Silver nanoparticles deposited on titanium oxide were produced by laser-liquid interaction of TiO₂ powder and AgNO₃ dissolved in isopropanol. Different weight ratios of TiO₂ and AgNO₃ were sonicated in isopropanol before being irradiated by the focused output of second harmonic (532 nm) of Nd:YAG laser operating at 10 Hz with the power of 25 mJ. Isopropanol was used as the reaction medium to prevent the aggregation of silver nanoparticles while the reaction proceeds. The sizes of the Ag nanoparticles produced by the laser-liquid interaction were found to be about 18~22 nm in diameter. Since the size, shape, and microstructure of Ag particles are dependent on some important parameters, such as power, frequency, and interaction time, those factors were considered in the laser-induced processes.

The UV-Vis measurements (200~800 nm; 300 nm/min.) of Ag/TiO₂ photocatalysts are shown in Figure 1. From the results, the absorption curve of TiO₂ indicates that TiO₂-only has no absorption in the spectral region above 420 nm. After the doping of Ag nanoparticles onto TiO₂, the absorption curves of (a)~(e) were obviously improved in the 380~780 nm spectral region. This additional absorption peak in the visible region occurs because of the surface plasmon resonance (SPR), the interference of electromagnetic field with the conduction electrons of silver particles dispersed on the TiO₂ matrix. The enhanced absorption is indicative of the greater probability of enhancing the photocatalytic efficiency of Ag/TiO₂ by broadening the light absorption in the visible region with Ag nanoparticles.

It has been reported that the doped metallic nanoparticles on TiO₂ are acting like electron traps [23], retarding the recombination of electron-hole pairs which were provoked by the photon absorption of the TiO₂ matrix. We found that among the various ratios of Ag/TiO₂ produced by laser-liquid interaction, the 2.0 wt% Ag/TiO₂ [Figure 1 curve (c)] presents the highest photon absorption in the visible light region with λ_max (maximum absorption wavelength) located at around 470 nm. Interestingly, the absorption enhancement of Ag/TiO₂ at visible light region does not correspond to the ratios of silver in the photocatalysts. Although silver nanoparticles help improve the visible light absorption, it is likely that some particles shield the interaction of light as more Ag is deposited on the TiO₂ matrix. It is also noted that a shifting of the maximum absorptions in the visible light region occurred for various ratios of Ag/TiO₂, just as curves (d) and (e) in Figure 1 are red-shifted to 500~520 nm, which is attributed to the size differences of the deposited Ag nanoparticles.

Figure 2 shows the XRD patterns of TiO₂-only and various ratios of Ag/TiO₂ produced by the laser-liquid interaction. Silver signals were clearly observed at 38°, 44.2°, 64.6° and 77.2° (2θ) of the patterns among other signals from TiO₂. As the weight percentages of Ag increase, the signals of deposited Ag are more distinct correspondingly. Compare the peaks of TiO₂ in Figure 2(a) with those in (b)~(f) from various Ag/TiO₂, these peaks looked identical to each other. It suggests that a large part of Ag particles were not incorporated in TiO₂ lattice, but deposited on the surface of the matrix instead. The crystallite size of nanoparticles can be calculated by applying to the Scherrer’s equation: (1) D   =   0.9 λ / ( B cos θ )where D is the average crystallite size, 0.9 is the shape factor of the grain, λ is the wavelength of X-ray which is 0.154051 nm for Cu Kα radiation, B is the FWHM of the diffraction peak, and θ is the incident angle of X-ray. By the diffraction data in Figure 2, the primary particle size can be measured according to the Scherrer analysis for diffraction peak width at 25.2° (2θ). The estimated sizes are about 17.7 nm, 18.3 nm, 19.0 nm, 19.0 nm, 17.7 nm, and 17.7 nm for TiO₂-only, 0.5 wt%, 1.0 wt%, 2.0 wt%, 5.0 wt%, and 10.0 wt% Ag/TiO₂, respectively. The crystallinity of photocatalysts can be estimated by the relative peak areas of XRD to standard of P25 TiO₂, which is considered to be 100% crystalline, because the intensity of the X-ray scattering is proportional to the peak area above the background scattering [24]. The degree of crystallinity for these photocatalysts are 86.3%, 101.0%, 88.8%, 97.8%, and 83.8% for 0.5 wt%, 1.0 wt%, 2.0 wt%, 5.0 wt%, and 10.0 wt% Ag/TiO₂, respectively. The results suggest that the use of various amount of silver doping leads to photocatalysts with different crystallinities.

The SEM image of 2.0 wt% Ag/TiO₂ nanoparticles is shown in Figure 3. Some porous surface dispersed among cauliflower-like clusters of grains was observed. The scope of particle size was measured to be approximately 30~45 nm. The results revealed that some aggregation of TiO₂ grains occurs during laser-liquid interaction if we compare the particle size from SEM and from the XRD data calculation.

Figure 4 shows the TEM image of 2.0 wt% Ag/TiO₂ nanoparticles. Small spherical Ag nanoparticles (some of these particles are indicated by arrows) were observed scattered on the surface of TiO₂. For Ag particles, the diameter ranges from 3 to 6 nm, and the diameter of TiO₂ was found to be in the scope of 20~40 nm. It suggests that small silver nanoparticles can be prepared by pulsed-laser irradiation applied to AgNO₃ and TiO₂ system, and that a good dispersion of these particles on the surface of the matrix is probable. The average diameter of particles from TEM image was found to be in accordance with the results of the SEM image.

We have also performed the EDX analysis on the 2.0 wt% Ag/TiO₂ catalyst. The EDX diagram of 2.0 wt% Ag/TiO₂ is shown in Figure 5, where the silver signals are found at around 3.00 keV [25]. Though the peaks of silver are insignificant due to its content in TiO₂ matrix, it can be indicative of the presence of Ag particles in catalyst.

In this article, methylene blue was used as a model pollutant for evaluation of the photocatalytical efficiency of the laser-induced Ag/TiO₂ nanoparticles. Methylene blue, with an absorption maximum at 668 nm in visible light region, as shown in Figure 6, is usually used in mixed indicators or as a redox indicator. Hence, the amount of MB was measured quantitatively with the absorption of light at 668 nm.

Figure 7 presents the light emitting spectrum of the halogen lamp applied in the degradation procedure, its emitting wavelength ranges from 350 nm to 800 nm with the predominant peak at 575 nm. The power of the halogen lamp is 150 W. Because the band gap of TiO₂ is 3.0~3.2 eV, a wavelength greater than 387~414 nm will be short of any absorption by TiO₂-only, as shown in Figure 1. If the photocatalysis of Ag/TiO₂ in visible-light range improves through halogen lamp irradiation, it can be attributed to the absorption of visible light by Ag particles doped in the TiO₂.

Before the photodegradation experiments were conducted, the possible reactions such as adsorption of MB by nanoparticles and direct light-degradation were studied as follows: the adsorption of MB by Ag/TiO₂ was carried out by putting the photocatalyst into the MB solution, an aliquot of solution was examined successively by UV-Vis at 20-min intervals. From our experiments, it took about 20 min. to reach the saturated adsorption of MB by Ag/TiO₂; the direct light-degradation test was carried out by illuminating the solution directly with a halogen lamp without photocatalysts. There is no obvious degradation of MB in solution in two hours, just as the result observed in Figure 8(a). These effects were taken into consideration for the photodegradation experiments of this work.

The pH value of the solution also plays an important role in the photodegradation process; it was found that the maximum rate of photocatalytic degradation by TiO₂ is achieved at pH 6.9 [26]. Hence, the pH condition of photodegradation experiments of this work was controlled before the degradation process for optimum results. Figure 8 shows the photodegradation of MB solution in the presence of Ag/TiO₂ initiated by the illumination of a halogen lamp. Line (a) in Figure 8 presents the degradation reaction of MB in the absence of catalyst, revealing no change in MB concentration under this condition. Rapid conversion of MB (about 28~35%) is found for the first 20 min. irradiation except line (f), which is about 10% conversion for 5.0 wt% Ag/TiO₂. The conversion of MB progressed gradually afterward, reaching 54, 45, 40, 45, 22, and 55% for 0.0, 0.5, 1.0, 2.0, 5.0, and 10.0 wt% Ag/TiO₂ at 1 h irradiation, respectively. The maximum efficiency was achieved by 2.0 wt% Ag/TiO₂ for an 82.3% conversion at 2 h irradiation. The degradation of MB displayed a first-order reaction, rate constants of various ratios of Ag/TiO₂ are listed in Table 1. However, zero order of the degradation reaction for higher concentrations was reported elsewhere [5].

In our photodegradation tests, the most efficient photocatalyt of all the ratios of Ag to TiO₂ tested was found to be the 2.0 wt% Ag/TiO₂. A maximum of 82.3% MB degradation by Ag/TiO₂ in aqueous solution was achieved. However, the photodegradation efficiency of Ag/TiO₂ did not exhibit the increasing trend with Ag content greater than 2.0 wt%, the efficiency decreases with higher Ag contents were observed instead. According to other studies [23], the detrimental effects of Ag in greater amount relative to TiO₂ can be explained as follows: (1) the increasing amount of Ag on TiO₂ becomes a barrier hindering the contact of dye molecules with TiO₂; (2) the Ag particles prevent the light absorption of TiO₂; (3) excess Ag on TiO₂ facilitates the recombination of electron-hole pairs of TiO₂ in bulk or at the surface; (4) greater contents of Ag particles inhibit the reaction of oxygen in the photocatalytic process. In view of this, it is important to adjust the Ag and TiO₂ composition to the proper ratio in order to acquire the optimum photodegradation efficiency of Ag/TiO₂ photocatalyst dealing with MB aqueous solution.

P25 TiO₂ was purchased from Degussa (Germany). Silver nitrate was purchased from Hayashi Pure Chemical Industry (99.5% purity). Isopropyl alcohol was purchased from Hayashi Pure Chemical Industry (99.9% purity). Methylene Blue was purchased from Panreac Quimica SA. Deionized water was used throughout all experiments. All reagents were used without further purification.

Silver nitrate was used as Ag precursor. The various ratios of AgNO₃ to TiO₂ were prepared by mixing 0.5, 1.0, 2.0, 5.0, and 10.0 weight percents of AgNO₃ in TiO₂, with total weight maintained at 1.0 g. Then 10 g of isopropanol was added to each mixture. Each solution was sonicated 30 min. for uniform mixing. A pulsed-laser (25 mJ, 532 nm, 10 Hz Nd:YAG) light was applied from top of the container to each sample for 60 min. Finally, the air-dry samples were annealed at 200 °C for 60 min.

The UV-Vis spectra of Ag/TiO₂ were carried out on a Hitachi U-3010 spectrophotometer with an integrated sphere, wavelength from 200 nm through 800 nm, with a scanning rate of 300 nm/min. The XRD analyses were carried out on a Shimadzu XD-D1 X-ray diffractometer, using Cu-Kα radiation with λ = 0.154051 nm, in a range of 20–80° (2θ). TEM images were recorded on a JEOL JEM 1200-EX electron microscope. SEM images were recorded on a Philips XL40 microscope.

The installation for photocatalytic degradation of MB by Ag/TiO₂ was assembled as the following: A 150 W halogen lamp, wavelength range from 350 nm to 800 nm with the predominant peak at 575 nm, was used as the light source placed on top of the setup. 1.0 g of Ag/TiO₂ was added in 100 mL of MB (7,000 mg/L) solution. After the mixture was sonicated for 30 min., the halogen lamp was turned on to initiate the reaction. During the irradiation periods, 5 mL of solution were taken out of the reactor and centrifuged to separate the solid from the solution at 20-min intervals. An UV-Vis spectroscopy was used to detect the MB concentration of each centrifuged solution, which was collected at 20-min interval for two hours of reaction time in all.