Size-Controlled Synthesis and Optical Properties of Small-Sized ZnO

The large ratio of the UV emission to visible emissions in PL spectra can be obtained in small-sized ZnO nanorods. Furthermore, the morphologic and op...
0 downloads 8 Views 7MB Size
Download PDF
J. Phys. Chem. C 2009, 113, 7497–7502

7497

Size-Controlled Synthesis and Optical Properties of Small-Sized ZnO Nanorods Y.W. Chen,† Q. Qiao,† Y.C. Liu,*,† and G.L. Yang‡ Center for AdVanced Optoelectronic Functional Materials Research, Key Laboratory for UV Light-Emitting Materials and Technology of Ministry of Education, Northeast Normal UniVersity, Changchun 130024, People’s Republic of China, and Department of Physics, Drexel UniVersity, Philadelphia, PennsylVania 19104 ReceiVed: NoVember 5, 2008; ReVised Manuscript ReceiVed: March 2, 2009

One-dimensional ZnO nanorods with very small diameters in the range of 10 ( 2 nm were fabricated by facile sol-gel and hydrothermal methods. The structural, morphologic and optical properties of ZnO nanorods were investigated using X-ray diffraction, electron microscopy, and photoluminescence (PL). A strong ultraviolet (UV) near-band-edge emission peak at 373 nm (3.32 eV) was observed, while the deep-level emission band was too weak to observe at room temperature. The UV emission peak has an obvious blue shift compared to the samples with a larger size, which was attributed to the size-decreased nanorods. The large ratio of the UV emission to visible emissions in PL spectra can be obtained in small-sized ZnO nanorods. Furthermore, the morphologic and optical properties of ZnO nanorods can be adjusted by controlling the reactant concentration and reaction time. The possible growth mechanism of the nanorods is proposed. Introduction Zinc oxide is a unique semiconductor material with a characteristic direct and wide bandgap (3.37 eV at room temperature1) and large exciton binding energy of 60 meV. It has been recognized as a valuable optoelectronic material in efficient ultraviolet (UV) laser, piezoelectricity, energy conversion, sensing, etc.2-6 In comparison with zero-dimensional (0D) nanostructures, one-dimensional (1D) semiconducting nanostructures can facilitate more efficient carrier transport due to decreased grain boundaries, surface defects and disorders, and discontinuous interfaces. To utilize 1D ZnO nanostructures for optoelectronic nanodevices, it is essential to have detailed information about their optoelectronic properties and architectures, including size, geometry, and morphology, and to have the ability to control these properties in the fabrication process. The application of the 1D nanostructures depends on this tenability and these parameters. At present, it is still a challenge to fabricate well-controlled and small width 1D ZnO nanostructures approaching the quantum size range and to fine-tune their properties. Synthesis of 1D ZnO nanostructures with controlled small sizes exhibits great potential in various optoelectronic technical applications and device miniaturization.7,8 It is expected to show some interesting and novel electrical and optical properties owing to quantum confinement. However, the large-sized 1D ZnO structures with a width in micrometer and sub-micrometer range are generally obtained, and the usual smaller size is in the order of several tens of nanometers in diameter or thickness.9,10 The capability of controlling the width of the 1D ZnO nanostructures and reaching the nanometer regime via simple and low-cost methods is normally difficult to obtain.11,12 This obstacle hinders the expected quantum confinement effect application in the development of future nanodevices. In general, it is rather difficult to synthesize small-sized ZnO with 1D morphology in aqueous or ethanol solution via a wet * Towhomcorrespondenceshouldbeaddressed.E-mail:[email protected]. Phone: (86)431-85099168. Fax: (86)431-85684009. † Northeast Normal University. ‡ Drexel University.

chemical method, because the growth speed is too fast for precursor to control the crystal size. Some kinds of organic substance used as structure-directing reagents, such as specific surfactants, organic molecules (ethylene glycol), catalysts, etc., are introduced into the reaction process in order to decrease the crystal size. However, it is not easy to control the reaction in such a complicated process, and the morphology of the products is complex. The prevalent approaches such as vapor phase transport, chemical vapor deposition and other synthesis methods also need to solve the similar problems of reduced size. The normally obtained width is about 50 nm, and it is difficult to further decrease the width of nanorods. Herein, we developed facile and mild sol-gel and hydrothermal methods without the catalysts to fabricate ZnO nanorods with small diameters in the nanoscale regime13 and reduce the diameters to the range of 10 ( 2 nm. This solution-grown method can potentially enable ZnO nanowire dye-sensitized solar cells14 or low-cost thin-film transistor (TFT) arrays/circuits via mass manufacturing.15 A strong UV near-band-edge emission at room temperature is observed for the small-sized ZnO nanorods, while the green emission is quite weak. It is rarely seen such a blue-shift UV emission peak position coming from ZnO nanorods obtained in aqueous solution. The crystal size, morphology, and optical properties could be well controlled by modulating the reactant concentration and the reaction time. It indicates a great potential for the nanoscale device fabrication. Experimental Section The ZnO nanorods were synthesized in hydrothermal process using seeded growth method to control the crystal size. The experimental procedure included two steps. First, ZnO nanoparticles were prepared by sol-gel method reported by Spanhel et al.16,17 1.10 g of zinc acetate dehydrate (Zn(CH3COO)2 · 2H2O, A. R, Fluka) was dissolved in 100 mL of ethanol and refluxed for 2 h. The solution was then cooled to room temperature, and 2.5 mL tetramethylammonium hydroxide ((CH3)4NOH, A. R, Fluka) was added into the solution. Then the solution was hydrolyzed in an ultrasonic bath for 30 min to obtain ZnO sols. The as-prepared nanoparticles have a diameter of about 4 nm

10.1021/jp809778w CCC: $40.75  2009 American Chemical Society Published on Web 04/09/2009

7498

J. Phys. Chem. C, Vol. 113, No. 18, 2009

Figure 1. Morphological characterization of ZnO nanorods. (a) SEM and (b) TEM image and selected-area electron diffraction pattern (inset) of as-grown nanorods (1 mM 3 h) fabricated via two-step hydrothermal reaction.

calculated from UV-visible absorption spectra.18,19 These nanoparticles were used as seed nuclei. Second, 1.0 mL of ZnO sols solution was directly dropped to 30 mL of equimolar (1 mM) aqueous solution of zinc nitrate hexahydrate (Zn(NO3)2 · 6H2O, A. R, Fluka) and hexamethylenetetramine (HMT, A. R, Beijing Chemical Reagent Corp.) in a Teflonlined autoclave. The autoclave was sealed and heated at 90 °C for 3 h and then cooled to room temperature. Control experiments were performed to examine the influence of different reactant concentration and reaction time on the formation of ZnO nanostructures. The comparison samples were synthesized by different reactant concentration (2 mM, 5 mM, and 10 mM) for 3 h. Zn(NO3)2 · 6H2O and HMT were kept the same 1:1 molar ratio. For samples obtained with different reaction time, the precursor concentration of Zn(NO3)2 · 6H2O and HMT solution was kept at 1 mM. The reaction time varied from 0.5 to 36 h. The white precipitate of samples was centrifugated, washed thoroughly, and dried at 30 °C in a vacuum oven for 12 h. The morphology of the nanorods was investigated by a Hitachi S 4800 field-emission scanning electron microscopy (FE-SEM). Transmission electron microscopy (TEM) images and the selected-area electron diffraction (SAED) pattern were obtained on a Hitachi H-8100IV microscope with an accelerating voltage of 200 KV. X-ray powder diffraction (XRD) analysis was conducted on a Rigaku D/max-2500 X-ray diffractometer with Cu KR radiation (λ ) 1.5418 Å). The room-temperature photoluminescence (PL) and low-temperature PL emission spectra were recorded with a HR-800 LabRam Infinity Spectrophotometer excited by a continuous He-Cd laser with a wavelength of 325 nm. Optical absorption spectra were measured at room temperature with a Lambda 900 UV-vis Spectrophotometer (Perkin-Elmer, American). Results and Discussion The morphology of the material was investigated by SEM and TEM. The as-synthesized ZnO nanorods fabricated via seeds growth method (1 mM and 3 h) are smooth and have a smaller size than that of usually obtained samples by other methods. As shown in SEM and TEM images in parts a and b of Figure 1, the nanorods have an average diameter and length of about 10 ( 2 and 100 nm, respectively. They are unlike other large nanorods reported in the literatures. After the two-step hydro-

Chen et al. thermal process, the size decreased apparently. The inset in Figure 1b is the SAED pattern. Electron diffraction analysis revealed that the obtained samples are single crystal wurtzite structure and grow mainly along the [0001] direction. There have been many reports on the fabrication and the characterization of 1D ZnO nanomaterials.20,21 However, diameters of these nanomaterials are normally much larger than 10 nm. This smaller size structures may have potential practical merit for high-performance ZnO based nano-optoelectronic applications. To learn more about the growth information of nanorods, a series of control experiments were performed by changing the reactant (zinc nitrate hexahydrate and HMT) concentration and the reaction time. SEM images of ZnO nanorods synthesized by different reactant concentration from 2 mM to 10 mM are shown in Figure 2. Compared with the size obtained by smaller concentration (1 mM and 3 h) shown in Figure 1, the average diameter and length increase with the reactant concentration in the solution from 10 to 40 nm and 100 nm to 1 µm, respectively. The growth rate is obviously faster at a higher concentration. Because of this dependence, the optical absorption feature has been observed variable. Figure 3 shows the UV-vis absorption spectra of ZnO nanorods obtained by different reactant concentration from 1 mM to 10 mM. It can be found that the absorption edge has a red shift with increasing reactant concentrations, indicating the enlarged size of ZnO nanorods. To form ZnO nanorods with smaller diameters, the concentrations of zinc nitride and HMT in the aqueous solution should decrease. However, there is a reactant concentration limit for these nanorods with a narrow size distribution. When the concentration of precursor reaches 10 mM, two kinds of ZnO nanorods with different sizes are obtained simultaneously. It can be confirmed from the SEM images. Similar to the seed layer on the substrate in the hydrothermal method,22 these nanorods can also precipitate out based on these dispersed nanoparticles in large quantity. In this process, it is not necessary to introduce any catalyst in the aqueous solution or other surfactant agents on the surface of the nucleation seeds. Because of the aggregation of the nanoparticles on the substrate, it is very difficult to make the crystal size decrease to a certain small range. When the seed size is controlled to several nanometers, the small size nanorods can heterogeneously grow on the crystal nuclei instead of homogeneous nucleation in the aqueous solution. XRD patterns were taken to examine the crystal structure of the ZnO nanorod samples as shown in Figure 4. For all samples, all of the diffraction peaks can be indexed as the pure hexagonal phase of ZnO with a typical wurtzite structure (a ) 3.25 Å and c ) 5.21 Å). The lattice constants a and c can be calculated according to Bragg’s law 2d sin θ ) nλ. The lattice constants were given by23

a)

λ √3sin θ

c)

λ sin θ

for the (100) orientation and (002) orientation, respectively. According to Figure 4, the full width half maximum (FWHM) of (002) peak is narrowest among diffraction peaks. It exhibits the evidence of the c-axis texture. This observation accords with the SAED results in Figure 1. There is no obvious characteristic diffraction peaks associated to secondary phases or clusters in any of our samples. The PL properties of nanostructures were investigated at room temperature. Figure 5 shows the typical room-temperature PL spectra of the ZnO nanorods with different reactant concentration. It was normalized to the UV emission in order to easily

Small-Sized ZnO Nanorods

J. Phys. Chem. C, Vol. 113, No. 18, 2009 7499

Figure 2. SEM images of the ZnO nanorods synthesized at different reactant concentrations. (a) 2 mM, (b) 5 mM, and (c) 10 mM.

Figure 3. The UV-vis absorption spectra of ZnO nanorods obtained at different reactant concentrations varied from 1 to 10 mM.

Figure 4. XRD patterns of ZnO nanorods obtained at different reactant concentrations.

compare. For the ZnO nanorods with reactant (zinc nitrate hexahydrate and HMT) concentration 1 mM, the prominent UV near-band-edge exciton emission peaks with strong intensity at around 373 nm (3.32 eV) can be observed. Compared to the samples synthesized with a higher reactant concentration, the UV emission peak has a blue shift, which is rarely observed from ZnO nanorods obtained in aqueous solution. It was deduced from the size-decreased ZnO nanorods with a small diameter. In the other reported ZnO nanobelts with a width of 5.5 nm and ZnO nanofibers with a diameter less than 10 nm,11,12 similar UV PL emission position at 3.32 eV is presented for these different size 1D nanomaterials. One possible reason is that the size is not small enough. The Bohr exciton radius (2.23 nm) of zinc oxide is too small, which is not comparable to the most of 1D ZnO nanomaterials. Size distribution may be another reason. For 1D nanostructures, quantum confinement effect has to depend on the change of the width or diameter of 1D ZnO

Figure 5. Room-temperature PL spectra (normalized to the UV emission peak) of ZnO nanorods obtained at different reactant concentrations.

nanostructure. It is very different from the ZnO nanoparticles, which is defined in three dimensions. In comparison, near-bandedge emission from a smaller value such as from 3.24 to 3.30 eV can be obtained in bulk ZnO material at room temperature. For these lower energy emission positions, it is generally ascribed to either high carrier concentration or no pure-excitonic emission (e.g., supposition of a free exciton peak and a free exciton-1LO peak). However, PL emission position such as 3.32 eV rarely can be obtained in 1D ZnO material. It should be emphasized that the weak emission in the visible region is almost negligible. The sharp and intense UV emission corresponding to the near band-edge emission is normally attributed to the exciton recombination.24,25 The deep-level transition is related to oxygen vacancies, surface states, and some structural defects.26 This visible emission resulted from the radiative recombination of a photogenerated hole with an electron occupying the oxygen vacancies.25 The relative PL intensity ratio of the UV near-band-edge emission to deep-level emission is related to the crystal quality of ZnO nanostructures.27 Currently, there are literatures reported that a higher surface area to volume ratio for ZnO nanostructures with a smaller size might favor a high-level surface defects, which accounts for the increase of the green emission relative to the UV emission.28-30 Although small-sized 1D ZnO nanomaterials have been reported, the deep-level emission was strong. In our case, the PL intensity ratio of the UV emission relative to the deeplevel emission increases obviously for the ZnO nanorods with a diameter about 10 nm, which indicates good crystalline, optical quality and a low defect density for the ZnO nanorods. When the reactant concentration is increased, the UV emission peak

7500

J. Phys. Chem. C, Vol. 113, No. 18, 2009

Chen et al.

Figure 6. SEM images of the ZnO nanorods synthesized at different reaction times: (a) 9 h, (b) 18 h, and (c) 36 h.

Figure 7. The UV-vis absorption spectra of ZnO nanorods obtained at different reaction time varied from 0.5 to 36 h.

has a red shift. It indicated the enlarged size, which is consistent with SEM images in Figure 2. For samples grown with a concentration of 10 mM, there is an obvious asymmetry shape for the UV emission peak. The possible reason is that nanorods with different sizes exist in the sample. They all have separate emission centers for the UV emission. As a result, the asymmetry shape is obtained. The sizes of these nanorods can also be adjusted by controlling the reaction time. To investigate the dependence on the different reaction time, the precursor concentration of Zn(NO3)2 · 6H2O and HMT solution was kept at 1 mM. The growth time varied from 0.5 to 36 h. The corresponding SEM images of the ZnO nanorods are shown in Figure 6. With increasing growth time, the average diameter and length of these nanorods are enlarged at growth time from 3 to 18 h. For samples obtained at growth time 36 h, some nanorods are shorter than that of samples obtained at 18 h. Moreover, the growth rate is not as fast as the samples at a higher reactant concentration and a lower reaction time, and a size distribution appears accompanying this growth process. This dependence can also be exhibited by the UV-vis absorption spectra shown in Figure 7. The absorbance of ZnO sol used as nuclei is also inserted as a contrastive line. According to Figure 7, the absorption edge shifts to long wavelength side with increasing growth time, which indicates the bigger average size obtained with higher growth time. It is consistent with the corresponding SEM images. Additionally, it can also be observed that the growth rate of nanorods is faster within the early stage. Because the products are synthesized in the aqueous solution, the precursor reacts very fast at a proper temperature and is consumed quickly.

Figure 8. XRD patterns of ZnO nanorods obtained by different reaction time from 3 to 36 h.

Figure 9. Room-temperature PL spectra of ZnO nanorods obtained at different reaction times (3-36 h). It was normalized to the UV emission peak.

The growth rate becomes slow with increasing reaction time so that the Zn2+ ion concentration decreases quickly. The samples obtained at growth time 3, 9, 18, and 36 h were chosen to examine the crystal structure and PL properties. XRD spectra were shown in Figure 8. As expected, all samples exhibited similar XRD patterns, and the diffraction peaks can be readily indexed to a typical hexagonal wurtzite structure. No diffraction peaks of other compounds are detected. Similar to Figure 4, the FWHM of (002) peak is narrowest among diffraction peaks, which demonstrates the nanorods grow along [0001] direction. The PL properties of the nanostructures were examined at room temperature as shown in Figure 9. The typical room-temperature PL spectra of the ZnO nanorods were obtained at different growth time. The prominent UV near-bandedge emission peaks with strong intensity of the exciton transition above 382 nm can be still observed at growth time

Small-Sized ZnO Nanorods

J. Phys. Chem. C, Vol. 113, No. 18, 2009 7501

SCHEME 1: Schematic Illustration for Nanorod Synthesisa

a (a) ZnO nanoparticles synthesized by sol-gel method are used as nuclei. These crystal seeds experience: process I, heterogeneous nucleation (dispersed in the solution) and form nanorods; process II, nanorod self-assembly from the nucleation sites of the ZnO seed layer deposited on the substrates. (b) Larger-sized nanorods come from the spontaneous nucleation in the aqueous solution.

from 3 to 36 h. The UV emission peaks of the samples at high growth time from 9 to 36 h have a red shift compared to the samples obtained at 3 h. This peak position shift is obvious for samples obtained at growth time 36 h. The line shape of UV emission peak does not fit well a symmetric Gauss shape. It can be confirmed that this kind of asymmetry comes from the different emission centers in UV PL emission peaks. The weak deep-level emission band is centered at around 550 nm. The PL intensity ratio of UV emission to deep-level emission decreases apparently for the samples obtained at growth time 36 h, revealing that the related surface defect enhanced. It can be concluded that the growth time must be controlled within a proper range in order to obtain good crystalline. The general chemical reactions in this hydrothermal deposition process31,32 involving HMT and zinc nitrate hexahydrate as reactants can be expressed as follows

(CH2)6N4+6H2O f 6HCHO + 4NH3

(1)

NH3+H2O T NH+ 4 + OH

(2)

2OH- + Zn2+ f Zn(OH)2 T ZnO + H2O

(3)

The detailed explanation for synthesizing nanorods from nanoparticles based on heterogeneous process and the contrastive process in normal hydrothermal method is depicted in Scheme 1. In this process, the uniform small nuclei play an important role in determining the growth kinetics of these small size nanorods. In the initial stage, it is the epitaxial heterogeneous nucleation on the existing nuclei that is the crystal growth kinetics process and not the spontaneous (homogeneous) nucleation of the nanorods occurring in the aqueous solution. The lattice matching between ZnO nanorods and the nucleus seeds (ZnO nanoparticles) is crucial to the crystal growth on the seed surface. They have the same crystal type, and provide the best matching and lowest energy barrier. Heterogeneous nucleation has a lower free energy barrier of activation than homogeneous nucleation.33 Therefore, the heterogeneous growth becomes the preferential crystal growth form. The precursors are consumed for the growth of ZnO on the nuclei seeds, and the precursor concentration decreases quickly. In this case, the remaining concentration is lower than the supersaturation degree. Because spontaneous nucleation is favored under a high supersaturation degree on the basis of classical crystallization theories, the solution concentration is not adequate to initiate spontaneous nucleation. The nanorods formation with larger size coming from spontaneous nucleation is prevented. Moreover,

as a polar crystal, the hexagonal wurtzite ZnO can be described schematically as hexagonal close packing of oxygen and zinc atoms in point group 3m and space group P63mc with zinc atoms in tetrahedral sites.34,35 O2- and Zn2+ ions are arranged alternatively along the c axis. ZnO has a polar ( (0001) planes and nonpolar (1000) planes with 3-fold coordinated atoms. The polar plane with a relatively higher surface energy is metastable, while the nonpolar plane with lower surface energy is the most stable.36 Under the hydrothermal conditions, the growth velocity of polar plane is faster than other planes to minimize surface energy. This fast anisotropic growth along [0001] direction of the wurtzite structure leads to the formation of ZnO nanorods, which was confirmed by XRD results. Size distribution of nanorods could be affected by the uniformity of nanoparticle size because these nanoparticles act as nucleation sites for nanorods. Larger size distribution of nanoparticles in turn leads to size distribution of nanorods. Consequently, the uniformity of nanoparticles obtained by sol-gel method is very important for narrow size distribution of nanorods. In the current hydrothermal process to obtain the nanorod arrays, ZnO seeds synthesized by sol-gel method is deposited on the substrates. Nanorods grow from the nucleation sites of the seed layer. Because of the existence of nanoparticle aggregation on the substrates, it is difficult to further decrease the crystal width to 50 nm. However, spontaneous nucleation occurs in the supersaturated aqueous solution if there is no crystal seeds limitation. As a result, the nanorods grow along [0001] direction at a fast rate and tend to form a larger width under hydrothermal conditions. Reaction time is an important factor which affects the morphology and size of the nanostructures. It involves the change of reactant concentration and reaction rate in the thermodynamics and kinetics processes. In the heated aqueous hydrothermal process with HMT and zinc nitrate as reactants in the deposition baths, zinc hydroxide phase as a middle reaction product hydrolyzes to form ZnO shown in eq 3. Therefore, the dissolution and reprecipitation both occur in the growth process. At the early growth period, there is a fast growth process for ZnO nanorods formation because of the high supersaturation at the initial reactant concentration and proper temperature. Both the diameter and length increase quickly. The growth velocity of length along [0001] direction is faster due to the anisotropic growth of ZnO. However, the reaction rate slows with time, and the dissolution processes exceed the precipitation at the longer reaction time when the most precursors are largely consumed and Zn2+ ion concentration decreases. As shown in Figure 6c, the nanorods dissolve from the top and make the length shorter as half of that at growth time 18 h in Figure 6b. After the longer growth time, the dissolution rate is faster than the crystal growth rate. The Ostwald ripening becomes a main thermodynamics when ZnO growth reaches certain equilibrium.36 Conclusions In conclusion, we have synthesized well-crystallized 1D ZnO nanorods with small diameters in the ranges of 10 ( 2 nm at low temperature by facile and mild sol-gel and hydrothermal methods. It is a new mean to conveniently obtain the wellcontrolled and small width 1D ZnO nanostructures in aqueous solution without the catalysts. Room-temperature PL measurements showed an intense ultraviolet emission at 373 nm (3.32 eV). It has an obvious blue-shift compared to the samples synthesized with a higher reactant concentration and reaction time. The weak visible emission related to deep-level emission

7502

J. Phys. Chem. C, Vol. 113, No. 18, 2009

is almost negligible. The large ratio of UV emission to visible emission in PL spectra can be demonstrated in the small-sized ZnO nanorods. The crystal size, morphology, and optical properties can be controlled by modulating the reactant concentration and reaction time. The possible growth mechanism of the nanorods is proposed. It is desirable for their widespread potential applications as key building blocks of nanoscale optoelectronic devices. Acknowledgment. This work was supported by the National High Technology Research and Development Program of China (2006AA03Z311), the National Natural Science Foundation of China Grant Nos. 50725205, 50802014, and 10704015, and the Science Foundation for Young Teachers of Northeast Normal University (20070203). References and Notes (1) Meyer, B. K.; Alves, H.; Hofmann, D. M.; Kriegseis, W.; Forster, D.; Bertram, F.; Christen, J.; Hoffmann, A.; Strassurg, M.; Dworzak, M.; Haboeck, U.; Rodina, A. V. Phys. Status Solidi B 2004, 241, 231. (2) Huang, M. H.; Mao, S.; Feick, H.; Yan, H.; Wu, Y.; Kind, H.; Weber, E.; Russo, R.; Yang, P. Science 2001, 292, 1897. (3) Wang, X. D.; Song, J. H.; Liu, J.; Wang, Z. L. Science 2007, 316, 102. (4) Martinson, A. B. F.; Elam, J. W.; Hupp, J. T.; Pellin, M. J. Nano Lett. 2007, 7, 2183. (5) Patolsky, F.; Lieber, C. M. Mater. Today 2005, 8, 20. (6) Liao, L.; Lu, H. B.; Li, J. C.; He, H.; Wang, D. F.; Fu, D. J.; Liu, C. J. Phys. Chem. C 2007, 111, 1900. (7) Xia, Y.; Yang, P.; Sun, Y.; Wu, Y.; Yan, H. AdV. Mater. 2003, 15, 353. (8) Ozgur, U.; Alivov, Y. I.; Liu, C. J. Appl. Phys. 2005, 98, 041301. (9) Vayssieres, L. AdV. Mater. 2003, 15, 464. (10) Park, W. I.; Yi, G.-C. AdV. Mater. 2004, 16, 87. (11) Wang, X. D.; Ding, Y.; Summers, C. J.; Wang, Z. L. J. Phys. Chem. B 2004, 108, 8773. (12) Fang, Y.; Pang, Q.; Wen, X.; Wang, J.; Yang, S. Small 2006, 2, 612.

Chen et al. (13) Tong, Y.; Dong, L.; Liu, Y.; Zhao, D.; Zhang, J.; Lu, Y.; Shen, D.; Fan, X. Mater. Lett. 2007, 61, 3578. (14) Law, M.; Greene, L. E.; Johnson, J. C.; Saykally, R.; Yang, P. D. Nat. Mater. 2005, 4, 455. (15) Ong, B. S.; Li, C.; Li, Y.; Wu, Y.; Loutfy, R. J. Am. Chem. Soc. 2007, 129, 2750. (16) Spanhel, L.; Anderson, M. A. J. Am. Chem. Soc. 1991, 113, 2826. (17) Meulenkamp, E. A. J. Phys. Chem. B 1998, 102, 5566. (18) Pesika, N. S.; Stebe, K. J.; Searson, P. C. AdV. Mater. 2003, 15, 1289. (19) Sui, X. M.; Shao, C. L.; Liu, Y. C. Appl. Phys. Lett. 2005, 87, 113115. (20) Kar, S.; Pal, B. N.; Chaudhuri, S.; Chakravorty, D. J. Phys. Chem. B 2006, 110, 4605. (21) Tak, Y.; Yong, K. J. Phys. Chem. B 2005, 109, 19263. (22) Greene, L. E.; Law, M.; Goldberger, J.; Kim, F.; Johnson, J. D.; Zhang, Y. F.; Saykally, R. J.; Yang, P. D. Angew. Chem., Int. Ed. 2003, 42, 3031. (23) Shan, F. K.; Kim, B. I.; Liu, G. X.; Liu, Z. F.; Sohn, J. Y.; Lee, W. J.; Shin, B. C.; Yu, Y. S. J. Appl. Phys. 2004, 95, 4772. (24) Van, D. A.; Meulenkamp, E. A. J. Lumin. 2000, 454, 87–89. (25) Vanheusden, K.; Warren, W. L.; Seager, C. H.; Tallant, D. R.; Voigt, J. A.; Gnade, B. E. J. Appl. Phys. 1996, 79, 7983. (26) Zhang, X. T.; Liu, Y. C.; Zhang, J. Y.; Lu, Y. M.; Shen, D. Z.; Fan, X. W.; Kong, X. G. J. Cryst. Growth 2003, 254, 80. (27) Izaki, M.; Watase, S.; Takahashi, H. Appl. Phys. Lett. 2003, 83, 4930. (28) Yang, P.; Yan, H. Q.; Mao, S.; Russo, R.; Johnson, J.; Saykally, R.; Morris, N.; Pham, J.; He, R. R.; Choi, H. J. AdV. Funct. Mater. 2002, 12, 323. (29) Huang, M. H.; Wu, Y. Y.; Yang, P. D. AdV. Mater. 2001, 13, 113. (30) Tang, Q.; Zhou, W.; Qian, Y. Chem. Commun. 2004, 6, 712. (31) Govender, K.; Boyle, D. S.; Kenway, P. B.; O’Brien, P. J. Mater. Chem. 2004, 14, 2575. (32) Li, Q.; Kumar, V.; Li, Y.; Zhang, H.; Marks, T. J.; Chang, R. P. H. Chem. Mater. 2005, 17, 1001. (33) Vayssieres, L.; Keis, K.; Lindquist, S.-E.; Hagfeldt, A. J. Phys. Chem. 2001, 105, 3350. (34) Kong, X. Y.; Ding, Y.; Yang, R.; Wang, Z. L. Science 2004, 303, 1348. (35) Kar, S.; Dev, A.; Chaudhuri, S. J. Phys. Chem. B 2006, 110, 17848. (36) Peterson, R. B.; Fields, C. L.; Gregg, B. A. Langmuir 2004, 20, 5114.

JP809778W