Rapid Microwave Synthesis of Porous TiO2 ... - ACS Publications

ARTICLE pubs.acs.org/JPCC

Rapid Microwave Synthesis of Porous TiO2 Spheres and Their Applications in Dye-Sensitized Solar Cells Hong-En Wang,†,‡ Ling-Xia Zheng,† Chao-Ping Liu,†,‡ Yan-Kuan Liu,†,‡ Chun-Yan Luan,†,‡ Hua Cheng,† Yang Yang Li,† Ludvik Martinu,†,§ Juan Antonio Zapien,*,†,‡ and Igor Bello*,†,‡ †

Department of Physics and Materials Science and ‡Center of Super-Diamond and Advanced Films (COSDAF), City University of Hong Kong, Hong Kong SAR, PR China ABSTRACT: Porous anatase TiO2 spheres with sizes ranging from 150 to 250 nm were synthesized by a rapid microwave treatment of spherical titanium glycolate precursors preformed via an ethylene glycol-mediated solgel process. The effects of various experimental conditions on the formation of titanium glycolate precursors and final TiO2 spheres were investigated. A dye-sensitized solar cell (DSSC) assembled with the as-synthesized porous TiO2 spheres as photoanodes exhibits a 5% energy conversion efficiency, which is almost 40% higher than that made of the standard commercial Degussa P25 TiO2 nanopowders.

1. INTRODUCTION Dye-sensitized solar cells (DSSCs) have been intensively investigated during the past two decades as a promising alternative to conventional silicon-based solar cells due to their potentially low production cost and high energy conversion efficiency.1 A typical DSSC consists of a dye-adsorbed photoanode and a Pt counter electrode filled with an iodide/triiodide redox electrolyte. Much effort has been made toward improving the performance of DSSCs by means of increasing light harvest, improving charge transport, and reducing recombination, which can be implemented via optimizing photosensitizers,24 photoanodes,514 redox electrolytes,1519 and counter electrodes.2023 Among these, the structure and morphology of photoanodes play critical roles in determining the light harvest and charge transport properties, which significantly influence the final cell performance. Conventionally, the photoanodes of DSSCs are prepared by spreading a viscous paste containing anatase TiO2 nanocrystals (∼20 nm) on transparent conducting substrates with a doctorblade or screen-printing method.24,25 These films possess high internal surface area for dye adsorption, thereby giving rise to high photocurrent and energy conversion efficiency. However, the electron transport in such photoanode films is slower than those in bulk single-crystal counterparts due to the random walkup of electrons and the trapping/detrapping events along the electron’s path due to defects, surface states, grain boundaries, and self-trapping.26 Oriented polycrystalline TiO2 nanotubes and single-crystal TiO2 nanowire/nanorod arrays have been devised and synthesized to tackle this issue.2732 However, most of the cell performance based on these array nanostructures reported so far has been inferior to those of porous TiO2 r 2011 American Chemical Society

nanocrystal films possibly due to a lower specific surface area and/or other limiting factors.2832 In addition, it is still under discussion whether such array structures can significantly improve charge transport properties.33,34 Another key issue regarding the nanocrystal-based photoanode films is that it is transparent and has a negligible light scattering ability, which leads to the loss of a large portion of the visible light at longer wavelengths. This issue can hardly be improved by simply increasing the film thickness due to the limited electron diffusion length. Additional scattering layers (scattering centers) or incorporation of photonic crystals have been adopted to increase the light absorption.6,25,3539 However, multilayer structures not only complicate the photoanodes but also may decrease the available internal surface area for dye molecules adsorption. It is necessary to increase the light absorption by scattering effects while maintaining a high internal surface area of the photoanode films for efficient dye adsorption. Recently, increased attention has been paid to the utilization of nanoporous and/or mesoporous TiO2 submicrometer-/microspheres as photoanodes in DSSCs,4044 and high energy conversion efficiencies (7.2%∼8.4%) have been obtained. Several routes have been developed to synthesize porous or hollow TiO2 nanostructures, such as templates,45 structural directing reagents,36 polymers,46 reflux,47 and hydrothermal/ solvothermal4042,48 effects. The use of templates and structural directing reagents usually complicates the preparation process Received: February 3, 2011 Revised: April 13, 2011 Published: May 03, 2011 10419

dx.doi.org/10.1021/jp2011588 | J. Phys. Chem. C 2011, 115, 10419–10425

The Journal of Physical Chemistry C

ARTICLE

and increases the production cost. The conventional hydrothermal/solvothermal processes are often very time- and energyconsuming. Therefore, controlled synthesis of porous TiO2 nanostructures with well-defined morphology, particle size, and pore structure by a rapid, reproducible, and simple method remains a significant challenge. In this paper, we report on the fabrication of porous TiO2 spheres by a rapid and simple microwave synthesis approach and their application in DSSCs. A photoelectric conversion efficiency (η) of 5% was achieved, which represents an almost 40% enhancement compared to that obtained in DSSCs assembled with standard commercial P25 nanopowders (η = 3.6%).

2. EXPERIMENTAL SECTION 2.1. Synthesis. Commercially available titanium(IV) isopropoxide (TIP, 97%), ethylene glycol (EG, 99.8%), acetone (g99.9%), and ethanol (g99.5%) were used as received. The synthesis of porous TiO2 spheres includes the preparation of spherical titanium glycolate precursors and subsequent transformation into porous anatase TiO2 spheres by microwave heating. The titanium glycolate spheres were synthesized according to the work of Jiang et al.49 with some modifications. Unless otherwise specified, TIP (0.5 mL) dissolved in EG (20 mL) was quickly poured into an acetone bath (100 mL) containing trace of water (0.5 mL) under strong stirring for 15 min. The resulting white precipitates were statically aged for 1 h and then centrifuged, followed by several times washing with ethanol, and finally dried in air at 60 °C. For the microwave synthesis, the titanium glycolates were dispersed into 10 mL of water and an equal volume of ethanol and then loaded into a 35 mL reaction vessel for microwave heating (Discover S-Class, CEM Corporation). The set temperature, pressure limit, and heating power were 150 °C, 150 psi, and 100 W, respectively. After heating at 150 °C for 10 min, the samples were harvested by centrifugation, followed by several times rinsing in ethanol, and finally dried at 60 °C. 2.2. Characterization. The crystallographic structure of the samples was recorded on a Siemens D500 diffractometer operated at 40 kV/30 mA. Raman characterization was conducted with a Renishaw 2000 laser Raman microscope equipped with a 633 nm heliumneon laser; spectra were collected at room temperature with a 10 s exposure time. Surface morphology of the samples was observed with a Philips XL30 FEG fieldemission scanning electron microscope (FESEM), while transmission electron microscopy (TEM) images and selected-area electron diffraction (SAED) patterns were taken with a Philips CM20 transmission electron microscope. High-resolution transmission electron microscopy (HRTEM) analysis was performed using a Philips CM200 FEG TEM operated at 200 kV. The specific surface area and average pore size of the samples were analyzed by nitrogen (N2) adsorptiondesorption isotherms at 77 K using a NOVA 1200e Surface Area & Pore Size Analyzer (Quantachrome Instruments). Prior to adsorption experiments, the samples were outgassed at 150 °C for 2 h. Ultravioletvisible (UVvis) spectra were recorded on a PerkinElmer Lambda 750 UV/vis spectrophotometer. 2.3. Photovoltaic Performance Studies. To prepare the photoanodes of DSSCs, the synthesized porous TiO2 spheres, hydroxypropyl cellulose (Mw ∼ 80 000), and terpinol with a mass ratio of 2:1:8, respectively, were added into ethanol under stirring to form a slurry. The slurry was then evaporated at 70 °C along

Figure 1. (a) XRD patterns and (b) Raman spectra of titanium glycolate precursors and TiO2 spheres.

with stirring until a viscous paste was produced. The paste was spread onto fluorine-doped tin oxide (FTO)-covered glass (15 Ω square1, Nippon sheet glass company) using a doctorblade method with adhesive tape to control the thickness. The resulting TiO2 films were scraped into 0.2 cm2 circles and sintered at 450 °C for 30 min in air with a temperature ramping rate of 2 °C min1. After cooling to 90 °C, the films were soaked with an absolute ethanol solution containing 0.5 mM Ruthenizer 535 bis-TBA (N719 dye, Solaronix) and kept at room temperature for 24 h. The Pt counter electrodes were prepared by dropping an absolute ethanol solution containing 5 mM H2PtCl6 onto an FTO-covered glass and subsequent annealing at 400 °C for 20 min. The two electrodes were separated by a 60 μm thick Thermoplast hot-melt sealing foil (Meltonix 117060PF, Solaronix) to form a sandwich-type cell. The internal space of the cell was filled with a liquid redox electrolyte (0.1 M LiI, 50 mM I2, and 0.6 M 1,2-dimethyl-3-propylimidazolium iodide (DMPII) dissolved in acetonitrile) and sealed by capillary forces. Current density (J)voltage (V) curves were measured by using a Keithley 2400 Source Meter under the illumination of simulated sunlight (100 mW cm2) provided by an Oriel solar simulator with an air mass 1.5 global (AM 1.5G) filter.

3. RESULTS AND DISCUSSION 3.1. Microstructure of the Titanium Glycolate Precursors and TiO2 Samples. Figure 1a shows the XRD patterns of the

synthesized titanium glycolate precursors and TiO2, respectively. It reveals that the titanium glycolate precursors are amorphous with no characteristic diffraction peaks. After microwave heating, all the diffraction peaks can be indexed to pure anatase structures 10420

dx.doi.org/10.1021/jp2011588 |J. Phys. Chem. C 2011, 115, 10419–10425


ARTICLE

Figure 3. (a) TEM image and SAED pattern (inset image), (b) energydispersive X-ray (EDX) spectrum, and (c) HRTEM image of porous TiO2 spheres. Figure 2. FESEM images of (a) titanium glycolate precursors and (b) porous TiO2 spheres.

(JCPDS Card No. 21-1272). The broadened diffraction peaks indicate a nanocrystalline structure. The crystallite size of the samples was estimated to be ∼8.8 nm based on the full width at half-maximum (fwhm) of the (101) peak of the anatase phase according to the Scherrer equation. Figure 1b presents the corresponding Raman spectra of the titanium glycolate precursors and TiO2 samples. Similar to the XRD results as shown in Figure 1a, no Raman lines are observed for the titanium glycolates, while four Raman lines at 148, 400, 518, and 640 cm1 can be assigned as the Eg, B1g, A1g or B1g, and Eg modes of the anatase phase, respectively.50 These results reveal that the amorphous titanium glycolates have been fully transformed into anatase TiO2 after microwave treatment. Figure 2 presents the FESEM images of the titanium glycolate precursors and TiO2 products after microwave treatment, respectively. Clearly, the titanium glycolate precursors exhibit a well-defined spherical morphology with smooth surfaces (Figure 2a). The particle size typically ranges from 150 to 250 nm, and some spheres stick to each other. After microwave heating at 150 °C for 10 min, the resultant TiO2 product consists of a large quantity of porous flocky spheres (Figure 2b), which is quite different from the titanium glycolate precursors. Figure 3a shows a TEM image of the prepared TiO2 product that appears to consist of porous spheres formed by interconnected nanocrystals with a size of about 10 nm. The concentric circles composed of bright dots in the SAED pattern (inset image of Figure 3a) demonstrate the polycrystalline structure of the TiO2 spheres. Energy-dispersive X-ray spectra (EDX) indicate

that only Ti and O elements are contained in the synthesized product, apart from the C and Cu peaks, which arise from the carbon-coated copper grid used for the TEM characterization (Figure 3b). Figure 3c further presents the HRTEM image of the porous TiO2 spheres, which reveals that the TiO2 spheres are constructed of packed nanocrystals that are closely interconnected with each other, suggesting good charge transport properties, and presenting multiple pores (indicated by black arrows) formed by interparticle voids, suggesting high internal surface area for dye adsorption and potential enhancement of light harvesting efficiency. Lattice images are also observed, suggesting that the spheres are crystalline. A lattice spacing of ∼0.35 nm can be ascribed to the lattice spacing of the (101) plane of the anatase phase. The BrunauerEmmettTeller (BET) specific surface area and average pore size of the porous TiO2 spheres after annealing at 500 °C for 1 h have been found to yield ∼60.2 m2 g1 and 3.7 nm, respectively, as determined by N2 adsorptiondesorption isotherms. 3.2. Effect of Experimental Parameters on the Synthesis of Porous TiO2 Spheres. The synthesis of titanium glycolate spheres is based on the well-developed EG-mediated solgel route, which was first reported by Xia’s group49 and further modified by others.5153 It is well-known that most titanium alkoxides are reactive with water, and thus white precipitates can be produced quickly when they are in contact with moisture. However, their reactivity can be significantly reduced by an alkoxy chain-exchange reaction with EG as expressed in the following equation: TiðOCHðCH3 Þ2 Þ4 þ 2HOðCH2 Þ2 OH f TiððOCH2 Þ2 Þ2 þ 4ðCH3 Þ2 CHOH 10421

ð1Þ



ARTICLE

Figure 4. FESEM images of titanium glycolate precursors synthesized under different conditions: (a) without water; (b) 0.25 mL of water; (c) 1 mL of water; and (d) 1 mL of TIP.

The resultant titanium glycolates are quite stable and can be kept for a long period of time. Next, titanium glycolates can be precipitated by pouring them into an acetone bath containing traces of water. The exact formation mechanism of the titanium glycolate spheres is still unclear and under discussion.51 As described below, we found that the water and TIP contents play important roles in determining the morphology and particle size of the final titanium glycolate products. When no water is contained in the acetone bath, few large spherical particles (500800 nm) encapsulated into a porous matrix are obtained (Figure 4a). When 0.25 mL of water is contained in the acetone bath, larger spherical particles with sizes of 450900 nm are produced (Figure 4b). When the water content is further increased to 1 mL, polydispersed spheres with a broad size distribution (100400 nm) are obtained (Figure 4c). When the TIP content is increased to 1 mL, as a result, deformed spherical TiO2 particles with sizes of 75150 nm are formed (Figure 4d). In addition, when the EG sol containing TIP is added dropwise into the acetone bath, deformed TiO2 nanoparticles with sizes ca. 50 nm are produced, revealing that the injection procedure of TIP sol is also vital for the formation of spherical titanium glycolates. On the basis of the aforementioned experimental results, we deduce that the trace water is indispensable in the synthesis of spherical titanium glycolates, and it helps the precipitation of titanium glycolates and formation of dispersed particles. In addition, the concentrations of water and TIP can both alter the formation rate of titanium glycolate units, while stirring promotes their coarsening process by crash and friction. When the content of water or TIP increases, more titanium glycolate units are produced during the initial stage, and titanium glycolate particles with a small size and a broad size distribution are obtained. It should be pointed out that the size distribution of our titanium glycolate spheres is not monodisperse which might be ascribed to the much higher TIP concentration used in our experimental processes compared to the literature.49,51,54 After the formation of titanium glycolate precursors, porous anatase TiO2 spheres can further be synthesized by a subsequent microwave-initiated reaction. It is noted that the water used in the solvents is essential to the synthesis of porous anatase spheres. Few products composed of amorphous aggregates would be obtained if the reaction solvent was substituted by

Figure 5. FESEM images of TiO2 products microwave synthesized in different solvents: (a) absolute ethanol and (b) pure water.

absolute ethanol (Figure 5a). In contrast, anatase spheres with similar porous structures can also be obtained solely employing water as reaction solvent (Figure 5b). We summarize that the amorphous titanium glycolates have a loose structure and can promptly react with water to lose the alkoxy groups and form initial TiO2 nuclei during the microwave heating process. The reaction equations can be formulated as follows TiððOCH2 Þ2 Þ2 þ 4H2 O f TiðOHÞ4 þ 2HOðCH2 Þ2 OH ð2Þ TiðOHÞ4 f TiO2 V þ 2H2 O

ð3Þ

In general, the density difference between titanium glycolates and TiO2 can result in the occasional formation of cracks on the spheres’ surface, which can facilitate the penetration of water and release the alkoxy groups from inside the spheres. This reaction will proceed rapidly and continuously along with the growth of the TiO2 nuclei under the microwave heating; this results in porous anatase TiO2 spheres consisting of interconnected nanocrystals. It is noted that the formation of porous anatase spheres is completed within 10 min under current microwave reaction conditions. A further increase of heating time to 30 min or 1 h does not appear to alter the crystallinity and morphology of the final products. This indicates that the microwave reaction used here is more efficient than the conventional hydrothermal/ solvothermal process.55 10422



Figure 6. FESEM images of porous spheres after being (a) treated in an ultrasonic bath for 30 min and (b) ground in an agate mortar for 20 min.

3.3. Structural Stability of the As-Synthesized Porous TiO2 Spheres. The structural stability of porous nanostructures is

important for practical device applications. Here, the as-prepared porous TiO2 spheres were subjected to two kinds of different treatments to assess their stability. Figure 6 shows the FESEM images of porous spheres after being treated in a strong ultrasonic bath for 30 min or ground in an agate mortar for 20 min, respectively. From Figure 6a, one can see that the spherical morphology and pore structures of the TiO2 product are well preserved after ultrasonic treatment. The morphology of the TiO2 sample ground in an agate mortar slightly deviates from the spherical shape (Figure 6b); however, no obvious pulverization was observed. These results suggest that the as-synthesized TiO2 product is constructed of interconnected nanoparticles with strong interaction forces between adjacent particles, giving rise to a robust porous nanostructure. This conclusion is important from the point of view of practical applications of these porous TiO2 spheres, for example, a photoanode in dye- or quantum dot sensitized solar cells and electrodes in lithium-ion batteries. In both kinds of devices, ultrasonic or grinding treatments are commonly used for the preparation of TiO2 paste or slurry. 3.4. Photovoltaic Performances of As-Synthesized Porous TiO2 Spheres. The as-synthesized porous TiO2 spheres and commercial P25 nanopowders (∼21 nm particle size, Degussa) were employed as photoanodes to assemble the DSSCs. Figure 7a shows the cross-section FESEM image of the photoanode film made of porous TiO2 spheres with a thickness of about 10 μm, while Figure 7b shows the JV characteristics and output power density of the as-assembled cells measured under

ARTICLE

Figure 7. (a) Cross-section FESEM images of the photoanode film made of porous TiO2 spheres. (b) JV curves of the DSSCs assembled with the porous TiO2 spheres and with commercial P25 nanopowders as photoanodes.

Table 1. Photovoltaic Parameters of the DSSCs Assembled with Porous TiO2 Spheres and Commercial P25 Nanopowders samples

Jsc/mA cm2

Voc/V

efficiency (%)

fill factor

P25

10.4

0.6

3.6

0.57

porous spheres

15.6

0.6

5.0

0.53

100 mW cm2. The fill factor (FF) and power conversion efficiency (η) can be determined by the following equations FF ¼ Vm 3 Jm =Voc 3 Jsc

ð4Þ

η ¼ Voc 3 Jsc 3 FF=Pin

ð5Þ

where Vm and Jm are the voltage and current density at maximum power output; Voc and Jsc are the open-circuit voltage and shortcircuit current density; and Pin is the incident light power density. The corresponding photovoltaic parameters of the two cells are summarized in Table 1. We can see that the cell made of commercial P25 nanopowders possesses a η of about 3.6%, which is similar to the result reported by Yu et al.56 In contrast, the cell made of porous TiO2 spheres shows a higher value of η ∼ 5%, which represents about a 40% enhancement compared to the former cell. Considering similar values of Voc and FF in these two cells, the improvement of the cell performance with porous TiO2 spheres arises from a larger Jsc value or, in other words, due to its improved light harvest efficiency. First, our BET measurement indicates that the porous TiO2 spheres exhibit a higher specific 10423



Figure 8. Diffused reflectance spectra of the photoanode films made of P25 nanopowders and porous TiO2 spheres.

surface area (60.2 m2 g1) than that of the commercial P25 nanopowders (∼45 m2 g1).56 Therefore, the photoanode made of porous TiO2 spheres possesses higher dye adsorption ability than the one made of P25 nanopowders at a similar particle packing density and film thickness. Second, the photoanode film made of porous spheres with a larger particle size (150250 nm) provides better light scattering properties.40,42 This is supported by the results shown in Figure 8 which feature diffused reflectance spectra of the photoanode films. Clearly, both films made of P25 nanopowders and porous TiO2 spheres show high diffused reflection abilities in the wavelength range between 400 and 450 nm. However, a fast drop in the diffused reflection ability was observed for the P25 nanopowder film as the wavelength increased from 500 to 800 nm. This is mainly attributed to their small particle size (∼21 nm), which is much smaller than the wavelength of visible light. In contrast, the porous TiO2 sphere film showed higher diffused reflection ability in the longer wavelength region (500800 nm), suggesting that the incident light was more efficiently scattered by the film made of porous TiO2 spheres due to their larger particle size that is comparable to the wavelength of visible light.41 The improved light scattering effect can thus increase the light traveling path and enhance the interaction between TiO2 and dye molecules, resulting in higher light harvest efficiencies and corresponding higher photocurrent. In addition, the charge transport through the photoanode films made of porous TiO2 spheres might be improved due to better interparticle connectivity caused by the close attachment of primary particles, as previously discussed in connection with the HRTEM in Figure 3c, which contributes to better charge collection and higher Jsc.42 However, it is noted that the efficiencies of our devices are still inferior to the results reported by some other groups.40,41 Especially, the assembled cells have fairly low Voc and FF values. On one hand, the low Voc may be caused by the lack of tertbutylpyridine (TBP) or similar compounds in the composition of electrolytes, which have been used to increase the cells’ Voc by shifting the TiO2 conduction band edge toward higher energies and increasing the electron lifetime in the TiO2 films.57 On the other hand, the TiO2 spheres possess a low crystallinity and small pore size. As a result, a poor electronic conductivity of TiO2 and inferior electrolyte diffusion into/outside the TiO2 spheres could be expected, which can increase the cells’ internal transport resistance and thus lower the Voc and FF. Last, but not least, the relatively low packing density of the TiO2 spheres and poor

ARTICLE

interconnectivity between adjacent spheres in the prepared photoanode films (see Figure 7a) might also account for the increase of charge transport resistance and decrease of FF. To further improve the photovoltaic properties of our devices, several aspects can be addressed. First, TBP or similar compounds may be added into the redox electrolytes to increase the cells’ Voc. Second, the crystallinity of the TiO2 can be enhanced, and the pore size of spheres might be enlarged by finely tuning the experimental parameters (e.g., reaction temperature, reaction pressure, and additional postannealing) to improve the electronic conductivity, dye anchoring/adsorption on the TiO2 surfaces, and regeneration of oxidized dye molecules by redox electrolytes. Third, the connectivity between adjacent TiO2 spheres, packing density of TiO2 spheres in the photoanode film, and the film thickness should be optimized to improve the charge transport and electrolyte diffusion within the photoanode film together with the dye adsorption. In addition, a bifunctional interface layer can be incorporated between the FTO substrates and TiO2 films to both suppress the back electron transfer from transparent conducting substrates into the redox electrolytes and reduce the interfacial resistance between the conducting substrate and TiO2.58 Finally, surface modification of the photoanode films with an ultrathin oxide layer (e.g., TiO259 or MgO60) may be further performed to passivate the surface states of TiO2 and increase the back electron transfer resistance at the TiO2/electrolyte interfaces. However, special care should also be taken that such a layer does not block the TiO2 layer, which can influence the dye adsorption and regeneration with redox electrolyte.

4. CONCLUSION Porous TiO2 nanostructured spheres were successfully synthesized by a facile and rapid microwave process. The particles exhibit a well-defined spherical morphology with abundant pores. The particle size and size distribution can be finely tuned by altering the experimental parameters, in particular (TIP and water contents). A dye-sensitized solar cell assembled with the porous TiO2 spheres as a photoanode exhibits a photoelectric conversion efficiency of 5%, which represents a ∼40% enhancement compared to the P25 nanopowders. On the basis of BET surface area and light scattering measurements, the improved photovoltaic performance can be explained by the increased light harvest efficiency due to the particles’ high specific surface area and large particle size. The present approach can be further extended to the synthesis of other porous metal oxide materials and systems. In addition, the as-prepared porous TiO2 spheres can also have possible applications in several other fields, such as lithium-ion batteries, photocatalysis, and catalyst support. ’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected] (J.A.Z.); [email protected] (I. B.). Notes §

On leave from the Department of Engineering Physics, Ecole Polytechnique de Montreal, Montreal, Quebec, Canada.

’ ACKNOWLEDGMENT This work was fully supported by GRF of Hong Kong under the project number CityU 110 209. The authors also thank Dr. Hao Tang for HRTEM characterization. 10424



’ REFERENCES (1) O’Regan, B.; Gratzel, M. Nature 1991, 353, 737. (2) Hardin, B. E.; Hoke, E. T.; Armstrong, P. B.; Yum, J.-H.; Comte, P.; Torres, T.; Frechet, J. M. J.; Nazeeruddin, M. K.; Gratzel, M.; McGehee, M. D. Nat. Photonics 2009, 3, 406. (3) Lee, K.; Park, S. W.; Ko, M. J.; Kim, K.; Park, N.-G. Nat. Mater. 2009, 8, 665. (4) Gratzel, M. Acc. Chem. Res. 2009, 42, 1788. (5) Cheung, K. Y.; Yip, C. T.; Djurisic, A. B.; Leung, Y. H.; Chan, W. K. Adv. Funct. Mater. 2007, 17, 555. (6) Wang, Z.-S.; Kawauchi, H.; Kashima, T.; Arakawa, H. Coord. Chem. Rev. 2004, 248, 1381. (7) Liu, B.; Boercker, J. E.; Aydil, E. S. Nanotechnology 2008, 19, 505604. (8) Oh, J.-K.; Lee, J.-K.; Kim, H.-S.; Han, S.-B.; Park, K.-W. Chem. Mater. 2010, 22, 1114. (9) Baek, I. C.; Vithal, M.; Chang, J. A.; Yum, J.-H.; Nazeeruddin, Md.K.; Gratzel, M.; Chung, Y.-C.; Seok, S. I. Electrochem. Commun. 2009, 11, 909. (10) Marco, L. D.; Manca, M.; Ginnuzzi, R.; Malara, F.; Melcarne, G.; Ciccarella, G.; Zama, I.; Cingolani, R.; Gigli, G. J. Phys. Chem. C 2010, 114, 4228. (11) Grinis, L.; Kotlyar, S.; Ruhle, S.; Grinblat, J.; Zaban, A. Adv. Funct. Mater. 2010, 20, 282. (12) Lee, S.; Cho, I.-S.; Lee, J. H.; Kim, D. H.; Kim, D. W.; Kim, J. Y.; Shin, H.; Lee, J.-K.; Jung, H. S.; Park, N.-G.; Kim, K.; Ko, M. J.; Hong, K. S. Chem. Mater. 2010, 22, 1958. (13) Yang, W.; Wan, F.; Wang, Y.; Jiang, C. Appl. Phys. Lett. 2009, 95, 133121. (14) Kang, S. H.; Choi, S.-H.; Kang, M.-S.; Kim, J.-Y.; Kim, H.-S.; Hyeon, T.; Sung, Y.-E. Adv. Mater. 2008, 20, 54. (15) Bach, U.; Lupo, D.; Comte, P.; Moser, J. E.; Weissortel, F.; Salbeck, J.; Spreitzer, H.; Gratzel, M. Nature 1998, 395, 583. (16) Bai, Y.; Cao, Y.; Zhang, J.; Wang, M.; Li, R.; Wang, P.; Zakeeruddin, S. M.; Gratzel, M. Nat. Mater. 2008, 7, 626. (17) Snaith, H. J.; Schmidt-Mende, L. Adv. Mater. 2007, 19, 3187. (18) Wang, P.; Zakeeruddin, S. M.; Moser, J. E.; Nazeeruddin, M. K.; Sekiguchi, T.; Gratzel, M. Nat. Mater. 2003, 2, 402. (19) Tang, Y.; Pan, X.; Zhang, C.; Dai, S.; Kong, F.; Hu, L.; Sui, Y. J. Phys. Chem. C 2010, 114, 4160. (20) Fang, X.; Ma, T.; Guan, G.; Akiyama, M.; Kida, T.; Abe, E. J. Electroanal. Chem. 2004, 570, 257. (21) Murakami, T. N.; Ito, S.; Wang, Q.; Nazeeruddin, M. K.; Bessho, T.; Cesar, I.; Liska, P.; Humphry-Baker, R.; Comte, P.; Pechy, P.; Gratzel, M. J. Electrochem. Soc. 2006, 153, A2255. (22) Sun, K.; Fan, B.; Ouyang, J. J. Phys. Chem. C 2010, 114, 4237. (23) Wang, M.; Anghel, A. M.; Marsan, B.; Ha, N.-L. C.; Pootrakulchote, N.; Zakeeruddin, S. M.; Gratzel, M. J. Am. Chem. Soc. 2009, 131, 15976. (24) Barbe, C. J.; Arendse, F.; Comte, P.; Jirousek, M.; Lenzmann, F.; Shklover, V.; Gratzel, M. J. Am. Ceram. Soc. 1997, 80, 3157. (25) Ito, S.; Murakami, T. N.; Comte, P.; Liska, P.; Gratzel, C.; Nazeeruddin, M. K.; Gratzel, M. Thin Solid Films 2008, 516, 4613. (26) Peter, L. M. Phys. Chem. Chem. Phys. 2007, 9, 2630. (27) Mor, G. K.; Varghese, O. K.; Paulose, M.; Shankar, K.; Grimes, C. A. Sol. Energy Mater. Sol. Cells 2006, 90, 2011. (28) Xu, C.; Shin, P. H.; Cao, L.; Wu, J.; Gao, D. Chem. Mater. 2010, 22, 143. (29) Feng, X.; Shankar, K.; Varghese, O. K.; Paulose, M.; Latempa, T. J.; Grimes, C. A. Nano Lett. 2008, 8, 3781. (30) Liu, B.; Aydil, E. S. J. Am. Chem. Soc. 2009, 131, 3985. (31) Kumar, A.; Madaria, A. R.; Zhou, C. J. Phys. Chem. C 2010, 114, 7787. (32) Wang, H. E.; Chen, Z.; Leung, Y. H.; Luan, C.; Liu, C.; Tang, Y.; Yan, C.; Zhang, W.; Zapien, J. A.; Bello, I.; Lee, S. T. Appl. Phys. Lett. 2010, 96, 263104. (33) Zhu, K.; Neale, N. R.; Miedaner, A.; Frank, A. J. Nano Lett. 2007, 7, 69.

ARTICLE

(34) Enache-Pommer, E.; Liu, B.; Aydil, E. S. Phys. Chem. Chem. Phys. 2009, 11, 9648. (35) Guldin, S.; Huttner, S.; Kolle, M.; Welland, M. E.; MullerBuschbaum, P.; Friend, R. H.; Steiner, U.; Tetreault, N. Nano Lett. 2010, 10, 2303. (36) Yang, L.; Lin, Y.; Jia, J.; Li, X.; Xiao, X.; Zhou, X. Micropor. Mesopor. Mat. 2008, 112, 45. (37) Nishimura, S.; Abrams, N.; Lewis, B. A.; Halaoui, L. I.; Mallouk, T. E.; Benkstein, K. D.; Lagemaat, J. V. D.; Frank, A. J. J. Am. Chem. Soc. 2003, 125, 6306. (38) Qiu, Y.; Chen, W.; Yang, S. Angew. Chem., Int. Ed. 2010, 49, 3675. (39) Huang, F.; Chen, D.; Zhang, X. L; Caruso, R. A.; Cheng, Y. B. Adv. Funct. Mater. 2010, 20, 1301. (40) Kim, Y. J.; Lee, M. H.; Kim, H. J.; Lim, G.; Choi, Y. S.; Park, N. G.; Kim, K.; Lee, W. I. Adv. Mater. 2009, 21, 3668. (41) Chen, D.; Huang, F.; Cheng, Y. B.; Caruso, R. A. Adv. Mater. 2009, 21, 2206. (42) Yang, W. G.; Wan, F. R.; Chen, Q. W.; Li, J. J.; Xu, D. S. J. Mater. Chem. 2010, 20, 2870. (43) Sauvage, F.; Chen, D.; Comte, P.; Huang, F.; Heiniger, L. P.; Cheng, Y. B.; Caruso, R. A.; Graetzel, M. ACS Nano 2010, 4, 4420. (44) Shao, W.; Gu, F.; Li, C.; Lu, M. Inorg. Chem. 2010, 49, 5453. (45) Kondo, Y.; Yoshikawa, H.; Awaga, K.; Murayama, M.; Mori, T.; Sunada, K.; Bandow, S.; Iijima, S. Langmuir 2008, 24, 547. (46) Eiden-Assmann, S.; Widoniak, J.; Maret, G. Chem. Mater. 2004, 16, 6. (47) Zhong, L. S.; Hu, J. S.; Wan, L. J.; Song, W. G. Chem. Commun. 2008, 1184. (48) Chen, D.; Cao, L.; Huang, F.; Imperia, P.; Cheng, Y. B.; Caruso, R. A. J. Am. Chem. Soc. 2010, 132, 4438. (49) Jiang, X.; Herricks, T.; Xia, Y. Adv. Mater. 2003, 15, 1205. (50) Zhang, W. F.; He, Y. L.; Zhang, M. S.; Yin, Z.; Chen, Q. J. Phys. D: Appl. Phys. 2000, 33, 912. (51) Pal, M.; Serrano, J. G.; Santiago, P.; Pal, U. J. Phys. Chem. C 2007, 111, 96. (52) Wang, H.; Li, B.; Yan, Z.; Lu, Z.; Cheng, R.; Qian, D. Rare Metals 2008, 27, 1. (53) Yu, H. K.; Yi, G. R.; Kang, J. H.; Cho, Y. S.; Manoharan, V. N.; Pine, D. J.; Yang, S. M. Chem. Mater. 2008, 20, 2704. (54) Wang, P.; Xie, T. F.; Li, H. Y.; Peng, L.; Zhang, Y.; Wu, T. S.; Pang, S.; Zhao, Y. F.; Wang, D. J. Chem.—Eur. J. 2009, 15, 4366. (55) Periyat, P.; Leyland, N.; McCormack, D. E.; Colreavy, J.; Corr, D.; Pillai, S. C. J. Mater. Chem. 2010, 20, 3650. (56) Yu, J.; Fan, J.; Lv, K. Nanoscale 2010, 2, 2144. (57) Boschloo, G.; Haggman, L.; Hagfeldt, A. J. Phys. Chem. B 2006, 110, 13144. (58) Wu, M. S.; Tsai, C. H.; Wei, T. C. Chem. Commun. 2011, 47, 2871. (59) Grinis, L.; Kotlyar, S.; Ruhle, S.; Grinblat, J.; Zaban, A. Adv. Funct. Mater. 2010, 20, 282. (60) Peng, T.; Fan, K.; Zhao, D.; Chen, J. J. Phys. Chem. C 2010, 114, 22346.

’ NOTE ADDED AFTER ASAP PUBLICATION This article was published ASAP on May 3, 2011. Several text changes have been made in the last paragraph of section 3.4. The corrected version was published on May 6, 2011.

10425
