Photocatalytic Degradation of Polystyrene Plastic under Fluorescent

Sep 5, 2003 - more severe. In this paper, solid-phase photocatalytic degradation of polystyrene (PS) plastic, one of the most common commercial plasti...
0 downloads 0 Views 176KB Size
Download PDF
Environ. Sci. Technol. 2003, 37, 4494-4499

Photocatalytic Degradation of Polystyrene Plastic under Fluorescent Light JING SHANG,† MING CHAI,† AND Y O N G F A Z H U * ,†,‡ Department of Chemistry and State Key Laboratory of C1 Chemical Technology, Tsinghua University, Beijing, 100084, P.R. China

Plastic is used widely all over the world, due to the fact that it is low cost, is easily processable, and has lightweight properties. However, the hazard of discarding waste plastic, so-called “white pollution”, is becoming more and more severe. In this paper, solid-phase photocatalytic degradation of polystyrene (PS) plastic, one of the most common commercial plastics, over copper phthalocyanine (CuPc) sensitized TiO2 photocatalyst (TiO2/CuPc) has been investigated under fluorescent light irradiation in the air. UV-vis spectra show that TiO2/CuPc extends its photoresponse range to visible light, contrasting to only UV light absorption of pure TiO2. The PS photodegradation experiments exhibit that higher PS weight loss rate, lower PS average molecular weight, less amount of volatile organic compounds, and more CO2 can be obtained in the system of PS-(TiO2/CuPc), in comparison with the PSTiO2 system. Therefore, PS photodegradation over TiO2/ CuPc composite is more complete and efficient than over pure TiO2, suggesting the potential application of dyesensitized TiO2 catalyst in the thorough photodegradation of PS plastic under fluorescent light. During the photodegradation of PS plastic, the reactive oxygen species generated on TiO2 or TiO2/CuPc particle surfaces play important roles in chain scission. The present study demonstrates that the combination of polymer plastic with dyesensitized TiO2 catalyst in the form of thin film is a practical and useful way to photodegrade plastic contaminants in the sunlight.

1. Introduction Plastic waste such as polystyrene (PS) disposal has been recognized as a worldwide environmental problem. Though various kinds of techniques have been proposed for the conversion of waste PS plastic, it is generally accepted that recycling material is not a long-term solution to the present problem (1). Due to their chemical stability and nonbiodegradability, waste PS plastic is being mainly disposed by incineration, which will release a lot of toxic byproducts. Thermal or catalytical degradation of waste PS plastic into fuel oils has been investigated in recent years (1-4). Unfortunately, this technique needs high temperature and * Corresponding author phone: +86-10-62783586; fax: +86-1062787601; e-mail: [email protected]. † Department of Chemistry. ‡ State Key Laboratory of C1 Chemical Technology. 4494

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 19, 2003

appropriate catalysts to produce hydrocarbons with narrow distribution, which cause high costs (4). Consequently, new technologies have to be developed for the degradation of plastic waste. Heterogeneous photocatalytic oxidation reactions can occur under moderate conditions, such as room temperature, one atmosphere pressure, and molecular oxygen as the only oxidant (5). Recently, photocatalytic techniques have been successfully applied in the disposal of air and wastewater pollutants. Therefore, it is worthwhile to study the solidphase photodegradation of waste plastic under the condition of the atmosphere and sunlight (sunlight is a very important issue, because sun is an unending resource). The composition of plastic and TiO2 particles has been proven to be a new and useful way to decompose solid polymer in open air (6). However, the light absorption region of anatase-typed TiO2 particles (λ e 385 nm) does not fit with the solar spectrum, because the solar energy above 3.0 eV (λ e 410 nm) only makes up less than 5% of the whole sunlight. Therefore, the development of low-band gap photoactive materials, that is, so-called visible light photocatalysts, is strongly urged for solving environmental and energy problems. Alternatively, in view of low cost and feasibility, it is desirable to incorporate TiO2 catalyst with the visible light response via the composition (7, 8). Dye sensitization is considered to be an efficient method to modify the photoresponse properties of TiO2 particles. The dyes used are erythrosine B (9), rose bengal (10), and metal phthalocyanines (Pc) (11-13), etc., among which metal Pc may be an appropriate candidate because of its good chemical stability and high absorption coefficient within the solar spectrum. Up to now, a number of studies regarding dye sensitization of semiconductors have focused on the solar energy conversion (14-17). However, relatively less research was done on the utilization of dye-sensitized photocatalyst in the detoxification of hazardous contaminants. In these studies dye sensitization was carried out in aqueous solutions with solar light as the energy source in order to seek an efficient method for water purification. To our best knowledge, there has been no research done on the solid-phase photodegradation of plastic over dye-sensitized TiO2 catalyst. In view of the environmental conservation and the sustainable development of human society, it is very important to find an ecofriendly disposal of plastic waste where they degrade to carbon dioxide and water under the sunlight irradiation without producing toxic byproducts. In this paper, solid-phase photocatalytic oxidation of PS over TiO2 or TiO2/CuPc photocatalyst under fluorescent light irradiation is investigated for the first time. Considerable degradation of PS over TiO2/CuPc is revealed, while the charge-transfer mechanism in TiO2/CuPc sample is also discussed.

2. Experiments 2.1. Preparation and Characterization of TiO2 and TiO2/ CuPc Photocatalysts. TiO2 nanoparticle powder was prepared by the Sol-Gel method with TiCl4 as the precursor (18). X-ray analysis (Bruker D8 Advance powder diffractometer with Cu KR source) of TiO2 powder confirmed the complete formation of anatase with the mean crystallite size of 13.2 nm, calculated by the Scherrer equation (19). CuPc was purchased from Aldrich (dye content 97%). TiO2/ CuPc composite was prepared as follows: 0.2 g of TiO2 powder was dispersed in 50 mL of ethanol solution, followed by stirring for 30 min at 333 K to obtain a uniform suspension. 10.1021/es0209464 CCC: $25.00

 2003 American Chemical Society Published on Web 09/05/2003

Then 100 mL of CuPc ethanol solution with the concentration of 2.4 × 10-5 mol/L was dropped into TiO2 suspension. The mixed solution was centrifugated, and the precipitate was washed for several times with distilled water and dried at 353 K. Finally, the bluish TiO2/CuPc photocatalyst was obtained with 0.7 wt % CuPc in the composite. UV-vis spectra of TiO2, CuPc, and TiO2/CuPc samples were measured using a UV-365 spectrophotometer. 2.2. Preparation and Characterization of PS-TiO2 and PS-(TiO2/CuPc) Composite Thin Films. One-off plastic appliance with the main component of PS was applied in the experiment. The PS-TiO2 or PS-(TiO2/CuPc) composite film was cast as follows: PS solution was prepared by dissolving 0.5 g of one-off plastic appliance in 10 mL of tetrahydrofuran (THF) under vigorous stirring for 30 min. Under a typical condition, 0.01 g of TiO2 or TiO2/CuPc powder was suspended uniformly in 10 mL of PS solution, with the ratio of TiO2 or TiO2/CuPc to PS 2.0 wt %. Then 10 mL of PS-TiO2 or PS(TiO2/CuPc) solution was spread on a poly(tetrafluoroethylene) plate (15 cm × 15 cm) and dried for 48 h at room temperature. The thickness of PS-TiO2 or PS-(TiO2/CuPc) film was measured to be roughly 35 µm by Scanning Electron Microscope (SEM) (KYKY 2000). The average molecular weight of the PS-TiO2 and PS(TiO2/CuPc) samples was determined by GPC (HP 1100) equipped with a PL mixed C (×2) column and a refractive index detector. Polystyrene standards (provided by polymer laboratory) with known molecular weight from 1000 to 1 000 000 were used for the Mw calibration. For GPC analysis, all the samples were dissolved in THF and then filtered through a 0.2 µm PTFE syringe filter in order to remove TiO2 particles. The surface morphologies of the composite samples were taken by SEM (KYKY 2000). 2.3. The Photocatalytic Oxidation of PS-TiO2 and PS(TiO2/CuPc) Composite Films. PS-TiO2 and PS-(TiO2/CuPc) composite films were irradiated under three 8 W fluorescent lamps with the wavelength range from 310 to 750 nm. The total light intensity is 1.75 mW/cm2 at the distance of 7 cm, while the intensity of ultraviolet light is about 0.05 mW/cm2 (the intensity of the sunlight in the afternoon in August of Beijing City, China, is about 0.65 mW/cm2). A typical area of the composite film was about 60 cm2. Photodegradation reactions were carried out under the ambient air in a lamphousing box (30 cm × 25 cm × 15 cm) where temperature was maintained around 298 K. The reaction systems were sealed when determining the concentrations of volatile organic compounds and CO2 during photocatalytic oxidations of PS-TiO2 and PS-(TiO2/CuPc) samples. The volatile organic compounds were detected by a gas chromatograph (GC) equipped with a flame ionization detector and GDX-403 steel column at 373 K. The volatile organic compounds were identified by fitting the reserve time of the standard samples to the GC results under the same condition. The concentration of volatile organic compounds was expressed as peak area. The total peak area was measured by setting time range beforehand and calculating by software the integral area under the curve within the given time range. CO2 concentration was measured by GC equipped with a thermal conductivity detector and GDX-403 steel column at 343 K.

3. Results and Discussion 3.1. The Optoelectronic Properties of TiO2 and TiO2/CuPc Photocatalysts. Figure 1 shows the UV-vis absorption spectra of TiO2, TiO2/CuPc, and CuPc samples. Obviously, there is no absorption above 400 nm for anatase TiO2, while CuPc shows visible light absorption from 500 to 650 nm. Consequently, TiO2/CuPc composite exhibits a broader absorption range for the solar spectrum than pure TiO2.

FIGURE 1. The UV-vis absorption spectra of TiO2, TiO2/CuPc, and CuPc samples.

FIGURE 2. The charge separation mechanism of TiO2/CuPc sample under visible and UV light irradiation. TiO2 particles can absorb UV light (λ < 387 nm) to create mobile electrons (e-) and holes (h+) in the conduction band (CB) and valence band (VB), respectively (eq 1). Adsorbed oxygen molecules can capture electrons, producing O2-, O, and O- species (eqs 2-4) (20, 21). At the same time, photogenerated holes can be trapped by hydroxyl ions or water adsorbed on the surface, producing hydroxyl radicals, •OH (eqs 5 and 6), which play important roles in photocatalytic reactions (22). hυ

TiO2 98 TiO2(e- + h+)

(1)

O2 ads + e- f O2ads

(2)

O2 ads f 2Oads

(3)

Oads + e- f Oads

(4)

OH- + h+ f •OH

(5)

H2Oads + h+ f •OH + H+

(6)

If the electrons and holes cannot be captured in time, they will recombine with each other within a few nanoseconds, which will reduce the photocatalytic efficiency of TiO2. However, TiO2/CuPc composite reveals a different case, which is shown in Figure 2. If TiO2/CuPc composite is irradiated by visible light with the energy lower than the band gap of TiO2 (3.2 eV) but higher than the band gap of CuPc (1.81 eV), CuPc will be excited with an electron in excited singlet state (S1) and a hole in the ground state (S0). Because the oxidation potential of CuPc (S1) is around -0.63 V vs NHE (normal hydrogen electrode) (13), and the conduction band of TiO2 is around -0.5 V vs NHE (20), the efficient VOL. 37, NO. 19, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4495

FIGURE 3. Weight loss of PS-TiO2 and PS-(TiO2/CuPc) samples at different irradiation times.

TABLE 1. Weight Average Molecular Weight (Mw) and Number Average Molecular Weight (Mn) of PS-(TiO2/CuPc) and PS-TiO2 at Different Irradiation Times irradiation time/h 0 5 10 30 200

PS-TiO2 (g/mol, ×104) Mw Mn

PS-(TiO2/CuPc) (g/mol, ×104) Mw Mn

24.8 24.3 23.9 22.8 22.5

25.9 24.2 23.4 21.6 21.2

8.49 8.26 8.17 7.98 7.91

8.56 7.59 7.26 7.08 6.92

charge separation at the interface between TiO2 and CuPc takes place:

CuPc(S0) + hυ f CuPc*(S1)

(7)

CuPc*(S1) + TiO2 f CuPc+ + TiO2(e-)

(8)

If TiO2 is optically excited, hole injection from the valance band of TiO2 to the ground state of CuPc is thermodynamically permitted:

TiO2 + hυ f TiO2(e- + h+)

(9)

CuPc + TiO2(e- + h+) f CuPc+ + TiO2(e-)

(10)

Therefore, whether CuPc or TiO2 absorbs the light and produces excitons, the dissociation of the excitons into free electrons and holes can occur at the interface between CuPc and TiO2. More importantly, due to the existence of the interface between CuPc and TiO2, separated electrons and holes have little possibility to recombine again, regardless of the existence of the charge-capturing species mentioned above. This ensures higher charge separation efficiency and better photooxidation capacity for the composite. 3.2. Photocatalytic Oxidation of PS-TiO2 and PS-(TiO2/ CuPc). Figure 3 shows the irradiation time dependence of the weight loss of PS-TiO2 and PS-(TiO2/CuPc) samples. It is observed that the weight loss rate for PS-(TiO2/CuPc) is much higher than PS-TiO2, for example, 6.9% weight loss for PS-(TiO2/CuPc) versus 4.1% for PS-TiO2 at 250 h. Table 1 presents weight average molecular weight (Mw) and number average molecular weight (Mn) of PS-(TiO2/CuPc) and PSTiO2 at different irradiation times. It can be seen that both Mw and Mn decrease with irradiation time. The decrease of Mw or Mn for PS-(TiO2/CuPc) is always faster than PS-TiO2. Moreover, both Mw and Mn decrease rapidly at the initial stage of the irradiation, and then the decrease becomes slow 4496

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 19, 2003

with the reaction proceeding. In polymer degradation where chain scission occurs at random, one always can see a rapid initial molecular weight decrease, which slows down as the process continues (23, 24). Figure 4 shows the surface morphologies of PS-TiO2 and PS-(TiO2/CuPc) thin films before and after irradiation. After irradiation in the air for 10 h, PS-TiO2 film exhibits only a little increase in cavity density (the number of cavities per square micron). Moreover, there are no cracks observed on the surface of PS-TiO2 film during the whole irradiation process. Therefore, it can be supposed that the photodegradation of PS mainly happens on the film surface where electrons or holes combine with adsorbed oxygen molecules or hydroxyl ion to produce O2- or •OH, two very important reactive oxygen species for the degradation of PS, as mentioned in eqs 1-6. Apparently, photocatalytic reaction first starts at the interface between PS and exposed TiO2 photocatalyst, which leads to the formation of cavities centered at TiO2 particles. The cavities grow up with the irradiation, extending to the inner film and spreading over the film surface simultaneously. Although some electrons and holes can be produced on TiO2 particles embedded in the film, most of them recombine with each other due to lack of appropriate trappers inside the film. Therefore, embedded TiO2 particles might not participate in the photocatalytic reaction until they are exposed to the air. In this sense, cavity density would not increase much after irradiation. However, the surface morphologies of PS-(TiO2/CuPc) exhibit an obvious difference before and after irradiation, as shown in Figure 4(B). After 5-h irradiation, the cavity density increases from 0.1 to 0.3 µm-2, and the average diameter of cavities increases to 1.0 µm. After 10-h irradiation, some cavities grow up to the diameter of 6.0 µm. Moreover, some cracks emerge on the surface of PS-(TiO2/CuPc) thin film after irradiation, different from the case of PS-TiO2. Based on the above observations, it is reasonable to assume that the photodegardation of PS can happen both on the surface and inside the film simultaneously. In the photocatalytic degradation of PS, not only O2- and •OH play the important roles but also the holes generated in the ground state of CuPc. As mentioned in section 3.1, efficient holes production occurs in the ground state of CuPc under fluorescent light illumination. Although holes in the ground state of CuPc have lower oxidative ability than those in valence band of TiO2 (13), as shown in Figure 2, it is energetically favorable for them to participate in the oxidation of PS polymer:

h+ + -(CH2CHPh)- f -(CH2CHPh)-+

(11)

-(CH2CHPh)-+ + O2- f -(C4 HCHPh)- + HO2• (12) -(CH2CHPh)-+ + O2- f -(CH2C4 Ph)- + HO2• HO2• + HO2• f H2O2 + O2 hυ

H2O2 98 2•OH

(13) (14) (15)

Therefore, embedded TiO2/CuPc particles can generate enough •OH to photodegrade inner PS, causing much cavity density increase and cracks. 3.3. The Formation of Volatile Organic Compounds and CO2 during the Photocatalytic Oxidation of PS-TiO2 and PS-(TiO2/CuPc). Curve e in Figure 5(A) presents the peaks of volatile organic compounds identified by GC. Peak 1 can be attributed to the air. Peaks 2-5 are characteristics of the volatile organic compounds. Ethene, acetaldehyde, formaldehyde, and ethanol standard samples feature curves a-d,

FIGURE 4. SEM images of PS-TiO2 (A) and PS-(TiO2/CuPc) (B) films. (A1): PS-TiO2 film before irradiation; (A2): PS-TiO2 film irradiated for 10 h; (B1): PS-(TiO2/CuPc) film before irradiation; (B2): PS-(TiO2/CuPc) film irradiated for 5 h; (B3): PS-(TiO2/CuPc) film irradiated for 10 h. respectively. It can be seen that the reserve time of curves a-d fit well to peaks 2-5, respectively. Even at different oven temperatures, the coincidence revealed in Figure 5(A) can be reproduced well. Therefore, the volatile organic compounds detected by GC can be attributed to ethene, acetaldehyde, formaldehyde, and ethanol. Figure 5(B) shows irradiation time dependence of the peak-area sum of these volatile organic compounds during the photodegradation of PS-TiO2 and PS-(TiO2/CuPc). Obviously, the total amount of volatile organic compounds released from the photodegradation of PS-(TiO2/CuPc) is always less than that from PS-TiO2 at a given irradiation time. Also, both of the two curves show the maximum amount at the point of about 40-h irradiation, revealing the relationship between the generation and photodegradation of the volatile organic compounds. At the initial stage of the irradiation, the generation of volatile organic compounds is dominant, so the total amount increases with the irradiation time. As the reaction proceeds, the generation of volatile organic compounds becomes slower and slower, while the photodegradation of these volatile organic compounds becomes faster and faster. As a result, the photodegradation

of the volatile organic compounds becomes predominant after 40-h irradiation, and the total amount of these intermediates starts to decrease. The formation of carbon dioxide and volatile organic compounds during the photocatalytic degradation results in weight loss and average molecular weight decrease. Figure 5(C) shows the irradiation time dependence of carbon dioxide concentration. It can be seen that the CO2 concentration increases with the irradiation times in these two cases, and the amount of CO2 released from PS-(TiO2/CuPc) is more than PS-TiO2. Moreover, these two plots show that CO2 generation is fast in the early stage and then decelerates gradually in the subsequent stage. The initial rapid CO2 generation can be ascribed to the reaction of reactive oxygen species with adjacent polymer chains. As TiO2 particlessurrounding polymer matrix is etched away, the further photooxidation has to depend on the slow desorption process of photooxidants, resulting in a slow CO2 formation. 3.4. Solid-Phase Photocatalytic Oxidation Mechanism of PS Plastic under Fluorescent Light. It has been indicated (25-28) that PS degradation under ultraviolet irradiation may be initiated directly by photons attacking PS to create excited VOL. 37, NO. 19, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4497

acetaldehyde, and ethanol (eq 25). These intermediates can be further oxidized to CO2 and H2O (eq 26) (30, 31).

-(CH2CHPh)- + •OH f -(CH2C4 Ph)- + H2O (16) -(CH2CHPh)- + •OH f -(C4 HCHPh)- + H2O

(17)

-(CH2C4 Ph)- + O2 f -(CH2C(OO4 )Ph)-

(18)

-(C4 HCHPh)- + O2 f -(CH(O4 O)CHPh)-

(19)

-(CH2C(OO4 )Ph)- + -(CH2CHPh)- f -(CH2C(OOH)Ph)- + -(CH2C4 Ph)- (20) -(CH(OO4 )CHPh)- + -(CH2CHPh)- f -(CH(OOH)CHPh)- + -(CH2C4 Ph)- (21) -(CH2C(OOH)Ph)- f -(CH2COPh)- + •OH (22) -(CH(OOH)CHPh)- f -(CHOCHPh)- + •OH (23) -(CH2COPh)- (or -(CHOCHPh)-) + O2 photocatalyst, hυ

98 intermediates such as PhCH2CHO, PhCHO, and PhCOOH (24) -(CH2CHPh)-, PhCH2CHO, PhCHO, and PhCOOH photocatalyst, hυ

98 phenyl ring detaching to form ethene, formaldehyde, acetaldehyde, and ethanol (25) photocatalyst, hυ

98 CO2 + H2O

(26)

As illustrated above, PS-(TiO2/CuPc) shows better photocatalytic degradation than PS-TiO2. The more reactive oxygen species generated, the faster plastic photodegradation goes on. Since the higher charge separation efficiency of TiO2/ CuPc photocatalyst results in the more reactive oxygen species generation both on film surface and inside thin film, PS presents a faster and more complete mineralization over TiO2/CuPc than over TiO2 photocatalyst. The present research put forward an eco-friendly technique to degrade plastic waste under the sunlight irradiation with little formation of toxic byproducts. Polymer-dye sensitized TiO2 composite gives a potential and promising way to solve “white pollution” for the plastic industry and environmental purification.

FIGURE 5. (A) Curves a-d represent GC results of ethene, acetaldehyde, formaldehyde, and ethanol standard samples, respectively; curve e is GC result of volatile organic compounds released from the photodegradation; (B) the peak-area sum of the volatile organic compounds for PS-TiO2 and PS-(TiO2/CuPc) at different irradiation times; and (C) CO2 concentrations at different irradiation times over PS-TiO2 and PS-(TiO2/CuPc).

states, followed by chain scission, branching cross-linking, and oxidation. It must be noted that under fluorescent light irradiation that contains little ultraviolet light, the photolysis of PS plastic might be neglected. During the photocatalytic degradation, PS is initiated by the reactive oxygen species such as •OH attacking neighboring polymer chains (eqs 16 and 17). Once introduced in polymer chain, the carboncentered radicals incorporate with oxygen molecules, leading to the chain cleavage and producing new reactive radicals (eqs 18-24) (25, 27). Phenyl ring can also be detached during the photoreaction (29), accompanied by the formation of volatile organic compounds such as ethene, formaldehyde, 4498

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 19, 2003

Acknowledgments This work was financed by the Chinese National Science Foundation (20071021), Excellent Young Teacher Program of MOE, P.R. China.

Literature Cited (1) Lee, S. Y.; Yoon, J. H.; Kim, J. R.; Park, D. W. Polym. Degrad. Stab. 2001, 74, 297. (2) Lee, S. Y.; Yoon, J. H.; Kim, J. R.; Park, D. W. J. Anal. Appl. Pyrol. 2002, 64, 71. (3) Faravelli, T.; Pinciroli, M.; Pisano, F.; Bozzano, G.; Dente, M.; Ranzi, E. J. Anal. Appl. Pyrol. 2001, 60, 103. (4) Dong, D.; Tasaka, S.; Inagaki, N. Polym. Degrad. Stab. 2001, 72, 345. (5) Peral, J.; Domenech, X.; Ollis, D. F. J. Chem. Technol. Biotechnol. 1997, 70, 117. (6) Cho, S.; Choi, W. J. Photochem. Photobiol. A 2001, 143, 221. (7) Linsebigler, A. L.; Lu, G. Q.; Yates, J. T. Chem. Rev. 1995, 95, 735. (8) Iwasaki, M.; Hara, M.; Kawada, H.; Tada, H.; Ito, S. J. Colloid Interface Sci. 2000, 224, 202. (9) Kamat, P. V.; Fox, M. A. Chem. Phys. Lett. 1983, 102, 379. (10) Ross, H.; Bendig, J.; Hecht, S. Sol. Energy Mater. Sol. Cells 1994, 33, 475.

(11) Rophael, M. W.; Khalil, L. B. Vacuum 1990, 41, 143. (12) Wohrle, D.; Schneider, G.; Stark, J.; Schulz-Ekolff, G. J. Mol. Catal. 1992, 75, L39. (13) Giraudeau, A.; Fan, F. F.; Bard, A. J. J. Am. Chem. Soc. 1980, 102, 5137. (14) Wang, H.; Chen, D. W.; Xu, G. Z.; Zhang, X. F.; Xu, H. J. Chin. Sci. Bull. 1994, 39, 424. (15) O’Regan, B.; Gratzel, M. Nature 1991, 353, 737. (16) Falaras, P.; Gratzel, M. J. Electrochem. Soc. 1993, 140, L92. (17) Kay, A.; Gratzel, M. J. Phys. Chem. 1993, 97, 6272. (18) Zhu, Y. F.; Zhang, L.; Yao, W. Q.; Cao, L. L. Appl. Surf. Sci. 2000, 158, 32. (19) Ogawa, H.; Abe, A. J. Electrochem. Soc. 1981, 128, 685. (20) Park, D. R.; Zhang, J. L.; Ikeue, K.; Yamashita, H.; Anpo, M. J. Catal. 1999, 185, 114. (21) Peral, J.; Ollis, D. F. J. Catal. 1992, 136, 554. (22) Zhu, Y. J.; Olson, N.; Beebe, T. P. Environ. Sci. Technol. 2001, 35, 3113. (23) Vallint, G.; Cortia, A.; Pera, A.; Solaro, R.; Cioni, F.; Chiellini, E. J. Gen. Appl. Microbiol. 1994, 40, 445.

(24) Imam, S. H.; Chen, L.; Gordon, S. H.; Shogren, R. L.; Weisleder, D.; Greene, R. V. J. Environ. Polym. Degrad. 1998, 6, 91. (25) Crawford, K. D.; Hughes, K. D. J. Phys. Chem. B 1997, 101, 864. (26) Torikai, A.; Kobatake, T.; Okisaki, F. J. Appl. Polym. Sci. 1998, 67, 1293. (27) Kuzina, S. I.; Mikhailov, A. I. Eur. Polym. J. 1998, 34, 291. (28) Kaczmark, H.; Kaminska, A.; Swiatek, M.; Sanyal, S. Eur. Polym. J. 2000, 36, 1167. (29) Wells, R. K.; Royston, A.; Badyal, J. P. S. Macromolecules 1994, 27, 7465. (30) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. Rev. 1995, 95, 69. (31) Augugliaro, V.; Voluccia, S.; Loddo, V.; Marchese, L.; Martra, G.; Palmisano, L.; Schiavello, M. Appl. Catal. B 1999, 20, 15.

Received for review September 19, 2002. Revised manuscript received May 19, 2003. Accepted July 14, 2003. ES0209464

VOL. 37, NO. 19, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4499