- Email: [email protected]
Development of surface coating technology of TiO2 powder and improvement of photocatalytic activity by surface modification
Thin Solid Films 475 (2005) 171 – 177 www.elsevier.com/locate/tsf
Development of surface coating technology of TiO2 powder and improvement of photocatalytic activity by surface modification T.K. Kima, M.N. Leea, S.H. Leeb, Y.C. Parkb, C.K. Jungb, J.-H. Boob,* b
a Environmental Research Laboratory, SolarTech. CO. Ltd., Seoul 153-023, Korea Institute of Basic Science and Department of Chemistry, Sungkyunkwan University, Suwon 440-746, Korea
Available online 25 August 2004
Abstract We have synthesized the titanium dioxide (TiO2) powders with the similar structure as a commercially available Degussa P-25 TiO2 powder for TiO2-based photocatalysts. To improve the photocatalytic activity, electronic modification on the TiO2-based photocatalysts by chemical solution deposition (CSD) coating was also carried out with metal oxides such as Fe2O3 and Al2O3. The structural and compositional changes as well as optical characteristics are mainly investigated by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), UV-visible spectroscopy, and ellipsometry measurements. In addition, electron spin resonance (ESR) studies have also been carried out to verify the existence of paramagnetic species such as OH and H2O radicals on UV-irradiated TiO2-based photocatalysts. ESR data showed that the hydroxy radicals could decompose organic pollutants into harmless products because they have high oxidizing power. From the reduction test of nitrobenzene, it is found that the photocatalytic effect of TiO2-based photocatalysts coated with Fe2O3 is twice better than that of commercially available noncoated TiO2 photocatalysts. In the case of photocatalytic oxidation reaction of phenol under UV irradiation, moreover, the experimental results showed a consistency with ESR data indicating that TiO2 coated with metal oxides would be one of the most effective photocatalysts. D 2004 Elsevier B.V. All rights reserved. Keywords: TiO2-based photocatalysts; Surface modification; Chemical solution deposition; Photocatalytic reduction–oxidation reaction; Electron spin resonance analysis
1. Introduction Titanium dioxide (TiO2) has been known as a useful photocatalytic material because it is photosensitive, stable, and inexpensive [1–5]. There is a growing interest in recent years to find new, efficient, and economic methods to clean up the environment of pollution materials. This is inspired by the potential application of TiO2-based photocatalysts for the total destruction of organic compounds in polluted air and wastewater. The bulk material of TiO2 is well known to have three main phases namely rutile, anatase, and brookite [6]. Among them, the TiO2 exists mostly as rutile and anatase
* Corresponding author. Tel.: +82 31 290 7072. fax: +82 31 290 7075. E-mail address: [email protected] (J.-H. Boo). 0040-6090/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2004.07.021
phases and both phases have tetragonal structures. Rutile is a high-temperature stable phase and has an optical energy band gap of 3.0 eV (415 nm), while anatase is formed at a lower temperature with an optical energy band gap of 3.2 eV (380 nm) as well as refractive index of 2.3 [7]. It is well known that generally, the TiO2-based photocatalyst with anatase phase shows more excellent photocatalytic effect than that with rutile phase, and the anatase phase can be transformed into the rutile phase at above 800 8C [8,9]. The role of the holes and electrons at the surface of TiO2 in heterogeneous photocatalysis has been investigated in aqueous suspension for the reduction–oxidation (redox) reaction of several organic compounds [10–14]. Electron spin resonance (ESR) studies have been carried out to verify the existence of paramagnetic species such as d OH and HO2d on UV-irradiation of TiO2 [5,14–19]. However, the photophysical mechanism of surface-modified TiO2 pow-
172
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
ders by metal oxides is not well understood yet. Electronic modification of the photocatalyst by coating of metal oxides such as Fe2O3, Al2O3, MoO2, and MoO3 has a strong effect on the photoreactivity of the system. The efficiency of electron-hole separation and the dynamics of interfacial electron transfer can be dramatically influenced. Various synthetic methods and coating techniques have been developed for development and modification of the TiO2-based photocatalysts including chemical vapor deposition (CVD), metal–organic chemical vapor deposition (MOCVD), and sol–gel methods [20,21]. In this study, therefore, we have synthesized the TiO2 powders with the similar structure as a commercially available Degussa P-25 TiO2-based photocatalysts using both slurry of metatitanic acid and sol–gel method. We have also developed a surface modification technology of TiO2based photocatalysts by chemical solution deposition (CSD) for improving their photocatalytic activity. To investigate the effects of surface modification by CSD and annealing, moreover, we mainly studied the structural and compositional changes of the TiO2 powders as well as optical characteristics with X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), ESR, UV-visible spectroscopy (UV-Vis), and ellipsometry measurements because the electronic modification of the photocatalysts by CSD has a strong effect on the photoreactivity of the TiO2-based photocatalysts.
2. Experimental TiO2 powders were synthesized by a slurry reaction of metatitanic acid [TiO(OH)2] and ethyl alcohol. After ball milling of them for 24 h, the slurry was kept at the temperature of 110 8C for further 24 h in the oven for evaporation of the alcohol, and then annealed at 600 8C for 2 h, resulting in a TiO2 powder formation with 58 m2/g of BET surface area and 87 nm of average particle size, respectively. The synthesized TiO2 powders have the similar structure and surface area as the commercially available Degussa P-25 TiO2 powders (rutile/anatase=3:7; surface area=55 m2/g) [20]. Moreover, a coating sol for TiO2-based photocatalysts was also prepared by the sol–gel method, using titanium tetraisopropoxide, acid (HCl, HNO3, HF, etc.), and water. Titanium tetraisopropoxide was slowly dropped into the 0.4% nitric acid solution after stirring it vigorously for 2 h at room temperature, and then heated at 80 8C for 24 h. During the reaction, isopropanol was removed by distillation and then the milky–bulk reaction solution was gradually changed to blue fine–milky solution. TiO2 coating sol was cooled at room temperature, then colloidal SiO2 and tetraethoxyorthosilicate (TEOS) were added to this solution, stirring for 5 h. To improve their photocatalytic activities, electronic modification on the TiO2-based photocatalysts by CSD was also carried out with metal oxides such as Fe2O3 and Al2O3. In addition, in
order to understand the detailed mechanism of photocatalytic reduction–oxidation reactions, electron spin resonance (ESR) studies have also been carried out.
3. Results and discussion Fig. 1(a) shows the X-ray diffraction pattern of TiO2 powders synthesized by a reaction of titanium tetraisopropoxide, acid (HCl, HNO3, HF, etc.), and water and then annealed at different temperatures in air. In the sample annealed at 300 8C, the powder has in part a nature of anatase crystal structure. With increasing annealing temperature to 600 8C, the crystallinity of the TiO2 powder was increased because the characteristic TiO2(101) diffraction peak with anatase structure becomes stronger than that of 300 8C. However, a new diffraction peaks due to a phase transition from anatase phase to rutile phase also appeared in the TiO2 powder after it was annealed at 1000 8C. By measuring the relative peak intensity of their main diffraction peaks of (b) and (c), the abundance ratio of rutile and anatase phases that were confirmed by XPS (not shown here) can be deducted to be rutile/anatase=5:5. Comparing Fig. 1(a) with Fig. 1(d), that is, XRD pattern of commercially available Degussa P-25 TiO2 powder, one can identify the difference of the abundance ratio with annealing temperature. It is well known that the abundance ratio of Degussa P-25 TiO2 powder, which is now the most excellent photocatalyst in the world, has a rutile to anatase ratio of 3:7. Thus, it is highly desirable for us to make the TiO2 powder with the same abundance ratio as Degussa P25 TiO2 powder as well as similar crystallinity by controlling the annealing temperature, reactive gas, and surface area with near by the same particle size. To make the best photocatalysts, we also did surface modification of the synthesized TiO2 powders by CSD coating with metal oxides such as Fe 2 O 3 (Fe 3+ ) and Al2O3(Al3+) for improving their photocatalytic activity for total destruction of organic compounds in waste solvents. Fig. 1(b) and (c) shows the XRD patterns of TiO2 powders obtained after 0.1% Fe2O3 (b) and 0.1% Al2O3(c) coatings annealed at different temperatures. Below 600 8C annealing temperature, highly oriented TiO2 powders with pure anatase structure could be observed from both Fig. 1(b) and (c). The abundance ratio obtained from the samples was about rutile/ anatase=3:7. At 1000 8C annealing temperature, however, we can see distinct difference of the abundance ratio of rutile to anatase phases after judging the relative main diffraction peaks. Although we obtained the same rutile to anatase ratio as Degussa P-25 TiO2 powder in the case of a TiO2 sample annealed at 600 8C, we cannot directly compare the photocatalytic activity between our synthesized TiO2 powders and commercially available Degussa P-25 TiO2 powder with rutile to anatase ratio only. Therefore, we measured the ESR spectra and tried surface modification of our synthesized TiO2 powders using annealing or chemical solution deposi-
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
173
Fig. 1. X-ray diffraction patterns of (a) pure TiO2, (b) Fe2O3-coated TiO2 and Al2O3-coated TiO2 (c) powders by CSD with annealing at various temperatures. (d) XRD pattern of the commercially available Degussa P-25 TiO2 powder.
tion (CSD). From these results, we made a conclusion that the Fe2O3- or Al2O3-coated TiO2 powders with rutile to anatase ratio of 3:7 have higher photocatalytic activity than that of commercially available P-25 TiO2 powder, suggesting that a surface modification technology by chemical solution deposition (CSD) with metal oxides will improve the photocatalytic activity of the TiO2-based photocatalysts. Titanium dioxide doped with metal ions is found to be one of the most effective photocatalysts. Coating of different metal oxides to the TiO2 is known to influence the charge carrier recombination and interfacial charge transfer rates of photogenerated carriers [2,11,19]. However, the photophysical mechanism of surface-modified TiO2 powders is not well understood yet. In order to elucidate the role of metal ions that are used as dopants, ESR studies on TiO2 samples coated with metal oxides were also carried out in this study. TiO2 samples coated with metal oxides (Al2O3 and Fe2O3) were annealed at the temperatures of 300, 600, and 1000 8C. The samples annealed at 300 and 600 8C existed as mixtures of amorphous and anatase TiO2, while the sample annealed at 1000 8C did as a mixture of anatase and rutile TiO2, respectively. Upon UV-irradiation at 196 8C, ESR spectra of TiO2 samples annealed at 600 8C showed {NTiIV–OHd } radicals with g 1 of 2.016 and g 2 of 2.002 that arise from hole trapping. In addition, another ESR peak with the g value of 1.991 which is attributed to {NTiIII–OH} generating from trapping of excited electrons increased in order of pure TiO2,
Al3+-, and Fe3+-coated TiO2. However, TiO2 samples annealed at 300 8C did not show the EPR peak due to {NTiIII–OH} UV-irradiation. They exhibited ESR peaks with the g values in the range of g of 2.016–2.014 which may be ascribed to O radicals. When 0.1 wt.% Fe3+-coated TiO2 was irradiated at 196 8C, a unique ESR peak with the g value of 2.002 was obtained and its intensity increased after UV-irradiation. This signal may be attributed to O2 that arises from the transfer of an electron to oxygen adsorbed on Ti near-oxygen vacancy. Fig. 2(a) shows a typical ESR spectrum of the Degussa P25 powder obtained after UV irradiation. Four strong peaks are observed at g of 2.023, 2.016, 2.002, and 1.991, respectively. Because the Degussa P-25 powder consists a mixture of anatase and rutile phases, two kinds of O or {NTiIV–OHd } radicals attributed to both phases are observed at g of 2.023 and 2.016, respectively. The peaks at g of 2.002, 1.991, and 1.978 are attributed to O2 , anatase Ti3+{NTiIII– OH}, oxygen deficiency Ti3+{NTiIII–OH}, respectively. Comparing these data with the ESR spectra of Fe2O3-coated TiO2 (Fig. 2(b)) and Al2O3-coated TiO2 (Fig. 2(c)) powders, one can identify two differences. One is that the strong peak ascribed to anatase Ti3+(NTiIII–OH) is due to the Al2O3 coating, reflecting trapping of excited electrons. In addition, the intensities of g of 2.016 and 2.002 peaks also increase due to the formation of many hydroxy {NTiIV–OHd } radicals. The other one is that in the case of Fe2O3-coated TiO2
174
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
we can see a red shift tendency with annealing temperature. From the reflectance data, an optical band gap can be calculated, and the band gaps of the samples decrease with increasing annealing temperature. Reflecting the data of optical band gap between anatase phases (3.2 eV) and rutile phase (3.0 eV), the samples of unannealed TiO2 and annealed TiO2 powders at below 600 8C have relatively higher band gap than those of annealed TiO2 powder at 1000 8C. Comparing the optical band gap for Degussa P-25, the annealed TiO2 powder at 1000 8C will have near by the same optical band gap through the broader range, indicating high photosensitivity due to wide range band gap. Considering the crystallinity, the degree of photosensitivity that can
Fig. 2. (a) ESR spectrum of the Degussa P-25 TiO2 powder obtained after UV irradiation. ESR spectra of the modified TiO2-based photocatalysts with Fe2O3-coated TiO2 (b) and Al2O3-coated TiO2 (c) powders obtained before and after UV irradiations, respectively.
powder, the intensity of g of 1.991 peak largely decreases compared with Al2O3-coated TiO2 powder, due to trapping of excited electrons by oxide coating. This result can be explained by that, because the coated oxides have transition metal ions with vacant d-orbitals, Fe3+ can easily trap the excited electrons rather than Al3+. Fig. 3(a) shows the variation of optical reflectance for the TiO2 powder with different annealing temperatures. Here,
Fig. 3. Variation of optical reflectance (a) and transmittance (b) of the synthesized TiO2 powders and TiO2 coating layers obtained at various annealing temperatures with different thickness. (c) Relationship between refractive index and coating layer thickness.
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
absorb the light is affected by the crystallinity of TiO2 powder. The better the crystallinity with increasing annealing temperature, the high the photoreactivity will be turned out. The reason of having broader band gap in the case of a Degussa P-25 TiO2 powder is thus due to the crystal nature of anatase and rutile mixture which can delay charge carrier recombination time. Fig. 3(b) shows the changes of optical transmittance obtained from TiO2 coating layers on glass substrates with different thickness. Although Fig. 3(b) shows that the transmittance increases with increasing film thickness, the arising point (i.e., initial region with a linear slope) of transmittance decreases with increasing film thickness, suggesting an optical band gap shift to low energy side. This means that with decreasing film thickness, the initial arising point (wavelength) of transmittance is shifted from the UV region (380 nm) to the visible region (about 400 nm). Because TiO thin films with visible range optical transmittance have shown relatively high photocatalytic effect rather than those with UV range optical transmittance [1,22], it is highly desirable for us to make TiO thin films with an optical band gap less than 4.25 nm (above 380 nm). Therefore, to satisfy these conditions, our data suggest a very thin film with below 100-nm thickness. The observed rising point of transmittance is 300 for 231-nm thickness, 350 for 140-nm thickness, and 380 for 98-nm thickness, corresponding to 4.50, 4.34, and 4.25 eV of optical band
175
gaps, respectively. This change indicates that there is a critical thickness to maximize the photoreactivity of the TiO2 coating. Fig. 3(c) shows the relationship between the refractive index measured by ellipsometry and TiO2 film thickness. With increasing film thickness, the refractive index decreases linearly from 2.2 to 1.4 with a slope of 0.06, indicating that the surface of the TiO2 coatings gets rougher with increasing film thickness, resulting in decreasing refractive index and in a larger surface area. This indicates that optical properties such as transmittance, optical band gap, and refractive index are highly influenced by the film thickness as well as the annealing temperature. Fig. 4(a) shows the typical UV-Vis absorbance spectra of 0.1% Fe2O3-coated TiO2 photocatalyst and annealed at 600 8C, obtained during a photocatalytic oxidation reaction with a waste solvent including high amount phenol. With increasing reaction time, most of phenol over 95% is rapidly decomposed within 2 h. Experimental results on decomposition of phenols under UV-irradiation with both the synthesized TiO2 powders with different annealing temperatures, and Al2O3-coated TiO2 and pure TiO2 powders as well as the Fe2O3-coated TiO2 photocatalysts showed a tendency of different reaction rates. For example, Fig. 4(b) shows the changes of phenol concentrations as a function of UV irradiation time. In Fig. 4(b), we can see a distinct behavior of the photocatalytic oxidation–reaction rates of phenol decomposition. For Fe2O3-coated TiO2
Fig. 4. In situ UV-Vis absorbance spectra (a) and changes of phenol concentrations (b) of phenol decomposition with UV irradiation time reacted with pure TiO2 powders that obtained at various annealing temperatures. Variations of nitrobenzene concentrations as a function of UV irradiation time reacted with (c) pure TiO2 powders and (d) both Fe2O3-coated TiO2 and Al2O3-coated TiO2 powders with various annealing temperatures as well as the commercially available Degussa P-25 TiO2 powder.
176
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
photocatalysts annealed at 600 8C, relatively high amounts of phenol can be decomposed within 2.5 h, while the TiO2 photocatalysts annealed at 1000 8C does not remove the phenol. This is an interesting result and it is consistent with the ESR and XRD results which showed that the best crystallinity and peak intensity of the hydroxy {NTiIV– OHd } radicals was very strong in the case of TiO2 photocatalysts annealed at 600 8C compared to the other TiO2 photocatalysts annealed at different temperatures. When we repeated the same experiments on the photocatalytic oxidation reactions of phenol with Al2O3-coated TiO2 powders under the same experimental condition, we also got similar consistency with the ESR results and we can make a conclusion that the photocatalytic oxidation– reaction rates of phenol were decreased in the order of: Fe2O3-, Al2O3-coated TiO2, and pure TiO2 powders, respectively. Conclusively, the hydroxy {NTiIV–OHd } radicals can decompose organic pollutants into harmless products because they have high oxidizing power (standard reduction potential: 2.80 eV). From the photocatalytic reduction test of nitrobenzene, it is also found that the photocatalytic effect of TiO2-based photocatalysts coated with metal oxides is much better than those of both commercially available Degussa P-25 TiO2 photocatalysts and pure TiO2 powders synthesized in this study. Exactly the same experimental process has been adapted for the same samples used in the experiments of Fig. 4(a) and (b). Fig. 4(c) shows the variation of nitrobenzene concentrations as a function of UV-irradiation time obtained during the photocatalytic reduction reaction. In Fig. 4(c), we can also see a different behavior of the photocatalytic reaction rates of nitrobenzene. In the photocatalytic oxidation reactions of phenol shown in Fig. 4(b), quite similar results are obtained in Fig. 4(c). In particular, the maximum photocatalytic reaction rates of nitrobenzene decomposition is observed from the TiO2 photocatalysts annealed at 600 8C and that the nitrobenzene is completely destroyed after UVirradiation for over 1 h. However, there is no strong photocatalytic effect on the TiO2 photocatalysts annealed at 1000 8C. This is in good agreement with the photocatalytic oxidation reactions of phenol with the synthesized TiO2 powder and with the ESR results. The same experiment was done for the Fe2O3- and Al2O3-coated TiO2 powders to compare the photocatalytic effect with the commercially available Deggusa P-25 TiO2 photocatalyst. Fig. 4(d) shows the changes of nitrobenzene concentrations as a function of UV-irradiation time reacted with both the Fe2O3- and Al2O3coated TiO2 powders after annealing the samples at various temperatures. Within 40 min, the nitrobenzene is nearly decomposed on both Al3+- and Fe3+-coated TiO2 powders obtained after annealing at 600 8C, respectively. However, the decomposition rate of the Fe3+-coated TiO2 powder obtained after annealing at 600 8C is 1.3 times higher than that of the Al3+-coated TiO2 powder obtained after annealing under the same temperature. This means that the Fe3+-coated TiO2 powders may have the highest photocatalytic efficiency
than both Al3+-coated TiO2 powders and pure TiO2 powders. Based on Fig. 4(c) and (d), we can see a distinct change between surface-modified TiO2 photocatalysts and pure TiO2 powders. In the case of surface-modified TiO2 photocatalysts (especially with Fe3+-coated TiO2), the nitrobenzene is completely removed within 20 min, while it needs 40 min or more for the uncoated TiO2 and the Degussa photocatalysts, respectively. This result can be explained with two facts. One is that, as the ESR data showed, an ESR peak with a g value of 1.991 attributed to {NTiIII–OH} generating is increased in order of pure TiO2, Al3+- and Fe3+-coated TiO2. Moreover, in the case of Fe2O3-coated TiO2 powder, the intensity of g of 1.991 peak largely decreases compared with Al2O3-coated TiO2 powder due to trapping of excited electrons by oxide coating. The other one is that, because the coated oxides have transition metal ions with vacant dorbitals, Fe3+ can easily trap the excited electrons rather than Al3+. This means that the anatase Ti3+{NTiIII–OH} and oxygen-deficient Ti3+{NTiIII–OH} species generated from a trapping of excited electrons are main parameters to influence the photocatalytic activity. Conclusively, the photocatalytic reaction rate of nitrobenzene decomposition on the Fe2O3-coated TiO2 powders are higher than those of both pure TiO2 and Al2O3-coated TiO2 powders, and that the photocatalytic effect of Fe2O3-coated TiO2 sample annealed at 600 8C is twice better than that of Degussa P-25 TiO2 powder.
4. Conclusions We have synthesized TiO2 powders with the similar structure (rutile/anatase=3:7, surface area: 58 m2/g) as the commercially available Degussa P-25 TiO2-based photocatalysts using both slurry of metatitanic acid and alcohol mixture. Moreover, a coating sol for TiO2-based photocatalysts was also prepared by the sol–gel method, using titanium tetraisopropoxide, acid (HCl, HNO3, HF, etc.), and water. For improving the photocatalytic activity of the synthesized TiO2-based photocatalysts, we also developed a surface modification technology by chemical solution deposition (CSD) with metal oxides such as Fe2O3 and Al2O3. In addition, we studied the structural and compositional changes of the TiO2 powders as well as optical characteristics with XRD, XPS, UV-Vis, and ellipsometry measurements in order to investigate the effects of surface modification by CSD coating and annealing. ESR studies on UV irradiation of the synthesized TiO2 photocatalysts have also been done to understand the photocatalytic reaction as well as to verify the existence of paramagnetic species. At 196 8C, the Fe2O3-coated TiO2 powders annealed at 600 8C showed the largest ESR peaks due to the hydroxy {NTiIV–OHd } radicals because they have high oxidizing power. Experimental results on decomposition of phenols under UV-irradiation with both Fe2O3- and Al2O3-coated TiO2 powders showed a consis-
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
tency with the ESR results, and the photocatalytic oxidation–reaction rates of phenol were decreased in order of: Fe2O3-, Al2O3-coated TiO2, and pure TiO2 powders, respectively. From the reduction test of nitrobenzene using these TiO2 samples, it was found that the photocatalytic effect of a Fe2O3-coated TiO2 sample annealed at 600 8C was twice better than that of Degussa P-25 TiO2 powder.
Acknowledgements Support of this research by the Ministry of Science and Technology is acknowledged. This work is also supported from the Center for Advanced Plasma Surface Technology in the SungKyunKwan University through the ERC project of the Korean Science and Engineering Foundation.
References [1] A. Fujishima, K. Honda, Nature 37 (1972) 238. [2] L. Pamisano, V. Augugliaro, A. Sclafani, M. Schianvello, J. Phys. Chem. 92 (1988) 6710. [3] J.C. Conesa, J. Soria, V. Augualiaro, L. Palmisano, Structure and Reactivity of Surfaces, Elsevier Science Pub., Amsterdam, 1989. [4] N. Schrauzer, T.D. Guth, J. Am. Chem. Soc. 99 (1977) 7198.
177
[5] N. Endoh, J.K. Leland, A.J. Bard, J. Phys. Chem. 90 (1986) 6233. [6] F.A. Grant, Rev. Mod. Phys. 31 (1959) 646. [7] G.S. Brady, Materials Handbook, 10th ed., McGraw-Hill, New York, 1971, p. 815. [8] M. Graezel (Ed.), Energy Resources through Photochemistry and Catalysis, Academic Press, New York, 1983. [9] R.J.G. Clark, The Chemistry of Titanium and Vanadium, Elsevier Science Pub., New York, 1968, p. 266. [10] N. Serpone, E. Pelizzetti, Photocatalysis, John Wiley & Sons Press, 1989. [11] M.R. Hoffmann, S.T. Martin, W. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995) 69. [12] D.F. Ollis, C.Y. Hsiaa, L. Budiman, C.L. Lee, J. Catal. 88 (1984) 89. [13] C.S. Turchi, D.F. Ollis, J. Catal. 119 (1989) 480. [14] G. Grabner, G.Z. Li, R.M. Quint, J. Chem. Soc., Faraday Trans. 87 (1991) 1097. [15] H. Noda, K. Oikawa, Y. Kamada, Bull. Chem. Soc. Jpn. 66 (1993) 455. [16] M. Anpo, T. Shima, Y. Kubokawa, Chem. Lett. (1985) 1799. [17] E. Kuino, U. Syuji, G. Masaya, I. Kenji, S. Tsuneo, F. Hiroshi, K. Yoshifumi, Langmuir 13 (1997) 2803. [18] M. Graezel, R.F. Howe, J. Phys. Chem. 94 (1990) 2566. [19] J. Soria, J.C. Conesa, V. Augualiaro, L. Palmisano, M. Schiavello, A. Sclafani, J. Phys. Chem. 95 (1991) 274. [20] Degussa Technical Bulletin, No. 56—Two Highly Dispersed Metallic Oxides by the Aerosil Process, 1990, p. 3–21. [21] T.K. Kim, PhD thesis (SungKyunKwan University, 1999). [22] T. Yamaki, T. Sumita, A. Miyashita, J. Cryst. Growth 237 (2002) 574.
Development of surface coating technology of TiO2 powder and improvement of photocatalytic activity by surface modification T.K. Kima, M.N. Leea, S.H. Leeb, Y.C. Parkb, C.K. Jungb, J.-H. Boob,* b
a Environmental Research Laboratory, SolarTech. CO. Ltd., Seoul 153-023, Korea Institute of Basic Science and Department of Chemistry, Sungkyunkwan University, Suwon 440-746, Korea
Available online 25 August 2004
Abstract We have synthesized the titanium dioxide (TiO2) powders with the similar structure as a commercially available Degussa P-25 TiO2 powder for TiO2-based photocatalysts. To improve the photocatalytic activity, electronic modification on the TiO2-based photocatalysts by chemical solution deposition (CSD) coating was also carried out with metal oxides such as Fe2O3 and Al2O3. The structural and compositional changes as well as optical characteristics are mainly investigated by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), UV-visible spectroscopy, and ellipsometry measurements. In addition, electron spin resonance (ESR) studies have also been carried out to verify the existence of paramagnetic species such as OH and H2O radicals on UV-irradiated TiO2-based photocatalysts. ESR data showed that the hydroxy radicals could decompose organic pollutants into harmless products because they have high oxidizing power. From the reduction test of nitrobenzene, it is found that the photocatalytic effect of TiO2-based photocatalysts coated with Fe2O3 is twice better than that of commercially available noncoated TiO2 photocatalysts. In the case of photocatalytic oxidation reaction of phenol under UV irradiation, moreover, the experimental results showed a consistency with ESR data indicating that TiO2 coated with metal oxides would be one of the most effective photocatalysts. D 2004 Elsevier B.V. All rights reserved. Keywords: TiO2-based photocatalysts; Surface modification; Chemical solution deposition; Photocatalytic reduction–oxidation reaction; Electron spin resonance analysis
1. Introduction Titanium dioxide (TiO2) has been known as a useful photocatalytic material because it is photosensitive, stable, and inexpensive [1–5]. There is a growing interest in recent years to find new, efficient, and economic methods to clean up the environment of pollution materials. This is inspired by the potential application of TiO2-based photocatalysts for the total destruction of organic compounds in polluted air and wastewater. The bulk material of TiO2 is well known to have three main phases namely rutile, anatase, and brookite [6]. Among them, the TiO2 exists mostly as rutile and anatase
* Corresponding author. Tel.: +82 31 290 7072. fax: +82 31 290 7075. E-mail address: [email protected] (J.-H. Boo). 0040-6090/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2004.07.021
phases and both phases have tetragonal structures. Rutile is a high-temperature stable phase and has an optical energy band gap of 3.0 eV (415 nm), while anatase is formed at a lower temperature with an optical energy band gap of 3.2 eV (380 nm) as well as refractive index of 2.3 [7]. It is well known that generally, the TiO2-based photocatalyst with anatase phase shows more excellent photocatalytic effect than that with rutile phase, and the anatase phase can be transformed into the rutile phase at above 800 8C [8,9]. The role of the holes and electrons at the surface of TiO2 in heterogeneous photocatalysis has been investigated in aqueous suspension for the reduction–oxidation (redox) reaction of several organic compounds [10–14]. Electron spin resonance (ESR) studies have been carried out to verify the existence of paramagnetic species such as d OH and HO2d on UV-irradiation of TiO2 [5,14–19]. However, the photophysical mechanism of surface-modified TiO2 pow-
172
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
ders by metal oxides is not well understood yet. Electronic modification of the photocatalyst by coating of metal oxides such as Fe2O3, Al2O3, MoO2, and MoO3 has a strong effect on the photoreactivity of the system. The efficiency of electron-hole separation and the dynamics of interfacial electron transfer can be dramatically influenced. Various synthetic methods and coating techniques have been developed for development and modification of the TiO2-based photocatalysts including chemical vapor deposition (CVD), metal–organic chemical vapor deposition (MOCVD), and sol–gel methods [20,21]. In this study, therefore, we have synthesized the TiO2 powders with the similar structure as a commercially available Degussa P-25 TiO2-based photocatalysts using both slurry of metatitanic acid and sol–gel method. We have also developed a surface modification technology of TiO2based photocatalysts by chemical solution deposition (CSD) for improving their photocatalytic activity. To investigate the effects of surface modification by CSD and annealing, moreover, we mainly studied the structural and compositional changes of the TiO2 powders as well as optical characteristics with X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), ESR, UV-visible spectroscopy (UV-Vis), and ellipsometry measurements because the electronic modification of the photocatalysts by CSD has a strong effect on the photoreactivity of the TiO2-based photocatalysts.
2. Experimental TiO2 powders were synthesized by a slurry reaction of metatitanic acid [TiO(OH)2] and ethyl alcohol. After ball milling of them for 24 h, the slurry was kept at the temperature of 110 8C for further 24 h in the oven for evaporation of the alcohol, and then annealed at 600 8C for 2 h, resulting in a TiO2 powder formation with 58 m2/g of BET surface area and 87 nm of average particle size, respectively. The synthesized TiO2 powders have the similar structure and surface area as the commercially available Degussa P-25 TiO2 powders (rutile/anatase=3:7; surface area=55 m2/g) [20]. Moreover, a coating sol for TiO2-based photocatalysts was also prepared by the sol–gel method, using titanium tetraisopropoxide, acid (HCl, HNO3, HF, etc.), and water. Titanium tetraisopropoxide was slowly dropped into the 0.4% nitric acid solution after stirring it vigorously for 2 h at room temperature, and then heated at 80 8C for 24 h. During the reaction, isopropanol was removed by distillation and then the milky–bulk reaction solution was gradually changed to blue fine–milky solution. TiO2 coating sol was cooled at room temperature, then colloidal SiO2 and tetraethoxyorthosilicate (TEOS) were added to this solution, stirring for 5 h. To improve their photocatalytic activities, electronic modification on the TiO2-based photocatalysts by CSD was also carried out with metal oxides such as Fe2O3 and Al2O3. In addition, in
order to understand the detailed mechanism of photocatalytic reduction–oxidation reactions, electron spin resonance (ESR) studies have also been carried out.
3. Results and discussion Fig. 1(a) shows the X-ray diffraction pattern of TiO2 powders synthesized by a reaction of titanium tetraisopropoxide, acid (HCl, HNO3, HF, etc.), and water and then annealed at different temperatures in air. In the sample annealed at 300 8C, the powder has in part a nature of anatase crystal structure. With increasing annealing temperature to 600 8C, the crystallinity of the TiO2 powder was increased because the characteristic TiO2(101) diffraction peak with anatase structure becomes stronger than that of 300 8C. However, a new diffraction peaks due to a phase transition from anatase phase to rutile phase also appeared in the TiO2 powder after it was annealed at 1000 8C. By measuring the relative peak intensity of their main diffraction peaks of (b) and (c), the abundance ratio of rutile and anatase phases that were confirmed by XPS (not shown here) can be deducted to be rutile/anatase=5:5. Comparing Fig. 1(a) with Fig. 1(d), that is, XRD pattern of commercially available Degussa P-25 TiO2 powder, one can identify the difference of the abundance ratio with annealing temperature. It is well known that the abundance ratio of Degussa P-25 TiO2 powder, which is now the most excellent photocatalyst in the world, has a rutile to anatase ratio of 3:7. Thus, it is highly desirable for us to make the TiO2 powder with the same abundance ratio as Degussa P25 TiO2 powder as well as similar crystallinity by controlling the annealing temperature, reactive gas, and surface area with near by the same particle size. To make the best photocatalysts, we also did surface modification of the synthesized TiO2 powders by CSD coating with metal oxides such as Fe 2 O 3 (Fe 3+ ) and Al2O3(Al3+) for improving their photocatalytic activity for total destruction of organic compounds in waste solvents. Fig. 1(b) and (c) shows the XRD patterns of TiO2 powders obtained after 0.1% Fe2O3 (b) and 0.1% Al2O3(c) coatings annealed at different temperatures. Below 600 8C annealing temperature, highly oriented TiO2 powders with pure anatase structure could be observed from both Fig. 1(b) and (c). The abundance ratio obtained from the samples was about rutile/ anatase=3:7. At 1000 8C annealing temperature, however, we can see distinct difference of the abundance ratio of rutile to anatase phases after judging the relative main diffraction peaks. Although we obtained the same rutile to anatase ratio as Degussa P-25 TiO2 powder in the case of a TiO2 sample annealed at 600 8C, we cannot directly compare the photocatalytic activity between our synthesized TiO2 powders and commercially available Degussa P-25 TiO2 powder with rutile to anatase ratio only. Therefore, we measured the ESR spectra and tried surface modification of our synthesized TiO2 powders using annealing or chemical solution deposi-
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
173
Fig. 1. X-ray diffraction patterns of (a) pure TiO2, (b) Fe2O3-coated TiO2 and Al2O3-coated TiO2 (c) powders by CSD with annealing at various temperatures. (d) XRD pattern of the commercially available Degussa P-25 TiO2 powder.
tion (CSD). From these results, we made a conclusion that the Fe2O3- or Al2O3-coated TiO2 powders with rutile to anatase ratio of 3:7 have higher photocatalytic activity than that of commercially available P-25 TiO2 powder, suggesting that a surface modification technology by chemical solution deposition (CSD) with metal oxides will improve the photocatalytic activity of the TiO2-based photocatalysts. Titanium dioxide doped with metal ions is found to be one of the most effective photocatalysts. Coating of different metal oxides to the TiO2 is known to influence the charge carrier recombination and interfacial charge transfer rates of photogenerated carriers [2,11,19]. However, the photophysical mechanism of surface-modified TiO2 powders is not well understood yet. In order to elucidate the role of metal ions that are used as dopants, ESR studies on TiO2 samples coated with metal oxides were also carried out in this study. TiO2 samples coated with metal oxides (Al2O3 and Fe2O3) were annealed at the temperatures of 300, 600, and 1000 8C. The samples annealed at 300 and 600 8C existed as mixtures of amorphous and anatase TiO2, while the sample annealed at 1000 8C did as a mixture of anatase and rutile TiO2, respectively. Upon UV-irradiation at 196 8C, ESR spectra of TiO2 samples annealed at 600 8C showed {NTiIV–OHd } radicals with g 1 of 2.016 and g 2 of 2.002 that arise from hole trapping. In addition, another ESR peak with the g value of 1.991 which is attributed to {NTiIII–OH} generating from trapping of excited electrons increased in order of pure TiO2,
Al3+-, and Fe3+-coated TiO2. However, TiO2 samples annealed at 300 8C did not show the EPR peak due to {NTiIII–OH} UV-irradiation. They exhibited ESR peaks with the g values in the range of g of 2.016–2.014 which may be ascribed to O radicals. When 0.1 wt.% Fe3+-coated TiO2 was irradiated at 196 8C, a unique ESR peak with the g value of 2.002 was obtained and its intensity increased after UV-irradiation. This signal may be attributed to O2 that arises from the transfer of an electron to oxygen adsorbed on Ti near-oxygen vacancy. Fig. 2(a) shows a typical ESR spectrum of the Degussa P25 powder obtained after UV irradiation. Four strong peaks are observed at g of 2.023, 2.016, 2.002, and 1.991, respectively. Because the Degussa P-25 powder consists a mixture of anatase and rutile phases, two kinds of O or {NTiIV–OHd } radicals attributed to both phases are observed at g of 2.023 and 2.016, respectively. The peaks at g of 2.002, 1.991, and 1.978 are attributed to O2 , anatase Ti3+{NTiIII– OH}, oxygen deficiency Ti3+{NTiIII–OH}, respectively. Comparing these data with the ESR spectra of Fe2O3-coated TiO2 (Fig. 2(b)) and Al2O3-coated TiO2 (Fig. 2(c)) powders, one can identify two differences. One is that the strong peak ascribed to anatase Ti3+(NTiIII–OH) is due to the Al2O3 coating, reflecting trapping of excited electrons. In addition, the intensities of g of 2.016 and 2.002 peaks also increase due to the formation of many hydroxy {NTiIV–OHd } radicals. The other one is that in the case of Fe2O3-coated TiO2
174
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
we can see a red shift tendency with annealing temperature. From the reflectance data, an optical band gap can be calculated, and the band gaps of the samples decrease with increasing annealing temperature. Reflecting the data of optical band gap between anatase phases (3.2 eV) and rutile phase (3.0 eV), the samples of unannealed TiO2 and annealed TiO2 powders at below 600 8C have relatively higher band gap than those of annealed TiO2 powder at 1000 8C. Comparing the optical band gap for Degussa P-25, the annealed TiO2 powder at 1000 8C will have near by the same optical band gap through the broader range, indicating high photosensitivity due to wide range band gap. Considering the crystallinity, the degree of photosensitivity that can
Fig. 2. (a) ESR spectrum of the Degussa P-25 TiO2 powder obtained after UV irradiation. ESR spectra of the modified TiO2-based photocatalysts with Fe2O3-coated TiO2 (b) and Al2O3-coated TiO2 (c) powders obtained before and after UV irradiations, respectively.
powder, the intensity of g of 1.991 peak largely decreases compared with Al2O3-coated TiO2 powder, due to trapping of excited electrons by oxide coating. This result can be explained by that, because the coated oxides have transition metal ions with vacant d-orbitals, Fe3+ can easily trap the excited electrons rather than Al3+. Fig. 3(a) shows the variation of optical reflectance for the TiO2 powder with different annealing temperatures. Here,
Fig. 3. Variation of optical reflectance (a) and transmittance (b) of the synthesized TiO2 powders and TiO2 coating layers obtained at various annealing temperatures with different thickness. (c) Relationship between refractive index and coating layer thickness.
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
absorb the light is affected by the crystallinity of TiO2 powder. The better the crystallinity with increasing annealing temperature, the high the photoreactivity will be turned out. The reason of having broader band gap in the case of a Degussa P-25 TiO2 powder is thus due to the crystal nature of anatase and rutile mixture which can delay charge carrier recombination time. Fig. 3(b) shows the changes of optical transmittance obtained from TiO2 coating layers on glass substrates with different thickness. Although Fig. 3(b) shows that the transmittance increases with increasing film thickness, the arising point (i.e., initial region with a linear slope) of transmittance decreases with increasing film thickness, suggesting an optical band gap shift to low energy side. This means that with decreasing film thickness, the initial arising point (wavelength) of transmittance is shifted from the UV region (380 nm) to the visible region (about 400 nm). Because TiO thin films with visible range optical transmittance have shown relatively high photocatalytic effect rather than those with UV range optical transmittance [1,22], it is highly desirable for us to make TiO thin films with an optical band gap less than 4.25 nm (above 380 nm). Therefore, to satisfy these conditions, our data suggest a very thin film with below 100-nm thickness. The observed rising point of transmittance is 300 for 231-nm thickness, 350 for 140-nm thickness, and 380 for 98-nm thickness, corresponding to 4.50, 4.34, and 4.25 eV of optical band
175
gaps, respectively. This change indicates that there is a critical thickness to maximize the photoreactivity of the TiO2 coating. Fig. 3(c) shows the relationship between the refractive index measured by ellipsometry and TiO2 film thickness. With increasing film thickness, the refractive index decreases linearly from 2.2 to 1.4 with a slope of 0.06, indicating that the surface of the TiO2 coatings gets rougher with increasing film thickness, resulting in decreasing refractive index and in a larger surface area. This indicates that optical properties such as transmittance, optical band gap, and refractive index are highly influenced by the film thickness as well as the annealing temperature. Fig. 4(a) shows the typical UV-Vis absorbance spectra of 0.1% Fe2O3-coated TiO2 photocatalyst and annealed at 600 8C, obtained during a photocatalytic oxidation reaction with a waste solvent including high amount phenol. With increasing reaction time, most of phenol over 95% is rapidly decomposed within 2 h. Experimental results on decomposition of phenols under UV-irradiation with both the synthesized TiO2 powders with different annealing temperatures, and Al2O3-coated TiO2 and pure TiO2 powders as well as the Fe2O3-coated TiO2 photocatalysts showed a tendency of different reaction rates. For example, Fig. 4(b) shows the changes of phenol concentrations as a function of UV irradiation time. In Fig. 4(b), we can see a distinct behavior of the photocatalytic oxidation–reaction rates of phenol decomposition. For Fe2O3-coated TiO2
Fig. 4. In situ UV-Vis absorbance spectra (a) and changes of phenol concentrations (b) of phenol decomposition with UV irradiation time reacted with pure TiO2 powders that obtained at various annealing temperatures. Variations of nitrobenzene concentrations as a function of UV irradiation time reacted with (c) pure TiO2 powders and (d) both Fe2O3-coated TiO2 and Al2O3-coated TiO2 powders with various annealing temperatures as well as the commercially available Degussa P-25 TiO2 powder.
176
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
photocatalysts annealed at 600 8C, relatively high amounts of phenol can be decomposed within 2.5 h, while the TiO2 photocatalysts annealed at 1000 8C does not remove the phenol. This is an interesting result and it is consistent with the ESR and XRD results which showed that the best crystallinity and peak intensity of the hydroxy {NTiIV– OHd } radicals was very strong in the case of TiO2 photocatalysts annealed at 600 8C compared to the other TiO2 photocatalysts annealed at different temperatures. When we repeated the same experiments on the photocatalytic oxidation reactions of phenol with Al2O3-coated TiO2 powders under the same experimental condition, we also got similar consistency with the ESR results and we can make a conclusion that the photocatalytic oxidation– reaction rates of phenol were decreased in the order of: Fe2O3-, Al2O3-coated TiO2, and pure TiO2 powders, respectively. Conclusively, the hydroxy {NTiIV–OHd } radicals can decompose organic pollutants into harmless products because they have high oxidizing power (standard reduction potential: 2.80 eV). From the photocatalytic reduction test of nitrobenzene, it is also found that the photocatalytic effect of TiO2-based photocatalysts coated with metal oxides is much better than those of both commercially available Degussa P-25 TiO2 photocatalysts and pure TiO2 powders synthesized in this study. Exactly the same experimental process has been adapted for the same samples used in the experiments of Fig. 4(a) and (b). Fig. 4(c) shows the variation of nitrobenzene concentrations as a function of UV-irradiation time obtained during the photocatalytic reduction reaction. In Fig. 4(c), we can also see a different behavior of the photocatalytic reaction rates of nitrobenzene. In the photocatalytic oxidation reactions of phenol shown in Fig. 4(b), quite similar results are obtained in Fig. 4(c). In particular, the maximum photocatalytic reaction rates of nitrobenzene decomposition is observed from the TiO2 photocatalysts annealed at 600 8C and that the nitrobenzene is completely destroyed after UVirradiation for over 1 h. However, there is no strong photocatalytic effect on the TiO2 photocatalysts annealed at 1000 8C. This is in good agreement with the photocatalytic oxidation reactions of phenol with the synthesized TiO2 powder and with the ESR results. The same experiment was done for the Fe2O3- and Al2O3-coated TiO2 powders to compare the photocatalytic effect with the commercially available Deggusa P-25 TiO2 photocatalyst. Fig. 4(d) shows the changes of nitrobenzene concentrations as a function of UV-irradiation time reacted with both the Fe2O3- and Al2O3coated TiO2 powders after annealing the samples at various temperatures. Within 40 min, the nitrobenzene is nearly decomposed on both Al3+- and Fe3+-coated TiO2 powders obtained after annealing at 600 8C, respectively. However, the decomposition rate of the Fe3+-coated TiO2 powder obtained after annealing at 600 8C is 1.3 times higher than that of the Al3+-coated TiO2 powder obtained after annealing under the same temperature. This means that the Fe3+-coated TiO2 powders may have the highest photocatalytic efficiency
than both Al3+-coated TiO2 powders and pure TiO2 powders. Based on Fig. 4(c) and (d), we can see a distinct change between surface-modified TiO2 photocatalysts and pure TiO2 powders. In the case of surface-modified TiO2 photocatalysts (especially with Fe3+-coated TiO2), the nitrobenzene is completely removed within 20 min, while it needs 40 min or more for the uncoated TiO2 and the Degussa photocatalysts, respectively. This result can be explained with two facts. One is that, as the ESR data showed, an ESR peak with a g value of 1.991 attributed to {NTiIII–OH} generating is increased in order of pure TiO2, Al3+- and Fe3+-coated TiO2. Moreover, in the case of Fe2O3-coated TiO2 powder, the intensity of g of 1.991 peak largely decreases compared with Al2O3-coated TiO2 powder due to trapping of excited electrons by oxide coating. The other one is that, because the coated oxides have transition metal ions with vacant dorbitals, Fe3+ can easily trap the excited electrons rather than Al3+. This means that the anatase Ti3+{NTiIII–OH} and oxygen-deficient Ti3+{NTiIII–OH} species generated from a trapping of excited electrons are main parameters to influence the photocatalytic activity. Conclusively, the photocatalytic reaction rate of nitrobenzene decomposition on the Fe2O3-coated TiO2 powders are higher than those of both pure TiO2 and Al2O3-coated TiO2 powders, and that the photocatalytic effect of Fe2O3-coated TiO2 sample annealed at 600 8C is twice better than that of Degussa P-25 TiO2 powder.
4. Conclusions We have synthesized TiO2 powders with the similar structure (rutile/anatase=3:7, surface area: 58 m2/g) as the commercially available Degussa P-25 TiO2-based photocatalysts using both slurry of metatitanic acid and alcohol mixture. Moreover, a coating sol for TiO2-based photocatalysts was also prepared by the sol–gel method, using titanium tetraisopropoxide, acid (HCl, HNO3, HF, etc.), and water. For improving the photocatalytic activity of the synthesized TiO2-based photocatalysts, we also developed a surface modification technology by chemical solution deposition (CSD) with metal oxides such as Fe2O3 and Al2O3. In addition, we studied the structural and compositional changes of the TiO2 powders as well as optical characteristics with XRD, XPS, UV-Vis, and ellipsometry measurements in order to investigate the effects of surface modification by CSD coating and annealing. ESR studies on UV irradiation of the synthesized TiO2 photocatalysts have also been done to understand the photocatalytic reaction as well as to verify the existence of paramagnetic species. At 196 8C, the Fe2O3-coated TiO2 powders annealed at 600 8C showed the largest ESR peaks due to the hydroxy {NTiIV–OHd } radicals because they have high oxidizing power. Experimental results on decomposition of phenols under UV-irradiation with both Fe2O3- and Al2O3-coated TiO2 powders showed a consis-
T.K. Kim et al. / Thin Solid Films 475 (2005) 171–177
tency with the ESR results, and the photocatalytic oxidation–reaction rates of phenol were decreased in order of: Fe2O3-, Al2O3-coated TiO2, and pure TiO2 powders, respectively. From the reduction test of nitrobenzene using these TiO2 samples, it was found that the photocatalytic effect of a Fe2O3-coated TiO2 sample annealed at 600 8C was twice better than that of Degussa P-25 TiO2 powder.
Acknowledgements Support of this research by the Ministry of Science and Technology is acknowledged. This work is also supported from the Center for Advanced Plasma Surface Technology in the SungKyunKwan University through the ERC project of the Korean Science and Engineering Foundation.
References [1] A. Fujishima, K. Honda, Nature 37 (1972) 238. [2] L. Pamisano, V. Augugliaro, A. Sclafani, M. Schianvello, J. Phys. Chem. 92 (1988) 6710. [3] J.C. Conesa, J. Soria, V. Augualiaro, L. Palmisano, Structure and Reactivity of Surfaces, Elsevier Science Pub., Amsterdam, 1989. [4] N. Schrauzer, T.D. Guth, J. Am. Chem. Soc. 99 (1977) 7198.
177
[5] N. Endoh, J.K. Leland, A.J. Bard, J. Phys. Chem. 90 (1986) 6233. [6] F.A. Grant, Rev. Mod. Phys. 31 (1959) 646. [7] G.S. Brady, Materials Handbook, 10th ed., McGraw-Hill, New York, 1971, p. 815. [8] M. Graezel (Ed.), Energy Resources through Photochemistry and Catalysis, Academic Press, New York, 1983. [9] R.J.G. Clark, The Chemistry of Titanium and Vanadium, Elsevier Science Pub., New York, 1968, p. 266. [10] N. Serpone, E. Pelizzetti, Photocatalysis, John Wiley & Sons Press, 1989. [11] M.R. Hoffmann, S.T. Martin, W. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995) 69. [12] D.F. Ollis, C.Y. Hsiaa, L. Budiman, C.L. Lee, J. Catal. 88 (1984) 89. [13] C.S. Turchi, D.F. Ollis, J. Catal. 119 (1989) 480. [14] G. Grabner, G.Z. Li, R.M. Quint, J. Chem. Soc., Faraday Trans. 87 (1991) 1097. [15] H. Noda, K. Oikawa, Y. Kamada, Bull. Chem. Soc. Jpn. 66 (1993) 455. [16] M. Anpo, T. Shima, Y. Kubokawa, Chem. Lett. (1985) 1799. [17] E. Kuino, U. Syuji, G. Masaya, I. Kenji, S. Tsuneo, F. Hiroshi, K. Yoshifumi, Langmuir 13 (1997) 2803. [18] M. Graezel, R.F. Howe, J. Phys. Chem. 94 (1990) 2566. [19] J. Soria, J.C. Conesa, V. Augualiaro, L. Palmisano, M. Schiavello, A. Sclafani, J. Phys. Chem. 95 (1991) 274. [20] Degussa Technical Bulletin, No. 56—Two Highly Dispersed Metallic Oxides by the Aerosil Process, 1990, p. 3–21. [21] T.K. Kim, PhD thesis (SungKyunKwan University, 1999). [22] T. Yamaki, T. Sumita, A. Miyashita, J. Cryst. Growth 237 (2002) 574.