Dd

SRa

b

a

ARRAA

KSSIEN

1

m[m[imscwamsr

0h

Colloids and Surfaces B: Biointerfaces 106 (2013) 126– 134

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces

jou rna l h om epa g e: www.elsev ier .com/ locate /co lsur fb

ual role of outer epicuticular lipids in determining the wettability ofragonfly wings

ong Ha T. Nguyena, Hayden K. Webba, Jafar Hasana, Mark J. Tobinb,ussell J. Crawforda, Elena P. Ivanovaa,∗

Faculty of Life and Social Sciences, Swinburne University of Technology, PO Box 218, Hawthorn, VIC 3122, AustraliaAustralian Synchrotron, 800 Blackburn Rd, Clayton, Australia

r t i c l e i n f o

rticle history:eceived 9 November 2012eceived in revised form 21 January 2013ccepted 21 January 2013vailable online 29 January 2013

eywords:elf-cleaninguperhydrophobicitynsect wingspicuticular lipidsanostructures

a b s t r a c t

Numerous natural surfaces possess superhydrophobicity and self-cleaning properties that would beextremely beneficial when applied in industry. Dragonfly wings are one example of such surfaces, andwhile their general surface structure is known, their precise chemical composition is not. Here, the epi-cuticular lipids of dragonfly wing membranes were characterized to investigate their significance incontributing to self-cleaning and superhydrophobic properties. After just 10 s of lipid extraction usingchloroform, the water contact angles exhibited by the wings decreased below the accepted thresholdfor superhydrophobicity (150◦). Infrared spectra collected at the Australian Synchrotron contained char-acteristic absorption bands of amide, ester and aliphatic hydrocarbons moieties on the wing surfaces,the latter of which was decreased post-extraction with chloroform. GC–MS data analysis revealed thatthe epicuticular wax components were dominated by n-alkanes with even-numbered carbons, especiallyn-hexacosane, and palmitic acid. SEM and AFM data analysis conducted on the untreated and chloroform-

extracted wing surfaces demonstrated that surface topography changed after extraction; the surfacenanostructure was progressively lost with extended extraction times. The data presented here indicatethat epicuticular lipids contribute not only to self-cleaning and superhydrophobic properties throughtheir inherent hydrophobic nature, but also by forming the physical structure of the wing surface. Thisknowledge will be extremely valuable for reconstruction of dragonfly wing structures as a biomimetictemplate.

. Introduction

Naturally occurring surfaces provide great inspiration for novelaterials with high potential for various industrial applications

1–7]. These surfaces have been evolved and optimized overillions of years in order to adapt to ever-changing environments

8]. Nature is therefore an excellent resource for the development ofnnovative technologies which can be applied widely in chemistry,

edicine and material science [9,10]. Insects are prime examples ofuch surfaces; their lifestyles often dictate that they remain in closeontact with wet and/or dirty environments [10]. In particular theirings frequently account for significant portions of the surface

rea, but only occupy a small fraction of their body mass. In order to

aintain functionality insect wings have evolved to possess highly

pecific structures and surface chemistries which allow them toemain clean and dry [4,11–20]. Their highly functional surfaces

∗ Corresponding author. Tel.: +61 3 9214 5137.E-mail address: eivanova@swin.edu.au (E.P. Ivanova).

927-7765/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.colsurfb.2013.01.042

© 2013 Elsevier B.V. All rights reserved.

prevent penetration of the water and enable the droplets to collectall the dirt and/or contaminants as they roll off the surface [21,22].This process requires that the surface is superhydrophobic and iscommonly referred to as self-cleaning.

The wettability of a surface is dependent on two factors: thechemical composition of the surface and its physical morphologyand roughness [23–32]. The Wenzel and Cassie–Baxter theoriesof wettability describe the role of surface roughness in determin-ing wettability [33,34]. In brief, the Wenzel theory describes howsurface roughness can increase or decrease the observed watercontact angles depending on the inherent hydrophobicity of thesurface material, and the Cassie–Baxter theory describes how a sur-face can exhibit enhanced wettability by trapping air underneatha water droplet between the surface features. Contact angle hys-teresis (CAH) is another parameter that can be used to describethe hydrophobicity of a surface. It can be determined by measur-ing the difference between the advancing and receding contact

angles of water droplets on a solid surface. A surface that exhibitslarge water contact angle and low CAH can be considered super-hydrophobic and self-cleaning [25,27,31,35,36]. Superhydrophobicsurfaces commonly adhere to the Cassie–Baxter wettability state

faces B

[yd

temwficaestcpa

ct[bniif[rfrb

eiabacukirpt[twpd[ctatw

2

2

sMmt

S.H.T. Nguyen et al. / Colloids and Sur

21,23,25,27,29,32,36–40], including insect wings, however it is notet established to what degree the surface chemistry of the wingetermines the wettability.

Insect cuticle has been the subject of much research, dueo their sophisticated structure and multi-functional surfaces,.g. anti-wetting, anti-icing and anti-fouling [9,20,27,41–46]. Theechanisms of many of these properties are still unclear, so muchork still needs to be conducted. Only recently it was shown for therst time that cicada wings possess bactericidal effects due to theirharacteristic nanostructure [20]. The antimicrobial effect of fattycids obtained from insects has also been investigated [47–51], forxample, the mixture of fatty acids extracted from the secretoryetae of biting midge (Forcipomyia nigra) larvae has been reportedo exhibit antimicrobial activity [47]. The principle antimicrobialomponents of this mixture were found to be pelargonic, capric andalmitoleic acids. The most common finding among these works isn abundance of palmitic and oleic acid [49,50,52–54].

There has been much research conducted into the hierar-hical structure of the surfaces of insects, under the rationalehat it is the main contributing factor for superhydrophobicity10,32,55,56]. Surface chemistry on the other hand, has typicallyeen studied independently from the superhydrophobicity phe-omenon, and put into the context of macrobiological studies, e.g.

ntra-species and inter-species chemical communication betweenndividuals, and between the insect and environment [57–61], oror construction of taxonomic profiles based on surface chemistry25,41,61–65]. In the case of insect wings, the majority of previousesearch into surface wettability focuses almost entirely on sur-ace topography [4,9–11,13,24,27,30,66–68] and chemical studiesarely address the concept of self-cleaning and/or superhydropho-icity [62,69–72].

The outer most layer of insect cuticle is known as thepicuticle, and is composed of a complex mixture of lipidsncluding wax esters, triglycerides, and other derivatives ofliphatic hydrocarbons [14,62,70,72–78]. These highly hydropho-ic organic compounds likely contribute to the superhydrophobicnd self-cleaning properties of insect wings. In particular, whileonsiderable research has been conducted into the overall cutic-lar chemical composition [14,57,58,64,77,79,80], to the authors’nowledge there are no studies that have characterized the chem-cal composition of the epicuticle alone. Several studies haveeferred to the presence of epicuticular waxes, but they do notresent detailed analysis of their chemical composition, or relatehe lipid composition to the special characteristics of insect cuticle12,13,81]. For example, a recent study investigating the con-ributions of surface architecture and surface chemistry to theettability of beetle elytra determined the fundamental elementsresent using X-ray photoelectron spectroscopy (XPS), howeverid not present an analysis of the specific chemical components82]. The current research aims to extract the epicuticular lipidomponents of insect wings, to determine their role in governinghe surface wettability, and in turn self-cleaning abilities. Throughnalysis of lipid extracts using GC–MS, the detailed identities ofhe chemical components of the very outer most layer of dragonflying Hemicordulia tau were found.

. Materials and methods

.1. Materials

Dragonfly (H. tau) wing samples, which commonly dwell in

uburban regions of Melbourne, Australia were provided by theelbourne Museum. The insect wings were obtained from early toid-2011 and preserved in sterile Petri dishes at room tempera-

ure (ca. 22 ◦C). When required, the wings were removed from the

: Biointerfaces 106 (2013) 126– 134 127

body aseptically and stored at room temperature in sterile plasticcontainers.

Chloroform (CHROMASOLV for HPLC), n-tetracosane (GC Grade)and bis-N,N-(trimethylsilyl) trifluoroacetamide (BSTFA for TMSiderivatives) were obtained from Sigma Aldrich (St. Louis, MO, USA),as were standard alkane solutions (C8–C20 and C21–C40) which wereemployed for GC–MS analyses.

2.2. Extraction of epicuticular lipids

About 1 cm in length of the dragonfly wings was used for step-wise extraction with chloroform. Epicuticular lipids were extractedfrom pieces of dragonfly wings approximately 1 cm × 1 cm in sizeusing extraction times of 10 s, 20 s, and 30 s in chloroform. Extrac-tion times were minimized so as to limit the extraction of lipids tothe epicuticle only. The lipid–chloroform mixture was then filteredthrough glass wool to remove any contaminating particles that mayhave been present on the wing surface. The chloroform extractedwing samples were retained for further surface characterization. Toeach of the chloroform fractions, n-tetracosane (20 �g) was addedas internal standard before performing GC–MS analysis.

2.3. Wettability of the wing surfaces

The contact angles of Milli-Q water on the wing surface wasmeasured using the sessile drop method [114–117]. The contactangle measurements were carried out in air using FTA1000 (FirstTen Ångstroms, Inc., Portsmouth, VA, USA) instrument. An averageof at least ten measurements was obtained for each of chloroformextracted samples, in duplicate. The evaluation of contact angleswas performed by recording 50 images in 2 s with a PelcomodelPCHM 575-4 camera using FTA Windows Mode 4 software. To elim-inate effect of the vein structure on the wing surface, the waterdroplet was placed on regions sufficiently large to accommodatethe droplet footprint. The measurement was taken at ambient con-ditions of 21 ◦C and relative humidity of 60–70%.

2.4. Synchrotron radiation Fourier-transform infraredspectrometry (SR-FTIR)

The distribution of organic functional groups present across thewing membrane of the dragonfly species before and after chlo-roform extraction was analyzed using SR-FTIR, on the Infraredmicroscopy beam-line at the Australian Synchrotron. The sam-ples were scanned in transmission mode over several areas ofapproximately 50 �m × 50 �m using a Bruker Hyperion 2000 FTIRmicroscope (Bruker Optic GmbH, Ettlingen, Germany). OPUS ver-sion 6.5 software was employed for operating the microscope andspectrometer and analyzing the data.

2.5. Gas chromatography–mass spectrometry (GC–MS)

Prior to GC–MS analysis, any hydroxyl-containing componentsof the wing extracts (e.g. alcohols and carboxylic acids) wereconverted to trimethylsilyl (TMSi) derivatives by means of N,N-bis-trimethylsilyl-triflouro-acetamide (BSTFA, C8H21NOSi2) in thepresence of pyridine for 45 min at 70 ◦C according to previouslyoptimized laboratory protocols [57,60,62,64,83,84]. The composi-tion of the solution was determined by injecting 1 �L of sampleextract, separation via a capillary GC column (Rxi-5SIL MS columnfused silica; column 30.0 m, 0.25 mm i.d., df = 0.25 �m, Restek Cor-poration, Bellefonte, PA, USA) with He as carrier gas at constant

flow of 1.4 mL min−1, and a mass spectrometric detector (GCMS-QP2010, Shimadzu Scientific Instruments, Columbia, USA). Thechromatograph was programmed as follows: on-column injectionat 50 ◦C, oven at 50 ◦C for 5 min, then increasing to 280 ◦C at the rate

1 faces B: Biointerfaces 106 (2013) 126– 134

3rfCmsi

R

wcatrc

2

ueGmS(

dmeeg

2

mPsasapaor(Akot‘ad

3

3

lirwf

Fig. 1. Loss of self-cleaning properties after extraction of the wing surfaces withchloroform. Average static water contact angle (gray) and average contact anglehysteresis (black), with example water droplet contact angles (above). The shaded

28 S.H.T. Nguyen et al. / Colloids and Sur

◦C min−1, and then held for 30 min at 280 ◦C. Mass spectra wereecorded from m/z 14 to 800. The percentage composition reportedor each component is based on the peak area from the Total Ionurrent (TIC) chromatogram without standardization. A standardixture of n-alkanes (C8–C40) in chloroform was used as external

tandard to verify retention time for each compound. The retentionndices (RI) for each compound were calculated based on:

I = 100

[n0 + RTx − RTn0

RTn1 − RTn0

](1)

here n0 is the number of carbons for an n-alkane standard and n1orresponds to the next n-alkane in the series. RTx, RTn0 and RTn1

re the retention times for the component being measured relativeo the two closest n-alkane standards. The RI was compared withetention index databases [85,86] to identify the components notontained in the Wiley Registry of Mass Spectral Data.

.6. Scanning electron microscopy (SEM)

The surface morphologies of the both chloroform extracted andntreated wings was observed using a field emission scanninglectron microscope (FeSEM-SUPRA 40VP, Carl Zeiss GmbH, Jena,ermany) at fixed voltage of 3 kV. The sample was attached toetallic substrata using conductive double-sided adhesive tape.

amples were sputter coated with gold using a JEOL NeoCoatermodel MP-19020NCTR) prior to imaging.

To observe cross-sections of the wing membranes, the wing wasipped in liquid nitrogen and then snapped perpendicular to theain veins. Broken wings were attached to metallic discs by one

nd, to enable the exposed end to orient away from the substratum,nabling better imaging. Samples were then sputter coated withold in an identical manner as above.

.7. AFM analysis

An atomic force microscope (Innova, Veeco, USA) in tappingode was employed to evaluate the surface topographic profiles.

hosphorus-doped silicon probes (MPP-31120-10, Veeco) with apring constant of 0.9 N m−1, radius of curvature (tip) of 8 nm and

resonance frequency of approximately 20 kHz were utilized forurface imaging. Scanning was carried out at a right angle to thexis of the cantilever at 1 Hz. The topographical data of the sam-les were processed with first order horizontal and vertical levelingnd surface topography. The image analysis and determinationf the surface roughness parameters, the average roughness (Sa),oot-mean-square (rms) roughness (Sq) and maximum peak heightSmax), were performed using the instrument software (SPM Labnalysis v.7.11). Two additional parameters, skewness (Ssk) andurtosis (Sku), which are useful for describing surface morphol-gy, have been also utilized. Skewness measures the symmetry ofhe height probability density function and kurtosis measures thepeakedness’ of the profile [87–90]. All images were produced usingn Innova atomic force microscope (Veeco) and by exporting rawata files to Gwyddion data processing software [91].

. Results and discussion

.1. Surface wettability

Removal of the low surface energy material in the form ofipids and waxes from the wing surface was expected to lead to an

ncrease in wettability. The water contact angle (CA) on the chlo-oform extracted wings was indeed observed to decrease, coupledith an increase in the contact angle hysteresis (CAH) (Fig. 1). In

act, just 10 s of extraction was enough to cause the water CA of the

area at the top of the graph represents the minimum water contact angle thresholdrequired for self-cleaning ability. Results were an average of at least 10 independentmeasurements of water CA and CAH. Error bars indicates standard deviation, n = 20.

wing to fall below the threshold for superhydrophobic materials(shaded area, Fig. 1). The untreated wing fulfilled the requirementto be considered self-cleaning and superhydrophobic, i.e. high CA(>150◦) [11,22,27,28,30,92], and low CAH (2.3◦) [1,25,28]. The lowCAH value of 2.3◦ highlights the self-cleaning ability of a surface;only a small tilting angle (greater than 2.3◦) is required for a waterdroplet in contact with the surface to run off the surface, collectingcontaminants and removing them as the droplet rolls off the sur-face. After 10 s of chloroform extraction the CA decreased to 130◦

and CAH increased to 13.7, demonstrating an increase in wettabil-ity, and indicating that the surface was no longer self-cleaning.Longer extraction caused further increase in wettability, i.e. after20 s (CA = 120◦, CAH = 22.6◦) and 30 s (CA = 105◦, CAH = 29.9) extrac-tions. Both wing samples were still relatively hydrophobic after theextraction, which suggests that the newly exposed outer layer ofthe wing is composed of a material that possesses a particularlylow surface energy. As observed for the wing subjected to chloro-form extraction for a period of 10 s, those exposed to chloroform forperiods of both 20 and 30 s did not meet the threshold requirementto be considered superhydrophobic, and thus are not able to beclassified as ‘self-cleaning’. Extensive changes in the hydrophobicparameters highlighted the significance of the work in identifyingthe relative contribution of surface topography and surface chem-istry to the behavior of the wing surfaces. This information wouldbe useful in the development of new materials that possess similarproperties as the dragonfly wings.

3.2. Surface chemistry of extracted wings

Synchrotron-radiation FTIR was employed in transmissionmode to determine the chemical composition of the wing mem-branes, prior to and post-lipid extraction. Spectra representingthe untreated and chloroform extracted wings each containedthree major groups of bands at 3480–3230 cm−1, 3000–2800 cm−1

and 1750–1480 cm−1 which correspond to hydroxyl, alkyl hydro-

carbons and amide groups (Fig. 2, Table 1). The spectra of allfour wings were dominated by amide I and amide II absorptionbands due to C O bond stretching coupled to N H bending(1610–1695 cm−1) and C N stretching coupled to N H bending

S.H.T. Nguyen et al. / Colloids and Surfaces B

Fig. 2. Chemical profiles of chloroform-extracted wings as inferred from infraredabsorption spectra. The intensity of the CH2 stretching bands (as indicated) is agte

(gmltOaww

TAI

ood approximation for the amount of aliphatic hydrocarbons, i.e. lipids present onhe wing surface. The intensity of these bands decreased successively with extendedxtraction time. The spectra were acquired at the Australian Synchrotron.

1480–1575 cm−1) respectively [93–96]. The presence of amideroups can be attributed to chitin and proteins, which are theajor structural components of insect cuticle [97–101]. The fiber-

ike structures of chitin and proteins impart the robustness ofhe wing’s bulk, and are unaffected by chloroform extraction.ther broad carbonyl bands sometimes overlap this region, such

s aliphatic and aromatic carboxylic acids, aldehydes and ketones,hich may contribute a minor ‘shoulder’ to the edge of the peakithin these regions [102]. A small ester carbonyl stretching band

able 1ssignment of chemical components to the major absorption bands present in the

R spectra of the wing membranes.

Band position (cm−1) Functional groups Characteristic

1480–1580 Amide II • Out-of-phase combination ofthe NH in plane bend and theCN stretching vibration• Carbonyl stretching band dueto ketone, carboxylic acid [112]

1570–1700 Amide ICarbonyl groups

• C O stretch vibration causesCCN deformation and the NHin-plane bend

1700–1750 Ester • Carbonyl stretching band[113]

2800–3000 Alkanes or alkylgroups

• CH2, CH3 symmetric andasymmetric stretch

3000–3100 Unsaturated andaromatic C H

• CH stretch (C is sp2) [103]

3200–3500 O H stretch fromfree alcohol orphenol

• O H and N H stretching,which can be due tocarbohydrates, carboxylicacids, amines oralcohols/phenols [103]

: Biointerfaces 106 (2013) 126– 134 129

was present at approximately 1735 cm−1 in all the recorded spec-tra. The broad, overlapping O H and N H stretching bands within3500–3200 cm−1 were most likely also a result of absorption by thechitin and protein components of the wings [103]. This region wasalso relatively unchanged by lipid extraction with chloroform.

The C H stretching region (2840–3000 cm−1) contains fourbands representing the symmetric (�s) and anti-symmetric (�as)stretching vibrations of CH2 and CH3 functional groups (Fig. 2). Thepresence of C H stretching bands with a prevalence of methylenebands is indicative of long-chain aliphatic hydrocarbons, whichwere previously reported to be components of the cuticular waxespresent in insect wings [104,105]. The C H stretching IR absorp-tion bands appeared with relatively high intensity in the spectra ofuntreated wings; however their intensity considerably decreasedas the extraction time increased. This observation suggests thatthe extraction of hydrophobic wax-like components present onepicuticular layer of the wing membrane using chloroform wassuccessful.

3.3. Composition of lipid extracts

The identities of the lipid components that were extracted fromthe epicuticular layer of H. tau dragonfly wings were determinedusing GC–MS (Supplementary Table S1). Data analysis confirmedthe presence of aliphatic hydrocarbons and fatty acids; the majorcomponents of all fractions were n-alkanes (Fig. 3). They occupieda large proportion (approximately 79%) of the total compoundspresent in the 10 s chloroform-extracted fraction and remained sta-ble after 20 and 30 s of extraction time (71% and 69%, respectively).Alkanes ranged in length from C10 to C34, and were dominatedby even numbers, with a minor amount of C19, C23, C25, andC29 (Table S1). Because the odd-numbered n-alkanes requiredlonger extraction times to be detectable by GC–MS they weremost likely localized to lower regions of the epicuticle. Amongthe n-alkanes, n-hexacosane (C26) alone accounted for 26.82% oftotal detected compounds in the 10 s-extracted fraction, and thisincreased slightly to approximately 29% in the 20 s fraction, anddropped to 18% in the 30 s fraction. Methylalkanes were also preva-lent among the aliphatic hydrocarbons detected in the epicuticularlayer of the wing membrane. They comprised up to 19% of thetotal compounds present in the 10 s fraction, however, their pro-portion decreased as the time of extraction increased. As only ashort time was required to extract these compounds they are mostlikely to be localized to the very outer layer of the wing epicuticle.The only oxygen-containing compound detected in 10 s fractionwas hexadecanoic acid (palmitic acid) which made up 7% of theextracted epicuticular lipids. The proportion of palmitic acid in thewing extracts increased to 18% and 14% after 20 s and 30 s extractiontime, respectively.

Further analysis of the aliphatic hydrocarbons revealed thatmore than 69% of the alkanes present in the 10 s extracted fractionranged from C19 to C26, which changed only slightly to 67% and72% after 20 and 30 s of chloroform extraction respectively (Fig. 3).Chain lengths between C10 and C18 occupied 22% of total alkanespresent in the 10 s extracted fraction, 27% after 20 s extraction and22% after 30 s. About 9% of total alkanes present in the 10 s fractionwere longer carbon chain length (C27–C34) which appeared withstable or slightly decreased proportions in the 20 and 30 s extracts.

Aliphatic hydrocarbons in total occupied about 93% of the 10 sextracted fraction, 82% of the 20 s fraction and 86% of the 30 sof extraction respectively. Their changes in proportions assistedin predicting the location of certain compounds to different lay-

ers of the dragonfly wing epicuticle. For example, even-numberedn-alkanes appeared with high proportions in 10 s fraction andremained stable or slightly decreased in fractions with longerextraction times. Long-chain hydrocarbon biosynthesis occurs by

130 S.H.T. Nguyen et al. / Colloids and Surfaces B: Biointerfaces 106 (2013) 126– 134

F gth ofT bon a

aemsaTlaioaoaa

rasoseeimeogh

3

tdcto

ig. 3. Relative proportion of all identified chemical classes (top row) and chain lenhe epicuticle was dominated by n-alkanes, which mostly ranged from 19 to 26 car

ddition of acetyl groups to a fatty acid precursor, therefore chainxtension occurs in multiples of C2, and the final length of theolecule is dependent on the fatty acid precursor [61,106]. Within

uch a short time as 10 s, most of those compounds were extractednd their proportions did not increase with longer extraction time.herefore, they are most likely to be located at the very outerayer of the wing epicuticle. The proportions of branching alkanesnd hexadecanoic acid on the other hand increased steadily withncreased extraction time so they were likely distributed through-ut the wing membrane. Removal of a large portion of n-alkanesppeared to be the cause of the loss of self-cleaning after 10 sf chloroform extraction. Further extraction led to loss of largermount of lipid components, and further decreases in water contactngles.

Previous studies on insect cuticle compositions have commonlyeported high proportions of fatty acids, especially palmitic acidnd oleic acid [49,50,52–54,107,108]. In the current work, a sub-tantial proportion of palmitic acid was also detected, howeverleic acid was below detectable levels in all lipid extracts. Thisuggests that of the two, only palmitic acid is located within thepicuticle, and oleic acid may be found in deeper cuticular lay-rs. Palmitic acid is the most common fatty acid in nature, ands ubiquitous not just among insects, but in animals, plants and

icroorganisms as well [106,109–111]. As palmitic acid has anven number of carbon atoms in its backbone chain, and was thenly detected fatty acid in lipid extracts it may be tentatively sug-ested that it also acts as the precursor molecule for long-chainydrocarbon biosynthesis.

.4. Morphology of extracted wings

The Wenzel and Cassie–Baxter theories of wettability statehat the roughness of a substratum may play a significant role in

etermining its observed wettability. As such, before it can be con-lusively stated that the extraction of the lipids in the epicuticle ofhe wing caused the loss of superhydrophobicity and self-cleaningf the surface, the morphology and roughness of the wings

aliphatic hydrocarbons (second row) present in the epicuticular layer of the wing.toms in length.

post-extraction must be assessed. High-resolution scanning elec-tron micrographs of the untreated wing surfaces and cross-sectionsrevealed three distinct layers: the top and bottom epicuticularlayers, and an intracuticle (Fig. 4). The epicuticular layers werecomposed of multi-column pillars with heights of about 200 nm.Surface-view micrographs revealed that the pillars have roundtops with a diameter of approximately 80 ± 20 nm, spaced approx-imately 180 ± 30 nm, although somewhat randomly distributed.These observations were similar to those reported previously byWatson et al. [11] for the dragonfly species Rhyothemis phyllis chloe.

After the wings were extracted with chloroform for 10 s, slightchanges to the pillar structures on the surface were observed. Thepillars structure still remained after extraction, however the roundtops that were seen on the tops of the pillar structures of intactwings disappeared (Fig. 4c and d). As a result the pillars appearedto become sharper, and in addition became more sparse, suggest-ing that the outermost portions of the pillars were dissolved bychloroform extraction. As the extraction time increased, the pro-gressive loss of surface structure became more obvious. After 20 s(Fig. 4e and f), the epicuticular layers of the wing appeared thin-ner from the cross-section view, and circular patterns were visiblein the surface images. Minimal structure was left scattered acrossthe surface after 30 s of chloroform extraction, while the internalstructure remained unaffected (Fig. 4g and h).

AFM analysis of lipid-extracted wings was also performed toprovide quantitative measurements of the surface topography.The changes in the surface morphologies observed in the electronmicrographs were also reflected in AFM analysis. Typical two-dimensional images and cross-sectional profiles of 5 �m × 5 �mAFM scan areas of each of the wing surfaces are presented in Fig. 5.Each cross-sectional profile was generated by selecting one linefrom the corresponding two-dimensional AFM scan and plotting x-translation vs. height data. The mean planes of corresponding scans

were defined as 0 in the cross-sectional profiles. The surface fea-tures in the profiles appeared to decrease in height and sharpnessas the time of lipid extraction increased. Untreated wing surfaceprofiles contained features ranging from 50 to 150 nm. After 10 s

S.H.T. Nguyen et al. / Colloids and Surfaces B: Biointerfaces 106 (2013) 126– 134 131

Fig. 4. Cross-section (a, c, e, g) and surface (b, d, f, h) morphology of dragonfly wing surfaces (Hemicordulia tau) after extraction with chloroform over various time intervals.Electron micrographs were recorded of untreated wings (a, b), wings after 10 s of chloroform extraction (c, d), after 20 s extraction (e, f) and after 30 s extraction (g, h).Progressive loss of surface structure is visible with extended extraction time, with minimal structure remaining after 30 s. Scale bars = 400 nm.

Table 2Roughness analysis of the untreated and chloroform extracted wing surfaces.a

Roughness statisticsb Untreated 10 s 20 s 30 s

Smax (nm) 814 ± 72 409 ± 40 285 ± 14 332 ± 14Sa (nm) 47.0 ± 2.6 38.4 ± 10.5 25.2 ± 2.7 40.2 ± 8.3Sq (nm) 61.0 ± 3.3 49.6 ± 10.7 32.4 ± 3.1 49.3 ± 9.9Ssk 0.616 ± 0.058 −0.170 ± 0.705 0.193 ± 0.035 −0.085 ± 0.236Sku 1.714 ± 0.435 1.83 ± 3.29 0.688 ± 0.460 −0.399 ± 0.279Sdr (%) 278 ± 51 157 ± 54 121 ± 3 143 ± 10

a Results presented are averages of three measurements taken from 25 �m2 scanning areas and errors are the standard deviations. The cut-off length (�c) is 0.83 �m.b A technical description of the roughness parameters used can be found in Crawford et al. [44].

132 S.H.T. Nguyen et al. / Colloids and Surfaces B: Biointerfaces 106 (2013) 126– 134

F riods

a orm el

osceawah

ig. 5. Surface topography of dragonfly wings (Hemicordulia tau) after various pereas) and cross-sectional profiles of the untreated (a) and 10 s, 20 s and 30 s chlorofower in height with increasing of extraction time. Scale bars = 1 �m.

f extraction with chloroform, the features appeared rounder andhorter (50–100 nm). Notably, while scanning the surface of thehloroform-extracted wings, the AFM tip experienced lessened lat-ral deflections, i.e. the tip was adhering less to the wing surface

fter lipid extraction, suggesting that less soft/sticky lipid materialas present on the surface. As time of extraction increased to 20

nd 30 s, the surface features decreased further to about 80 nm ineight, appeared broader and spaced further apart (Fig. 5c and d,

of extraction with chloroform. Typical two-dimensional AFM scans (5 �m × 5 �mxtracted wing surfaces (b, c, d) are presented. Surface features became broader and

Supplementary Fig. S1). A roughness was also performed on theuntreated and chloroform extracted wings. Six statistical parame-ters, Smax, Sa, Sq, Ssk, Sku and Sdr, were calculated for each surface(Table 2). Detailed definitions for each parameter can be found in

Crawford et al. [44] briefly however, Smax is a measure of the maxi-mum height variation, Sa and Sq are measures of the typical heightvariation, Ssk and Sku describe the shape of the height distributionand Sdr is a measure of the surface area relative to the projected area

faces B

oaSslttcttiotdltrp

4

tehssrstpffcFoacttmli

A

UTo

A

f2

R

S.H.T. Nguyen et al. / Colloids and Sur

f the scan. Compared to the untreated wing surface, Sdr decreaseds the extraction time increased. A similar trend was observed formax, Sa, and Sq. The average roughness (Sa) of the untreated wingurface was found to be 47.0 nm, which gradually decreased to asow as 25.2 nm after 20 s of chloroform extraction. The roughness ofhe wings subjected to the chloroform extraction for 10 s appearedo decrease, however the changes in roughness were not signifi-ant. However, in contrast to that observed for the wings subjectedo chloroform extraction for 10 and 20 s, the roughness parame-ers of the wing samples exposed to chloroform extraction for 30 sncreased. The longer extraction time may have led to the removalf more cuticular wax material from between the pillars than onheir upper regions. This would enable the AFM tip to penetrateeeper into the wing surface, causing the pillar height to appear

arger. The loss of surface nanostructure with extended extractionime highlights the fact that the components of the wing that areapidly soluble in chloroform, i.e. the epicuticular lipids and waxeslay a major role in forming the wing nanoarchitecture.

. Conclusions

The epicuticular lipids of H. tau wing membranes were extractedo determine the effect on surface wettability. Wings that werextracted with chloroform were found to have decreased in surfaceydrophobicity, falling below the water CA threshold that definesuperhydrophobicity after just 10 s of lipid extraction. The loss ofurface structure on the wing surface due to extraction with chlo-oform leads to two important conclusions. The first is that it is notimply the loss of low surface energy material from the epicuticlehat caused the wing to lose its superhydrophobic and self-cleaningroperties; loss of structure was almost certainly a contributingactor. Secondly, the removal of the wing nanostructures via chloro-orm extraction enables the conclusion that the nanostructures areomposed of the materials that were identified in the lipid extracts.or the first time, it can be stated that the nanoscale pillars presentn the surface of H. tau wing membranes are composed primarily ofliphatic hydrocarbons, especially n-alkanes with even-numberedhain lengths between C19 and C26, and a relatively small propor-ion of palmitic acid. These components are vital in determininghe superhydrophobic and self-cleaning properties of H. tau wing

embranes, as they not only present a low-energy hydrophobicipid layer to the external environment, but are the primary build-ng materials forming the nanostructures on the wing surface.

cknowledgements

The authors would like to thank Dr. Peter Mahon, at Swinburneniversity for his advice on acquiring and interpreting GC–MS data.his work was funded in part by the Advanced Manufacturing Co-perative Research Centre AMCRC.

ppendix A. Supplementary data

Supplementary data associated with this article can beound, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.013.01.042.

eferences

[1] W. Barthlott, C. Neinhuis, Planta 202 (1997) 1.[2] W. Barthlott, C. Neinhuis, D. Cutler, F. Ditsch, I. Meusel, I. Theisen, H. Wilhelmi,

Bot. J. Linn. Soc. 126 (1998) 237.[3] J.L. Boeve, D. Voigt, S.N. Gorb, Arthropod. Struct. Dev. 40 (2011) 186.

[4] D. Byun, J. Hong, Saputra, J.H. Ko, Y.J. Lee, H.C. Park, B.K. Byun, J.R. Lukes, J.

Bionic. Eng. 6 (2009) 63.[5] H.M.S. Hu, G.S. Watson, B.W. Cribb, J.A. Watson, J. Exp. Biol. 214 (2011) 915.[6] Y.T. Cheng, D.E. Rodak, C.A. Wong, C.A. Hayden, Nanotechnology 17 (2006)

1359.

: Biointerfaces 106 (2013) 126– 134 133

[7] K. Koch, B. Bhushan, H.J. Ensikat, W. Barthlott, Philos. Transact. A. Math. Phys.Eng. Sci. 367 (2009) 1673.

[8] P. Kenrick, P.R. Crane, Nature 389 (1997) 33.[9] G.S. Watson, B.W. Cribb, J.A. Watson, ACS Nano 4 (2010) 129.

[10] G.S. Watson, B.W. Cribb, J.A. Watson, PLoS One 6 (2011), Art. No. e24368.[11] G.S. Watson, J.A. Watson, S. Hu, C.L. Brown, B.W. Cribb, S. Myhra, Int. J.

Nanomanuf. 5 (2010) 112.[12] M. Sun, G.S. Watson, Y. Zheng, J.A. Watson, A. Liang, J. Exp. Biol. 212 (2009)

3148.[13] Y.-l. Wan, Q. Cong, J.F. Jin, X.-j. Wang, Jilin Daxue Xuebao (Gongxueban)/J. Jilin

Univ. (Engineering and Technology Edition) 39 (2009) 732.[14] M. Gołebiowski, M.I. Bogus, M. Paszkiewicz, P. Stepnowski, Anal. Bioanal.

Chem. 399 (2011) 3177.[15] M.I. Bogus, M. Czygier, M. Golbiowski, E. Kdra, J. Kucinska, J. Mazgajska, J.

Samborski, W. Wieloch, E. Wloka, Exp. Parasitol. 125 (2010) 400.[16] M. Gołebiowski, E. Malinski, M.I. Bogus, J. Kumirska, P. Stepnowski, Insect.

Biochem. Mol. Biol. 38 (2008) 619.[17] S.L. Jarrold, D. Moore, U. Potter, A.K. Charnley, Mycol. Res. 111 (2007) 240.[18] C. De Pasquale, S. Guarino, E. Peri, G. Alonzo, S. Colazza, Anal. Bioanal. Chem.

389 (2007) 1259.[19] R. Escobar Galindo, R. Gago, D. Duday, C. Palacio, Anal. Bioanal. Chem. 396

(2010) 2725.[20] E.P. Ivanova, J. Hasan, H.K. Webb, V.K. Truong, G.S. Watson, J.A. Watson, V.A.

Baulin, S. Pogodin, J.Y. Wang, M.J. Tobin, C. Löbbe, R.J. Crawford, Small 8 (2012)2489.

[21] A. Marmur, Langmuir 20 (2004) 3517.[22] N.A. Patankar, Langmuir 20 (2004) 8209.[23] A. Lafuma, D. Quéré, Nat. Mater. 2 (2003) 457.[24] T. Sun, L. Feng, X. Gao, L. Jiang, Acc. Chem. Res. 38 (2005) 644.[25] B. Bhushan, Y.C. Jung, J. Phys. Condens. Mater. 20 (2008).[26] M. Nosonovsky, B. Bhushan, J. Phys. Condens. Mater. 20 (2008).[27] B. Bhushan, Y.C. Jung, K. Koch, Philos. Transact. A. Math. Phys. Eng. Sci. 367

(2009) 1631.[28] B. Bhushan, Y.C. Jung, K. Koch, Langmuir 25 (2009) 3240.[29] J.P. Rothstein, Annu. Rev. Fluid Mech. 42 (2010) 89.[30] B. Bhushan, C.J. Jung, M. Nosonovsky, Lotus effect: surfaces with roughness-

induced superhydrophobicity, self-cleaning, and low adhesion, in: SpringerHandbook of Nanotechnology, 3rd ed., Springer, New York, 2010, pp. 1437.

[31] B. Bhushan, Beilstein. J. Nanotechnol. 2 (2011) 66.[32] B. Bhushan, Langmuir 28 (2012) 1698.[33] R.N. Wenzel, Ind. Eng. Chem. 28 (1936) 988.[34] A.B.D. Cassie, S. Baxter, Trans. Faraday Soc. 40 (1944) 546.[35] Y.C. Jung, B. Bhushan, Nanotechnology 17 (2006) 4970.[36] M. Ma, R.M. Hill, Curr. Opin. Colloid Interface Sci. 11 (2006) 193.[37] V.K. Truong, H.K. Webb, E. Fadeeva, B.N. Chichkov, A.H.F. Wu, R. Lamb, J.Y.

Wang, R.J. Crawford, E.P. Ivanova, Biofouling 28 (2012) 539.[38] M. Nosonovsky, B. Bhushan, Microsyst. Technol. 11 (2005) 535.[39] R. Fürstner, W. Barthlott, C. Neinhuis, P. Walzel, Langmuir 21 (2005) 956.[40] L. Feng, S. Li, Y. Li, H. Li, L. Zhang, J. Zhai, Y. Song, B. Liu, L. Jiang, D. Zhu, Adv.

Mater. 14 (2002) 1857.[41] J.F. Ferveur, Behav. Genet. 35 (2005) 279.[42] H. Mei, D. Luo, P. Guo, C. Song, C. Liu, Y. Zheng, L. Jiang, Soft Matter 7 (2011)

10569.[43] H.K. Webb, J. Hasan, V.K. Truong, R.J. Crawford, E.P. Ivanova, Curr. Med. Chem.

18 (2011) 3367.[44] R.J. Crawford, H.K. Webb, V.K. Truong, J. Hasan, E.P. Ivanova, Adv Colloid Inter-

face Sci. 179–182 (2012) 142.[45] M. Nosonovsky, B. Bhushan, Adv. Funct. Mater. 18 (2008) 843.[46] T. Wagner, C. Neinhuis, W. Barthlott, Acta Zool. 77 (1996) 213.[47] A. Urbanek, R. Szadziewski, P. Stepnowski, J. Boros-Majewska, I. Gabriel, M.

Dawgul, W. Kamysz, D. Sosnowska, M. Gołeogonekbiowski, J. Insect Physiol.58 (2012) 1265.

[48] N.M. Carballeira, Prog. Lipid Res. 47 (2008) 50.[49] R. Crespo, M. Patricia Juárez, L.F.R. Cafferata, Mycologia 92 (2000) 528.[50] K.E. Ekpo, A.O. Onigbinde, I.O. Asia, Afr. J. Pharm. Pharmacol. 3 (2009) 051.[51] F.Z. Kkbay, E. Kuyumcu, T. Bilenler, B. Yldz, Nat. Prod. Res. 26 (2012) 1668.[52] M. Cunningham, A. Gonzalez, M. Dreon, D. Castro, R. Pollero, J. Parasitol. 87

(2001) 1251.[53] M.A. Madariaga, A.M. Municio, A. Ribera, Comp. Biochem. Physiol. 36 (1970)

271.[54] J.E. McFarlane, Comp. Biochem. Physiol. 24 (1968) 377.[55] M. Sun, A. Liang, G.S. Watson, J.A. Watson, Y. Zheng, J. Ju, L. Jiang, PLoS One 7

(2012), Art. no. e35056.[56] Y.Y. Yan, N. Gao, W. Barthlott, Adv. Colloid Interface Sci. 169 (2011) 80.[57] M. Gołebiowski, M. Paszkiewicz, A. Grubba, D. Gasiewska, M.I. Bogus, E. Włóka,

W. Wieloch, P. Stepnowski, Bull. Entomol. Res. 102 (2012) 453.[58] F.R. Dani, Annales Zoologici Fennici 43 (2006) 500.[59] D.H. Choe, S.R. Ramírez, N.D. Tsutsui, J. Chem. Ecol. 38 (2012) 176.[60] M. Gołebiowski, E. Malinski, J. Nawrot, P. Stepnowski, J. Stored Prod. Res. 44

(2008) 386.[61] R.W. Howard, G.J. Blomquist, Annu. Rev. Entomol. 50 (2005) 371.

[62] G.D. Yocum, J.S. Buckner, C.L. Fatland, Comp. Biochem. Physiol. B, Biochem.

Mol. Biol. 159 (2011) 163.[63] S. Martin, F. Drijfhout, J. Chem. Ecol. 35 (2009) 1151.[64] S.F. Geiselhardt, S. Geiselhardt, K. Peschke, Chemoecology 19 (2009) 185.[65] I. Saïd, G. Costagliola, I. Leoncini, C. Rivault, J. Insect Physiol. 51 (2005) 995.

1 faces B

[113] P. Lasch, M. Boese, A. Pacifico, M. Diem, Vib. Spectrosc. 28 (2002) 147.

34 S.H.T. Nguyen et al. / Colloids and Sur

[66] G.S. Watson, S. Myhra, B.W. Cribb, J.A. Watson, Biophys. J. 94 (2008) 3352.[67] Y. Su, B. Ji, K. Zhang, H. Gao, Y. Huang, K. Hwang, Langmuir 26 (2010) 4984.[68] Z. Burton, B. Bhushan, Nano Lett. 5 (2005) 1607.[69] D.R. Nelson, T.P. Freeman, J.S. Buckner, K.A. Hoelmer, C.G. Jackson, J.R. Hagler,

Comp. Biochem. Physiol. B, Biochem. Mol. Biol. 136 (2003) 343.[70] D.R. Nelson, C.L. Fatland, J.S. Buckner, T.P. Freeman, Comp. Biochem. Physiol.

B, Biochem. Mol. Biol. 123 (1999) 137.[71] D.R. Nelson, L.D. Charlet, Comp. Biochem. Physiol. B, Biochem. Mol. Biol. 135

(2003) 273.[72] K.H. Lockey, Comp. Biochem. Physiol. B, Biochem. Mol. Biol. 81 (1985) 263.[73] H.K. Lockey, J. Exp. Biol. 37 (1959) 316.[74] C.L. Soliday, G.J. Blomquist, L.L. Jackson, J. Lipid. Res. 15 (1974) 399.[75] K.H. Lockey, Comp. Biochem. Physiol. B, Biochem. Mol. Biol. 65 (1980) 457.[76] J.S. Buckner, M.C. Mardaus, D.R. Nelson, Comp. Biochem. Physiol. B, Biochem.

Mol. Biol. 114 (1996) 207.[77] S.L. Lapointe, W.B. Hunter, R.T. Alessandro, Agric. For. Entomol. 6 (2004) 251.[78] C. Buschhaus, R. Jetter, J. Exp. Bot. 62 (2011) 841.[79] K.D. Cameron, M.A. Teece, L.B. Smart, Plant Physiol. 140 (2006) 176.[80] D.R. Nelson, R.E. Lee Jr., Comp. Biochem. Physiol. B, Biochem. Mol. Biol. 138

(2004) 313.[81] Y.-l. Wan, Q. Cong, X.-j. Wang, Z. Yan, J. Bionic. Eng. 5 (2008) 40.[82] M. Sun, A. Liang, G.S. Watson, J.A. Watson, Y. Zheng, L. Jiang, PLoS One 7 (2012),

Art. no. e46710.[83] C. Buschhaus, H. Herz, R. Jetter, Ann. Bot. 100 (2007) 1557.[84] M. Gołebiowski, M.I. Bogus, M. Paszkiewicz, P. Stepnowski, J. Insect Physiol.

56 (2010) 391.[85] The Sadtler Standard Gas Chromatography Retention Index Library, Sadtler

Research Laboratories, Philadelphia, PA, 1984.[86] V. Pacakova, K. Feltl, Chromatographic Retention Indices: An Aid to the Iden-

tification of Organic Compounds, Ellis Horwood, London, 1992.[87] E.P. Ivanova, V.K. Truong, J.Y. Wang, C.C. Bemdt, R.T. Jones, I.I. Yusuf, I. Peake,

H.W. Schmidt, C. Fluke, D. Barnes, R.J. Crawford, Langmuir 26 (2010) 1973.[88] E.S. Gadelmawla, M.M. Koura, T.M.A. Maksoud, I.M. Elewa, H.H. Soliman, J.

Mater. Process. Technol. 123 (2002) 133.[89] S.F. Lamolle, M. Monjo, S.P. Lyngstadaas, J.E. Ellingsen, H.J. Haugen, J. Biomed.

Mater. Res. A 88 (2009) 581.

: Biointerfaces 106 (2013) 126– 134

[90] V.K. Truong, S. Rundell, R. Lapovok, Y. Estrin, J.Y. Wang, C.C. Berndt, D.G.Barnes, C.J. Fluke, R.J. Crawford, E.P. Ivanova, Appl. Microbio. Biot. 83 (2009)925.

[91] D. Necas, P. Klapetek, Cent. Eur. J. Phys. 10 (2012) 181.[92] H.J. Ensikat, P. Ditsche-Kuru, C. Neinhuis, W. Barthlott, Beilstein. J. Nanotech-

nol. 2 (2011) 152.[93] W. Sajomsang, P. Gonil, Mater. Sci. Eng. C 30 (2010) 357.[94] Z. Ganim, S.C. Hoi, A.W. Smith, L.P. Deflores, K.C. Jones, A. Tokmakoff, Acc.

Chem. Res. 41 (2008) 432.[95] J. Brugnerotto, J. Lizardi, F.M. Goycoolea, W. Argüelles-Monal, J. Desbrières,

M. Rinaudo, Polymer 42 (2001) 3569.[96] F. Caruso, D.N. Furlong, K. Ariga, I. Ichinose, T. Kunitake, Langmuir 14 (1998)

4559.[97] A.C. Neville, D.A. Parry, J. Woodhead-Galloway, J. Cell. Sci. 21 (1976) 73.[98] P. Kreuz, W. Arnold, A.B. Kesel, Ann. Biomed. Eng. 29 (2001) 1054.[99] K.J. Kramer, S. Muthukrishnan, Insect. Biochem. Mol. Biol. 27 (1997) 887.

[100] R.H. Hackman, M. Goldberg, J. Insect Physiol. 33 (1987) 39.[101] S.N. Gorb, Microsc. Res. Tech. 37 (1997) 583.[102] M. Jackson, H.H. Mantsch, Crit. Rev. Biochem. Mol. Biol. 30 (1995) 95.[103] C. Coury, A.M. Dillner, Atmos. Environ. 42 (2008) 5923.[104] S. Merk, A. Blume, M. Riederer, Planta 204 (1998) 44.[105] J. Zeier, L. Schreiber, Planta 209 (1999) 537.[106] L. Samuels, L. Kunst, R. Jetter, Annu. Rev. Plant Biol. 59 (2008) 683.[107] J.S. Buckner, M.M. Hagen, Arch. Insect Biochem. Physiol. 53 (2003) 66.[108] H.M. Womeni, M. Linder, B. Tiencheu, F.T. Mbiapo, P. Villeneuve, J. Fanni, M.

Parmentier, OCL-Oleagineux Corps Gras Lipides 16 (2009) 230.[109] S. Smith, FASEB J. 8 (1994) 1248.[110] G. Milligan, M. Parenti, A.I. Magee, Trends Biochem. Sci. 20 (1995) 181.[111] W. Dowhan, Annu. Rev. Biochem. 66 (1997) 199.[112] A. Barth, Biochim. Biophys. Acta 1767 (2007) 1073.

[114] R.J. Crawford, L.K. Koopal, J. Ralston, Colloids Surf 27 (1987) 57.[115] C.J. Van Oss, R.J. Good, M.K. Chaudhury, Langmuir 4 (1988) 884.[116] D.W. Guy, R.J. Crawford, D.E. Mainwaring, Fuel 75 (1996) 238.[117] D. Öner, T.J. McCarthy, Langmuir 16 (2000) 7777.