Functionalization of ceramic tile surface by sol–gel technique

Journal of Colloid and Interface Science 334 (2009) 195–201

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science

www.elsevier .com/locate / jc is

Functionalization of ceramic tile surface by sol–gel technique

F. Bondioli a,*, R. Taurino a, A.M. Ferrari b

a Università di Modena e Reggio Emilia, Dipartimento di Ingegneria dei Materiali e dell’Ambiente, Via Vignolese 905, 41100 Modena, Italyb Università di Modena e Reggio Emilia, Dipartimento di Scienze e Metodi dell’Ingegneria, Via Amendola 2, 42100 Reggio Emilia, Italy

a r t i c l e i n f o

Article history:Received 19 December 2008Accepted 23 February 2009Available online 7 April 2009

Keywords:FunctionalizationTileTitaniaPhotocatalicitySol–gelContact angleScratch resistanceSelf-cleaning

0021-9797/$ - see front matter � 2009 Elsevier Inc. Adoi:10.1016/j.jcis.2009.02.054

* Corresponding author. Fax: +39 059 205 6243.E-mail address: [email protected] (F. B

a b s t r a c t

The aim of this investigation was the surface functionalization of industrial ceramic tiles by sol–gel tech-nique to improve at the same time the cleanability of unglazed surfaces. This objective was pursuedthrough the design and preparation of nanostructured coating that was deposited on polished unglazedtiles by air-brushing. In particular TiO2–SiO2 binary film with 1, 2 or 5 wt% of titania were prepared byusing tetraethoxysilane and titania nanoparticles as precursors. The obtained films were characterizedby scratch tests to verify the adhesion of the coatings to the polished tiles. To mainly evaluate the effectof the thermal treatment (temperature range 100–600 �C) on the photocatalicity of the coatings, the filmswere studied under UV exposure by contact angle measurements and cleanability test. Particular atten-tion has been paid to preserve the aesthetical aspect of the final product and the obtained hue variationwas evaluated by means of UV–visible spectroscopy and colorimetric analysis.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction

The unglazed fine porcelainized stoneware tile, also called por-celainized stoneware or grés, is a product used both for internaland external applications in building field. Low-porosity, highmechanical, abrasion, chemical and stain resistance make thismaterial ideal to flooring areas with elevated public use [1,2]. Itshigh technical properties are mainly due to an extremely sinteredbody composed of different crystalline phases (anorthite, mulliteand quartz) immersed in a vitreous matrix [3,4]. In the ceramictiles industrial field, the porcelainized tiles have become moreand more important with regard to its spread from very few mar-ket shares limited as to their application fields to more and morediversified ones; the result has been a clear increase in productionvolumes. This product, which was formerly considered only from atechnical standpoint, nowadays shows high aesthetical potentiali-ties allowing its use for over-refined purposes. In particular, thehigh surface hardness of the unglazed porcelainized tiles allowsmirror-polished surfaces that give to this product a high aestheti-cal quality. The main problem is that this material, even if it has avery low open porosity (about 0.1% as to the absorbed water, and0.5% according to the mercury porosimetry), has an internal closedporosity of around 6%, with pore sizes range from 1 to 10 lm. Thisporosity appears during the polishing phase, where about 0.5–1 mm of the superficial layer is removed, causing a superficial

ll rights reserved.

ondioli).

microporosity that increases the tendency to the product dirtiness.There are many ways of intervention (some of them have alreadybeen widely adopted) as for exemplum using transparent poly-meric layers, however the mechanical properties and chemicaldurability of these coatings are often very poor [5].

The development of new easy-to-clean or even self-cleaningsurfaces has recently been under the focus of nanotechnology, i.e.by investigating different surface structures. Some models of self-cleaning surfaces are available in nature, such as lotus plant leaves[5] and the wings of insects [6]. Recently to improve surface clean-ability properties [7–9] the photocatalicity of TiO2 nanoparticleshas been used. For exemplum, the production of photocatalyticallyactive building materials, allows to obtain self-cleaning and self-sterilizing surface that, moreover, might degrade several organiccontaminants in the surrounding environment by UV radiationactivation [10].

Moreover, one of the main interesting properties of nanoparti-cles, characterized by a mean diameter below the light wavelength,is their transparency if applied on a substrate or dispersed in a ma-trix. To take advantage of this property, the aim of this study was ori-ented in design and synthesis of silica (SiO2) and titania (TiO2)multifunctional inorganic coating, to apply on the polished unglazedsurfaces. The goal was the obtainment of a transparent physical bar-rier to enhance the cleanability of tile exploiting at the same time thephotocatalicity of titania in order to obtain a multifunctional coat-ing. This objective was pursued through the design and preparationof nanostructured sol–gel coating that was deposited on polishedunglazed tiles by air-brushing. The obtained films were fully charac-

196 F. Bondioli et al. / Journal of Colloid and Interface Science 334 (2009) 195–201

terized to mainly evaluate the effect of the thermal treatment (tem-perature range 100–600 �C) on scratch resistance, cleanability, andphotocatalicity of the coatings. Particular attention has been paidto preserve the aesthetical aspect of the final product and the ob-tained hue variation was evaluated by means of UV–visible spectros-copy and colorimetric analysis.

2. Experimental section

2.1. Coating preparation

Silica sols were prepared and referred as sol X_Y where X isthe final concentration of titania oxide with respect to the silicacontent and Y is the treatment temperature. Sols were preparedby mixing tetraethylorthosilicate (TEOS, Sigma Aldrich), and crys-talline anatase nanoparticles (TiO2, mean particle size 18 nm)(NAMA 07-P704, kindly provided by Colorobbia Italia, I) sus-pended in ethanol (EtOH) at 5 wt%, in different weight ratio, toobtain a theoretical final silica/titania ratio of 99/1, 98/2 and95/5 (wt/wt). Differently from the work of other authors thatinvestigated titania-rich films, [11,12] these compositions werechosen to evaluate the effect of small amount of titania nanopar-ticles on silica films.

TEOS was dissolved in ethanol (Sigma Aldrich) and then titaniasuspension was slowly added to obtain a final mixture concentra-tion (TEOS + TiO2) of about 5 wt%.

Table 1 lists in detail the prepared compostiions (the final metaloxide content was calculated assuming the completion of thesol–gel reactions of the metal alkoxide precursors) and the heattemperature of the samples.

The application of sols were performed by air-brushing [13] onpolished porcelainized stoneware tile (kindly provided by MarazziGroup, I). This deposition technique was chosen, on a laboratoryscale, taking into account the industrial applicability and the pos-sible technological solutions necessary to implement these surfacetreatments in the industrial traditional process. During the air-brushing application, the substrates were kept at a distance ofabout 14 cm from the air-brush applicator. Each sample wascoated with 0.1 ml/cm2 of solution. After coating, the surfaces wereair-dried for about 1 min with a hairdryer, then dried at room tem-perature for 24 hours and finally heat-treated at 100, 200, 400 and600 �C for 5 min.

2.2. Coating characterizations

In order to verify the adhesion of the coatings to the polishedtiles, scratch tests (Micro-Combi tester) with linearly increasing

Table 1Chemical composition of the prepared silica-based sols, including heat temperature.

Sample ID SiO2/TiO2 (wt/wt) Heat temperature (�C)

0_100 100/0 1000_200 100/0 2000_400 100/0 4000_600 100/0 6001_100 99/1 1001_200 99/1 2001_400 99/1 4001_600 99/1 6002_100 98/2 1002_200 98/2 2002_400 98/2 4002_600 98/2 6005_100 95/5 1005_200 95/5 2005_400 95/5 4005_600 95/5 600

load (0.1–30 N, scratch speed of 1 mm/min) were performed onthe samples using a Rockwell indenter with spherical tip, 100 lmradius. At least, three scratches were performed on each coating,with the minimum distance between two scratches set at 4 mmto achieve results representative of the average response overgreater surfaces. The critical loads Lc1 (first crack) and Lc2 (edgespallation) were determined by optical microscopy.

The effect of the coating on the tile color was determined byperforming color measurements on both uncoated and coated tilesby UV–vis spectroscopy (model Lambda 19, Perkin Elmer) usingthe CIELab method in order to obtain L*, a* and b* values [14].The method defines a color through three parameters, L*, a* andb*, measuring brightness, red/green and yellow/blue color intensi-ties, respectively [15]. The method allows, moreover, to define acolor difference as DE*, based on the relationship:

DE� ¼ ðDL�Þ2 þ ðDa�Þ2 þ ðDb�Þ2h i1

2 ð1Þ

where DL*, Da* and Db* measure the differences in luminosity andin chromaticity between two color. In this way the hue variationdue to the coating step were determined.

To evaluate the effect of the coating on the tile appearance, thegloss of the samples was measured by a Novo Gloss Trio apparatus;measurements were performed by using 60� as standard geometry.

The microstructure of the samples was investigated by scanningelectron microscopy (SEM) using a XL 30 instrument (Philips) overgold-coated samples.

In order to qualitatively examine the crystalline phases devel-oped in the coatings as a function of the temperature treatment,powdered samples were obtained, in the same conditions of thecoatings, using a refractory crucible. The X-ray diffraction mea-surements (XRD) were carried out using a conventional Bragg–Brentano diffractometer (X’PERT PRO, Philips Research Laborato-ries) with Ni-filtered Cu Ka radiation. The patterns were recordedon the grounded samples (<25 lm in size) in the 5–80� 2 h range atroom temperature, with a scanning rate of 0.001�/s and a step sizeof 0.02�. Fourier Transform Infrared spectroscopy (FT-IR analysis)was performed on the powders in the attenuated total reflectancemode with an Avatar 330 spectrometer (Thermo Nicolet, Ger-many). A minimum of 32 scans with a resolution of 4cm�1 wasused.

To evaluate the photocatalytic activity of the obtained coating,contact angle measurements and cleanability test were performed.Static water contact angles (CA) were measured by the sessile dropmethod [16] using a conventional drop shape technique OCA 20apparatus (DataPhysics Instrument GmbH, Filderstadt, Germany).To avoid any surface contamination, all specimens were rinsed intetrahydrofuran, THF, and accurately air-dried just before mea-surement. Static CAs were determined on the basis of at least 10measurements and a drop volume of 15 ll. All CA measurementswere carried out at ambient conditions, under UV irradiation. UVlight with wavelength range 325–390 nm and light intensity5.5 mW/cm2 was used as light source [12]. Determination of CAwas based on the Young–Laplace equation and performed every30 s for 10min. The result was the mean of the drop on five repli-cate samples.

Photo-degradation of methylene blue solution (500 ppm) wasused to assess the photocatalytic activity of the coating. The coatedtiles were treated with the methylene blue solution, briefly pressedwith paper and then irradiated with the UV lamp described above.The photocatalytic decomposition of blue was monitored, every30min for 240min, by color measurement using a portable UVspectrophotometer (Color Quality, COROB, I) equipped with Stan-dard Illuminant D65 and expressed as color variation (DE*) withrespect to the initial color of the tile.

F. Bondioli et al. / Journal of Colloid and Interface Science 334 (2009) 195–201 197

3. Results and discussion

To verify the adhesion of the coatings to the polished tiles,scratch tests with linearly increasing load were performed on thesamples [17]. In Table 2 the first (coating failure) and second (coat-

Table 2First critical load (LC1), second critical load (LC2) and friction coefficients of theobtained films.

Sample ID LC1 (N) LC2 (N) Friction coefficient

Uncoated tile 5.7 ± 0.17 13.4 ± 1.0 0.059 ± 0.017

0_100 6.1 ± 0.2 12.4 ± 0.6 0.064 ± 0.0140_200 6.2 ± 0.7 11.9 ± 1.0 0.067 ± 0.0350_400 6.8 ± 0.7 13.1 ± 0.8 0.060 ± 0.0150_600 7.6 ± 0.6 12.4 ± 0.2 0.050 ± 0.018

1_100 8.5 ± 0.3 15.2 ± 0.1 0.063 ± 0.0121_200 8.1 ± 0.2 15.8 ± 0.7 0.077 ± 0.0191_400 7.8 ± 0.6 15.2 ± 0.1 0.050 ± 0.0241_600 7.6 ± 0.1 15.6 ± 0.7 0.045 ± 0.009

2_100 8.2 ± 0.1 15.9 ± 0.8 0.062 ± 0.0182_200 7.5 ± 0.1 15.8 ± 0.2 0.049 ± 0.0172_400 7.7 ± 0.5 15.2 ± 0.1 0.055 ± 0.0252_600 7.7 ± 0.3 15.8 ± 0.3 0.050 ± 0.018

5_100 8.2 ± 0.4 15.9 ± 0.5 0.069 ± 0.0135_200 7.7 ± 0.1 16.2 ± 0.1 0.102 ± 0.0265_400 9.1 ± 0.2 16.3 ± 0.3 0.058 ± 0.0125_600 9.2 ± 0.3 16.1 ± 0.5 0.041 ± 0.013

Fig. 1. Penetration depth (Pd) in a progressive load

Fig. 2. Optical image of the samples a

ing detachment) critical load values are reported for the obtainedsamples. For the sample 0_Y, independently from the temperatureof the thermal treatment, the normal load needed to remove thecoating from the substrate is in the range of 12–13 N while, addingtitania to the coatings, this value increases in the range 15–16 N.Moreover, a consistent improvement of penetration resistance byincreasing titania content can be noted in terms of penetrationdepth (Pd). In Fig. 1 the behaviour of the X_600 samples chosenas representative are reported: the slope of the penetration curvesdecreases progressively with the increase of titania content. Thecoatings with a titania content lower than 5 wt% do not show anincrease of their adhesion strength with the increase of the tem-perature of the thermal treatment. Instead, an increase of load atwhich the scratch track appears was obtained with titania contentof 5 wt%, as showed by the optical images reported in Fig. 2. In par-ticular, the sample 5_600 presents the higher critical loads becauseit failures at 9.2 N and it is removed at 16.1 N (Table 2).

Thus the obtained results show that the film scratch resistancegenerally increases as both titania content and temperature are in-creased. This behaviour could be attributed to an high structuralintegrity of the coatings related to the temperature of the thermaltreatment and to an increase of the final E modulus by increasingtitania content.

Finally, the results obtained on the uncoated tiles show that thedeposition of the nanostructured coatings allows to a increase ofthe scratch resistance as underlined by the increase of the criticalloads. Moreover, Table 2 shows the average values of friction coef-ficient to improve the accuracy in the interpretation of the in-

scratch test of the samples treated at 600 �C.

fter the progressive scratch test.

Table 3DE* and gloss values of the prepared coatings.

Sample ID DE* Gloss (GU)

Polished grès / 78.0 ± 0.4

0_100 1.0 57.0 ± 2.10_200 0.8 41.3 ± 1.10_400 0.7 40.4 ± 1.00_600 1.1 44.9 ± 1.8

1_100 0.7 75.7 ± 0.91_200 0.6 75.2 ± 0.61_400 0.3 72.2 ± 0.71_600 0.6 70.2 ± 0.8

2_100 0.3 76.0 ± 0.72_200 0.9 71.8 ± 0.52_400 0.7 70.2 ± 0.92_600 0.7 74.4 ± 0.3

5_100 0.8 79.9 ± 0.35_200 0.9 69.8 ± 0.85_400 0.8 66.6 ± 0.25_600 1.0 63.9 ± 1.2

198 F. Bondioli et al. / Journal of Colloid and Interface Science 334 (2009) 195–201

volved mechanism. It is possible to note that the friction coefficientis inversely correlated to the scratch resistance of these thin films.For exemplum, the films 5_600 exhibits the lower friction coeffi-cient. As reported [17], the effect of friction is to add a compressivestress to the front edge of the contact and then intensify the tensilestress at the back edge with fracture events.

In Table 3, DE* between the coated tile surfaces and the un-treated tile are reported. The Table clearly shows that in all thestudied cases the value is <1, indicating that there is not a visiblehue variation due to the applied coatings. This result underlines

Fig. 3. SEM micrographs of uncoated (a) a

that the titania nanoparticles are well dispersed in the coatingsthat thus are effectively transparent.

In Table 3, the gloss measured on the prepared samples is alsoreported. This parameter that is correlated to the surface rough-ness of the samples is directly correlated to the temperature ofthe thermal treatment and decreases as the temperature is in-creased. The behavior is also related to the titania content andthe gloss decreases as the titania nanoparticles are increased inthe coatings.

In Fig. 3, the SEM images of uncoated and 5_600 sample sur-faces. The images of the polished uncoated tile (Fig. 3a) evidencethe porosity of the sample, with pore sizes range from 1 to10 lm, appeared during the polishing phase. Instead, the imagesof the 5_600 sample, chosen as representative (Fig. 3b), show a de-crease of the porosity due to the nanostructured coating. The filmcoats the pore with some exception for the bigger one where alsosome crack, probably due to the drying process, are visible. By pro-ducing a fracture in the coating (Fig. 3c) it is possible to measurethe film thickness that is about 2–3 lm. Nevertheless, SEM mea-surements did not allow a definite evaluation of the thickness of0_600 sample that is presumably lower than 0.5–1 lm. The figurealso shows the porosity decrease and the roughness of the coatedsurface that is slightly higher of that of uncoated tile in agreementwith the gloss data.

To understand these behaviors and verify the structure andmicrostructure of the as obtained coatings, several characteriza-tions were performed on powdered samples obtained in the sameconditions utilized for the coatings.

In Fig. 4 the XRD patterns of 0_600 and 5_600 powders are re-ported. The Figure clearly shows that the silica precursor TEOS isnot crystallized also at the higher temperature used, developingan amorphous matrix in which the anatase nanoparticles are dis-

nd 5_600 (b and c) sample surfaces.

Fig. 4. XRD pattern of 0_600 (a) and 5_600 (b) powders. The bars correspond to anatase (ICDD # 00-001-0562).

F. Bondioli et al. / Journal of Colloid and Interface Science 334 (2009) 195–201 199

persed. The main peak of anatase (ICDD # 00-001-0562), in fact, isidentifiable at around 25.3� 2h.

Fig. 5. IR spectra of 5 wt% titania film annealed at 200 �C (a) a

In Fig. 5a, the IR spectra of the powders with 5 wt% of anataseannealed at 200 �C, chosen as representative. The main band at

nd as a function of annealing temperature (b, particular).

Fig. 6. Contact angle of 5 wt% titania films as a function of annealing temperature and UV irradiation time: (a) uncoated grés, (b) 5_100, (c) 5_200, (d) 5_400, (e) 5_600.

Fig. 7. Color variation (DE*) of 5 wt% titania films as a function of annealing temperature and UV irradiation time. (a) uncoated grés, (b) 5_100, (c) 5_200, (d) 5_400, (e) 5_600.

200 F. Bondioli et al. / Journal of Colloid and Interface Science 334 (2009) 195–201

1050 cm�1 and the 800 cm�1 peak are assigned to Si–O–R stretch-ing vibration of ethoxy groups directly bonded to silicon [18]. Withthe increase in the annealing temperature (Fig. 5b, particular)these bands decrease in intensity indicating a densification of thesilicon-oxide skeleton. This is in agreement with the intensity de-crease as the temperature is increased of the peak at 924 cm�1.This band is attributed to the stretching vibration of Si–OH or SiO�

groups [19] overlapped onto that by Si–O–Ti stretching [20]. Fur-thermore, the broad band between 3100 and 3600 cm�1 and thepeak at 1620 cm�1, observed for the samples heated below400 �C, are attributed to stretching vibration of OH group andH2O molecules, respectively. This shows that the films heated be-low 400 �C contain hydroxyl or water molecules while higher tem-peratures lead to a decrease/elimination of hydroxyl content.

The IR results explain the correlation between the mechanicalperformances (scratch resistance) and the annealing temperaturefor the 5_Y samples. In fact when the coating is treated at the max-imum temperature, 5_600 sample, LC1 and LC2, increase signifi-cantly thanks to an increase of the Si–O–Si backbone of the

inorganic network formed by condensation of TEOS as showed bythe decrease of the signal at 1050 cm�1 and the 800 cm�1.

Finally, to evaluate the photocatalytic activity of the obtainedcoating, contact angle measurements and cleanability test wereperformed. The contact angles of water for the 5 wt% titania filmstreated at different temperatures as a function of UV-light irradia-tion time are shown in Fig. 6. All the coatings have a lower CAswith respect to the untreated tile and the value decreases as theirradiation time is increased that is the tile surface becomes morehydrophilic by UV-light irradiation. Regarding the effect of thethermal treatment, the CAs decrease as the annealing temperatureis increased suggesting that also silica structure plays an importantsynergistic role for surface hydrophilicity of titania–silica mixedfilms.[12] In particular, in 5_100 sample the CA decreases mono-tonically with a behavior that is comparable to that of uncoatedtile. This is due to the effect of evaporation and thus the CA linearlydecreases as evaporation, accelerated by UV irradiation, proceeds,as already evaluated in water-glass system [21]. The CAs, instead,decrease drastically in the 5_200, 5_400 and 5_600 samples. It is

F. Bondioli et al. / Journal of Colloid and Interface Science 334 (2009) 195–201 201

noticed that the CA of 0�, called super hydrophlicity, is observed inthe sample annealed at 600 �C irradiated only 3 min. This behaviorconfirms that also very small amounts of anatase nanoparticles in-duce a strong hydrophilicity of surfaces and that the silica matrixdoes not affect the anatase photocatalytic activity.

Photo-degradation of methylene blue solution (500 ppm) wasused to assess the photocatalytic activity of the samples containing5 wt% of titania. In Fig. 7 the color variations (DE*) as a function ofboth annealing temperature and UV-light irradiation are reported.It is important to notice that also on the untreated sample there isa slightly color variation probably due to the degradation effect ofthe UV-light on the methylene blue. All the coatings have, how-ever, an higher DE* with respect to the untreated tile and the valueincreases as the irradiation time is increased. The behavior is notlinear and after approximately 60 min the degradation kineticchanges becoming slower. Regarding the effect of the thermaltreatment, the DE* decrease as the annealing temperature is in-creased. This behavior is in agreement with the gloss data that de-crease as the annealing temperature is increased suggesting theimportant role of surface roughness in the photodegradation pro-cess [22]. In general rough surfaces were easy to soil and hard toclean.

4. Conclusion

The sol–gel process was used in this work for the preparation ofcoatings for ceramic tiles. This study investigated the effect ofannealing temperature and titania content on the final propertiesof the obtained films. The titania-based films are transparent, donot modify the glossy of the uncoated tiles and show a good adhe-sion. The higher performance was obtained with 5_600 film thatshow the higher condensation degree of TEOS as determined byFTIR. The surface photocatalicity were optimized with the higherthermal treatments (200, 400 and 600 �C) even if photodegradationprocess is clearly affected by the sample surface roughness.

Acknowledgments

Support from Prin2007 project ‘‘Made in Italy in the ceramicindustry for building use. Nanopowders and nanotechnologies forthe aesthetic innovation and functionalization of the ceramic sur-faces” (20073MZALE) and experimental help of Dr. Francesco Boz-za are gratefully acknowledged.

References

[1] FINE PORCELAINIZED STONEWARE, ed. SACMI in web site www.sacmi.com.[2] T. Manfredini, G.C. Pellacani, M. Romagnoli, L. Pennisi, Am. Ceram. Soc. Bull. 74

(1995) 76.[3] C. Leonelli, F. Bondioli, P. Veronesi, M. Romagnoli, T. Manfredini, G.C. Pellacani,

V. Cannillo, Eur. Ceram. Soc. 21 (2001) 785.[4] W. Barthlott, C. Neinhuis, Planta 202 (1997) 1.[5] R. Kuisma, L. Fröberg, H. Kymäläinen, E. Pesonen-Leinonen, M. Piispanen, P.

Melamies, M. Hautala, A.M. Sjöberg, L. Hupa, J. Eur. Ceram. Soc. 27 (2007) 101.[6] G.S. Watson, J.A. Watson, Appl. Surf. Sci. 235 (2004) 139.[7] R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojima, A. Kitamura, M.

Shimohigoshi, T. Watanabe, Nature 388 (1997) 431.[8] A. Fujishima, T.N. Rao, D.A. Tryk, J. Photochem. Photobiol. C1 (2000) 1.[9] A. Fujishima, T.N. Rao, D.A. Tryk, Electrochim. Acta 45 (2000) 4683.

[10] R. Benedix, F. Dehn, J. Quaas, M. Orgass, Lacer 5 (2000) 157.[11] Z. Luo, H. Cai, X. Ren, J. Liu, W. Hong, P. Zhang, Mat. Sci. Eng. B 138 (2007) 151.[12] M. Maeda, S. Yamasaki, Thin solid Films 483 (2005) 102.[13] R. Taurino, E. Fabbri, M. Messori, F. Pilati, D. Pospieck, A. Syntaska, J. Colloid

Interface Sci. 325 (2008) 149.[14] CIE, Recommendations of Uniform Color Spaces, Color Difference Equations,

Phychometrics Color Terms. Supplement No. 2 of CIE Publ. no. 15 (E1-1.31)1971, Bureau Central de la CIE, Paris, 1978.

[15] R.S. Hunter, J. Opt. Soc. Am. 48 (1958) 985.[16] H. Tim Muster, A. Clive Prestidge, Int. J. Pharm. 234 (2002) 43 .[17] A. Kupicka, Prog. Org. Coat. 46 (2003) 32.[18] L. Smith, Spectrochim. Acta 16 (1960) 87.[19] X.M. Du, R.M. Almeida, J. Mater. Res. 11 (1996) 353.[20] X. Gao, I.E. Wachs, Catal. Today 51 (1999) 233.[21] A.K. Panwar, S.K. Barthwal, S. Ray, J. Adhesion Sci. Technol. 17 (2003)

1321.[22] M. Dondi, G. Ercolani, G. Guarini, C. Melandr, M. Raimondo, E. Rocha e

Almendra, P.M. Tenorio Cavalcante, J. Eur. Ceram. Soc. 25 (2005) 357.