1 mmol) was dissolved in water (15 mL) by warming to 60 °C until a clear solution was formed. Nitro compounds (0.1 mmol) were added to this solution which was sonicated for 10 min. Commercial TiO₂ (P25) (30 mg), ammonium formate (25 mg) and triethyl orthoformate (0.2 mL) for formylation or anhydride (0.12 mmol) were added into a round-bottom Pyrex flask (25 mL). The reaction mixture was degassed by Ar gas (20 min) and sealed with a septum. Afterwards, the flask was irradiated under stirring with sunlight according to the data in Table 3. The organic material was extracted with ethyl acetate. The organic phase was then dried (anhydrous MgSO₄), filtered, and the solvent was removed under vacuum. Pure products were obtained after recrystallization or by column chromatography on silica using n-hexane and ethyl acetate mixture as an eluent (SiO₂; n-hexane–EtOAc 5 : 1).

Acknowledgements

The authors acknowledge the support by IASBS Research Council of this work.

Notes and references

1. CRC Handbook of Organic Photochemistry and Photobiology, ed. A. G. Griesbeck, M. Oelgemöller and F. Ghetti, 2012, vol. 1; B. Pohlmann, H. D. Scharf, U. Jarolimek and P. Mauermann, Solar Energy, 1997, 61, 159; E. Haggiage, E. E. Coyle, K. Joyce and M. Oelgemöller, Green Chem., 2009, 11, 318.
2. X. Lang, X. Chen and J. Zhao, Chem. Soc. Rev., 2014, 43, 473; H. Kisch, Angew. Chem., Int. Ed., 2013, 52, 812; O. S. Mohamed, A. E. M. Gaber and A. A. Abdel-Wahab, J. Photochem. Photobiol., A, 2002, 148, 205; S. Higashimoto, N. Kitao, N. Yohida, T. Sakura, M. Azuma, H. Ohue and Y. Sakata, J. Catal., 2009, 266, 279; S. Yurdakal, V. Augugliaro, V. Loddo, G. Palmisano and L. Palmisano, New J. Chem., 2012, 36, 1762.
3. A. Fujishima, K. Hashimoto and T. Watanabe, TiO2 Photocatalysis: Fundamentals and Applications, BKC Inc., Tokyo, Japan, 1999; C. Liao, C. Ch. W. Huang and J. C. S. Wu, Catalysts, 2012, 2, 490.
4. F. Mahdavi, T. C. Bruton and Y. Li, J. Org. Chem., 1993, 58, 744; J. L. Ferry, J. Phys. Chem. B, 1998, 102, 2239; J. L. Ferry, Langmuir, 1998, 14, 3551; S. Füldner, T. Mitkina, T. Trottmann, A. Frimberger, M. Gruber and B. König, Photochem. Photobiol. Sci., 2011, 10, 623; S. Liu, M. Q. Yang and Y. G. Xu, J. Mater. Chem. A, 2014, 2, 430.
5. E. S. Chan-Shing, D. Boucher and J. Lessard, Can. J. Chem., 1999, 77, 687; D. C. Gowda, Synth. Commun., 2003, 33, 281; B. K. Banik, C. Mukhopadhyay, M. S. Venkatraman and F. F. Becker, Tetrahedron Lett., 1998, 39, 7243.
6. A. Hakki, R. Dillert and D. Bahnemann, Catal. Today, 2009, 144, 154; H. Wang, R. E. Partch and Y. Li, J. Org. Chem., 1997, 62, 5222.
7. A. Chanda and V. V. Fokin, Chem. Rev., 2009, 109, 725.
8. C. C. Chen, W. H. Ma and J. C. Zhao, Chem. Soc. Rev., 2010, 39, 4206.
9. W. T. Nichols and S. Yoon, Appl. Surf. Sci., 2013, 285, 517.
10. G. Wang, L. Yang, F. Wu and N. Deng, J. Chem. Technol. Biotechnol., 2013, 89, 297; X. Zhang, F. Wu and N. Deng, J. Hazard. Mater., 2011, 185, 117; S. H. Toma, J. A. Bonacin, K. Araki and H. E. Toma, Surf. Sci., 2006, 600, 4591; S. Vasudevan and R. Chalasani, ACS Nano, 2014, 7, 4093.
11. L. Lu, F. Wu and N. Deng, Appl. Catal., B, 2004, 53, 87.
12. P. Eskandari, F. Kazemi and Z. Zand, J. Photochem. Photobiol., A, 2014, 274, 7; P. Eskandari, F. Kazemi and Y. Azizian-Kalandaragh, Sep. Purif. Technol., 2013, 120, 180; Z. Zand, F. Kazemi and S. Hosseini, Tetrahedron Lett., 2014, 55, 338; S. Abedi, B. Karimi, F. Kazemi, M. Bostina and H. Vali, Org. Biomol. Chem., 2013, 11, 416; M. Abdollahi Kakroudi, F. Kazemi and B. Kaboudin, J. Mol. Catal. A: Chem., 2014, 392, 112.
13. B. C. Chen, M. S. Bednarz, R. Zhao, J. E. Sundeen, P. Chen, Z. Shen, A. P. Skoumbourdis and J. C. Barrish, Tetrahedron Lett., 2000, 41, 5453; K. Kobayashi, S. Nagato, M. Kawakita, O. Morikawa and H. Konishi, Chem. Lett., 1995, 575; B. Kaboudin and M. Khodamorady, Synlett, 2010, 2905.
14. F. Kazemi and M. Abdollahi Kakroudi, Bull. Korean Chem. Soc., 2013, 34, 500; B. Kaboudin, R. Mostafalu and T. Yokomatsu, Green Chem., 2013, 15, 2266.
15. I. Shown, M. Ujihara and T. Imae, J. Colloid Interface Sci., 2010, 352, 232.
16. M. H. Du, J. Feng and S. B. Zhang, Phys. Rev. Lett., 2007, 98.
17. X. Zhang, X. Li and N. Deng, Ind. Eng. Chem. Res., 2012, 51, 704.
18. D. Bahnemann, A. Henglein, J. Lilie and L. Spanhel, J. Phys. Chem., 1984, 88, 709.
19. T. Tachikawa, S. Tojo, M. Fujitsuka and T. Majima, Chem.–Eur. J., 2006, 12, 7585; T. Rajh, L. X. Chen, K. Lukas, T. Liu, M. C. Thurnauer and D. M. Tiede, J. Phys. Chem. B, 2002, 106, 10543.
20. B. Casu, M. Reggiani, G. G. Gallo and A. Vigevani, Tetrahedron, 1968, 24, 803.
21. X. B. Lou, L. He, Y. Qian, Y. M. Liu, Y. Cao and K. N. Fan, Adv. Synth. Catal., 2011, 353, 281.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, ¹H NMR and ¹³C NMR spectra of the synthesized compounds, UV-vis spectra, TGA, Raman and BET analysis. See DOI: 10.1039/c4ra08059a

---