Ti plates (99.9% in purity, Titan Engineering, Singapore) with sizes of 12.5 × 15 × 1 cm³ were pickled in the mixture of HF, HNO₃ and H₂O by volume ration of 1:3:6 at ambient temperature. After ultrasonically cleaning in deionized water, each piece of the Ti plate was soaked in a 50 ml H₂O₂ solution with different concentrations. The reactants were kept at 353 K with ultrasonication for a certain time in water bath. Then, the Ti plate was taken out and gently rinsed with deionized water and dried in oven. Ultrasonication and drying time, as well as concentration of H₂O₂ were tested systematically to study their influence on the photoctalysts activity. The morphology of as-prepared photocatalyst was observed via field emission scanning electron microscopy (FE-SEM, JEOL 6340F, Japan). Samples were characterized on X-ray diffraction (XRD, Bruker, Germany) with Cu Κα1 radiation (λ = 1.5406 Å). Surface elemental analysis was conducted on energy-dispersive X-ray detector equipped on the FESEM.

The photocatalytic activity of the prepared Ti plates was conducted in a glass reactor in a blackbox at room temperature. The light source was a 9 W UVC spot light source. A piece of prepared TiO₂ plate (4.5 × 5 cm²) was immersed into 100 mL of phenol solution (100 mg/L) for certain time. The reaction solution was magnetically stirred continuously during the photocatalytic reaction. Temperature variation throughout the reaction is less than 3 K. Phenol concentration change was analyzed by a high performance liquid phase chromatography (HPLC, LC-20AD, Shimadzu, Japan) with UV detector at a wavelength of 272 nm, using a 250 mm C15 RP column with particle size at 5 µm. The mobile phase employed was methanol/water (60/40) with flow rate at1 mL/min. Parallel experiments were carried out to ensure the reproducibility. The adsorption of phenol on the TiO₂ plate was assessed by the dark condition reaction. The photocatalytic activity of the Ti plate was defined as follows:

where C₀ is the phenol initial concentration and C_t is the concentration at reaction time (t).

The morphology of the aged Ti plate is shown in Figure 1. Uniform photocatalyst thin layers were generated via H₂O₂ oxidization on both sides of the Ti plates (Figure 1(a)). Figure 1(b) shows densely packed microfibers with uniformly distributed nanoflakes which look like sea corals. High magnification FESEM images in Figure 1(c) and Figure 1(d) show the surface layer consisted of well-aligned nanoflakes with tens of nm in width and hundreds of nm in length. The XRD pattern in Figure 3 shows that this surface layer is composed of well-crys- tallized anatase phase TiO₂.

As shown in Figure 1(d) of the cross-sectional FESEM image, the grown thin film has a depth of ca. 0.2 - 0.3 µm. The porosity and surface-to-volume ratios of the sample were greatly enhanced by these primary nanostructures, which could induce efficient diffusion and increased mass transfer of organic pollutant molecules within the microfibers architecture. The EDX spectra shows that the TiO₂ nanoflakes were elementally composed of only pure Ti and O with no extraneous elements from intermediates observed in the final anatase product.

Figure 2 shows the FESEM images of the samples prepared with and without ultrasonication. The nanostructure and density of the thin film prepared in water bath with ultrasonication is obviously finer than that obtained without ultrasonication. It could be attributed to the higher heating uniformity, lower temperature variation and higher heat transfer speed introduced by ultrasonication.

The as-prepared Ti plates were either dried in air or aged in oven at 80˚C for 8 h, respectively. The XRD patterns of the completed products in Figure 3 shows that substantial crystallization of TiO₂ in an atasephase took place in the aging process, indicating that aging is not only helpful for surface stabilization, but also for the atomic diffusion and crystallization.

Figure 3(b) shows the XRD pattern of the TiO₂ nanoflakes harvested after aging. The obtained diffraction peaks are well matched with the JCPDS Card No.: 21-1272 indicating the body-centered tetragonal anatase phase TiO₂. The diffraction peaks at 2θ = 25.25˚ (101), 47.98˚ (200), 53.59˚ (106), 37.82˚ (004) and 62.36˚ (215) are all corresponding to the lattice plane of pureanatase phase. The absence of peaks at 2θ = 27.5˚ (110) and 30.8˚ (121) corresponding to rutile and brookite phase confirms the synthesized TiO₂ was in pure anatasephase [23]. The aged TiO₂ shows a wide (101) peak with more background disturbance indicating that the amorphous nature of the synthesized nanomaterials. However, no obvious peak was obtained for TiO₂ prepared without ultrosonication. The crystallite sizes are calculated from the broadening of the anatase (101) peaks using Scherer’s formula.

D = Kλ/(βcosθ) (2)

where D is the crystallite size (nm), K is the shape constant (0.9), k is the wave length of Cu Κα1 radiation (1.5406 Å), θ is the diffraction angle (˚) and β is the full width at half maximum peak height. The crystallite size of the air dried and aged TiO₂ is calculated as 14.8 and 4.65 nm, respectively. The result confirms that aging process helps crystallization of synthesized TiO₂ nanoflakes.

From the above investigation, it is clear that low-temperature aging of samples promoted the crystallinity of anatase formed and enhanced polymorph stability. Based on these results, the effect of H₂O₂ concentration was further studied. The oxidization reactions were conducted at 80˚C using four different concentrations (5%, 10%, 20% and 35%) of H₂O₂. Figure 4 illustrates the SEM images of the surface morphology of the four samples. The uniformity and density of the grown nanoflakes are increased with increasing H₂O₂ concentration. Figure 4(d) demonstrates the largest density and contents of TiO₂ was obtained by 35 wt% H₂O₂.

In order to evaluate the photocatalytic activity of the prepared photocatalyst, photocatalytic oxidation of phenol was carried out using catalysts prepared with different conditions as shown in Figure 5. Dark condition experiment with photocatalyst was carried out to examine the adsorption of phenol molecules on the catalyst surface. The control experiment was carried out with nophotocatalyst but with light only to evaluate the photolysis of phenol. Three replicates of all experiments were performed at identical conditions to perform error analysis.

Figure 5 shows that only a modest removal of phenol occurred in the control experiment (<20%) due to direct photolysis. While under dark condition with the synthesized photocatalyst, it is clear that limited amount of phenol was removed by adsorption. Comparing between samples prepared with/without ultrasonication, after 6 hours reaction, 94% and 72% of phenol was removed, respectively, which is in line with the surface morphology analysis.

Figure 6 shows the result of photocatalytic degradation of phenol with the air-dried and aged photocatalysts. The aged photocatalyst shows much better photocatalytic activity than that of the air-dried photocatalyst. The result can be attributed to the difference in UV light utilization by these two synthesized photocatalysts. Similar

results have been reported by other researchers on TiO₂ photocatalysts [24]. In general, photocatalytic activity depends on the efficiency of utilization of the incident radiation absorbed by the catalyst [25]. Due to amorphous nature of the air-dried TiO₂ photocatalyst, the scattering of incident UV light energy was more than the absorbed energy (results in decreased formation of electron and hole pairs) as compared to that of the aged catalyst, which may be resulted in decreased photocatalytic activity.