Introduction Over the past several decades, inorganic hollow spheres have received an enormous amount of attention owing to their wide application in catalysis, water treatment, lithium-ion batteries, drug-delivery controlled carriers, sensors, antibacterial materials and so on.1–4 Up to now, many methods have been developed for the synthesis of hollow micro/nanospheres.5 Among these methods, template-assisted synthesis is extensively applied as a facile and effective pathway to fabricate designed functional products with specific architectures through replication of the structures and morphologies of the templates, such as silica particles, polymer latex spheres, carbon spheres, emulsion droplets, micelles/vesicles and gas bubbles.6–14 Due to the improved physical and chemical properties compared to single-component hollow spheres, hybrid hollow spheres are of great importance to a potentially broader range of applications in materials science.15–18 Titanium oxide (TiO2), with a wide band gap of 3.2 eV, is a very promising photocatalytic material due to its nonphotocorrosive properties, nontoxicity, low-cost and photostability.19 However, because of the poor utilization of solar energy and the short diffusion length of a photogenerated electron–hole pair, TiO2 shows low quantum efficiency in photocatalytic reactions. Therefore, it is of great importance to extend the photoresponse of TiO2 into the visible region.20 Many efforts have been attempted by modifying TiO2 with metals, such as Pt, Au, Ru, Pd, and Ag, so as to efficiently extend the photo-response from UV to visible-light.21–23 Among them, owing to the low cost and nontoxicity relative to other noble metals, Ag-doped TiO2 nanohybrids are particularly suitable for industrial applications. Moreover, the deposition of Ag onto the surface of TiO2 has been shown to be beneficial for maximizing the efficiency of photocatalytic reactions. Ag particles can act as electron acceptor centers, causing electron and hole pair separation. Simultaneously, their Fermi level is below the conduction band of TiO2, leading to a drastically improved photocatalytic activity of TiO2 and an increased quantum yield for photocatalytic processes.24,25 Recently, a variety of strategies have been carried out to prepare TiO2–Ag nanocomposites. The widely used methods include photodeposition, chemical deposition and a conventional impregnation method.26–29 To the best of our knowledge, there have been few reports on the synthesis of TiO2–Ag hybrid hollow nanostructures. Due to their facile mobility, low density, high specific surface area and superior catalytic activity, hollow structures have received considerable attention in the field of photocatalysis.30 Herein, TiO2–Ag hybrid hollow spheres with a highly uniform morphology and good structural stability were facilely synthesized via a one-pot hydrothermal method using carbon spheres as templates followed by an annealing treatment. The detailed procedure for the synthesis of TiO2–Ag hybrid hollow spheres is illustrated in Scheme 1. Furthermore, the loading amount of the anchored Ag nanocrystals is conveniently controlled by varying the concentration of silver nitrate (AgNO3). As expected, the as-prepared hollow hybrids exhibit excellent photocatalytic activity for the degradation of rhodamine B (RhB) and methyl orange (MO) dyes when exposed to visible-light irradiation. Accordingly, optimum matching for the best photocatalytic activity was investigated thoroughly and a reasonable mechanism was further proposed to explain the role of Ag nanocrystals in the TiO2–Ag hybrids. 2. Materials and methods All reagents were analytical grade, purchased from the Shanghai Chemical Reagent Factory and used as received without further purification. Synthesis of the TiO2–Ag hybrid hollow spheres Monodisperse carbon spheres were synthesized in a similar manner to what we presented previously.31 Briefly, 3.96 g of glucose was dissolved in 40 mL of distilled water, forming a clear solution. The solution was transferred into a 50 mL autoclave with a Teflon seal, maintained at 180 °C for 8 h and finally black products were obtained. For the preparation of the TiO2–Ag hybrid, 0.2 g of carbon spheres, 0.096 g of titanium sulfate (Ti(SO4)2), 0.288 g of urea and a predetermined amount of AgNO3 were dissolved in 40 mL of distilled water with the assistance of ultrasonication sonication for 10 min. The mixture was then transferred into a 50 mL Teflon-lined autoclave and maintained at 150 °C for 10 h. After collection by centrifugation, the products were washed with ethanol and distilled water three times before being dried at 80 °C for more than 6 h. The final hybrid hollow spheres were obtained after calcination of the above samples at 400 °C in static air for 2 h. The products prepared with different amounts of AgNO3 (0, 0.00025, 0.0005, 0.00075, 0.00100 and 0.00125 M) after the hydrothermal route were assigned sample codes of pure TiO2, Sa, Sb, Sc, Sd, and Se, respectively. Characterization Powder X-ray diffraction (XRD) measurements of the samples were performed with a Philips PW3040/60 X-ray diffractometer using Cu–Kα radiation at a scanning rate of 0.06° s−1 . Scanning electron microscopy (SEM) was performed with a Hitachi S-4800 scanning electron microanalyzer with an accelerating voltage of 15 kV. Transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM) and selected area electron diffraction (SAED) were conducted at 200 kV with a JEM-2100F field emission TEM. Energy dispersive X-ray spectrometry (EDS) was performed with a spectroscope attached to an HRTEM. The samples used for the TEM measurements were prepared by dispersing the products in ethanol and placing several drops of the suspension on holey carbon films supported by copper grids. Further evidence for the composition of the product was inferred from X-ray photoelectron spectroscopy (XPS), using an ESCALab MKII X-ray photoelectron spectrometer with Mg–Kα X-rays as the excitation source. PL spectra were recorded on an Edinburgh FLSP920 fluorescence spectrometer and the absorption spectra were measured using a PerkinElmer Lambda 900 UV-vis spectrophotometer at room temperature. Photocatalytic activity of the TiO2–Ag hollow hybrid spheres The photocatalytic activities of the TiO2–Ag hybrids were evaluated by the degradation of RhB, MO and benzoic acid under visible-light irradiation from a 500 W Xe lamp (CEL-HXF300)with a 420 nm cut-off filter.

The reaction cell was placed in a sealed black box with the top opened and the cut-off filter was placed to provide visible-light irradiation. In a typical process, 0.05 g of the photocatalyst was added into 100 mL of a solution containing one of the above organic dyes with a concentration of 5 mg L−1 . After being dispersed in an ultrasonic bath for 5 min, the solution was stirred for 2 h in the dark in order to reach an adsorption equilibrium between the catalyst and the solution and was then exposed to visible-light irradiation. The samples were collected by centrifugation at given time intervals to measure the RhB, MO and benzoic acid degradation concentration by UV-vis spectroscopy. To study the regeneration and reusability of the photocatalyst, the photocatalyst was collected by centrifugation and then washed with ethanol and distilled water three times before being redispersed in a aqueous solution of a fresh dye for use in the next cycle. ˙OH radical reactions were performed as follows. 5 mg of the different hybrid hollow sphere samples were suspended in 8 mL of an aqueous solution containing 10 mM of NaOH and 5 mM of terephthalic acid (TA), respectively. Before exposure to visible-light, the suspension was stirred in the dark for 30 min. After being irradiated for 30 min, the solutions were centrifuged for fluorescence spectroscopy measurements. A fluorescence spectrophotometer was used to measure the fluorescence signal of the 2-hydroxy-terephthalic acid (TAOH) generated. The wavelength of the excitation light used for recording the fluorescence spectra was 320 nm. 3. Results and discussion The XRD patterns of the as-prepared TiO2–Ag hybrid hollow spheres obtained with different concentrations of Ag+ ions are shown in Fig. 1. For pure TiO2 hollow spheres (pattern a), all the diffraction peaks can be indexed as body-centered tetragonal TiO2 (JCPDS standard card no. 89-4921) with lattice constants of a = 3.777 Å and c = 5.9.501 Å. No impurity peaks are detected which shows that the products are pure phase. Patterns b, c, d, e and f belong to the as-prepared TiO2–Ag hybrid hollow spheres with different concentrations of AgNO3. In addition to the obvious TiO2 patterns, broadened diffraction peaks at 2θ values of 38.1°, 44.3° and 64.5° are observed, which match well with the (111), (200) and (220) crystal planes of face-centered cubic (fcc) Ag (JPCDS standard card No. 89- 3722, a = 4.085 Å). Because the diffraction peak of the Ag (111) crystal plane is very adjacent to that of the TiO2 (004) plane, they overlap each other. With the increase of the Ag+ concentration, the diffraction intensity of the Ag crystal plane gradually becomes strong, which indicates that the Ag nanocrystals grow larger and larger. In addition, the average size of the Ag nanocrystals as the concentration of AgNO3 increases, calculated using the Debye–Scherrer equation based on the full width at half-maximum of the diffraction peak, are 12.7, 18.1, 19.5, 21.1 and 23.1 nm, respectively. A typical SEM image of the TiO2 hollow spheres and the TiO2–Ag hybrid hollow spheres synthesized using different concentrations of AgNO3 are shown in Fig. 2. Fig. 2a shows pure TiO2 hollow spheres of about 700 nm in diameter with a smooth surface. For the TiO2–Ag hybrid hollow spheres, in spite of the morphology the hollow interior of the TiO2 hollow spheres are unchanged after hydrothermal treatment and Ag deposition results in all the hybrid hollow spheres having quite a rough surface. Moreover, our experimental results reveal that the size and loading amount of Ag nanocrystals on the surface of the TiO2 hollow spheres depend on the AgNO3 concentration. From Fig. 2b–f, on increasing the concentration of AgNO3, the surface of the hollow spheres became rougher and rougher, implying an increasing conversion from TiO2 hollow spheres to TiO2–Ag hybrid hollow spheres and this result is consistent with XRD. The geometrical structure and hollow interior of the as-obtained TiO2–Ag hybrid hollow spheres are further elucidated by the TEM results, as shown in Fig. 3. From these images, it can be seen that a large quantity of Ag nanocrystals is well distributed on the surface of the TiO2 hollow spheres and increase and grow larger on increasing the Ag+ concentration. Furthermore, the HRTEM image (inset in Fig. 3c) of the surface layer of the hybrid hollow sphere shows an individual Ag crystal size is about 20 nm, which is in agreement with the XRD pattern. The EDS pattern of the surface layer of the as-prepared hybrid hollow sphere is shown in the inset of Fig. 3d, which further confirms the coexistence of Ag and TiO2 components. To further investigate the distribution of the Ag and TiO2 nanocrystals, the HRTEM image of the edge of a single hybrid hollow sphere (sample Sc) is displayed in Fig. 4. The clear lattice fringe of d = 0.35 nm matches that of the (101) plane of the anantase structure of TiO2 and the fringe spacing at 0.24 nm corresponds to the (111) plane of the cubic phase of Ag. No other impurities such as silver oxides are detected. Additionally, the border of the Ag and TiO2 nanocrystals is almost invisible, which indicates that composition fusion may occur between the interface of the Ag and TiO2. More information on the elemental composition and chemical state of the as-prepared pure TiO2 hollow spheres and TiO2–Ag hybrid hollow spheres (sample Se) is provided by XPS. The fully scanned spectra (Fig. 5a) show that the pure TiO2 hollow spheres contain Ti, O and C elements and the hybrid sample contain Ag, Ti, O and C elements. Herein, the C element may be ascribed to an adventitious carbon-based contaminant. To obtain further evidence about the interaction between the Ag nanocrystals and TiO2 support, high resolution XPS spectra of Ti 2p and Ag 3d are displayed in Fig. 5b–d. From Fig. 4b, the peak position for Ti 2p in the TiO2–Ag hybrids shifts to a higher binding energy band than that in pure TiO2. This result confirms a lower electron density for the Ti atoms in the TiO2–Ag hybrids and there is a strong interaction between metallic Ag and TiO2. Additionally, the XPS peak for Ti 2p3/2 of the TiO2–Ag sample can be fitted into two components, one located at 458.9 eV, attributed to a Ti(IV) species, and the other located at 458.2 eV, assigned to a Ti(III) species (in Fig. 4c), further indicating the strong interaction formed between Ag and the TiO2 species. Ti(III) oxide is a narrow band-gap semiconductor and its energy level is located between the valence band and the conduction band of TiO2, which may be advantageous to the higher photocatalytic activity of the TiO2–Ag hybrids driven by visible-light.24,32 The high resolution XPS spectrum of Ag 3d in the hybrids (in Fig. 5d) shows two individual peaks located at 367.8 and 373.8 eV, which could be assigned to Ag 3d5/2 and Ag 3d3/2. Compared to bulk Ag (368.3 eV for Ag 3d5/2 and 374.3 eV for Ag 3d3/2), an obviously negative shift is observed, which further indicates that some electrons may migrate from the TiO2 to metallic Ag and there is a strong interaction between the Ag nanocrystals and TiO2 support. Because the Ag nanocrystals can act as electron acceptors and help to separate the photoexcited electron–hole pairs, the hybrid structure can inhibit the recombination of excited electrons and holes and then enhance the photocatalytic activity.20 The photocatalytic activities of the as-prepared TiO2–Ag hybrid hollow spheres are evaluated by the degradation of the organic dyes RhB and MO under visible-light irradiation, as shown in Fig. 6. Fig. 6a presents the visible-light photodegradation behaviors of RhB using the as-prepared TiO2–Ag hybrid hollow spheres obtained with different concentrations of a Ag precursor as catalysts. For comparison, commercial Degussa P25 is also used as a photocatalytic reference to quantitatively understand the photocatalytic activity of the TiO2–Ag hybrid catalysts. It can be seen that the concentration of RhB changes little using P25 as the catalyst under visible-light irradiation for 60 min, indicating the weak photo-response of P25 in the visible region. As expected, all the TiO2–Ag hybrid hollow spheres exhibit remarkable improvement in their photocatalytic activity under visible-light illumination. When the concentration of AgNO3 was less than 0.0075 M, the photoactivity was enhanced steadily with an increase in the concentration of AgNO3. However, further increases in the concentration of AgNO3 during the synthesis process results in decreasing photocatalytic activity. Specifically, sample Sc (MAg+ = 0.0075 M) displays the best activity among all the samples. The order of the photocatalytic activity for the samples can be summarized as follows: Sc > Sb > Sd > Sa > Se > pure TiO2 hollow spheres >P25. Fig. 6b shows the photodegradation behaviors of MO catalyzed by P25 and different hybrids under visiblelight illumination, which exhibits a similar regularity to the degradation of RhB. The superior photocatalytic performance of the TiO2–Ag hybrid hollow spheres may be ascribed as follows and a probable schematic diagram representing the charge transfer process in the hybrids is illustrated in Scheme 2. Firstly, the Ag nanocrystals are photoexcited owing to plasmon resonance under visible-light illumination and the photoexcited electrons migrate from the surface of the Ag nanocrystals to the conduction band of TiO2. 33 Due to the high crystallinity of the Ag nanocrystals, the resistance in the electron migration is relatively low, which leads to a reduction in the recombination of excited electrons and holes and then increases the photocatalytic performance. Secondly, the strong interaction between Ag and TiO2 will also result in a higher molar ratio of Ti(III) species on the surface, as revealed by the XPS spectra. Because Ti(III) oxide is a narrow band gap semiconductor and its energy level is located between the valence band and the conduction band of TiO2, the electrons could be excited into the conduction band of TiO2 under visible-light irradiation.19 However, over-accumulation of the Ag content can also act as recombination centers, leading to a decrease in the concentration of photogenerated charge carriers and the photocatalytic activity of the photocatalyst.34 Furthermore, higher surface loadings may decrease the catalytic efficiency of the semiconductor due to the reductive availability of the semiconductor surface for light absorption and pollutant adsorption.35 Therefore, the photocatalytic activity for the sample Sc (MAg+ = 0.0075 M) photocatalyst displays a larger activity than the other samples. Fig. 6c displays the photodegradation behavior of benzoic acid in the presence of pure TiO2 hollow spheres and sample Sc. It can be seen that benzoic acid can be degraded in the presence of TiO2–Ag hybrid hollow spheres. However, there is almost no obvious change in the presence of pure TiO2. Thus, this result further validates the TiO2–Ag nanohybrids as photocatalysts which can be excited by visible-light. We have also evaluated the reusability of the TiO2–Ag photocatalyst for the photodegradation of RhB, as shown in Fig. 6d. After six cycled runs of the photodegradation of RhB, the photocatalytic activity of the as-prepared TiO2–Ag hybrid hollow spheres does not show any obvious deterioration. It clearly demonstrates that the samples are quite stable and have great application potential in water treatment. Additionally, the photocatalytic effect of the hollow sphere structures for the degradation of RhB is further described in Fig. S1 (see ESI†), which clearly shows that the TiO2–Ag hybrid hollow spheres (sample Sc) are more efficient than the as-prepared P25-Ag and TiO2–Ag under the same conditions. Furthermore, the ˙OH radicals formed during photocatalysis with different hybrid hollow spheres could be probed using a method described previously.36 It is well known that ˙OH reacts with TA in basic solutions to generate TAOH, which emits a unique fluorescence signal with its peak centered at 426 nm.37 Fig. 7 shows significant fluorescent signals associated with TAOH, generated upon visible-light irradiation of the different catalysts suspended in a TA solution for 30 min, which clearly demonstrate that the photoexcited holes are powerful enough to oxidize surface-adsorbed hydroxyl groups and water to generate ˙OH radicals. Additionally, most ˙OH radicals are formed when using sample Sc as the catalyst in the photoreaction process, and this result is in a good agreement with that of the photodegradation of the dyes. Thus, in these hybrid systems, Ag nanocrystals might be responsible for the visible-light induced photocatalytic degradation by improving the photogenerated electron and hole separation as well as charge migration, allowing both the electrons and holes to partake in the overall photocatalytic reaction. 4. Conclusions In summary, highly uniform TiO2–Ag hybrid hollow spheres, about 700 nm in diameter, have been successfully synthesized by a facile one-pot hydrothermal method using carbon spheres as templates followed by an annealing treatment. Using this method, the loading amount of the Ag nanocrystals could be varied or controlled by the concentration of AgNO3. As expected, these hybrid hollow spheres exhibit much higher visible-light-driven photocatalytic activities for the degradation of organic pollutes, which is mainly attributed to surface plasmon resonances of the Ag nanocrystals excited by visiblelight. Additionally, the presence of Ti(III) in the TiO2–Ag hybrids might contribute to higher photocatalytic activity. Furthermore, the optimal synergistic effect between TiO2 and Ag was investigated via the ˙OH radicals formed during photocatalysis with different hybrids. This work not only demonstrates a facile synthesis of TiO2–Ag hybrid hollow spheres for a photocatalytic application, but can also be extended to synthesize other hybrid hollow nanostructures for different applications. Acknowledgements Financial support from the Natural Science Foundation of China (21171146) and Zhejiang Provincial Natural Science Foundation of China (Y4110304) is gratefully acknowledged. J. F. Chen acknowledges the financial support from the Anhui Provincial Natural Science Foundation (11040606M53). References 1 B. Wang, J. S. Chen, H. B. Wu, Z. Y. Wang and X. W. Lou, J. Am. Chem. Soc., 2011, 133, 17146. 2 H. M. Chen, J. H. He, H. M. Tang and C. X. Yan, Chem. Mater., 2008, 20, 5894. 3 P. Jiang, J. F. Bertone and V. L. Colvin, Science, 2001, 291, 453. 4 Y. F. Zhu, J. L. Shi, W. H. Shen, X. P. Dong, J. W. Feng, M. L. Ruan and Y. S. Li, Angew. Chem., 2005, 117, 5213. 5 Q. Zhang, W. S. Wang, J. Goebl and Y. D. Yin, Nano Today, 2009, 4, 494. 6 S. J. Ding, J. S. Chen, G. G. Qi, X. N. Duan, Z. Y. Wang, E. P. Giannelis, L. A. Archer and X. W. Lou, J. Am. Chem. Soc., 2011, 133, 21. 7 C. Ge, D. Z. Zhang, A. L. Wang, H. B. Yin, M. Ren, Y. M. Liu, T. S. Jiang and L. B. Yu, J. Phys. Chem. Solids, 2009, 70, 1432. 8 X. M. Sun, J. F. Liu and Y. D. Li, Chem.–Eur. J., 2006, 12, 2039. 9 C. J. Martinez, B. Hockey, C. B. Montgomery and S. Semancik, Langmuir, 2005, 21, 7937. 10 Y. Liu, Y. Hu, C. F. Guo and X. Hu, J. Am. Ceram. Soc., 2011, 94, 1667. 11 K. Nam, J. Shim, H. Ki, S. I. Choi, G. Lee, J. Jang, Y. Jo, M. H. Jung, H. Song and J. Park, Angew. Chem., Int. Ed., 2008, 47, 9504. 12 Y. Hu, J. F. Chen, W. M. Chen, X. H. Lin and X. L. Li, Adv. Mater., 2003, 15, 726. 13 S. S. Kim, W. Zhang and T. Pinnavaia, Science, 1998, 282, 1302. 14 J. Liu, S. Z. Qiao, S. B. Hartono and G. Q. Lu, Angew. Chem., Int. Ed., 2010, 49, 4981. 15 F. Teng, T. Xu, S. Liang, G. Buergen, W. Yao and Y. Zhu, Catal. Commun., 2008, 9, 1119. 16 W. Zhang, J. Hu, Y. Guo, S. Zheng, L. Zhong, W. Song and L. Wan, Adv. Mater., 2008, 20, 1160. 17 P. Wu, N. Du, H. Zhang, C. X. Zhai and D. R. Yang, ACS Appl. Mater. Interfaces, 2011, 3, 1946. 18 Y. Liu, L. Yu, Y. Hu, C. F. Guo, F. M. Zhang and X. W. Lou, Nanoscale, 2012, 4, 183. 19 M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann, Chem. Rev., 1995, 95, 69. 20 J. X. Li, J. H. Xu, W. L. Dai and K. N. Fan, J. Phys. Chem. C, 2009, 113, 8343. 21 R. E. Rettew, N. K. Allam and F. M. Alamgir, ACS Appl. Mater. Interfaces, 2011, 3, 147. 22 Y. Y. Song, Z.D. Gao and P. Schmuki, Electrochem. Commun., 2011, 13, 290. 23 X. H. Li, G. Y. Chen, L. B. Yang, Z. Jin and J. H. Liu, Adv. Funct. Mater., 2010, 20, 2815. 24 C. Y. Su, L. Liu, M. Y. Zhang, Y. Zhang and C. L. Shao, CrystEngComm, 2012, 14, 3989. 25 H. M. Sung-Suh, J. R. Choi, H. J. Hah, S. M. Koo and Y. C. Bae, J. Photochem. Photobiol., A, 2004, 163, 37. 26 Y. H. Zheng, C. Q. Chen, Y. Y. Zhan, X. Y. Lin, Q. Zheng and K. M. Wei, J. Phys. Chem. C, 2008, 112, 10773. 27 B. M. Wen, C. Y. Liu and Y. Liu, Inorg. Chem., 2005, 44, 6503. 28 M. Cha, C. Shin, D. Kannaiyan, Y. H. Jang, S. T. Kochuveedu, D. Y. Ryu and D. H. Kim, J. Mater. Chem., 2009, 19, 7245. 29 B. Cheng, Y. Le and J. G. Yu, J. Hazard. Mater., 2010, 177, 971. 30 J. Zhang, S. W. Liu, J. G. Yu and M. Jaroniec, J. Mater. Chem., 2011, 21, 14655. 31 Y. Hu, Y. Liu, H. S. Qian, Z. Q. Li and J. F. Chen, Langmuir, 2010, 26, 18570. 32 H. W. Chen, Y. Ku and Y. L. Kuo, Chem. Eng. Technol., 2007, 30, 1242. 33 X. Z. Li and F. B. Li, Environ. Sci. Technol., 2001, 35, 2381. 34 H. Tahiri, Y. A. Ichou and J. M. Herrmann, J. Photochem. Photobiol., A, 1998, 114, 219. 35 I. M. Arabatzis, T. Stergiopoulos, D. Andreeva, S. Kitova, S. G. Neophytides and P. Falaras, J. Catal., 2003, 220, 127. 36 Y. Liu, Y. Hu, M. J. Zhou, H. S. Qian and X. Hu, Appl. Catal., B., 2012, 21, 425. 37 G. Liu, P. Niu, L. C. Yin and H. M. Cheng, J. Am. Chem. Soc., 2012, 134, 9070.