actants with argon as the carrier gas. Growth time varied from 0.5 to 2.0 h, so as to obtain different thickness of films. ZnO nanoneedle arrays were prepared on Si substrates at the same growth temperature using the same MOCVD system [7]. Prior to the nanoneedle growth, no metal catalyst was deposited on the substrates. Highly dense nanoneedle arrays were vertically grown on the entire substrate. Surface morphologies of both ZnO thin films and nanoneedle arrays were characterized using field-emission gun scanning electron microscopy (Philips XL30S). Details of the film and nanoneedle growth and structural characterization are described elsewhere [7,8,10].
Photocatalytic activities of the samples were determined by measuring the decoloration of Orange II [4-(2-hydroxy-1-naphthylazo)benzenesulfonic acid] solution, which was selected as a test compound. The as-grown ZnO films and nanoneedle arrays were immersed into 5 ml of an aqueous solution of Orange II (0.01 mM, pH 6.2), and were irradiated with a 200 W mercury lamp through a Pyrex filter to remove wavelengths shorter than 300 nm. The incident light intensity was measured to be 1.1 mW cm ±2 using a power meter. For each set of photolysis experiments, four samples and a control (without ZnO) were simultaneously irradiated under ultraviolet light. The irradiation time varied between 1 and 15 h, and was 5 h in most cases. The dye solution was sampled periodically during the irradiation in order to check the degree of decoloration, which was done by measuring the absorbance at 486 nm as a function of irradiation time using a UV-vis spectrophotometer (Shimadzu, 2401PC). The decomposed dye contents were calculated using the following equation: 100(1ÀI/I 0 ) where I and I 0 are the absorbance peak intensity of the sample and the control, respectively. This equation is valid if there is a linear relationship between the absorbance peak intensity and the dye concentration. We experimentally confirmed the linear relationship in the dye concentration range below 0.02 mM.