Synthesis of Ag/Ag₂CO₃ heterostructures with high length–diameter ratios for excellent photoactivity and anti-photocorrosion

Abstract

Promising photocatalysts, Ag/Ag₂CO₃ one-dimensional heterostructures with high length–diameter ratios, are synthesized on a large scale by a simple one-pot method. The photocatalysis results prove that they have excellent photoactivity and stability under visible light. We propose a possible mechanism to explain the photocatalytic performance of the as-prepared heterostructures.

---

Introduction

Metal–semiconductor hybrid nanostructures have been attracting considerable research attention because of their unique properties which are extremely different from those of the single components.¹ They have been widely applied in diverse fields, including optics, electronics, catalytic reactions, etc.² Metal–semiconductor hybrid nanostructures in particular, exhibit interesting advantages as photocatalysts for organic pollutant degradation and water splitting because of plasmonic effects and rapid photoexcited electron–hole separation.^3–5 In the photocatalytic process, metal nanoparticles not only promote visible light absorption effectively, but also act as good electron acceptors to prevent photoexcited electrons recombining with holes.^6,7 So far, significant effort has been made toward a variety of hybrid photocatalysts. Among those, silver compounds loaded with noble metals (mainly Ag) were obtained by extensive synthesis routes since the breakthrough discovery of the Ag₃PO₄ semiconductor that exhibited high photocatalytic capabilities for water-splitting and organic dye decomposition under visible light.⁸ For example, Ye and co-workers successfully utilized Ag nanoparticles and nanowires to promote the Ag₃PO₄ photocatalytic performance.^9–11

Additionally, as a silver compound with a narrow band gap (2.3 eV), the Ag₂CO₃ photocatalyst has also been found to exhibit an excellent visible light response to degrade organic pollutants and bactericidal activity with high efficiency.^12–15 However, it is well-know that Ag₂CO₃ encounters the problem of severe photo-corrosion, which manifests as Ag⁺ being reduced to metallic Ag by photoexcited electrons in the photocatalytic process. At present, two approaches have been exploited for photocorrosion inhibition. One way is to introduce a sacrificial reagent to be an electron acceptor in the photocatalytic system, such as AgNO₃;¹² the other method is synthesis of various Ag₂CO₃ nanostructures to improve the photochemical activity and anti-photocorrosion ability of Ag₂CO₃, such as Ag₂CO₃ nanorods,¹⁶ Ag₂CO₃/AgX (X = Cl, Br, I) complexes,¹⁷ nanofibers containing Ag₂CO₃ nanoparticles,¹⁸ Ag₂CO₃/semiconductor composites^19–22 and graphene–Ag₂CO₃.²³ However, compared with other silver compounds such as Ag₃PO₄ and Ag₃VO₄, the investigation of Ag₂CO₃ photocatalysts is still insufficient. In particular, there are no reports about Ag nanoparticle loaded Ag₂CO₃ photocatalysts formed by a simple one-pot method. Although, Hu¹⁷ et al. utilized visible-light to irradiate Ag₂CO₃ nanorods in order to obtain Ag₂CO₃/Ag. A “two-step” method was used in their research, which is a frequently-used strategy for obtaining metal–semiconductor hybrid nanostructures. More importantly, Ag₂CO₃/Ag hybrid nanostructures with super high length–diameter ratios (nanowires) have not been successfully prepared previously. However, many physical properties, chemical behaviors and chemical reactivities can be improved significantly on the nanowire surface, which benefits their photocatalytic performance.²⁴

Herein, we successfully developed a facile one-pot strategy to fabricate Ag/Ag₂CO₃ hybrid nanostructures assisted by PVP and urea. The as-prepared Ag/Ag₂CO₃ heterostructured photocatalysts exhibit extremely high activity and stability. The morphology can be controlled via changing of the reaction conditions, which influences the photocatalytic properties of Ag₂CO₃ in the degradation of MO and RhB. The experimental results show that a higher length–diameter ratio in the hybrid nanostructure improves the photocatalytic performance; moreover, its photocurrent response time is very fast under visible light in photoelectrochemical reactions. We also discuss in detail the relationship between morphology and photoactivity. The study shows promise for the application of Ag/Ag₂CO₃ hybrid nanostructures in photodegradation and water splitting under visible light.

Experimental section

The typical synthesis of Ag/Ag₂CO₃ hybrid structures

Ag/Ag₂CO₃ hybrid structures were prepared using a simple one-pot hydrothermal synthesis method. Typically, 2 mmol of AgNO₃ and 0.4 g of PVP was dissolved into 20 mL of deionized water at room temperature with vigorous stirring. Then 2 mmol of urea was added gradually into the above mixed solution, and continuously stirred until a homogeneous solution was obtained. The resulting mixture was transferred into a Teflon-lined autoclave and heated at 120 °C for 8 h. After cooling naturally to room temperature, the products were separated by centrifugation and washed a few times with deionized water and ethanol alternatingly to remove ionic remnants. The final products were dried in a vacuum drying oven for 12 h.

Photocatalyst characterization

An XRD-6000 X-ray diffractometer with Cu Kα radiation (λ = 1.54060 Å) was used to obtain X-ray powder diffraction (XRD) spectra with a scanning rate of 0.1° s⁻¹. The morphology of the products was analyzed by scanning electron microscopy (SEM, HITACHI S-4800) with an attached energy-dispersive X-ray (EDX) system and transmission electron microscopy (TEM) images were also acquired (FEI TECNAI G20). X-ray photoelectron spectroscopy (XPS) was carried out on a Physical Electronics 5400 ESCA spectrometer with a base pressure of 109 Torr and Mg Kα X-ray radiation at a power of 200 W. UV-vis diffuse reflectance spectra were obtained on a UV-vis spectrophotometer (UV-24500, SHIMADZU).

Photocatalytic performance measurements

The organic dye photodecomposition measurements were taken on RhB and MO solutions. Firstly, RhB or MO powder was dissolved into distilled water with a concentration of 10 mg L⁻¹. 0.1 g of the as-prepared Ag/Ag₂CO₃ hybrid structure was dispersed into the above organic dye solutions. Then to complete the adsorption–desorption equilibrium between the dye and the photocatalyst, they were magnetically stirred for 30 min in the dark. Visible-light was obtained from a 300 W Xe arc lamp (PLS-SXE 300, Beijing), and UV light was optically filtered out. The above suspensions were irradiated under visible-light with magnetic stirring. Monitoring of the absorbance of the dye solutions in the degradation process was carried out on a UV-vis spectrophotometer (UV-4100, HITACHI) every 2 min after removal of the photocatalyst from the photodecomposition systems by centrifugal separation. Radical-trapping experiments were executed by the above process, but 3 mL of DMSO was added into the photocatalytic system.

Photoelectrochemical measurements

The photocurrents were measured with an electrochemical workstation (Zahner-Zennium Germany) in a standard three-electrode system with 0.5 M Na₂SO₄ electrolyte. A platinum wire was used as a counter electrode with Ag/AgCl (saturated KCl) as a reference electrode and ITO as a working electrode. 5 mg of sample powder was dispersed ultrasonically in a mixed solution of 850 μL of distilled water, 50 μL of Nafion solution and 100 μL of ethylene glycol. 20 μL of the dispersed colloid was spin-coated onto a piece of ITO with a fixed area of 1.5 × 1.5 cm², and then dried completely in air at room temperature. To obtain the photocurrent analysis, the ITO slice, as a working electrode, was irradiated and shielded at intervals of 60 s under a 500 W Xe arc lamp (San-ei Electric Japan).

Results and discussion

The following are the main reactions, which are interlaced with each other in the closed higher temperature and pressure system:

CO(NH₂)₂ + H₂O → NH₃ + CO₂

NH₃ + Ag⁺ → Ag(NH₃)₂⁺

Ag(NH₃)₂⁺ + CO₃²⁻ ⇋ Ag₂CO₃ + NH₃

The crystalline structure and phase purity of the as-prepared samples were examined by X-ray powder diffraction (XRD), as shown in Fig. 1a. The result is in agreement with the standard pattern of Ag₂CO₃ (JCPDS, card no. 70-2184). All of the diffraction peaks can be indexed to Ag₂CO₃ and metallic Ag without other peaks being found in the pattern, suggesting that the as-prepared products are high purity and crystallinity, with metallic Ag existing in them. The Ag/C molar ratio (2.72) was determined by energy dispersive X-ray spectrometry (Fig. 1b), which further indicates the existence of metallic Ag. To further prove the presence of Ag⁰ in the products, X-ray photoelectron spectroscopy (XPS) was introduced to reveal the valence state and elementary components. It was easily found that the products were composed of Ag, C and O elements, from the spectrum shown in (Fig. 1c). The higher resolution XPS data shows only one O 1s peak located at 531.1 eV, which is assigned to Ag₂CO₃, and not to Ag₂O, according to previous studies.^25–27 Fig. 1d shows the XPS spectrum of Ag 3d in the products. It clearly shows two individual peaks located at 368.2 eV and 374.2 eV, which were assigned to the Ag 3d_5/2 and Ag 3d_3/2 binding energies, respectively. Four peaks can be obtained by further dividing Ag 3d_5/2 and Ag 3d_3/2. The strong peaks at 368.0 eV and 374.0 eV are assigned to Ag⁺ in Ag₂CO₃ and the bands at 368.9 eV and 374.9 eV are attributed to Ag⁰ species.²⁸ The XPS results confirm that the as-prepared products are a Ag and Ag₂CO₃ hybrid structure without other impurities, which is consistent with the XRD and EDX results.

Fig. 1 Analysis of the crystalline structure and elemental composition of the as-prepared product. (a) XRD, (b) EDX and (c) and (d) XPS.

The morphology of the typical product was revealed by scanning electron microscopy (Fig. 2a and b). It was clearly found that many nanoparticles were homogeneously loaded onto the one-dimensional nanowires with a diameter of about 100 nm and a length of 5–10 μm. There are no impurity particles or aggregates in the products. The structure details were further investigated by TEM. The TEM image in the dark field (Fig. 2c) shows that the nanowires are composed of nanoparticles and that they are actually porous structures, which is somewhat different to the SEM results. However, we also clearly observe that many small nanoparticles with sizes in the range of 5–10 nm are present on the surface of the nanowires. Moreover, photocorrosion phenomena did not occur when Ag₂CO₃ was illuminated in the electron beam of the TEM, which shows that the hybrid nanostructures are more stable than Ag₂CO₃ particles deposited at room temperature (Fig. S1, ESI†). The HR-TEM image (inset in Fig. 2d) reveals that the lattices of Ag and Ag₂CO₃ are in close contact, indicating that strong interactions were established between Ag and Ag₂CO₃ during the synthesis process.

Fig. 2 Electron microscopy images of a typical product. (a and b) SEM images and (c and d) TEM images (the inset image is a HR-TEM image of Ag nanoparticles).

To investigate the growth process of hybrid nanowires, we executed many parallel experiments. Fig. S2† shows the morphologies of products synthesized with different amounts of PVP. We could not obtain nanowires without PVP in the reaction, as shown in Fig. S2a.† A great many nanorods appeared and scattered particles were not found in the products with an increasing amount of PVP. Fig. S2b† shows a SEM image of the nanorods synthesized with 0.2 g of PVP. With an increasing amount of PVP, the nanorods changed to a more uniform and longer form (Fig. S2c†). However, a large number of disharmonic nanoparticles appeared when the amount of PVP exceeded 0.4 g (Fig. S2d†). The influence of urea quantity on the morphology and composition of the products was also studied (Fig. S3†). Many non-uniform nanoparticles were observed when we did not add urea into the reaction (Fig. S3a†) and no wires were found in the products. The EDX pattern shows that the products formed are pure Ag nanoparticles (inset in Fig. S3a†). More and more nanowires appeared and the Ag nanoparticle content in the products increased with an increase in urea until the particles changed totally to wire-shaped structures. When the amount of urea was over 0.24 g, a great many nanorods with a non-uniform size appeared in the products.

The reaction time also influences the morphology of the as-synthesized catalysts. Fig. S4† shows the SEM images of the different products formed from 2 mmol of urea, 2 mmol of AgNO₃ and 0.4 g of PVP, but with different reaction times. It was easily found that 1-dimensional structures do not appear in the products from shorter reaction times, while a great many spindle-like nanostructures are obtained when the reaction time is increased to 12 h. We presume that Ag₂CO₃ goes through several recrystallization phenomena as the reaction time is prolonged. To study the relationship between catalyst morphology and photocatalytic performance, we chose the as-synthesized samples with different morphologies and bulk Ag₂CO₃ prepared by precipitating Ag⁺ and CO₃²⁻ at room temperature for comparison. Based on these results, we summarize the relationship between the reaction conditions and the morphology in Table 1.

Table 1 The typical morphology of the as-prepared products formed under different synthesis conditions

Reaction condition	Morphology	Length–diameter ratio	Characterization results
murea : mpvp = 0.3 : 1 heat for 8 h	Nanowires	50–100 : 1	Fig. 1 and 2
murea : mpvp = 0.6 : 1 heat for 8 h	Nanorods	10 : 1	Fig. S5a, c, 6a and b
murea : mpvp = 0.3 : 1 heat for 20 h	Nanospindles	10 : 1	Fig. S5b, d, 6c and d

---

Fig. S5† shows SEM and TEM images and XRD and EDX patterns of the as-prepared samples formed under different reaction conditions. It is easily found that they are rod and spindle shaped structures with lengths of ∼1 μm and diameters of ∼100 nm. XRD and EDX data (inset in Fig. S5†) display peaks all corresponding to an elemental composition of Ag and Ag₂CO₃, which indicates that both of the structures are Ag/Ag₂CO₃ hybrid products. It is clear that C 1s, O 1s, and Ag 3d and 3p all appear in the XPS results (Fig. S6†), which further proves that the products synthesized using the reaction conditions outlined in Table 1 are Ag/Ag₂CO₃ hybrid products.

The photoactivity of the as-prepared products was studied by degradation of organic pollutants under visible light irradiation (λ > 400 nm). Taking MO and RhB as model molecules, we did not introduce any sacrificial reagents into the photocatalytic experiments. Fig. 3 shows the UV-vis spectra of RhB and MO after exposure to visible light for different lengths of time in the presence of typical wire-shaped photocatalysts. These spectra show that the intensities of the absorption peaks at 550 nm and 460 nm decrease as the illumination time increases, in other words, the concentrations of the organic dyes in the aqueous solutions fall by 95% over 12 min and 16 min respectively. Fig. 3c and d directly present the photocatalytic performance of different catalysts for RhB and MO degradation respectively; C_o is the concentration of dye after completion of the adsorption/desorption equilibrium in the dark for 30 min. We easily find out the following results from the graphs: firstly, it is very difficult for P25 to degrade organic dyes under visible light because its band gap is too large (∼3.2 eV) to absorb light of lower energy. Secondly, the photoactivities of the as-prepared catalysts are superior to that of bulk Ag₂CO₃ precipitated by AgNO₃ and NaHCO₃ at room temperature, because they possess Ag nanoparticles and a high ratio of length to diameter. Thirdly, the samples with larger length–diameter aspect ratios exhibit the best performance among all the catalysts because of their higher surface area and larger amount of active interfaces. Finally, the rate of RhB degradation is faster than MO, which should be attributed to the different photocatalytic processes.

Fig. 3 UV-vis spectra of RhB (a) and MO (b) exposed to visible light for different lengths of time in the presence of nanowire photocatalysts; photocatalytic activities of P25, pure bulk Ag₂CO₃, and Ag/Ag₂CO₃ nanorods, nanospindles and nanowires for RhB (c) and MO (d) degradation under visible-light irradiation.

It is important to test the stability of photocatalysts in light. Fig. 4 shows the photocatalytic stability of the four Ag/Ag₂CO₃ structures prepared under different synthesis conditions. Notably, the rod, spindle and wire shaped nanostructures exhibit better photoactivity than bulk Ag₂CO₃, as they still decompose nearly 90% of MO and RhB; the nanowires in particular show degradation beyond 90% even after three successive cycles. Meanwhile the photoactivity of bulk Ag₂CO₃ reduces greatly because of serious photocorrosion after irradiation for 30 min. The results indicate that the hybrid one-dimensional structures improve the separation and migration efficiency of hole-carriers generated by visible light excitation. In comparison to Ag₂CO₃ nanorods and nanospindles, the nanowires possess a shorter transfer path for photo-induced electrons and more reaction active sites because of their smaller diameter and larger specific surface area, which effectively facilitates electron transfer out of Ag₂CO₃.

Fig. 4 Comparison of the photocatalytic stability of bulk Ag₂CO₃, and Ag/Ag₂CO₃ nanowires, nanospindles and nanorods in recycling reactions. (a) RhB and (b) MO.

Furthermore, photocurrent analysis was executed to clarify the improved photoactivity of wire-shaped Ag/Ag₂CO₃ hybrid nanostructures in a three-electrode PEC cell. Fig. S7a† displays the current density–potential curve (j–v) recorded in 0.5 M Na₂SO₄ aqueous solution. It is easily found that the dark current is very weak up to 1.0 V (vs. Ag/AgCl), but that the current in visible light increases obviously from 0.868 V (vs. Ag/AgCl), where electrocatalytic oxygen evolution could start. The photoelectric response and stability of the photocatalyst is also observed from the current–time (i–t) curve of on/off light cycles. The i–t investigation was carried out with a xenon lamp time interval of 60 s that turned on and turned off at a potential of 0.6 V (vs. Ag/AgCl). Fig. S7b (ESI†) shows that the anodic current increased rapidly in 0.39 s when the light was turned on, which indicates that photoinduced holes migrate to the surface for surface state filling (non-faradaic current) and to oxidize water (faradaic current). The photocurrent remains steady after a quick decline because of faradaic current decay. When the irradiation is over, a cathodic current spike appears rapidly in 0.55 s due to photoexcited electrons recombining with the holes stored in the surface states.²⁹ Moreover, the Ag/Ag₂CO₃ nanowires have a steady performance after 30 min illumination under visible light.

According to the above photocatalytic results, we propose a possible mechanism to explain the better photocatalytic performance of Ag/Ag₂CO₃ one-dimensional hybrid nanostructures than bulk Ag₂CO₃ under visible light, especially the wire-shaped structure with the biggest length–diameter ratio. This is schematically illustrated in Fig. 5. Firstly, Ag nanoparticles loaded in situ onto the one-dimensional Ag₂CO₃ semiconductor greatly improve the absorption in the visible light region. UV-visible diffuse reflectance spectra were used to investigate the optical absorbance of the different photocatalysts, which are shown in Fig. S8a.† The bulk Ag₂CO₃ has an absorption edge around 480 nm and its absorption range extends from the UV to the visible region. However, the as-prepared one-dimensional hybrid nanostructures have interesting strong absorption in the visible light region, especially the wire-shaped structures due to the surface plasmonic resonance (SPR) effect. Considering the fact that Ag₂CO₃ is an indirect bandgap semiconductor,²¹ the plots of the transformed Kubelka–Munk function of light energy (αhν)^1/2 versus energy (hν) of bulk Ag₂CO₃ and the as-synthesized hybrids are shown in Fig. S7b.† The bandgap width of bulk Ag₂CO₃ is estimated to be 2.4 eV, while those of the Ag/Ag₂CO₃ nanowires, nanospindles and nanorods are 1.6 eV, 1.8 eV and 2.1 eV, respectively.

Fig. 5 (a) Schematic illustrations of the photo-induced charge separation of the Ag/Ag₂CO₃ heterostructures. E_F and E_F′ refer to the Fermi levels of Ag₂CO₃ before and after reaching equilibrium. (b) The possible decomposition mechanism of dyes on the surface of Ag/Ag₂CO₃.

Secondly, a great many photoexcited electrons (e⁻) and photoinduced holes (h⁺) are produced in Ag₂CO₃ semiconductors by illumination with visible light. The close contact of the Ag nanoparticles and Ag₂CO₃ equilibrates their Fermi levels, which causes an energy level close to the conduction band of the Ag₂CO₃ semiconductor. Ag nanoparticles are good electron acceptors because their excellent conductivity greatly promotes interfacial charge-transfer kinetics among Ag and Ag₂CO₃, which further accelerates the effective separation of photoexcited electrons and photoinduced holes to decrease the recombination opportunity (Fig. 5a).³⁰ This is why the as-prepared Ag/Ag₂CO₃ one-dimensional hybrid nanostructures are more stable than bulk Ag₂CO₃. The electrons in the conduction band of Ag₂CO₃ can migrate quickly through the Schottky barrier into metallic Ag, which effectively facilitates their participation in the multiple electron reduction reaction.¹³

O₂ + e⁻ → ˙O₂⁻

˙O₂⁻ + 2H⁺ + 2e⁻ → H₂O₂

H₂O₂ + e⁻ → ˙OH + OH⁻

While the remaining holes on Ag₂CO₃ oxidize OH⁻ from the above reaction and water. The oxidation reactions are as follows:

OH⁻ + h⁺ → ˙OH

H₂O + h⁺ → ˙OH + H⁺

The highly reactive ˙O₂⁻ and ˙OH species from the electron reduction and hole oxidation are sufficient to destroy the structure of a dye molecule, which is an important reason for the decomposition of the pollutants.

Thirdly, Ag/Ag₂CO₃ one-dimensional hybrid nanostructures provide high surface areas and large numbers of active interfaces, especially the nanowires with their high length–diameter ratio. The larger specific surface area of the nanostructures can promote charge transfer across material interfaces, such as solid–solid and solid–liquid, which can provide a large number of active sites for the quick degradation of organic molecules.³¹

We can explain the difference in degradation rate between RhB and MO by combining the above photocatalytic mechanismwith Chen’s study;¹³ the oxyradicals mainly oxidizing RhB and MO are ˙O₂⁻ and ˙OH species respectively. It is presumed that the faster degradation rate of RhB compared to MO is because the ˙O₂⁻ concentration is higher than the ˙OH concentration, which indicates that a great many photoexcited electrons are transferred to Ag nanoparticles and can further reduce O₂ to ˙O₂⁻. We also performed radical trapping experiments, which are shown in Fig. S19.† The degradation rates of the dyes decrease obviously when DMSO is introduced into the photocatalytic system, indicating that ˙O₂⁻ and ˙OH play important roles in the photooxidation of RhB and MO. However, it is clearly found that the rate decrease of MO is faster than that of RhB, which shows that the radicals oxidizing MO and RhB are mainly ˙OH and ˙O₂⁻, respectively.

Conclusions

In summary, we successfully prepared one-dimensional Ag/Ag₂CO₃ hybrid nanostructures with high length–diameter ratios via a facile and low-cost one-pot method. They exhibit superior photocatalytic activities and stabilities in RhB and MO degradation processes under visible light compared to P25 and bulk Ag₂CO₃. This should be attributed to many Ag nanoparticles existing in the as-synthesized nanostructures and their special one-dimensional morphology, which improves the light absorption and quick separation of the photoexcited electrons and photoinduced holes.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (21271009, 21471006), the Doctoral Fund of the Ministry of Education of China (20123424110002), the Programs for Science and Technology Development of Anhui Province (1501021019), the Program for Postgraduate Innovative Research of Anhui Normal University (2014yks072), the Program for Innovative Research Team of Anhui Education Committee, the Program for Talents Team of University in Anhui and the Graduate Student Research Innovation Project of Anhui Normal University (2015cxsj131zd).

Notes and references

1. C. Gong, L. Colombo, R. M. Wallace and K. Cho, Nano Lett., 2014, 14, 1714–1720; R. B. Jiang, B. X. Li, C. H. Fang and J. F. Wang, Adv. Mater., 2014, 26, 5274–5309; Y. P. Du, B. Chen, Z. Y. Yin, Z. Q. Liu and H. Zhang, Small, 2014, 10, 4727–4734; D. W. Ding, K. Liu, S. N. He, C. B. Gao and Y. D. Yin, Nano Lett., 2014, 14, 6731–6736; Y. Z. Chen, D. Q. Zeng, M. B. Cortie, A. Dowd, H. Z. Guo, J. B. Wang and D. L. Peng, Small, 2015, 11, 1460–1469; L. Nahar, R. J. A. Esteves, S. Hafiz, U. Ozgür and I. U. Arachchige, ACS Nano, 2015, 9, 9810–9821.
2. U. Banin, Y. B. Shahar and K. Vinokurov, Chem. Mater., 2014, 26, 97–110.
3. S. K. Ghosh and T. Pal, Chem. Rev., 2007, 107, 4797–4862.
4. D. B. Ingram and S. Linic, J. Am. Chem. Soc., 2011, 133, 5202–5205.
5. H. Tada, T. Kiyonaga and S. Naya, Chem. Soc. Rev., 2009, 38, 1849–1858.
6. J. Chen, J. Wu, P. Wu and D. Tsai, J. Phys. Chem. C, 2011, 115, 210–216.
7. P. Wang, B. Huang, Z. Lou, X. Zhang, X. Qin, Y. Dai, Z. Zheng and X. Wang, Chem.–Eur. J., 2010, 16, 538–544.
8. Z. Yi, J. Ye, N. Kikugawa, T. Kako, S. Ouyang, H. Stuart-Williams, H. Yang, J. Cao, W. Luo, Z. Li, Y. Liu and R. Withers, Nat. Mater., 2010, 9, 559–564.
9. Y. P. Bi, H. Y. Hu, S. X. Ouyang, Z. B. Jiao, G. X. Lu and J. H. Ye, J. Mater. Chem., 2012, 22, 14847–14850.
10. Y. P. Bi, H. Y. Hu, S. X. Ouyang, Z. B. Jiao, G. X. Lu and J. H. Ye, Chem.–Eur. J., 2012, 18, 14272–14275.
11. H. Y. Hu, Z. B. Jiao, T. Wang, J. H. Ye, G. X. Lu and Y. P. Bi, J. Mater. Chem. A, 2013, 1, 10612–10616.
12. G. Dai, J. Yu and G. Liu, J. Phys. Chem. C, 2012, 116, 15519–15524.
13. H. J. Dong, G. Chen, J. X. Sun, C. M. Li, Y. G. Yu and D. H. Chen, Appl. Catal., B, 2013, 134, 46–54.
14. J. J. Buckley, P. L. Gai, A. F. Lee, L. Olivi and K. Wilson, Chem. Commun., 2008, 34, 4013–4015.
15. C. Xu, Y. Liu, B. Huang, H. Li, X. Qin, X. Zhang and Y. Dai, Appl. Surf. Sci., 2011, 257, 8732–8736.
16. J. J. Li, W. L. Yang, J. Q. Ning, Y. J. Zhong and Y. Hu, Nanoscale, 2014, 6, 5612–5615.
17. H. Xu, J. X. Zhu, Y. X. Song, T. T. Zhu, W. K. Zhao, Y. H. Song, Z. L. Da, C. B. Liu and H. M. Li, J. Alloys Compd., 2015, 622, 347–357; J. J. Li, Y. L. Xie, Y. J. Zhong and Y. Hu, J. Mater. Chem. A, 2015, 3, 5474–5481.
18. G. Panthi, S. J. Park, T. W. Kim, H. J. Chung, S. T. Hong, M. Park and H. Y. Kim, J. Mater. Sci., 2015, 50, 4477–4485; P. S. Saud, B. Pant, Z. K. Ghouri, G. Panthi, S. J. Park, W. D. Han, M. Park and H. Y. Kim, Fibers Polym., 2015, 16, 1336–1342.
19. Y. F. Li, L. Fang, R. X. Jin, Y. Yang, X. Fang, Y. Xing and S. Y. Song, Nanoscale, 2015, 7, 758–764; H. T. Ren, S. Y. Jia, S. H. Wu, T. H. Zhang and X. Han, Mater. Lett., 2015, 142, 15–18.
20. J. Wang, C. Dong, B. B. Jiang, K. L. Wu, J. Sun, X. Z. Li, W. J. Zhang, B. Zhang and X. W. Wei, Mater. Lett., 2014, 131, 108–111.
21. C. L. Yu, G. Li, S. Kumar, K. Yang and R. C. Jin, Adv. Mater., 2014, 26, 892–898.
22. C. L. Wu, Mater. Lett., 2014, 136, 62–264; C. L. Wu, Mater. Lett., 2015, 152, 76–78.
23. C. Dong, B. B. Jiang, K. L. Wu, Y. Hu, S. H. Xia and X. W. Wei, Mater. Lett., 2015, 146, 37–39; C. Dong, K. L. Wu, X. W. Wei, X. Z. Li, L. Liu, T. H. Ding, J. Wang and Y. Ye, CrystEngComm, 2014, 16, 730–736.
24. A. I. Hochbaum and P. D. Yang, Chem. Rev., 2010, 110, 527–546.
25. J. F. Weaver and G. B. Hoflund, Chem. Mater., 1994, 6, 1693–1699.
26. C. T. Campbell and M. T. Paffett, Surf. Sci., 1984, 143, 517–535.
27. R. W. Bigelow, Appl. Surf. Sci., 1988, 32, 122–140.
28. C. H. An, J. Z. Wang, W. Jiang, M. Y. Zhang, X. J. Ming, S. T. Wang and Q. H. Zhang, Nanoscale, 2012, 4, 5646–5650; Y. F Li, K. Li, Y. Yang, L. J. Li, Y. Xing, S. Y. Song, R. C. Jin and M. Li, Chem.–Eur. J., 2015, 21, 17739–17747.
29. D. W. Kim, S. C. Riha, E. J. DeMarco, A. B. F. Martinson, O. K. Farha and J. T. Hupp, ACS Nano, 2014, 8, 12199–12207.
30. H. Yamada, A. Bhattacharyya and J. Maier, Adv. Funct. Mater., 2006, 16, 525; P. Kamat, J. Phys. Chem. C, 2007, 111, 2834; T. Tatsuma, K. Takada and T. Miyazaki, Adv. Mater., 2007, 19, 1249; M. Antoniadoua and P. Lianos, Appl. Catal., B, 2010, 99, 307; Y. Bi and J. Ye, Chem. Commun., 2010, 46, 1532.
31. F. E. Osterloh, Chem. Soc. Rev., 2013, 42, 2294–2320.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental section, SEM, TEM, XRD, XPS, UV-vis diffuse reflectance spectra, j–v and i–t curves. See DOI: 10.1039/c6ra21325a

---