Introduction

In current research on photoelectrochemical (PEC) applications, such as photovoltaics and photocatalysis, semiconducting nanomaterials, such as TiO₂ and ZnO, have been the subject of great interest due to their size-tunable physicochemical properties, high activities, and lower prices.^1-3 Previous studies have proved that such semiconductors display excellent activities and stabilities. However, the fast recombination rate of the photogenerated electron/hole pairs and the limited photoresponding range hinder the commercialization of these materials.⁴ In order to enhance the photogenerated charge separation and the photoresponding range of such semiconductors, the following strategies can be adopted: phase and morphological control, doping, surface sensitization, composite materials.⁵

Recently, considerable interest has focused on nanocomposite films and powders, such as TiO₂-SiO₂,⁶ TiO₂-WO₃,^7,8 TiO₂-CdSe,⁹ and TiO₂-ZnO,¹⁰ which have been considered as effective semiconductors. Generally, inorganic semiconductor nanocomposites can be fabricated by two routes:¹¹ physical mixing (often with surfactant-assistance) two or more phases of nanoparticles,^12,13 or chemical synthesis via various synthetic routes.¹⁴ Compared to the physical-mixing nanocomposites, the chemical-fabricated nanocomposites are normally prepared into core/shell structures with spherical, beltlike, or rodlike morphologies,^15-19 and they allow a better mixing or closer contact among the different phases. It is generally accepted that properties of composites obtained often cannot be considered as a simple superposition of the properties of individual components due to strong surface interactions between the closely packed nanoparticles in the binary oxide systems.²⁰ Apart from alternation of electronic states by creating interfacial regions (such as p-n junctions),^19,21 the organization pattern and shape of each component in an integrated nanocomposite further determine its ultimate physicochemical properties and thus performance.^11,22

In this context, many attempts have been made to synthesize coupled bicomponent ZnO-TiO₂ nanocomposites by a physical or chemical process with two aims:^20,23-26 (i) extending the light adsorption spectrum and improving the using efficiency of the light, and (ii) suppressing the recombination of photogenerated electron/hole pairs. These nanocomposites may increase the PEC conversion efficiency by increasing the charge separation and extending the photoresponding range. In addition, one-dimensional (1D) ZnO nanostructures are expected to enhance photo-electron efficiency as well as gas-sensing and photonic performance due to their enhanced surface-to-volume ratio and quantum confinement effect,^27-29 and they have been used as better candidates for PEC applications. There has been considerable effort made toward the investigation of ZnO-TiO₂ nanocomposites; however, to the best of our knowledge, little attention has been paid to the coupling of ZnO 1D nanostructures with TiO₂ nanoparticles. Taking into account the fact that spherical nanocomposites are very easy to aggregate, which will reduce their activity, we believe that the fabrication of coupled nanocomposites consisting of ZnO 1D nanorods and TiO₂ nanoparticles would be very advantageous for PEC applications.

In this study, coupled bicomponent nanocomposites consisting of TiO₂ nanoparticles and ZnO 1D nanorods were prepared by a one-step hydrothermal method. The crystal structures and surface properties of coupled ZnO-TiO₂ nanocomposites were characterized by various techniques including transmission electron microscopy (TEM), X-ray diffraction (XRD), and Brunauer-Emmett-Teller (BET) techniques. The photocurrent actions and photocatalytic activities of these nanocomposites were subsequently investigated, and the relationship between their nanoscale structures and PEC properties was also addressed. This is of notable significance for understanding the unique properties that result from the coupled nanocomposites and designing new nanocomposites of advanced functions in PEC applications.

Experimental Section

Preparation of ZnO-TiO₂ Nanocomposites. The coupled bicomponent ZnO-TiO₂ nanocomposites were prepared using a simple hydrothermal method. Typically, ZnCl₂ and TiCl₄ in the molar ratios of 1:2, 1:1, and 2:1 were dissolved in a mixed solvent of 10 mL of ethanol and 10 mL of water under stirring for the preparation of the coupled ZnO-TiO₂ nanocomposites with the Zn/Ti molar ratios of 1:2, 1:1 and 2:1, labeled as ZT12, ZT11, and ZT21, respectively. Then 10 mL of 0.6 M urea aqueous solution was added dropwise to the stirred solution. After stirring for several minutes, the obtained transparent solution was transferred into a Teflon-lined stainless-steel autoclave. The autoclave was maintained at 180 C for 16 h and then cooled to room temperature naturally. The resulting white precipitate was recovered by centrifugation and washed with deionized water several times, and the precipitate finally was calcinated at 450 C for 2 h in the air. For comparison, the TiO₂ nanoparticles and ZnO nanorods were prepared using the same procedure as mentioned above except that the starting materials were TiCl₄ for TiO₂ and ZnCl₂ for ZnO, respectively.

Material Characterizations. Power X-ray diffraction (XRD) was performed on a Bruker D8-Advance X-ray powder diffractometer with monochromatized Cu K radiation ( = 1.5418 Å). The 2 range used in the measurements was from 20 to 70. Transmission electron microscopy (TEM) images and selected-area electron diffraction (SAED) were taken with a Hitachi model H-800 transmission electron microscope, using an accelerating voltage of 200 kV. Scanning electron microscopy (SEM) images were acquired on a field emission (FE) scanning electron microscope (JEOL JSM-6700F) operated at 1.0 kV. Specific surface areas were measured by Brunauer-Emmett-Teller (BET) nitrogen adsorption-desorption (Shimadzu, Micromeritics ASAP 2010 Instrument). Mott-Schottky (MS) spectra were measured with a three-electrode cell, using the sintered photoanode as the working electrode, a platinum wire as the counter electrode, and a standard Ag/AgCl in saturated KCl as the reference electrode. The electrolyte was 0.1 M KCl aqueous solution. Impedance spectra were obtained with a PARSTAT-2273 Advanced Electrochemical System (Princeton Applied Research) equipped with an impedance analyzer with ZSimpWin software and controlled by a computer.

Photocurrent Measurements. The synthesized ZnO-TiO₂ nanocomposite paste for the fabrication of a photoanode was obtained by mixing 2 mL of ethanol and 300 mg of ZnO-TiO₂ nanocomposite powder homogeneously. The obtained paste was spread on the FTO conducting glass (15 /square F-doping SnO₂) with a glass rod, using adhesive tapes as spacers. After the films were dried under ambient conditions, they were sintered in air at 450 C for 2 h. The film thickness measured with a profilometer was about 3 m. Photocurrent action spectra were measured in a two-electrode configuration home-built experimental system, where the sintered photoanode served as the working electrode with the active area of about 1 cm² by using Teflon tape and a platinum wire was used as the counter electrode.³⁰ A 500-W Xe lamp with a monochromator was used as the light source. The photoelectrochemical cell was illuminated from the FTO side of the photoanode electrode by incident light. The generated photocurrent signal was collected by using a lock-in amplifier (Stanford instrument SR830 DSP) synchronized with a light chopper (Stanford instrument SR540). The monochromatic illuminating light intensity was about 15 W/cm² estimated with a radiometer (Photoelectronic Instrument Co. IPAS). The illumination area of the photoanode was about 0.12 cm². All measurements were done after bubbling N₂ for 20 min and controlled automatically by a computer.

Photocatalytic Measurements. Aqueous suspensions of methylene blue (MB) (1 × 10^-5 M) and the sintered nanomaterial electrodes were placed in a 3-mL quartz-glass vessel. The photoreaction vessel was exposed to UV-vis irradiation under ambient conditions with an average intensity of 35 mW cm^-2 produced by a 100-W high-pressure mercury lamp, which was positioned 12 cm away from the vessel. The irradiance intensity was measured by a radiometer (FZ-A, Photoelectric Instrument Factory of Beijing Normal University, China). At given time intervals, the photoreacted solution was analyzed with a UV-vis spectrophotometer (Cary-500, VARIAN, USA) by recording variations of the absorption band maximum (660 nm) in the UV-vis spectrum of MB.

Results and Discussion

Preparation of ZnO-TiO₂ Nanocomposites. A series of coupled ZnO-TiO₂ nanocomposites were successfully synthesized using a one-step synthetic method under hydrothermal conditions based on the starting precursor materials of ZnCl₂ and TiCl₄ with different reaction molar ratios. In view of the same reaction conditions and experimental processes, the effects of these different reactions on the final products were supposed to be completely dependent on the molar ratios of the starting precursor materials (ZnCl₂ and TiCl₄). The structures of the obtained products were examined by TEM. Figure 1 shows the typical TEM images of these nanostructures. When only TiCl₄ or ZnCl₂ was used as the starting precursor material, we obtained pure TiO₂ nanoparticles with diameters of about 10-20 nm and pure ZnO nanorods with diameters of about 50 nm and lengths of up to a few hundred nanometers, respectively (Figure 1, parts A and B). When TiCl₄ and ZnCl₂ were used as the starting precursor material and other conditions were kept unchanged, coupled bicomponent nanocomposites consisting of ZnO 1D nanorods and TiO₂ nanoparticles were formed. Figure 1, parts C-E, shows the typical TEM images of coupled ZnO-TiO₂ nanocomposites obtained at various Zn/Ti molar ratios. It was observed that TiO₂ nanoparticles were well formed on the ZnO nanorods support. For the Zn/Ti molar ratio of 1:2, the ZnO nanorods were densely surrounded by the TiO₂ nanoparticles. When the Zn/Ti molar ratio was extended to 1:1 and 2:1, the distributed TiO₂ nanoparticles over the ZnO nanorods apparently decreased. Also, during the formation of the ZnO-TiO₂ nanocomposites, the original structure and shape of the TiO₂ nanoparticles and ZnO nanorods were well preserved, which is confirmed in the related TEM images. The representative SAED patterns shown in the insets of Figure 1 indicate excellent crystallinity for the prepared individual TiO₂ nanoparticles (anatase phase) and ZnO nanorods (wurtzite phase), and they also show a superimposition of two different sets of diffraction patterns for the prepared ZnO-TiO₂ nanocomposites, which demonstrates the crystallographic relationships among the TiO₂ nanoparticles and ZnO nanorods template. For the nanocomposites, the first set of diffraction spots belongs to the zone of anatase TiO₂, while the second set of diffraction spots is reflected in the detection of wurtzite ZnO. Furthermore, the superimposition of wurtzite ZnO and anatase TiO₂ was completely dependent on the Zn/Ti molar ratio. In addition, FE-SEM images provided further information involving the structures and surface morphologies of the as-prepared nanomaterials. As shown in Figure 2, nanoparticles with diameters of about 10-20 nm for TiO₂ (Figure 2A) and nanorods with irregular shapes for ZnO (Figure 2B) can be observed, respectively. As for the as-prepared ZnO-TiO₂ nanocomposite (ZT21), it can be clearly seen that TiO₂ nanoparticles were well distributed over ZnO nanorods (Figure 2C), which also indicated that the original structure and shape of the TiO₂ nanoparticles and ZnO nanorods were well preserved during the preparation of the ZnO-TiO₂ nanocomposites.

	Figure 1 Representative transmission electron microscopy (TEM) images of TiO2 nanoparticles (A), ZnO nanorods (B), ZT12 nanocomposites (C), ZT11 nanocomposites (D), and ZT21 nanocomposites (E). Insets show the selected-area electron diffraction (SAED) pattern.
	Figure 2 FE-SEM images of TiO2 nanoparticles (A), ZnO nanorods (B), and ZT21 nanocomposites (C).

Figure 3 shows the XRD patterns for TiO₂ nanoparticles, ZnO nanorods, and ZnO-TiO₂ nanocomposites, which provide further insight into the crystallinity of the products. For the preprared TiO₂ nanoparticles, all diffraction peaks can be well indexed as phase-pure anatase TiO₂. In the case of the prepared ZnO nanorods, nine characteristic peaks (2 = 31.77, 34.42, 36.25, 47.54, 56.60, 62.86, 66.38, 67.96, and 69.101), marked by their Miller indices ((100), (002), (101), (102), (110), (103), (200), (112), and (201)) were observed. This reveals that the resultant ZnO nanorods are in the wurtzite phase (JCPDS no. 36-1451). Compared to the monocomponent nanomaterials (TiO₂ nanoparticles or ZnO nanorods), the coupled nanocomposites indeed consist of both anatase TiO₂ and wurtzite ZnO, which is clearly shown in the XRD pattern. Furthermore, with the increase of the Zn/Ti molar ratio (from the value of 1:2 to 2:1), the characteristic peaks of anatase TiO₂ gradually decreased, and the characteristic peaks of wurtzite ZnO gradually increased in contrast. These XRD results further confirmed the successful preparation of the coupled bicomponent nanocomposites consisting of anatase TiO₂ and wurtzite ZnO.

Figure 3 XRD patterns of TiO2, ZnO, ZT12, ZT11, and ZT21 nanoparticles. Solid squares: anatase TiO2; solid circles: ZnO.

Photocurrent Actions. In order to examine the structure-specific PEC properties of the prepared nanomaterials, measurements of the photocurrent action spectra were performed in a home-built PEC experimental system. The results of photocurrent actions for all the samples are shown in Figure 4, in which the spectra were calibrated with the intensity of the monochromatic incident light.

Figure 4 Photocurrent action spectra of different nanoparticles thin film electrodes: TiO2, ZnO, ZT12, ZT11, ZT21, and P25.

As shown, the TiO₂ nanoparticles electrode showed a photocurrent spectrum with the maximum wavelength at 345 nm corresponding to the band gap of nanocrystalline TiO₂, which was blue-shifted from the band gap of bulk TiO₂ (387 nm, 3.2 eV) due to the quantum confinement effect.³¹ Interestingly, the photoelectrode with ZnO nanorods showed a broader photocurrent spectrum with a similar intensity covering the range of 315-400 nm. Also, its maximum signal appeared at around 375 nm, slightly red-shifted from the band gap of bulk ZnO (368 nm, 3.37 eV),³² most likely due to the effect of one-dimensional morphology. In the case of the prepared coupled ZnO-TiO₂ nanocomposites, the photocurrent was largely enhanced with several orders of magnitude higher photocurrent intensities than that of the monocomponent nanomaterials (TiO₂ nanoparticles or ZnO nanorods). Furthermore, because of the synergistic effect, the photocurrent spectrum of the coupled ZnO-TiO₂ nanocomposite was slightly blue-shifted from that of the ZnO nanorods, and the maximum wavelength remained at around 370 nm. In addition, the generated photocurrent of the coupled ZnO-TiO₂ nanocomposite varied with the Zn/Ti molar ratio. It was found that the resultant coupled ZnO-TiO₂ nanocomposite with the Zn/Ti molar ratio of 2:1 showed the highest photocurrent signal, which was even higher than that of the commercial TiO₂ P25 (Degussa P25, which consists of about 30% rutile and 70% anatase with a particle size of about 20 nm).

The above results indicate that the coupling of TiO₂ nanoparticles and ZnO nanorods resulted in considerable improvement in photocurrent generation. Compared to the monocomponent nanomaterials (TiO₂ nanoparticles or ZnO nanorods), the larger photocurrent must stem from some combination of more efficient electron injection into the film of coupled ZnO-TiO₂ nanocomposites and more efficient electron collection by FTO. The origin of this effect can be multifold. On the one hand, TiO₂ nanoparticles deposited on the surface of ZnO nanorods could effectively passivate surface recombination sites and act as a radial energy barrier that can repel electrons from the surface of ZnO nanorods. On the other hand, electron transport within single-crystalline ZnO nanorods could be much faster, and surface fields within each nanorod could be used to enforce charge separation and thereby ensure that faster transport results in a longer diffusion length.³³ Meanwhile, the preserved excellent crystallinity of TiO₂ nanoparticles and ZnO nanorods could also enable more efficient electron injection and transport within the coupled ZnO-TiO₂ nanocomposites. In addition, with the introduction of TiO₂ nanoparticles into the surface of ZnO nanorods, the increase in surface area was appreciable. The surface area of the monocomponent nanoparticles (TiO₂ nanoparticles or ZnO nanorods) was less than 20 m²/g, while that of the coupled ZnO-TiO₂ nanocomposites was 60-90 m²/g, which was larger than that of TiO₂ P25 nanoparticles (47.9 m²/g). Thus, the enhancement of photocurrent activities was also in part caused by the increase in surface area, which could enhance the light harvest and charge transfer within the coupled ZnO-TiO₂ nanocomposites photoelectrode.

Furthermore, to further characterize the electron-transfer properties of the synthesized coupled ZnO-TiO₂ nanocomposites, Mott-Schottky measurements were performed by using the impedance technique.^30,34,35 Figure 5 shows the MS plots of the electrodes based on the prepared TiO₂ nanoparticles, ZnO nanorods, and coupled ZnO-TiO₂ nanocomposites. Reversed sigmoidal plots were observed with an overall shape consistent with that typical for n-type semiconductors, and the reproducible flat-band potentials, that is, the potentials corresponding to the situation in which there is no charge accumulation in the semiconductor so that the energy bands show no bending, could be obtained from the x-intercepts of the linear region. Compared to the TiO₂ nanoparticles or ZnO nanorods-based electrode, the coupled ZnO-TiO₂ nanocomposites-based electrode showed a large negative shift of the conduction band. In addition, the slope of the linear region for the coupled ZnO-TiO₂ nanocomposites sample showed a relatively lower value, clearly indicating a higher donor density for the nanocomposites electrode. It is well-known that the presence of a large number of surface states can lead to a considerable change of the band position.^30,36 Also, a multitude of surface states within the coupled ZnO-TiO₂ nanocomposite might have caused the observed broadening and enhancement of photocurrent.

Figure 5 Mott-Schottky (MS) plots of TiO2 nanoparticles (hollow squares), ZnO nanoparticles (hollow circles), and ZT21 nanoparticles (solid triangles) thin film electrodes. Mott-Schottky measurements were done at the frequency of 1 kHz in the aqueous solution of 0.1 M KCl aqueous solution.

Photocatalytic Activities. The photocatalytic activities of as-prepared TiO₂ nanoparticles, ZnO nanorods, and coupled ZnO-TiO₂ nanocomposites were measured with photocatalytic degradation of active methylene blue (MB) as a model reaction, and the experimental results are shown in Figure 6. The temporal evolution of the spectral changes accompanying the photodegradation of MB over as-prepared coupled ZnO-TiO₂ nanocomposite (Sample ZT21) is shown in Figure 6A. The MB dye initially showed a major absorption band at 660 nm, while a gradual decrease in absorption with a slight shift of the band to shorter wavelengths was observed with light irradiation through an aqueous MB solution containing the ZT21 electrode. In this case, the dye solution became nearly transparent after 180 min of irradiation, consistent with facile destruction of the chromophoric structure of the MB.

Figure 6 (A) UV-visible absorption spectral changes of methylene blue (MB) aqueous solution over the ZT21 nanoparticles thin film electrode as a function of irradiation time (curves from top to bottom represent different irradiation times: 0, 30, 60, 90, 120, 150, and 180 min, respectively). (B) Absorption changes ( = 660 nm) plot for the photocatalytic degradation of MB with TiO2 (), ZnO thin film electrode (), ZT12 (), ZT11 (), ZT21 (right-facing solid triangle), and P25 thin film electrode (). (C/C0 is the normalized concentration of the solution).

During the degradation process, the dye was photodegraded in a stepwise manner with the color of the solution changing from an initial deep blue to nearly transparent. Figure 6B shows the temporal concentration changes of MB over TiO₂, ZnO, ZT12, ZT11, ZT21, as well as P25 nanostructures-based electrodes. The normalized concentration of the solution (C/C₀) is proportional to the normalized maximum absorbance (A/A₀), and therefore we used C/C₀ instead of A/A₀. As shown in Figure 6B, the photodegradation of MB catalyzed by the nanostructures follows a first-order rate law, -ln(C/C₀) = Kt, where K is the apparent rate constant of the degradation. In our experiment, K was found to be 0.516%, 0.555%, 0.746%, 0.9%, 1.371%, and 1.027 min^-1 for the TiO₂, ZnO, ZT12, ZT11, ZT21, and P25 nanostructures, respectively. It is clear that the reaction rates of the coupled ZnO-TiO₂ nanocomposite samples were much higher than that of as-prepared TiO₂ nanoparticles or ZnO nanorods. Especially, the photocatalytic activity of the coupled ZnO-TiO₂ nanocomposite with the Zn/Ti molar ratio of 2:1 was even higher than that of commercial TiO₂ P25. These results are in accordance with the photocurrent results.

A factor contributing to the enhanced photocatalytic activity of coupled ZnO-TiO₂ nanocomposites is its higher specific surface (60-90 m²/g), compared to less than 20 m²/g of the TiO₂ nanoparticles or ZnO nanorods. The higher specific surface area would be responsible for providing the higher adsorption ability of the catalytic surface toward target molecules and the higher ability of generating photoinduced electron-hole pairs of active sites.^37-39 The heightened photocatalytic activity of coupled ZnO-TiO₂ nanocomposites could be further ascribed to the enhanced charge separation derived from the coupling of TiO₂ with ZnO, which would promote interfacial charge-transfer relative to that of charge-carrier recombination.^40,41 In addition, the surface areas of ZT12 (87 m²/g) and ZT11 (84 m²/g) sample were larger than that of the ZT21 sample (64 m²/g), but their photoactivities were lower than that of the ZT21 sample. The fact that surface area and photoactivity followed opposite trends should alert one that the surface area was not the only factor responsible for determining the adsorption affinity and photoactivity of the coupled ZnO-TiO₂ nanocomposite. It is well-known that adsorption on the surface of nanostructures is ultimately affected by structural and electronic properties of surface states including coordinatively unsaturated surface cations, bridging oxygens, hydroxyl groups, and surface adsorbed species.^42,43 These considerations suggest that the surface properties and morphologies of ZT21 were more favorable for the adsorption of MB than those of ZT12 and ZT11, most probably due to the different Zn/Ti molar ratios employed to produce specific nanocomposite architectures.

Conclusions

In summary, novel coupled bicomponent nanocomposites consisting of wurtzite ZnO 1D nanorods and anatase TiO₂ nanoparticles were successfully prepared using a simple one-step hydrothermal method. The characterization data for the coupled ZnO-TiO₂ nanocomposites indicate that the majority of the TiO₂ nanoparticles were located around the surface of ZnO nanorods instead of doping into the ZnO lattice. Compared with the mere TiO₂ nanoparticles or ZnO nanorods, the coupling of TiO₂ nanoparticles and ZnO nanorods produced a significant effect on its properties, such as surface morphologies, surface areas, electronic properties, and PEC properties. It was demonstrated that the PEC performance of coupled ZnO-TiO₂ nanocomposites (including their photocurrent action and photocatalytic abilities) was significantly enhanced, and it was maximized for the coupled ZnO-TiO₂ nanocomposite with the Zn/Ti molar ratio of 2:1. This enhancement was, on the one hand, due to an increase of the surface area, which can enhance the light harvest and the ability of generating photoinduced electron-hole pairs of active sites and, on the other hand, due to the favorable electron-transfer properties of the heterojunctions TiO₂/ZnO in the coupled ZnO-TiO₂ nanocomposites. Thus, the present work provided a simple alternative method to effectively improve the PEC activities of nanostructures through the coupling of 0D nanostructures and 1D nanostructures. The importance of this study also lies in the fact that it reveals the possibility of expanding the related work from the coupling of TiO₂ nanoparticles and ZnO nanorods to other oxide semiconductors for developing new functional nanomaterials with promising applications in solar energy conversion, photocatalysis, and sensors.

Acknowledgment

This work was financially supported by the National Natural Science Foundation of China (No. 20435010, No. 20628303), 863 Project (2006AA05Z123), and National Basic Research Program of China (No. 2007CB310500).