Surface & Coatings Technology 204 (2010) 1445–1451

Contents lists available at ScienceDirect

Surface & Coatings Technology

j ourna l homepage: www.e lsev ie r.com/ locate /sur fcoat

Stable TiO2 dispersions for nanocoating preparation

N. Veronovski a,⁎, P. Andreozzi b, C. La Mesa b, M. Sfiligoj-Smole a

a University of Maribor, Characterization and Processing of Polymers Laboratory, Faculty of Mechanical Engineering, University of Maribor, Smetanova 17, SI-2000 Maribor, Sloveniab Sapienza University, Department of Chemistry, Cannizzaro Building, P.le A. Moro 5, I-00185 Rome, Italy

⁎ Corresponding author. Tel.: +386 3 42 44 103; fax:E-mail address: nikaveronovski@gmail.com (N. Vero

0257-8972/$ – see front matter © 2009 Elsevier B.V. Aldoi:10.1016/j.surfcoat.2009.09.041

a b s t r a c t

a r t i c l e i n f o

Article history:Received 15 July 2009Accepted in revised form 15 September 2009Available online 22 September 2009

Keywords:TiO2 P25Cationic and anionic surfactantsCoatingsSize and ζ-potential characterizationSEM analysisSelf-cleaning

In this research the preparation and characterization of titanium (TiO2) coated self-cleaning cellulosematerialsstarting from TiO2 P25 powder (Degussa, Germany) was studied. The aim of the research was to decrease highaggregation of TiO2 P25 nanoparticles, using surfactant species as dispersant and/or stabilisers (consideringthe balance between repulsive and attractive forces), in view of the fact that TiO2 nanoparticles, whenoptimally separated into smaller particle populations, present the best properties in the system they are usedin (coatings). For this purpose cationic alkanediyl-α,ω-bis-N-dodecyl-N, N′-dimethyl-ammonium bromide(Gemini) and anionic sodium dodecyl sulphate (SDS) surfactants were applied, with concentrations undertheir CMSs. Size and zeta-potential (ζ-potential) characterization of stable colloidal dispersions wereperformed. For stable 0.5, 2.5 and 5.0 mg/mL TiO2 dispersions in the presence of 250×10−6mol/L Geminisurfactant (ζ-potential~40mV) only two scattering populations were determined, at 78–95nm and at~280nm. As a proof of stabilized TiO2 P25-surfactant colloidal dispersions uniform coatings were obtained,generated at the fibre surfaces, which were analyzed by scanning electron microscopy (SEM). With the usageof proper amounts of surfactants, homogeneous thin TiO2 coatings were formed. Superior dense coatings onthe fibre surfaces were formed after treatment in 5.0 mg/mL aqueous TiO2 P25 dispersions, in the presence of250×10−6mol/L Gemini surfactant in 5.0×10−3mol/LKBr at 25 °C. In addition to that, self-cleaning test wasperformed. Higher photocatalytic activity was determined for samples with denser coatings. Fibres with thincoatings had a lack of photocatalytic activity.

+386 3 42 44 182.novski).

l rights reserved.

© 2009 Elsevier B.V. All rights reserved.

1. Introduction

The production of inorganic particles in the form of powder, as wellas their applicationfield, has beendeveloped rapidly in the last decade.Inorganic particles are frequently used to obtainmaterials with specialproperties such as composites, ceramics and coatings. Self-cleaningapplications using semiconducting powders or thin films have becomea subject of increasing interest especially in the last 10 years. Self-cleaning materials were developed by coating glass, ceramic tiles,plastics or fibres with highly photoactive semiconducting photocata-lyst titanium dioxide (TiO2) [1–6].

Redispergation and stabilization of powders in liquid media is stilla big problem of different technological processes, since numerousproperties of the final product depend strongly on colloidal stability ofparticles and their distribution in a certain volume [7].

One of the most important and controllable properties of suchsuspensions is their stability. In the macroscopic sense this term ex-presses an even distribution of components throughout the wholevolume and resistance of the phases to separate from each other withtime [8]. The sedimentation behaviour is often seen as the crucial

stability criterion. According to the Derjaguin, Landau, Verwey andOverbeek theory (DLVO theory), there are two basic forces controllingthe stability of colloidal suspensions: the van-der-Waals and theelectrostatic forces [9]. If the value of the total potential energy ispositive andhigh enough, particleswill repel each other; otherwise theyexperience a mutual attraction. The van-der-Waals force is attractivebetween particles of the same kind and can be repulsive or attractivebetween two different particles. The electrostatic force is related to theparticle charge. For multicomponent system it is either attractive orrepulsive depending on the sign of charge of the respective particlekinds. The zeta potential (ζ-potential) represents the effective measureof the particle charge. Formost one component suspensions high valuesof this parameter indicate high stability, while low values implycoagulation. One of the most effective ways to affect the properties ofcolloidal suspensions is the addition of surfactants, which adsorb at thesolid–liquid interface. Surfactants are known to play a vital role inmanyprocesses of interest in both fundamental and applied science.

In order to obtain self-cleaning coatings, TiO2 P25 photocatalyst wasapplied in the present research. TiO2 P25, containing a mixture of rutileand anatase crystalline forms, is themostwidely used photocatalyst andhas proven tobe the best photocatalyst towards a broad range of organicpollutants [10–15]. Like in other catalytic applications, the surface areawhich is available for reactions plays a part in photocatalysis also [16].Since the level of TiO2 P25 nanoparticles aggregation is rather high

1446 N. Veronovski et al. / Surface & Coatings Technology 204 (2010) 1445–1451

[16,17], surfactants were introduced into the process to control nano-particles aggregation.

The adsorption of conventional surfactants on TiO2 and othersurfaces [7,18–28] aswell asGemini adsorption at solid/liquid interfacesand on solid surfaces was already investigated by several researchers[29–31]. However, the aim of our research was to determine theinfluence of surfactants on the nanocoating formation. Gemini consist-ing of two surfactant units show significant surface active efficiency andform micelles at much lower concentrations then the correspondinghomologues [32,33]. High surfactant efficiency and low critical micelleconcentration (CMC) values have suggested the use of Geminisurfactants. They form micelles at concentrations much lower thancorresponding monomeric species. Higher effectiveness of Geminicauses more stable dispersions, due to their surface tension at theCMC (γCMC) [34]. For this purpose Gemini and SDS surfactants wereselected with concentrations below the critical micelle concentration(CMC).

2. Experimental

2.1. Materials

In the research TiO2 P25 nanoparticles were used (kindly providedby Degussa, Germany) with average diameter of 21nm and specificsurface area close to 55±15 m2/g and refractive index above 2.5[35,36], which were dispersed in filtered double-distilled water or in5.0×10−3mol/LKBr solution.

As cationic surfactants alkanediylα,ω-bis (N-dodecyl-N, N′-dimethy-lammonium bromides), alkylammonium Geminis, were used, withdodecyl groups linked to both ends of α, ω-N,N′-dimethylamine chainsseparatedby twoor sixmethyleneunits,which act as spacers between thepolar head groups. 12-6-12Gemini surfactantwas usedwith CMCs belowthe milimolar range, 5.0×10−4mol/L in 5.0×10−3mol/LKBr at 25 °C.

As anionic surfactant sodium dodecyl sulphate (SDS) was usedwith CMC 8.0×10−3mol/L in 5.0×10−3mol/LKBr at 25 °C.

TiO2 P25 concentrations used were 0.5, 2.5 and 5.0 mg/mL, SDSconcentration used was 5.0×10−3mol/L. In the meantime cationicGemini surfactant concentrations used were 1.0 and 250×10−6mol/L.

Regenerated cellulose Lyocell fibres (1.17 dtex; Lenzing, Austria)were coated.

Fig. 1. Stabilizat

2.2. Preparation of stable colloidal dispersions

2.2.1. Preparation of KBr solutionKBr solutionwas prepared using 5.0×10−3mol/L KBr and deionized

water (pH~7).

2.2.2. Preparation of Gemini solutionsWater solution of 5.0×10−4mol/L Gemini surfactant in deionized

water was prepared (pH~6). During a research, diluted solutions withproper surfactant concentrations were prepared from an initialsolution. The Gemini surfactant concentrations used were: 1.0 and250×10−6mol/L.

2.2.3. Preparation of SDS solutionsWhile stabilization by cationic Gemini surfactants was investigat-

ed using different concentrations, SDS was used at a singleconcentration only. Water solution of 5.0×10−3mol/L SDS surfactantin deionized water was prepared (pH~7).

2.2.4. Preparation of TiO2 P25 suspensions0.5, 2.5 and 5.0 mg/mL TiO2 P25 was mixed with 5.0×10−3mol/

LKBr solution (pH~4.5). The suspensions were dispersed for 1h usingultrasound device Branson 5200 (Branson, Danbury, CT), operating at47kHz and 185W, maintained at ~25–30 °C, to re-disperse large ag-glomerates before addition of surfactant, to separate big agglomeratesinto smaller units with the goal to obtain more uniform particledistribution in aqueous suspension. In this way, higher available surfacearea of TiO2 P25 for more efficient surfactant adsorption is obtained.

2.2.5. Preparation of TiO2 P25 dispersions with surfactantsAfter 1h of TiO2 P25 suspension sonication in ultrasound device,

drop-wise addition of Gemini/SDS solutions followed. The resultingmixtureswere prepared at 25 °C and kept under stirring for two days toensure equilibration.

Stabilization process began with mixing and stirring of twosolutions, Gemini/SDS and TiO2 P25, which resulted in formation ofstable colloid dispersion. Adsorption of surfactant molecules on theTiO2 P25 nanoparticles surfaces, led to formation of one or morelayers, which prevented further TiO2 P25 aggregation. An expectedstabilization process in the case of Gemini surfactant (cationic Gemini

ion process.

Table 1Samples designation.

Designation Dispersion

0 5.0 mg/mL TiO2 P25 aq. suspension without surfactantA 0.5 mg/mL TiO2 P25 aq. suspension with 1.0×10−6mol/L GeminiB 0.5 mg/mL TiO2 P25 aq. suspension with 250×10−6mol/L GeminiC 2.5 mg/mL TiO2 P25 aq. suspension with 250×10−6mol/L GeminiD 5.0 mg/mL TiO2 P25 aq. suspension with 250×10−6mol/L GeminiE 0.5 mg/mL TiO2 P25 aq. suspension with 5.0×10−3mol/L SDSF 5.0 mg/mL TiO2 P25 aq. suspension with 5.0×10−3mol/L SDS

1447N. Veronovski et al. / Surface & Coatings Technology 204 (2010) 1445–1451

molecules adsorption on the surface of negatively charged TiO2 P25nanoparticles) is presented in Fig. 1.

Sample designation is presented in Table 1.

2.2.6. Preparation of TiO2 P25 dispersions for ζ-potential and DLS AnalysisTo avoid experimental difficulties and get the samples free fromdust

and large particles, which would strongly scatter light, the dispersionswere filtered with 0.22 μm pore size membranes (Millipore).

2.3. Coating process

After equilibration,fibreswere treated in stable dispersions for 1h. Inaddition, fibres were rinsed with water and dried at room temperature.

2.4. Zeta potential (ζ-potential) analysis

For ζ-potential analysis a Malvern laser-velocimetry Doppler utility(Zetasizer Nano series HT (Malvern, UK), was used for determination ofthe electro-phoretic mobility, μ (m2/sV), of the TiO2 P25 dispersions. μof the dispersions was transformed into ζ-potential according to [37]

ζ =4πημε

� �ð1Þ

where ε is a dielectric constant of the dispersing medium and η thesolvent viscosity.

2.5. Particle size distribution analysis

To determine the particle size distribution, dynamic light scattering(DLS) measurements were carried out. The analysis was performedusing aMalvern light scattering unit, Zetasizer Nano series HT (Malvern,UK) [38]. Individual peaks in particle size distributions were derivedfrommulti-modal correlation functions. The datawerefitted byCONTINalgorithms. The apparatus performances were controlled by measuringthe size of 100nm polystyrene latex spheres, stabilized by surface

Fig. 2. ζ-potential distribution plot for 5.0 mg/mL aqueous TiO2 P25 dispersion without surfsurfactants.

sulfate groups (Alfa Aesar) [39]. The measuring temperature was fixedat 25.0±0.1 °C, and controlled by a Peltier unit. Experimentswere run afew minutes after thermal equilibrium was reached, to minimizeeventual drifts. According to the experiments, particle sizes are constantto within a few percent.

In terms of prime principles care should be taken to operate inconditions where the diffusive contribution is much higher than thedensity gradient sensed by the particles. This condition is expressedby Peclet's number, Pe, according to [40]

Pe =43πr

4ΔρgKBT

" #ð2Þ

where r is the particles radius,Δρ the density gradient, g the gravity andKBT the thermal energy. When the latter term is ≫(4πr4Δρg/3),sedimentation is immaterial and only diffusive contributions arerelevant. The above conditionswere usually met in the present systemsand this was also proven by an almost constant scattering intensity.

2.6. Surface observations

Fibre surface morphologies after surface modification in stabilizedcolloidal dispersions were studied by SEM analysis, using LEO 1450 VPScanning Electron Microscope, with a maximum resolution up to3.5 nm at 30 kV.

2.7. Self-cleaning test

Self-cleaning test was performed by observing the photodegrada-tion oxidation of organic dye solution, which was spoiled on thesurface of untreated and TiO2 treated fabric. A drop of red beet sapwasused for staining samples, however the same results were obtained ifany other stain was analysed, e.g. wine-stain. Samples were exposedto direct day light for 33 days. Colour changes of the stain werefollowed visually and colorimetrically using a Datacolor internationalMicroflash 200d apparatus.

3. Results and discussion

3.1. Zeta potential (ζ-potential) analysis

ζ-potential measurements confirmed the instability of TiO2 P25aqueous dispersions (ζ-potential=14.2 mV). In Fig. 2 an increasingstability after the addition of Gemini or SDS surfactant can be seen. Anincrease in the absolute value of ζ-potentialwas observed after additionof cationic and anionic surfactant, respectively. Cationic and anionicsurfactants shifted ζ-potential to values higher than ±30mV and

actants and for 0.5, 2.5 and 5.0 mg/mL aqueous TiO2 P25 dispersions in the presence of

Table 2ζ-potentials of different TiO2 P25 colloidal dispersions in 5×10−3mol/LKBr at 25.0 °C.

Suspension ζ-potential [mV]

0 14.2A 36.8B 40.5C 39.4D 43.6E −45.7F −43.4

Table 3The mean particle sizes at each peak of 0.5, 2.5 and 5.0 mg/mL TiO2 P25 colloidaldispersions in 5.0×10−3mol/LKBr at 25.0 °C.

Suspension Peak1 Peak2 Peak3[nm] [nm] [nm]

0 175 676 4694A 276 1248 /B 78 281 /C 95 279 /D 112 320 /E 177 555 4821F / 1047 5502

1448 N. Veronovski et al. / Surface & Coatings Technology 204 (2010) 1445–1451

caused the increase of dispersion stability. The highest stability of 0.5,2.5 and 5.0 mg/mL TiO2 P25 aq. suspensionwas reached after addition of250×10−6mol/L Gemini surfactant (ζ-potential~40 mV). The resultingstability is due to the transfer of surfactant molecules from solution tobinding sites on titanium. An adsorption mechanism of cationic surfac-tants on TiO2 nanoparticles is described in details in the literature[41,42]. High stability was obtained in the presence of 5.0×10−3mol/LSDS surfactant, likewise (ζ-potential −43.4 and −45.7 mV).

Results in Fig. 2 and Table 2 indicate that the stability is reflected bythe ζ-potential of the TiO2 dispersions in the presence of surfactants.This method enables the estimation of the stabilization progress andthe interaction between surfactants and TiO2 P25 nanoparticles.Results were confirmed by size distribution results (see Fig. 3).

ζ-potential values of different TiO2 P25 colloidal dispersions in5.0×10−3mol/LKBr at 25.0 °C are listed in Table 2.

3.2. Particle size distribution analysis

To determine the particle size distribution, DLS measurementswere carried out. Size distribution plot for 5.0 mg/mL P25 aqueoussolution without surfactants and for 0.5, 2.5 and 5.0 mg/mL TiO2 P25aqueous dispersions in the presence of 1.0 and 250×10−6mol/LGemini, as well for 0.5 mg/mL P25 aqueous dispersion in the pre-sence of 5.0×10−3mol/L SDS is presented at Fig. 3. According to theplot, the samples contain different scattering populations. The thirdpeak in the case of 5.0 mg/mL P25 aqueous suspension withoutsurfactants and 0.5 mg/mLTiO2 P25 aqueous suspension in thepresenceof 5.0×10−3 mol/L SDS, for instance, proves the presence of largeagglomerates. Such behaviour was confirmed by ζ-potential measure-

Fig. 3. Size distribution plot for 5.0 mg/mL P25 aqueous suspension without surfactants and250×10−6 mol/L Gemini and for 5.0×10−3 SDS in 5.0×10−3mol/LKBr at 25.0 °C.

ments. These disappear after the addition of sufficient amounts ofsurfactant. In such conditions, only two populations are present, whichare stable against sedimentation. In the latter conditions, only twooverlapping populations occur. As a result of Gemini addition we canobserve a decrease in aggregation, which was the highest in the case of0.5 and 2.5 mg/mL TiO2 P25 aqueous dispersionwhen 250×10−6 mol/LGemini was used, when the system was almost completely dispersed.Two scattering populations were determined; at 78–95 and at 279–281nm. Large agglomerates of size ~4700 nm disappeared. Satisfyingresults were obtained for 5.0 mg/mL TiO2 P25 aqueous dispersion in thepresence of 250×10−6mol/L Gemini, likewise. Results indicate that thepresence of 1.0×10−6mol/L Gemini wasn't enough for good particledistribution in 0.5 mg/mL TiO2 P25 aqueous dispersion.

Table 3 indicates the mean particle sizes in 0.5, 2.5 and 5.0 mg/mLTiO2 P25 colloidal dispersions at each peak.

3.3. Surface observations

Treatment with 5.0 mg/mL TiO2 P25 aqueous dispersion resultedin high agglomeration of TiO2 P25 nanoparticles. Fibre surface in Fig. 4isn't coated entirely. This result is in accordance with the particle sizedistribution and ζ-potential analyses, where an analysis demonstrat-ed the presence of big agglomerates in the investigated suspension.

The use of TiO2 P25-surfactant colloidal dispersions resulted information of more homogeneous coatings with more uniform particledistribution on the fibre surface. 0.5 mg/mL TiO2 P25 aqueous dispersionin addition of 1.0×10−6mol/L Gemini yielded nanocoatings with poordensity. At the surface of fibre some small agglomerates occurred,

for 0.50, 2.5 and 5.0 mg/mL TiO2 P25 nanoparticle dispersions in the presence of 1.0 and

Fig. 4. SEM image of surface morphology of fibre treated with 5.0 mg/mL TiO2 P25aqueous suspension without surfactants.

Fig. 5. SEM images of surfacemorphologies of fibres treatedwith stable TiO2 dispersions— 0.5P25 aq. dispersion in addition of 250×10−6mol/L Gemini (B), 2.5 mg/mL TiO2 P25 aq. dispersionof 250×10−6mol/L Gemini (D), 5.0 mg/mL TiO2 P25 aq. dispersion in addition of 5.0×10−3molat magnification of 25×103.

1449N. Veronovski et al. / Surface & Coatings Technology 204 (2010) 1445–1451

however the surface was completely covered with sufficiently dispersednanoparticles (Fig. 5A). All the examinedfibreswere covered in the sameway. Any agglomerates which can't be seen at the surface of the fibre,when treatedwith 0.5 mg/mL TiO2 P25 aqueous dispersion in addition of250×10−6mol/L Gemini surfactant, occurred (Fig. 5B). We can observemore mono-dispersed nanoparticles. Coatings are homogeneous. Afterthe fibre surface treatment with 2.5 mg/mL TiO2 P25 aqueous dispersionin addition of 250×10−6mol/L Gemini surfactant, several smallerclusters occurred. Coating wasn't regular, particle distribution wasn'tuniform (Fig. 5C). After the treatment in 5.0 mg/mL TiO2 P25 aqueousdispersion in addition of 250×10−6mol/L Gemini, fibre surface wasentirely covered, the level of density of yielded coatingwashigh (Fig. 5D).SEM image of fibre surface treated with 5.0 mg/mL TiO2 P25 aqueousdispersion in addition of 5.0×10−3mol/L SDS surfactant shows veryhomogeneous coating (Fig. 5E). However, some fibres didn't havecomplete coverage; low amount of TiO2 P25 nanoparticles was attached.Particle distribution wasn't very dense, still it was continuous. Homoge-neous coating resulted from the treatment of the fibre surface with

mg/mL TiO2 P25 aq. dispersion in addition of 1.0×10−6mol/L Gemini (A), 0.5 mg/mL TiO2

in addition of 250×10−6mol/L Gemini (C), 5.0 mg/mL TiO2 P25 aq. dispersion in addition/L SDS (E), 0.5 mg/mL TiO2 P25 aq. dispersion in addition of 5.0×10−3mol/L SDS (F); taken

Fig. 6. Stain photodegradation as a function of time for untreated sample and samplestreated with 0.5 mg/mL TiO2 P25 aqueous dispersion in addition of 250×10−6mol/LGemini surfactant (B treatment) and with 5.0 mg/mL TiO2 P25 aqueous dispersion inaddition of 250×10−6mol/L Gemini surfactant (D treatment).

1450 N. Veronovski et al. / Surface & Coatings Technology 204 (2010) 1445–1451

0.5 mg/mL TiO2 P25 aqueous dispersion in addition of 5.0×10−3mol/LSDS surfactant. The stability of used colloid dispersion was high(−48.4 mV). Particle distribution in the yielded coating was denseagain (Fig. 5F), just like in the treatment of fibre surface with 5.0 mg/mLTiO2 P25 in addition of 250×10−6mol/L Gemini surfactant.

3.4. Self-cleaning test

Self-cleaning test was performed by observing the photodegradationoxidation of organic dye solution (red beet sap), which was spoiledon the untreated and treated fabrics. Colour changes of the stain weredetermined visually and colorimetrically using a Datacolor internationalMICROFLASH200d apparatus. Self-cleaning propertieswere determinedfor untreated samples and samples treated with 0.5 mg/mL TiO2 P25aqueous dispersion in addition of 250×10−6mol/L Gemini surfactant (Btreatment) andwith 5.0 mg/mL TiO2 P25 aqueous dispersion in additionof 250×10−6mol/L Gemini surfactant (D treatment).

Sampleswere exposed to direct daylight for 2h before staining themwith a drop of red beet sap. Colour characteristics of stained samples

Fig. 7. Stain photodegradation on the surface of the untreated (A) and on the surface of sampGemini (B) as a function of time (t=5days).

were determined using colorimetric measurements (lightness (L⁎)component of CIE Lab system was investigated). At the beginning, L⁎was determined for unstained samples. In addition, stained sampleswere exposed to daylight for another period of time. From the graphL⁎= f (t) in Fig. 6 we can notice, that a major change in colour of stainoccurred after thefirst hour of exposure to daylight. Themost significantchange in L⁎ of stain occurred in the case of stained sample treatedwith5.0 mg/mL TiO2 P25 aqueous dispersion in addition of 250×10−6mol/LGemini surfactant (ΔL⁎=37.3). Change in the colour of stain wasnoticed for stained untreated sample exposed to daylight, as well. Thisindicates that UV light, presented in daylight, is responsible for partialdegradation of colour substance in stain. Treatment B (0.5 mg/mL TiO2

P25 in addition of 250×10−6mol/L Gemini) yielded a nanocoatingwithlower photocatalytic activity (ΔL⁎=31.27) compared with nanocoat-ing, which was formed during the treatment D (5.0 mg/mL TiO2 P25 inaddition of 250×10−6mol/LGemini). The difference in the colour of thestain on untreated stained sample was minor (ΔL⁎=28.61).

After33days of exposure todaylight, colour characteristics of stainedsamples approached to those of unstained samples. The degradation ofstain was the most significant on the sample treated with 5.0 mg/mLTiO2 P25 aqueous dispersion in addition of 250×10−6mol/L Gemini.The difference between unstained and stained sample, 33days exposedto direct daylight was ΔL⁎=0.36. The colour difference of stain onuntreated samples, 33days exposed to direct daylight (L⁎33=87.17)differed from the one in the dark (L⁎33=81.80) for ΔL⁎=5.37.

The results of the present research indicated that by increasing theamount of TiO2 from 0.5 to 5.0 mg/mL better self-cleaning properties oftreated samples were obtained. The reason for lower photocatalyticactivity of sample treated with 0.5 mg/mL TiO2 P25 aqueous dispersionin addition of 250×10−6mol/L Gemini surfactant is presumably athinner coating of TiO2 P25 nanoparticles. Thickness of TiO2 coating isrelated to the TiO2 loading. Higher TiO2 loading means more availablesites,where reactions of photocatalysis could take place. Thicker coatingenhances photocatalytic activity due to higher TiO2 concentration, sincemore active oxygen species (hydroxyl radicals, hydrogen peroxide) aregenerated in the reaction of photocatalysis, which are able to destroypollutants. The surface of coated fibres shows the presence of welldispersed TiO2 nanoparticles, so an increased number of individual TiO2

nanoparticles of highly reduced dimensions are available, allowingbetter photocatalytic performance. These results are in agreement withTobaldi et al. [43]. Hence, to achieve nanocoatings with higherphotocatalytic activity, higher TiO2 P25 concentration has to be used.

The result of visual determination of self-cleaning, based on thephoto-catalytic degradation of the organic dye dropped on the surface

les treated with 5.0 mg/mL TiO2 P25 aqueous dispersion in addition of 250×10−6mol/L

1451N. Veronovski et al. / Surface & Coatings Technology 204 (2010) 1445–1451

of the untreated and on the surface of samples treated with 5.0 mg/mLTiO2 P25 aqueous dispersion in addition of 250×10−6mol/L Gemini,is demonstrated in Fig. 7. No decolouration of the dyestuff after 5daysof exposure to direct daylight was observed when the untreatedsample (A) was used. However, sample treated with 5.0 mg/mL TiO2

P25 aqueous dispersion in addition of 250×10−6mol/L Gemini (B)displayed self-cleaning effect.

The influence of self-cleaning test (photocatalytic activity) on thefibre stability was taken into consideration, as well. Fibre mechanicalproperty determination was performed after self-cleaning test. Theresults indicated that the photocatalytic activity hasn't got significantinfluence on fibre properties. SEM analyses performed on treatedfibres after self-cleaning test revealed that TiO2 coating was stillpresent on the fibre surface and that fibres remained undamaged.

4. Conclusions

Surfactant adsorption at the solid/solution interface was employedto modify a surface of TiO2 P25 nanoparticles and hence colloidal sta-bility was improved.

Results obtained by ζ-potential analysis revealed that usingoptimal TiO2 P25 nanoparticles and surfactant concentrations resultedin stable colloidal dispersions. The results of ζ-potential agree verywell with the results obtained by DLS analysis. The adsorption ofsurfactants on TiO2 particles caused an increase in stability.

SEM analysis of the fibre surface morphologies has confirmed thatTiO2 P25 nanoparticles have been successfully and uniformly immo-bilized on to the fibre surfaces, when stable colloidal dispersions wereused. Particle distribution is homogeneous and the extent of particleaggregation decreased in the presence of optimal addition of thesurfactant. In themeantime fibre surface treatmentwith unstable TiO2

P25 aqueous dispersion yielded coatings with huge agglomerates.The results suggest that the dispersion stability as a dominating

factor in deciding the nanocoating formation, is influenced by thesurfactant and its amount used.

Results indicate very important influence of coating thickness onself-cleaning properties. Fibres with thin coatings, prepared fromdispersions with low TiO2 P25 content, had a lack of photocatalyticactivity. Therefore sufficient coating thickness is required for obtain-ing the highest photocatalytic activity.

References

[1] D.M. Tobaldia, A. Tuccib, G. Camera-Rodac, G. Baldid, L. Espositob, J. Eur. Ceram.Soc. 28/14 (2008) 2645.

[2] C. Euvananont, C. Junin, K. Inpor, P. Limthongkul, C. Thanachayanont, Ceram. Int.34 (2008) 1067.

[3] R. Benedix, F. Dehn, J. Quaas, M. Orgass, Lacer 5 (2000) 157.[4] K.H. Qi, W.A. Daoud, J.H. Xin, C.L. Mak, W.S. Tang, W.P. Cheung, J. Mater. Chem. 16

(2006) 4567.

[5] H. Honda, A. Ishizaki, R. Soma, K. Hashimoto, A. Fujishima, J. Illum. Eng. Soc.(Winter 1998) 42.

[6] A. Fujishima, T.N. Rao, D.A. Tryk, J. Photochem. Photobiol. C Photochem. Rev. 1(2000) 1.

[7] Z.M. Yaremko, N.H. Tkachenko, C. Bellmann, A. Pich, J. Colloid Interface Sci. 296(2006) 565.

[8] D.J. Shaw, Introduction to Colloid and Surface Chemistry, Butterworth-Heine-mann, Boston, 1992.

[9] R.J. Hunter, Zeta Potential in Colloid Science, Academic Press, 1981.[10] C. Guillard, J. Disdier, J.-M. Herrmann, C. Lechaut, T. Chopin, S. Malato, J. Blanco,

J. Cat. Today 54/2–3 (1999) 217.[11] O. Heintz, D. Robert, J.V. Weber, J. Photochem. Photobiol., A Chem. 135 (2000) 77.[12] R. Vormberg, Elements, Degussa, Sci. Newsl. 9 (2004) 21.[13] R.R. Bacsa, J. Kiwi, Appl. Catal., B Environ. 16/1 (1998) 19.[14] R.I. Bickley, T. Gonzalez-Carreno, J.S. Lees, L. Palmisano, R.J.D. Tilley, J. Solid State

Chem. 92 (1991) 178.[15] A.K. Datye, G. Riegel, J.R. Bolton, M. Huang, M.R. Prairie, J. Solid State Chem. 115

(1995) 236.[16] J. Winkler, Titanium Dioxide, Hannover: Vincentz, p. 80 and 101–114, 2003.[17] F. Bajd, A Seminar: Interakcije med nanodelci (Interactions between nanoparti-

cles), University in Ljubljana, Faculty for Mathematics and Physics, Ljubljana,2007.

[18] K. Paciejewska, F. Babick, M. Stintz, R. Lange, Conditioning of Binary Suspensions ofTitania and Silica Particles with Cationic Surfactant, Partec, Nuremberg, Germany,2007.

[19] N.H. Tkachenko, Z.M. Yaremko, C. Bellmann, M.M. Soltys, J. Colloid Interface Sci.299 (2006) 686.

[20] A. Jada, J. Hoffstetter, B. Siffert, P. Dumas, Colloids Surf., A Physicochem. Eng. Asp.149 (1999) 315.

[21] K. Esumi, S. Uda, M. Goino, Y. Ishiduki, T. Suhara, H. Fukui, Y. Koide, Langmuir 13(1997) 2803.

[22] K. Esumi, M. Goino, Y. Koide, J. Colloid Interface Sci. 183 (1996) 539.[23] L.G.T. Erikssson, P.M. Claesson, J.C. Eriksson, V.V. Yaminsky, J. Colloid Interface Sci.

181 (1996) 476.[24] M.L. Fielden, P.M. Claesson, R.E. Verrall, Langmuir 15 (1999) 3924.[25] R. Atkin, V.S.J. Craig, P.G. Hartley, E.J. Wanless, S. Biggs, Langmuir 19 (2003) 4222.[26] V. Dutschk, K.G. Sabbatovskiy, M. Stolz, K. Grundke, V.M. Rudoy, J. Colloid Interface

Sci. 267 (2003) 456.[27] G. Bai, Y.Wang, H. Yan, R.K. Thomas, J.C.T. Kwak, J. Phys. Chem., B 106 (2003) 2153.[28] T. Sato, S. Kohnosu, J. Colloid Interface Sci. 143 (1991) 434.[29] P.J. Dale, J. Kijlstra, B. Vincent, Langmuir 21 (2005) 12250.[30] M.J. Rosen, F. Li, J. Colloid Interface Sci. 234 (2001) 418.[31] A.R. Tehrani-Bagha, H. Bahrami, B. Movassagh, M. Arami, S.H. Amirshahi, F.M.

Mengerc, Colloids Surf., A Physicochem. Eng. Asp. 307 (2007) 121.[32] R. Zana, Langmuir 12 (1996) 1208.[33] R. Zana, Structure–Performance Relationships in Surfactants, 2nd ed, Surfactant

Sci. Ser., vol. 547, Marcel Dekker, New York, 2003.[34] M. Chorro, C. Chorro, O. Dolladille, S. Partyka, R. Zana, J. Colloid Interface Sci. 210

(1999) 134.[35] T. Imae, K. Muto, S. Ikeda, Colloid Polym. Sci. 269 (1991) 43.[36] A.R. Tehrani-Bagha, H. Bahrami, B. Movassagh, M. Arami, S.H. Amirshahi, F.M.

Mengerc, Colloids Surf., A Physicochem. Eng. Asp. 30 (2007) 121.[37] A. Bonincontro, E. Spigone, M. Ruiz-Pena, C. Letizia, C. La Mesa, J. Colloid Interface

Sci. 304 (2006) 342.[38] A. Bonincontro, C. La Mesa, C. Proietti, G. Risuleo, Biomacromolecules 8 (2007)

1824.[39] P. Andreozzi, A. Bonincontro, C. La Mesa, J. Phys. Chem., B 112 (2008) 3339.[40] J. Zhu, M. Li, R. Rogers, W.V. Meyer, R.H. Ottewill, W.B. Russel, P.M. Chaikin, Nature

387 (1997) 883.[41] P.J. Dale, J. Kijlstra, B. Vincent, Langmuir 21 (2005) 12250.[42] J. Zajac, J.L. Trompette, S. Partyka, Langmuir 12 (1996) 1357.[43] D.M. Tobaldi, A. Tucci, G. Camera-Roda, J. Eur. Ceram. Soc. 28 (2008) 2645.