Preillumination of TiO 2 and Ta 2 O 5 Photoactive Thin Films As a Tool to Tailor the Synthesis of...

Preillumination of TiO2 and Ta2O5 Photoactive Thin Films As a Toolto Tailor the Synthesis of Composite Materials

Juan R. Sanchez-Valencia,† Ana Borras,†,‡ Angel Barranco,*,† Victor J. Rico,†

Juan P. Espinos,† and Agustın R. Gonzalez-Elipe†

Instituto de Ciencia de Materiales de SeVilla (CSIC-UniV. SeVilla), AVda, Americo Vespucio 49,41092 SeVilla, Spain, and Nanotech@surfaces Laboratory, EMPA Materials Science and Technology,

Feuerwerkerstrasse 39, CH-3602 Thun, Switzerland

ReceiVed March 11, 2008. ReVised Manuscript ReceiVed June 5, 2008

Illumination of TiO2 thin films with UV light is known to induce the transformation of the surface of this materialfrom partially hydrophobic into fully hydrophilic. The present work shows that this transformation is accompaniedby other effects that may be used to control the synthesis of composite materials. For this purpose, TiO2 and Ta2O5

transparent thin films with a columnar structure and open pores were prepared by electron evaporation at glancingangles. Transparent TiO2 thin films with micropores (i.e., pores smaller than 2 nm) prepared by plasma enhancedchemical vapor deposition (PECVD) were also used. All these films became hydrophilic upon UV illumination.Rhodamine 6G and Rhodamine 800 dyes were irreversibly adsorbed within the columns of the TiO2 and Ta2O5 thinfilms by immersion into a water solution of these molecules. Isolated and aggregated molecules of these two dyeswere detected by visible absorption spectroscopy. The infiltration adsorption efficiency was directly correlated withthe acidity of the medium, increasing at basic pHs as expected from simple considerations based on the concepts ofthe point of zero charge (PZC) in colloidal oxides. The infiltration experiments were repeated with columnar TiO2

and Ta2O5 thin films that were subjected to preillumination with UV light. It was found that this treatment produceda modification in the type (isolated or aggregated) and amount of dye molecules incorporated into the pores. Moreover,the selective adsorption of a given dye in preilluminated areas of the films permitted the lithographic coloring of thefilms. Preillumination also controls the UV induced deposition of silver on the surface of the microporous TiO2 thinfilms. It was found that the size distribution of the formed silver nanoparticles was dependent on the preilluminationtreatment and that a well-resolved surface plasmon resonance at around 500 nm was only monitored in the preilluminatedfilms. A model is proposed to account for the effects induced by UV preillumination on the TiO2 and Ta2O5 oxidesurfaces. The possibilities of this type of light treatment for the tailored synthesis of nanocomposite thin films (i.e.,dye-oxide, metal nanoparticles-oxide) are highlighted.

Introduction

Since the pioneering work of Wang et al.,1 it is well-establishedthat TiO2 becomes hydrophilic upon illumination with UV light.This original finding has given rise to a wide series of workstrying to explain the reasons of this change in surface properties.2–6

Although most works attribute this effect to an amphiphilictransformation of the surface of TiO2 where some regions wouldbecome enriched in -OH groups, there are also some authorsthat attribute the change in wetting angle to the photo-oxidativeremoval of carbonaceous contamination.6 It has been also reportedthat UV preillumination of TiO2 powders is an effective treatmentfor enhancing the photoactivity of this material in subsequentphotocatalytic processes.7 Recently, we have also shown thatthin films of Ta2O5, another wide band gap semiconductor (Eg

) 4.4 eV), become hydrophilic when illluminated with UV light.8

However, despite the ample literature existing about this wettingeffect and other photocatalytic properties of these semiconductoroxides,9 to our knowledge, no studies have tried to use UV-preillumination as a tool for the synthesis of new materials. Inthis paper, we present a series of experiments showing thatpreilluminated TiO2 and Ta2O5 thin films behave differentlytoward, to the adsorption in their pores, surfaces of cationic dyemolecules and/or the formation of small silver clusters on itssurface. This investigation has been carried out with columnarTiO2 and Ta2O5 thin films prepared by physical vapor deposition(PVD) and with homogeneous microporous TiO2 thin filmsprepared by plasma enhanced chemical vapor deposition(PECVD).10,11 All the utilized films presented optical quality(i.e., they did not disperse the light) so that the final compositefilms might be integrated directly into optical or photonicdevices.12

Two types of composite thin films have been synthesized. Thefirst one incorporates the cationic dye molecules Rhodamine 6G(Rh6G) and Rhodamine 800 (Rh800) into the pores of thecolumnar thin films. These materials are prepared by immersion

* To whom correspondence should be addressed. E-mail: [email protected].

† Instituto de Ciencia de Materiales de Sevilla (CSIC-Univ. Sevilla).‡ Nanotech@surfaces Laboratory.(1) Wang, R.; Hashimoto, K.; Fujishima, A.; Chikuni, M.; Kojima, E.; Kitamura,

A.; Shimohigoshi, M.; Watanabe, T. Nature 1997, 388, 431–432.(2) Stevens, N.; Priest, C. I.; Sedev, R.; Ralston, J. Langmuir 2003, 19, 3272–

3275.(3) Zeman, P.; Takabayashi, S. J. Vac. Sci. Technol., A 2002, 20, 388.(4) Miyauchi, M.; Kieda, N.; Hishita, Sh.; Mitsuhashi, T.; Nakajima, A.;

Watanabe, T.; Hashimoto, K. Surf. Sci. 2002, 511, 401.(5) Borras, A.; Lopez, C.; Rico, V.; Gracia, F.; Gonzalez-Elipe, A. R.; Richter,

E.; Battiston, G.; Gerbasi, R.; McSporran, N.; Sauthier, G.; Gyorgy, E.; Figueras,A. J. Phys. Chem. C 2007, 111, 1801–1808.

(6) Zubkov, T.; Stahl, D.; Thompson, T. L.; Panayotov, D.; Diwald, O.; Yates,J. T. J. Phys. Chem. B 2005, 109, 15454.

(7) Murabayashi, M.; Itoh, K.; Zaho, L.-H.; Kitahara, K. J. AdV. Oxid. Technol.2002, 5, 67.

(8) Rico, V.; Lopez, C.; Borras, A.; Espinos, J. P.; Gonzalez-Elipe, A. R. Sol.Energy Mater. Sol. Cells 2006, 90, 2944.

(9) Carp, O.; Huisman, C. L.; Reller, A. Prog. Solid State Chem. 2004, 31,33.

(10) Gracia, F.; Holgado, J. P.; Gonzalez-Elipe, A. R. Langmuir 2004, 20,1688.

(11) Borras, A.; Cotrino, J.; Gonzalez-Elipe, A. R. J. Electrochem. Soc. 2007,154, 152.

(12) Zhu, X.-L.; Lo, D. Appl. Phys. Lett. 2002, 80, 917.

9460 Langmuir 2008, 24, 9460-9469

10.1021/la800773v CCC: $40.75 2008 American Chemical SocietyPublished on Web 07/19/2008

of the films in solutions of these molecules with controlled pHs.Different results are obtained depending on whether the filmswere preilluminated. A second type of composite film is basedon microporous TiO2 thin films immersed in a silver cationsolution and, after extraction from the solution, UV-illluminatedin air. Again, differences are obtained depending on whether themicroporous films were subjected to a preillumination treatment.

Rh6G molecules are known to diffuse within the interlayerspace of silicates and other layered materials where they becomeincorporated.13–17 The incorporation of these molecules intosol-gel silica is another related topic that has deserved muchattention during the last years.18 The main interest of thesematerials relies on the fluorescence emission of the dye moleculesthat can be modified after their incorporation into a solid matrixacting as a host. Thus, from these studies, it is known thatfluorescent Rh6G molecules can be adsorbed in the interlayerspace of silicates in the form of isolated molecules, dimers (eitherH or J dimers, with the first one not presenting fluorescentemission), or even higher aggregates.18,19 A composite materialformed by Rh6G dye molecules incorporated in mesoporousTiO2 thin films has been proposed as a good candidate for laserapplications.19,20 The Rh800 laser dye has been less studied thanother rhodamine dyes, although reports exist in the literatureabout its fluorescent behavior upon aggregation in aqueoussolutions21,22 or its use for labeling organic fluids (blood, serum,etc.).21–23

On the other hand, formation of silver and gold particles onthe surface of UV-illuminated TiO2 from solutions of theirrespective ions is a well-studied topic because of its implicationsin photography, for the recovery of metals, and/or the preparationof nanoparticles.8,24,28 In this latter case, much interest exists inobtaining a narrow distribution of particle sizes yielding well-resolved surface resonance plasmon structures. These plasmonstructures are originated in the enhancement of the electric fieldinside the Ag nanocluster. Changes in the wavelength maximumof their surface plasmon resonances are the basis of the use ofmetal nanoparticles as sensors, as “chameleonic” materials, and,when they are integrated into films, for nonlineal optic applica-tions.27–31

In relation to some of these previous works using clay mineralsor oxide powders as a host or support of the dyes or metal

nanoparticles, a clear advantage of the thin films used here is thatthey do not disperse the visible light and therefore can be directlyused for optical and/or photonic applications.12

A first objective of the present work is to develop experimentalstrategies that permit the incorporation of dye molecules intofilms or the formation on their surface of a narrow distributionof particle sizes of silver. A second objective is to control thissynthesis by preilluminating the TiO2/Ta2O5 thin films. As aresult of the different experimental evidence obtained in thecourse of this work, it is shown that preillumination of TiO2/Ta2O5 thin films is a critical step for the tailored synthesis ofcomposite materials containing dyes or silver nanoparticles. Toaccount for the singular behavior of the preilluminated surfaces,a model is proposed based on the suggestions of Wang et al.1

about the hydrophilic transformation of UV-illuminated TiO2.The capacity of the surface of oxides in water dispersions tobecome either positively or negatively charged depending on thepH of the solution32,33 is also taken into consideration within thismodel. We believe that the reported findings open new ways totailor the synthesis of composite materials formed by oxide thinfilms acting as a host and dyes or metal particles as an opticallyactive component.

Experimental SectionMaterials. Composites of the type dye-TiO2 or dye-Ta2O5 were

prepared by using columnar thin film oxides as host materials. Forthis purpose, transparent and amorphous TiO2 and Ta2O5 thin films,hereafter called TiO2-I and Ta2O5, were prepared by PVD at roomtemperature on quartz and silicon substrates. Evaporation was carriedout in a electron bombardment evaporator by using TiO and Ta2O5

pellets as target materials. Stoichiometric thin films of TiO2 andTa2O5 were obtained by performing the evaporation in 10-4 Torrof O2. Evaporation was carried out by placing the substrates atglancing angle geometries (70°) with respect to the evaporator source.This geometry produces a tilted columnar microstructure for thefilms.34–36 Films with a thickness around 400 nm for TiO2-I and 500nm for Ta2O5 were prepared by this method. A common characteristicof all these films is that they are very porous and, therefore, arecharacterized by relatively low refractive indices (n smaller than 1.9and 1.5 for the TiO2 and Ta2O5 thin films, respectively).

Another kind of transparent TiO2 thin film was used for thesynthesis of Ag-TiO2 composite materials. They were prepared onquartz and silicon substrates by PECVD according to experimentalprotocols previously reported.10,11,37 These thin films were char-acterized by n values between 1.95 and 2.0. As it will be shownbelow, they present a homogeneous microstructure without observ-able defects or inhomogeneities. A thickness around 100 nm wasselected for these samples. In these thin films, only micropores (i.e.,pores smaller than 2 nm) could be detected by water adsorptionisotherms with a quartz crystal monitor.37 Hereafter, these thin filmsamples will be called TiO2-II.

After preparation, all the films were stored in the dark for at least1 month before their use for the synthesis of the composite films.As-prepared films are hydrophilic due to the effect of the UV lightof the plasma/evaporator sources,38 and prolonged storage in thedark is required to make the surface of the films hydrophobic. This

(13) Martınez-Martınez, V.; Lopez-Arbeloa, F.; Banuelos-Prieto, J.; Arbeloa-Lopez, T.; Lopez-Arbeloa, I. Langmuir 2004, 20, 5709.

(14) Gemeay, A. H. J. Colloid Interface Sci. 2002, 235, 251.(15) Bujdak, J.; Iyi, N.; Sasai, R. J. Phys. Chem B 2004, 108, 4470.(16) Bujdak, J.; Iyi, N. J. Phys. Chem. B 2005, 109, 4608.(17) Martınez, V.; Salleres, S.; Banuelos, J.; Lopez-Arbeloa, F. J. Fluoresc.

2006, 16, 233.(18) del Monte, F.; Mackenzie, J. D.; Levy, D. Langmuir 2000, 16, 7377.(19) Sasai, R.; Fujita, T.; Iyi, N.; Itoh, H.; Takagi, K. Langmuir 2002, 18,

6578.(20) Vogel, R.; Meredith, P.; Kartini, I.; Harvey, M.; Riches, J. D.; Bishop,

A.; Heckenberg, N.; Trau, M.; Rubinsztein-Dunlop, H. ChemPhysChem 2003, 4,595.

(21) Sekiguchi, K.; Yamaguchi, Sh.; Tahara, T. J. Phys. Chem. A 2006, 110,2601.

(22) Abugo, O. O.; Fair, R.; Lakowicz, J. R. Anal. Biochem. 2000, 279, 142.(23) Sakanoue, J.; Ichikawa, K.; Nomura, Y.; Tamura, M. J. Biochem. 1997,

121, 29.(24) Fernandez, A.; Gonzalez-Elipe, A. R. Appl. Surf. Sci. 1993, 69, 285.(25) Fernandez, A.; Caballero, A.; Gonzalez-Elipe, A. R.; Herrmann, J.-M.;

Dexpert, H.; Villain, F. J. Phys. Chem. 1995, 99, 3303–3309.(26) Yonezawa, Y.; Kometani, N.; Sakaue, T.; Yano, A. J. Photochem.

Photobiol., A 2005, 171, 1–3.(27) Friedmann, D.; Hansing, H.; Bahnemann, D. Z. Phys. Chem. 2007, 221,

329.(28) Ohko, Y.; Tatsuma, T.; Fujii, T.; Naoi, K.; Niwa, Ch.; Kubota, Y.;

Fujishima, A. Nat. Mater. 2003, 2, 29.(29) Del Coso, R.; Requejo-Isidro, J.; Solis, J.; Gonzalo, J.; Afonso, C. N.

J. Appl. Phys. 2004, 95, 5.(30) Ballesteros, J. M.; Solis, J.; Serna, R.; Afonso, C. N. Appl. Phys. Lett.

1999, 74, 2791.(31) Noguez, C.; Barrera, R. G. Phys. ReV. B 1998, 57, 302.

(32) Kosmulski, M. Chemical Properties of Material Surfaces; Marcel DekkerInc.: New York, 2001.

(33) Hiemenz, P. C.; Rajagopalan, R. Principles of Colloids and SurfaceChemistry; Marcel Dekker Inc.: New York, 1997.

(34) Robbie, K.; Brett, M. J. J. Vac. Sci. Technol., A 1997, 15, 1460.(35) Wang, S.; Xia, G.; He, H.; Yi, K.; Shao, J.; Fan, Z. J. Alloys Compd. 2007,

431, 287.(36) Kiema, G. K.; Colgan, M. J.; Brett, M. J. Sol. Energy Mater. Sol. Cells

2005, 85, 321.(37) Borras, A.; Barranco, A.; Gonzalez-Elipe, A. R. J. Mater. Sci. 2006, 41,

5220.(38) Han, J.-B.; Wang, X.; Wang, N.; Wei, Z.-H.; Yu, G.-P.; Zhou, Z.-G.;

Wang, Q.-Q. Surf. Coat. Technol. 2006, 200, 4876.

Preillumination of TiO2 and Ta2O5 Thin Films Langmuir, Vol. 24, No. 17, 2008 9461

permits us to start the preillumination experiments with filmspresenting an equivalent state of hydrophobicity.

Methods. The microstructure of the TiO2 and Ta2O5 thin filmsamples prepared on silicon was examined by field emission scanningelectron microscopy (FESEM) in a Hitachi S5200 microscope. Cross-sectional views, obtained by cleaving the silicon substrates, weretaken for the different samples.

UV-visible absorption spectra in the transmission mode wererecorded with a Perkin-Elmer Lambda 12 spectrometer for the thinfilms deposited on quartz plates. Although the absorbance is thetypical magnitude used for presenting UV-vis absorption spectraof dyes in liquid solutions, in this work, we have preferred to representthe percentage of transmitted light. This is so because it is the usualway of presenting these data when dealing with optical thin films.39

Films were preilluminated in air under ambient conditions for 20min with the full spectrum of a 125 W Xe lamp. Calibration of thelight intensity at the position of the samples yielded a photon intensityof 2 W cm-2. This preillumination was enough to render fullyhydrophilic the surface for all the films (i.e., water contact anglesless than 10°). Immediately after the preillumination treatment, thefilm samples were immersed in the different solutions as describedbelow. The preilluminated samples deposited on Si(100) wereanalyzed by Fourier transform infrared (FT-IR) spectroscopy in aNicolet 510 spectrometer and by X-ray photoelectron spectroscopy(XPS) in a VG ESCALAB 210 spectrometer. These experimentsyielded complementary information about the state of the samples,and the main results are reported in the Supporting Information.

Refractive indices, determined by UV-vis absorption spectros-copy and by variable angle spectroscopic ellipsometry (VASE), andwetting angles, measured by the Young method, were used forcharacterization of the films. A detailed description of theseexperiments can be found in previous works.8,10,11,37

Infiltration of Dye Molecules to Prepare Dye-TiO2 (Ta2O5)Composite Films. Rhodamine 6G and Rhodamine 800 fluorescentdyes supplied from Aldrich were used for these infiltrationexperiments. A 10-4 M solution of these dyes in water at controlledvalues of pH was used for these experiments. The pH of the solutionwas controlled between 3.8 and 6.6 by adding defined amounts ofHCl or NaOH. The TiO2-I or Ta2O5 films were immersed in one ofthese solutions and maintained there for 5 min. The samples werethen taken out from the solution and washed with water at the samepH as the dye solutions. With this washing, any dye molecule thatis not irreversibly incorporated into the thin films was removed fromtheir surfaces. The films were then dried by blowing nitrogen ontotheir surfaces. After these treatments, the samples presented thecharacteristic pink or blue color of the two dye molecules used forinfiltration. The intensity of the color changed with the pH of thesolution, a feature that pointed out that the dye adsorption degreeis dependent on this parameter.

Experiments were also carried out by immersing the thin filmsamples in 10-4 M solutions of the two dyes.

The amount of dye molecules incorporated into the films wasassessed after extracting them by prolonged stirring of the filmsamples in ethanol until total film discoloration. The measuredextinction coefficient of the dye solutions resulting from this stirringwas then compared with the values of a calibration curve determinedfrom the measurements of a series of reference solutions with knownconcentrations of the dye molecules. These results have been includedin the Supporting Information.

Adsorption/Photoreduction of Silver to Prepare Ag-TiO2

Composite Films. Another set of experiments was carried out tryingto synthesize Ag-TiO2 composite thin films depicting a well-definedsurface resonance plasmon structure.29,31 For these experiments,we used the TiO2-II thin films prepared by PECVD.

The film samples, either stored in the dark or preilluminated withthe Xe lamp, were introduced in a 10-3 M aqueous solution ofAgNO3 for 12 h. Once the samples were taken out from the solution,they were dried by blowing nitrogen onto their surfaces and then

exposed to the light of the UV lamp in air for 20 min. The sampleswere subjected to several cycles of immersion plus illuminationuntil a well-defined plasmon structure was observed in the UV-visspectrum of the preilluminated films.

Results

Thin Film Microstructure and Hydrophilic Conversion.Figure 1 shows cross-sectional FESEM micrographs corre-sponding to Ta2O5 and TiO2-I thin films prepared by PVD at anangle of 70° with respect to the evaporation source, and to aTiO2-II film prepared by PECVD. The films selected for theFESEM analysis were thicker than those used for the infiltrationsto make the observation of their microstructural features easier.Films of this kind have been used as hosts for the differentadsorption/deposition experiments discussed in the next sections.Samples Ta2O5 and TiO2-I are highly porous as deduced fromthe FESEM micrographs in Figure 1 and from the value of theirrefraction indices estimated as, respectively, 1.47 and 1.86. Thesevalues are a proof of their high porosity, since they are muchsmaller than those of the corresponding bulk materials (i.e 2.2for Ta2O5 and between 2.4 and 2.7 for TiO2

40). Samples TiO2-IIhad refractive indices between 1.95 and 2.0, indicating that theyare less porous. In previous studies on this sample, we showedthat its porosity mainly consisted of micropores (i.e., pores smaller

(39) Gracia, F.; Holgado, J. P.; Caballero, A.; Gonzalez-Elipe, A. R. J. Phys.Chem. B 2004, 108, 17466.

(40) Lide, D. R., Ed. CRC Handbook of Chemistry and Physics; CRC Press:Boca Raton, FL, 1990.

Figure 1. Cross-sectional FESEM micrographs of the TiO2-I (top), Ta2O5

(middle), and TiO2-II (bottom) thin films.

9462 Langmuir, Vol. 24, No. 17, 2008 Sanchez-Valencia et al.

than 2 nm37). The exclusive presence of micropores agrees withthe aspect of this sample in Figure 1 where no hints of anydefined defect and/or structure can be seen. Meanwhile, in samplesTa2O5 and TiO2-I, the existence of mesopores (i.e., pores largerthan 2 nm) can be deduced from their two micrographs presentedin Figure 1. It is expected that accessibility of large moleculesis restricted in samples TiO2-II but it will be feasible in the othertype of films. In turn, it is likely that control of surface processeswill be better achieved in samples TiO2-II where, according toFigure 1, there are no accessible paths to the interior of its structure.

The wetting angles on these three thin film samples beforeillumination were 95°, 85°, and 80° for Ta2O5, TiO2-I, and TiO2-II, respectively. The contact angles became practically zero afterillumination (i.e., values smaller than 10°) with UV light for avariable period of time depending on the sample. This is a well-known behavior for TiO2.1–6 For Ta2O5, the possibility ofconverting its surface to hydrophilic upon UV irradiation hasbeen recently reported by our group.8

Dye Adsorption and pH. Differences in the adsorption of thedye molecules within the pores of the two host thin films canbe found by changing the pH of the solution. This is clearlyillustrated by the UV-vis absorption spectra reported in Figures2 and 3 for, respectively, TiO2-I and Ta2O5 thin films and theRh6G and Rh800 dyes. The spectra of the infiltrated films areshown for three values of pH (3.8, 5.5, and 6.6). The two curvespresented in each panel of these figures correspond to the originalfilms as prepared and after their immersion in the solutions withthe dye molecules. The spectra of the original TiO2-I thin filmsare characterized by the typical oscillations found for thin filmsof a different refractive index than that of the substrate.10,39 Thefact that similar oscillating structures are not so clearly observedfor the Ta2O5 thin films is because they are characterized by arefraction index very similar to that of the quartz substrate. Inaddition to the oscillations, the spectra of the immersed filmspresent well-defined bands with different shapes and intensitiesthat must be attributed to the incorporated dye molecules. Infact, these bands present the typical shape and maxima of theRh6G and Rh800 molecules infiltrated in the interlayer space ofclay and other mesoporous host materials13–18,41 or in solu-tion.21–23 However, the most interesting evidence for theobjectives of the present investigation are the changes in the

intensity of these bands according to the pH and the type of dyeand thin film material. Thus, it is found that no infiltration ofRh6G dye occurs at pH 3.8 in TiO2-I thin films. Some infiltrationalready occurs at a pH of 5.5 (i.e., close to the point of zerocharge (PZC) for TiO2

32,33 and is further enhanced at a pH of6.6. The Rh800 molecules present a different behavior, sincethey present a considerable adsorption already at a pH of 3.8.The adsorption capacity remains practically unmodified at a pHof 5.5 and slightly increases at a pH of 6.6. A similar tendencyas a function of pH is found for the Ta2O5 thin films (cf. Figure3) where it is found a general increase in the intensity of theRh6G dye absorption bands as the pH increases. In this case,some adsorption of Rh6G already occurs at a pH of 3.8, a resultthat might be connected with the fact that the value of the PZCof this oxide is around 3.32,33 Also a common result with theTiO2-I thin films is that the adsorption of Rh800 is less affectedby the value of the pH (cf. Figure 3, right panel). Extraction ofthe dye molecules into an ethanol solution as explained in theexperimental section provides a way of estimating the amountof incorporated molecules within the films. The measured valuesare shown in the Supporting Information (see Table T1).

According to the point of zero charge (PZC) theory of colloids,immersion of an oxide at a pH higher than its PZC leads to thedevelopment of a negative charge on its surface by dissociationof the-OH surface groups. This charge is compensated by cationsin a double layer of charge.32,33 The previous results suggest thatincorporation of dye molecules into the pores of the thin filmsis favored when there is an excess of negative charge on thesurface of the columns defining their microstructure. Thisrequirement seems to be critical for Rh6G and less strict for theRh800 molecules.

The spectra of the samples immersed in the Rh6G solution atpHs 5.5 and 6.6 depict well-defined absorption bands with maximaat around 499 and 529 nm which are characteristic of this dyemolecule.13–19,42 In previous works about the intercalation ofRh6G molecules in layered clays and other materials, it has beenrecognized that the relative intensity of these two bands dependson the degree of aggregation of these molecules in the adsorbedor intercalated state.13–19,42 Meanwhile, the spectra of the samplesimmersed in the Rh800 solution are well-characterized by

(41) Shinozaki, R.; Nakato, T. Langmuir 2004, 20, 7583.(42) Toptygin, D.; Packard, B. Z.; Brand, L. Chem. Phys. Lett. 1997, 277,

430–435.

Figure 2. UV-vis transmission spectra recorded for the TiO2-I thinfilms before and after immersion in the indicated dye solutions atincreasing pHs: (left) spectra taken from the experiments with Rh6Gand (right) spectra taken from the experiments with Rh800.

Figure 3. UV-vis transmission spectra recorded for the Ta2O5 thinfilms before and after immersion in the indicated solutions with dyesat increasing pHs: (left) Rh6G dye solution and (right) Rh800 dye solution.


absorption bands with maxima at around 629 and 693 nm typicalof this dye molecule.21 The different relative intensities of thesebands and a shoulder at around 570 nm also observed in thespectra in Figure 3 must also be attributed to a different degreeof aggregation of this molecule as studied for its aqueoussolutions.21

Another qualitative assessment by comparing the shape of thespectra of the dye molecules in Figures 2 and 3 and that reportedin previous works13–23,42 is that isolated Rh6G and Rh800molecules are predominant at acid pHs, while at higher pHs therelative concentration of dimers and other aggregates increases.However, a thorough analysis of the agglomeration degree canonly be carried through the analysis of the fluorescence yield ofthe composite films, a subject that is outside the scope of thiswork primarily devoted to ascertain the effect of UV preillu-mination on the adsorption of dyes in TiO2 and Ta2O5 photoactivethin films.

Dye Adsorption Experiments with Preilluminated Samples.Figures 4 and 5 show the UV-vis absorption spectra for TiO2-Iand Ta2O5 thin film samples that have been preilluminated withthe UV lamp and then immersed in a solution of Rh6G or Rh800

dyes at the three pHs considered in this work. The spectra of thesamples immersed without preillumination (i.e., the absorptionbands of dyes in Figures 2 and 3) are included again forcomparison. It is apparent from these figures that, for the twofilms, preillumination produces a decrease in the Rh6G dyeadsorption degree. This decrease is particularly important forRh6G and Ta2O5 at pH 3.8 where adsorption is suppressed inthe preilluminated sample. By contrast, the adsorption of Rh800remains practically unaltered if we compare the preilluminatedsamples with the nonillluminated ones. Only a small change inthe shape of the spectrum of the dye molecules can be observedfor the Ta2O5 thin films. According to the different shape of thespectra, the observed change indicates an increase in the relativeconcentration of adsorbed dye aggregates.19,42 To account forthe factors that may contribute to the observed decrease in dyeadsorption in the illuminated samples, they were analyzed byinfrared (FT-IR) spectroscopy. This analysis showed a slightincrease in the intensity of the broad band around 3000 cm-1

attributed to -OH groups on the surface (see Figure S1 in theSupporting Information). In another related experiment, a TiO2

Figure 4. UV-vis transmission spectra recorded for preilluminated andnonilluminated TiO2-I thin films after immersion in the dye solutionsat increasing pHs: (left) Rh6G dye solution and (right) Rh800 dye solution.

Figure 5. UV-vis transmission spectra recorded for preilluminated andnonilluminated Ta2O5 thin films after immersion in the dye solutions atincreasing pHs: (left) Rh6G dye solution and (right) Rh800 dye solution.

Figure 6. UV-vis transmission spectra of TiO2-I (left) and Ta2O5 (right)thin films immersed in a solution of Rh6G and Rh800 at pH ) 6.6. Thespectra are recorded for preilluminated and nonilluminated films asindicated for immersion times of 5 min (top panels) and 3 days (bottompanels). The image in the right panel corresponds to a Ta2O5 thin filmilluminated with UV-light covering the left part of the sample with ametal mask and subsequently immersed for 3 min in the solution of thetwo dyes (i.e., equivalent to the situation of the spectra in the top rightpanel).

Figure 7. UV-visible transmission spectra of the TiO2-II thin filmssubjected to successive treatments involving immersion in a AgNO3

solution followed by UV illlumination. Results are shown for preillu-minated (bottom) and nonilluminated films (top).


thin film sample was measured after and before UV illuminationby XPS. The O1s spectrum after irradiation showed thedevelopment of a shoulder in the O1s peak at around 531.5 eVthat is congruent with the formation of new -OH groups on thesurface (see Figure S2 in the Supporting Information). Thesetwo experiments support that illumination under ambientconditions produces a certain increase in the hydroxylation degreeof the surfaces of our samples. These results are in agreementwith the early proposal of Wang et al.1 that hydroxylation is aconsequence of the UV irradiation of TiO2.

The decrease of the Rh6G adsorption in the preilluminatedfilms persists even if the films are immersed in a solutioncontaining the two dyes. Some results obtained in experimentswhere the films where immersed in a solution of the two dyesat pH ) 6.6 are reported in Figure 6. The two series of spectrain this figure are obtained for the films immersed in this solutionfor 5 min (i.e., the conventional protocol used for our immersionexperiments) and for equivalent films left in the solution for amuch longer time (i.e., 3 days). The experiment is carried outfor both preilluminated and nonilluminated films. The spectra ofthe nonilluminated Ta2O5 thin films immersed in the solution fora short time are characterized by the superposition of theabsorption bands due to the two dyes. Meanwhile, the spectrumof the preilluminated sample of this oxide only depicts the bandscorresponding to Rh800 without any trace of Rh6G. This resultis in agreement with those reported in Figure 5 obtained forseparate solutions of each dye. For the TiO2-I thin films, a similareffect is obtained, although in this case a small amount of Rh6Gmolecules are also incorporated in the preilluminated sample.This behavior also agrees with the results in Figure 4 with solutionscontaining only one type of dye molecule. As a result of thedifferent adsorption capacity of the films for the two dyes afterpreillumination, the films may become differently colored as

illustrated by the image of a double colored film inserted inFigure 6. This image has been taken for a Ta2O5 thin film whoseleft part was covered during the preillumination treatment. Tothe eye, the preilluminated part presents a blue color (typical ofRh800), while the nonilluminated part had a pink-blue aspect(typical of the contribution of both the Rh800 and Rh6G spectra).

Another interesting effect that arises by comparing the intensityof the bands of the two dyes in either the nonilluminated or thepreilluminated films of Ta2O5 and TiO2-I is that their sum ispractically constant. This suggests that the number of incorporateddye molecules is roughly the same and probably depends on thetotal porosity of the films and/or the accessibility of their poresto the relatively high amount of Rhodamine molecules. The factthat, in the preilluminated films, the spectrum is dominatedby the features of the Rh800 molecules indicates that, in thesesamples, adsorption of this dye is favored with respect to Rh6G.

On the other hand, when the films were left in the solution for3 days, the spectra of the nonilluminated and the preilluminatedfilms were very similar, with the two cases consisting of similarspectral contributions of the Rh6G and Rh800 molecules. Thismeans that any preference for the incorporation of Rh800 in thepreilluminated films is lost after this long immersion period,with the films behaving as nonilluminated. This result is consistentwith the known loss of superhydrophilicity reported for il-luminated TiO2 films when they are stored for a long period oftime. A rough estimation of the transformation time in this caseis of some hours. This period of time was required to observea color change in the films left in the solution of the two dyes.After this period of time, the blue color attributed to Rh800 issubstituted by a mixed pink+blue color due to the incorporationof the two dyes into the films. The slight differences betweenthe two spectra suggest that the adsorption mechanism of thedyes is time dependent and/or that the characteristics of the

Figure 8. FESEM planar view micrographs of the TiO2-II thin films immersed in a AgNO3 solution and then UV illluminated in air for two (top)and three times (bottom): (left) preilluminated film and (right) nonilluminated film.


surfaces of the two oxides change when they remain immersedin the solutions for a long period of time.

Silver Reduction upon UV Illumination. Preillumination ofTiO2 can be also used to control the formation of metal particlesby photoreduction of cations or complex anions of noble metalssuch silver, gold, and so forth.24–26,43 In the present work,photoreduction of Ag+ has been intended with TiO2-II thin filmsthat were immersed in the dark in an AgNO3 solution and thensubjected to UV illumination. A similar experiment on the TiO2-Ithin film led to the incorporation of large amounts of silver andto a noncontrollable process. This result was expected becauseof the open porous structure of these films that makes accessibletheir whole thickness to the silver solution. This situation is nottrue in sample TiO2-II where, because of its nonaccessiblemicrostructure, reduction processes only occurring at the topmostsurface of the films are expected. After removal of the films fromthe solution, they were transparent, thus indicating that nophotoreduction of silver had taken place during immersion and/or drying. However, when thereafter they were illuminated inair, they became colored or dark depending on the preilluminationtreatment. This sequence of steps was repeated several timesuntil the coloring/darkening was appreciable. This evidenceindicates that some silver cations remain adsorbed on the surfacesof the films and are photoreduced by UV illumination. Mostinteresting by this experiment was that the final aspect of thefilms depended on whether they were subjected to the preillu-mination treatment according to the standard procedure describedin the Experimental Section. Figure 7 shows the UV-visabsorption spectra recorded for preilluminated and nonilluminatedsamples after successive immersion/photoreduction treatments.It is apparent in this figure that while the nonilluminated samplepresents a flat spectrum depicting absorption in the whole rangeof wavelengths, the spectrum of the preilluminated sample ischaracterized by a well-defined band with a maximum shiftingbetween 504 and 522 nm for samples immersed/illuminated oneand three times, respectively. This band is characteristic of thesurface plasmon resonance structure of small silver particles andis responsible for the coloring found for these thin films.27,29,30,43,44

It is important to mention that this behavior was only found forTiO2-II thin films with a microporous structure. For otherpreilluminated samples such as the TiO2-I thin films preparedby PVD or TiO2 thin films with columnar microstructures preparedby PECVD,37 the obtained spectra were similar to those in thetop panel of Figure 7.

The two types of spectra in Figure 7 point to the existence ofa wide distribution of particle sizes in the nonillluminated samplesand to a relatively sharp distribution of sizes in the preilluminatedones. This different particle size distribution was effectivelyconfirmed by FESEM analysis of the samples. Figure 8 showsa series of FESEM micrographs of the surface of the two typesof films after two and three successive immersion/photoreductioncycles. In Figure 9, we present the particle size histograms forthese two types of samples. It is apparent from these figures thatin the nonilluminated samples the particle size distribution iswider with small and large particles of up to 100 nm in size. Sucha wide distribution of particle sizes produces an extendedabsorption of light for all the wavelengths of the spectrum. Bycontrast, in the preilluminated sample, the silver particles presenta narrower size distribution that must be responsible for theappearance of the surface plasmon resonance peak in Figure 7.It is also interesting in Figure 9 that the number of silver particlesin the preilluminated sample is higher than that in the nonillu-

minated one. Effectively, after counting the number of particles(particles µm-2) in the images in Figure 8, we obtained valuesof 250 and 600 for the nonilluminated and preilluminated samples,respectively.

Discussion

Thin Film Microstructure and Functionality of the ThinFilms As Host Materials. The previous experiments have beenperformed with thin films that have well-controlled microstruc-tures and porosities. In particular, TiO2-I and Ta2O5 thin filmsamples are characterized by a columnar microstructure withlarge and open pores which are very accessible to species comingfrom the medium. Owing to these characteristics, they are idealsystems to study diffusion processes of relatively large moleculessuch as Rh6G and Rh800 dyes. Experiments dealing with theintercalation of dye molecules in layered clays and other porousmaterials are a common topic in colloidal chemistry andfluorescence studies.13–18,41 Our columnar systems have incommon with the layered materials their accessibility to moleculesof the medium, although they differ in the distribution ofelectrostatic charges through their microstructural features. Claymaterials may have some negative charge in their layers that iscompensated by cations present in the interlayer space. Theamount of negative charge can be adjusted by modifying thetype of cations forming the structure of the layers. Typically,cationic dyes become intercalated into the interlayer space ofclays by an ion exchange mechanism.13–17 By contrast, withoutbeing immersed in a solution, the columns in the TiO2-I or Ta2O5

samples are electrically neutral and the mechanism by which theRh6G and the Rh800 cationic dyes incorporate into the filmsmust be different as discussed in the next subsection.

Previous experiments have also shown that thin films with avery open columnar microstructure do not develop well-definedplasmon structures upon photoreduction of silver. In fact, a largevariation of particle sizes of silver was obtained with thesecolumnar thin films independently of whether they were subjectedto preillumination. By contrast, samples TiO2-II, characterizedby a homogeneous microstructure where only micropores arepresent, have resulted in ideal films for the preparation of Ag/TiO2 nanocomposites with a well-defined surface plasmonresonance structure. According to the micrographs in Figure 8

(43) Yu, J.; Xiong, J.; Cheng, B.; Liu, S. Appl. Catal., B 2005, 60, 211–221.(44) Tanahashi, I.; Yamazaki, F.; Hamada, K. Chem. Lett. 2006, 35, 454–455.

Figure 9. Size distributions of silver particles for the nonilluminated(top) and preilluminated (bottom) TiO2-II thin films. The distributionsshown correspond to the samples subjected to two and three immersionplus illlumination cycles. The insets correspond to the same distributionin an enlarged y-scale to appreciate the differences in the formation oflarge particles in each case.


and their analysis in Figure 9, a narrower distribution of particlesizes is obtained when preilluminated TiO2-II thin films are usedfor the experiment. In this regard, it is interesting to mention thateven if these films have some residual porosity consisting ofmicropores,37 the growth of particles mainly occurs on theirsurface.

Incorporation of Dye Molecules into Columnar Thin FilmOxides. The immersion experiments described in Figures 2 and3 have pointed out that a significant adsorption of Rh6G molecularcations into the pores of the TiO2-I/Ta2O5 thin films only occurswhen the pH of the solution is above the PZC32,33 of these twooxides. In aqueous solution at these pHs, the surfaces of theoxides become negatively charged and this negative charge iscompensated by an equivalent concentration of cations in thedouble layer of charge, so that the system is electrically neutral.An electrostatic interaction develops between the negative chargeof the surface and the compensating cations. The observedadsorption of the Rh6G cations under these conditions suggeststhat a negative charge on the surface is a requirement for aneffective anchoring of dye molecules of Rh6G. An electrostaticinteraction between the positively charged Rh6G molecules andthe negatively charged surface seems to be the main drivingforce for this adsorption. Similar electrostatic interactions areassumed to account for the adsorption and intercalation of thisdye molecule in laminar silicates,13 although in this case thenegative charge on the surface is not caused by pH effects butis a feature of the structure of these clays (i.e., the negativecharge in the lamellae is compensated by cations in theinterlaminar space).

Another interesting information drawn from the spectra inthis figure is that the relative amount of Rh6G molecules in theform of aggregated species increases with the pH and, therefore,with the net incorporation of dye molecules into the film. Asimilar behavior consisting of an increase in the concentrationof aggregated molecules with the amount of adsorbed moleculeshas been reported for Rh6G incorporated into clays or adsorbedon the surfaces of SiO2 xerogels.13–18

The Rh800 dye molecule presents a slightly different behavior,since incorporation into either TiO2-I or Ta2O5 thin films is little

affected by the pH (cf. Figures 2 and 3). This fact suggests thatthe adsorption mechanisms are not completely equivalent for thetwo rhodamine molecules. The fact that Rh800 is a flat moleculewhile the Rh6G molecule contains a carboxyphenyl group thatcan rotate and tends to orient perpendicular to the xanthene ringseems to be an important factor making the adsorption behaviorof this latter molecule more sensitive to electrostatic interactionswith the surface. From the shape of the bands in Figure 3, it isalso apparent that the Rh800 molecules have a definite tendencyto agglomerate22 when incorporated in the columnar films andthat the relative concentration of these aggregated species slightlyincreases with the pH.

Preillumination and Selective Adsorption of Dyes. Theresults in Figures 4 and 5 show that, at any pH value,preillumination produces a decrease in the number of Rh6Gmolecules that become irreversibly adsorbed into the films. Bycontrast, preillumination does not modify significantly theconcentration of incorporated Rh800 molecules. For this molecule,preillumination only produces a slight increase in the relativeamount of aggregated species, even though the total intensity ofthe adsorption spectrum remains practically constant as comparedwith that of a nonilluminated sample.

The dependence of the adsorption capacity of the thin filmson the preillumination must be related with surface changesinduced by the UV light in both the TiO2 and Ta2O5 surfaces.Since the pioneering work of Wang et al.,1 TiO2 is known tobecome hydrophilic when it is UV illuminated. Recently, wehave shown that Ta2O5 thin films also become hydrophilicwhen they are subjected to UV illumination.8 In this pioneeringwork by Wang et al.,1 it is assumed that TiO2 becomeshydrophilic because the UV light favors the formation ofsurface domains with a high concentration of -OH groups.These ideas have been further supported by the work of otherresearchers.2–5 This situation is schematically described in Figure10a, showing that after UV illumination in air the TiO2 surfaceis covered by extra -OH groups formed by the heterolyticaladsorption of water (a process that is neutral from the point ofview of electrostatic charge). This hypothesis is also supportedby the FT-IR and XPS results shown in the Supporting

Figure 10. Scheme of the proposed model to account for the experimental evidence. Initially, the TiO2 and Ta2O5 surfaces are hydrophobic, containinga percentage of intrinsic hydroxyl groups (OHi in the figure). After the UV-vis treatment, the oxide surfaces are enriched in -OH groups and becomehydrophilic. When these oxides are immersed at pH > PZC, the hydroxyl groups generated by the illumination process are not deprotonated at thestudied pHs and the net negative charge on the surface is smaller than that on a nonilluminated surface.


Information. Recently, Yates et al.6 have claimed that hydro-philicity is favored by the photochemical oxidative removal ofcarbonaceous rests existing on the surfaces of TiO2 materialsexposed to the air. Without discarding some contribution of thislatter effect, our results with the Rh6G molecules can beconsistently explained by assuming the hydroxylation modelschematized in Figure 10. In fact, according to the concepts ofthe PZC in oxides,32,33 when a thin film of TiO2 or Ta2O5 isimmersed in a relatively basic solution (i.e., by basic this meansa pH above the PZC of the oxides), charged -O- groups areformed onto their surfaces by dissociation of the intrinsic -OHgroups. For the columnar thin film materials used here, this processmust occur on both the external and the internal surfaces of thepores of the films. Some heterogeneity in the extension of thisprocess is expected from the amorphous and little orderedcharacter of the columnar thin films used here (cf. Figure 1).However, for the preilluminated films, the formation of -O-

groups at basic pHs will mainly affect the intrinsic hydroxylgroups of the surface and not those generated by the UVillumination. This situation is described in the scheme in Figure10b. According to it, after immersion in the basic solution, theformation of charged-O- groups from the intrinsic-OH groupswill build up a zone of negative charge compensated by cations(M+) forming a double layer of charge. The interfacial space ofcharge would develop less effectively in the zones enriched in-OH groups as a result of the UV preillumination treatment.Implicitly, we assume that these -OH groups generated by theillumination have a different pK value than that of the intrinsic-OH groups of the native surface of TiO2. Adsorption of cationicRh6G molecules will predominantly occur in the more effectivelycharged zones of the surface (i.e., by substituting some of thecompensating M+ cations of the double layer of charge) and willbe less favored in the less charged regions. As an overall effect,the total amount of adsorbed Rh6G molecules would be higherin the nonilluminated thin films than in the preilluminated ones.

In comparison with Rh6G, the adsorption capacity of the thinfilms toward Rh800 is practically not affected by the preillu-mination. Its adsorption was not significantly affected by the pHof the solution. This different behavior suggests that theincorporation of this dye molecule into the films occurs througha different mechanism. We can conjecture that either the differentelectronegativities of the two cationic molecules or their differentstructures are factors conditioning their different adsorptionbehaviors. However, this point is still open, and further studiesand experiments (e.g., by looking at the polarization behaviorof the fluorescence emission of the two adsorbed molecules) arestill required to clarify it.

This different method to become incorporated into thepreilluminated films of the Rh800 and Rh6G dyes is the causeof the preferential adsorption of the latter from solutions containingthe two molecules (cf. Figure 6). The extent of this preferentialadsorption is nearly complete in Ta2O5 and less pronounced inTiO2. The different surface chemistry and PZCs32,33 of the twooxides must be the reason for this different behavior. From apractical point of view, experiments like those described in Figure6 suggest the possibility of using the preillumination ofsemiconductor oxides such as TiO2 or Ta2O5 as a lithographictool for tailoring the synthesis of new photonic materials.

Preillumination and Development of Surface PlasmonStructures. In the preilluminated TiO2-II thin films, a relativelynarrower distribution of silver particles is formed after severalimmersion/illlumination cycles. These particles yield a well-defined surface plasmon resonance structure. In contrast, nowell-defined surface plasmon resonance develops if the film

is not subjected to any preilllumination treatment. The FESEMmicrographs reveal that in the preilluminated film the surfacedensity of particles is higher and its size distribution is narroweras compared with the case of the silver particles in thenonilluminated sample (cf. Figure 8). In addition, a largernumber of silver particles are formed on the preillluminatedfilms. Such a different behavior can be tentatively accounted forby the surface model schematized in Figure 10. From the classicalliterature on the photoactivity of TiO2, it is well-known that thecontrolling step is the capture of the photoholes by the -OHgroups on the oxide surface.45 According to the scheme in Figure10, certain zones of the surfaces of the preilluminated TiO2 filmsare enriched in -OH groups. This assumption agrees with theenhanced photocatalytic activity reported for preilluminatedsamples of TiO2.7 A comparatively higher photoactivity of thesezones of the surface would favor the nucleation of silver and theformation of a relatively higher number of silver particles.Nucleation will be restricted in the nonilluminated samples dueto the relatively lower concentration of -OH surface groups atthe surface. Subsequent immersion/illumination treatments wouldcontribute to the growth of the nuclei initially formed on thepreilluminated samples but will permit a less controlled growthof the nuclei formed in the nonilluminated films. Within thisscheme, it is also likely that silver diffusion is favored on thehydrophilic patches existing in the preilluminated sample, thusfavoring that, after several immersion/UV illumination cycles,clusters more similar in size spread all over the surface.

Conclusions

The previous results and their discussion have revealed a seriesof interesting properties of TiO2 and Ta2O5 thin films that canbe used for the tailored synthesis of new materials. Thedependence on pH of the adsorption efficiency of dye moleculesconfirms the formation of a double layer of charge in the internalsurface of the columnar thin films prepared by glancing angleevaporation. In this regard, our work has pointed out thatconsiderations based on the PZC concepts developed for oxidesin the form of colloids32,33 can be applied to thin films. The openmicrostructure of the TiO2-I and Ta2O5 films used in ourexperiments is a requirement for the detection of this effect, atleast with Rh6G dye molecules.

The dependence of the adsorption degree on the pH is alsoa function of the type of dye molecule and thin film materialconsidered. It is likely that the different structure, size, capacityof aggregation, and charge distribution around the dye moleculeare critical factors that control their different behaviors. However,further studies are still required to properly account for theobserved differences.

The decrease in the capacity of Rh6G adsorption onto thepreilluminated samples has been tentatively explained byassuming the amphiphilic model of Wang et al.1 to accountfor the hydrophilic character of these samples. According tothe scheme in Figure 10, Rh6G will become adsorbed on thepatches where a double layer of charge develops at basic pHs,thus accounting for the observed decrease in adsorption capacity.This model is sustained by the silver photoreduction experiments,whereby a higher concentration of silver particles with a narrowparticle size distribution is formed on the preilluminated sam-ples.

Finally, it is interesting to remark that a critical factor for thecontrol of the infiltration processes or, alternatively, the formationof silver particles is the microstructure of the films. In this regard,it is particularly interesting that dye adsorption experiments areeffective in porous thin films prepared by glancing angle


evaporation. Meanwhile, silver reduction to yield well-definedplasmon structures requires homogeneous samples without largepores. The fact that all the samples are transparent and do notscatter the light opens the way to use this new type of thin filmsfor the tailored synthesis of optical materials based on infiltrationand preillumination methodologies.

Acknowledgment. We thank EU (Contract Nos. 033793 and032583) and MECD (Project NAN2004-09317-C04-01) forfinancial support.

Supporting Information Available: Table with the concentra-tion of the adsorbed dyes as a function of the pH, and XPS and FT-IRspectra of a TiO2-I sample before and after illumination with UV light.This material is available free of charge via the Internet at http://pubs.acs.org.

LA800773V

(45) Gonzalez-Elipe, A. R.; Munuera, G.; Soria, J. J. Chem. Soc., FaradayTrans. 1 1979, 75, 748.


Preillumination of TiO 2 and Ta 2 O 5 Photoactive Thin Films As a Tool to Tailor the Synthesis of...

Documents

Transcript of Preillumination of TiO 2 and Ta 2 O 5 Photoactive Thin Films As a Tool to Tailor the Synthesis of...