H

Ma

b

c

a

ARR2AA

KHHUAZM

1

fsWetwCaabtaiieEf[

f

0h

Applied Surface Science 264 (2013) 344– 348

Contents lists available at SciVerse ScienceDirect

Applied Surface Science

jou rn al h om epa g e: www.elsev ier .com/ locate /apsusc

ydrophobic/hydrophilic switching on zinc oxide micro-textured surface

yo Tay Zar Myinta, Nithin Senthur Kumarb, Gabor Louis Hornyaka, Joydeep Duttaa,c,∗

Center of Excellence in Nanotechnology, School of Engineering and Technology, Asian Institute of Technology, Klong Luang, Pathumthani 12120, ThailandSchool of Engineering, Vanderbilt University, Nashville, TN 37240, USAChair in Nanotechnology, Water Research Center, Sultan Qaboos University, 123 Al-Khoudh, Oman

r t i c l e i n f o

rticle history:eceived 25 July 2012eceived in revised form4 September 2012ccepted 4 October 2012vailable online 16 October 2012

a b s t r a c t

Switchable wettability of zinc oxide (ZnO) microrod coated surfaces was controlled in two different ways:(1) by physical geometry (surface coverage area SA: the area covered by solid) and (2) by irradiation withultraviolet (UV) light followed by infrared (IR) or furnace heating. In the first approach, the thresholdcoverage area for achieving hydrophobic surfaces was found to be <40%, which is in good agreementwith predicted values in the literature leading to a metastable Cassie–Baxter regime. The transformationof hydrophobic to hydrophilic surfaces was studied by alternating cycles of 3 h exposure to ultravio-

eywords:ydrophobicydrophilicltravioletnnealinginc oxideicrorod

let (�peak ∼ 253 nm) light followed by 1 h of annealing or IR irradiation alone. Three different annealingtemperatures (120 ◦C, 200 ◦C and 250 ◦C) were utilized. Results of this work can be applied for designingsurfaces with controlled wettability.

© 2012 Elsevier B.V. All rights reserved.

. Introduction

The wettability of a surface is influenced by its morphology (sur-ace roughness, micro-nanostructure) and the material dependenturface energy [1–4]. However, hydrophilic (water contact angle,

CA < 90◦) materials can be turned hydrophobic (WCA > 90◦) andven superhydrophobic (WCA > 150◦) by micro and nanostruc-uring the surfaces. The phenomenon of surface wettability isell understood and explained by Wenzel model [2] and the

assie–Baxter model [5–8]. In the Wenzel model, it is presumed thatll topographical features of the surface are completely wet by thepplied liquid. In other words, the surface is completely saturatedy the liquid since only solid and liquid interface are considered. Inhe Cassie–Baxter state [1], the applied liquid is considered to formn interface that lies on top of the topographical facets (e.g. touch-ng only the tips of the structured surfaces) without penetrationnto the valleys whereby solid, liquid and air interfaces are consid-red in this regime, which is commonly referred to as the “Fakirffect” [9]. Depending on the surface structure and material usedor structuring, transitions between these two states are possible

7].

In 2008, theoretical studies of Duez et al. showed that a sur-ace of parallel striped structures with lateral length scale from

∗ Corresponding author at: Tel.: +66 2524 5680; fax: +66 2524 5617.E-mail addresses: joy@ait.asia, dutta@squ.edu.om (J. Dutta).

169-4332/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.apsusc.2012.10.024

1 to 10 �m and less than 40% solid fraction (60% was comprisedof air) is hydrophobic [10]. Shastry et al. demonstrated that theheight to pillar diameter (h/a) is an important factor for deter-mining hydrophobicity [11]. We have previously reported thatthe surface hydrophobicity follows the pillar height when spac-ing between pillars is kept constant [12]. Das et al. have recentlyshown that WCA can be changed from 104◦ to 135◦ by the sim-ple manipulation of surface morphology [13]. As early as 2007,Bhushan et al. proposed that the spacing factor, the ratio betweendiameter of pillar to the pitch distance, affects the hydrophobicityin the Cassie–Baxter regime [14]. A tailored superhydrophobic sur-face with dual roughness surface has also been reported by He et al.where unitary (single roughness surfaces) or dual roughness sur-faces fabricated on silicon substrates resulted in WCA to abruptlychange from 53◦ for unitary structures, to 156◦ in the case of binarystructures [15]. Ning et al. reported on superhydrophobic surfacesfabricated on zinc substrate [16]. Hexagonal cavities covered withplatinum nanoparticles of 50 nm (average diameter) led to the dualroughness structures which gave a WCA of 171◦. Hou et al. applieda low surface energy material on ZnO microrods and reported achange in the wettability from a WCA of 40◦ to 153◦, characteristicof a superhydrophobic surface [17].

Since the report of Sun et al. in 2001 on the photoinduced sur-

face wettability conversion on ZnO and titanium dioxide (TiO2)thin films, there has been much activity concerning the inter-action of water droplets on nanostructured coatings formed bymetal oxide particles consisting of nanorods and microrods [18].

urface

RanUsues2ntitdpfai

swseZarbrpWCe

2

2

azAaoupd

2

hadrn(awo

ptrng

M.T.Z. Myint et al. / Applied S

ecently, Khranovskyy et al. reported photoinduced surface wett-bility on different ZnO nanostructures. It showed the effects ofanostructure size upon switching rate and wetting nature underV illumination [19]. Switchable wettability of ZnO microrods was

hown by Han and Gao in which WCA was varied from 0◦ to 151◦

nder cycles of ultraviolet exposure and darkness [20]. In 2010, Lint al. reported the controlled wettability of titanium oxide (TiOx)urfaces using UV–vis (Ultraviolet–Visible) light irradiation [21]. In010, Lv et al. proposed the surface wettability switching on ZnOanorods with different growth times. The proportion of nonpolaro polar surface was mainly attributed to the wettability switchingn the longer growth of ZnO nanorods [22]. Lü et al. have proposedhat the surface wetting transition from hydrophobic to superhy-rophilic on the sodium (Na) doped ZnO thin film occur due to theresence of Na which creates more photo active sites on the sur-ace that can be reversed by annealing over extended times in thembient [23]. The mechanism of light-induced wettability changesn ZnO nanostructure has been reported by many groups [24–28].

We report here an experimental design to observe the effect ofurface coverage of ZnO microrods on hydrophobicity, switchingettability (by applying UV/IR irradiation) of ZnO microrods array

urfaces and optimization of the heating temperature for recov-ry to its original state (hydrophobic state). The purpose of usingnO microrods was to develop a surface that reduces the contactrea between solid and water (in this case solid, liquid and airegime) compared to a flat surface (solid and liquid only) that cane achieved by coating with ZnO nanocrystallites. We also compareesults with our previous work [12] in which surfaces consisting ofatterned microbumps showed superhydrophobic behavior withCA exceeding 150◦ – a straightforward demonstration of the

assie–Baxter case and surface behavior as hypothesized by Duezt al. [10].

. Experimental

.1. Chemicals

Analytical grade zinc acetate dihydrate (Zn(CH3COO)2·2H2O)nd sodium hydroxide (NaOH) (both from MERCK, Germany),inc nitrate hexahydrate (Zn(NO3)2·6H2O) (APS Ajax Finechem,ustralia), hexamethylenetetramine ((CH2)6N4) (Aldrich, USA),nd isopropanol ((CH3)2CHOH) (Lab Scan, Ireland) were used with-ut further purification. Standard microscope glass slides weresed as substrates for zinc oxide microrod growth. The seedingrocess to grow ZnO microrods was accomplished by two methodsescribed below.

.1.1. Preparation of ZnO nanocrystallite for seeding processMethod 1: ZnO nanocrystallite seed stock was prepared by a

ydrolytic procedure described in detail elsewhere [29]. Briefly, solution of 20 mM sodium hydroxide in isopropanol was addedrop by drop to 1 mM zinc acetate solution under continuous stir-ing. The mixture was hydrolyzed at 60◦C for 2 h to form ZnOanocrystallites. Nanocrystallites were then deposited dropwisesolution dropping method) on clean glass substrates kept constantt 60◦C. Following the coating with ZnO crystallites, the substratesere annealed at 250 ◦C in air to remove any un-reacted chemicals

n the surface.

Method 2: ZnO nanocrystallites were also formed by the decom-

osition of zinc acetate directly on glass substrates by thermalreatment in air. When the temperature of the glass substrateeached ca. 350 ◦C, zinc acetate decomposed into ZnO to formanocrystallites. These nanocrystallites were used as seeds for therowth of ZnO microrods [30].

Science 264 (2013) 344– 348 345

2.1.2. Hydrothermal growth of ZnO microrodsFollowing seeding on glass substrates (that are naturally

hydrophilic), microrods were grown by a hydrothermal process ina chemical bath containing equimolar (20 mM) solutions of zincnitrate hexahydrate and hexamethylenetetramine at 90 ◦C [30].The hydrothermal growth process of ZnO micro and nanorods arewell understood with several reviews available in the literature[31–34]. Depending upon the concentration of the reactants andthe temperature of growth in the hydrothermal bath, the formationof regularly shaped rods with varying widths are observed [30].The growth was carried out for 20 h with the precursor solutionreplaced every 5 h to replenish zinc ions [29]. ZnO microrod coatedsubstrates were carefully rinsed several times with deionized (DI)water and annealed at 250 ◦C for 1 h to remove un-reacted organicmatter. As-prepared samples were dried at 100 ◦C for a few hoursprior to all measurements.

2.2. Water contact angle (WCA) measurements and surface areaestimation

Water droplet of 5 �L was used for all WCA measurements.A minimum of five measurements for each surface on differentportion of the substrates were conducted to provide a mean value.WCA was recorded with a customized contact angle instrumentequipped with Dino Lite Pro AM 413T digital microscope camera.WCA values were determined and analyzed with ImageJ Analysissoftware following a method described by Stalder et al. [35]. Thedimensions of ZnO microrod were estimated from scanning elec-tron microscope (SEM) images. Coverage area (SA) of ZnO microrodswas also estimated from SEM images using the image analysis soft-ware.

The wettability conversion experiments were conducted usingcommercial UV lamp equipped with 6 W dual lamps yielding about1.0 mW/cm2 intensity. A standard furnace was used for heating.Both the experiments were conducted under ambient atmospheric(air) conditions with relative humidity (RH) ∼ 65%.

3. Results and discussion

3.1. Wettability vs. surface morphology

In our laboratory findings, the WCA of a solid thin layer of ZnOformed by nanocrystallites deposited on glass substrates was foundto be 26◦ ± 3◦ (this WCA is called intrinsic contact angle), whichconfirm that continuous, un-structured and untreated ZnO surfaces(so called flat surfaces) are hydrophilic in nature [12].

Fig. 1a shows scanning electron micrograph (SEM) image of ZnOmicrorods grown on a seeded glass substrate following Method1 [29]. The ZnO microrods were ca. 5.0 ± 0.5 �m in height witha width of 1.2 ± 0.1 �m as estimated from the cross section (notshown here) and top view of SEM images (Fig. 1a). The purpose ofusing 5.0 �m long ZnO rods is that hydrophobicity of unitary (singleroughness) structured ZnO surfaces saturate for longer rod arrays[12]. Coverage area, the area covered by ZnO microrods (SA) of sam-ples was found to be less than 39% in these samples (Fig. 1a) whichis also reported in our previous work [12]. On the other hand, theremaining area of 61% was air. However, the estimated SA of theglass substrate covered with ZnO microrods was certainly overes-timated because ZnO microrods were not all positioned normal tothe surface. The static WCA was found to be ca. 123◦ ± 3◦ which isa demonstration of the formation of hydrophobic surface through

the microstructuring of a native hydrophilic material (ZnO).

Fig. 1b represents a typical SEM image of ZnO microrods grownon seeded substrate synthesized following Method 2. On an aver-age, the surface coverage was found to be higher than 60% (in the

346 M.T.Z. Myint et al. / Applied Surface Science 264 (2013) 344– 348

Fig. 1. (a) Scanning electron microscopy (SEM) image of hydrothermally grown ZnO microrods on glass substrate (low density growth of ZnO microrods due to the ZnOn icroroo eededs

pis

lsti(

alcaye

c

w

c

wch

CWluFtmaC

Fa

with published results as shown in Fig. 3. In this case, types ofmaterials (for e.g. silicon [38–41], PDMS (polydimethylsiloxane)[42], polyurethane [43], and ZnO [6,12]) used to structure the sur-faces are not considered and only microstructured surfaces are

anocrystallites seeded layer: Method 1) (inset: 5 �L water droplet sits on the ZnO mn glass substrate (high density growth of ZnO microrods due to the zinc acetate substrate).

resent example, the surface coverage was estimated to be ca. 62%)n which less than 40% area was placed with air. On this surface, thetatic WCA was found to be ca. 16◦ ± 3◦.

Duez et al. predicted that for rough surfaces with coverage (SA)ess than 40% should show hydrophobic character (i.e. the WCAtarts to exceed 90◦) [10]. Our results are in good agreement to theheoretical prediction of Duez et al. [10] as increased hydrophobic-ty is observed with decreasing surface coverage of the substratesFig. 1a and b).

In the above case, the classical Cassie–Baxter criteria can bepplied to the so called “Fakir” surface (solid-air composite with noiquid penetrating into the pillar supports): considering an intrinsicontact angle of �i with the solid area fraction ϕ and the remainingrea fraction of air given by (1 − ϕ), which leads to a perfect surfaceielding water contact angle of 180◦, the equilibrium Cassie–Baxterquation can be expressed by,

os �F = ϕ(cos �i + 1) − 1 (1)

here �F is the apparent contact angle of the Fakir state.Cassie–Baxter stable state holds true when,

os �i <ϕ − 1r − ϕ

(2)

here r is the roughness factor. High roughness factor with lowoverage area fraction is the key factor that leads to a stable stateydrophobic regime [12,36].

The low solid area (<40%) surfaces followed the metastableassie–Baxter state (Eq. (2) & Fig. 2a) that showed the apparentCA of 123◦ ± 3◦. The water droplet slightly invades into the pil-

ars, but air packets still exist under water droplet, to cause thenstable state with high adhesion as schematically represented inig. 2a. The estimated solid area fraction following Eq. (1) was found

o be 0.24 compared to the experimental value of 0.39 (overesti-

ated due to the slanted ZnO microrods) for the solid coveragerea of <40% which is thus reasonably explained by the metastableassie–Baxter equation.

ig. 2. Schematic representation of (a) metastable Cassie–Baxter regime (solid, liquidnd air regime) and (b) hemiwicking or water impregnating the solid texture.

ds coated glass substrate). (b) SEM image of hydrothermally grown ZnO microrods layer: Method 2) (inset: 5 �L water droplet sits on the ZnO microrods coated glass

For high solid area fraction (>40%), the surface was completelywet and water penetrates into the pillars. It is hard to conclude thatthe wetting state is in the Wenzel regime [2] because the theoreticalvalue does not fit to our experimental estimates based on roughnessfactor (r = 10.33) and intrinsic contact angle (�i = 26◦). However, theexperimental scenario can be described by the so called “hemi-wicking” state (Fig. 2b) [36,37]. In this case, the surface wettabilitylies between spreading and imbibitions whereby the intermediateWCA of � = 0◦ and � < �/2 respectively. In this hemiwicking case, thecritical angle can be expressed as:

cos �C = ϕ − 1r − ϕ

(3)

The rougher the surface, the larger the critical angle �c observed,leading to hemiwicking.

From our previous work, micropatterned arrayed surfaces withdiameter of single microbump of around 150 �m were built on glasssubstrates was reported [12]. From surface morphology analysis,the coverage area was found to be ca. 4.3% (close to the coverageestimated for natural lotus leaf surface 3–4%) and WCA of 153◦ ± 3◦

was achieved. The calculated value of roughness factor was foundto be 5.8 which is equivalent to an area fraction of 0.046. The esti-mated static WCA was ca. 156◦ which is in good agreement withthe experimental observation (153◦ ± 3◦). Moreover, the theoret-ical estimation and experimental results match reasonably well

Fig. 3. Summarized values of water contact angle with respect to coverage areaof micro/nanorod coated surfaces and the resulting decrease in WCA. The valuesderived from the previous reports and the results indicate that the physical propertyof the coverage area of a surface directly impacts the water adhesion on surfaces(trend line is a guide to the eyes).( , Barberoglou et al. [38]; , Bhushan et al. [40]; , Feng et al. [43]; , Kim et al.[42]; , Kwon et al. [39]; , Li et al. [6]; , Zhu et al. [41]; , Myint et al. (thiswork and previous work [12]) (averaged over five measurements within the errorbar: ±3◦)) (experimental result (>40% coverage area) is shown in Fig. 1b)

M.T.Z. Myint et al. / Applied Surface Science 264 (2013) 344– 348 347

Fim

cim

3

zhasanmt(ficrateWem

3

ouhmsqbovscebh[ctg[

Fig. 5. Hydrophilic surfaces were annealed in the furnace after prolonged irradiationunder UV light at different temperatures to restore original hydrophobic properties.The wettability of the sample substrates was measured at regular intervals throughwater contact angle measurements( : annealed at 120 ◦C; : annealed at 200 ◦C; : annealed at 250 ◦C) (errorbar = ± 3◦).

ig. 4. The contact angle changes on the ZnO microrods surfaces with increasing UVllumination time (inset: typical optical images of the droplets sitting on the ZnO

icrorods surfaces) (error bar = ±3◦).

onsidered to calculate the coverage area. The purpose of target-ng microstructure is that the formation of hydrophobic surfaces

ainly depends on the microstructure.

.2. Photoinduced phenomenon on ZnO microstructured surfaces

Semiconductor materials such as titanium dioxide (TiO2) andinc oxide (ZnO) surfaces can be turned from hydrophobic toydrophilic upon irradiation with UV light [44–46]. The charge cre-tion on surfaces of titania and ZnO have quite a lot of similarityince they are both metal oxide semiconductors and the surfacesre generally hydrolyzed in the ambient. We applied this phe-omenon to the surface formed with ZnO microrods. When ZnOicrorods are irradiated by UV light with photon energy higher

han or equal to the bandgap of ZnO (3.37 eV), electron (e−)–holeh+) pair forms. The migration of holes to the ZnO microrod sur-aces take place upon the recombination with trapped electronsn O2 whereby releasing O2− from the surface (O2− + h+ → O2 (g))reating oxygen vacancies on ZnO surfaces. Some of the electronseact with Zn2+ ions to form Zn+ surface defects, often referred tos surface trapped electrons. In the other hand, unpaired free elec-rons are trapped by re-adsorption of O2 on the ZnO surface until anquilibrium state of release and re-adsorption of O2 [47] is reached.hen UV illumination is taken off, holes quickly recombine with

xcited electrons but excess electrons remain trapped on the ZnOicrorods surface.

.3. Wettability vs. UV/IR irradiation

The wetting contact angle on ZnO microrods surface decreasedver time following exposure to UV light as shown in Fig. 4. Grad-al increase in hydrophilicity (∼180 min irradiation to achieve aydrophilic surface as shown in Fig. 4) can be attributed to the for-ation of oxygen vacancies on the surface of ZnO crystals due to

low release of oxygen during UV illumination [23,48,49]. Conse-uently, unpaired Zn ions on the ZnO surface may coordinativelyind to oxygen atom of water molecule leading to the formationf a thin water layer on (0 0 1) surface of ZnO crystal [50]. Theseacant sites can then interact with water molecules causing the dis-ociative adsorption of water molecules on the ZnO surfaces whichan lead to increased water adsorption on the surface [18]. How-ver, the dissociation of water molecule on the surface is triggeredy a second water molecule from neighboring lattice site throughydrogen bonding which leads to slow hydrophilization process50,51]. Finally, the defect sites (oxygen vacancy) are more kineti-

ally favorable for the attachment of hydroxyl group which distorthe surface electronic and geometric structure rendering it ener-etically unstable whereby the surface becomes more hydrophilic23].

Fig. 6. The reversible wettability conversion (hydrophobic–hydrophilic) of ZnOmicrorod coated glass substrate under UV illumination (1.0 mW/cm2) and annealing(heating) in the ambient.

Hydrophobicity was restored by heating at temperatures abovethe boiling point of water for approximately 60 min. Heatingtemperatures of 120 ◦C and 200 ◦C were found to be suitablefor regeneration of the hydrophobicity on ZnO microrod surface(Fig. 5). This is due to the availability of sufficient energy to removehydroxyl groups from the surface at these temperatures. A favor-able adsorption of oxygen from the ambient replaces the vacantsites leading to a relaxation of the electronic states on the sur-face. Upon heating the ZnO microrods at higher temperature suchas 250 ◦C or above, the oxygen vacancy sites or defective sitesare more favorable for adsorption of hydroxyl group than oxygen[52]. Moreover, the surface defects are also created at tempera-tures higher than 230 ◦C on ZnO surface [18]. From steady-statephotoluminescence measurements of ZnO nanorods annealed atdifferent temperatures we observed that the maximum defect orig-inated emission at 590 nm, attributed to the surface defects wereobserved in case of the ZnO nanoparticles annealed at 250 ◦C. Thisindicates the highest concentration of the Vo++ defect states in theZnO crystal at 250 ◦C which would affect the orientation of adsorbedwater molecules on the ZnO nanorod surfaces hence leading to thehydrophilicity of these surfaces. The consequence of prefereableadsorption of hydroxyl groups on the defective surfaces maintainthem hydrophilic (Fig. 5).

The irradiated wettability (hydrophobic–hydrophilic) switchingmechanism of ZnO microrod coated glass substrates was carried outseveral times and showed good reversibility as shown in Fig. 6.

4. Conclusion

We have utilized simple seeding techniques to grow ZnO micro-rods on glass substrates followed by a hydrothermal growth

3 urface

pehfa

Ubaau

A

faaNSE

R

[

[[

[

[[[[

[

[

[[[

[

[

[

[[[

[[[[

[

[

[

[[[

[[[[

[[

[[

[

[

[

48 M.T.Z. Myint et al. / Applied S

rocess to fabricate arrays of microrods with different surface cov-rage. The ideal coverage area (solid area fraction) for preparingydrophobic surfaces with ZnO microrods on glass substrates was

ound to be less than 40% with higher surface coverage leading to hydrophilic surface.

We also demonstrated that exposure of ZnO microrods to 3 h ofV light led to the generation of hydrophilic surfaces, which coulde restored to its original hydrophobic state by heating at 120 ◦Cnd 200 ◦C for 1 h. Based on these results, it is feasible to implement

simple hydrophobic/hydrophilic UV/IR based switching surfacessing ZnO microrods.

cknowledgements

The authors would like to acknowledge partial financial supportrom the National Nanotechnology Center of the National Sciencend Technology Development Agency (NSTDA), Ministry of Sciencend Technology (MOST), Thailand and the Centre of Excellence inanotechnology at the Asian Institute of Technology, Thailand, andheikh Saqr Al Qasimi Graduate Research Fellowship, United Arabmirates.

eferences

[1] A.B.D. Cassie, S. Baxter, Trans. Faraday Soc. 40 (1944) 546–551.[2] R.N. Wenzel, Ind. Eng. Chem. 28 (1936) 988–994.[3] W. Barthlott, C. Neinhuis, Planta 202 (1997) 1–8.[4] W. Pfleging, R. Kohler, M. Torge, V. Trouillet, F. Danneil, M. Stüber, Appl. Surf.

Sci. 257 (2011) 7907–7912.[5] H. Shao, X. Qian, B. Huang, Mater. Sci. Semicond. Process. 10 (2007) 68–76.[6] Y. Li, W. Cai, G. Duan, B. Cao, F. Sun, F. Lu, J. Colloid Interface Sci. 287 (2005)

634–639.[7] C.I. Park, H.E. Jeong, S.H. Lee, H.S. Cho, K.Y. Suh, J. Colloid Interface Sci. 336

(2009) 298–303.[8] J. Wu, J. Xia, W. Lei, B.-P. Lei, Mater. Lett. 64 (2010) 1251–1253.[9] L. Mahadevan, Nature 411 (2001) 895–896.10] C. Duez, C. Ybert, C. Barentin, C. Cottin-Bizonne, L. Bocquet, J. Adhes. Sci. Technol.

22 (2008) 335–351.11] A. Shastry, M.J. Case, K.F. Böhringer, Langmuir 22 (2006) 6161–6167.12] M.T.Z. Myint, R. Kitsomboonloha, S. Baruah, J. Dutta, J. Colloid Interface Sci. 354

(2011) 810–815.13] S.N. Das, J.H. Choi, J.P. Kar, J.M. Myoung, Appl. Surf. Sci. 255 (2009)

7319–7322.14] B. Bhushan, M. Nosonovsky, C.J. Yong, J. R. Soc. Interface 4 (2007) 643–648.15] Y. He, C. Jiang, H. Yin, W. Yuan, Appl. Surf. Sci. 257 (2011) 7689–7692.16] T. Ning, W. Xu, S. Lu, J. Colloid Interface Sci. 361 (2011) 388–396.17] X. Hou, F. Zhou, B. Yu, W. Liu, Mater. Sci. Eng. A 452-453 (2007) 732–736.

[

[[

Science 264 (2013) 344– 348

18] R.D. Sun, A. Nakajima, A. Fujishima, T. Watanabe, K. Hashimoto, J. Phys. Chem.B 105 (2001) 1984–1990.

19] V. Khranovskyy, T. Ekblad, R. Yakimova, L. Hultman, Appl. Surf. Sci. 258 (2012)8146–8152.

20] J. Han, W. Gao, J. Electron. Mater. 38 (2009) 601–608.21] Z. Lin, G.-H. Lee, C.-M. Liu, I.-S. Lee, Surf. Coat. Technol. 205 (2010) S391–S397.22] J. Lv, J. Zhu, K. Huang, F. Meng, X. Song, Z. Sun, Appl. Surf. Sci. 257 (2011)

7534–7538.23] J. Lü, K. Huang, X. Chen, J. Zhu, F. Meng, X. Song, Z. Sun, Appl. Surf. Sci. 257

(2011) 2086–2090.24] E.L. Papadopoulou, M. Barberoglou, V. Zorba, A. Manousaki, A. Pagkozidis, E.

Stratakis, C. Fotakis, J. Phys. Chem. C 113 (2009) 2891–2895.25] J. Lü, K. Huang, X. Chen, J. Zhu, F. Meng, X. Song, Z. Sun, Appl. Surf. Sci. 256

(2010) 4720–4723.26] M. Miyauchi, A. Shimai, Y. Tsuru, J. Phys. Chem. B 109 (2005) 13307–13311.27] D. Sirbu, A.P. Rambu, G.I. Rusu, Mater. Sci. Eng. 176 (2011) 266–270.28] G. Kenanakis, E. Stratakis, K. Vlachou, D. Vernardou, E. Koudoumas, N. Kat-

sarakis, Appl. Surf. Sci. 254 (2008) 5695–5699.29] S. Baruah, J. Dutta, Sci. Technol. Adv. Mater. 10 (2009) 013001–013010.30] S. Baruah, J. Dutta, J. Sol-Gel Sci. Technol. 50 (2009) 456–464.31] S. Baruah, J. Dutta, J. Cryst. Growth 311 (2009) 2549–2554.32] S. Baruah, C. Thanachayanont, J. Dutta, Sci. Technol. Adv. Mater. 9 (2008)

025008–025016.33] R. Kitsomboonloha, S. Baruah, M.T.Z. Myint, V. Subramanian, J. Dutta, J. Cryst.

Growth 311 (2009) 2352–2358.34] A. Sugunan, H.C. Warad, M. Boman, J. Dutta, J. Sol-Gel Sci. Technol. 39 (2006)

49–56.35] A.F. Stalder, G. Kulik, D. Sage, L. Barbieri, P. Hoffmann, Colloids Surf. A 286 (2006)

92–103.36] J. Bico, U. Thiele, D. Quéré, Colloids Surf. A 206 (2002) 41–46.37] D. Quéré, Annu. Rev. Mater. Res. 38 (2008) 71–99.38] M. Barberoglou, V. Zorba, E. Stratakis, E. Spanakis, P. Tzanetakis, S.H. Anas-

tasiadis, C. Fotakis, Appl. Surf. Sci. 255 (2009) 5425–5429.39] Y. Kwon, N. Patankar, J. Choi, J. Lee, Langmuir 25 (2009) 6129–6136.40] B. Bhushan, Y. Chae Jung, Ultramicroscopy 107 (2007) 1033–1041.41] L. Zhu, Y. Feng, X. Ye, Z. Zhou, Sens. Actuators A 130–131 (2006) 595–600.42] D.S. Kim, B.K. Lee, J. Yeo, M.J. Choi, W. Yang, T.H. Kwon, Microelectron. Eng. 86

(2009) 1375–1378.43] J. Feng, B. Huang, M. Zhong, J. Colloid Interface Sci. 336 (2009) 268–272.44] D.C. Hennessy, M. Pierce, K.-C. Chang, S. Takakusagi, H. You, K. Uosaki, Elec-

trochim. Acta 53 (2008) 6173–6177.45] X. Zong, P. Wang, Physica E: Low Dimens. Syst. Nanostruct. 41 (2009) 757–761.46] X. Feng, L. Feng, M. Jin, J. Zhai, L. Jiang, D. Zhu, J. Am. Chem. Soc. 126 (2004)

62–63.47] Y. Li, F. Della Valle, M. Simonnet, I. Yamada, J.J. Delaunay, Appl. Phys. Lett. 94

(2009) 023110–023112.48] T. Watanabe, S. Fukayama, M. Miyauchi, A. Fujishima, K. Hashimoto, J. Sol-Gel

Sci. Technol. 19 (2000) 71–76.49] A. Nakajima, K. Hashimoto, T. Watanabe, Monatshe. Chem. 132 (2001) 31–41.

50] B. Meyer, D. Marx, O. Dulub, U. Diebold, M. Kunat, D. Langenberg, C. Wöll,

Angew. Chem. Int. Ed. 43 (2004) 6642–6645.51] O. Dulub, B. Meyer, U. Diebold, Phys. Rev. Lett. 95 (2005) 136101–136104.52] R. Wang, N. Sakai, A. Fujishima, T. Watanabe, K. Hashimoto, J. Phys. Chem. B

103 (1999) 2188–2194.