Effects of Electron Transfer between TiO2 Films and Conducting

An electron transfer is indicated to occur in the interfaces between TiO2 films and conducting substrate Al or ITO, which results in an Ohm contact or...
0 downloads 0 Views 366KB Size
Download PDF
13470

J. Phys. Chem. B 2006, 110, 13470-13476

Effects of Electron Transfer between TiO2 Films and Conducting Substrates on the Photocatalytic Oxidation of Organic Pollutants Wenxin Dai, Xuxu Wang, Ping Liu, Yiming Xu, Guangshe Li, and Xianzhi Fu* Research Institute of Photocatalysis, Chemistry and Chemical Engineering College, Fuzhou UniVersity, Fuzhou 350002, P. R. China ReceiVed: March 10, 2006; In Final Form: May 12, 2006

TiO2 films on Al alloy (Al), indium-tin oxide glass (ITO/glass), and glass were prepared by a dip-coating method. ITO is found to have a higher work function, while the work function for Al is lower than that of TiO2 films. An electron transfer is indicated to occur in the interfaces between TiO2 films and conducting substrate Al or ITO, which results in an Ohm contact or Schottky barrier under the transient equilibrium UV radiation conditions. Photocatalytic measurements showed that the TiO2 films on Al have a higher activity for photocatalytic oxidation of C2H4, but the activity for photocatalytic degradation of oleic acid is lower as compared with TiO2 films on glass. Alternatively, TiO2 films on ITO give completely contrary photocatalytic performance to those on Al. These observations could be associated with the electron transfer, in which Al acts as an electron donor and offers electrons to TiO2, allowing photocatalytic oxidation of ethylene to proceed by the photogenerated electrons, while ITO could be an acceptor for the photogenerated electrons, which is beneficial to photocatalytic degradation of oleic acid by the photogenerated holes. This electron-transfer model could be extended to other photocatalytic systems.

Introduction Titanium dioxide has found a wide application in the fields of environmental treatment and functional material due to its strong photocatalytic effect and photoinduced hydrophilicity.1-6 TiO2 films on tiles, glass, stainless steel, and alloy possess some special functions such as deodorization, antibacterial, and selfcleaning, and therefore have been used as the advanced materials for outer walls, building windows, kitchens, offices, and medical facilities.7-13 In general, TiO2 films coated on metal substrates (such as stainless steel) display strikingly higher photocatalytic performance than those coated on inert substrates (such as glass).10 Such effects have been suggested as a result of the chemical interactions between TiO2 and metal substrates formed during high-temperature sample calcinations. However, an electron transfer between conductor and semiconductor is also expected to occur because of the difference in their work functions. Such an electron transfer has been successfully applied in electrolighting (EL),14 optoelectronics apparatus (such as solar cells),15 and electrochromic devices,16 in which the direction and the efficiency of the electron transfer are strongly dependent on the work functions of substrates.17,18 Recently, TiO2 films on various conducting substrates prepared by radio frequency magnetron sputtering deposition methods have been reported, showing that an irradiation can lead to an electron transfer into the substrates, with its efficiency being dominated by the work functions of the substrates.19,20 From this viewpoint, the electron-transfer effect could play key roles in recombination of the electron-hole pairs generated on the irradiated TiO2 and, consequently, in the photocatalytic activity of the TiO2 films. Ma and co-workers have also reported on the enhancement of the photocatalytic degradation of rhodamine B over TiO2 films on ITO.21 However, the relevant mechanism still remains unclear. * Corresponding author. E-mail: [email protected]. Telephone and Fax: +86-591-83738608.

Superoxide radicals (O2•-) that are formed from reduction of oxygen by the photogenerated electrons on TiO2 surface have been proposed as the main reactive species for the gas-phase photooxidation of organic pollutants on TiO2,4,22 whereas in liquid-phase photooxidation, the organic reaction is mainly initiated by hydroxyl radicals that are produced from oxidation of surface hydroxyl groups via the photogenerated holes.4,22-24 Depending on the fates of the photogenerated electrons and holes, TiO2 films on the same conducting substrate can give rise to a distinct photocatalytic activity for the organic oxidation in gas phase or in liquid phase. To understand the mechanism of photocatalytic oxidation of organic pollutants, it is crucial to examine the effects of electron transfer between TiO2 films and substrates. In this work, transparent TiO2 films coated on an Al alloy and an indium-tin oxide-covered glass (ITO/glass) were prepared. The photocatalytic performances of these films were examined by degradation of oleic acid and C2H4 under UV irradiation. Roles of the substrates in the photocatalytic reactions were explored. Two models about the electron transfer between conducting substrate and TiO2 films are proposed for the distinct photoactivities of these films. Experimental Section Preparation of TiO2 Sol. HNO3 (1.1 mL of 68 vol %) was diluted with 150 mL of H2O into which 12.5 mL of titanium tetraisopropoxide (analytic grade from Shanghai Chemistry Reagent Co. Ltd., China) was added slowly under vigorous stirring for hydrolysis. To obtain the clear TiO2 sol, the suspension solution obtained was allowed to stay at 40 °C under stirring for 48 h and was further dialyzed to pH ) 3.5. Pretreatment of Al Alloy, ITO, and Glass. Before coating, the Al alloy sheet (106H24, Al-Mg alloy, 60 mm × 25 mm × 0.2 mm thick) was pretreated as follows. The sheet was soaked in a diluted solution of sodium hydroxide at 50 °C for 3 min and then rinsed in tap water, diluted nitric acid, tap water, and

10.1021/jp061483h CCC: $33.50 © 2006 American Chemical Society Published on Web 06/16/2006

Photocatalytic Oxidation of Organic Pollutants

Figure 1. Schematic illustration of photoconductivity measurements. (a) ITO electrode, (b) TiO2 film, and (c) funds of samples.

deionized water in sequence. The pretreated Al alloy sheet was denoted as Al. The ITO glass plate (South Glass Group Co. Ltd., Shenzhen, China) and microscope slides (Sail brand, China) were pretreated with acetone, ethanol, and deionized water in sequence. These pretreated substrates were dried in ambient air and denoted as ITO/glass and glass, respectively. Preparation of TiO2 Films on Substrates. TiO2 films were prepared by dip-coating of the TiO2 sol onto the substrates in an ambient atmosphere. The withdrawing rate of the films was fixed at 1 cm/min. The coating process was repeated three times, then the obtained TiO2 films were calcined at 450 °C in air at a ramping rate of 2 °C /min for 2 h. The resulting transparent TiO2 films with a thickness of 300 ( 20 nm on Al, ITO/glass, and glass were denoted as TiO2/Al, TiO2/ITO/glass, and TiO2/ glass, respectively. Preparation of Pt-TiO2 and KI-TiO2 Films on Glass. An aqueous solution of potassium iodide (0.5 wt %) or chloroplatinic acid (0.5 wt %) was added dropwise into the TiO2 sol to reach the mass ratio of TiO2/KI ) 1/0.005 or TiO2/Pt ) 1/0.005. The obtained sol was then used for preparation of KITiO2/glass or Pt-TiO2/glass following a similar procedure to those for TiO2/Al and others. The resulting films were further treated with H2 (10-20 mL/min) in a fixed-bed reactor at 350 °C for 2 h at a ramp rate of 5 °C /min. Characterization. The work functions of the samples were measured by ultraviolet photoelectron spectroscopy (UPS). The measurements were performed on an ADES-400 UPS system (VG, England) with a monochromatic He(I) source (21.2 eV) at an ultrahigh vacuum ( TiO2/glass > TiO2/Al. That is, in comparison with TiO2 films coated substrate glass, the TiO2 films coated on ITO/glass had higher activity, while the TiO2 films coated on Al had lower activity of photocatalytic degradation of oleic acid. Controlled experiments on blank substrate (ITO/glass, glass and Al) showed that oleic acid also degraded under similar conditions, which was most likely due to the direct photolysis of oleic acid. It is noted that the rate of such direct photolysis was very slow as compared with the TiO2 photocatalytic degradation. Photocatalytic oxidation of ethylene was also studied. Figure 10 illustrates the survival ratio of ethylene as a function of illumination time over three samples under UV illumination. The oxidation rate of ethylene on TiO2/Al was faster than that on TiO2/glass, but the rate on TiO2/ITO/glass was lower than that on TiO2/glass. This trend in photoactivity was the reverse of that observed for photocatalytic degradation of oleic acid. It should be mentioned that no degradation of ethylene was observed on three pure substrates (ITO/glass, glass, and Al).

Photocatalytic Oxidation of Organic Pollutants

J. Phys. Chem. B, Vol. 110, No. 27, 2006 13473

Figure 5. SEM images of (a) TiO2/Al, (b) TiO2/glass, and (c) TiO2/ITO/glass.

Figure 6. Ti2p high-resolution XPS spectra of (a) TiO2/ITO/glass, (b)TiO2/Al, and (c) TiO2/glass.

Figure 8. FT-IR spectra of oleic acid as a function of illumination time of (a) 0 min, (b) 60 min, (c) 120 min, and (d) 180 min.

Figure 7. UV-Vis diffuse reflection spectra of (a) TiO2/Al, (b) TiO2/ ITO/glass, (c) TiO2/glass, and (d) ITO/glass.

Figure 9. Survival ratio of oleic acid as a function of illumination time over samples: (4) Al, ()) glass, (0) ITO/glass, (2) TiO2/Al, (() TiO2/glass, and (9) TiO2/ITO/glass.

These photocatalytic reactions were also performed on films of TiO2-Pt/glass, TiO2-KI/glass, and TiO2/glass, respectively. As shown in Figure 11, in comparison with TiO2/glass, TiO2Pt/glass enhanced the photocatalytic degradation of oleic acid, but decreased the photocatalytic oxidation of ethylene, which was consistent with the result reported previously by Fu et al.26 TiO2-KI/glass showed a completely contrary photocatalytic degradation to TiO2-Pt/glass. Discussion In comparison with the TiO2 films on glass, TiO2 films on Al had a lower photoactivity for the degradation rate of oleic acid, but a higher photocatalytic oxidation rate of ethylene. This observation was a totally reverse of the results obtained when

using TiO2 films on ITO/glass. Because the TiO2 films coated on different substrates had similar morphology, phase structure, and optical properties, the differences in photoactivity cannot be solely attributed to the structural effect. Although XPS analysis indicated the presence of the elements of substrates in the TiO2 films, no reaction can be expected to take place between TiO2 and substrates. The differences observed in UPS and photoconductivity may confirm the presence of electrontransfer effects. Introduction of Pt and other noble metals into the TiO2 matrix has been used for improving the photocatalytic activity of TiO2 in many organic oxidations.4,22 A Schottky barrier effect at the Pt/TiO2 interface induces electron transportation across the interface to noble metal, which explains the improved photo-

13474 J. Phys. Chem. B, Vol. 110, No. 27, 2006

Figure 10. Survival ratio of ethylene as a function of illumination time over samples: (4) Al, ()) glass, (0) ITO/glass, (2) TiO2/Al, (() TiO2/glass, and (9) TiO2/ITO/glass.

efficiencies.22 However, the Schottky barrier is only formed when the metal or conductor has a higher work function than that of TiO2. Once the metals or conductors have lower work functions than TiO2, the electrons would transfer from the metals or conductors to TiO2, and consequently, an Ohm contact may be formed at the interfaces. Consequently, different electrontransfer effects would exist in the interfaces of TiO2/Al and TiO2/ITO under UV illumination. The changes observed in photoconductivity (Figure 3) are thus attributed to the differences in work functions between Al and ITO. Because ITO has a higher work function than TiO2 (Table 1), electrons would transfer from TiO2 to ITO until a thermodynamic equilibrium is reached by forming a depletion layer (i.e., Schottky barrier) at the TiO2/ITO interface. When TiO2/ ITO is further exposed to UV light, the photogenerated electrons can cause a shift of Fermi level (EF) of TiO2 to form a new quasi-Fermi level (EF*) (see Figure 12). Meanwhile, the previous thermodynamic equilibrium state for electron transfer is destroyed, and consequently, the photogenerated electrons would continuously transfer across the TiO2/ITO interface to ITO, approaching a nonequilibrium illumination condition (also a transient equilibrium).27 In reference to the electron-transfer model under a thermodynamic equilibrium condition, we assumed that a Schottky barrier could also be formed at the TiO2/ITO interface under transient equilibrium illumination (Figure 12). For the TiO2/Al sample, however, because Al has a lower work function than TiO2 (Table 1), the electrons would transfer from Al to TiO2 until an accumulation layer (i.e., Ohm contact) is formed at the TiO2/Al interface. When the film is further exposed to UV light, the electrons would continuously transfer from Al to TiO2 to reach a nonequilibrium illumination, something like the case of TiO2/ITO. Such a formation process for ideal Ohm contact under transient equilibrium is illustrated in Figure 13. However, for the TiO2/ glass sample, there is no electron transfer between TiO2 and glass, which is most likely that glass is an insulator and has a larger work function and a wider band gap than TiO2. This electron-transfer model could explain the results of photoconductivity in Figure 3. For TiO2/ITO/glass, when a positive bias is applied to the ITO substrate, the Schottky barrier height (depletion barrier Φb) of the TiO2/ITO interface lowers, which enhances the photogenerated electron transfer from TiO2 to ITO surface and therefore favors the production of the photovoltage. However, as a passive bias is applied to the ITO substrate, the Schottky barrier heightens, resulting in the circuit opening and the absence of photovoltage. With regard to TiO2/

Dai et al. Al, although an Ohm contact instead of a Schottky barrier is formed at the TiO2/Al interface, the applied bias could also contribute to the accumulation barrier (Φb). Under a positive bias, the accumulation barrier height becomes low, which enhances the free electron transfer from Al to TiO2. However, this transfer direction of free electrons is opposite to that of the photogenerated electrons under a positive bias voltage, which explains the counteracting photovoltage. As a passive bias voltage is applied to the Al substrate, the photogenerated electrons would transfer from TiO2 to other electrode ITO, producing a photovoltage by the transfer of both free electrons and photogenerated electrons. Meanwhile, because of the increase in accumulation barrier height, the free electron transfer from Al to TiO2 becomes more difficult than that of photogenerated electrons when transferring from TiO2 to ITO in the TiO2/ITO/glass, which is believed to be the main reason for the weak photovoltage signals observed for Al. This electron-transfer process would play significant roles in trapping or the recombination of photogenerated electrons and holes because the self-existent state of photogenerated electrons or holes can be modified. Because the extent of electron transfer and its direction are strongly dependent on the work functions of the substrates,19,20 the distinct electron transfer that occurs in the interfaces of TiO2/ITO and TiO2/Al can give rise to different effects on trapping or recombination of photogenerated electrons and holes. For TiO2/ITO/glass, the photogenerated electrons transferring from TiO2 to ITO would increase the number of the photogenerated holes trapped by the surfaces of TiO2 films but decrease the number of the trapped photogenerated electrons. Although the transferred electrons might return to the valence band of TiO2 and further recombine with the photogenerated holes, the lifetime of holes would be sufficiently prolonged. For TiO2/Al, after transferring across Al/ TiO2 interface to the conductor band of TiO2, the free electrons may either act as the photogenerated electrons or recombine with the photogenerated holes in the valence band of TiO2. Therefore, it is reasonable that the number of the trapped photogenerated electrons could increase, while the number of the trapped photogenerated holes can be decreased. Because no electron transfer is expected between TiO2 and glass, the number of photogenerated electrons or holes in TiO2/glass would not change comparatively. Such a change in the number of the trapped photogenerated electrons and holes in TiO2/Al and TiO2/ ITO/glass could give rise to the distinct photocatalytic oxidation of ethylene and photocatalytic degradation of oleic acid. Concerning the photocatalytic degradation of oleic acid at the semiliquid-solid interface on TiO2 films, the reaction mechanism is not available in the literature. Here, it is proposed that the hydroxyl radicals produced by photogenerated hole reaction with surface hydroxyl groups28 can be the main oxidant for the photocatalytic degradation of carboxylic acids at the liquid-solid interface of TiO2 catalysts. When the surface of TiO2 films is covered with oleic acid, the photocatalytic degradation of oleic acid is primarily initiated by direct oxidation of holes or hydroxyl radicals. On the basis of this assumption, that the photoactivity of TiO2/Al is lowered than that of TiO2/ glass for photocatalytic degradation of oleic acid is ascribed to the decreased number of the trapped holes. Conversely, the higher photoactivity of TiO2/ITO/glass than that of TiO2/glass is due to the increased number of the photogenerated holes. Photooxidation of ethylene would be induced by the trapped photogenerated electrons because the reactions of the trapped photogenerated electrons with molecular oxygen can produce the superoxide radicals that could act as the dominant active

Photocatalytic Oxidation of Organic Pollutants

J. Phys. Chem. B, Vol. 110, No. 27, 2006 13475

Figure 11. Effect of the diference of Pt and KI on the photocatalytic decomposition of oleic acid (a) and ethylene (b) on the TiO2 films: (2) TiO2-KI/glass, (() TiO2/glass, and (9) TiO2-Pt/glass.

Figure 12. Schematic of forming an ideal Schottky barrier between ITO and TiO2 under transient equilibrium UV radiation.

Figure 13. Schematic of forming an ideal Ohm contact between Al and TiO2 under transient equilibrium UV radiation.

sites29,30 for the photooxidation process of ethylene at the gassolid interface on TiO2 catalysts. Consequently, the photoactivity of TiO2/Al for the photooxidation of ethylene that is higher than that of TiO2/glass originated from the increased number of the trapped photogenerated electrons on TiO2/Al, whereas the decreased number of the trapped photogenerated electrons accounted for the lower photoactivity of TiO2/ITO/glass. It should also be mentioned that the work functions of substrates may vary with their compositions,31 and the photocatalytic property of TiO2 films would vary accordingly. On the basis of the electron-transfer model, it is probable that the Al substrate of TiO2/Al would be gradually oxidized during UV light illumination because of the continuous electron transfer from Al to TiO2. This would lead to a decrease in the electronic donation ability of Al to TiO2 and an increase in the work function of Al with a significant photocorrosion of the Al surface. This speculation was confirmed when a TiO2/Al surface containing small amounts of water droplets was exposed

to UV light, the smooth surfaces of TiO2/Al became worse at the spots underneath the water drop, whereas the surfaces remained the same for Al, TiO2/ITO/glass, TiO2/glass, and glass under similar conditions. Therefore, not all building materials could be photocatalytically functionalized by coated TiO2 films owing to the potential photocorrosion. However, because most of the other metallic substrates have work functions higher than TiO2, the direction of electron transfer would be completely opposite to that on TiO2/Al. Therefore, no photocorrosion can be expected for other metallic substrates. To further confirm the electron-transfer model and reaction mechanisms, Pt as an electron acceptor and KI as an electron donor22 were deposited onto TiO2 films. The Schottky barrier (or an Ohm contact) existing at the interfaces of TiO2/Pt or TiO2/ I- did give rise to the consistent photocatalytic performances with those when using TiO2/ITO/glass or TiO2/Al (Figure 11). The Schottky barrier presented in the interface of Pt and TiO2 could decrease the recombination of photogenerated electronhole pairs, consequently prolonging the lifetime of the lonely photogenerated electrons or holes trapped on the TiO2 surface and enhancing the photocatalytic activity of TiO2.22 In this regard, Pt deposited on TiO2 would result in enhancement in all photocatalytic reactions. In fact, not all photocatalytic reaction could benefit from introduction of Pt into TiO2 catalysts. For example, the photocatalytic oxidation of CO was not enhanced on Pt/TiO2 (110) as compared with the reaction on pure TiO2.32 However, the Schottky barrier associated with the electrontransfer model predicts that Pt could selectively improve the photocatalytic activity of TiO2, depending on the relevant photocatalytic reactions. That is, Pt is beneficial to the reactions that are mainly induced by photogenerated holes, but is harmful to those induced by photogenerated electrons. So, this electrontransfer model offers a complementary explanation to the Schottky barrier effect previously reported.22 This electrontransfer model may be further extended to other photocatalytic systems: First, it would be possible to estimate the effects of photogenerated holes or electrons on photocatalytic reaction based on the influence of conductive substrate or doping substance on activity of TiO2. Second, the activity of TiO2 catalysts is probably enhanced by offering electrons to TiO2 or accepting electrons from TiO2 according to the known photocatalytic reaction mechanism. Conclusions This work reported on the electron transfer that occurred in the interfaces of semiconductor TiO2 films and conductive

13476 J. Phys. Chem. B, Vol. 110, No. 27, 2006 substrates of ITO or Al. These interfaces were described as a Schottky barrier (or Ohm contact) under transient equilibrium UV radiation conditions. The work function of ITO was higher and that of Al was lower than that of TiO2 films. TiO2 films on Al had a lower activity for photocatalytic degradation of oleic acid but a higher activity for the photocatalytic oxidation of C2H4 as compared with TiO2 films on glass. However, TiO2 films on ITO presented different photocatalytic performances contrary to those of the TiO2 films on Al. Al acted as the electron donor to offer electrons to TiO2 in enhancing the photocatalytic oxidation of ethylene via the photogenerated electrons, whereas ITO was found to be an electron acceptor to draw the photogenerated electrons in improving the photocatalytic degradation of oleic acid via the photogenerated holes. This electron-transfer effect may be extended to other photocatalytic systems. Acknowledgment. This work was financially supported by the National Natural Science Foundation of China (nos. 20373011, 20473017, 20573020, and 20537010), the National Key Basic Research Special Foundation, China (2004CCA07100), Key Foundation of the Education Bureau of Fujian Province, China (JA05176, K04015), and the Natural Science Foundation of Fujian (no. Z0513006). We are also grateful to Prof. Shining Bao of Zhejiang University for the measurements of UPS. References and Notes (1) Negishi, N.; Iyoda, T.; Hashimoto, K.; Fujishima, A. Chem. Lett. 1995, 841. (2) Ohko, Y.; Fujishima, A. J. Phys. Chem. B 1998, 102, 1724. (3) Ohko, Y.; Tryk, D. A.; Hashimoto, K.; Fujishima, A. J. Phys. Chem. B 1998, 102, 2699. (4) Hoffmann, M. R.; Martin, S. T.; Choi, W. Chem. ReV. 1995, 95, 69. (5) Bount, M. C.; Kim, D. H.; Falconer, J. L. EnViron. Sci. Technol. 2001, 35, 2988. (6) Yu, J. G.; Yu, J. C.; Ho, W. K.; Jiang, Z. T. New J. Chem. 2002, 26, 607. (7) Liu, P.; Ling, L.; Lin, H. X.; Fu, X. Z. Chem. J. Chin. UniV. 2000, 21, 462.

Dai et al. (8) Sakai, N.; Fujishima, A.; Watanabe, T.; Hashimoto, K. J. Phys. Chem. 2001, 105, 3023. (9) Hur, J. S.; Koh, Y. Biotechnol. Lett. 2002, 24, 23. (10) Yu, J. C.; Ho, W.; Lin, J.; Yip, H.; Wong, P. K. EnViron. Sci. Technol. 2003, 37, 2296. (11) Ha, H. Y.; Anderson, M. A. J. EnViron. Eng. 1996, 122, 217. (12) Zhu, Y.; Zhang, L.; Wang, L.; Fu, Y.; Cao, L. L. J. Mater. Chem. 2001, 11, 1864. (13) Zhu, Y.; Zhang, L.; Wang, L.; Tan, R. Q.; Cao, L. L. Surf. Interface Anal. 2001, 32, 218. (14) Mo¨nch, W. Surf. Sci. 1994, 299/300, 928. (15) Gra¨tzel, M. Nature 2001, 414, 338. (16) Campus, F.; Bonhote, P.; Gra¨tzel, M.; Heinen, S.; Walder, L. Sol. Energy Mater. Sol. Cells 1999, 56, 281. (17) Huang, C. H.; Li, F. Y.; Huang, Y. Y. Ultrathin Films for Optics and Electronics; Peking University Press: Beijing, 2001. (18) Ishii, H.; Sugiyama, K.; Ito, E.; Seki, K. AdV. Mater. 1999, 11, 605. (19) Anpo, M.; Kitano, M.; Kikuchi, H.; Hosoda, T.; Takeuchi, M.; Matsuoka, M.; Thomas, J. M. 13th International Congress on Catalysis; Paris, 2004, Abstract 2, O4-013. (20) Kikuchi, H.; Kitano, M.; Takeuchi, M.; Matsuoka, M.; Anpo, M.; Kamat, P. V. J. Phys. Chem. B 2006, 110, 5537. (21) Ma, Y.; Qiu, J. B.; Cao, Y. A.; Guan, Z. S. Chemosphere 2001, 44, 1087. (22) Linsebigler, A. L.; Lu, G. Q.; Yates, J. T. Chem. ReV. 1995, 95, 735. (23) Chen, J.; Ollis, D. F.; Rulkins, W. H.; Bruning, H. Water Res. 1999, 33, 1173. (24) Fox, M. A.; Dulay, M. T. Chem. ReV. 1993, 93, 341. (25) Lv, B.; Lv, P.; Shi, S. R.; Zhang, J. H.; Tang, J. X.; Luo, H.; He, B. M.; BAO, S. N. Acta Phys. Sin. 2002, 51, 2644. (26) Fu, X.; Clark, L. A.; Zeltner, W. A. J. Photochem. Photobiol., A 1996, 97, 181. (27) Bisquert, J.; Zaban, A.; Salvador, P. J. Phys. Chem. B 2002, 106, 8774. (28) Carraway, E. R.; Hoffman, A. J.; Hoffmann, M. R. EnViron. Sci. Technol. 1994, 28, 786. (29) Yamazaki, S.; Tanaka, S.; Tsukamoto, H. J. Photochem. Photobiol., A 1999, 121, 55. (30) Ohtani, B.; Ueda, Y.; Nishimoto, S.; Kagiya, T.; Hachisuka, H. J. Chem. Soc., Perkin Trans. 2 1990, 11, 1955. (31) Ishii, R.; Matsumura, K.; Sakai, A.; Sakata, T. Appl. Surf. Sci. 2001, 169-170, 658. (32) Linsebigler, A.; Rusu, C.; Yates, J. T., Jr. J. Am. Chem. Soc. 1996, 118, 5284.