Films - American Chemical Society

Apr 13, 2010 - house or office windows. When irradiated with UV light, stearic acid decomposes ..... 1249–1254. (8) Wu, M.; Lin, G.; Chen, D.; Wang,...
12 downloads 0 Views 3MB Size
7646

J. Phys. Chem. C 2010, 114, 7646–7652

Photocatalytic Performance of Hybrid SiO2-TiO2 Films Samuele Gardin,† Raffaella Signorini,*,† Anna Pistore,‡ Gioia Della Giustina,‡ Giovanna Brusatin,‡ Massimo Guglielmi,‡ and Renato Bozio† Department of Chemical Science, UniVersity of PadoVa, Via Marzolo 1, 35131, PadoVa, Italy, and Department of Mech. Eng.-Materials Section, UniVersity of PadoVa, Via Marzolo 9, 35131, PadoVa, Italy ReceiVed: December 3, 2009; ReVised Manuscript ReceiVed: March 16, 2010

In this paper, we have investigated the properties of a composite SiO2-TiO2/TiO2 nanoparticle system. Hybrid films have been prepared using a sol-gel process, starting from tetraisopropoxy titanate and 3-glycidoxipropyltrimethoxysilane as precursors and performing the synthesis at room temperature. The spin-coated films have been thoroughly characterized with UV-vis and FT-IR spectroscopies, TEM analysis, profilometry, and ellipsometry to evaluate their physical and structural properties. Particular attention has been devoted to the study of their photocatalytic action. 1. Introduction Among different existing photocatalysts, titanium dioxide based films have been intensively investigated for their promising mechanical, chemical, electrical, and optical properties. In particular, great attention has been devoted to the study of the photocatalytic properties of TiO2 powders and thin films useful for the purification of air and water and the provision of selfcleaning surfaces.1-3 This activity can be obtained due to its ability to mineralize a wide range of organic contaminants, such as aromatics, alkanes, alcohols, haloalkanes, dyes, insecticides, and surfactants, and to the photoinduced superhydrophilic effect. TiO2 works as a catalyst for the photodecomposition of organic compounds; the oxidation reaction is represented by the equation hν g Eg(TiO2)

organic + O2 98 CO2 + H2O + mineral acids

where mineral acids are generated when heteroatoms, such as S, N, and Cl, are present in the organic components. In semiconductor photocatalysis, photons of energy above the band gap generate electron-hole pairs that can either recombine or react with surface species. In the second case, the photogenerated electrons reduce the oxygen, while the photogenerated holes mineralize the organic. The latter process probably involves the initial oxidation of surface OH- groups to hydroxyl radicals that then oxidize the organic and any subsequent intermediates.4 It has been demonstrated that the photocatalytic activity strongly depends on the physical properties of the TiO2, such as the crystal structure (amorphous, anatase, rutile, or brookite), the surface area, the particle size, the surface hydroxyls, and so on. In particular, the crystalline structure seems to be the crucial parameter to determine the photocatalytic activity: anatase seems to be the most active phase, whereas amorphous TiO2 shows negligible activity.5 The specific surface area also plays an important role in determining the catalytic activity: smaller TiO2 nanoparticles show a larger surface area and possess greater activity. * To whom correspondence should be addressed. Fax: +39 049 8275239. Tel: +39 049 8275118. E-mail: [email protected]. † Department of Chemical Science, University of Padova. ‡ Department of Mech. Eng.-Materials Section, University of Padova.

Different methods have been employed to prepare the nanoparticles, such as chemical precipitation, microemulsion, hydrothermal crystallization, and sol-gel synthesis.6-10 In the sol-gel processes, TiO2 is usually prepared by hydrolysis and polycondensation reactions of titanium alkoxides, Ti(OR)n, to form oxopolymers, which are then transformed into an oxide network. Because of the high reactivity of titanium alkoxides, some chelating reagents, such as diols, carboxylic acids, or diketonate compounds, are added during the hydrolysis step. After the condensation step, a calcination treatment at 400 °C or more is then required for removing the organic molecules from the final products and completing the crystallization.11,12 By this way, it is possible to obtain films showing high photocatalytic efficiency.13 A different alternative approach consists in the ex situ synthesis of TiO2 nanoparticles, performed at low temperature.14-16 Here, we propose a new method for the synthesis of films containing nanocrystalline TiO2 particles. This process is performed at room temperature, without any calcination step, and allows obtaining films of nanoparticles embedded in a silica-titania amorphous and hybrid organic-inorganic network. By exposing them to the UV light, it is possible to obtain transparent, homogeneous, and crack-free purely inorganic films, having a thickness of hundreds of nanometers. The spin-coated films have been characterized by means of UV-vis and FT-IR spectroscopies, TEM analysis, profilometry, and ellipsometry to evaluate their physical and structural properties. FT-IR spectroscopy has been first exploited to observe the evolution of the film microstructure with increasing UV treatment time. Moreover, stearic acid has been used, combined with FT-IR measurements, to measure the photocatalytic activity of the preirradiated films toward the photodegradation of organic compounds. The obtained photoactivity does not reach the highest values reported in the literature for sol-gel films but is comparable to that of commercial films.17 However, the absence of any calcination process and the soft conditions for the preparation of samples make this material a suitable candidate for a lowcost coating on thermally unstable substrates. 2. Experimental Section 2.1. Materials. Basic catalyzed hybrid silica-titania sols (G7Ti3) were prepared employing tetraisopropoxytitanate (Ti-

10.1021/jp911495h  2010 American Chemical Society Published on Web 04/13/2010

Photocatalytic Performance of Hybrid SiO2-TiO2 Films ISOP) and 3-glycidoxipropyltrimethoxysilane (GPTMS) as precursors. They are all purchased from Aldrich and used without further purification. 2-Metoxyethanol (2-MeOEtOH) was employed as solvent, bidistilled water for hydrolysis, and sodium hydroxide (1M) as catalyst. The sol was prepared in two different steps. In the first one, GPTMS was mixed with water (H2O/GPTMS ) 3) and the solution was stirred at room temperature for one night. This prehydrolysis step is necessary to compensate for the higher reactivity of titanium alkoxides, with respect to that of silicon alkoxides. In the second step, 2-MeOEtOH and NaOH (0.3% M in GPTMS) were added in that order. Separately, TiISOP was added to 2-MeOEtOH at a 1:1 volume ratio, and the solution was stirred for about 10 min. 2-MeOEtOH plays two different roles, acting as solvent as well as stabilizer of titanium alkoxide toward the hydrolysis-precipitation reaction.18,19 In fact, it has the ability to coordinate TiISOP, decreasing its hydrolysis rate. The two solutions were then mixed, reaching a concentration of 150 g/L SiO2 + TiO2 (that is, the oxide content if the sol was dried and calcined); then, 2-MeOEtOH was added to reach a sol concentration of 100 g/L SiO2 + TiO2. Finally, the sol was sonicated for 30 min and then left to react under stirring for 5 h at room temperature. The final Si/Ti molar ratio is 70: 30. All the sols were filtered with a microporous membrane (0.2 µm Millipore) before use. Cleaned silicon wafers and sodalime or quartz slides were used as substrates. Hybrid silica-titania coating films were obtained from the prepared sol by spin-coating and drying in an oven at 60 °C for 30 min (RT sample). Before the photocatalytic activity tests, the film has been exposed for 10 min to the UV radiation to decompose the organic components present inside the film and obtain a completely inorganic SiO2-TiO2 network. 2.2. Film Characterization. The film microstructure has been analyzed by infrared absorption spectra, in the range of 400-4500 cm-1, recorded by a Fourier transform infrared spectroscope (Jasco FT-IR-620), with the accuracy of (1 cm-1. TEM-EDS analysis has been performed in order to investigate the film structure and the nature of the titania clusters. The optical properties of the G7Ti3 films have been analyzed by means of ellipsometry and UV-vis absorption spectroscopy. Ultraviolet-visible (UV-vis) absorption spectra from films deposited on quartz slides were measured in the range of 200-800 nm by a spectrophotometer (JASCO V-570), with the accuracy of (0.3 nm. The refractive index of the coatings on quartz substrates was measured by ellipsometry, which allows also the determination of the film morphology. The film thickness has been examined also with a Tencor T-10 profilometer. 2.3. Photocatalytic Performance Measurements. The photocatalytic performances of the films toward the degradation of organic compounds have been measured using stearic acid (SA) as model material. It is a well-known reference material, and it provides a reasonable model compound for the type of solid organic films that deposit on exterior glass surfaces, such as house or office windows. When irradiated with UV light, stearic acid decomposes according to the following reaction hν g Eg semic.

CH3(CH2)16CO2H + 26O2 98 18CO2 + 18H2O

a process involving a transfer of 104 electrons.

J. Phys. Chem. C, Vol. 114, No. 17, 2010 7647

Figure 1. UV absorption spectrum (green line) of the G7Ti3 sample, showing an absorption edge at around 300 nm. The emission profiles of the 6 W UV lamp (blue line) and Hamamatsu LC5 lamp (red line) are also reported. It can be clearly seen that the 6 W lamp emits outside of the absorption wavelength of the active sample, whereas the LC5 lamp presents some emission peaks below 325 nm, where the sample still absorbs.

A film of SA has been deposited on the photocatalytic substrate, under test, by spin-coating at 1500 rpm a solution 8.8 × 10-3 M in methanol. This results in an initial stearic acid coverage of approximately 1.8 × 1015 molecules/cm2, calculated from the integrated area of the SA peaks in the 2800-3000 cm-1 range in the FT-IR spectra.20,21 The SA has three peaks in this range: the peaks at 2958, 2923, and 2853 cm-1 due to the asymmetric in-plane C-H stretching mode of the CH3 groups and to the asymmetric and symmetric C-H stretching modes of the CH2 groups, respectively. From the integrated area under these peaks, it is also possible to estimate the surface density of SA as a function of the increasing UV exposure time and so to obtain its photodecomposition rate. A single 6 W UV lamp with a maximum emission at 365 nm (∼1.5 mW/cm2) and a Hamamatsu LC5 UV mercury-xenon lamp with a large emission spectrum (with an emission intensity of 3500 mW/cm2 at a 1 cm distance or ∼30 mW/cm2 at 20 cm) have been used as irradiation sources. The emission profiles of these lamps are shown in Figure 1, together with the absorption spectrum of the photocatalytic film. 3. Results and Discussion 3.1. FT-IR Spectra. The FT-IR spectra of the RT film before and after 1-4 sequential 150 s steps of exposure to the LC5 UV lamp (d ∼ 3.5 cm, E ∼ 450 mW/cm2) are reported in Figure 2. A 30 min thermal treatment, at 500 °C on an unexposed film, and 600 s UV exposed samples are also reported for comparison. The FT-IR spectra of all the samples show two frequency ranges of great interest. In the first interval, between 3500 and 2500 cm-1, there are the large ν(O-H) band and the two peaks at 2930 and 2870 cm-1, relative to the νas(CH2) and νs(CH2), respectively. The ν(O-H) band receives four main contributions,22 reported in Figure 3. The type (a) silanol group is characterized by a sharp absorption peak around ∼3740 cm-1, and it never appears in any sample, indicating that the silanol groups are mainly hydrogen-bonded. The RT sample shows a narrower peak (with respect to the UV and 500 °C treated samples), indicating a narrower distribution of silanol groups (only (c) and (d) groups), probably with predominance of type (c) at ∼3380 cm-1 over type (d)

7648

J. Phys. Chem. C, Vol. 114, No. 17, 2010

Figure 2. FT-IR spectra of the RT sample before (red line) and after different UV exposure times with the LC5 lamp from a distance of 3.5 cm, with a corresponding incident intensity (considering the entire emission spectrum) of ∼450 mW/cm2. The spectra of samples treated at 500 and 800 °C are also reported (brown and dark green spectra, respectively).

Figure 3. Schematic representation of the four different possible silanol groups, free (a) and linked with different hydrogen bonds (b-d): the large circle is a silicon atom, the shaded circle is an oxygen atom, and the small circle is a hydrogen atom.

Gardin et al. groups at ∼3280 cm-1. A contribution of the peaks comes probably also from the ν(OH) group of 2-MeOEtOH that absorbs at 3420 cm-1.23 The UV exposure induces an initial (for the first two steps) increase of the band, due to the increase of Si-OH groups and to the molecular water released as a result of the decomposition of the organic components. After 450 s of UV exposure, when the organic component is almost entirely decomposed, the peak intensity then returns to the initial value, but there is an increase of the (d) groups and a lower wavenumber shoulder appears due to the presence of (b) silanol groups. The thermal treatment at 500 °C induces the analogous effect of the 600 s UV exposure concerning the organic component decomposition. However, in this case, the band is less intense, probably because the water from the thermal oxidation of hybrid groups is quickly evaporated at this temperature. The two sharp bands at 2930 and 2870 cm-1 can be used to control the organic chain evolution under different UV and thermal treatments: it can be clearly seen that the propylic chain becomes progressively weaker for longer UV exposure times and completely disappears after UV exposure for 600 s. The same effect can be obtained with 30 min thermal treatment at 500 °C. The second representative region, in the 1300-700 cm-1 range, contains a great number of superimposed peaks. To evaluate the different peak contributions to the resulting spectra, deconvolutions have been performed with a multipeak Gaussian fitting, as shown in Figure 4, and the results are reported in Table 1. The RT spectrum is reported in the top portion of Figure 4. The intense peak [4] at 1107 cm-1 is related to νAS(C-O-C) bonds of the propylic chain, whereas the sharp peak [1] at 1200 cm-1 is related to the symmetric stretching vibration of the CH2 groups of the

Figure 4. FT-IR absorption spectra in the region of 1300-800 cm-1 of the RT sample, 600 s UV exposed sample, UV + thermal 500 °C annealed sample, and 800 °C annealed sample. The black dotted lines are the recorded spectra, the green lines are the deconvoluted peaks, and the red lines are the curves resulting from the fitting.

Photocatalytic Performance of Hybrid SiO2-TiO2 Films

J. Phys. Chem. C, Vol. 114, No. 17, 2010 7649

TABLE 1: Peak Deconvolution for the RT Sample in Figure peaks RT 600 s UV UV + 500 °C 800 °C

position area position area position area position area

-1

max [cm ] max [cm-1] max [cm-1] max [cm-1]

4,a

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

1199 0.90

1171 0.30 1172 2.18 1192 1.20 1190 2.43

1145 1.88 1120 1.80 1133 2.24 1145 0.85

1107 8.21

1054 0.16

1045 9.53 1045 7.93 1048 8.31 1072 10.89

968 0.55 943 4.45 947 5.05 954 4.14

912 4.33 898 1.00 904 1.31 916 2.05

856 0.40

a The multiple peaks have been fitted as the accumulation of Gaussian functions. The resulting coefficient of determination R2 and reduced χ2 are 0.99982 and 4.3 · 10-7, respectively. The peak positions are reported together with the integrated area, width, and height.

propilyc chain; both are related to the presence of the organic component in the RT films. These bands disappear after UV exposure or thermal treatments, indicating a photo- or thermal decomposition of the organic components. The four shoulders at lower ([2], [3]) and higher ([5], [6]) frequencies, with respect to the main band at 1110 cm-1, are assigned, respectively, to the longitudinal (LO) and transverse (TO) optical components of the asymmetric Si-O-Si stretching vibrations νAS(Si-O-Si).13 We fit the LO shoulder with a two-peak deconvolution, and not with only one, as reported by some authors, because a better fitting results. The LO/TO ratio can be correlated with the volume fraction of the residual porosity. In fact, the longitudinal optical component band of νAS(Si-O-Si) in normal incidence transmission spectroscopy is activated only thanks to the pore light scattering of the IR radiation, such that a fraction of the absorbed light is effectively obliquely incident.24,25 According to this hypothesis, the LO/TO ratio increases from 0.22 for the RT sample to 0.65 and 0.56 after the 450 and 600 s UV treatments, respectively, as a consequence of the decomposition of the organic component that leaves pores inside the film. With thermal treatment at 500 °C, the LO/TO ratio becomes 0.41, indicating a film densification and decreases to 0.30 for the 800 °C treatment. The C-H stretching of the epoxy ring and the presence of residual 2-MeOEtOH are evidenced by the presence of the shoulder [9] at 856 cm-1 that disappears after only one 150 s step of UV exposure. Finally, the broad peak [8] centered at 912 cm-1 is probably the result of the superimposition of the Si-O-Ti, Si-OH, and Si-O- groups’ stretching vibrations and of the two 2-MeOEtOH bands at 880 and 960 cm-1. After the UV or thermal treatment at 500 °C, this peak shifts to lower frequencies, probably because the solvent contribution disappears. The band is now well-fitted with two peaks, the first one centered at 900 cm-1 and assigned to the Si-Ostretching vibration, while the main one, at 940 cm-1, is assigned to the Si-O-Ti and Si-OH stretching vibrations.26-28 In the sample treated at 800 °C, this peak can be attributed only to Si-O-Ti vibrations because Si-OH groups are no longer present in the film, as confirmed from the disappearance of the band centered at 3400 cm-1. After the thermal treatment, at 800 °C, the Si-O-Si band shifts to 1070 cm-1 and increases significantly, while the Si-O-Ti band decreases. This means that the number of Si-O-Si bonds has increased, while that of the Si-O-Ti bonds has decreased, suggesting a segregation of the titanium ions, which moved away from the silicon ions and migrated into a titania-rich region, allowing the formation of additional siloxane bridges. The remaining Si-O-Ti bonds are then mostly located on the surface of the titania cluster. To estimate

Figure 5. Ellipsometric determination of the refractive index of the RT and treated films in the range of 400-1200 nm. An enhancement in the refractive index for increasing UV exposure time is shown. The index increases even more with thermal treatment at 500 and 800 °C.

the Si-O-Ti connectivity (Ti dispersion in the Si matrix), it is interesting to use the parameter D(Si-O-Ti), defined as29,30

D(Si-O-Ti) )

S(Si-O-Ti) χSi · SSi-O-Si) χTi

where S(Si-O-Ti) and S(Si-O-Si) are the deconvoluted peak area of the ν(Si-O-Ti) at 940 cm-1 and νAS(Si-O-Si) at 1120 and 1170 cm-1, respectively; χSi and χTi are the molar proportions of Si and Ti. respectively. D(Si-O-Ti) gives a qualitative estimate of the fraction of Si-O-Ti species over the total Ti content and thus a sort of mixing efficiency. The results obtained are 2.6, 3.4, and 1.8 for the sample irradiated with UV light for 600 s and the samples with 500 and 800 °C thermal treatments, respectively. The higher D value seems to confirm the migration of the Ti ions, suggested above, even if this result cannot be considered as quantitative because it is not possible to estimate the Si-OH contribution to the Si-O-Ti peak. 3.2. Ellipsometric Measurements. The ellipsometric measurements of the RT sample and after different UV and thermal treatments reported in Figure 5 confirm what is outlined above. The film thickness decreases and the refractive index increases for increasing UV exposure time, according to the organic component decomposition and the film densification. The film porosity for the 600 s UV and for the thermally treated samples has been estimated from the Lorentz-Lorentz equation, assuming a pore refractive index of 131

(nf2 - 1) (nf2 + 2)

)

(1 - Vp)(nm2 - 1) (nm2 + 2)

7650

J. Phys. Chem. C, Vol. 114, No. 17, 2010

Gardin et al.

TABLE 2: RT Sample Film Thicknessa sample RT 300 600 500 600 800

s UV s UV °C × 30 min s UV + 500 °C × 30 min °C × 30 min

thickness (nm)

shrinkage (%)

porosity (%)

800 393 281 261 231 203

/ 51 65 67 71 75

10 7 6 3.7

a The changes in thickness and porosity after different treatments are also reported. All the data are obtained from the ellipsometric analysis.

Figure 6. UV-visible spectra of the RT sample and 600 s UV and thermal annealed films.

where Vp is the volume fraction of pores, nf is the measured refractive index, and nm is the effective index of an inorganic fully densified matrix, assumed to be 1.68 by extrapolation of Schroeder data.32 These results are also reported in Table 2. The sharp absorption band in the UV region (Figure 6) is indicative of the tendency of titania, in silica-titania sol-gel films, to form a separated phase, composed of titania or titanium-oxo clusters of nanometric sizes.23,33 This band does not move to longer wavelengths after UV or thermal treatments, suggesting that the cluster size and nature do not change. The UV-visible spectra of the RT and treated samples show no remarkable absorption in the 400-800 nm range (not reported in Figure 6). 3.3. TEM Analyses. To investigate the presence and the nature of the TiO2 clusters, TEM analyses have been performed. Figure 7 reports TEM images of an unirradiated and a UVirradiated film, where it is possible to see some crystalline particles of titania, well-dispersed in an amorphous background. The particle diameter is in the range of of 2-6 nm. The energydispersive spectroscopy (EDS) analysis shows the presence of a high amount of Ti and O in the nanoparticle-rich zone of the

sample, while the amount of Ti is much lower when a sample zone without nanoparticles have been analyzed. EDS has confirmed that the particles shown in the TEM micrographs are TiO2: the plane distance from the TEM picture corresponds to a brookite crystalline form of TiO2. All particles with a crystalline nature together with the great amount of hydroxyl groups and with the residual film porosity are responsible for the photocatalytic behavior of the film reported below. 3.4. Photocatalytic Tests. The first test has been performed with the 6 W UV lamp whose emission spectrum is shown in Figure 1 superimposed to the absorption spectrum of the G7Ti3. From the absorption spectra, it is clear that G7Ti3 films do not absorb at the emission wavelength of the lamp, so it is reasonable to suppose that they do not show any photocatalytic action toward the decomposition of the organic compounds. This fact is confirmed by the FT-IR spectra of the 600 s UV exposed film spin-coated with stearic acid (top panel of Figure 8). No remarkable differences in the spectra are visible also after 8 h of exposure with the UV lamp. Successive tests have been performed with the Hamamatsu LC5 UV lamp, possessing several emission bands in a wide spectral range, which enter also into the absorption band of the G7Ti3 photocatalytic film (see Figure 1). In particular, this lamp shows four peaks (centered at 290, 297, 303, and 313 nm) in the spectral region below 325 nm, where G7Ti3 film starts to absorb. This lamp has been used at a distance of 10 cm from the sample, giving an irradiance of ∼12 mW/cm2. After only 1 h of exposure time, the stearic acid IR peaks disappear, indicating the complete decomposition of the spin-coated acid film (center panel of Figure 8). The trend of SA concentration as a function of UV exposure time is reported in the bottom part of Figure 8. To confirm the G7Ti3 photocatalytic action, a stearic acid film spin-coated directly on a Si substrate has also been exposed to the same treatment. In that case, also after 2 h of exposure, the AS peaks only slightly decrease. To compare this result with others present in the literature, the formal quantum efficiency (FQE) and the quantum yield (QY) have also been calculated and reported in Table 3.8,34,35 The quantum yield of the G7Ti3 film is comparable with that of the ActivTM film (0.01-0.02), but is lower with respect to the pure titania sol-gel film (0.15-0.18).36 This extraordinary high photoactivity of sol-gel films can probably be ascribed to their great porosity that determines an increase of the contact area between the active film and the organic component to be destroyed. Nevertheless, it should be considered that only a small fraction of our film is constituted of active titanium clusters, while the great part is a SiO2-like or mixed SiO2-TiO2 inert matrix.

Figure 7. TEM analysis of the unexposed sample (left panel) and 600 s UV exposed sample, showing the titanium clusters or radius ranging from 2 to 6 nm (right panel).

Photocatalytic Performance of Hybrid SiO2-TiO2 Films

J. Phys. Chem. C, Vol. 114, No. 17, 2010 7651 By increasing the relative amount of titania in this film, it will be probably possible to improve its photocatalytical activity, approaching or even reaching the value reported above. Moreover, the great advantage of this film is the absence of a thermal annealing; by this way, they can be spin-coated even on thermally unstable substrates. 4. Conclusions We have reported the room-temperature synthesis of hybrid SiO2-TiO2 sol-gel films with nanocrystalline TiO2 clusters homogeneously dispersed inside the amorphous matrix. The structure of these films has been characterized with TEM micrographs and the induced UV edge absorption changes, confirming the presence of crystalline nanoparticles. Moreover, we have tested their photocatalytic activities using stearic acid as the reference material. This system seems to be really promising for future applications, such as UV-treated inorganic films or photocatalytic coatings, to be deposited on thermally degradable substrates. Acknowledgment. This work was supported by the grant PRIN 2007 (2007LN873M) and the CNR-INSTM agreement project “PROMO”. References and Notes

Figure 8. Photocatalytic test performed with a 6 W UV lamp (with a maximum emission at 365 nm, I ∼1.5 mW/cm2) (top). The irradiation up to 8 h on the RT and on the preirradiated active film spin-coated with stearic acid does not produce any remarkable difference in the FT-IR band intensity. Photocatalytic test performed with a LC5 lamp, at a distance of ∼10 cm between the sample preirradiated and the lamp, giving I ∼ 12 mW/cm2 energy (center). The decomposition of the AS is complete after 100 min of UV exposure (bottom).

TABLE 3: Formal Quantum Efficiency (FQE) and the Quantum Yield (QY) Calculated for the G7Ti3 Sample fraction incident of light intensity × 1018 rate, Ri × 1013 FQEa × 10-4 2 2 photon/cm /min mol/cm /min mol/photon absorbed, f b QYc × 10-2 1.11

2.05

0.18

0.138

0.013

a FQE (δ) calculated as δ ) rate of stearic acid destruction (molecules removed/cm2)/incident light intensity (photons cm-2). b Fraction of light absorbed (f) calculated using f ) (1 - 10-Abs(λ)) for the four main peaks, from an analysis of the overlap of the UV-vis spectra of the film with that of the emission spectrum of the lamp (using the data in Figure 1). c QY (φ) calculated as φ ) δ/f.

(1) Mo, J. H.; Zhang, Y. P.; Xu, Q. J.; Lamson, J. J.; Zhao, R. Y. Atmos. EnViron. 2009, 43, 2229–2246. (2) Euvananont, C.; Junin, C.; Inpor, K.; Limthongkul, P.; Thanachayanont, C. Ceram. Interfaces 2008, 34, 1067–1071. (3) Sheng, G. D.; Li, J. X.; Wang, S. W.; Wang, X. K. Prog. Chem. 2009, 21, 2492–2504. (4) Mills, A.; LeHunte, S. J. Photochem. Photobiol., A 1997, 108, 1– 35. (5) Ohtani, B.; Ogawa, Y.; Nishimoto, S. J. Phys. Chem. B 1997, 101, 3746–3752. (6) Scolan, A.; Sanchez, C. Chem. Mater. 1998, 11, 3217–3223. (7) Lal, M.; Chabra, V.; Ayub, P.; Maitra, A. J. Mater. Res. 1998, 13, 1249–1254. (8) Wu, M.; Lin, G.; Chen, D.; Wang, G.; He, D.; Feng, S.; Xu, R. Chem. Mater. 2002, 14, 1974–1980. (9) Wang, C.-C.; Ying, J. Y. Chem. Mater. 1999, 11, 3113–3120. (10) Addamo, M.; Bellardita, M.; Di Paola, A.; Palmisano, L. Chem. Commun. 2006, 4943–4945. (11) Yoon, K. H.; Noh, J. S.; Kwon, C. H.; Muhammed, M. Mater. Chem. Phys. 2006, 95, 79–83. (12) Addamo, M.; Augugliaro, V.; Di Paola, A.; Garcia-Lopez, E.; Loddo, V.; Marci, G.; Palmisano, L. Thin Solid Films 2008, 516, 3802– 3807. (13) Palmisano, G.; Gutierrez, M. C.; Ferrer, M. L.; Gil-Luna, M. D.; Augugliaro, V.; Yurdakal, S.; Pagliaro, M. J. Phys. Chem. C 2008, 112, 2667–2670. (14) Bischof, B. L.; Anderson, M. A. Chem. Mater. 1995, 7, 1772– 1778. (15) Yu, J.; Yu, J. C.; Leung, M. K.-P.; Ho, W.; Cheng, B.; Zhao, X.; Zhao, J. J. Catal. 2003, 217, 69–78. (16) Yuwono, A. H.; Xue, J.; Wang, J.; Elim, H. I.; Ji, W.; Li, Y.; White, T. J. J. Mater. Chem. 2003, 13, 1475–1479. (17) Mills, A.; Hill, G.; Bhopal, S.; Parkin, I. P.; O’Neil, S. A. J. Photochem. Photobiol., A 2003, 160, 185–194. (18) Pirson, A.; Mohsine, A.; Marchot, P.; Michaux, B.; Cantfort, O.; Pirard, J. P.; Lecloux, A. J. J. Sol-Gel Sci. Technol. 1995, 4, 179–185. (19) Pabon, E.; Retuert, J.; Quijada, R.; Zarate, A. Microporous Mesoporous Mater. 2004, 67, 195–203. (20) Paz, Y.; Luo, Z.; Rabenberg, L.; Heller, A. J. Mater. Res. 1995, 10, 2842–2848. (21) Del Monte, F.; Larsen, W.; Mackenzie, J. D. J. Am. Ceram. Soc. 2000, 83, 628–634. (22) Fidalgo, A.; Ilharco, L. M. J. Non-Cryst. Solids 2001, 283, 144– 154. (23) Galo Cardenas, T.; Ricardo Oliva, C. Mater. Res. Bull. 1998, 33, 1599–1608. (24) Seco, A. M.; Gonc¸alves, M. C.; Almeida, R. M. Mater. Sci. Eng. 2000, B76, 193–199. (25) Almeida, R. M.; Vasconcelos, H. C.; Ilharco, L. M. Proc. SPIE 1994, 2288, 678–686.

7652

J. Phys. Chem. C, Vol. 114, No. 17, 2010

(26) Gao, X.; Wachs, I. E. Catal. Today 1999, 51, 233–254. (27) Fidalgo, A.; Ilharco, L. M. J. Non-Cryst. Solids 2001, 283, 144– 154. (28) Que, W.; Hu, X. Thin Solid Films 2003, 436, 196–202. (29) Dutoit, D. C. M.; Schneider, M.; Baiker, A. J. Catal. 1995, 153, 165–176. (30) Klein, S.; Thorimbert, S.; Maier, W. F. J. Catal. 1996, 163, 476– 488. (31) Brusatin, G.; Guglielmi, M.; Innocenzi, P.; Martucci, A.; Battaglin, G.; Pelli, S.; Righini, G. J. Non-Cryst. Solids 1997, 220, 202– 209.

Gardin et al. (32) Sorek, Y.; Reisfeld, R.; Weiss, A. M. Chem. Phys. Lett. 1995, 244, 371–378. (33) Soppera, O.; Croutxe´-Barghorn, C.; Lougnot, D. J. New J. Chem. 2001, 25, 1006–1014. (34) Fretwell, R.; Douglas, P. J. Photochem. Photobiol., A 2001, 143, 229–240. (35) Mills, A.; Lepre, A.; Elliott, N.; Bhopal, S.; Parkin, I. P.; O’Neil, S. A. J. Photochem. Photobiol., A 2003, 160, 213–224. (36) Mills, A.; Elliott, N.; Hill, G.; Fallis, D.; Durrant, J. R.; Willis, R. L. Photochem. Photobiol. Sci. 2003, 2, 591–596.

JP911495H