Ss

AJa

b

c

d

a

ARR1AA

KBGMSH

1

cfivwshoa

igsaodut

0h

Applied Surface Science 289 (2014) 586– 591

Contents lists available at ScienceDirect

Applied Surface Science

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

urface treatment of polymer microfibrillar structures for improvedurface wettability and adhesion

mirpasha Peyvandia, Saqib Ul Abideenb, Yue Huangc, Ilsoon Leed, Parviz Soroushiana,ue Lub,∗

Department of Civil and Environmental Engineering, Michigan State University, East Lansing, MI 48824-1226, United StatesTechnova Corporation, 1926 Turner Street, Lansing, MI 48906, United StatesLabSys LLC, 1432 Glenhaven Ave, East Lansing, MI 48823, United StatesDepartment of Chemical Engineering and Materials Science, Michigan State University, East Lansing, MI 48824-1226, United States

r t i c l e i n f o

rticle history:eceived 12 September 2013eceived in revised form1 November 2013ccepted 12 November 2013vailable online 18 November 2013

a b s t r a c t

The effects of altering the polymer surface characteristics on adhesion qualities of bio-inspired fibrillaradhesives were found to be significant. Treatment of fibril tip surfaces in polymer fibrillar adhesivesimproved their wettability and adhesion capacity. Surface modifications of fibril tips involved UV/Ozoneand oxygen plasma treatments for making the fibril tips more hydrophilic. These surface treatmenteffects, however, tend to degrade over time (rendering hydrophobic recovery). The stability of treated(hydrophilic) surfaces was improved, while retaining their wettability, through coating with a polyelec-

eywords:io-inspired adhesiveseckoicrofibrillar structures

urface treatmentydrophilic

trolyte such as polyethyleneimine (PEI) via self-assembly.© 2013 Elsevier B.V. All rights reserved.

. Introduction

Used by lizards and insects (including geckos, spiders, beetles,rickets and flies) to climb vertical and even inverted surfaces, thene-hair (fibrillar) adhesive system is an excellent example of con-ergent evolution in biology. The density of surface hairs increasesith the body weight of animal, and gecko has the highest hair den-

ity and the finest (nano-scale) hairs among all animal species thatave been studied [1]. Gecko’s nano-scale fibrillar structure devel-ps high adhesion capacity with broad ranges of surface materialst high reliability levels [1–3].

The structure and unique capabilities of the gecko-foot haventrigued biologists and engineers for many years. Several investi-ations have been conducted into various aspects of the gecko-foottructure and behavior [4–6]. Current efforts to mimic gecko’sdhesion mechanism can be divided into two broad categories:ne uses arrays of relatively soft elastomeric fibrils such as poly-

imethylsiloxane (PDMS) and polyurethane (PU) [6,7]; the otherses arrays of high-modulus and high-aspect-ratio carbon nano-ube or nanofibers [8]. While many investigators identify van der

∗ Corresponding author. Tel.: +1 517 485 1402; fax: +1 517 485 1402.E-mail address: juelu66@yahoo.com (J. Lu).

169-4332/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.apsusc.2013.11.048

Waals interactions as the primary adhesion mechanism of fibril-lar structures, some debate that capillary force, also contribute togecko-foot adhesion to a variety of surfaces over a broad rangeof relative humidity [9,10]. Theoretical investigations indicate thatcapillary force can make important contributions to the adhesioncapacity of fibrillar arrays [11,12], as this cohesive force is verylarge compared to van der Waals interaction [13]. In addition, adhe-sion against rough surfaces may benefit more from the capillaryeffect, as the capillary water bridging two surfaces is not signifi-cantly affected by the surface roughness [14]; the van der Waalsforce, on the other hand, is a distinctly short-distance force thatis significantly affected by surface roughness. However, the highlyhydrophobic nature of polymer microfibrillar structures is a disad-vantage in building capillary bridges with the substrate [14].

Different surface modification techniques can be used totransform hydrophobic polymeric surfaces into hydrophilic ones.Examples of such techniques include multicomponent polyaddi-tion reaction, corona discharges, oxygen plasma treatment, andUV irradiation with or without ozone [15,16]. UV/Ozone (UVO)treatment has been extensively applied to natural and synthetic

polymers towards modification of the surface chemistry and wet-ting characteristics. It has been used for enhancement of interfacialadhesion in adhesive joints and composites [17]. UVO treatment isa photosensitized oxidation process which causes excitation and

rface

daoTfiasrtoppnnaseppctobwmSw

wtast

2

2

2

pvdp2fppNpppTfp

2

B1w(eS

A. Peyvandi et al. / Applied Su

issociation of polymer molecules by short-wavelength UV radi-tion; in addition, UV light dissociates oxygen to generate atomicxygen which easily reacts with oxygen molecules to form ozone.he highly reactive ozone reacts with polymers to form polarunctional groups, such as peroxyl, hydroxyl, carbonyl, etc., thusncreasing the wettability of polymer surfaces [18]. Anotherpproach to improvement of the water-wettability of polymerurfaces involves oxygen plasma treatment [16,19], which is envi-onmentally friendly and easy to implement. Oxygen plasmareatment causes breakage of chemical bonds due to bombardmentf the polymer surface with ions of high energy, and introducesolar groups to the polymer surface. UVO or plasma treatment ofolymer surfaces can, under proper operating conditions, generateanostructured surfaces [20]. This feature can be used to produceano-fibrillar structures on micro-fibril tips, thus producing hier-rchical structures which can, similar to gecko-foot, conform tourface roughness at different scales. However, hydrophobic recov-ry of oxidized polymer surfaces, caused by reversible relaxationrocesses of polar groups, can occur over time after UVO or oxygenlasma treatment of polymer surfaces. Other mechanisms can alsoontribute to this trend; in the case of PDMS, for example, migra-ion of free siloxanes from the bulk to the surface through a porousr cracked hydrophilic silica-like layer contributes to hydropho-ic recovery [21]. The hydrophobic recovery of polymer surfacesill affect its adhesion. Menezes and Doi studied adhering poly-er molecules of desired qualities to treated polymer surfaces [15].

tability of hydrophilic surfaces can be improved through coatingith a polyelectrolyte with nano-thickness via self-assembly [22].

In this study, the polymer microfibrillar structures were treatedith UVO and oxygen plasma to render stable changes in the fibril

ip morphology and the hydrophilic qualities, and their effect ondhesion properties of polymer microfibrils were examined. Thetable hydrophilic surfaces of polymer microfibrils were achievedhrough further coating of a thin layer of polyethyleneimine (PEI).

. Materials and methods

.1. Fabrication process of fibrillar arrays

.1.1. Fabrication of silicon templateFibrillar arrays were produced by soft-molding of elastomeric

recursors on a micro-fabricated silicon template [23]; thisersatile approach enables fabrication of fibrils with differentimensions. Photolithography was used to create the master tem-late with pattern of holes with a diameter of 20 �m, a height of0 �m and a center-to-center distance of 30 �m. For the purpose ofabricating the lithographic SU-8 templates, the silicon wafer wasretreated using a 3-step RCA cleaning process. The first step iserformed by immersing the silicon wafer in a 1:1:5 solution ofH4OH + H2O2 + H2O at 75 ◦C, and rinsing with DI water. Since thehotoresist is hydrophobic, surface dehydration is performed toromote adhesion by baking in oven at 200 ◦C for 20 min. The SU-8hotoresist was then spin-coated on the silicon wafer at 2000 rpm.he thickness of the SU-8 film was verified using a DekTak3 sur-ace profiler. The patterns were created by the photo-lithographicrocess.

.1.2. Soft-molding on the lithographic templatesFibrillar arrays were made with polyurethane (PU) (ST-3040,

JB Enterprises, Inc) or polydimethylsiloxane (PDMS) (Sylgard84, supplied by Dow Corning). The templates were silanized

ith heptadecafluoro-1,1,2,2-tetrahydrooctyltrichlorosilane

hepta-fluorosilane). Gas-phase silanization was performed in anvacuated desiccator for 1 h, followed by baking at 95 ◦C for 1 h.ylgard 184 prepolymer and cross-linker (at a ratio of 10:1) or PU

Science 289 (2014) 586– 591 587

ST-3040 A and B (20:17 by weight) was mixed, degassed, and thenpoured on the silanized template. The PU and PDMS were curedat room temperature and 65 ◦C, respectively, in light vacuum over24 h, and then carefully demolded to avoid rupture of the polymermicrofibrillar array. The total thickness of the elastomer sheet(with fibrillar surface) was approximately 1 mm.

2.2. Surface treatment of fibrillar arrays

A Photo Surface Processor (Model: PL16-110, SEN Lights Cor-poration) was used for UVO treatment of PU fibrillar arrays. Thedegree of UVO treatment can be controlled by the time of expo-sure and the distance to the sample. The fibrillar structures weretreated for 30 min; this duration was selected through a trial-and-adjustment approach. Oxygen plasma treatment was carried outon PU and PDMS surfaces for 1–5 min at 1000 mTorr and an oxygenflow rate of 18.3 ml/min, using a Harrick Plasma Cleaner. To coatpolymer fibrils with a thin layer of polyelectrolyte, a solution of1 wt% polyethyleneimine (Aldrich) was prepared, with pH adjustedto 6.5, and the polymer fibrillar arrays, after oxygen plasma treat-ment, were placed in this solution over a 1-h period; it was thenremoved from the solution, rinsed with DI water, and dried undervacuum.

2.3. Surface characterization

Scanning electron microscopy (SEM, JEOL 6300F) and opti-cal microscopy (DC5-163 digital compound microscope, NationalOptical) were used to study the morphology of polymer microfib-rillar structures. Atomic force microscopy (AFM) surface imaging(Nanoscope IV Multimode SPM from Veeco) was also used to inves-tigate the effect of surface treatment on fibril tips. Samples werescanned in tapping mode using an NSC 15 probe with nominalfrequency of ∼300 kHz and scan rate of about 0.5 Hz. Scans wereperformed at different scales in the range of 500 nm to 100 �m.Water contact angle measurement was performed using the sessiledrop technique with a manual goniometer (Rame Hart, Inc Model100-00 115 NRL-C.A.)

2.4. Adhesion test

Tension and shear adhesion capacities of 1 cm × 1 cm specimensof fibrillar arrays were measured against the substrates, includingglass slides and polyvinyl chloride (PVC) films in a closed room withcontrolled humidity. A pulley setup built in house was used for theadhesion test. The substrates were sonicated in distilled water for15 min, and then blow dried with N2 gas. A preload pressure of5 N/cm2 was applied for 1 min to establish adhesion prior to per-formance of tension and shear adhesion tests. At least three sampleswere evaluated for each condition, and the standard deviation wascalculated.

3. Results and discussion

3.1. UVO surface treatment

Fig. 1(a) shows a SEM image of PU fibrillar array with fibrildiameter and length of 20 �m fabricated from the lithographic tem-plate. Fig. 2 compares the water contact angles of PU plain film andfibrillar array prior to and after UVO treatment. The water contactangle of the plain PU film is 73◦, which is consistent with valuesreported in the literature (in the range of 65◦ to 75◦) [18]. Therefore,

the nature of PU surface is relatively hydrophobic. The PU fibril-lar array shows a much higher water contact angle (120◦), whichcan be attributed to the patterned surface morphology, similar tothe lotus effect [24]. UVO treatment of PU plain film and fibrillar

588 A. Peyvandi et al. / Applied Surface Science 289 (2014) 586– 591

F fore at f PU fi

a1tnactstapfttohi

Fp

to UVO treatment. These results could be explained by the greater

ig. 1. Morphologies of PU fibrillar array with 20 �m fibril diameter and length bereatment, (c) optical image of PU fibrils after 15-min UVO; and (d) optical image o

rray dramatically reduced their water contact angles to 15◦ and7◦, respectively; UVO-treated PU surfaces are thus highly wet-able. This effect, as noted above, results from the highly reactiveature of ozone gas, which induces chain scission and function-lization of PU polymer with polar functional groups. Fig. 1(b–d)ompares the side view of PU fibrils before and after different UVOreatment times. The edges of fibrils are rounded after UVO expo-ure more than 30 min. Study by Kuang et al. argued that the UVOreatment only breaks the long chain (–C–C–) bonds and insertstomic oxygen and ozone molecules at the chain ends to createolar functional groups without removing material from the sur-ace [18], which is not consistent with our observation. It is clearhat UVO treatment has etching effects, similar to oxygen plasma

reatment. Fig. 3 shows an atomic force microscope (AFM) imagef a PU fibril tip after UVO treatment. The UVO-treated fibril tipas increased roughness and grain size, embodies fine morpholog-

cal features which could benefit the adhesion capacity of fibrillar

ig. 2. Comparison of the water contact angles of PU plain film and fibrillar arrayrior to and after UVO treatment.

nd after UVO treatment: SEM (a) and optical (b) images of PU fibrillar array beforebrils after 30-min UVO.

arrays by improving the fibril tip conformability against surfaceroughness. The UVO treatment of PDMS microfibrillar arrays wasunsuccessful with the same process, even with much extendedexposure time, which is possibly due to limited ozone generation,though surface modification of PDMS to hydrophilic surface wasreported in the literature [25].

The adhesion capacities of PU fibrillar arrays prior to and afterUVO treatment were evaluated against glass at a relatively humid-ity of 50% (Fig. 4). UVO treatment is observed to significantlyimprove the adhesion capacity of the PU fibrillar array. The shearand tensile adhesion capacities of PU fibrillar arrays after UVO treat-ment were about three and seven times, respectively, those prior

contributions of capillary effect to the adhesion capacity of fibril-lar arrays with more hydrophilic surfaces. Alteration of fibril tip

Fig. 3. AFM image of UVO-treated PU fibril tip.

A. Peyvandi et al. / Applied Surface Science 289 (2014) 586– 591 589

FU

mt

ir(gwwaa

3

amsPcd2oipPssfiab

FU

Fig. 6. Comparison of the water contact angles of PDMS and PU fibrillar array priorto and after oxygen plasma treatment.

with UVO or oxygen plasma, which is possibly due to the lossof low molecular weight organic material created on the surface

ig. 4. Shear and tensile adhesion capacities of PU fibrillar arrays prior to and afterVO treatment.

orphology upon UVO treatment could be another factor con-ributing to improve adhesion capacity.

It is known that capillary force depends on both relative humid-ty and the surface energy of contacting surfaces. The effects ofelative humidity on the shear and tensile adhesion capacitiesagainst the glass substrate) of PU fibrillar arrays were investi-ated (Fig. 5). Both shear and tensile adhesion capacities improvedith increasing relative humidity (RH) up to a peak value, beyondhich further RH rise produces lower adhesion capacities. Shear

nd tensile adhesion capacities against the glass substrate peakedt approximately 50% and 75% relative humidity, respectively.

.2. Oxygen plasma surface treatment

The other pretreatment approach, plasma treatment, is easy, fastnd environmentally friendly. The effect of oxygen plasma treat-ent on both PDMS and PU microfibrillar structures were also

tudied. Fig. 6 presents change in surface water contact angles ofDMS and PU before and after oxygen plasma treatment. The waterontact angles of PDMS and PU microfibrillar structures slightlyropped after 30 s plasma treatment, and further dropped to below0◦ after 1 min, indicating a completely hydrophilic surface. Theptical microscope observation showed the tips of the fibrils werentact (images are not shown). Fig. 7 shows the effects of oxygenlasma treatment on the shear and tensile adhesion capacities ofDMS and PU fibrillar arrays, respectively, against the glass sub-trate at 50% relative humidity. Both shear and tensile adhesion

lightly increased after oxygen plasma treatment, indicating bene-t of the hydrophilic surface on the adhesion capacity of PU fibrillarrrays; though when compared with UVO treatment, the contri-ution of oxygen plasma treatment to adhesion capacity is less

ig. 5. Effect of relative humidity on the shear and tensile adhesion capacity ofVO-treated PU fibrillar array against glass.

Fig. 7. Shear and tensile adhesion capacities of PDMS and fibrillar arrays prior toand after oxygen plasma treatment.

pronounced, which confirms that alteration of fibril tip morphol-ogy resulting from UVO treatment results in a greater improvedadhesion capacity of the microfibrils.

3.3. Improvement of the hydrophilic stability of treated fibrillararrays

Hydrophobic recovery is common for polymer surfaces treated

during the treatment and reorientation of side groups from thebulk to the surface [21]. As shown in Fig. 8, the water contact

Fig. 8. Change in water contact angles of PDMS microfibrillar array after oxygenplasma treatment with and without further PEI coating as a function of time.

590 A. Peyvandi et al. / Applied Surface

Ft

aw9sifcfhsossai

taipfifotPtaota

Fp

ig. 9. Change in water contact angles of PU microfibrillar array after oxygen plasmareatment with and without further PEI coating as a function of time.

ngle of PDMS microfibrils after 5 min oxygen plasma treatmentas increased from 10◦ to 36◦ in 72 h, and further increased to

5◦ in seven days. The plasma treated PU microfibrils showed aame trend as shown in Fig. 9, with hydrophobicity recoveredn a week. In order to improve the stability of hydrophilic sur-ace of fibrillar arrays after oxygen plasma treatment, they wereoated with a thin layer of polyethyleneimine (PEI) via electrostaticorce with the negative charged surface. As shown in Fig. 8, theydrophobic recovery of PDMS after oxygen plasma treatment wasignificantly inhibited by the PEI coating. The water contact anglef the PEI coated PDMS fibrils remained under 60◦ for two-weektorage. The coating of PEI is more effective on the PU fibrils to pre-erve the hydrophilic surface, as shown in Fig. 9. The water contactngle of PU fibrils barely changed during seven months of storagen air.

The shear adhesion capacity test results for oxygen plasma-reated followed by PEI-coated PU fibrillar arrays (against the glassnd polyvinyl chloride (PVC) substrates at 50% relative humid-ty), measured immediately and over time after treatment, areresented in Fig. 10. The adhesion of oxygen-plasma-treated PUbrillar arrays against the glass substrate significantly decreased

or the first week, which may result from the aging effect of PUver time, as freshly prepared PU fibrillar arrays were used in theest; it then slightly decreased over 7-month storage. The maturedU fibrillar arrays after oxygen plasma treatment largely preservedheir improved adhesion capacity against PVC after PEI coating,

s shown in Fig. 10. Therefore, significant gains in the stabilityf adhesion capacity over time are realized upon introduction ofhin self-assembled PEI layer on oxygen plasma-treated PU fibrillarrrays.

ig. 10. Adhesion capacities of PU microfibrillar arrays over time after oxygenlasma treatment and PEI coating.

[

[

[

[

[

[

[

[

Science 289 (2014) 586– 591

4. Conclusion

The adhesion capacity of elastomeric (PU and PDMS) fibril-lar arrays, which results from their ability to conform to surfaceroughness, benefits from UV/Ozone or oxygen plasma treatment.The combination of molecular dissociation and tailored surfacemorphology can be used to explain the improved adhesion capac-ity of treated fibrillar arrays. Treatment of fibril tip surfacesrenders them more hydrophilic, and the adhesion capacity oftreated fibrillar arrays is sensitive to the relative humidity ofthe environment. These observations point to the contributionof capillary effect to the adhesion capacity of treated fibrillararrays. Treated fibrillar arrays have a tendency towards hydropho-bic recovery and loss of adhesion capacity over time in air,which results from reaction of treated surfaces with moleculesavailable in air. Self-assembly of thin layers of selected poly-electrolytes on treated surfaces can hinder hydrophobic recovery,and thus stabilize the hydrophilic surface of treated fibrillararrays.

Acknowledgment

The authors wish to acknowledge the support of U.S. Depart-ment of Defense Small Business Innovation Research Program(Contract nos. FA8651-07-C-0092, and W911NF-10-C-0060) for theproject reported herein.

References

[1] B. Bhushan, Adhesion of multi-level hierarchical attachment systems in geckofeet, Journal of Adhesion Science and Technology 21 (2007) 1213–1258.

[2] K. Autumn, N. Gravish, Gecko adhesion: evolutionary nanotechnology, Philo-sophical Transactions of the Royal Society of London, Series A 366 (2008)1575–1590.

[3] K. Autumn, A. Dittmore, D. Santos, M. Spenko, M. Cutkosky, Frictional adhesion:a new angle on gecko attachment, Journal of Experimental Biology 209 (2006)569–3579.

[4] M.P. Murphy, B. Aksak, M. Sitti, Gecko-inspired directional and controllableadhesion, Small 5 (2009) 170–175.

[5] A. Del Campo, C. Greiner, E. Arzt, Contact shape controls adhesion of bioinspiredfibrillar surfaces, Langmuir 23 (2007) 10235–10243.

[6] A. Del Campo, E. Arzt, Design parameters and current fabrication approachesfor developing bioinspired dry adhesives, Macromolecular Bioscience 7 (2007)118–127.

[7] S. Kim, M. Sitti, Biologically inspired polymer microfibers with spatulate tipsas repeatable fibrillar adhesives, Applied Physics Letters 89 (2006).

[8] L. Ge, S. Sethi, L. Ci, P.M. Ajayan, A. Dhinojwala, Carbon nanotube-based syn-thetic gecko tapes, Proceedings of the National Academy of Sciences of theUnited States of America 104 (2007) 10792–10795.

[9] J.B. Puthoff, M.S. Prowse, M. Wilkinson, K. Autumn, Changes in materialsproperties explain the effects of humidity on gecko adhesion, Journal of Exper-imental Biology 213 (2010) 3699–3704.

10] W.X. Sun, P. Neuzil, T.S. Kustandi, S. Oh, V.D. Samper, The nature of the geckolizard adhesive force, Biophysical Journal 89 (2005) L14–L17.

11] M.F. Yanik, Towards gecko-feet-inspired bandages, Trends in Biotechnology 27(2009) 1–2.

12] B. Zhao, N. Pesika, H. Zeng, Z. Wei, Y. Chen, K. Autumn, K. Turner, J. Israelachvili,Role of tilted adhesion fibrils (Setae) in the adhesion and locomotion of gecko-like systems, Journal of Physical Chemistry B 113 (2009) 3615–3621.

13] C. Gogelein, M. Brinkmann, M. Schroter, S. Herminghaus, Controlling the for-mation of capillary bridges in binary liquid mixtures, Langmuir 26 (2010)17184–17189.

14] P.J. van Zwol, G. Palasantzas, J.T.M. De Hosson, Influence of roughness oncapillary forces between hydrophilic surfaces, Physical Review E: Statistical,Nonlinear, and Soft Matter Physics 78 (2008).

15] Doi Menezes, Highly stable hydrophilic surfaces of PDMS thin layer obtained byUV radiation and oxygen plasma treatments, Physica Status Solidi (C) CurrentTopics in Solid State Physics 7 (2010) 189–192.

16] T. Yamamoto, J.R. Newsome, D.S. Ensor, Modification of surface-energy,dry-etching, and organic film removal using atmospheric-pressure pulsed-

corona plasma, IEEE Transactions on Industry Applications 31 (1995)494–499.

17] M.L. Sham, J. Li, P.C. Ma, J.K. Kim, Cleaning and functionalization of polymersurfaces and nanoscale carbon fillers by UV/Ozone treatment: a review, Journalof Composite Materials 43 (2009) 1537–1564.

rface

[

[

[

[

[

[

[

A. Peyvandi et al. / Applied Su

18] P. Kuang, J. Lee, C.H. Kim, K.M. Ho, K. Constant, Improved surface wettability ofpolyurethane films by ultraviolet ozone treatment, Journal of Applied PolymerScience 16 (2010) 3024–3033.

19] Park, Koh, Pae, Effects of surface modification by Ar irradiation on wettabil-ity of surfaces of poly(ethylene terephthalate) films, Polymer Engineering andScience 28 (1998) 1185–1192.

20] A.D. Tserepi, M.E. Vlachopoulou, E. Gogolides, Nanotexturing of

poly(dimethylsiloxane) in plasmas for creating robust super-hydrophobicsurfaces, Nanotechnology 17 (2006) 3977–3983.

21] A. Olah, H. Hillborg, G.J. Vancso, Hydrophobic recovery of UV/ozone treatedpoly(dimethylsiloxane): adhesion studies by contact mechanics and mecha-nism of surface modification, Applied Surface Science 239 (2005) 410–423.

[

Science 289 (2014) 586– 591 591

22] S. Mehrotra, S.C. Hunley, K.M. Pawelec, L.X. Zhang, I. Lee, S. Baek, C. Chan, Celladhesive behavior on thin polyelectrolyte multilayers: cells attempt to achievehomeostasis of its adhesion energy, Langmuir 26 (2010) 12794–12802.

23] A. Peyvandi, P. Soroushian, J. Lu, A new self-loading locomotion mechanismfor wall climbing robots employing biomimetic adhesives, Journal of BionicEngineering 10 (2013) 12–18.

24] K. Koch, B. Bhushan, W. Barthlott, Diversity of structure, morphology and wet-

ting of plant surfaces, Soft Matter 4 (2008) 1943–1963.

25] K. Efimenko, W.E. Wallace, J. Genzer, Surface modification of Sylgard-184poly(dimethyl siloxane) networks by ultraviolet and ultraviolet/ozonetreatment, Journal of Colloid and Interface Science 254 (2002)306–315.