Complex Langmuir−Blodgett Films of SiO2 and ZnO Nanoparticles

Langmuir 2008, 24, 12575-12580

12575

Complex Langmuir-Blodgett Films of SiO2 and ZnO Nanoparticles with Advantageous Optical and Photocatalytical Properties ´ braha´m,† O ¨ rs Sepsi,† Erzse´bet Hild,† Didier Cot,‡ Lı´via Nasza´lyi Nagy,†,‡ No´ra A ‡ Andre´ Ayral, and Zolta´n Ho´rvo¨lgyi*,† Department of Physical Chemistry and Materials Science, Centre for Colloid Chemistry, Budapest UniVersity of Technology and Economics, H-1521 Budapest, Hungary, and Institut Europe´en des Membranes, UMR n° 5635 CNRS-ENSCM-UMII, cc047, UniVersite´ Montpellier II, Place Euge`ne Bataillon, F-34095 Montpellier cedex 5, France ReceiVed June 6, 2008. ReVised Manuscript ReceiVed September 1, 2008 Multifunctional Langmuir-Blodgett (LB) films were fabricated on the surface of glass substrates using sol-gel derived ZnO and SiO2 particles. ZnO particles of 6 and 110 nm diameter were synthesized according to the methods of Meulenkamp and Seelig et al. (Meulenkamp, E. A. J. Phys. Chem. B 1998, 102, 5566; Seelig, E. W.; Tang, B.; Yamilov, A.; Cao, H.; Chang, R. P. H. Mater. Chem. Phys. 2003, 80, 257). Silica particles of 37 and 96 nm were prepared by the Sto¨ber method (Sto¨ber, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62). Alternate deposition of monoparticulate Langmuir films of SiO2 and ZnO nanoparticles provided complex (six- and ninelayered) LB films with both antireflective and photocatalytic properties. The LB films were investigated with scanning electron microscopy (morphology and structure) and UV-vis spectroscopy (optical properties and stability). The photocatalytic activity was measured by immersing the UV-irradiated films into an aqueous solution of Methyl Orange and following the photodegradation of the dye by optical spectroscopy. Adding ZnO particles to the silica films slightly lowered the antireflection property but ensured strong photocatalytic activity. Both the photocatalytic activity and antireflection properties were proved to be sensitive to the sequence of the silica and ZnO layers, with optimum properties in the case of nine-layered films with a repeated (SiO2-ZnO-ZnO) structure.

Introduction Nanoparticulate thin films on solid supports can be candidates for various modern applications.4 Layer-by-layer (LBL) fabrication makes possible to build up thin films with precisely designed physical and physicochemical properties. Moreover, preparing the films from particles of different size and/or chemical composition, one can get multifunctional films with unique properties. The Langmuir-Blodgett (LB) technique, which is well-known for preparing molecular5-9 and, recently, nanoparticulate films,10-16 can be accomplished precisely by the LBL manner. Coatings with photocatalytic (self-cleaning and antibacterial) properties can be prepared by depositing TiO2, ZnO, or other semiconductor particles17-19 onto solid substrates. Considering the requirements of enhanced transparency in many applications, the combination of antireflective and photocatalytic properties seems very attractive. Although monolayers and multilayers have been fabricated by the Langmuir-Blodgett technique from silica11-15 and from * To whom correspondence should be addressed. E-mail: zhorvolgyi@ mail.bme.hu. Telephone: +36 1 463 2911. Fax: +36 1 463 3767. † Budapest University of Technology and Economics. ‡ Universite´ Montpellier II.

(1) Meulenkamp, E. A. J. Phys Chem. B 1998, 102, 5566. (2) Seelig, E. W.; Tang, B.; Yamilov, A.; Cao, H.; Chang, R. P. H. Mater. Chem. Phys. 2003, 80, 257. (3) Sto¨ber, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62. (4) Fendler, J. H.; Meldrum, F. C. AdV. Mater. 1995, 7, 607. (5) Tokareva, I.; Hutter, E. J. Am. Chem. Soc. 2004, 126, 15784. (6) Tokareva, I.; Minko, S.; Fendler, J. H.; Hutter, E. J. Am. Chem. Soc. 2004, 126, 15950. (7) Ulman, A. Ultrathin Organic Films from LB to Self Assembly; Academic Press: New York, 1991. (8) Mohai, M.; Kiss, E´.; To´th, A.; Szalma, J.; Berto´ti, I. Surf. Interface Anal. 2002, 32, 772. (9) Marek, T.; Szeles, Cs.; Su¨vegh, K.; Kiss, E´.; Ve´rtes, A.; Lynn, K. G. Langmuir 1999, 15, 8189.

ZnO16 nanoparticles, there is no report in the literature about the fabrication and examination of complex silica and ZnO LB films. The main purpose of the present work is to produce multifunctional coatings with enhanced light transmittance and self-cleaning properties on transparent substrates in which the silica layers provide good antireflection (AR) properties and the ZnO content ensures photocatalytic behavior. The LB technique allows us to design the antireflection (AR) properties of the films by choosing a suitable sequence and number of different particulate layers, in which the relatively large surface area ensures good opportunity for photocatalytic reactions. We have chosen ZnO as the photocatalytic component for several reasons: The particles can be prepared in a relatively broad size range (from a few nanometers up to several hundred nanometers) with narrow size distribution, which is beneficial for the LB fabrication method.1,2 The refractive index of ZnO (either wurtzite type or amorphous) is somewhat lower than that of titania (anatase or rutile), and this is useful when one needs to prepare AR coatings on glass substrates. On the other hand, the photocatalytic utilization of ZnO nanoparticles can be very advantageous in several applications where titania can only perform moderately.20,21 (10) Szekeres, M.; Kamalin, O.; Schoonheydt, R. A.; Wostyn, K.; Clays, K.; Persoons, A.; De´ka´ny, I. J. Mater. Chem. 2002, 12(11), 3268. (11) Szekeres, M.; Kamalin, O.; Grobet, P. G.; Schoonheydt, R. A.; Wostyn, K.; Clays, K.; Persoons, A.; De´ka´ny, I. Colloids Surf., A 2003, 227, 77. (12) Reculusa, S.; Ravaine, S. Appl. Surf. Sci. 2005, 246, 409. (13) Tolnai, Gy.; Nagy, P. M.; Keresztes, Zs.; Lucz, P.; Ka´lma´n, E. Mater. Sci. Forum 2005, 473-474, 279. (14) Dea´k, A.; Sze´kely, I.; Ka´lma´n, E.; Keresztes, Zs.; Kova´cs, A. L.; Ho´rvo¨lgyi, Z. Thin Solid Films 2005, 484, 310. (15) Dea´k, A.; Bancsi, B.; To´th, A. L.; Kova´cs, A. L.; Ho´rvo¨lgyi, Z. Colloids Surf., A. 2006, 278(1-3), 10. (16) Nasza´lyi, L.; Dea´k, A.; Hild, E.; Ayral, A.; Kova´cs, A. L.; Ho´rvo¨lgyi, Z. Thin Solid Films 2006, 515, 2587. (17) Nasza´lyi, L.; Bosc, F.; El Mansouri, A.; van der Lee, A.; Cot, D.; Ho´rvo¨lgyi, Z.; Ayral, A. Sep. Purif. Technol. 2008, 59, 304.

10.1021/la801766y CCC: $40.75  2008 American Chemical Society Published on Web 10/11/2008

12576 Langmuir, Vol. 24, No. 21, 2008

The LB layers were produced from SiO2 and ZnO alcosols, which were synthesized by wet colloidal chemistry routes. The silica particles were prepared by the Sto¨ber method.3 ZnO particles of 6 and 110 nm diameter were synthesized according to the methods of Meulenkamp and Seelig et al.1,2 The mixture of alcosols and chloroform was used to spread the particles at the water-air interface in a film balance. The Langmuir films of SiO2 or ZnO particles were transferred onto solid substrates by the Langmuir-Blodgett technique. The opposite surface charge of silica (zero point of charge (ZPC) at pH ≈ 2) and ZnO (ZPC at pH ≈ 9) at the conditions of film balance experiments (pH ≈ 6) is supposed to ensure good adhesion between the particles of successive layers as it was observed before in preparing selforganized multilayers.22-24 Two-, six-, and nine-layered complex film structures were designed and deposited with a view to achieve a different gradient refractive index inside the films depending on the sequence of silica and ZnO layers. The thickness of the samples was in the optimum range for reflection-reducing (80-200 nm) in the visiblenear-infrared light range.14 The moderate ZnO content and its relatively homogeneous in-depth distribution were supposed to keep the average refractive index of the film near the optimum (∼1.2). The complex films built up from silica and ZnO layers were characterized by scanning electron microscopy (SEM) and optical (UV-vis) spectroscopy. The photocatalytic properties of the complex LB films were studied under UV light on films in contact with an aqueous solution of Methyl Orange. Absorbance of the dye solution and transmittance of the complex films were measured to investigate photodegradation and photocorrosion phenomena.

Experimental Details Materials. For particle preparation, zinc acetate dihydrate (>98% ACS reagent, Aldrich), lithium hydroxide monohydrate (>99% Sigmaultra, Aldrich), absolute ethanol (a.r. >99.7%, Reanal), and diethylene glycol (DEG, purum, Reanal) were used as received. For the film balance experiments, chloroform (ultraresi analyzed, >99.8%, Baker) as the spreading liquid was used in a volume ratio of 1:1 for ZnO sols and 2:1 for silica sols. Distilled water was purified in a Millipore Simplicity 185 system (18.2 MΩ cm). For the preparation of LB films, microscope glass slides and silicon substrates were used. The substrates were thoroughly cleaned prior to use. The glass slides were immersed into chromic-sulfuric acid (Reanal) for 1 h, then rinsed with distilled water and ethanol, and finally dried at room temperature. The silicon wafers were immersed into 2% aqueous HF solution (diluted from 40% HF, a.r., Reanal) for 30 s, then washed with distilled water and ethanol, and finally dried at room temperature. Experimental Methods. Preparation of Alcosols. Nanocrystalline ZnO sols in ethanol with particle diameters of 3-6 nm were prepared as previously described.16 This synthesis is based on Meulenkamp’s method:1 A total of 1.1 g of zinc acetate dihydrate was dissolved in 50 mL of boiling ethanol. After cooling to 0 °C, 0.29 g of lithium hydroxide monohydrate in 50 mL of ethanolic solution was slowly added to it. A transparent colloidal sol was obtained. ZnO particles of 110 ( 18 nm diameter were synthesized in diethylene glycol according to the procedure of Seelig et al.2 First, 2.2 g of zinc acetate dihydrate was rapidly heated to 160 °C in 100 (18) Bosc, F.; Ayral, A.; Guizard, C. J. Membr. Sci. 2005, 265, 13. (19) Szabo´, T.; Ne´meth, J.; De´ka´ny, I. Colloids Surf., A 2004, 230, 23. (20) Sakthivel, S.; Neppolian, B.; Shankar, M. V.; Arabindoo, B.; Palanichamy, M.; Murugesan, V. Sol. Energy Mater. Sol. Cells 2003, 77, 65. (21) Wang, C.; Xu, B. Q.; Wang, X.; Zhao, J. J. Solid State Chem. 2005, 178, 3500. (22) Kotov, N. A.; De´ka´ny, I.; Fendler, J. K. J. Phys. Chem. 1995, 99, 13069. (23) Szu¨cs, A.; Haraszti, T.; De´ka´ny, I.; Fendler, J. K. J. Phys. Chem. B 2001, 105, 10579. (24) Hornok, V.; Erdo˜helyi, A.; De´ka´ny, I. Colloid Polym. Sci. 2006, 284, 611.

Naszályi Nagy et al.

Figure 1. Structure and denomination of complex LB films. The number of the layers in the film is given first. The second symbol refers to the arrangement of the layers: “a” means that the proportion of ZnO (light gray spheres) increases from the substrate toward the air, “b” is for alternate sequences of silica (dark gray spheres) and ZnO, and “c” denotes decreasing ZnO proportion from the substrate toward the air. In samples 6/x and 9/x, 37 nm silica particles and 6 nm ZnO particles were used. The two-layered films are built up from 96 nm silica and 110 nm ZnO particles.

Figure 2. Blank substrate and ZnO free LB films and their denomination. Sample name is composed of “B” and the number of silica layers in the film. In B1, 96 nm silica particles were used, while B2 and B3 were composed of 37 nm silica particles.

mL of DEG. After 3 h, a white slurry formed that was separated from the supernatant in a centrifuge (4200 rpm, 3 h). The procedure continued with a second reaction by rapidly heating up the same amount of starting reagents to 150 °C. Next, 10 mL of the primary supernatant was added to it, and the temperature was allowed to rise up to 160 °C and kept there for 3 h. Finally, the resulting stable, white sol was cooled down and stored at room temperature. The silica particles used in this study had a mean diameter and standard deviation of 37 ( 5 nm and 96 ( 13 nm, respectively.3,15 Images (not presented here) taken by using a JEOL JEM-100 CXII transmission electron microscope (TEM) were used for the determination of particle size. For that, samples were taken from the Langmuir films of particles at moderate surface pressures using Formvar coated copper grids. Preparation of LB Films. The trough of the Wilhelmy film balance was filled with Millipore water, and the water surface was cleaned. The substrate was immersed into the water prior to the spreading of the particles. Slowly decreasing the available area, a continuous Langmuir film was formed on the water surface. Withdrawal of the substrate was launched at ∼50% of the collapse pressure. The deposited layers were dried at ambient temperature. In the case of multilayered LB films, the procedure was repeated 1-8 times. The obtained films were finally dried at 105 °C in air. The complex LB films were denominated showing the total number of layers and the sequence of layers in the film (Figure 1): the ZnO free (blank) samples were called “BX”, where “X” is the number of the silica layers in the film (Figure 2). Structural Characterization of LB Films. Information about the LB film structure and morphology was obtained from the images taken with a Hitachi S-4500 scanning electron microscope. For the SEM investigations, films deposited on either glass slides or silicon wafers were used.

Complex LB Films of SiO2 and ZnO Nanoparticles

Langmuir, Vol. 24, No. 21, 2008 12577

Table 1. Calculated Total ZnO Mass and Surface Area in Complex LB Films Composed of 6 nm (in 6/x and 9/x) or 110 nm (in 2/x) Diameter ZnO Particlesa 6/x 9/x 2/x a

ZnO mass (kg)

ZnO surface (cm2)

2 × 10-9 4 × 10-9 1.2 × 10-8

7.08 14.16 4.12

Particles were assumed to be spherical and close-packed.

Optical Characterization of the LB Films. An Agilent 8453 UV-vis spectrophotometer was used to collect the transmission spectra of the LB films in the wavelength range between 300 and 1100 nm. The reference was air in every case. In the transparent range (400-1100 nm), the optical parameters of the film were determined from transmittance spectra by means of an optical model, which takes in-depth inhomogeneity into account.25 In this model, the in-depth variation of the refractive index was supposed to follow a smooth and monotonous function that could be approximated with the geometrical optical (GO) profile:25

n(z) )

n(0) (1 + z ⁄ ξ)2

(1)

where z is the distance from the film-air interface, n(0) is the surface refractive index, and ξ is the inhomogeneity parameter of the film. The transmittance and reflectance of the inhomogeneous film can be expressed by the optical admittance, which is obtained in the closed form in the case of GO profiles.26,27 Using this solution to simulate the transmittance spectrum of a thick transparent substrate covered by identical nonabsorbing layers on both sides, we get the optical parameters of the films by an appropriate fitting procedure. Our fitting procedure resulted in the film thickness d, the average refractive index of the film Nav, and the inhomogeneity factor g defined as

g)

 N(0) N(d)

Figure 3. SEM pictures of monolayered LB films prepared with (a) 6 nm ZnO particles and (b) 37 nm silica particles used in complex LB films 6/x and 9/x.

(2)

UV-vis spectroscopy was also used to follow the changes induced in the film composition during photocatalytic use. For that, the absorption edge of ZnO appearing near 360 nm was used where the absorption was proportional to the ZnO content of the film. Transmittance spectra were taken first after film preparation, a second time when the photocatalytic test was finished, and a third time after heat treatment of the sample at 500 °C for 1 h in air. Characterization of the PhotoactiVity of LB Films. The photocatalytic activity of complex LB films was investigated during the degradation of Methyl Orange (MO) (spec., Reanal) in 5.5 mg/L aqueous solution illuminated by an UV lamp (Philips, CleoHPA 400 W, max intensity at 375 nm). The supported catalyst film (of ∼6 cm2 area) was put in a thermostatted beaker. A total of 10 mL of the MO solution was poured into the beaker and kept at 25 °C under magnetic stirring during the experiment. The absorbance of the organic dye in solution was measured at 464 nm as a function of the irradiation time. Before each absorbance measurement, the volume of the MO solution was readjusted to 10 mL with distilled water because there is slight evaporation of the solvent under the UV lamp. The photocatalytic activity of the layers was evaluated in terms of the surface and mass of ZnO in each sample. The calculated values are shown in Table 1 (contact areas between spheres were neglected). These were obtained assuming ideal spherical shape of the particles and hexagonal close-packed LB structure. Small ZnO ¨ .; A ´ braha´m, N.; Ho´rvo¨lgyi, Z. (25) Hild, E.; Dea´k, A.; Nasza´lyi, L.; Sepsi, O J. Opt. A: Pure Appl. Opt 2007, 9, 920. (26) Jacobsson R. In Progress in optics; Wolf, E., Ed.; North Holland: Amsterdam, 1965; Vol. 5, pp 247-286. (27) Hild, E.; Evans, M. W. J. Appl. Phys. 1986, 59, 1822.

Figure 4. SEM pictures of (a) 6/a, (b) 6/b, (c) 6/c, and (d) 2/a complex LB films.

particles were found to be compact.16 The mass of the 110 nm diameter particles was estimated from density measurements with a pycnometer; the surface area was evaluated from N2 adsorption/ desorption isotherms by the Brunauer-Emmett-Teller (BET) method.28 The photostability of the LB films was investigated keeping the samples under UV light for 3 h both in air and in water.

Results and Discussion Structure of the LB Films. SEM pictures of LB monolayers of the ZnO and SiO2 particles used in samples 6/x and 9/x are shown in Figure 3. Though they form a continuous and homogeneous layer, the tiny ZnO particles seem to form aggregates in the LB layer. The larger silica particles constitute a well-defined, close-packed Langmuir-Blodgett film. The presence of the ZnO layers in the complex films is not directly observable between the silica layers at the actual magnification, but their effect on the close-packed silica structure is visible. For 6/a, the arrangement of silica particles is undisturbed, since the ZnO layers are deposited on the top (Figure 4a). The arrangement of silica particles is apparently perturbed, however, by the ZnO layers deposited between and under silica layers in the case of layers 6/b and 6/c (Figure 4b,c). Figure 4d shows the surface view of sample 2/a with larger silica and ZnO particles. The 110 nm ZnO particles form an irregular top layer on the close-packed layer of 96 nm silica particles. Optical Properties of LB Films. Transmittance spectra of six-layered and nine-layered samples are given in Figure 5a and b. Antireflectivity originating from the special structure of the silica LB films is maintained in these complex films while the position of maximum transmittance shifts toward higher wavelengths. (28) Brunauer, S.; Emmett, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60, 309.

12578 Langmuir, Vol. 24, No. 21, 2008

Naszályi Nagy et al. Table 2. Calculated Thickness d, Average Refractive Index Nav, and Inhomogeneity Factor g of Complex LB Filmsa no. of sample

d (nm)

Nav

g

χ2 (10-7)

6/a 6/b 6/c 9/a 9/b 9/c 2/a 2/c

116 119 115 132 127 145 165 175

1.300 1.329 1.341 1.346 1.344 1.311 1.371 1.356

1.104 1.054 1.006 1.061 1.034 1.001 1.057 1.044

6.7 2.3 2.8 1.1 1.8 0.5 1.1 0.5

a

χ2 is the sum of deviation squares.

Transmittance exceeds 95% in almost all the visible and NIR ranges. The UV light, however, is absorbed by the ZnO content of the samples that can be advantageous in applications coupled with UV protection. The absorption intensity depends on both the number of ZnO layers in the sample and the size of the ZnO particles. In Figure 5c, the transmittance spectra of two-layered samples are shown. In this case, the effect of ZnO particles is dominant in the transmittance spectrum. The transmission exceeds the transmission of substrate only above 600 nm, and a maximum is reached at about 1000 nm. An upper bound for the thickness of the films can be estimated as the sum of the particle diameters in the consecutive layers. Therefore, the thickness of samples 6/x, 9/x, and 2/x should not exceed 129, 147, and 206 nm, respectively. Fitting the UV-vis spectra with the model of Hild et al.25 gave the film parameters summarized in Table 2. As a matter of fact, the thickness of multilayered LB films was always lower than the sum of the diameter of the constituting particles. This is in good agreement with previous observations for silica LB films.29 Thus, the subsequent layers interpenetrate to some extent. Hence, the regular arrangement of the particles in the films shown in Figure 1 is an abstraction. Otherwise, interlocking layers allow a more smooth variation of the refractive index, especially for large differences between the sizes of the

silica and ZnO particles. The presence of a high refractive index ZnO (2.0130) increased significantly the average refractive index compared to the one of a single silica LB film (1.22).29 The average refractive index of the films is between 1.30 and 1.37 (highest values were observed for two-layered LB films containing large ZnO particles with more important mass). The inhomogeneity factor being always higher than 1 indicates a refractive index that increases in the layer from the layer-substrate interface to the layer-air interface (unfavorable for antireflectivity). A lower value of g means a smaller difference between the refractive index values at the two interfaces. According to our expectations, structure “c” (decreasing ZnO content from the substrate to the air) yields the highest transmittance with the lowest refractive index and lowest g values. In the last column of Table 1, χ2 is the mean-square deviation of the fitting. The values of g show that the gradient refractive index could readily be established with three layers of 6 nm diameter ZnO integrated in the silica structure. However, at the best disposition of 6 layers of tiny ZnO particles in nine-layered complex LB films (sample 9/c), the value of the inhomogeneity factor could be minimized. Twolayered samples gave rise to a high inhomogeneity factor. These results indicate that, in complex LB films built up from different types of particles, a greater difference in the particle sizes contributes to a finer tuning of the refractive index through a better interlocking of the layers. Photocatalytic Activity and Capacity of the LB Films. The results of photocatalytic investigations are summarized in Figure 6, where the specific degradation values (total degradation/LB film area) are plotted as function of the irradiation time. Methyl Orange is a negatively charged organic dye, and therefore, it is expected to adsorb on the positively charged surface of zinc oxide below pH 9, facilitating the degradation of the dye molecules. Slight degradation of Methyl Orange in 5.5 mg/L aqueous solution is observed under UV light (3.7% of loss after 3 h). In the presence of B2, total degradation rises to nearly 6%. Yet, these values are negligible considering the total degradation of 72% (in 3 h) for sample 6/a, 90% (in 2 h) for sample 9/a, and finally 80% (in 3 h) for sample 2/a. For samples 6/x and 9/x, the position of the tiny ZnO particles has a definite effect on the photodegradation activity. The degradation values after 120 min (D) for each film are summarized in Figure 7. The photocatalytic performance is better for those complex films in which the ZnO layers are situated in the outer part of the films (2/a, 6/a, 9/a). Inner ZnO layers are probably less accessible for the dye solution. One can also observe that the degree of degradation increases with increasing number of ZnO layers but not with increasing ZnO mass (see Table 2), showing the importance of the surface area of the photoactive component in the degradation.

(29) Dea´k, A.; Hild, E.; Kova´cs, A. L.; Ho´rvo¨lgyi, Z. Mater. Sci. Forum 2007, 537-538, 329.

(30) Norton, D. P.; Heo, Y. W.; Ivill, M. P.; Ip, K.; Pearton, S. J.; Chisholm, M. F.; Steiner, T. Mater. Today 2004, 7, 34.

Figure 5. Transmittance spectra of (a) six-layered, (b) nine-layered, and (c) two-layered complex LB films and blank samples.

Complex LB Films of SiO2 and ZnO Nanoparticles

Figure 6. Photodegradation of Methyl Orange in 5.5 mg/L aqueous solution in contact with (a) six-layered, (b) nine-layered, and (c) twolayered complex LB films and blank samples.

Langmuir, Vol. 24, No. 21, 2008 12579

Figure 8. Photostability experiments. Transmittance spectra of (a) sixlayered, (b) nine-layered, and (c) two-layered complex LB films after preparation (e.g., 6/a), a second time when the photocatalytic test was finished (e.g., 6/aM), and a third time after heat treatment of the film at 500 °C for 1 h (e.g., 6/aMD). Table 3. Transmittance Values at the Band Edge 360 nm (Local Minimum on the Derivative Spectra) after (1) Preparation, (2) Photodegradation Test, and (3) Heat Treatment at 500 °C 1 2 3

Figure 7. Comparison of the photocatalytic activity of complex LB films after 120 min of UV irradiation (D) in 5.5 mg/L Methyl Orange aqueous solution. ZnO mass and surface area are given in Table 2 for each sample.

Moreover, it was found that an increase in the surface area of ZnO via the number of LB layers did not always produce a proportional increase in photoactivity. This phenomenon is probably due to the screening effect of outer ZnO layers, which prevents inner ZnO layers (for 9/a) from being equally effective. Finally, the D values for films containing 110 nm diameter particles seem to be too high as compared to those containing tiny ZnO particulate films considering the higher theoretical

6/a

6/b

6/c

9/a

9/b

9/c

2/a

2/c

74% 89% 91%

74% 88% 98%

81% 85% 90%

68% 76% 76%

75% 78% 79%

69% 79% 78%

45% 53% 54%

46% 52% 53%

surface area of ZnO in the latter (Table 1). We assume that either the surface area of large particles was underestimated (more open porosity may exist there) or that of the tiny particles was overestimated. Since the surface area of large ZnO particles was evaluated from reliable N2 adsorption/desorption measurements, it is more plausible to think that aggregation of 3 nm diameter particles causes a decrease in the available catalyst surface for related films (the contact areas between spheres are not negligible). On the whole, 9/a was found to be the most active for the photodegradation of Methyl Orange; 9/b and 9/c show, at the same time, a better transparency in the visible region. Thus, 9/b is the best candidate for optimum AR and photocatalytical behavior in the present study. We also investigated the effect of the photocatalytic test on the film composition (ZnO content). UV-vis spectra of the samples were recorded before and after the photocatalytic test and finally after thermal treatment at 500 °C (Figure 8). The absorption edge of ZnO (360 nm) is notably diminished after the

12580 Langmuir, Vol. 24, No. 21, 2008

Naszályi Nagy et al.

lifetime (photocatalytic capacity) of these multifunctional coatings, but on the other hand it abates their antireflectivity. Photocatalytic experiments in this study were carried out at the native pH of the Methyl Orange solution (6.5), but we suppose, on the basis of other research works,32 that addition of a base (pH ≈ 8) would prevent the dissolution of the catalyst without deteriorating the adsorption of the dye (the ZPC of ZnO is near pH 9).

Conclusions

Figure 9. Photocatalytic capacity experiments. Photodegradation experiments repeated two or three times on samples (a) 6/a (three layers of 37 nm silica particles and three layers of 3 nm ZnO particles) and (b) 2/a (one layer of 96 nm silica particles and one layer of 110 nm ZnO particles).

test for each sample (Table 3). It implies a significant amount of ZnO dissolved in water (pH of the Methyl Orange solution was ∼6.5,31 and the ZPC of ZnO surpace is pH ∼ 9). Additional experiments confirmed our hypothesis: 6/a samples were irradiated for 3 h in air and in pure distilled water. We could see a significant absorbance decrease only for samples which were irradiated in water. Due to the film alteration, a 10-14% increase in transmittance at 360 nm was observed for all complex films (loss of ZnO). It means, however, the quasi-total decomposition of the ZnO content for samples 6/x. This observation was further examined by repeating the photocatalytic test on samples 6/a and 2/a (Figure 9). It is clear that the photocatalytic activity of sample 6/a is decreased after only 3 h of use. The activity of sample 2/a is unchanged up to 6 h of testing, but thereafter it is reduced. These observations lead us to conclude that the increasing mass of ZnO in complex LB films has a positive effect on the (31) Dome´nech, J.; Pietro, A. J. Phys. Chem. 1986, 90, 1123.

The preparation and investigation of multifunctional LB films composed of ZnO and SiO2 layers with different nanoparticle sizes are reported in this study. The LB technique has been found to be an efficient method for the tailoring of complex layer structures: small (3-6 nm) ZnO nanoparticles could be integrated in antireflective LB films of SiO2 particles without deterioration of the film transparency in the visible region. All complex films have a gradient refractive index according to the optical analysis. Films with the lowest inhomogeneity factor exhibit the best AR effect. Such structures ensure the most homogeneous mixing of the different types of particles with decreasing ZnO amounts from the substrate toward the air. The presence of ZnO provides photocatalytic activity for the complex film. The degradation of Methyl Orange in aqueous solution increases with a higher available surface of ZnO in the film. The lifetime of the samples could be extended by increasing the ZnO content in the film. The ZnO content, however, cannot be raised exceedingly, because it is unfavorable for the antireflection properties. An optimum should be defined for each application depending on its requirements. In our study, aiming for both the best transparency and good photocatalytical activity, sample 9/b is considered optimal. We also conclude that the photostability of the LB films should be improved. Therefore, our future work will be focused on this topic. Acknowledgment. This work was supported by the Hungarian National Scientific Foundation for Research (OTKA T 049156). It was carried out in the scope of COST Action 540 and a cooperation program, BALATON (Te´T F-20/04), sponsored by the Hungarian Ministry of Science and Technology and the French ¨ veges Program Embassy in Budapest. Thanks to KPI for the O accorded to Z.H. and to the French Ministry of Foreign Affairs for the mobility grant of L.N.N. The authors are very indebted to Dr. Attila Kova´cs for TEM, to Prof. Miklo´s Kubinyi, and to Dr. Pe´ter Baranyai for spectroscopy measurements. LA801766Y (32) Chen, C. C. J. Mol. Catal. A: Chem. 2006, 264, 82.