Role of Hydroxyl Radicals and Mechanism of Escherichia coli

Ecology and Environmental Engineering (MOE) and State Key Laboratory of Fine ... of Western Australia, 35 Stirling Highway, Crawley, Western Austr...
0 downloads 0 Views 3MB Size
Article pubs.acs.org/est

Role of Hydroxyl Radicals and Mechanism of Escherichia coli Inactivation on Ag/AgBr/TiO2 Nanotube Array Electrode under Visible Light Irradiation Yang Hou,† Xinyong Li,*,†,‡ Qidong Zhao,† Guohua Chen,*,‡ and Colin L. Raston§ †

Key Laboratory of Industrial Ecology and Environmental Engineering (MOE) and State Key Laboratory of Fine Chemical, School of Environmental Science and Technology, Dalian University of Technology, Dalian 116024, China ‡ Department of Chemical and Biomolecular Engineering, The Hong Kong University of Science & Technology, Clear Water Bay, Kowloon, Hong Kong § Centre for Strategic Nano-Fabrication, School of Biomedical, Biomolecular and Chemical Sciences, University of Western Australia, 35 Stirling Highway, Crawley, Western Australia 6009, Australia S Supporting Information *

ABSTRACT: A ternary Ag/AgBr/TiO2 nanotube array electrode with enhanced visible-light activity was synthesized by a two-step approach including electrochemical process of anodization and an in situ photoassisted deposition strategy. The dramatically enhanced photoelectrocatalytic activity of the composite electrode was evaluated via the inactivation of Escherichia coli under visible light irradiation (λ>420 nm), whose performance of complete sterilization was much superior to other reference photocatalysts. PL, ESR, and radicals trapping studies revealed hydroxyl radicals were involved as the main active oxygen species in the photoelectrocatalytic reaction. The process of the damage of the cell wall and the cell membrane was directly observed by ESEM, TEM, and FTIR, as well as further confirmed by determination of potassium ion leakage from the killed bacteria. The present results pointed to oxidative attack from the exterior to the interior of the Escherichia coli by OH•, O2•−, holes and Br0, causing the cell to die as the primary mechanism of photoelectrocatalytic inactivation.



INTRODUCTION A growing number of countries around the world have irrigation and drinking water supply problems. Many water sources are not only polluted by hazardous chemicals but also by pathogenic microorganisms and, therefore, have to be disinfected before use.1 Traditional water disinfection methods such as chlorination and ozonation have shown disadvantages related to the formation of potentially hazardous disinfection byproduct (DBPs) with carcinogenic and mutagenic potential.2 Therefore, the development of more effective disinfection technologies has become an urgent issue. In 1985, Matsunaga et al. discovered for the first time the bactericidal activity of TiO2 as a photocatalyst.3 Since then, numerous studies related to the bactericidal effect of TiO2 photocatalyst have been conducted to inactivate bacteria, viruses, and cancer cells.4,5 However, two drawbacks hinder the practical application of TiO2 for photocatalytic inactivation of bacteria: its relatively low efficiency of light utilization due to its large band gap and poor quantum efficiency.6 To address these problems, considerable efforts have been recently taken involving doping with nonmetal anions,7 metallization,8 and combination with narrow-band gap semiconductors.9 However, doped TiO2 always has dissatisfactory quantum efficiency because dopant © 2012 American Chemical Society

usually can act as a recombination center for the photogenerated electrons and holes.10 Recently, coupling TiO2 with narrow-band gap semiconductor and metal composite has been observed to be effective in enhancing the visible-light activity and improving the charge separation efficiency simultaneously, which has attracted a tremendous amount of attention.11,12 Ag/AgBr composite is well-known as a photosensitive material and regarded as a promising candidate of highly efficient and stable visible-light photocatalyst.13 Hu et al. synthesized visible-light-active Ag/AgBr/TiO2 nanoparticles by a deposition-precipitation process and demonstrated their enhanced photocatalytic activity in destroying bacteria.14 Elahifard et al. reported that Ag/AgBr/TiO2-covered apatite had a high ability for adsorbing bacteria in the dark and exhibited a significantly improved antibacterial activity under visible light.15 However, irregular photoinduced charge transfer behavior in the powder-form catalysts limits the promotion in the photoconversion efficiency. Besides, the drawback that it is Received: Revised: Accepted: Published: 4042

November 15, 2011 January 28, 2012 March 3, 2012 March 4, 2012 dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

difficult to separate and recycle the composite powder from the reaction system further hinders their practical application. The combination of TiO2 nanotube arrays (TiO2−NA) and Ag/ AgBr nanoparticles as a recyclable electrode may provide an ideal system to overcome the above problems, because the oriented heterojunction structures with visible-light activity would possess the superior capability of enhancing the energy conversion efficiency. To date, no work on the bactericidal capability of the Ag/AgBr/TiO2 nanotube arrays (Ag/AgBr/ TiO2−NA) has been reported to our best knowledge. In this work, we report the synthesis of Ag/AgBr nanoparticles decorated TiO2−NA with enhanced visible-light activity by a two-step approach, including electrochemical anodization technique followed by in situ photoassisted deposition strategy. The dramatically improved photoelectrocatalytic (PEC) activity and excellent photostability of the Ag/ AgBr/TiO2 −NA were evaluated in the inactivation of Escherichia coli (E. coli). The role of hydroxyl radicals and mechanism of PEC inactivation of E. coli on the Ag/AgBr/ TiO2−NA electrode under visible light irradiation was studied in detail.

Figure 1. (a) Top surface view and (b) cross-section view of TiO2− NA. (c) Top surface view and (d) high-magnification top surface view of Ag/AgBr/TiO2−NA.



has a close-packed structure with an average nanotube diameter of about 90 nm and the tube length around 550 nm. After the photoassisted deposition process, Ag/AgBr nanoparticles with a diameter of approximate 20 nm are not only distributed on the inner but also outer walls of the TiO2−NA (Figure 1c-d), and no obvious aggregates of Ag/AgBr are observed at the entrances of TiO2−NA. The nanotubular structure of TiO2 maintained its integrity without significant morphological changes. The XRD results show that both anatase TiO2 and cubic AgBr are present in the fresh Ag/AgBr/TiO2−NA (Figure S1). No diffraction peaks assigned to metallic Ag are observed, probably due to its low content and high dispersity. As compared with fresh Ag/AgBr/TiO2−NA, the XRD pattern of used Ag/AgBr/TiO2−NA after the inactivation of E. coli exhibits similar XRD pattern to fresh Ag/AgBr/TiO2−NA, indicating the stable phase structure of AgBr. This observation also suggests that the metallic Ag should exist in the composite system because it can effectively inhibit the decomposition of AgBr. Due to the surface plasmon resonance (SPR) of metallic Ag, the photogenerated electrons remained in the Ag nanoparticles without being transferred to the Ag+ ions of the AgBr lattice, avoiding the photocorrosion of AgBr.17 To investigate the surface chemical states of Ag, the high-resolution XPS spectrum of the Ag 3d region is provided in Figure S2. Two typical peaks of Ag 3d located at about 374.1 and 367.9 eV can be attributed to the Ag 3d5/2 and Ag 3d3/2 binding energies, respectively. These two peaks could be further deconvoluted into several peaks at 373.7, 374.4 eV and 367.6, 368.4 eV, respectively, where those at 374.4 and 368.4 eV are ascribed to the metallic Ag0 and those at 373.7 and 367.6 eV are ascribed to Ag+ of Ag/AgBr/TiO2−NA. Similar results are also reported by other researchers.18,19 The calculated surface mole ratio of the metallic Ag0 to Ag+ is about 1:4.1. Therefore, the combination of XRD and XPS analysis confirms that there are both Ag and AgBr species in the Ag/AgBr/TiO2−NA system. Additionally, PL, DRS, and photocurrent investigations verify that the loading of Ag/AgBr nanoparticles not only enhances the visible-light absorption of TiO2−NA but also improves the separation efficiency of the photogenerated electron−hole pairs. The photoinduced electrons from metallic Ag and AgBr

EXPERIMENTAL SECTION Preparation of Ag/AgBr/TiO2−NA. The highly ordered TiO2−NA was synthesized by anodic oxidation in a HF electrolyte, similar to that described previously.16 Ag/AgBr nanoparticles were then deposited into the TiO2−NA by an in situ photoassisted deposition strategy. Typically, the TiO2−NA was ultrasonicated for 30 min in 100 mL of distilled water. Then, 1.2 g of cetylmethylammonium bromide was added to the suspension, and the mixture was stirred magnetically for another 30 min. After that, 2 mL of 0.6 mol L −1 diamminesilver(I) hydroxide (Ag(NH3)2OH) was quickly added to the mixture with the aid of irradiation provided by a 500 W xenon lamp, and the resulting suspensions were stirred at room temperature for 5 h. In this process, under the alkaline and photoassisted condition, cationic surfactant cetylmethylammonium bromide could adsorb onto the surface of TiO2 to limit the number of nucleation sites for an Ag/AgBr island to grow, achieving a uniform deposition of Ag/AgBr nanopsrticles into the TiO2−NA. The deposited sample was then placed on a hot plate at 70 °C for 1 h. After three cycles and heat treatment, the prepared sample was further calcined in air at 500 °C for 2 h. Characterization. The characterization of the Ag/AgBr/ TiO2−NA included environmental scanning electron microscopy (ESEM), transmission electron microscopy (TEM), X-ray diffraction (XRD), UV−vis diffuse reflectance spectroscopy (DRS), X-ray photoelectron spectroscopy (XPS), Electron spin resonance (ESR) and Photoluminescence (PL). Detailed information can be found in the Supporting Information. Antibacterial Activity Tests. Details about E. coli incubation, PEC inactivation, E. coli colonies count, total organic carbon (TOC), Fourier transform infrared (FTIR) experiment for investigating the inactivation process and the monitoring of K+ leakage from the inactive E. coli and eluted Ag+ from the Ag/AgBr/TiO2−NA electrode are provided in the Supporting Information.



RESULTS AND DISCUSSION Characterization of Photocatalysts. Figure 1a-b shows a typical ESEM image of the highly ordered TiO2−NA, which 4043

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

Ag+, was eluted from Ag/AgBr/TiO2−NA electrode, and there was no more Ag+ leakage even after 2 h of PEC reaction. More importantly, no obvious bactericidal effect was observed in the presence of 0.3 mg L−1 Ag+ even after 2 h. These facts suggested that photocatalyst itself and the eluted Ag+ were not toxic to E. coli (Figure 2b). As shown in Figure 2a, direct photolysis of E. coli in 80 min with visible light alone was almost absent except for oscillation increases, which may be attributed to the E. coli growing with the help of the visible light irradiation.20 In the electrochemical process, it exhibited low bactericidal activity, and only 0.83 logs of E. coli was inactivated after 80 min, affirming that the bias potential of 0.6 V was too low to induce E. coli inactivation. However, compared to direct photolysis, the Ag/AgBr/TiO2− NA electrode (without a bias potential) inactivated 47.3% of E. coli during the same time, implying that the Ag/AgBr/TiO2− NA electrode was highly effective at the killing of E. coli under visible light irradiation. Clearly, the E. coli with the initial concentration of 1.2 × 107 colony forming units per milliliter (cfu mL−1) was completely killed within 80 min of irradiation, when the bias potential of 0.6 V was applied in photoelectrocatalysis process. Moreover, the viable colonies on the plates reduced significantly as the irradiation times increased in the photoelectrocatalysis process, and after 80 min irradiation, a complete inactivation of E. coli could also be observed in Figures S6−S7. Additionally, the inactivation efficiency of E. coli in the photoelectrocatalysis process (100%) was even higher than the sum of electrochemical process (11.7%) and photocatalysis (47.3%), indicating the presence of a significant synergic effect between electrochemical process and photocatalysis, which is consistent with the results of our previous work.16 Figure 2c compares the PEC antibacterial activity of E. coli for different samples under visible light irradiation. It is clear that Ag/AgBr/TiO2−NA electrode exhibited the highest PEC activity for inactivation of E. coli and an almost complete inactivation of E. coli was observed after 80 min irradiation, which is higher than those of the N-doped TiO2−NA electrode (2.77 logs removal of E. coli) and TiO2−NA electrode (1.49 logs removal of E. coli) by a factor of 2.56 and 4.72. For comparison, a porous TiO2 electrode was prepared according to Baram et al.,21 but a very low inactivation efficiency of E. coli was obtained, which was even slower than that of TiO2−NA electrode. In the present work, only 0.51 logs E. coli was killed by porous TiO2 electrode after 80 min reaction because of the limitation of poor visible-light absorption capability of porous TiO2 electrode. Moreover, Ag/AgBr/TiO2 and apatite-covered Ag/AgBr/TiO2 electrodes were also prepared according to previous studies to make further comparison.14,15 With the Ag/AgBr/TiO2 electrode, about 3.33 logs E. coli was killed after irradiation for 80 min. However, a higher PEC antibacterial activity of 5.39 logs reduction of E. coli was observed for the apatite-covered Ag/AgBr/TiO2 electrode due to the enhanced adsorption capacity of the Ag/ AgBr/TiO2 electrode after the introduction of apatite, which was in good agreement with the result reported by Elahifard et al.15 Interestingly, in spite of that, the bactericidal activity of the apatite-covered Ag/AgBr/TiO2 electrode was still lower than that of the Ag/AgBr/TiO2−NA electrode. The maximum antibacterial activity of the Ag/AgBr/TiO2−NA electrode was about 1.29 and 2.08 times higher than that of the apatitecovered Ag/AgBr/TiO2 electrode and the Ag/AgBr/TiO2 electrode under the same experiment conditions, respectively.

nanoparticles are easily transferred to the Ti substrate via the walls of TiO2 nanotubes (Figure S3−5). PEC Inactivation of E. coli. Figure 2a compared the inactivation of E. coli on the Ag/AgBr/TiO2−NA electrode

Figure 2. (a) Inactivation of E. coli by direct photolysis, electrochemical process, photocatalysis, and photoelectrocatalysis, respectively, with the Ag/AgBr/TiO2−NA electrode under visible light irradiation. (b) Ag+ leakage from Ag/AgBr/TiO2−NA electrode vs time plotted in the PEC inactivation of E. coli process and inactivation of E. coli vs time plotted in aqueous dispersion containing 0.3 mg L−1 Ag+. (c) Inactivation of E. coli by PEC technology with porous TiO2, TiO2−NA, N/TiO2−NA, Ag/AgBr/TiO2, apatite-covered Ag/AgBr/ TiO2, and Ag/AgBr/TiO2−NA electrodes under visible light irradiation (λ > 420 nm, I0 = 25.3 mW cm−2, 0.6 V vs SCE).

between different killing processes, including direct photolysis, electrochemical process, photocatalysis, and photoelectrocatalysis. As a comparison, the dark control and inactivation effect of Ag+ leakage from the Ag/AgBr/TiO2−NA electrode were carried out, and corresponding results showed no bactericidal effect. Clearly, only a little amount of Ag+, about 0.27 mg L−1 4044

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

Figure 3. (a) DMPO spin-trapping ESR spectra recorded at ambient temperature in aqueous dispersion: a, TiO2−NA and b, Ag/AgBr/TiO2−NA under visible light irradiation (λ = 532 nm, 10 Hz, the signals were recorded after visible light irradiation for 40 s). (b) DMPO spin-trapping ESR spectra recorded after different periods of irradiation with a pulse laser (λ = 532 nm, 10 Hz) for the Ag/AgBr/TiO2−NA. (c) PL spectra of TiO2− NA and Ag/AgBr/TiO2−NA in a 5 × 10−4 M basic solution of terephthalic acid under visible light irradiation at a fixed 20 min. (d) PL spectral change with irradiation time on Ag/AgBr/TiO2−NA in a 5 × 10−4 M basic solution of terephthalic acid. (e) Plots of photogenerated carriers trapping for PEC inactivation of E. coli over Ag/AgBr/TiO2−NA.

To the best of our knowledge, the Ag/AgBr/TiO2−NA electrode prepared herein has the highest performance among the Ag/AgBr decorated TiO2 systems reported so far, even though many other factors including the length of the TiO2 tubes and the mass per unit area of the deposited Ag/AgBr were not optimized in this work. The high bactericidal activity of the Ag/AgBr/TiO2−NA electrode in this work is mainly attributed to its large specific surface area (nanotubes) and the modification of Ag/AgBr. Meanwhile, the special geometry of nanotube arrays also provides a direct path for electron transport toward Ti substrate, promoting photogenerated electron−hole pairs separation. Recently, Kang et al. achieved an almost complete inactivation of E. coli within 60 min using CdS/Pt-TiO2 nanotube electrode, a Xe lamp source (74 mW cm−2), and an applied bias of 0.6 V for bacterial solution with a concentration of 5 × 108 cfu mL−1.22 Comparatively, although we just used a visible-light source with a lower light intensity in the present work, the inactivation efficiency of E. coli over the Ag/AgBr/TiO2−NA electrode was still close to that of the previous CdS/Pt-TiO2 nanotube electrode, implying that the

Ag/AgBr/TiO2−NA electrode may have a higher PEC activity for E. coli inactivation. Moreover, compared to the Ag/AgBr/ TiO2−NA electrode, the CdS/Pt-TiO2 system suffered from the drawbacks of higher cost (from Pt), high level of CdS coverage and instability (CdS photocorrosion), which limited its practical application. An in-depth understanding on the role of active radicals during bacterial inactivation is essential to devise and apply a strategy to kill efficiently a wide range of microorganisms practically. To understand the role of active oxygen species involved in the PEC process, the ESR spin-trap technique (with DMPO) was employed to characterize the photogenerated reactive oxygen species over Ag/AgBr/TiO2−NA under visible light irradiation. The characteristic four peaks of DMPO−OH• with intensity 1:2:2:1 were clearly observed in the ESR signal (Figure 3a), which was similar to the spectra reported by other research for the OH• adduct,23 elucidating that OH• was really generated on the Ag/AgBr/TiO2−NA under visible light irradiation. In contrast, no OH• signals were detected in the TiO2−NA systems under the otherwise identical conditions. In 4045

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

Figure 4. ESEM images of E. coli (a) untreated and (b) after PEC inactivation treatment with Ag/AgBr/TiO2−NA electrode for 80 min. TEM images of E. coli (c) untreated and after PEC inactivation treatment with Ag/AgBr/TiO2−NA electrode for (d) 40 min and (e) 80 min.

addition, the change of signal intensity of OH• (Ag/AgBr/ TiO2−NA) at different reaction times under visible light irradiation was shown in Figure 3b. No ESR signals were observed when the reaction was performed in dark. However, the intensity of signal further strengthened with increasing irradiation time and became stable in 60 s; therefore, the intensity of the OH• adduct peak produced in 80 s irradiation was nearly similar to those in 60 s irradiation. We also employed the PL technique using terephthalic acid as a probe to further investigate the generation of OH•.24 Figure 3c shows the comparison of PL intensity for TiO2−NA and Ag/AgBr/TiO2−NA electrode in the 5 × 10−4 M terephthalic acid solution with a concentration of 2 × 10−3 M NaOH at 20 min. No PL signal was observed upon irradiation of the TiO2−NA electrode because it was not activated by visible light, indicating no production of OH• radicals. However, a gradual increase in PL intensity at about

425 nm was observed with increasing irradiation time for Ag/ AgBr/TiO2−NA electrode, which implied the production of OH• radicals (Figure 3d). Usually, the PL intensity was proportional to the amount of produced OH• radicals. The above results showed that the OH• radicals were the main active oxygen species in the photoelectrochemical process of the E. coli/Ag/AgBr/TiO2−NA system, which was in agreement with that revealed by the ESR analysis. The radicals and holes trapping experiments were designed to further elucidate the PEC inactivation of E. coli process over Ag/AgBr/TiO2−NA. As shown in Figure 3e, the PEC inactivation of E. coli was slightly retarded with about 16.6% reduction by the injection of a scavenger for holes (EDTA),25 which implied the minor role of holes either acting as the oxidizing agent or the origination of the •OH radicals in this process. With the addition of tBuOH,26 an efficient OH• radicals quencher, the inactivation efficiency of E. coli was 4046

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

depressed markedly and only 51.1% of the E. coli was killed after 80 min irradiation, indicating OH• radicals was the primary oxidant in the bactericidal reaction. In the presence of p-benzoquinone (super oxide (O2•−) radicals scavenger),27 a moderate suppression about 67.3% removal of E. coli was observed within 80 min. These results further confirmed that both OH• and O2•−radicals were the primary active species in this system, while holes were also involved. Evidence of E. coli Damage. To better understand the mechanism of E. coli inactivation by the reactive species generated by Ag/AgBr/TiO2−NA electrode, the morphology and microstructure of E. coli before and after PEC inactivation treatments was examined by ESEM and TEM studies (Figure 4). Figure 4a shows a representative ESEM image of E. coli before PEC inactivation treatment. The untreated E. coli exhibited damage-free and well-preserved cell walls, indicating that the cells were healthy before they were treated with the Ag/AgBr/TiO2−NA electrode. However, after complete inactivation of the E. coli by the PEC treatment, the morphology of cell wall showed obvious damages. Many E. coli lost parts of the cell wall and the cell membrane or even material inside, so that deep ‘holes’ appeared (Figure 4b). These results showed that treated cells with Ag/AgBr/TiO2− NA electrode were damaged, forming pits and holes in their cell walls. While in a previous study by Wu et al., the formation of pits and holes in E. coli cells has been proposed.28 Figure 4c-e shows the TEM images of E. coli at different reaction stages of PEC treatment. The features of live E. coli were a well-defined cell wall and an evenly colored interior of the cells, which was corresponding to the presence of proteins, enzymes, and DNA (Figure 4c).29 Meanwhile, the lipopolysaccharide layer of the outer membrane played an important role in providing a barrier of selective permeability for the live E. coli.30 After 40 min PEC treatment, the cell wall of E. coli was decomposed, and an electron translucent region appeared at the central of the cells, indicating that the integrity of the cell has been damaged, leading to a leakage of the interior component (Figure 4d). Obviously, the destruction of the E. coli cells was more serious and the whole bacteria cells were fragmented after 80 min irradiation (Figure 4e), leading to the ultimate cell death. Based on the ESEM and TEM investigations, the active species generated from the visible light-excited Ag/AgBr/TiO2−NA electrode could destroy the cell wall and membrane of the E. coli, leading to the leakage of the intracellular substances. K+ existed virtually in bacteria and played a role in the regulation of polysome content and protein synthesis, quickly leaked from the inactivated bacteria because of the permeability change of membrane, resulting in the loss of cell viability.31 The leakage of K+ from inactivated E. coli in different killing processes was measured by a Perkin-Elmer Analyst 700 atomic absorbance spectrometer, and the result was shown in Figure 5a. There was nearly no significant leakage of K+ from E. coli in the direct photolysis (0.23 ppm) and electrochemical process (0.98 ppm), indicating that the control experiments had no bactericidal effect or low bactericidal activity. However, in the photocatalysis process, K+ immediately leaked out from E. coli, and the leakage concentration gradually increased in parallel with reaction time, reaching a stable value of 3.28 ppm at the end of reaction, which was much higher than those in the control experiments. Interestingly, much higher concentration of K+ was released and increased to about 8.23 ppm with the inactivation of E. coli during the photoelectrocatalysis process. These results implied that the cell membrane permeability had

Figure 5. (a) K+ leakage from E. coli under different conditions: direct photolysis, electrochemical process, photocatalysis, and photoelectrocatalysis. (b-c) Changes of FTIR spectra of E. coli during the PEC inactivation process. (b) Bands in the spectral region of 4000−2500 cm−1 and (c) bands in the spectral region of 2000−1000 cm−1.

been disrupted and some intracellular substances had been leaked from the inactivation of E. coli via the PEC reaction, which agrees with the above results of ESEM and TEM analysis. Figure 5b-c shows the FTIR spectra of E. coli during the PEC treatment in the presence of the Ag/AgBr/TiO2−NA electrode. In the region 4000−2500 cm−1 (Figure 5b), the characteristic peaks at 3297 and 3066 cm−1 were assigned to amide A and amide B, respectively.30 The νa(CH3) vibration bands were shown at 2960 cm−1, νa(CH2) at 2923 cm−1, νs(CH3) at 2870 cm−1, and νs(CH2) at 2851 cm−1. With increasing the irradiation time, these characteristic peaks of FTIR spectra decayed in intensity or even disappeared. Figure 5c shows the FTIR spectra of E. coli in the region of 2000−1000 cm−1. The amide I bands at 1656 cm−1, arising principally from ν(CO) stretching vibrations, were observed, and the characteristic 4047

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

peaks at 1543 cm−1 were assigned to amide II primarily due to N−H bending with contributions from the C−N stretching vibrations of the peptide group. The bands around at 1237 cm−1 arose from the nucleic acids bands overlapping with the asymmetric stretching mode νas(PO2−) of the phospholipid phospho-diester.32 In addition, the peak of 1083 cm−1 well matched the assignment for the vibrations of the sugar rings of lipopolysaccharides. Significant decrease of the amide I band and amide II band was detected, while the decay of the PO2− band and sugar rings was observed as a function of time. After 80 min, the new peaks appeared at 1730 and 1334 cm−1, which might be due to the formation of the α, β unsaturated aldehydes during the breakdown of hydroperoxides or lipid endoperoxides.33 These results also indicated that the cell wall and membrane were destroyed by the attack of the reactive species, resulting in leakages of the carboxylic acid and then cell death. These vibrational mode assignments were summarized in Table S1. More details about both monitoring of intermediates and effect of DBPs for PEC inactivation of E. coli can be found in the Supporting Information. PEC Inactivation Mechanism Discussion. Based on the previous reports22,34 and the above experimental results, the mechanisms of bactericidal reaction over the Ag/AgBr/TiO2− NA system are proposed as follows (Scheme S1): Under visible light irradiation (λ>420 nm), the photogenerated electron− hole pairs came from both plasmon-excited Ag nanoparticles and photoexcited AgBr semiconductor. Due to the dipolar character of the SPR of Ag nanoparticles,17 the photogenerated electrons from the plasmon-excited Ag nanoparticles (1) transferred across the interface of the Ag/TiO2 to the surface of the TiO2 nanotubes farthest away from the Ag/AgBr interface (electron transfer I: Ag→TiO2) (2). Although the detailed mechanism of such transferal is still unclear at present, the electrons injection from Ag to TiO2 on SPR excitation has been proven to occur in Ag/TiO2 and Au/TiO2 systems.35,36 One reasonable explanation suggested that electrons oscillating collectively on the SPR excitation might lead to interband excitation, giving enough energy to electrons moving to the TiO2 interface via overcoming the Schottky barrier on Ag/ TiO2.37 Simultaneously, the leftover holes diffused into the surface of the AgBr particles and caused the oxidation of Br− ions to Br0 atoms because the surface of AgBr particles was negatively charged and most likely terminated by Br− ions due to the xenon lamp-induced reduction of partial Ag+ ions (3).38 On the other hand, AgBr could be excited by visible light and generated electron−hole pairs (4). Metallic Ag species on the surface of the Ag/AgBr/TiO2−NA mainly acted as electron traps, enhancing the electrons and holes separation (electron transfer II: AgBr→Ag) (5) and the subsequent transfer of the trapped electrons to the TiO2 conduction band (electron transfer III: (AgBr→Ag→TiO2) (6). Thus, simultaneous electron transfer I, II, and III (vectorial electron transfer of AgBr→Ag→TiO2) should occur as the result of visible-light excitation of both Ag and AgBr. Some of the photogenerated holes left in the AgBr valence band could combine with Br− to form Br0 atoms. As Br0 atoms were reactive radical species, they were able to kill E. coli and become reduced to Br− again, according to the previous report (7).39 Of course, the detailed roles of Br0 radical species requires further clarification in future work. Whereas the other holes accumulated in AgBr could directly react with E. coli or interact with surface-bound H2O or OH− to produce the OH• radical species (8, 9), which was an extremely strong oxidant for

inactivation of E. coli. Meanwhile, the photogenerated electrons at the surface of the TiO2 could travel along the TiO2 nanotubes, passed through the interface between TiO2 and Ti to the external circuit under the external electric field (10). Furthermore, photoelectrons arrived at the counter electrode surface could react with the adsorbed molecular oxygen to yield O2•− (11). The generated O2•− then further combine with H+ to produce HO2• (12),40 which could react with the trapped electrons to generate OH• radicals (13).41 These theoretical speculations are validated by the obtained experimental results on the active radicals. The reactive species, such as Br0, OH•, O2•− and holes, could attack the cell membrane and wall, disrupt membrane integrity, or destroy the molecules in the cell surface, which brought massive cell kills and lyses (14). Based on the above analysis, the relevant reactions at the composite electrode surface can be expressed as follows Ag + hν → Ag *

(1)

Ag * + TiO2 → Ag +• + TiO2(e)

(2)

Ag +• + Br − → Ag + Br 0

(3)

AgBr + hν → AgBr(e + h) → AgBr + heat

(4)

AgBr(e) + Ag → AgBr + Ag (e)

(5)

Ag (e) + TiO2 → TiO2(e) + Ag

(6)

AgBr(h) + Br − → AgBr + Br 0

(7)

AgBr(h) + H2O → H+ + OH •

(8)

AgBr(h) + OH − → OH •

(9)

TiO2(e) + external electrostatic field → external circuit (10)

Counter electrode(e) + O2 → O2•−

(11)

O2•− + H+ → HO2•

(12)

2(e) + HO2• + H+ → OH • + OH −

(13)

Br 0 , OH •, O2•−, h + E. coli → inactivated E. coli + Br − (14)

Stability of Ag/AgBr/TiO2−NA Electrode. The stability of the Ag/AgBr/TiO2−NA electrode was investigated by repeatedly PEC inactivation of E. coli experiments (Figure S8). The results clearly showed that the Ag/AgBr/TiO2−NA electrode did not exhibit any significant loss of activity even after four cycles of repeated experiments. The corresponding bacterial test results for E. coli were also shown in Figure S9, as photographs. In summary, the Ag/AgBr/TiO2−NA electrode exhibited great potential for PEC inactivation of bacteria. Remarkable improvement of bactericidal capability for the Ag/AgBr/TiO2− NA electrode benefited from enhanced visible-light harvesting and reduced recombination of photogenerated electron−hole pairs due to the synergistic effect of Ag/AgBr nanoparticles and TiO2 nanotubes. The inactivation mechanism of E. coli was suggested to be initial oxidative lesions to the cell wall and the cell membrane by the attack of the OH•, O2•−, holes and Br0 reactive species, resulting in leakages of the interior carboxylic 4048

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

E. C.; Falaras, P. Photo-induced treatment of breast epithelial cancer cells using nanostructured titanium dioxide solution. J. Photochem. Photobiol., A 2010, 214 (2−3), 215−223. (6) Palombari, R.; Ranchella, M.; Rol, C.; Sebastiani, G. V. Oxidative photoelectrochemical technology with Ti/TiO2 anodes. Sol. Energy Mater. Sol. Cells 2002, 71 (3), 359−368. (7) Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Visiblelight photocatalysis in nitrogen-doped titanium oxides. Science 2001, 293 (5528), 269−271. (8) Chen, X.; Zheng, Z. F.; Ke, X. B.; Jaatinen, E.; Xie, T. F.; Wang, D. J.; Guo, C.; Zhao, J. C.; Zhu, H. Y. Supported silver nanoparticles as photocatalysts under ultraviolet and visible light irradiation. Green Chem. 2010, 12 (3), 414−419. (9) Hou, Y.; Li, X. Y.; Zhao, Q. D.; Quan, X.; Chen, G. H. Electrochemically assisted photocatalytic degradation of 4-chlorophenol by ZnFe2O4-modified TiO2 nanotube array electrode under visible light irradiation. Environ. Sci. Technol. 2010, 44 (13), 5098−5103. (10) Chen, X.; Mao, S. S. Titanium dioxide nanomaterials: Synthesis, properties, modifications, and applications. Chem. Rev. 2007, 107 (7), 2891−2959. (11) Tada, H.; Mitsui, T.; Kiyonaga, T.; Akita, T.; Tanaka, K. Allsolid-state Z-scheme in CdS-Au-TiO2 three-component nanojunction system. Nat. Mater. 2006, 5 (10), 782−786. (12) Daskalaki, V. M.; Antoniadou, M.; Puma, G. L.; Kondarides, D. I.; Lianos, P. Solar light-responsive Pt/CdS/TiO2 photocatalysts for hydrogen production and simultaneous degradation of inorganic or organic sacrificial agents in wastewater. Environ. Sci. Technol. 2010, 44 (19), 7200−7205. (13) Zhang, L. S.; Wong, K. H.; Yip, H. Y.; Hu, C.; Yu, J. C.; Chan, C. Y.; Wong, P. K. Effective photocatalytic disinfection of E. coli K-12 using AgBr-Ag-Bi2WO6 nanojunction system irradiated by visible light: The role of diffusing hydroxyl radicals. Environ. Sci. Technol. 2010, 44 (4), 1392−1398. (14) Hu, C.; Lan, Y. Q.; Qu, J. H.; Hu, X. X.; Wang, A. M. Ag/AgBr/ TiO2 visible light photocatalyst for destruction of azodyes and bacteria. J. Phys. Chem. B 2006, 110 (9), 4066−4072. (15) Elahifard, M. R.; Rahimnejad, S.; Haghighi, S.; Gholami, M. R. Apatite-coated Ag/AgBr/TiO2 visible-light photocatalyst for destruction of bacteria. J. Am. Chem. Soc. 2007, 129 (31), 9552−9553. (16) Hou, Y.; Li, X. Y.; Zou, X. J.; Quan, X.; Chen, G. C. Photoeletrocatalytic activity of a Cu2O-loaded self-organized highly oriented TiO2 nanotube array electrode for 4-chlorophenol degradation. Environ. Sci. Technol. 2009, 43 (3), 858−863. (17) Zhang, Y. H.; Tang, Z. R.; Fu, X. Z.; Xu, Y. J. Nanocomposite of Ag-AgBr-TiO2 as a photoactive and durable catalyst for degradation of volatile organic compounds in the gas phase. Appl. Catal., B 2011, 106 (3−4), 445−452. (18) Cheng, H. F.; Huang, B. B.; Wang, P.; Wang, Z. Y; Lou, Z. Z.; Wang, J. P.; Qin, X. Y.; Zhang, X. Y.; Dai, Y. In situ ion exchange synthesis of the novel Ag/AgBr/BiOBr hybrid with highly efficient decontamination of pollutants. Chem. Commun. 2011, 47, 7054−7056. (19) Tian, G. H.; Chen, Y. J.; Bao, H. L.; Meng, X. Y.; Pan, K.; Zhou, W.; Tian, C. G.; Wang, J. Q.; Fu, H. G. Controlled synthesis of thorny anatase TiO2 tubes for construction of Ag-AgBr/TiO2 composites as highly efficient simulated solar-light photocatalyst. J. Mater. Chem. 2012, 22, 2081−2088. (20) Yu, H. T.; Quan, X.; Zhang, Y. N.; Ma, N.; Chen, S.; Zhao, H. M. Electrochemically assisted photocatalytic inactivation of Escherichia coli under visible light using a ZnIn2S4 film electrode. Langmuir 2008, 24 (14), 7599−7604. (21) Baram, N.; Starosvetsky, D.; Starosvetsky, J.; Epshtein, M.; Armon, R.; Ein-Eli, Y. Enhanced photo-efficiency of immobilized TiO2 catalyst via intense anodic bias. Electrochem. Commun. 2007, 9 (7), 1684−1688. (22) Kang, Q.; Lu, Q. Z.; Liu, S. H.; Yang, L. X.; Wen, L. F.; Luo, S. L.; Cai, Q. Y. A ternary hybrid CdS/Pt-TiO2 nanotube structure for photoelectrocatalytic bactericidal effects on Escherichia coli. Biomaterials 2010, 31 (12), 3317−3326.

acid and potassium ion due to the severe morphological and structural injuries, eventually causing the cell to die.



ASSOCIATED CONTENT

S Supporting Information *

Auxiliary information on experimental procedures, characterization of samples, photoelectrochemical measurements, antibacterial activity tests, and preparation of E. coli for ESEM and TEM; XRD pattern (Figure S1); XPS spectrum and TEM images (Figure S2); PL spectra (Figure S3); UV−vis diffuse reflectance spectra (Figure S4); Short-circuit photocurrent density vs time (Figure S5); Images of E. coli colonies (Figure S6) and photographs of E. coli (Figure S7); Stability (Figure S8 and Figure S9); TOC removal (Figure S10); PEC degradation of oxamic acids, oxalic acids, and DBPs on the Ag/ AgBr/TiO2−NA electrode under visible light irradiation (Figure S11−13); PEC inactivation of E. coli as affected by different light intensity (Figure S14 and Figure S21); PEC inactivation of E. coli with the Ag/AgBr/TiO2−NA electrode (Figure S15), Images of E. coli colonies (Figure S16), photographs of E. coli (Figure S17), TEM images (Figure S18) and changes of FTIR spectra (Figure S19) of E. coli under simulated sunlight irradiation; PEC inactivation of E. coli with different samples under simulated sunlight irradiation (Figure S20); PEC mechanism graph (Scheme S1); Assignment of the FTIR bands (Table S1). This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*Phone: +86-411-8470-7733 (X.Y.L.); +852-2358-7138 (G.H.C.). Fax: +86-411-8470-7733 (X.Y.L.); +852-2358-0054 (G.H.C.). E-mail: [email protected] (X.Y.L.); [email protected] (G.H.C.). Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported financially by the National Nature Science Foundation of China (No. 20877013, No. 20837001, NSFC-RGC 21061160495), the National High Technology Research and Development Program of China (863 Program) (No. 2007AA061402), and the Major State Basic Research Development Program of China (973 Program) (No. 2007CB613306).



REFERENCES

(1) Shannon, M. A.; Bohn, P. W.; Elimelech, M.; Georgiadis, J. G.; Marinas, B. J.; Mayes, A. M. Science and technology for water purification in the coming decades. Nature 2008, 452 (7185), 301− 310. (2) Muellner, M. G.; Wagner, E. D.; McCalla, K.; Richardson, S. D.; Woo, Y. T.; Plewa, M. J. Haloacetonitriles vs. regulated haloacetic acids: Are nitrogen-containing DBPs more toxic? Environ. Sci. Technol. 2007, 41 (2), 645−651. (3) Matsunaga, T.; Tomoda, R.; Nakajima, T.; Wake, H. Photoelectrochemical sterilization of microbial-cells by semiconductor powders. FEMS Microbiol. Lett. 1985, 29 (1−2), 211−214. (4) Shen, X. C.; Zhang, Z. L.; Zhou, B.; Peng, J.; Xie, M.; Zhang, M.; Pang, D. W. Visible light-induced plasmid DNA damage catalyzed by a CdSe/ZnS-photosensitized nano-TiO2 film. Environ. Sci. Technol. 2008, 42 (14), 5049−5054. (5) Lagopati, N.; Kitsiou, P. V.; Kontos, A. I.; Venieratos, P.; Kotsopoulou, E.; Kontos, A. G.; Dionysiou, D. D.; Pispas, S.; Tsilibary, 4049

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050

Environmental Science & Technology

Article

(23) Ai, Z. H.; Yang, P.; Lu, X. H. Degradation of 4-chlorophenol by microwave irradiation enhanced advanced oxidation processes. Chemosphere 2005, 60 (6), 824−827. (24) Yu, J. G.; Wang, W. G.; Cheng, B.; Su, B. L. Enhancement of photocatalytic activity of mesporous TiO2 powders by hydrothermal surface fluorination treatment. J. Phys. Chem. C 2009, 113 (16), 6743− 6750. (25) Pan, C. S.; Zhu, Y. F. New type of BiPO4 oxy-acid salt photocatalyst with high photocatalytic activity on degradation of dye. Environ. Sci. Technol. 2010, 44 (14), 5570−5574. (26) Xu, T. L.; Cai, Y.; O’Shea, K. E. Adsorption and photocatalyzed oxidation of methylated arsenic species in TiO2 suspensions. Environ. Sci. Technol. 2007, 41 (15), 5471−5477. (27) Yin, M. C.; Li, Z. S.; Kou, J. H.; Zou, Z. G. Mechanism investigation of visible light-induced degradation in a heterogeneous TiO2/Eosin Y/Rhodamine B system. Environ. Sci. Technol. 2009, 43 (21), 8361−836. (28) Wu, P. G.; Imlay, J. A.; Shang, J. K. Mechanism of Escherichia coli inactivation on palladium-modified nitrogen-doped titanium dioxide. Biomaterials 2010, 31 (29), 7526−7533. (29) Wu, Y.; Yao, S. J.; Guan, Y. X. Inactivation of microorganisms in carbon dioxide at elevated pressure and ambient temperature. Ind. Eng. Chem. Res. 2007, 46 (19), 6345−6352. (30) Hu, C.; Guo, J.; Qu, J. H.; Hu, X. X. Photocatalytic degradation of pathogenic bacteria with AgI/TiO2 under visible light irradiation. Langmuir 2007, 23 (9), 4982−4987. (31) Ren, J.; Wang, W. Z.; Zhang, L.; Chang, J.; Hu, S. Photocatalytic inactivation of bacteria by photocatalyst Bi2WO6 under visible light. Catal. Commun. 2009, 10 (14), 1940−1943. (32) Kiwi, J.; Nadtochenko, V. Evidence for the mechanism of photocatalytic degradation of the bacterial wall membrane at the TiO2 interface by ATR-FTIR and laser kinetic spectroscopy. Langmuir 2005, 21 (10), 4631−4641. (33) Gericke, A.; Huhnerfuss, H. Investigation of Z-unsaturated and E-unsaturated fatty-acids, fatty-acid esters, and fatty alcohols at the airwater-interface by infrared-spectroscopy. Langmuir 1995, 11 (1), 225− 230. (34) An, C. H.; Peng, S. N.; Sun, Y. G. Facile synthesis of sunlightdriven AgCl:Ag plasmonic nanophotocatalyst. Adv. Mater. 2010, 22 (23), 2570−2574. (35) Wen, Y. Y.; Ding, H. M.; Shan, Y. K. Preparation and visible light photocatalytic activity of Ag/TiO2/graphene nanocomposite. Nanoscale 2011, 3, 4411−4417. (36) Tian, Y.; Tatsuma, T. Mechanisms and applications of plasmoninduced charge separation at TiO2 films loaded with gold nanoparticles. J. Am. Chem. Soc. 2005, 127, 7632−7637. (37) Awazu, K.; Fujimaki, M.; Rockstuhl, C.; Tominaga, J.; Murakami, H.; Ohki, Y.; Yoshida, N.; Watanabe, T. A plasmonic photocatalyst consisting of silver nanoparticles embedded in titanium dioxide. J. Am. Chem. Soc. 2008, 130, 1676−1680. (38) Wang, P.; Huang, B. B.; Zhang, X. Y.; Qin, X. Y.; Jin, H.; Dai, Y.; Wang, Z. Y.; Wei, J. Y.; Zhan, J.; Wang, S. Y.; Wang, J. P.; Whangbo, M. H. Highly efficient visible-light plasmonic photocatalyst Ag@AgBr. Chem.Eur. J. 2009, 15 (8), 1821−1824. (39) Wang, P.; Huang, B. B.; Qin, Xi. Y.; Zhang, X. Y.; Dai, Y.; Whangbo, M. H. Ag/AgBr/WO3·3H2O: Visible-light photocatalyst for bacteria destruction. Inorg. Chem. 2009, 48 (22), 10697−10702. (40) Iliev, V.; Tomova, D.; Bilyarska, L.; Prahov, L.; Petrov, L. Phthalocyanine modified TiO2 or WO3-catalysts for photooxidation of sulfide and thiosulfate ions upon irradiation with visible light. J. Photochem. Photobiol., A 2003, 159 (3), 281−287. (41) Zheng, S.; Cai, Y.; O’Shea, K. E. TiO2 photocatalytic degradation of phenylarsonic acid. J. Photochem. Photobiol., A 2010, 210 (1), 61−68.

4050

dx.doi.org/10.1021/es204079d | Environ. Sci. Technol. 2012, 46, 4042−4050