Photocatalytic bactericidal activity of silver-sensitized titanium dioxide on Micrococcus lylae

ABSTRACT Under UV irradiation, titanium dioxide (TiO2) exhibits a strong bactericidal activity through the generation of hydroxyl radicals (•OH). Silver (Ag) sensitization is an effective way to enhance photocatalytic activity of TiO2. In the present study, Micrococcus lylae was used as a model bacterium to compare the bactericidal activity of Agsensitized TiO2 (in two different Ag/TiO2 molar ratios) and pure TiO2 (P25). When the concentration of photocatalysts was fixed on 0.2 mg/ml with 300 rpm stirring, no obvious difference observed among the three photocatalysts. However, the Ag-sensitized photocatalysts with higher Ag/TiO2 ratio showed better bactericidal efficiency when their concentration decreased (0.1 mg/ml) or the stirring speed increased (380 rpm). The results indicated that optimizing the phosico-chemical conditions of reaction promoted the efficiency of photocatalyst. Moreover, transmission electron microscopy (TEM) was used to observe the sub-cellular structural changes of M. lylae during photocatalytic oxidation (PCO). According to the TEM images, the disruption of cell wall occurred at a relatively long time after the cell death. The cause of cell death was the destruction of plasma membrane induced by membrane permeable •OH. These results supported that both modification on photocatalyst properties and optimization on reaction conditions enhanced the bactericidal efficiency of PCO

Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 47 - Photocatalytic bactericidal activity of silver-sensitized titanium dioxide on Micrococcus lylae H.Y. Yip 1,2 , Jimmy C.M. Yu 2 , S.C. Chan 3 , L.Z. Zhang 2 and P.K. Wong 1 1 Biology Department and 2 Chemistry Department, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong SAR, China. 3 EnvironmentalCare Limited, Tam Kon Shan Road, North Tsing Yi, Hong Kong SAR, China E-mail: pkowng@cuhk.edu.hk (P.K. Wong) ABSTRACT Under UV irradiation, titanium dioxide (TiO 2 ) exhibits a strong bactericidal activity through the generation of hydroxyl radicals (•OH). Silver (Ag) sensitization is an effective way to enhance photocatalytic activity of TiO 2 . In the present study, Micrococcus lylae was used as a model bacterium to compare the bactericidal activity of Ag- sensitized TiO 2 (in two different Ag/TiO 2 molar ratios) and pure TiO 2 (P25). When the concentration of photocatalysts was fixed on 0.2 mg/ml with 300 rpm stirring, no obvious difference observed among the three photocatalysts. However, the Ag-sensitized photocatalysts with higher Ag/TiO 2 ratio showed better bactericidal efficiency when their concentration decreased (0.1 mg/ml) or the stirring speed increased (380 rpm). The results indicated that optimizing the phosico-chemical conditions of reaction promoted the efficiency of photocatalyst. Moreover, transmission electron microscopy (TEM) was used to observe the sub-cellular structural changes of M. lylae during photocatalytic oxidation (PCO). According to the TEM images, the disruption of cell wall occurred at a relatively long time after the cell death. The cause of cell death was the destruction of plasma membrane induced by membrane permeable •OH. These results supported that both modification on photocatalyst properties and optimization on reaction conditions enhanced the bactericidal efficiency of PCO. KEYWORDS Photocatalytic oxidation, Bactericidal activity, Silver sensitized titanium dioxide INTORDUCTION Photocatalytic oxidation (PCO) has been accepted as a promising technology for the detoxification and disinfection of water and wastewater. In the presence of water and oxygen molecules, photocatalyst, such as titanium dioxide (TiO 2 ) generates reactive hydroxyl radicals (•OH) through a series of charge carriers (electron-hole pairs) mediated reactions when irradiated with near UV (λ < 385 nm). Hydroxyl radical is a powerful oxidizing species. It is highly toxic towards microorganisms and very reactive in the oxidation of toxic chemicals such as phenol and polychlorinated biphenyls (Huang and Hong, 2000; Zhang et al. 2001; Dunlop et al., 2002; Wolfrum et al., 2002). Photocatalytic inactivation of bacteria (Bacillus subtilis, Escherichia coli, Pseudomonas aeruginosa, and strains of Streptococci), bacterial spores (B. subtilis spores and Clostridium perfringens spores), fungal spore (Aspergillus niger spores) and virus (Phage Qβ and poliovirus 1) have been investigated (Saito et al., 1992; Watts et al., 1995; Butterfield et al., 1997; Lee et al., 1997; Vidal et al., 1999; Amézaga-Madrid et al., 2002; Dunlop et al., 2002; Wolfrum et al., 2002; Yu et al., 2002). In these inactivation studies, UV (300 – 400 nm) or solar light was used as energy source for the photocatalytic disinfection processes. The mechanism of bactericidal activity of PCO was first proposed by Matsunaga et al. (1988). They believed that the killing of bacteria by PCO was related to the amount of oxidized coenzyme A. The increase in dimeric form of coenzyme A was the root cause of decreases in metabolic activities. Later, Saito et al. (1992) reported that destruction of cell membrane led to the leakage of potassium ions, protein and genetic materials that Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 48 - paralleled cell death. In general, the different sensitivity of microorganisms to PCO was due to their structural differences, particularly in the thickness and complexity of cell wall. Silver ion (Ag + ) is a bactericidal agent (Pedahzur et al., 1995). The binding of Ag + and respiratory enzymes initiates changes in permeability of plasma membranes of bacteria and results in changes in cell body (Semikina and Skularhev, 1990). Coating (Absorbing or other words) Ag + on TiO 2 surface may result in a highly effective disinfectant. Firstly, TiO 2 acts as a substrate for supporting (Fixing) Ag + . The gradually release of Ag + from the surface can kill bacteria directly. Besides, during PCO, the recombination of electron-hole pairs on TiO 2 surface can be retarded as Ag + which acts as an electron acceptor. Therefore, the time for •OH generation can be extended and the effectiveness in killing bacteria increased. However, in the present study, TiO 2 was sensitized with metallic form of Ag because Ag + is easily released from the photocatalyst surface and lowers the reusability of sensitized photocatalyst. As Ag is a very good conductor, sensitization of it on TiO 2 surface can promote the transfer of electrons from the hole. Better charge separation results in less recombination and therefore, improving the photocatalytic activity. Reusable photocatalyst can minimized the cost in practical applications and environmental friendly bactericidal method is of strong current interest. In this study, the bactericidal activities of Ag sensitized TiO 2 with different Ag/TiO 2 molar ratios and pure TiO 2 under near-UV irradiation were compared. The effects of photocatalyst concentrations and stirring speeds were also investigated. Lastly, structural changes in bacteria during PCO were observed by transmission electron microscopy. METHODS AND MATERIALS Preparation of photocatalysts TiO 2 P25 was purchased from Degussa Corporation (Frankfurt, Germany) and used as a standard photocatalyst for comparison with Ag sensitized TiO 2 . Ag sensitized TiO 2 was prepared by suspending 1.0 g TiO 2 P25 in 0.1% P123 (triblock copolymer) ethanol solution with addition of 0.1 M silver ammonia complex ions afterwards. After stirring under ambient light for 1 h, the resulting Ag sensitized TiO 2 was recovered by centrifugation and washed by deionized water and ethanol. The Ag sensitized TiO 2 in two different Ag/TiO 2 molar ratio, namely Ag1 (5.07 x 10 -4 ) and Ag2 (1.17 x 10 -3 ), were prepared. Preparation of bacterial culture Micrococcus lylae, a Gram positive bacterium that was isolated in our laboratory, was used as a model bacterium in the experiment. It was incubated in 10% Trypticase soy broth (TSB) at 30°C and agitated at 200 rpm for 24 h. The culture was washed with 0.9% saline by centrifugation at 21,000×g for 5 min at 25°C and the pellet was resuspended in saline. The cell suspension was adjusted in centrifuged tube to the required cell concentration (3 x 10 7 cfu ml -1 ). Photocatalytic reaction Bactericidal effects of the three powdered photocatalysts, including P25, Ag1 and Ag2, were tested separately. The photocatalyst was added to 0.9% saline in a conical flask and homogenized by sonication. The suspension was then sterilized by autoclaving at 121°C for 20 min, allowed to cool, and mixed with the prepared cell suspension. The final photocatalyst concentration was adjusted to 0.1–0.2 mgml -1 and the final bacterial cell concentration was 3x10 6 cfu ml -1 . The photocatalytic reaction was started by irradiating the mixture with near UV light (maximum emission at 365 nm) and stopped by switching off the light. Each set of experiment was performed in triplicate. The light source used was 15W UV lamp (UV intensity: 2.6 Wm -2 ) mounted closely on one side of the flask. The reaction mixture was stirred with a magnetic stirrer to prevent settling of the photocatalyst. A bacterial suspension without photocatalyst was irradiated as a control and the Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 49 - reaction mixture with no UV irradiation was used as a dark control. Before and during the light irradiation, an aliquot of the reaction mixture was immediately diluted with 0.9% saline and plated on TSB agar. The colonies were counted after incubation at 37°C for 48 h. The survival of bacterial population during PCO was calculated by the equation: Survival (%) = (P T / P I ) x 100% where P I represented the initial population and P T represented the population after irradiation time (T). Transmission electron microscopy (TEM) Aliquots of reaction suspension were sampled for TEM study after 0, 30, 60 and 120 min of photocatalytic reaction. Samples which consisted of TiO 2 and M. lylae were centrifuged and pre- fixed in 3% glutaraldehyde for 2 h, washed two times with 0.1 M phosphate buffer (PBS) (pH 7.2) and post-fixed for 2 h in 1% osmium tetraoxide. After washing with PBS, the specimens were dehydrated in a graded series of ethanol and embedded in Spurr for polymerization. Ultra-thin sections (70 nm) were made with an ultra-microtome using a diamond knife, stained with 2.5% uranyl acetate and 2% lead citrate and examined by transmission electron microscopy (JEM- 1200EXII, JOEL Ltd., Tokyo, Japan) at 100 kV accelerating voltage. RESULTS AND DISCUSSION Bactericidal effect of Ag-sensitized photocatalyst In order to investigate the bactericidal effect of photocatalyst itself towards M. lylae, a dark control experiment was performed. Results showed that, bacterial survival was not affected by mixing with the three photocatalysts up to 60 min (Fig. 1). No Ag + released from Ag1 and Ag2 might take part in killing bacteria. It is also possible that the amount of Ag + released is well below the threshold level for triggering a bactericidal response under the present coating method. It has been confirmed by inductively coupled plasma atomic emission spectrometric measurement that the concentration of Ag + in the working solutions with Ag1 or Ag2 powder after 1 h of stirring in darkness was well below the detection limit (3 µg l -1 ). The antibacterial effect of Ag + is highly dependent on the ionization efficiency of Ag from TiO 2 surface and diffusion efficiency to the bacterial cell membrane (Keleher et al., 2002). Irradiation Time (min) 0 10203040506070 Survival (%) 0 20 40 60 80 100 120 P25 Dark Control Ag1 Dark Control Ag2 Dark Control Figure 1. Bactericidal effect of the three photocatalysts (0.2 mg ml -1 each) on M. lylae (3 x 10 6 cfu ml -1 ) survival over time. Test conducted in darkness with 300 rpm magnetic stirring. Each data point and error bar represents the mean and standard deviation of independent triplicates. Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 50 - Effect of Ag content in photocatalysts Bactericidal activities of the three photocatalysts were much more obvious under UV irradiation (Fig. 2). After 60 min of irradiation, about 80% of bacterial cells were killed. The decrease in bacterial population could only be due to the addition of photocatalysts since M. lylae was found to be UV resistant in the UV light control experiment. There was a difference of at least 10% in bactericidal activity between Ag2 and the other two photocatalysts after 60 min of UV irradiation. This result showed that Ag sensitization on TiO 2 could surely promote bactericidal activity but the Ag/TiO 2 molar ratio should be as high as that in the Ag2 sample. No difference in bactericidal activity was found between Ag1 and P25. Irradiation Time (min) 0 10203040506070 Survival (%) 0 20 40 60 80 100 120 P25 0.1 mg/ml Ag1 0.1 mg/ml Ag2 0.1 mg/ml UV light control Figure 2. Bactericidal effect of the three photocatalysts (0.1 mg ml -1 each) on M. lylae (3 x 10 6 cfu ml -1 ) survival over time. Test conducted in darkness with 300 rpm magnetic stirring. Each data point and error bar represents the mean and standard deviation of independent triplicates. Effect of photocatalyst concentration After increasing the photocatalyst concentration from 0.1 mg ml -1 to 0.2 mg ml -1 , there was no obvious difference in bactericidal activity among the three photocatalysts (Fig. 3). Ag sensitized TiO 2 did not show higher efficiency than pure TiO 2 in higher concentration. Ag2 showed better performance in lower concentration (0.1 mgml -1 ), but it just showed similar bactericidal activity to Ag1 and pure TiO 2 in 0.2 mg/ml. Irradiation Time (min) 0 10203040506070 Survival (%) 0 20 40 60 80 100 120 P25 0.2 mg/ml Ag1 0.2 mg/ml Ag2 0.2 mg/ml UV light control Figure 3. Bactericidal effect of the three photocatalysts (0.2 mg ml -1 each) on M. lylae (3 x 10 6 cfu ml -1 ) survival over time. Test conducted in darkness with 300 rpm magnetic stirring. Each data point and error bar represents the mean and standard deviation of independent triplicates. Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 51 - Increasing the concentration of photocatalysts did not increase the bactericidal activity. It might be because Ag2 at 0.1 mg ml -1 already provided the optimal concentration for the reaction and photocatalyst concentration was not the limiting factor for Ag1 and pure TiO 2 . Obviously, an advantage of using Ag2 is that smaller amount (0.1 m gml -1 ) of it is enough to provide the same bactericidal activity as in 0.2 mg ml -1 . It is important in cost saving in real application. Some previous studies showed that high concentration of photocatalyst was usually not favorable for PCO processes. San et al. (2001) performed a kinetic study on PCO of organic compound and found that at high TiO 2 concentration, aggregated particles reduced the interfacial area between reaction solution and the photocatalyst, and the number of active sites on TiO 2 surface that available for oxidation decreased. Moreover, addition of a high dose of TiO 2 increased the opacity and decreased light penetration by scattering of photons, which in turn lowered the energy provided for the initiation of PCO reactions (Chen et al., 2001). Besides these organic degradation studies, the negative effect of high dose photocatalyst was also observed in bactericidal studies, Saito et al. (1992) reported that there was only a 2.5 times increase in bacterial population when the concentration of TiO 2 was increased from 1 mgml -1 to 10 mgml -1 . In a previous E. coli disinfection study, it was found that TiO 2 concentration greater than 1mgml -1 actually decreased the killing efficiency (Maness et al., 1999). It might be due to the shading effect which caused light in TiO 2 cell suspension to become limiting. Usually, this limitation can be solved by increasing the UV intensity since higher UV intensity would provide more energy for •OH generation on the photocatalyst surface. Thus, increasing the UV intensity directly should enhance the oxidation efficiency of photocatalyst. Because of the limitation on adding UV lamp in the reactor of present study, the enhancement on bactericidal activity by UV intensity could not be tested. Under the condition of fixed UV intensity, adjustment of photocatalyst concentration is a critical step for maximizing the reaction efficiency but minimizing the shading effect. Effect of stirring speed In addition to the effect of photocatalyst concentrations, the effect of stirring speed in bactericidal efficiency was investigated (Fig. 4). It was found that increasing the stirring speed from 300 to 380 rpm improved bactericidal activity of Ag2, but it did not affect Ag1 and pure TiO 2 as photocatalyst. Irradiation Time (min) 0 10203040506070 Survival (%) 0 20 40 60 80 100 120 P25 increased stirring Ag1 increased stirring Ag2 increased stirring UV light control Figure 4. Bactericidal effect of the three photocatalysts (0.2 mg ml -1 each) on M. lylae (3 x 10 6 cfu ml -1 ) survival over time. Test conducted under UV irradiation with 380 rpm magnetic stirring. Each data point and error bar represents the mean and standard deviation of independent triplicates. Stirring is important for preventing the catalyst-cell slurry from settling. In addition, it promotes bactericidal action by increasing contact between photocatalyst particles and bacterial cells. As photocatalytic reactions can occur only on the surface of photocatalyst, increasing collision between Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 52 - photocatalyst and target compound advances the oxidation (San et al., 2001). Increasing the stirring speed improved the bactericidal efficiency of Ag2. However, mixing was not the limiting factor for the bactericidal action of Ag1 and pure TiO 2 since increasing the stirring speed did not seem to enhance the bactericidal activity of the two photocatalysts. Transmission electron microscopy (TEM) As Ag2 showed better performance among the three photocatalysts, it was used for TEM study under near-UV irradiation. Fig. 5 shows the TEM images of M. lylae after PCO. After 30 min of irradiation, morphological changes were observed in some cells (Fig. 5b). Some electron translucent portions appeared but destruction of cell wall was not observed. After 60 min, the hollow regions were extended and the morphological changes appeared in most of the cells (Fig. 5c). After 120 min, many cells were disrupted (Fig. 5d). (a) (b) (c) (d) Figure 5. TEM images of M. lylae mixed with 0.2 mg ml -1 of Ag2 before (0 min, (a)) and after 30 min (b), 60 min (c) and 120 min (d) of UV irradiation. The TEM images and experimental results demonstrated that the cell death was not due to disruption of cell wall. The survival ratio of bacterial population began to decrease when at the onset of irradiation. A previous study found that the photocatalyst was unable to attack cell wall as it was protected by the outer peptidoglycan layer in the earlier stage of reaction (Saito et al., 1992). Instead, the plasma membrane was firstly attacked by reactive oxygen species (e.g. superoxide radicals) which were generated from water molecules around the photocatalyst and were able to penetrate the outer layer of bacteria. These reactive species oxidized the membrane and broke the main permeability barrier of bacteria. The slow leakage of intracellular materials, including RNA, protein and minerals, explained the loss of cell viability at the early stage of PCO and the appearance of electron-translucent region after 30 min of irradiation. Cell wall destruction was believed to be a secondary phenomenon after the loss of cell viability. CONCLUSION Based on the experimental results, disinfection of water by PCO is feasible. The effectiveness of bactericidal action can be improved by optimization of physico-chemical conditions. Ag2 was more sensitive than Ag1 and TiO 2 in response to the change in photocatalyst concentration and stirring speed. According to the TEM images, the plasma membrane of M. lylae was the first target of reactive oxygen species and it was the main reason of cell death. The cell wall destruction was only a secondary phenomenon after the loss of cell viability. In the application of PCO for disinfection of water, photocatalyst is the major expenditure besides the capital cost in machinery. As the photocatalyst is reusable, it should be regenerated. Sensitization of metallic Ag on TiO 2 surface can enhance the photocatalytic activity and lowers the expenditure on catalyst. Moreover, operational cost in electricity can be reduced as the irradiation time is shortened by the use of more effective photocatalyst. In order to develop PCO as an effective treatment of bacteria in environmental samples, further studies are required to explore the Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 53 - possibility of lowering the cost with maximum effectiveness and convenient application such as using sunlight as energy source or immobilizing photocatalysts for easier collection and reuse. ACKNOWLEDGMENT This research project was supported by a research grant from the Innovational and Technology Fund (ITF) to Jimmy Yu and P.K. Wong, and a Direct Grant from the Research Committee of The Chinese University of Hong Kong to P.K. Wong. Critical review of this manuscript by Prof. CK Wong is greatly appreciated. REFERENCES Amézaga-Madrid P., Nevárez-Moorillón G.V., Orrantia-Borunda E. and Miki-Yoshida M. (2002) Photoinduced bactericidal activity against Pseudomonas aeruginosa by TiO 2 based thin films. FEMS Microbiology Letters 211, 183-188. Butterfield I.M., Christensen P.A., Curtis T.P. and Gunlazuardi J. (1997) Water disinfection using an immobilized titanium dioxide film in a photochemical reactor with electric field enhancement. Water Research 31, 675-677. Chen Y.X., Sun Z.S., Yang Y. and Ke Q. (2001) Heterogeneous photocatalytic oxidation of polyvinyl alcohol in water. Journal of Photochemistry and Photobiology A: Chemistry 142, 85-89. Dunlop P.S.M., Byrne J.A., Manga N. and Eggins B.R. (2002) The photocatalytic removal of bacterial pollutants from drinking water. Journal of Photochemistry and Photobiology A: Chemistry 148, 355-363. Huang Q.D. and Hong C.S. (2000) TiO 2 photocatalytic degradation of PCBs in soil-water systems containing fluoro surfactant. Chemosphere 41: 871-879. Keleher J., Bashant J., Heldt N., Johnson L. and Li Y.Z. (2002) Photocatalytic preparation of silver- coated TiO 2 particles for antibacterial applications. World Journal of Microbiology and Biotechnology 18, 133-139. Lee S., Nishida K., Otaki M. and Ohgaki S. (1997) Photocatalytic inactivation of phage Qβ by immobilized titanium dioxide mediated photocatalyst. Water Science and Technology 35 (11- 12), 101-106. Maness P.C., Smolinski S., Blake D.M., Huang Z., Wolfrum E.J. and Jacoby W.A. (1999) Bactericidal activity of photocatalytic TiO 2 reaction: toward an understanding of its killing mechanism. Applied and Environmental Microbiology 65, 4094-4098. Matsunaga T., Tomoda R., Nakajima T., Nakamura N. and Komine T. (1988) Continuous- sterilization system that uses photosemicoductor powders. Applied and Environmental Microbiology 54, 1330-1333. Pedahzur R., Lev O., Fattal B. and Shuval H.I. (1995) The interaction of silver ions and hydrogen peroxide in the inactivation of E. coli: A preliminary evaluation of a new long acting residual drinking water disinfectant. Water Science and Technology 31 (5-6), 123-129. Saito T., Iwase T., Horie J. and Morioka T. (1992) Mode of photocatalytic bactericidal action of powdered semiconductor TiO 2 on mutans streptococci. Journal of Photochemistry and Photobiology B: Biology 4, 369-379. San N., Hatipoğlu A., Koçtϋrk G. and Çinar Z. (2001) Prediction of primary intermediates and the photodegradation kinetics of 3-aminophenol in aqueous TiO 2 suspensions. Journal of Photochemistry and Photobiology A: Chemistry 139, 225-232. Journal of Water and Environment Technology, Vol.3, No.1, 2005 - 54 - Semikina A.L. and Skulachev V.P. (1990) Submicromolar silver increases passive sodium permeability and inhibits the respiration-supported formation of sodium gradient in Bacillus FTU vesicles. FEBS Letter 269, 69-72. Vidal A., Díaz A.I., El Hraiki A., Romero M., Muguruza I., Senhaji F. and Gonzalez J. (1999) Solar photocatalysis for detoxification and disinfection of contaminated water: pilot plant studies. Catalysis Today 54, 283-290. Watts R.J., Kong S., Orr M.P., Miller G.C. and Henry B.E. (1995) Photocatalytic inactivation of coliform bacteria and viruses in secondary wastewater effluent. Water Research 29, 95-100. Wolfrum E.J., Huang J., Blake D.M., Maness P.C., Huang Z. and Fiest J. (2002) Photocatalytic oxidation of bacteria, bacterial and fungal spores, and model biofilm components to carbon dioxide on titanium dioxide-coated surfaces. Environmental Science and Technology 36, 3412-3419. Yu J.C., Tang H.Y., Yu J.G., Chan H.C., Zhang L.Z., Xie Y.D., Wang H. and Wong S.P. (2002) Bactericidal and photocatalytic activities of TiO 2 thin films prepared by sol-gel and reverse micelle methods. Journal of Photochemistry and Photobiology A: Chemistry 153, 211-219. Zhang L.F., Kanki T., Sano N. and Toyoda A. (2001) Photocatalytic degradation of organic compounds in aqueous solution by a TiO 2 -coated rotating-drum reactor using solar light. Solar Energy 70, 331-337. . uses photosemicoductor powders. Applied and Environmental Microbiology 54, 133 0- 133 3. Pedahzur R., Lev O., Fattal B. and Shuval H.I. (1995) The interaction. a TiO 2 -coated rotating-drum reactor using solar light. Solar Energy 70, 33 1 -33 7.