A novel tri-layered photoanode of hierarchical ZnO microspheres on 1D ZnO nanowire arrays for dye-sensitized solar cells

Abstract

A novel tri-layered ZnO nanostructure, which consists of ZnO nanowires (ZnO NWs) as the underlayer, small ZnO hierarchical microspheres (S-ZnO HMSs) as the intermediate layer and large ZnO hierarchical microspheres (L-ZnO HMSs) as the overlayer, was designed and studied as the photoanode for dye-sensitized solar cells (DSSCs). The photoanode was characterized by X-ray diffraction, scanning electron microscopy and its optical properties were measured by UV-vis spectrophotometry. The performance of the DSSC based on the photoanode was characterized by the incident-photon-to-current conversion efficiency (IPCE) and voltage–current curve (I–V). The results indicate that ZnO hierarchical microspheres show a good bifunctional character (light scattering effect and dye-loading ability), which causes the corresponding DSSCs based on the tri-layered ZnO photoanode to obtain the highest photovoltaic conversion efficiency (3.21%) compared with the DSSC based on the double layered ZnO photoanode and single layer ZnO NWs photoanode.

---

Introduction

The dye-sensitized solar cells (DSSCs), developed by Grätzel and coworkers, have attracted the considerable attention of many research groups in the past two decades owing to their high efficiencies and low cost.^1,2 In general, the DSSC is a photoelectrochemical system that contains a nanometer-sized semiconductor film, a sensitizer, a counter electrode and an electrolyte. As an important part of the DSSC, the nanostructured semiconductor film should have a high surface area, suitable band gap and high electron mobility. Among all the wide band gap semiconductors, TiO₂ completely accords with the afore-mentioned conditions and has already been widely used in DSSCs and achieved a very good effect.^3–6 However, it is difficult to further increase the photovoltaic conversion efficiency because of the recombination between electrons and either the oxidized dye molecules or electron-accepting species in the electrolyte during the charge transport process.^7,8

To solve this issue, ZnO, which has an energy-band structure and physical properties similar to TiO₂, was studied. What's more, it possesses a higher electronic mobility that would be favorable for electron transport^9,10 and the richest family of nanostructures compared with TiO₂.^11,12 Recently, various ZnO nanostructures have been prepared to improve photovoltaic properties of ZnO-based DSSCs, such as: nanoparticles,^13,14 nanowires (or nanorods),^15,16 nanosheets,^17,18 nanotubes¹⁹ and microspheres.^20,21 In the initial research stages, the DSSCs based on ZnO nanoparticles only obtain the fairly low photoelectric conversion efficiencies (<3%),^22,23 which due to the limitation of the size and shape of ZnO nanoparticles. Then Keis et al. reported a compression method for preparing ZnO film with a size of ∼150 nm ZnO nanoparticles, which played a key role in the generation of light scattering, obtained a high photovoltaic conversion efficiency (5%).²⁴ Besides, the one-dimensional ZnO structures (nanowires/nanorods/nanotubes) were also reported, which were a confirmation that the one-dimensional structure offer better electron transport compared to nanoparticles.^15,16,19 However, the insufficient surface area of one-dimensional structure limited the amount of dye adsorption thereby possibly impacting the photovoltaic conversion efficiency of the DSSCs. Recently, a novel hierarchically ZnO nanosheet was fabricated, which owned a high surface area and a good light scattering effect.^21,25 Li et al. reported porous nanosheet-assembled ZnO microspheres and achieved an excellent photovoltaic conversion efficiency (5.16%) owe to this novel nanostructure which could generates a prominent aggregation-induced light scattering center and further enhanced the light absorption and propagation.²⁰

It is well-known that good dye absorption, electron transfer, electron collection, electron recombination inhibition and pronounced light-scattering effect are indispensable to a high performance photoanode, but these factors are often incompatible with one another. Compared with one-dimensional and three dimensional structures, the nanoparticles could have better dye absorption, but they could have lower electron transport and light scattering. Based on these considerations, a novel bilayer structure ZnO-based DSSCs were studied.^26–29 Zheng et al. reported a significantly improvement of cell performance based on a double light-scattering layer ZnO structures composed of sub-micrometer-sized plate like ZnO overlayer and ZnO monodispersed aggregates underlayer. The research results indicate that the double ZnO structures obtained the better photovoltaic conversion efficiency (3.44%) compared to ZnO monodispersed aggregates film (0.81%).²⁹ These novel double nanostructures had basically met the above condition which could maintain the adsorption amount of dye, meanwhile owned a good light scattering effect. Recently, Wu reported a novel tri-layered TiO₂ photoanode, which contained one-dimensional (1D) TiO₂ nanotubes, three dimensional (3D) TiO₂ hierarchical microsized spheres, as well as zero-dimensional (0D) nanoparticles with a large surface area.³⁰ This novel tri-layered photoanode collects the advantages of efficient charge-collection, light-harvesting, as well as high dye-loading capability. However, this tri-layered nanostructure is rarely reported, especially ZnO.

In this article, a novel tri-layered ZnO nanostructure as photoanode for DSSCs was fabricated and studied. In the novel photoanode, the ZnO nanowires (ZnO NWs) were used as the underlayer because of its good electron transport rate, a small ZnO hierarchical microspheres (S-ZnO HMSs) were used as the intermediate layer because of its good dye-loading and light scattering and a large size ZnO hierarchical microspheres (L-ZnO HMSs) were used as overlayer due to its good light scattering. For comparison, a double layered photoanode which contains ZnO NWs as the underlayer and S-ZnO HMSs as the overlayer and the monolayered ZnO NWs photoanode were also fabricated and studied. As a result, the photovoltaic conversion efficiency of DSSC based on the tri-layered ZnO photoanode obtained the highest photovoltaic conversion efficiency (3.21%) compared with the DSSCs based on the double layered ZnO photoanode (1.67%) and monolayered ZnO NWs photoanode (0.43%). These results mainly due to the ZnO hierarchical microspheres have good light scattering effect and dye-loading ability.

Experiment

Materials and characterization

All solvents and other chemicals were reagent grade and used without further purification. Zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O), hexamethylenetetramine (C₆H₁₂N₄), potassium nitrate (KNO₃), zinc chloride (ZnCl₂), ammonium hydroxide (NH₃, 28%) and CO(NH₂)₂ were purchased from Aldrich. The electrolyte consisted of 0.6 M 1,2-dimethyl-3-propylimidazolium iodide (DMPII), 0.025 M LiI, 0.04 M I₂, 0.05 M guanidium thiocyanate (GuSCN), and 0.28 M 4-tertbutylpyridine (TBP) in dry acetonitrile and N719 dye were purchased from heptachroma. The image of scanning electron microscope (SEM) was measured with JEOL JSM-7600F. X-Ray Diffraction patterns (XRD, SHIMADZU XRD-7000, Cu Kα radiation) was used to determine the crystalline phase. Absorption spectra were measured with SHIMADZU (model UV2550) UV-vis spectrophotometer. Diffuse reflectance of the films have been collected by a UV-vis spectrophotometer (U-4100, Hitachi) by using an integrating sphere. The photocurrent–voltage characteristics of the DSSCs were performed on an electrochemical workstation (CHI 660C, Shanghai Chenhua) under AM 1.5 G simulated solar light (100 mW cm⁻²) (CHF-XM-500W, Trusttech Co. Ltd., Beijing, China) and the active area was 0.16 cm² for all of the cells. The incident-photon-to-current conversion efficiency (IPCE) spectra of the DSSCs were measured on a monochromator (Model 77890, Newport). The electrochemical impedance spectra (EIS) were carried out on an electrochemical workstation (CHI660C, Shanghai) at a bias potential of −0.6 V in a dark condition with the frequency range from 0.1 Hz to 100 kHz.

Preparation of ZnO NWs

ZnO NWs were prepared via a two-step synthetic approach.^15,16 First, the ZnO seed layer fabricated by constant-current electro-deposition. In detail, fluorine-doped SnO₂ (FTO)-coated glass plates were used as cathode after rinsed ultrasonically, successively in acetone, ethanol and deionized water. Pt sheet, as the anode electrode, was also treated similar to FTO and an Ag/AgCl electrode was used as reference electrode. The preparation of electrolyte was as follows: 0.05 M zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O), 0.01 M hexamethylenetetramine (C₆H₁₂N₄) and 0.1 M potassium nitrate (KNO₃) were dissolved in deionized water and then the mixed solution was stirring 10 min at 70 °C. A piece of FTO glass substrate and Pt sheet substrate were vertically immersed in the electrolyte and Ag/AgCl electrode placed near anode, then a current density of 0.4 mA cm⁻² was applied. The ZnO seed layer was succeeded deposition on FTO after 30 min. The ZnO NWs were synthesized by a hydrothermal method similar to that described by Gao et al.^16,31 The process as follows: 100 mL aqueous solution, which contains 0.27 g zinc chloride (ZnCl₂), 0.28 g hexamethylenetetramine, and 5 mL ammonium hydroxide (NH₃, 28%), was prepared first. Then the resulting solution was transferred to a Teflon-lined stainless steel autoclave and the seeded substrate was placed vertically to the bottom of Teflon-lined stainless steel autoclave with the backside against the beaker wall. The reaction temperature was maintained at 95 °C for 16 h. After deposition, the film was rinsed with deionized water and dried at 50 °C for 10 h.

Preparation of S-ZnO HMSs

The S-ZnO HMSs were prepared via a direct precipitation of zinc salt in an alkaline environment at room-temperature.^21,25 Typically, 0.20 M zinc nitrate hexahydrate and 1.0 M sodium hydroxide were dissolved by 100 mL deionized water separately. Then the zinc nitrate hexahydrate aqueous solution was added into the sodium hydroxide aqueous solution with stirring about 20 min. The white precipitate, which called as S-ZnO HMSs, was filtered and washed with deionized water for several times.

Preparation of L-ZnO HMSs

The L-ZnO HMSs were prepared via a hydrothermal method has been reported.²⁰ Typically, 7.14 g of Zn(NO₃)₂·6H₂O and 4.32 g of CO(NH₂)₂ were added to 60 mL deionized water for stirring about 10 min. Then the resulting solution was transferred to a Teflon-lined stainless steel autoclave maintained at 120 °C for 3 h. After completion of the reaction, the autoclave was naturally cooled at room temperature and then the white precipitate was obtained. The precipitate was filtered and washed with deionized water for several times and dried in air naturally. Finally, the resulting powder was calcined at 500 °C for 0.5 h in air with a heating rate of 5 °C min⁻¹.

Fabrication of the DSSCs

The ZnO paste was prepared by mixing 0.50 g ZnO powder (S-ZnO HMSs or L-ZnO HMSs) with 0.75 g absolute alcohol and 0.5 g deionized water, then followed by ultrasonic treatment until the particles were dispersed homogeneously. For fabricating the double layer ZnO-based DSSCs, the S-ZnO HMSs paste was coated onto the ZnO NWs-grown substrate via by a doctor-blade technique. With similar way, the tri-layered ZnO nanostructured photoanode was fabricated by a doctor-blade technique to cover with L-ZnO HMSs on the prepared double layered photoanode. The thickness of the tri-layered film is 25–30 μm. Then the photoanodes were annealed at 450 °C for 30 min after being dried at 125 °C for 30 min in air. While being cooled to room temperature, the photoanodes were immersed into 0.3 mM N719 ethanol solution for 12 h at room temperature. The counter electrode was a magnetron sputtered platinum mirror. The photoanode and counter electrode constituted a sandwich-like open cell. The employed liquid electrolyte was a solution of 0.3 M HMII, 0.5 M LiI, 0.05 M I₂ and 0.5 M TBP in MPN.

Fabrication of the samples for measuring the dye adsorbed amount and absorption spectra on photoanode films

The measurement of the dye adsorbed amount was according to the previous literature³² and the adopted photoanode films were same as the fabrication of DSSCs. The measurement procedure as follows: the three kinds of ZnO photoanode films (area: 6 × 6 mm) were sensitized for 12 h in N719 bath and further employed for the measurement of the dye adsorbed amount. The three kinds of ZnO photoanode films were sensitized for 12 h in a N719 bath, which were adopted for absorption spectra measurement of the dyes on ZnO surface.

Results and discussion

Morphology and crystal structure of different photoanodes

The S-ZnO HMSs and L-ZnO HMSs were placed on the surface of ZnO NWs successively by a doctor-blade technique to get the tri-layered ZnO photoanode, which shown as in Fig. 1a. Fig. 1b is the cross-section SEM images of the tri-layered ZnO photoanode, which can be observed clearly its three layer structure. For the underlayer, which clearly demonstrate the underlayer structure as we designed (Fig. 1c). It was the typical of ZnO nanowires with the average length and diameter around 3–4 μm and 200–400 nm, respectively. For the intermediate layer, it was made by S-ZnO HMSs with the diameters in the range of 3–5 μm, which was synthesized by a direct precipitation method. It also could observe that the microspheres were constructed with densely interlaced nanosheets, which indicates that the S-ZnO HMSs has a high surface area and surface roughness (Fig. 1d). Fig. 1e is the SEM image of L-ZnO HMSs, consisting of nanoporous ZnO nanosheets with the diameters in the range of 12–14 μm, which not only generate a prominent aggregation-induced light scattering center to enhance the light absorption and propagation, but also enhance the transport of injected electrons due to the single crystal nature of the nanosheet provide a direct electron pathway.

Fig. 1 (a) Schematic diagram of the tri-layered ZnO photoanode, (b) the cross-section SEM images of the tri-layered ZnO photoanode, the SEM images of the (c) ZnO nanowires, (d) small ZnO hierarchical microspheres, (e) large ZnO hierarchical microspheres.

Fig. 2 shows the XRD patterns of the monolayered, the double layered and the tri-layered ZnO photoanodes. It can be seen that besides the three peaks at (110), (200) and (211) from the crystal orientation of SnO₂ due to the FTO substrate, all other peaks can be assigned to wurtzite type ZnO (Joint Committee on Powder Diffraction Standards (JCPDS) card file 36-1451), which correspond to (100), (002), (101), (102), (110), (103), (112) and (004) planes of the ZnO. It could be observed that the (002) peak of ZnO NWs film in intensity was much higher than those of other peaks, which is due to the ZnO NW arrays crystalline structure and their c-axis orientation is perpendicular to the substrate.¹⁶

Fig. 2 XRD patterns of the monolayered, double layered and tri-layered ZnO photoanodes.

Optical properties and device performances

Fig. 3a shows the reflectivity spectra of the photoanodes, which were fabricated by using the monolayered, the double layered and the tri-layered films, respectively. It could be observed that the three kinds of photoanodes have similar reflectance in the 400–800 nm regions. The tri-layered photoanode and the double layered photoanode both show better light-scattering performance than that of the monolayered photoanode, which indicates that the former two films have better light-scattering ability due to the ZnO microspheres. More importantly, the tri-layered photoanode shows the best performance in light-scattering than those of the two other photoanodes. This should be attributed to the ZnO microspheres with different sizes in the tri-layered photoanode. In addition, the broad and deep channels between adjacent nanosheets can extend the pathway into the porous inner architecture of hierarchical flowers for photon localization enhancement.²¹

Fig. 3 Optical properties of the monolayered, the double layered and the tri-layered ZnO photoanodes, (a) UV-vis reflectance spectra, (b) UV-vis absorbance spectra.

To further investigate the absorption of these ZnO photoanodes, UV-vis absorption spectra of the monolayered, the double layered and the tri-layered ZnO photoanodes were performed. The absorption spectra of ZnO photoanodes without adsorbed N719 show in Fig. 3b. The intensities of light absorbance were found to increase in the spectral range of 400–800 nm with the order the tri-layer > double layer > monolayer. The possible reason for the increased light absorbance is the addition of scattering layers, which resulted in an increase in the number of light pathways in the photoanode. And the photoanode with more scattering layers show the better light absorption, so, the tri-layered film obtained the strongest light absorption eventually. The dye adsorption amount and light scattering effect are considered as the two main influence factors on the efficiencies of DSSCs. So, the UV-vis absorption spectra of the photoanodes, which have been adsorbed N719, also were performed. It can be seen the light absorption intensity of double layered photoanode was significantly higher than that of monolayered photoanode, that is mainly due to S-ZnO HMSs have larger specific surface area and the more N719 adsorption amount can be obtained.²¹ And the light absorption intensity of tri-layered photoanode was higher than that of double layered photoanode, that also indicates the L-ZnO HMSs can increase the N719 adsorption amount.

Fig. 4 shows the I–V curves of DSSCs fabricated with different ZnO photoanodes and the according photovoltaic parameters are summarized in Table 1. The DSSC based on the monolayered ZnO photoanode (cell 1) shows lowest photovoltaic conversion efficiency (0.43%) with an open circuit voltage (V_oc) of 400 mV, a short circuit current density (J_sc) of 2.11 mA cm⁻² and a fill factor (FF) of 51.5%. And when adding the second layer, the photovoltaic conversion efficiency of cell 2 obtains a larger increase (1.67%) with an open circuit voltage (V_oc) of 528 mV, a short circuit current density (J_sc) of 5.15 mA cm⁻² and a fill factor (FF) of 61.5%. Further adding the third layer, the DSSC based on the tri-layered ZnO photoanode (cell 3) exhibits the highest photovoltaic conversion efficiency (3.21%) with an open circuit voltage (V_oc) of 536 mV, a short circuit current density (J_sc) of 10.0 mA cm⁻² and a fill factor (FF) of 59.6%. Compared with the cell 1, the DSSC based on double layered ZnO photoanode (cell 2) shows a higher J_sc because of the more adsorbed N719 dye (1.12 × 10⁻⁷ mol cm⁻²) compared with ZnO NWs (2.01 × 10⁻⁸ mol cm⁻²) and the light scattering effect. Moreover, the V_oc is increasing to 528 mV, which mainly due to the improvement of the direct contact of the electrolyte to FTO substrate and imperfect dye coverage.³³ When the larger ZnO hierarchical microspheres were coated on the double-layer film by a doctor-blade technique, the DSSC based on the tri-layered ZnO photoanode (cell 3) exhibits the highest photovoltaic conversion efficiency due to the further increase of the N719 adsorption amount (1.64 × 10⁻⁷ mol cm⁻²) and the excellent light scattering properties of L-ZnO HMSs.

Fig. 4 (a) Photocurrent–voltage characteristics of the DSSCs based on different photoanodes, (b) IPCE spectra of the DSSCs based on different photoanodes.

Table 1 Photovoltaic parameters of the as-fabricated DSSCs

Samples	Voc (mV)	Jsc (mA cm^−2)	FF (%)	η (%)	Absorbed dye (mol cm^−2)	Rs (Ω)	R1 (Ω)	R2 (Ω)
Cell 1	400 ± 2	2.11 ± 0.1	51.5 ± 0.3	0.43 ± 0.03	2.01 × 10^−8	16.7	1.30	13.6
Cell 2	528 ± 1	5.15 ± 0.2	61.5 ± 0.1	1.67 ± 0.07	1.12 × 10^−7	16.4	3.06	38.6
Cell 3	536 ± 1	10.0 ± 0.1	59.6 ± 0.1	3.21 ± 0.02	1.64 × 10^−7	15.8	3.05	78.7

---

The incident-photon-to-current conversion efficiency (IPCE) spectra offer detailed information on the light harvest of the DSSCs (Fig. 4b). The cell 2 and cell 3 show higher IPCE than that of cell 1 over the entire wavelength region, this result is consistent with the result obtained by current, which was due to light scattering and dye-loading ability. Different photoanodes show different light scattering effect. Compared with cell 1, the cell 2 has a higher IPCE due to light scattering of the S-ZnO HMSs and more dye adsorption amount. Further, the tri-layered ZnO photoanode obtain the highest IPCE, most likely contribute to the best light scattering and the most dye adsorption amount.

Electrochemical impedance spectroscopy (EIS) analysis

Fig. 5 shows the electrochemical impedance spectra of the above different DSSCs under dark light. In general, three semicircles extending from total resistance can be recognized and fitted according to an equivalent circuit. The semicircle in the high frequency region (1 × 10⁵ to 1 × 10² Hz) represents the impedance corresponding to charge transfer at the counter electrode (R₁), while those in intermediate frequencies (1 × 10² to 1 × 10 Hz) give information on the impedance at the TiO₂/electrolyte interface related to the charge transport/recombination (R₂), and the low-frequency region (10–0.1 Hz) represents the diffusion resistance of I₃⁻/I⁻ in the electrolyte (R₃), respectively.^34–36 The higher R_s values seem to indicate that a lower FF, and the R₁ values for the DSSCs are all similar because of the same counter electrodes, the fitted R₂ of the cells in the order cell 3 (78.7 Ω) > cell 2 (38.6 Ω) > cell 1 (13.6 Ω), which reflecting the ease of the electron recombination process in the corresponding photoanodes. Two factors can be invoked to explain the phenomenon of the higher R₂ value of the photoanode: a larger surface area resulting in a smaller recombination resistance and the single crystal nature of the nanosheet provide a direct electron pathway giving rise to a smaller transport resistance.^37,38

Fig. 5 Nyquist plots from impedance spectra of DSSCs based on the monolayered, the double and the tri-layered photoanodes.

Conclusions

In summary, the novel tri-layered ZnO photoanode was successfully fabricated on a FTO glass substrate which contains ZnO nanowires (ZnO NWs) as the underlayer, small ZnO hierarchical microspheres (S-ZnO HMSs) as the intermediate layer, and large ZnO hierarchical microspheres (L-ZnO HMSs) as the overlayer. The results demonstrated that ZnO microspheres have the good light scattering effect and dye-loading ability, and the photoanode of the tri-layered film was better than the photoanode of double layer film for light scattering effect and dye-loading ability. As a result, the DSSC based on the tri-layered ZnO photoanode obtained the highest photovoltaic conversion efficiency (3.21%), compared with the DSSCs based on double layered ZnO photoanode (1.67%) and the ZnO NWs photoanode (0.43%).

Acknowledgements

We thank the National Natural Science Foundation of China (Grant no. 21272033, 21402023), National Science and Technology Major Project of the Ministry of Environmental Protection of China (2012ZX07203-003-Z04) and Key Laboratory of Functional Inorganic Material Chemistry (Heilongjiang University), Ministry of Education of China for financial support.

Notes and references

1. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
2. S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, Nat. Chem., 2014, 6, 242–247.
3. M. Grätzel, Nature, 2001, 414, 338–344.
4. S. Ito, S. M. Zakeeruddin, R. Humphry-Baker, P. Liska, R. Charvet, P. Comte, M. K. Nazeeruddin, P. Péchy, M. Takata, H. Miura, S. Uchida and M. Grätzel, Adv. Mater., 2006, 18, 1202–1205.
5. G. K. Mor, K. Shankar, M. Paulose, O. K. Varghese and C. A. Grimes, Nano Lett., 2006, 6, 215–218.
6. S. H. Kang, S.-H. Choi, M.-S. Kang, J.-Y. Kim, H.-S. Kim, T. Hyeon and Y.-E. Sung, Adv. Mater., 2008, 20, 54–58.
7. M. Grätzel, J. Photochem. Photobiol., A, 2004, 164, 3–14.
8. J. Nissfolk, K. Fredin, A. Hagfeldt and G. Boschloo, J. Phys. Chem. B, 2006, 110, 17715–17718.
9. D. C. Looka, D. C. Reynoldsa, J. R. Sizeloveb, R. L. Jonesb, C. W. Littonb, G. Cantwellc and W. C. Harschc, Solid State Commun., 1998, 105, 399–401.
10. L. Forro, O. Chauvet, D. Emin, L. Zuppiroli, H. Berger and F. Lévy, J. Appl. Phys., 1994, 75, 633–635.
11. Z. L. Wang, Mater. Today, 2004, 7, 26–33.
12. Q. Zhang, C. S. Dandeneau, X. Zhou and G. Cao, Adv. Mater., 2009, 21, 4087–4108.
13. N. Memarian, I. Concina, A. Braga, S. M. Rozati, A. Vomiero and G. Sberveglieri, Angew. Chem., Int. Ed., 2011, 50, 12321–12325.
14. G. Pérez-Hernández, A. Vega-Poot, I. Pérez-Juárez, J. M. Camacho, O. Arésa, V. Rejón, J. L. Peña and G. Oskam, Sol. Energy Mater. Sol. Cells, 2012, 100, 21–26.
15. M. Law, L. E. Greene, J. C. Johnson, R. Saykally and P. Yang, Nat. Mater., 2005, 4, 455–459.
16. Y. F. Gao, M. Nagai, T. C. Chang and J. J. Shyue, Cryst. Growth Des., 2007, 7, 2467–2471.
17. E. Hosono, S. Fujihara, I. Honna and H. S. Zhou, Adv. Mater., 2005, 17, 2091–2094.
18. J. Qiu, M. Guo and X. Wang, ACS Appl. Mater. Interfaces, 2011, 3, 2358–2367.
19. A. B. F. Martinson, J. W. Elam, J. T. Hupp and M. J. Pellin, Nano Lett., 2007, 7, 2183–2187.
20. Z. D. Li, Z. Y. Zhou, G. G. Xue, T. Yu, J. G. Liu and Z. G. Zou, J. Mater. Chem., 2012, 22, 14341–14345.
21. C. Zhu, Y. Shi, C. Cheng, L. Wang, K. K. Fung and N. Wang, J. Nanomater., 2012, 2012, 212653.
22. G. Redmond, D. Fitzmaurice and M. Grätzel, Chem. Mater., 1994, 6, 686–691.
23. S. Rani, P. Suri, P. Shishodia and R. Mehra, Sol. Energy Mater. Sol. Cells, 2008, 92, 1639–1645.
24. K. Keis, E. Magnusson, H. Lindstrom, S. E. Lindquist and A. Hagfeldt, Sol. Energy Mater. Sol. Cells, 2002, 73, 51–58.
25. Y. T. Shi, C. Zhu, L. Wang, W. Li, C. Cheng, K. M. Ho, K. K. Funga and N. Wang, J. Mater. Chem., 2012, 22, 13097–13103.
26. K. Mahmood, H. W. Kang, R. Munir and H. J. Sung, RSC Adv., 2013, 3, 25136–25144.
27. J. J. Tian, L. L. Lv, X. Y. Wang, C. B. Fei, X. G. Liu, Z. X. Zhao, Y. J. Wang and G. Z. Cao, J. Phys. Chem. C, 2014, 118(30), 16611–16617.
28. J. Xu, K. Fan, W. Shi, K. Li and T. Peng, Sol. Energy, 2014, 101, 150–159.
29. Y.-Z. Zheng, X. Tao, L.-X. Wang, H. Xu, Q. Hou, W.-L. Zhou and J.-F. Chen, Chem. Mater., 2010, 22, 928–934.
30. W.-Q. Wu, Y.-F. Xu, H.-S. Rao, C.-Y. Su and D.-B. Kuang, J. Phys. Chem. C, 2014, 118, 16426–16432.
31. Y. G. Wei, C. Xu, S. Xu, C. Li, W. Z. Wu and Z. L. Wang, Nano Lett., 2010, 10(6), 2092–2096.
32. H. Tian, X. Yang, R. Chen, R. Zhang, A. Hagfeldt and L. Sun, J. Phys. Chem. C, 2008, 112, 11023–11033.
33. F. D. Nayeri and F. Salehi, Renewable Energy, 2013, 60, 246–255.
34. Q. Wang, J. E. Moser and M. Grätzel, J. Phys. Chem. B, 2005, 109, 14945–14953.
35. R. Kern, R. Sastrawan, J. Ferber, R. Stangl and J. Luther, Electrochim. Acta, 2002, 47, 4213–4225.
36. L. Y. Han, N. Koide, Y. Chiba and T. Mitate, Appl. Phys. Lett., 2004, 84, 2433–2435.
37. K. Yan, Y. Qiu, W. Chen, M. Zhang and S. Yang, Energy Environ. Sci., 2011, 4, 2168–2176.
38. G. Zhu, L. K. Pan, J. Yang, X. J. Liu, H. C. Suna and Z. Sun, J. Mater. Chem., 2012, 22, 24326–24329.

---