hierarchical micro and nano structured, hydrophilic, superhydrophobic and superoleophobic surfaces...

Microelectronic Engineering 132 (2015) 135–155

Contents lists available at ScienceDirect

Microelectronic Engineering

journal homepage: www.elsevier .com/locate /mee

Review Article

Hierarchical micro and nano structured, hydrophilic, superhydrophobicand superoleophobic surfaces incorporated in microfluidics, microarraysand lab on chip microsystems

http://dx.doi.org/10.1016/j.mee.2014.10.0020167-9317/� 2014 Elsevier B.V. All rights reserved.

⇑ Corresponding author.E-mail address: [email protected] (E. Gogolides).

Evangelos Gogolides ⇑, Kosmas Ellinas, Angeliki TserepiInstitute of Nanoscience and Nanotechnology, NCSR Demokritos, Aghia Paraskevi, Attiki 15310, Greece

a r t i c l e i n f o

Article history:Received 31 July 2014Received in revised form 17 September 2014Accepted 5 October 2014Available online 19 October 2014

Keywords:SuperhydrophobicPlasma nanotexturingMicroarraysCell arraysDroplet microfluidicsHydrophilic

a b s t r a c t

Control of wetting properties at the extremes of wetting states (superhydrophilic and superhydrophobic)is important for many applications, such as self-cleaning, anti-fogging, anti-icing, and antibacterialaction. While significant effort has been devoted to develop and characterize such open surfaces forvarious applications, their incorporation in sensors, microfluidics, and labs on chip, offers new functionaldevices and systems, and poses different requirements compared to open-area surfaces. In this paper,dedicated to the 30 year anniversary of Microelectronic Engineering, we aim to review the extremewetting states of surfaces, their fabrication processes focusing on plasma processing technology, andtheir incorporation into devices and systems. We start with an introduction and terminology forsuperhydrophilic, superhydrophobic, and superoleophobic surfaces, and continue with a review of thefabrication of such surfaces by plasma processing. We then review how such surfaces are incorporatedin microdevices and microsystems, and their applications. We address (a) Hydrophilic capillary pumpsand superhydrophobic valves, (b) Drag reduction in superhydrophobic microchannels and slip lengthincrease, (c) Superhydrophobic surfaces for droplet manipulation, applied to chemical and biologicalanalysis, (d) Biomolecule adsorption control on nanostructured surfaces, and (e) Cell adhesion on suchsurfaces. Finally, we conclude with perspectives and challenges.

� 2014 Elsevier B.V. All rights reserved.

1. Introduction and terminology We start with the definitions of super wetting states for which

Control of wetting properties at the extremes of wetting stateshas acquired an increased interest for many applications, such asself-cleaning, anti-fogging, anti-icing, and antibacterial action.Such surfaces are usually structured at the micro and nano-scaleand possess high or low surface energy. On one extreme lie super-hydrophilic and hydrophilic surfaces, while on the other extremesuperhydrophobic and superoleophobic surfaces are encountered.While significant effort has been devoted to develop and character-ize such open surfaces for various applications, their incorporationin sensors, microfluidics and labs on chip, offers new functionaldevices and systems, and poses different requirements comparedto open-area surfaces. In addition, it narrows the possible fabrica-tion steps, so that compatibility with microsystem fabricationtechnologies is possible.

In this paper, we aim to review the extreme wetting states ofsurfaces, their fabrication processes focusing on plasma processingtechnology, and their incorporation into devices and systems.

often multiple terms and some confusion may exist [1,2]. Most ofthe words are combinations of Greek or Latin roots with thefollowing meanings: ‘‘hydro = water’’, ‘‘oleo = elaion = oil’’, ‘‘philic =friendly, attracting’’, ‘‘phobic = afraid of, repelling’’, ‘‘motus = fear’’,‘‘amphi = both (i.e. water and oils for this application)’’, ‘‘omni = all,everything’’. We essentially agree with the classification andterminology proposed in [1], but we prefer to separate oils (typicaledible oils have surface tensions larger than 30 mN/m) from lowersurface tension liquids (e.g. alkanes). We keep the term oleophobicfor oils, and the term omniphobic for all liquids including liquids oflower surface tension compared to oils, considering that ‘‘omni’’ isa root with stronger meaning encompassing all liquids, while‘‘amphi’’ has a weaker meaning referring both to oils and water.Table 1 below summarizes the surface classifications and theterminology used for such surfaces. The reason for such detailedclassification is that the lower the surface tension the moredifficult it is to obtain a superamphiphobic state. In addition asdiscussed in [2], there are some counter-intuitive cases where asurface may be oleophobic, but not hydrophobic, hence the needfor the prefix amphi or omni.

http://crossmark.crossref.org/dialog/?doi=10.1016/j.mee.2014.10.002&domain=pdf

http://dx.doi.org/10.1016/j.mee.2014.10.002

mailto:[email protected]

http://dx.doi.org/10.1016/j.mee.2014.10.002

http://www.sciencedirect.com/science/journal/01679317

http://www.elsevier.com/locate/mee

Table 1Definitions of extreme wetting states.

State Superhydrophilic Superoleophilic Hydrophilic Oleophilic Hydrophobic Oleophobic Omniphobic Superhydrophobic Superoleophobic Superamphiphobic Superomniphobic

Liquid Water Oils Water Oils Water Oils (c >

30 mN/m)

Oils and

liquids with

c < 30 mN/m

Water Oils Both water and

oils

All, water, oils,

and low surface

tension liquids

(c < 30 mN/m)

Static

contact

angle

<10� <10� 10� <

h < 90�

10� <

h < 90�

90� <

h < 150�

90� <

h < 150�

90� <

h < 150�

>150� >150� >150� >150�

Hysteresis na na na na >10�

(sticky

surface)

>10�

(sticky

surface

>10�

(sticky

surface)

<10�

(slippery

surface

<10�

(slippery

surface

<10�

(slippery

surface)

<10�

(slippery

surface)

136 E. Gogolides et al. / Microelectronic Engineering 132 (2015) 135–155

Fig. 1 shows the spectrum of contact angles obtained for dropssitting on a rough super-wetting or super-anti-wetting surfaces.On one hand completely wetted superhydrophilic/superoleophil-ic/superamphiphilic states exist on rough surfaces. On the otherhand at the ‘‘phobic’’ side of contact angles various states arefound, starting from those where liquid wets the rough elements(Wenzel states, or ‘‘sticky’’ hydrophobic states) and ending in thesuper-anti-wetting states with very high contact angles. It isclearly understood that liquid drops easily roll-off on such super-anti-wetting surfaces. A drop can slip on a surface in two ways:(a) by contacting only a small solid fraction of the nanostructuredsurface, while the air fills in the rest of the surface asperities (the socalled Cassie-Baxter or ‘‘fakir’’ state), or (b) by sliding on an oillayer impregnating the surface asperities the so called SLIP mode(slippery liquid-infused porous surface) [3]. While we make noteof the SLIP mode, we shall mostly focus on the ‘‘fakir’’ state, as itinvolves more demanding micro–nano-fabrication. Both modesenable complete control of droplet movement in droplet basedmicrofluidics. However, similar advantages exist in open or closedmicrofluidics, when for example a slip boundary condition isapplied, rather than the well-known no-slip boundary condition.In addition, the known strong relation of wetting properties tobiomolecule/cell adhesion allows ‘‘intelligence’’ and complexityto be added in microfluidics by incorporation of such surfaces.We attempt here a review of such efforts.

2. Fabrication of nanostructured hydrophilic,superhydrophobic and superoleophobic surfaces

The theory, design, fabrication and characterization of suchsurfaces have received enormous attention in recent years. Severalreviews have appeared for superhydrophobic surfaces. Examples ofrecent reviews include a review by Shirtcliffe et al. for polymericsurfaces [4], one by Celia et al. for the design and fabrication ofsuperhydrophobic surfaces [5], one by Yan for the theoreticalaspects of superhydrophobic surfaces, their biomimetic aspect,and their fabrication [6], another by Liu and Jiang [7] for bioin-spired multiscale (hierarchical) structures, yet another by Yaoet al. for self-cleaning and antireflective properties [8], and yetanother by Grinthal and Aizenberg which reviews the SLIPSsurfaces and applications [9]. We note that since water is a liquidwith a large surface tension compared to other common liquidsand oils, it is easy to achieve superhydrophobicity with a ‘‘modest’’topography (e.g. small height topography, not necessarilyhierarchical, with no overhang structures etc.) and an initiallyhydrophobic material without coating or a ‘‘modest’’ hydrophobiccoating (i.e. with initial contact angles in the range 90–105�).However, superoleophobicity, superamphiphobicity, and superom-niphobicity are increasingly more difficult to achieve, as thesurface tensions of oils and alkanes are low, and thus their spread-ing on the surface is easier. Nevertheless, great progress has

recently taken place for superoleophobic surfaces. Recent reviewson the topic include a detailed review by Liu et al. on design,fabrication and application of superoleophobic surfaces [10], ashorter review from the same group by Xue et al. focusing onsuperoleophobic polymers [11], one by Valipour et al. stressingapplications [12], another by Bellanger et al. focusing on the phys-ics and chemistry needed and on the theoretical background forsuch surfaces [13], yet another by Bae et al. which emphasizesthe role of hierarchy on bioinspired structures [14], and yetanother by Chu and Seeger for superamphiphobic surfaces [15].

In the above mentioned publications a panorama of fabricationtechnologies is reviewed and compared. One important technologyfor fabrication of such ‘‘smart’’ surfaces is plasma processing. Thistechnology is most well suited for microsystems and labs on chip,as it is often a technology used in their fabrication, while it allowsthe creation of the whole spectrum of wetting regimes dependingon the plasma chemistry used. Already, a few reviews haveappeared on plasma technology for super-anti-wetting surfaces.Vourdas et al. [16], and Gogolides et al. [17] reviewed the workon plasma nanotexturing of polymers for superhydrophobicityand antireflectivity, Kylian et al. reviewed nanostructured, low-pressure, plasma deposited polymers [18], while Jafari et al.reviewed the plasma technology for superhydrophobicity [19].

2.1. Randomly nanostructured surfaces via plasma etching ordeposition

Plasma technology – usually at low pressure – can be used invarious modes, such as plasma etching, plasma deposition, andsputtering with inert gases. Plasma etching or sputtering of poly-mers with oxygen or noble-gas plasmas was found early-on tocause roughening of the polymers. Mora et al. [20] in 1988observed that oxygen plasma treatment of Poly(tetrafluoroethyl-ene) (PTFE) at a low pressure of 2 Pa created topographical andwetting changes on the material and observed the transition to asuperhydrophobic-like state at long (15 min) etching. A few yearslater in 1993 Ogawa et al. [21] roughened glass surfaces in CHF3/O2 plasmas and coated them with a perfluorinated monolayer torender glass superhydrophobic. In 1999 Youngblood et al. treatedPoly(ethylene) (PE) and PTFE simultaneously in an Ar plasma,and observed roughening of PE and superhydrophobic behaviorafter 100 min of treatment [22]. Plasma deposition was also usedas early as 1982 to deposit highly non-wettable films, such as PTFE[23], and in 2000 to deposit superhydrophobic fluorocarbon(PTFE-like) polymer in CH4/C4F8 plasmas [24]. These early effortswere reviewed in 2001 by Nakajima et al. [25].

Etching of organic crystalline polymers at relatively large pres-sures was shown to create roughness on their surface. The rough-ening step can be followed by a hydrophobic deposition layereither in the plasma or by other methods. This way Teshimaet al. [26] produced superhydrophobic Poly(Ethylene terephthal-

Fig. 1. A drop on rough (micro and nanostructured) and smooth surfaces with varying wetting properties: (a) a rough porous superhydrophilic (superoleophilic) surface,(b) a smooth hydrophilic (oleophilic) surface, (c) a smooth hydrophobic (oleophobic) surface, (d) a rough hydrophobic (oleophobic) porous sticky surface where the drop is ina Wenzel ‘‘sticky’’ state, (e) a hydrophobic SLIPS surface (slippery liquid infused porous surface), (f) a rough superhydrophobic (superoleophobic) surface, where the drop is ina Cassie-Baxter, slippery ‘‘fakir’’ state.

E. Gogolides et al. / Microelectronic Engineering 132 (2015) 135–155 137

ate) PET at 10 Pa after 10 min etching introducing the termnanotexture to describe the nanoroughness created, while Fresnaiset al. [27] fabricated superhydrophobic poly(ethylene) PE in amicrowave plasma of O2 and CF4.

In 2005 Tserepi et al. described a short (1–3 min) plasma etch-ing step in a high density helicon plasma reactor to create wavystructures of enhanced surface area or high aspect ratio polymericnanofilaments [28], that could grow several microns in heightwithin a few minutes etching, thus greatly increasing the surfacearea of the polymer. The process used etching gases appropriatefor the polymer being etched (SF6 for Si containing polymers suchas Poly(dimethyl)siloxane PDMS, or O2 for organic polymers suchas Poly(methyl)methacrylate PMMA). This process was alsodescribed as micro and nanotexturing of polymers, and gavesuperhydrophilic nanostructured surfaces, while a subsequentfluorocarbon plasma deposition [29] could render them superhy-drophobic and antireflective as described in Tserepi et al. [30],Vourdas et al. [31] and Kontziampasis et al. [32]. Plasma nanotex-turing was also used in low density plasma etching reactors bymany other groups, see for example the work by Wohlfart et al.[33], [34] for PET, Balu for paper [35], Fernández-Blázquez [36]for various organic polymers, Palumbo et al. [37], Di Mundo et al.[38], Vietro et al. for superhydrophobic polycarbonate forautomotive applications [39], Tarrade et al. for Poly(ethyleneterephthalate) [40] and many others.

Fig. 2 shows various micro and nanotextured surfaces createdby plasma etching: Fig. 2(i) shows PET [26], Fig. 2(ii) showsNanotextured PMMA [41], Fig. 2(iii) shows PET etched at variouspressures [34], Fig. 2(iv) shows PC (left) and PS (right) surfaces[38], Fig. 2(v) shows PDMS surfaces [42], while Fig. 2(vi) shows fastsuperhydrophobic surface fabrication on organic and inorganicpolymers as evidenced by the abrupt contact angle change versustime of etching for each polymer [17]. The polymeric micro andnanofilaments created by a long plasma nanotexturing processcan be fragile against capillary, bending, or adhesion forces. For thisreason Gnannapa et al. proposed a wetting-drying method tostructurally stabilize them prior to hydrophobization [43].

While in some cases plasma etching and nanotexturing may bedue to differential etching of components of the material (e.g.crystalline versus amorphous phase), in most cases, nanotexturingis due to the presence of etching inhibitors coming from the reactorwalls or present in the material, and the prevailing anisotropicetching conditions. This has been discussed in detail by Tsougeni[44] for organic polymers and Vlachopoulou et al. [45,46] forPDMS. A detailed study of the role of reactor walls was presented

by Gogolides et al. [47], while simulation of nanotexture formationand growth was presented by Kokkoris et al. [48].

Plasma etching and nanotexturing is not only possible withpolymers, but also for silicon and glass [21]. ‘‘Black’’ Silicon isone such example of random silicon structures which leads tosuperamphiphobicity [49], [50] and overhanging nanostructures[51]. Black Silicon structures can also be transferred to polymersto render them superamphiphobic [52,53].

Plasma deposition was also used to deposit randomly nanostruc-tured fluorocarbon coatings with superhydrophobic properties (seeFig. 3(i) and (ii)). In addition to the early works mentioned above, wenote also the work of Favia [54], and Milella for modulated plasmasexploring the mechanisms to deposit ribbon-like or bumpy fluoro-carbon layers [55], as reviewed also by the same author later on[56]. Vacuum deposition of superhydrophobic coatings was alsoreviewed by Cigala et al. [57], and Kylian et al. [18]. One of theadvantages of plasma deposition is that it can be used at atmo-spheric pressures as well. The early work of Tsoi et al. who depositedSiO2 networks [58] was followed by several more, as for exampleLee et al. who deposited superhydrophobic hydrocarbon polymers[59], and Wang et al. who used atmospheric plasma jet to depositsuperhydrophobic TiO2 trees (see Fig. 3(ii)); however a longdeposition time was necessary (over 1 h) [60]. Mechanisms ofplasma deposited nanotexture formation and transition from bumpto ribbon structures were discussed by Milella [55].

2.2. Ordered hierarchical, micro–nanostructured surfaces

It is often required that the micro and nano scale structures areclearly separated, providing a constant, well-defined distancebetween the microscale elements. Such requirements may beposed by the application (e.g. inkjet printer heads) [61], the fabri-cation method (top down technologies), or by the need for highthermodynamic and mechanical stability offered by suchstructures [62]. Such length scale separation is accomplished bydoing a first lithographic step followed by the etching and thenanotexturing steps. Numerous such examples exist in the litera-ture (see Fig. 4). Marquez-Velasco et al. [63] created SU8 photore-sist microposts, and they subsequently nanotextured their surfaceusing oxygen plasma (see Fig. 4(i)). Bormashenko et al. [64] hotembossed PE and then nanotextured it in CF4 plasma to createsuperoleophobic PE surfaces (see Fig. 4(ii)). Cortese et al. [65] cre-ated micropatterns on PDMS and then nanotextured them in theplasma to create hierarchical superhydrophobic surfaces. Similareffects have appeared for Silicon using Lithography and Deep Si

Fig. 2. (i) Nanotextured PET, 200 W, 10 min Oxygen plasma, Reprinted with permission from Ref. [26] Copyright 2005 Elsevier. (ii) Nanotextured PMMA in high densityoxygen plasma, by Skarmoutsou et al., Conditions: 1900 W, �100 V bias, 0.75 Pa, 100 sccm O2. Reprinted with permission from Ref. [41] Copyright 2012 IOP Publishing. (iii)PET surfaces etched at various pressures. Nanotexturing appears at low pressures and is a sign of anisotropic etching mechanisms, reprinted with permission from Ref. [34]Copyright 2010 John Wiley and Sons. (iv) PC (left) and PS (right) surfaces etched and nanotextured in Oxygen Plasma. Similar structures are obtained in CF4 plasmas, reprintedwith permission from Ref. [38] Copyright 2012 John Wiley and Sons. (v) PDMS etched in SF6 plasmas, reprinted with permission from Ref. [42] Copyright 2011 Elsevier. (vi)Fast superhydrophobic surface fabrication on organic and inorganic polymers using oxygen and SF6 plasmas respectively, followed by 30 s fluorocarbon polymer deposition.Contact angle change versus time of etching for each polymer. Reprinted with permission from Ref. [17] Copyright 2010 Inderscience.


plasma etching. Kwon et al. fabricated micropillars, which weresubsequently nanotextured using XeF2 etching [66]. Nguyen et al.fabricated also micropillars and compared two nanotexturing pro-cesses, metal assisted electroless etching, and chemical vapordeposition of silicon nanowires, the first showing only superhy-drophobicity, the second showing superamphiphobicity [67] (seeFig. 4(iii)). All such works demonstrate the advantages of hierarchi-cal topography; enhancement of hydrophobicity to superhydrop-hobicity and mostly the significant increase of the stability of theCassie state of such surfaces.

The idea that the slope of the microstructure is important forthe drop suspension was first discussed theoretically by Extrand[68,69]. Tuteja et al. in two successive publications demonstratedthat for superoleophobicity a re-entrant microstructure profile is

needed [70,71], the angle of the profile defining one of their designcriteria. They fabricated ‘‘micro-hoodoo’’, mushroom-like Si struc-tures to prove the concept. At the same time, Ahuja et al. [72] pro-duced Si nanonail superoleophobic surfaces (see Fig. 4(iv)), and byapplying a voltage on suspended drops caused transition from Cas-sie-Baxter to Wenzel regime. Zhao et al. demonstrated that whenSi microstructures are being etched with a Bosch process, the rip-ple itself induces a re-entrant profile leading to superoleophobicity[73]. When an overhang ‘‘capping layer’’ is added superoleophobic-ity is favored as the thickness of this layer gets smaller (seeFig. 4(v)) [61]. Overhang structures where produced in Silicon bySusarrey et al. [74] to study drop evaporation, and by Zeniouet al. to study Si nanowire mechanical stability and hydrophobicity(see Fig. 4(vi)) [75].

Fig. 2 (continued)


Colloidal micro–nanosphere lithography has early enough beenalso proposed as an ‘‘alternative’’ cost-effective lithographicmethod [76]. When combined with plasma etching and nanotex-turing, it can lead to nanostructured surfaces spanning the rangefrom superhydrophilic to superoleophobic. Shiu et al. [77] proposedthe use of colloidal lithography for superhydrophobic surfaces.Ellinas et al. [78] proposed ‘‘mesh-assisted’’ colloidal lithographyfor more uniform large area applications on silicon and polymers.He also designed the shrinking and simultaneous etching of PSspheres on top of PMMA to create triple-scale, overhangingsuperoleophobic polymeric surfaces [79]. These surfaces (seeFig. 4(vii)) were the analog of ‘‘mushroom-like’’ surfaces forpolymers. An isotropic etching step enhances the re-entrant profileof these structures. Such surfaces were shown to be scratchresistant and chemically resistant upon immersion to water andhexadecane for periods of months [62]. An analogous approach,where plasma etching controls the geometrical characteristics hasbeen proposed by Zhang et al. [80] for Si using silica microspheres.Hirai et al. created ordered Si nanostructures by self-assembly of apolymer film, followed by Si anisotropic plasma etching [81].Zeniou et al. used also colloidal lithography and a pulsed plasmaetching process to create superhydrophobic Si nanowires of ultrahigh aspect ratio (see Fig. 4(vi) as discussed above). Controlling

the first pulsing step allows control of the Si nanowire diameterunder the sphere, and results in oleophobicity [75].

While plasma nanotexturing has been described above (seeSection 2.1) as leading to random nanostructure formation onpolymers, under specific conditions the plasma itself can directthe assembly of small, quasi-ordered, organized nanostructures:

On one hand, this is possible using non-etching, or mildlyetching plasmas, for example noble gas mixtures for organicpolymers, which lead to their surface graphitization, or Oxygenrich mixtures for PDMS, which lead to surface oxide formation.Both the graphitized layer and the oxide layer may developstresses on the interface with the un-modified polymer leadingto bump or ripple formation (see Fig. 4(viii)). The mechanism ofsurface oxidation of Silicon containing polymers has beendiscussed in detail by Eon et al. [85] when etching Poly(hydraloligomeric Silsesquioxane) POSS polymers in oxygen plasmas,while roughness formation and superhydrophobic PDMS forma-tion in Oxygen plasmas were discussed by Tserepi et al. [86,87].The same group has also shown that performing oxygen plasmatreatment on pre-patterned PDMS leads to the development ofnew self-aligned structures [82]. The structures in PDMS (seeFig. 4(viii)) present order and self-alignment perpendicular to thepre-existing patterns, despite the lack of any lithographic step

Fig. 3. (i) Micro-ribbon like superhydrophobic coatings deposited by plasma processing, reprinted with permission from Ref. [54] Copyright 2003 Elsevier. (ii) Nanotrees byatmospheric plasmas, reprinted with permission from Ref. [60] Copyright 2011 Elsevier.


[87]. For photoresist etching in fluorocarbon/Ar plasmas, nano-structure formation has been discussed in detail by Shimiya et al.in a series of papers [88–90] as due to the surface layer formedduring the first 3 s of etching (see Fig. 4(ix)).

On the other hand, in addition to the stress mechanismdescribed above for oxides and surface graphitized or cross-linkedlayers, the etch inhibitor mechanism can also lead to organizednanostructure formation on surfaces. Xu et al. [91] observednanocone formation in Si when etching in Hydrogen containingmixtures, while Gogolides et al. [92], Vourdas et al. [84] andKontziampasis et al. [93] described ordered nanodot formationon oxygen plasma etched polymers at zero bias conditions (seeFig. 4(x)). Order was verified by a sharp peak in the Power SpectralDensity (PSD) of the surface heights. These phenomena lead to theconclusion that plasmas can direct the formation of organizedstructures, hence the term plasma directed organization was used.

3. Nanostructured, hydrophilic, superhydrophobic andsuperoleophobic surfaces incorporated in microdevices,microsystems and applications

The unique properties of nanostructured surfaces withextremes in wettability have inspired applications in microfluidics,sensors, microarrays and biomicrosystems. Numerous applicationsare possible such as creation of hydrophilic/superhydrophobicpatterns for confinement of drops, or droplet manipulation,patterning superhydrophilic channels for liquid transport, creatingpaths for cell growth or cell microarrays, and many more openmicroarray-like applications as they have been recently reviewedby Ueda and Levkin [94]. On the other hand microfluidics (closedor open) is also becoming a mature field with many diverse appli-cations requiring multiple functionalities. The progress in the fieldfocusing on biomedical research applications has been recentlyreviewed by Sackmann et al. [95], who stressed the need for simplesolutions to incorporate functionality and operate microfluidics.We will review below applications resulting from incorporationof nanostructured hydrophilic and superhydrophobic surfaces onlabs on chip focusing on the following aspects: Hydrophilicpumping and superhydrophobic valving for liquid movementcontrol, droplet microfluidics control, drag reduction surfaces,

antifouling and cell-repelling surfaces, microarrays and spots forsingle molecule detection, cell arrays and cell culture surfaces.

3.1. Hydrophilic capillary pumps and superhydrophobic valves

The use of hydrophilic microchannels for capillary pumping isan obvious application in microfluidics. In addition, the use of ahydrophobic patch in such a hydrophilic microchannel introducesa passive valve. The topic of valves on chip, including passivevalves, has been reviewed by Oh et al. [96].

However, recently intricate microfluidic geometry designs havebeen proposed to fabricate elegant and accurate pumps withoutmoving parts, to be used for point-of-care diagnostics (POC). TheIBM group in Zurich has produced several publications focusing on‘‘silicon-based’’ capillary microfluidic pumps [97], and their applica-tion as autonomous microfluidics for diagnostics [98–103] (seeFig. 5(i)). First, they discussed hydrophilic capillary pumps, delayvalves, and trigger valves all with intricate geometrical designs fabri-cated on Si [97]. They integrated this with on-chip mixers, reactors,and detection zones on an autonomous system [102]. They laterincluded controlled release of reagents [99], and integrated the com-ponents in a capillary driven autonomous chip for one-step immuno-assays [101] (see Fig. 5(ii)). They recently reviewed their efforts forPOC labs on chip [103]. Later, the same group proposed also simpler,hand-pressure actuated soft valves for filling in microfluidics [100].This platform has also been proposed for open microfluidics to studyreaction on biointerfaces in a microfluidics format [98]. The conceptof geometric shrinkage was also proposed for valving in microchan-nels with hydrophilic and hydrophobic walls by Zhang et al. [104].

Incorporation of a hydrophobic or superhydrophobic area-patchis the simplest way to create a valve on chip [105]. First efforts tocreate such valves were done as early as 2000 by Handique et al.[106] who proposed a hydrophobic patch inside a hydrophilicmicrochannel in order to construct a nanoliter measuring device.Andersson et al. [107] proposed a plasma deposited Teflon hydro-phobic valve, while hydrophobic patters where also proposed bySuk et al. to control capillary flow [108]. Londe et al. [109] one yearlater proposed a superhydrophobic valve with a thermosensitivebehavior (see Fig. 5(iii)), while Washe et al. proposed electrochem-ical actuation of such valves [110] .

Fig. 4. (i) SU8 microposts after oxygen plasma nanotexturing to create hierarchical dual scale morphology. Detail of the nanotexturing is shown as inset. Reprinted withpermission from Ref. [63] Copyright 2010 Elsevier. (ii) Imprinted PE with a metalic stamp to create micro structures. Details of the top of microstructures before and after CF4

plasma nanotexturing and hydrophobization. Reprinted with permission from Ref. [64] Copyright 2013 Elsevier. (iii) Hierarchical Si surface with Si nanowires grown on Simicropillars. Evidence of re-entrant geometry, which causes superoleophobicity, reprinted with permission from Ref. [67] Copyright 2014 Elsevier. (iv) Scanning electronmicroscopy (SEM) image of 2-lm-pitch nanonails. The nail head diameter is about 405 nm, the nail head thickness is about 125 nm, and the nanonail stem diameter is about280 nm, reprinted with permission from Ref. [72] Copyright 2008 American Chemical Society. (v) SEM micrographs of�3-lm-diameter Si pillar array etched with a Boschprocess producing a rippled pillar surface, and hydrophobized with FOTS (b) 6 lm, and (d) 12 lm center-to-center spacing. (Insets: sessile drops of water and hexadecane onthe pillar array surfaces). Reprinted with permission from Ref. [61] Copyright 2012 American Chemical Society. (vi) Ultra high aspect ratio, superhydrophobic Siliconnanowires etched with a pulsing process using a colloidal lithography mask, Reprinted with permission from Ref. [75] Copyright 2014 IOP Publishing. (vii) PS microspherelithography followed by plasma etching and nanotexturing to create overhang superoleophobic surfaces. Reprinted with permission from Ref. [62] Copyright 2014 AmericanChemical Society. (viii) Quasi-ordered Topography on initially flat PDMS after 7 min processing in Oxygen Plasmas. Aligned nano-topography formation on a similar initiallypre-patterned sample, after 10 min etching. Reprinted with permission from Ref. [82] Copyright 2008 Elsevier. (ix) Size of nanostructure formation on various polymers afterplasma etching due to surface layer formation, reprinted with permission from Ref. [83] Copyright 2011 John Wiley and Sons. (x) Plasma directed organization on PMMApolymer (top), transferred to the Si substrate (middle), and Power spectrum of surface morphology showing quasi-order. Reprinted with permission from Ref. [84] Copyright2010 IOP Publishing.


Fig. 4 (continued)

Fig. 4 (continued)


Fig. 5. (i) Capillary pump with delay valves and trigger valves with different threshold. Reprinted with permission from Ref. [97] Copyright 2008 Springer. (ii) Capillary-driven microfluidic chip for multiparametric immunoassays. (a) Photograph of a chip having various microfluidic functional elements (see insets) for holding and dissolvingdAbs, mixing them in a stream of sample and splitting the sample into 6 streams passing over cAb areas at flow rate conditions determined by the resistance of channelsconnecting to individual capillary pumps. (b) Equivalent electrical model of the chip used to calculate flow rates and filling times. (c) Illustration of the release and evendistribution by mixing of dAbs in a stream of sample passing through the different components of the microfluidic chip and that can result in variable amount of signal in thecase of a sandwich fluorescence immunoassay. Reprinted with permission from Ref. [101] Copyright 2011 Elsevier. (iii) A schematic of a superhydrophobic valve connectingtwo outlets and one inlet. Reprinted with permission from Ref. [109] Copyright 2008 Elsevier. Here the valve opens as a result of temperature responsive properties of thepolymer. (iv) Schematic of the fabrication process of a nanotextured hydrophilic polymeric microchannel with superhydrophobic patches acting as valves: (a) a thin inorganicphotoresist (ORMOCER) is deposited as etching mask on a PMMA substrate, (b) lithography on photoresist polymer using mask exposure, (c) photoresist development, (d)deep oxygen plasma etching of PMMA, substrate, with simultaneous plasma nanotexturing. Alternative to steps (a)–(c) one could hot emboss a microchannel and use step (d)only to nanotexture the channel and make it hydrophilic. (e) optional fluorocarbon plasma deposition of Teflon-like polymer through a stencil mask to define asuperhydrophobic valve, and (f) sealing with pressure-sensitive adhesive lamination film. Reprinted with permission from Ref. [111] Copyright 2013 Springer. (v)Hydrophobic recovery of PEEK (black circles) and PMMA (blue triangles) plates after nanotexturing in O2 plasmas for 20 min (bias voltage: �100 V, pressure: 0.75 Pa, power:1900 W, O2 flow: 100 sccm). For comparison, hydrophobic recovery of smooth PMMA (red triangles) plates after treatment in O2 plasmas for 1 min, under mild conditions,typical for plasma modification processes (pressure: 13.33 Pa, power: 100 W, O2 flow: 50 sccm RIE) is shown. Untreated PMMA and PEEK contact angles are also indicated.Notice the large delay in hydrophobic recovery for the plasma nanotextured surfaces which preserves superhydrophobicity from 10 days to 1 month at least. Reprinted withpermission from Ref. [105] Copyright 2010 The Royal Society of Chemistry. (vi) A red dye–water solution in a PMMA microchannel with integrated superhydrophobic,hydrophobic and superhydrophilic stripes. An optical microscope image of the microchannel with the variable wetting characteristics. The images show the water contactangle on the hydrophobic, superhydrophobic and superhydrophilic stripes in the microchannel inner surface, reprinted with permission from Ref. [105] Copyright 2010 TheRoyal Society of Chemistry. (vii) Operation of a passive superhydrophobic microvalve. Photograph of the microchannel inside the chip holder. The hydrophobic patch is placedin the middle of the microchannel, and its SEM image is shown as inset. (a) Stopped fluid flow (red dye) at the nanotextured hydrophobic patch, (b) filled microchannel afterthe valve opens. Reprinted with permission from Ref. [122] Copyright 2014 Springer. (viii) A capillary valve without any treatment (a) cannot stop the mixed solutioncontaining 0.1 wt% BSA and 1 wt% food dye after protein blocking. Also with (b) CYTOP or (c) CYTOP-polyaniline treatment, the capillary valve can hold the mixed solution.Reprinted with permission from Ref. [113] Copyright 2009 American Institute of Physics. (ix) Schematic representation of the methodology employed for the patterning of thesurfaces, where hydrophilic/superhydrophilic channel-like regions can be imprinted onto superhydrophobic surfaces. The fluorescent microscopy image highlights thesection of a superhydrophilic channel on the SH surface after depositing and drying a FITC solution along the channel. Reprinted with permission from Ref. [114] Copyright2010 APEX/JJAP/OR. (x) Patterned Al substrate (a1) and mask (a2) used for demonstrating a multi-step functional surface device capable of performing pumpless liquidbridging and draining. Elevated end view sequence displaying liquid bridging and draining on the design shown on the left applied on a horizontal paper substrate. Reprintedwith permission from Ref. [116] Copyright 2014 The Royal Society of Chemistry. (xi) Image of a hexadecane droplet moving along a predefined micropath because of thepresence of virtual walls in a top-covered device. The virtual wall comprises plasma etched holes with undercut profile. Reprinted with permission from Ref. [118] Copyright2013 The American Chemical Society. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)


Fig. 5 (continued)


Inspired by these early works on hydrophobic valves, and inparallel to the elegant works where pumping and valving on Sichips is controlled by geometry, Tsougeni et al. [105], Gogolideset al. [17], and Papageorgiou et al. [111] proposed plasma etchingand simultaneous plasma nanotexturing (roughening) of poly-meric microfluidics to control their wettability. This technology(see Fig. 5(iv)) allows the fabrication of nanostructured superhy-drophilic polymeric microchannels, which can be used as stablein time (non-ageing) hydrophilic pumps (see in Fig. 5(v) the longterm stability of contact angles); Vourdas et al. observed that suchmicrochannels could also be used for increased electroosmoticflow due to their higher electroosmotic mobility [112]. When adepositing fluorocarbon plasma [29] was ignited over such chipscovered with a stencil mask, area-selective superhydrophobicpatches can be incorporated on such a chip (see Fig. 5(vi)).

Ellinas et al. recently demonstrated that microchannel dimen-sions (width, height) determine opening pressure of such valves,while plasma nanotexturing time (i.e. height and spacing of thenanostructures) allows fine tuning of the opening threshold (seeFig 5(vii)). A microfluidic switchboard was proposed to allowsequential filling of three wells (to be used as detection areas in aPOC), by having different valves in front of each well. In this way,they created the analogous of sPROMs (structurally programmablemicrofluidic systems) proposed by Ahn’s group simply by alteringthe etching duration [96]. In sPROMs, Ahn used an intricate design

with constrictions to act as delay valves, and a detonator on chip, totake advantage of the lack of stable (non-ageing) capillary pumping.A combination of a geometric design and a superhydrophobiccoating was also proposed to incorporate superhydrophobic valveson Lab on Disk systems by He et al. [113], who used this technologyfor an ELISA on disk application (see Fig. 5(viii)).

The concept of capillary pumping on nanostructured surfacescan also be used for open microfluidics, where the walls of thechannels can be superhydrophobic areas. This concept has beenrealized by Oliveira et al. [114] who created rough superhydropho-bic PS slides by phase separation. Following a UV and ozone or aplasma exposure through a stencil mask, hydrophilic paths werecreated (see Fig. 5(ix)). A similar approach was used for creatinghydrophilic stripes on superhydrophobic paper [115]. Recentlythe same concept was used by Ghosh et al. [116], who fabricatednanostructured superhydrophobic titania, and created superhydro-philic areas by illuminating through a photomask. The differencebetween sticky hydrophobic and superhydrophobic areas has beenused to extract liquid drops from liquid volumes on surfaceshaving patterns of sticky wax on superhydrophobic paper (seeFig. 5(x)) [117]. While superhydrophobic surfaces can confinewater drops and allow their movement in superhydrophilicstripes, things are more difficult when oils or low surface tensionliquids have to be confined in oleophilic stripes. As has beendiscussed in Section 2 above, re-entrant microstructures favor

Fig. 5 (continued)


superoleophobicity. Therefore, Almeida et al. in order to confinehexadecane inside a shallow microchannel they created virtualsuperoleophobic walls with posts or holes, allowing the liquid toslide past them (see Fig. 5(xi)) [118].

Finally it should be mentioned that a superhydrophobic andsimultaneously oleophilic porous surface could be used as a filterof oil/hydrocarbon suspensions in water solution: Hydrocarbonscan pass through, while water is repelled [119–121].


3.2. Drag reduction in superhydrophobic microchannels and slip lengthincrease

Laminar flow in microchannels with superhydrophobic wallsresults in reduced pressure drop (drag reduction). This is due tothe fact that on the superhydrophobic wall a slip boundarycondition is applied, and a non-zero slip velocity exists. The firststudies to incorporate superhydrophobic surfaces in microchan-nels used model surfaces with microposts or microribs, whichcould not sustain large pressures. Later, Choi and Kim [123] usedsuperhydrophobic ‘‘black Silicon’’ nanoneedles, while Journetet al. [124] and Joseph et al. [125] incorporated superhydrophobiccarbon nanotubes inside microchannels and studied the flow pastsuch surfaces. Papageorgiou et al. [111] used plasma nanotex-tured superhydrophobic PMMA as a bottom superhydrophobicwall, and observed pressure reduction by 20% comparing thehydrophilic nanostructured and the superhydrophobic nanostruc-tured microchannels. The superhydrophobic state was preservedfor pressures higher than 3 bar, as confirmed by staining withred dye, which stained only the top part of the roughness asper-ities. Slip lengths of larger than 1.2 lm were estimated. The abil-ity to control slip by creating bubbles in a microchannel, whichprotrude a specific angle into the flow, was proposed by Karatayet al. [126], who measured slips lengths of 3–4 lm (see Fig. 6(i)).Lee et al. fabricated microchannels with microposts and blacksilicon at the bottom. This robust superhydrophobic surfaceresisted pressures up to 7 bar, by allowing regeneration of theair layer via electrolysis (see Fig. 6(ii)) [127]. Recently, Zhouet al. studied superhydrophobic microchannels theoretically[128], while Dey et al. [129] studied the flow with lPIV in super-hydrophobic microchannels with lotus leaf internal surface andfound a critical flow rate above which the superhydrophobicCassie state collapses into a Wenzel state, which induces flowdisturbances eventually penetrating into the liquid and causingsome mixing (see Fig. 6(iii)).

In addition to the above discussed velocity increase for pressuredriven flow Vourdas et al. [112] observed that electroosmoticvelocity were increased in plasma etched superhydrophobic walls,even more compared to superhydrophilic walls. This suggestedincreased electroosmotic mobilities in superhydrophobic micro-channels, almost by a factor of 2 compared to untreated PMMAand by 50% compared to hydrophilic nanotextured PMMA (seeFig. 6(iv)).

The challenge in this particular application is probably toachieve superhydrophobicity on all walls, and not only in the bot-tom wall. To this end Liu et al. [130] create nanostructured PDMSfilm, which they incorporate inside large (mm wide) microchan-nels. Chakraborty et al. [131] produced PS microchannels withmicro–nano textured pillars using oxygen plasma etching andstrain recovery deformations of a PS sheet.

3.3. Superhydrophobic surfaces for droplet manipulation in digitalmicrofluidics for chemical and biological analysis

One of the first observations of depositing drops on superhydro-phobic surfaces is that drop retains its spherical shape during mostof the evaporation process, until it is shrank down to a very smalldrop [74]. If the drop contains water solution of a chemical or bio-logical molecule, its concentration is constantly increasing insidethe drop, while water is evaporating. At the same time, since thedrop contact angle hysteresis is minimal, no residues are leftbehind as the drop contact area is shrank, and no ‘‘coffee-ring’’effects are observed. As a result, all the chemical or biological mol-ecules can be deposited onto a ‘‘single point’’, facilitating detectionfrom highly sensitive methods, such as X-ray, FTIR, micro-Ramanor other techniques as proposed by an Italian group [132–137].

The same group have applied this concept also for cells and cellexosomes [138,139], while they have recently improved the designof the superhydrophobic pattern on which the drop will condenseby proposing non-periodic designs [140] (see Fig. 7(i)). Superhy-drophobic nanostructured Silicon surfaces have also been used astargets for Matrix-Assisted or Matrix-less Laser Desorption Ioniza-tion spectrometry (MALDI). For example Ressine et al. proposed asuperhydrophobic nanoporous silicon with hydrophilic dropanchor points as a MALDI target [141].

Superhydrophobic surfaces offer minimal resistance to dropmovement with drops sliding on them without friction [145]. Thisproperty has been used for droplet transport using for exampleelectric or magnetic fields. Seo et al. used superhydrophobic mag-netic elastomer and magnetic forces to control the movement ofdroplets towards digital microfluidics [146]. Draper et al. incorpo-rated prepatterned superhydrophobic surfaces inside microfluidicsto allow movement, acceleration, merging and other control func-tions for droplets [147]. The field of droplet microfluidics (alsocalled digital microfluidics) has been reviewed recently by Sharmaet al. [148] focusing on dielectrophoresis (DEP) and electrowettingon dielectric (EWOD), while applications in chemistry werereviewed by Choi et al. [149].

Probably the most well-known application of hydrophobic sur-faces and fluorocarbon coatings for droplet manipulation is elec-trowetting on dielectric EWOD, where an electric field changesthe water contact with a surface, while an array of electrodes per-mits droplet movement. Spin-on or plasma deposited fluorocarboncoatings can be used. Bayati et al. proposed plasma deposited fluo-rocarbon layers for biofluid transport via electrowetting[29,150,151]. The use of superhydrophobic surfaces (micropillarsor nanonails) for EWOD was proposed by Krupenkin [152], andAhuja et al. [72] to induce Cassie-Baxter to Wenzel transition ondroplet by voltage application for several liquids. Reversible transi-tion was achieved by Krupenkin et al. [142] (see Fig. 7(ii)) byapplying a short electrical pulse, which causes a momentary evap-oration of a thin liquid layer adjacent to the substrate, to move thedrop back to the Cassie state. A recent review summarizes the useof superhydrophobic surfaces for EWOD [153] The topic was lateralso discussed by Lifton and Simon [119], while Dynamic EWODexperiments were conducted by Lee et al. on stretched andunstreched superhydrophobic Teflon [154].

Superhydrophobic surfaces often have defects on which a dropcan be pinned while sliding. Elegant applications may arise byclever use of such defects. Indeed, artificial defects may be createdto control drop movement. The combination of gravity and electricforces was used recently by T’ Mannetje et al. who allowed EWODforces to act as defects and pin a drop movement [143] (seeFig. 7(iii)). Control of drop movement can also be done on a poroussurface when vacuum or overpressure is applied on the backsideallowing drop rolling or pinning as described by Vourdas et al.[155]. The concept of using an array of ‘‘defects’’ rather islands ofhydrophilic sites on a superhydrophobic surface was recently pro-posed by Neto et al. [144]. Mimicking the ‘‘rose-petal’’ effect theyproduced an array of microindentations on a superhydrophobicsurface. Each indentation was used to deposit a droplet, whichserved as a microreactor for chemical or biological reactions orfor cell culture. In the case of cells, they turned the chip upside-down producing spheroids to be used as micro-tissues for drugscreening tests (see Fig. 7(iv)).

3.4. Biomolecule and other molecule adsorption control, fouling andantifouling surfaces

Nanostructured surfaces offer increased surface area, which ifcombined with appropriate surface chemistry may greatly facili-tate (bio)molecule adsorption and allow control of their binding.

Fig. 6. (i) (left) Controllable microfluidic bubble mattress and computational bubble unit cell. (A) Optical image of the microfluidic device with integrated gas (G) and liquid (L)channels, with the inlets and outlets indicated. (B) Scanning electron microscopy image of a representative microfluidic device, showing two main microchannels for gas (Pg)and liquid (Qw) streams connected by gas-filled side channels. (C) Bright-field microscopy image of bubbles protruding 35� ± 3.3� into the liquid microchannel with a height, H.The shear-free fraction, u, is defined as Lg = L = Lg = Ls + Lg, where Lg is the width of the gas gap, and Ls is the width of the solid boundary. (right) Effective slip length beff andeffective friction factor Cf as a function of the protrusion angle h obtained by lPIV measurements and numerical calculations. (A) Experimental and numerical beff results foru = 0.54 and u = 0.38. (B) Experimental and numerical Cf results for u = 0.54 and u = 0.38. In A and B, the solid line (—) and the circles (d) indicate the numerical andexperimental results for u = 0.54. The dashed line (—) and the squares (j) indicate the numerical and experimental results for u = 0.38. The horizontal black dashed linerepresents the value Cf = 0:8 obtained for the no-slip condition b = 0. Reprinted with permission from Ref. [126] Copyright 2013 Proceedings of the National Academy ofSciences. (ii) Proposed scheme to restore gas film underwater, shown for micropost. (a) Estimated range of height-to pitch ratio (H = L) that allows formation of a gas filmbetween posts as function of their gas fraction with CAbrec as a parameter. Red line represents the minimum H = L and black line the maximum H = Lacceptable. Microposts ofvarying gas fraction and pitch (while height is fixed at 50 m), denoted with d, have been fabricated on both smooth and nanostructured bottom surfaces and tested. The gas filmrestoration was achieved only when the acceptable zone was expanded due to the increased CArec on the nanostructured bottom surface, qualitatively agreeing with the theory.(b) Schematic representation of the successful gas film formation. (c) Scanning electron microscopy images of a test surface fabricated based on the above criteria. Reprintedwith permission from Ref. [127] Copyright 2011 American physical society. (iii) (a) A confocal microscopy image (at a 20� magnification with a Leica TCS-SP8; obtained forindependently performed experiments) showing the entrapped air-pockets between the lotus leaf replica and the liquid due to the Cassie–Baxter state, as in Regime I. The insetshows the close-up of an entrapped air-pocket. (b) A confocal microscopy image (at a 40� magnification; obtained for independently performed experiments) showing thelotus leaf replica–liquid interfacial condition due to the Wenzel state, as in Regime II. (c) Variation of the critical flow rate, i.e. the maximum sustainable flow rate in the RegimeI, with varying confinement effects, as represented by the ratio of the lotus leaf replica feature size to the microchannel hydraulic diameter. The theoretical estimation of thecritical flow rate agrees quite well with the experimentally evaluated values. The ranges of the in-flow rate, necessary for controlling the flow in the desired regime, can be alsoestimated from this graphical representation. Reprinted with permission from Ref. [129] Copyright 2014 The Royal Society of Chemistry. (iv) EOF velocity versus appliedelectric field (E) of (a) a typical smooth PMMA microfluidic (50 lm wide and 85 lm from the literature) lEOF = 2.07 ± 0.07 10�4 cm2/V s, (b) a plasma fabricated PMMAmicrofluidic (180 lm wide and 30 lm deep, rough and hydrophilic), lEOF = 2.83 ± 0.13 10�4 cm2/V s, and (c) a plasma fabricated PMMA microfluidic (180 lm wide and 30 lmdeep) with SH walls on its engraved part (not on the lid), lEOF = 3.89 ± 0.3 10�4 cm2/V s. Reprinted with permission from Ref. [112] Copyright 2008 Elsevier.


Fig. 7. (i) (left) A drop evaporating on a superhydrophobic surface, (right) The cartoon illustrates chemical analysis on a superhydrophobic surface. (A) The solution ispositioned upon the superhydrophobic surface. (B) The device is constituted by micro-pillars. (C) Each micropillar is covered by a layer of nanoporous silicon. (D) Due tosuperhydrophobicity, the substrate concentrates the solute into a small area. (E) The nanoporous silicon film would capture (and thus select) the molecules strictly smallerthan the pores size; spectroscopy techniques would consequently measure these molecules with good resolution. Reprinted with permission from Ref. [134] Copyright 2011Elsevier. (ii) Demonstration of electrically induced reversible transitions between different wetting states of a liquid on a nanostructured substrate. To induce a transitionbetween a rolling ball and an immobile droplet, a voltage was applied between the droplet (contacted through a Pt wire) and the substrate. To reverse the transition andconvert the immobile droplet back to the rolling ball state, a short pulse of electrical current was transmitted through the highly conductive 7.4 lm surface layer of thesubstrate. (a) With no voltage applied, a water droplet formed a highly mobile rolling ball on the nanograss substrate. (b) With the application of about 35 V, the water dropleton the nanograss substrate underwent a sharp transition to the immobile droplet state. (c) After a short pulse of electrical current was transmitted through the nanograsssubstrate, the droplet returned to the original rolling ball state. Reprinted with permission from Ref. [142] Copyright 2007 American Chemical Society. (iii) Drop trapping atelectrically tunable wetting defect. (a) Snapshots of sliding drop (volume: 60 ml; inclination angle: 4.3 a) Top: applied voltage 200 V smaller than the critical trapping voltage.Bottom: 400 V, larger than critical. (b) Schematic view of the setup illustrating inclined plane and the electrodes forming the electrical trap along with equivalent electricalcircuit diagram. (c) Schematic view of the potential energy landscape versus drop position for zero voltage (red line) and a finite voltage U (black), reprinted with permissionfrom Ref. [143] Copyright Nature publishing group. (iv) (a) Scheme of SH surfaces patterned with micro-indentations able to suspend arrays of droplets containing cells; upon24 h, spheroids are formed and drug-screening tests may be performed on the individual droplets. (b) Fluorescent images of L929 spheroids obtained from a confocalmicroscope 24 h after the addition of various concentrations of doxorubicin. (c) Percentage of live (green)/dead (red) cells per spot in the different conditions shown in (b).Reprinted with permission from Ref. [144] Copyright 2014 John Wiley and Sons. (For interpretation of the references to color in this figure legend, the reader is referred to theweb version of this article.)


Vlachopoulou et al. observed that SF6 plasma nanotextured PDMSshowed a 5� increase in protein adsorption compared to flatsurfaces [156,157], an optimal etching time being 6 min. They

proposed the use of such nanotextured substrates for proteinmicroarrays [42]. Rucker et al. also observed increased biomoleculeadsorption on oxygen plasma nanotextured organic polymers


[158]. Tsougeni et al. observed a similar trend in Oxygen Plasmananotextured PMMA [159], with optimal etching time being5 min. They also proposed the use of nanotextured PMMA forprotein arrays [160], and for DNA arrays [161] (see Fig. 8(i)). Ashas been demonstrated by Vlachopoulou et al. and Tsougeniet al. above, such microarrays offer an increased sensitivity by afactor of 100�, due to stronger signal (3–10�) and better spot con-finement and definition. It is worth mentioning at this point thatalthough these nanostructured surfaces are hydrophilic (not super-hydrophilic) microarray spots remain well defined on them, andthere is no need to surround them with a superhydrophobic region,although this is of course easy to achieve by simply using a stencilmask. This lack of a need for superhydrophobic confinementreduces the processing steps. Ren et al. nanotextured PDMS withpreformed microlens structures in CF4 and CF4/O2 plasmas andnoticed that while the first plasma produced superhydrophobicsurfaces, the second plasma produced hydrophilic nanotexturedareas on which DNA could be enriched [162] (see Fig. 8(ii)).

The concept of increased surface-area and controlled surfacechemistry is not only applicable for polymers, but also for Si, Auand possibly other substrates and can be also used on these sur-faces for increased biomolecule adsorption. For example Evans-Nguyen et al. [163] created a porous nanostructured Au surface.The surface after methylation (CH3) became superhydrophobic,while islands with carboxylic (COOH) termination becamehydrophilic and protein anchoring points. Thus, protein bindingspots with confined areas of diameter of 500 lm were formed(see Fig. 8(iii)).

This work by Evans-Nguyen et al. [163] also demonstrated thatsuperhydrophobic surfaces do not allow biomolecule adsorptionand binding. This was also observed by Neto et al. [164] whofabricated hydrophilic spots surrounded by superhydrophobicareas and proposed these spots as a platform for high throughputanalysis of the interactions of biomaterials, proteins and cells [164].

Koc et al. [165] fabricated superhydrophobic surfaces withmicro or nanoscale topography and hydrocarbon or fluorocarbonchemical termination. They observed, that when surfaces weresubjected to flow most of the protein adsorbed was removed fromthe surfaces, which may be a useful concept for antifouling micro-fluidics. This was further proven by Tsougeni et al. who fabricatedpolymeric microchannels comprising nanostructured hydrophilicand superhydrophobic areas created via stencil mask depositionof a fluorocarbon coating [166] (see Fig. 8(iv)). A fluorescent pro-tein solution was passed through the microchannels, which weresubsequently rinsed. It was observed that no protein bindingoccurred on the superhydrophobic part, allowing on-off proteinbinding on these surfaces, and confirming the antifouling characterof superhydrophobic surfaces. The concept of using superhydro-phobic surfaces for antifouling control and self-cleaning upon rins-ing was also proposed by Shirtcliffe et al. [167]. An advantage oftheir method is that they can make all the walls of the microchan-nel superhydrophobic. The topic of biomolecule adsorption onsuperhydrophilic and superhydrophobic surfaces has been recentlyreviewed by Song and Mano [168].

A different strategy for selective biomolecule binding involvedthe use of a lithographic pattern on a substrate (e.g. Photoresistpattern on glass), or in general a pattern of chemical contrast ofone material on a different substrate (e.g. SiO2 islands on Si) (seeFig. 8(v)). The pattern may be defined by lithography, or lithogra-phy and etching, or colloidal lithography. A plasma etching stepsubsequently modifies one of the materials by nanorougheningand appropriate chemical modification and allows selective pro-tein binding on one of the two materials. This way Bayati et al.[169] allowed selective protein adsorption on SiO2 over Si, creatingself-aligned’’ biomolecule spots on the patterns formed, while Mal-ainou et al. recently extended this concept to nanoscale patterns

using colloidal lithography [78,170]. The same authors also trans-ferred this concept to glass, showing selective protein adsorptionon rough photoresist patterns on glass. Using small (a few microns)photoresist pattern arrays, rather than a large photoresist spot theygreatly increased the uniformity of fluorescence, and the spothomogeneity in a microarray (see Fig. 8(v)).

Finally, nanostructured surfaces can be useful for gas or liquidsensors. Andreeva et al [171] used a superhydrophobic surface filmfor formaldehyde and toluene sensing. They observed increasedsensitivity, due to larger surface area and low water adsorption.Valsesia et al. [172] created islands of poly(acrylic acid) for biosen-sing applications, using colloidal lithography and plasma etching ofa poly(acrylic acid) film on which polystyrene microspheres weredeposited.

3.5. Cells on nanostructured surfaces and antifouling

Wettability does not only affect biomolecule adsorption andbinding, but also cell adhesion. Cell distribution across microchan-nels depended on their wetting properties due to microflow inertia[174]. Two reviews on the role of rough superhydrophilic andsuperhydrophobic surfaces on cell adhesion were recentlyauthored by Song and Mano [168] and Oliveira et al. [175]. The roleof plasma processing for nanobiotechnology applications has beenreviewed by Rossi et al. [176] and Colpo et al. [177]. It wasobserved by several authors that superhydrophobic surfaces donot favor the attachment of cells, as opposed to superhydrophilicsurfaces, allowing the creation of cell patterns [178] (seeFig. 9(i)). However, the same authors have shown that after 72 hof culture, cells start also to attach to the superhydrophobic areas(see Fig. 9(i)). Song et al. studied the effect of roughness and Argonplasma treatment on smooth and rough poly(L-lactic acid) surfaces.Untreated surfaces were hydrophobic (smooth) or superhydropho-bic (rough) (see Fig. 9(ii)). Treated surfaces were hydrophilic tosuperhydrophilic. They showed that plasma treatment increasedcell adhesion both on smooth and rough surfaces, while no adhe-sion was present on the superhydrophobic surfaces [179]. Oliveiraet al. produced superhydrophilic islands in superhydrophobicpolystyrene, and showed strong cell adhesion selectively on thehydrophilic regions [180]. However after 6 days culture cells,started to grow also on the superhydrophobic areas (seeFig. 9(iii)). Similar results were observed by Yang et al. with bloodsamples on TiO2 nanorods. In superhydrophobic areas no bloodcells adhered as opposed to hydrophilic and superhydrophilic areasleading the authors to propose such surfaces for implants [181]. Inanother work by Lai et al. TiO2 nanotube arrays were reversiblypatterned with superhydrophobic and superhydrophilic areas ona microfluidic chip by combinations of SAMs and photocatalyticlithography. Several applications were envisioned such as 2D pat-terned cell scaffolds, biomolecule adsorption, 3D surface patternsfor sensing or antifouling action [182] (see Fig. 9(iv)).

Similar on-off behavior, as well as increased adhesion on super-hydrophilic areas was also observed recently by Tsougeni et al.[183] in a plasma nanotextured PMMA microfluidic device withsuperhydrophilic and superhydrophobic areas using a cancer cellline adhering selectively on the hydrophilic areas (see Fig. 9(v)).The reduced adhesion of cells on superhydrophobic surfaces canbe useful for reducing bacteria attachment on such surfaces as pro-posed by Poncin-Epaillard et al. [184]. Reduced adhesion of mouseosteoblastic cells (MC3T3-E1) was also observed on superhydro-phobic Polystyrene [185].

Despite these general trends, there are some works which showincreased adhesion under some conditions (e.g. static non-flowconditions) on superhydrophobic surfaces. Di Mundo et al. haveshown that SaOs2 cells are inhibited on nanotextured hydrophobicsurfaces, while they seem to adhere on slippery superhydrophobic

Fig. 8. (i) Oxygen Plasma Nanostructured PMMA protein array. An array of two proteins is shown (BSA, RgG). Reprinted with permission from Ref. [160] Copyright 2010American Chemical Society. (ii) Fluorescent images. Labeled ssDNA adsorbed on PDMS regular structure surface treated via CF4/O2 plasma. Black area shows that nomolecules are adsorbed. Reprinted with permission from Ref. [162] Copyright 2009 The Royal Society of Chemistry. (iii) B) Schematic illustration and SEM images ofevaporated gold films with porous gold deposited on the surface at (B1) 350�, (B2) 1000�, and (B3) 10000�magnification and (C) after patterning with mercaptoundecanoicacid and dodecanethiol SAMs. The white bar in each image represents 10 lm. (D) A schematic illustrating exposure of protein-functionalized porous gold patterns to acomplex mixture, where species from the solution bind on COOH terminated areas of the nanostructured gold surface. Reprinted with permission from Ref. [163] Copyright2008 American Chemical Society. (iv) Fluorescence image of a microchannel with patterned wettability (from superhydrophobic, contact angle CA �150 to superhydrophilic,CA < 10) formed in PMMA by O2 plasma etching, deposition of a hydrophobic Teflon-like film in the left zone, and adsorption of AF488 labeled goat anti-rabbit IgG antibody.Notice the abrupt increase in fluorescence from left to right, and the large intensity in the superhydrophilic rough areas. Fluorescence intensity of b-BSA spots on smooth(dotted lines) and rough 20-min etched (solid lines) PMMA microchannel walls as a function of concentration of AF546-labeled streptavidin. Detection limit on smooth wallsis 720 ng/ml, while detection limit on rough walls is 6 ng/ml (i.e. 120� increased). Each point is the mean value of 6 measurements ± SD. Reprinted with permission from Ref.[166] Copyright 2012 Elsevier. (v) Process flow for (a) substrate preparation and (b) protein or DNA immobilization. (c) Schematic of creation of dense microarrays onphotolithographically patterned glass substrates. Notice the difference in size between the deposited protein droplets and the lithographically/plasma determined protein/DNA spots (colored circles). (down) Fluorescence images of glass substrates patterned with different photoresists spots after b-BSA immobilization and reaction with AF 546labeled streptavidin, (a)–(c) prior to any treatment, (d)–(f) after 2-min O2 and (g)–(i) after 2-min SF6 plasma treatment, for each resist. In the last column of the table theaverage fluorescence intensity values obtained from patterned photoresist/glass substrates (j) prior to any treatment (k) after 2-min O2 and (l) after 2-min SF6 plasmatreatment, upon adsorption of b-BSA and detection with AF 546-labeled streptavidin are provided. Each value is the mean of 25 measurements ± SD. Reprinted withpermission from Ref. [173] Copyright 2012 Elsevier.


surfaces with larger nanofeatures [186]. Similarly Cha et al. [187]have shown that stems cells adhere on lotus leave micro–nanotextured polystyrene. They predict that such surfaces could be

potentially used for an efficient increase of adipogenic differentia-tion of stem cells, which is important in the cosmetic and aestheticindustry [187]. Recently, these efforts have been extended also to

Fig. 8 (continued)


3D. A patterned superhydrophobic platform was proposed as a 3-dimensional miniaturized porous scaffold by Oliveira et al. [188].

The above controversies necessitate further studies for potentialuse of superhydrophobic surfaces in medical devices, a field of vastsocial and economical importance. Indeed, in vivo medical devices(i.e. vascular grafts, hemo-dialysis membranes, catheters, heartvalves, intravascular stents) which can work in contact with bloodare very essential for the treatment of several disorders. However,clogging that can result to thrombosis is a common problem thatsuch devices present due to poor biocompatibility of the materialsused in their fabrication [189]. Some recent studies suggest thatthese problems can be solved by surface passivation or incorpora-tion of anti-thrombosis and other anti-sticking agents [189]. Analternative scenario is to incorporate superhydrophobic surfacesin such devices in order to eliminate such clogging problems.A recent review presents the progress in the surface modification

of polymers that can be used in blood contact applications byblending the base polymer with surface additives [190]. Addition-ally, Zhu et al. reviewed the most common techniques used tocontrol the chemical composition and the surface geometric struc-tures in order to tune the adhesion behavior of superhydrophobicsurfaces [191]. This is an area where certainly more research isneeded.

4. Perspectives and challenges

There is a great potential in using both hydrophilic andsuperhydrophobic nanotextured surfaces in microdevices andMicrosystems/Labs on Chip. New functionalities and smart combi-nations are possible. The potential for innovation seems unlimited,and the range of chemical, biochemical, biological, environmental,and other applications is huge. It is the authors’ opinion that

Fig. 9. (i) Phase-contrast microscopic image of micropatterning of NIH 3T3 fibroblast cells cultured on the micropatterned superhydrophobic/superhydrophilic surface for 24 h(left lines, middle circles). On the right, numbers of cells cultured on superhydrophobic and superhydrophilic surfaces for 24 and 72 h are shown. Reprinted with permissionfrom Ref. [168] Copyright 2010 American Chemical Society. (ii) Fluorescent microscopy images of smooth and rough poly(L-lactic acid) surfaces treated with argon plasma oruntreated (blue DAPI staining for the nuclei of cells). (A) Smooth PLLA, untreated. (B) Rough PLLA, untreated (superhydrophobic). (C) Smooth surface treated with argon plasmafor 50 s (hydrophilic). (D) Rough surface treated with argon plasma for50 s (superhydrophilic). Insets: SEM images of cells attached to the rough surfaces. Inset scale bar: 20 lm.Reprinted with permission from Ref. [179] Copyright 2009 John Wiley and Sons. (iii) Schematic representation of the patterning of superhydrophilic regions onsuperhydrophobic PS surfaces by UV-Ozone irradiation and using a hollowed mask; the two cell seedings were made by immersion and in open-air. Fluorescent staining of theSaOs2 cell nucleus with DAPI: (A) on samples where cells were seeded over the whole surface, after 6 days in culture; (B) and in open-air culture where 7 mL of the cellularsuspension was dropped on the superhydrophilic region, after 2 days in culture. Reprinted with permission from Ref. [180] Copyright 2011 The Royal Society of Chemistry. (iv)SEM images of 3T3 cell adhesion on the superhydrophilic–superhydrophobic TNA surfaces micropatterned with square windows for 24 h. The pattern distances betweensuperhydrophilic regions were 45 lm in (c). (d) Magnified image of cell on the superhydrophilic area. The inset of (c) shows the schematic illustration of the correspondingwetting patterns. The inset of (d) shows the schematic diagram of the site-selective cell layer on 2D flat TNA scaffold. Reprinted with permission from Ref. [182] Copyright 2013John Wiley and Sons. (v) Fluorescence image of HT1080 cells cultivated on a microchannel with variable wetting characteristics (from superhydrophilic, contact angle CA < 10�,to superhydrophobic, CA > 150�) formed in PMMA by O2 plasma etching and deposition of a hydrophobic Teflon-like film in the middle zone. The cells were stained withphalloidin-Atto 488 to visualize the cytoskeleton and DAPI for staining the nucleus. Reprinted with permission from Ref. [183] Copyright 2014 Elsevier.


especially plasma nanotextured surfaces have a bright future ascost-effective, high quality microarray, cell array substrates, andcell culture or antifouling substrates. In addition, authors believethat plasma nanotexturing is ideal for embedding functionalitiesin microfluidics. These technologies are now mature enough forlarger scale use.

However, there are still many technological problems to beaddressed. Concerning plasma nanotexturing, plasma reactorsneed to be commercialized, which produce fast, uniform, andreproducible nanotexturing over large areas. The existing plasmareactors are mainly designed for smooth etching processes, andnanotexturing is a side-effect of electrode material sputtering.Probably, high density reactors are those which produce fast andcost-effective nanotexturing of surfaces. However, their cost ishigher than simpler low density plasma equipment.

To be accepted by biologists, chemists and professionals inthe Life Science area, plasma nanotextured surfaces will have

to be compared with commercial chemically modified sur-faces used routinely as substrates for microarrays. This willnecessitate stronger interaction among microtechnology/plasma processing groups and biology/diagnostic/bioanalyticgroups. This is necessary as very few labs can master multipletechnologies.

Concerning microfluidics applications, a major challenge is tomake all walls of a microfluidics possess the same functionality.While a great progress has been made in this area, the proposedsolutions are not attractive for mass production. The ability tobond a rough microfluidic channel and a rough lid with the sametopography and chemical functionality might solve the problem.However, no reliable bonding technologies exist, which wouldmake this feasible. In addition, bonding technologies for sealingmicrofluidics need to be further developed, to avoid destroyingchemical functionality or nanostructures. Beyond lamination,bonding technologies are today a problem for microfluidics.

Fig. 9 (continued)


Finally, for hydrophilic microfluidics, ageing or simple generic andfast methods for reactivation remain a challenge.

Fortunately, therefore, there is plenty of work to do for the next30 years of the field and of the journal.

Acknowledgments

The authors would like to thank all graduate students, post-doctoral fellows and collaborators inside NCSR Demokritos whohave contributed in the work done by our group and team, as wellas the funding of several EU and National projects. This work was

supported by: (a) the Research Excellence Project 695 ‘‘PlasmaDirected Assembly and Organization: PlasmaNanoFactory’’ which isimplemented under the ‘‘ARISTEIA I’’ Action of the ‘‘OPERATIONALPROGRAMME EDUCATION AND LIFELONG LEARNING’’ co-fundedby the European Social Fund (ESF) and National Resources,and (b) the Project ‘‘THALIS-DESIgn and fabrication of RobustsupErhyDROPhobic/philic surfaces and their application in therealization of ‘‘smart’’ microfluidic valves’’, co-funded by Hellenicand European Regional Development Funds (ERDF) under theHellenic National Strategic Reference Framework (NSRF)2007–2013.


References

[1] T.-S. Wong, T. Sun, L. Feng, J. Aizenberg, MRS Bull. 38 (2013) 366–371.[2] A.K. Kota, W. Choi, A. Tuteja, MRS Bull. 38 (2013) 383–390.[3] T.-S. Wong, S.H. Kang, S.K.Y. Tang, E.J. Smythe, B.D. Hatton, A. Grinthal, J.

Aizenberg, Nature 477 (2011) 443–447.[4] N.J. Shirtcliffe, G. McHale, M.I. Newton, J. Polym. Sci., Part B: Polym. Phys. 49

(2011) 1203–1217.[5] E. Celia, T. Darmanin, E. Taffin de Givenchy, S. Amigoni, F. Guittard, J. Colloid

Interface Sci. 402 (2013) 1–18.[6] Y.Y. Yan, N. Gao, W. Barthlott, Adv. Colloid Interface Sci. 169 (2011) 80–105.[7] K. Liu, L. Jiang, Nano Today 6 (2011) 155–175.[8] L. Yao, J. He, Prog. Mater Sci. 61 (2014) 94–143.[9] A. Grinthal, J. Aizenberg, Chem. Mater. 26 (2014) 698–708.

[10] K. Liu, Y. Tian, L. Jiang, Prog. Mater Sci. 58 (2013) 503–564.[11] Z. Xue, M. Liu, L. Jiang, J. Polym. Sci., Part B: Polym. Phys. 50 (2012) 1209–

1224.[12] N. Valipour M, F.C. Birjandi, J. Sargolzaei, Colloids Surf., A 448 (2014)

93–106.[13] H. Bellanger, T. Darmanin, E. Taffin de Givenchy, F. Guittard, Chem. Rev. 114

(2014) 2694–2716.[14] W.-G. Bae, H.N. Kim, D. Kim, S.-H. Park, H.E. Jeong, K.-Y. Suh, Adv. Mater. 26

(2014) 675–700.[15] Z. Chu, S. Seeger, Chem. Soc. Rev. 43 (2014) 2784.[16] N.E. Vourdas, M.E. Vlachopoulou, A. Tserepi, E. Gogolides, Int. J. Nanotechnol.

6 (2009) 196–207.[17] E. Gogolides, M. Vlachopoulou, K. Tsougeni, N. Vourdas, A. Tserepi, Int. J.

Nanomanuf. 6 (2010) 152–163.[18] O. Kylián, A. Choukourov, H. Biederman, Thin Solid Films 548 (2013)

1–17.[19] R. Jafari, S. Asadollahi, M. Farzaneh, Plasma Chem. Plasma Process. 33 (2012)

177–200.[20] M. Morra, E. Occhiello, F. Garbassi, Langmuir 5 (1989) 872–876.[21] K. Ogawa, M. Soga, Y. Takada, I. Nakayama, Jap. J. Appl. Phys. 32 (1993).[22] J.P. Youngblood, T.J. McCarthy, Macromolecules 32 (1999) 6800–6806.[23] B. Washo, Organic Coatings Applied Polymer Science Proceedings, 47 (1982)

69–72.[24] Y. Matsumoto, M. Ishida, Sens. Actuators, B 83 (2000) 179–185.[25] A. Nakajima, K. Hashimoto, T. Watanabe, Monatshefte fur Chemie 132 (2001)

31–41.[26] K. Teshima, H. Sugimura, Y. Inoue, O. Takai, A. Takano, Appl. Surf. Sci. 244

(2005) 619–622.[27] J. Fresnais, J.P. Chapel, F. Poncin-Epaillard, Surf. Coat. Technol. 200 (2006)

5296–5305.[28] A. Tserepi, E. Gogolides, K. Misiakos, M.-E. Vlachopoulou, N. Vourdas, Greek

Patent no 1006890, publication WO2007031799 (2005).[29] P. Bayiati, A. Tserepi, E. Gogolides, K. Misiakos, J. Vac. Sci. Technol., A 22

(2004) 1546–1551.[30] A.D. Tserepi, M.E. Vlachopoulou, E. Gogolides, Nanotechnology 17 (2006)

3977–3983.[31] N. Vourdas, A. Tserepi, E. Gogolides, Nanotechnology 18 (2007) 125304.[32] D. Kontziampasis, G. Boulousis, A. Smyrnakis, K. Ellinas, A. Tserepi, E.

Gogolides, Microelectron. Eng. 121 (2014) 33–38.[33] E. Wohlfart, J.P. Fernández-Blázquez, E. Knoche, A. Bello, E. Pérez, E. Arzt, A.n.

del Campo, Macromolecules 43 (2010) 9908–9917.[34] E. Wohlfart, J.P. Fernández-Blázquez, E. Arzt, A. del Campo, Plasma Processes

Polym. 8 (2011) 876–884.[35] B. Balu, V. Breedveld, D.W. Hess, Langmuir 24 (2008) 4785–4790.[36] J.P. Fernández-Blázquez, D. Fell, E. Bonaccurso, A.d. Campo, J. Colloid Interface

Sci. 357 (2011) 234–238.[37] F. Palumbo, R. Di Mundo, D. Cappelluti, R. d’Agostino, Plasma Processes

Polym. 8 (2) (2011) 118–126.[38] R. Di Mundo, M. Troia, F. Palumbo, M. Trotta, R. d’Agostino, Plasma Processes

Polym. 9 (10) (2012) 947–954.[39] N.D. Vietro, L. Belforte, V.G. Lambertini, F. Fracassi, Appl. Surf. Sci. 307 (2014)

698–703.[40] J. Tarrade, T. Darmanin, E. Taffin de Givenchy, F. Guittard, D. Debarnot, F.

Poncin-Epaillard, Appl. Surf. Sci. 292 (2014) 782–789.[41] A. Skarmoutsou, C.A. Charitidis, A.K. Gnanappa, A. Tserepi, E. Gogolides,

Nanotechnology 23 (2012) 505711.[42] M.-E. Vlachopoulou, A. Tserepi, P.S. Petrou, E. Gogolides, S.E. Kakabakos,

Colloids Surf. B 83 (2011) 270–276.[43] A.K. Gnanappa, D.P. Papageorgiou, E. Gogolides, A. Tserepi, A.G.

Papathanasiou, A.G. Boudouvis, Plasma Process Polym 9 (2012) 304–315.[44] K. Tsougeni, N. Vourdas, A. Tserepi, E. Gogolides, C. Cardinaud, Langmuir 25

(2009) 11748–11759.[45] M.E. Vlachopoulou, G. Kokkoris, C. Cardinaud, E. Gogolides, A. Tserepi, Plasma

Processes Polym. 10 (2013) 29–40.[46] M.E. Vlachopoulou, A. Tserepi, K. Beltsios, G. Boulousis, E. Gogolides,

Microelectron. Eng. 84 (2007) 1476–1479.[47] E. Gogolides, V. Constantoudis, G. Kokkoris, D. Kontziampasis, K. Tsougeni, G.

Boulousis, M.-E. Vlachopoulou, A. Tserepi, J. Phys. D Appl. Phys. 44 (2011)174021.

[48] G. Kokkoris, V. Constantoudis, P. Angelikopoulos, G. Boulousis, E. Gogolides,Phys. Rev. B 76 (2007) 193405.

[49] N. Atthi, J. Supadech, G. Dupuy, O.U. Nimittrakoolchai, A. Pankiew, S.Supothina, W. Jeamsaksiri, C. Hruanun, A. Poyai, C.H. Brachais, J. Nanosci.Nanotechnol. 12 (2012) 4919–4927.

[50] A.Y. Vorobyev, C. Guo, Laser Photonics Rev. 7 (2013) 385–407.[51] R.T.R. Kumar, K.B. Mogensen, P. Bøggild, J. Phys. Chem. C 114 (2010) 2936–2940.[52] S.E. Lee, D. Lee, P. Lee, S.H. Ko, S.S. Lee, S.U. Hong, Macromol. Mater. Eng. 298

(2013) 311–317.[53] V. Jokinen, P. Suvanto, A.R. Garapaty, J. Lyytinen, J. Koskinen, S. Franssila,

Colloids Surf. A 434 (2013) 207–212.[54] P. Favia, G. Cicala, A. Milella, F. Palumbo, R. Rossini, R. d’Agostino, Surf. Coat.

Technol. 169 (2003) 609–612.[55] A. Milella, F. Palumbo, P. Favia, G. Cicala, R. d’Agostino, Pure Appl. Chem. 77

(2005) 399–414.[56] A. Milella, R. Di Mundo, F. Palumbo, P. Favia, F. Fracassi, R. d’Agostino, Plasma

Processes Polym. 6 (2009) 460–466.[57] G. Cicala, Surf. Eng. 28 (2012) 141–148.[58] S. Tsoi, E. Fok, J.C. Sit, J.G.C. Veinot, Langmuir 20 (2004) 10771–10774.[59] S.H. Lee, Z.R. Dilworth, E. Hsiao, A.L. Barnette, M. Marino, J.H. Kim, J.G. Kang,

T.H. Jung, S.H. Kim, ACS Appl. Mater. Interfaces 3 (2011) 476–481.[60] D. Wang, Q. Yang, Y. Guo, X. Liu, J. Shi, J. Zhang, Mater. Lett. 65 (2011) 2526–2529.[61] H. Zhao, K.-C. Park, K.-Y. Law, Langmuir 28 (2012) 14925–14934.[62] K. Ellinas, S.P. Pujari, D.A. Dragatogiannis, C.A. Charitidis, A. Tserepi, H.

Zuilhof, E. Gogolides, ACS Appl. Mater. Interfaces 6 (2014) 6510–6524.[63] J. Marquez-Velasco, M.E. Vlachopoulou, A. Tserepi, E. Gogolides,

Microelectron. Eng. 87 (2010) 782–785.[64] E. Bormashenko, R. Grynyov, G. Chaniel, H. Taitelbaum, Y. Bormashenko, Appl.

Surf. Sci. 270 (2013) 98–103.[65] B. Cortese, S. D’Amone, M. Manca, I. Viola, R. Cingolani, G. Gigli, Langmuir 24

(2008) 2712–2718.[66] Y. Kwon, N. Patankar, J. Choi, J. Lee, Langmuir 25 (2009) 6129–6136.[67] T.P.N. Nguyen, R. Boukherroub, V. Thomy, Y. Coffinier, J. Colloid Interface Sci.

416 (2014) 280–288.[68] C.W. Extrand, Langmuir 18 (2002) 7991–7999.[69] C.W. Extrand, Langmuir 20 (2004) 5013–5018.[70] A. Tuteja, W. Choi, M. Ma, J.M. Mabry, S.A. Mazzella, G.C. Rutledge, G.H.

McKinley, R.E. Cohen, Science 318 (2007) 1618–1622.[71] A. Tuteja, W. Choi, J.M. Mabry, G.H. McKinley, R.E. Cohen, Proc. Natl. Acad. Sci.

105 (2008) 18200–18205.[72] A. Ahuja, J.A. Taylor, V. Lifton, A.A. Sidorenko, T.R. Salamon, E.J. Lobaton, P.

Kolodner, T.N. Krupenkin, Langmuir 24 (2008) 9–14.[73] H. Zhao, K.-Y. Law, V. Sambhy, Langmuir 27 (2011) 5927–5935.[74] A. Susarrey-Arce, Á.G. Marín, H. Nair, L. Lefferts, J.G.E. Gardeniers, D. Lohse, A.

van Houselt, Soft Matter 8 (2012) 9765.[75] A. Zeniou, K. Ellinas, A. Olziersky, E. Gogolides, Nanotechnology 25

(2014).[76] J. Zhang, Y. Li, X. Zhang, B. Yang, Adv. Mater. 22 (2010) 4249–4269.[77] J.Y. Shiu, C.W. Kuo, P.L. Chen, C.Y. Mou, Chem. Mater. 16 (2004) 561–564.[78] K. Ellinas, A. Smyrnakis, A. Malainou, A. Tserepi, E. Gogolides, Microelectron.

Eng. 88 (2011) 2547–2551.[79] K. Ellinas, A. Tserepi, E. Gogolides, Langmuir 27 (2011) 3960–3969.[80] X.M. Zhang, J.H. Zhang, Z.Y. Ren, X. Li, X. Zhang, D.F. Zhu, T.Q. Wang, T. Tian, B.

Yang, Langmuir 25 (2009) 7375–7382.[81] Y. Hirai, H. Yabu, Y. Matsuo, K. Ijiro, M. Shimomura, J. Mater. Chem. 20 (2010)

10804–10808.[82] K. Tsougeni, G. Boulousis, E. Gogolides, A. Tserepi, Microelectron. Eng. 85

(2008) 1233–1236.[83] T.-Y. Chung, D.B. Graves, F. Weilnboeck, R.L. Bruce, G.S. Oehrlein, M. Li, E.A.

Hudson, Plasma Processes Polym. 8 (2011) 1068–1079.[84] N. Vourdas, D. Kontziampasis, G. Kokkoris, V. Constantoudis, A. Goodyear, A.

Tserepi, M. Cooke, E. Gogolides, Nanotechnology 21 (2010) 085302.[85] D. Eon, V. Raballand, G. Cartry, C. Cardinaud, N. Vourdas, P. Argitis, E.

Gogolides, J. Vac. Sci. Technol., B 24 (2006) 2678–2688.[86] A. Tserepi, E. Gogolides, K. Tsougeni, V. Constantoudis, E.S. Valamontes, J.

Appl. Phys. 98 (2005) 113502-1–113502-6.[87] K. Tsougeni, A. Tserepi, G. Boulousis, V. Constantoudis, E. Gogolides, Plasma

Processes Polym. 4 (2007) 398–405.[88] M. Sumiya, R. Bruce, S. Engelmann, F. Weilnboeck, G.S. Oehrlein, Journal of

Vacuum Science & Technology B: Microelectronics and Nanometer Structures26 (2008) 1637.

[89] M. Sumiya, R. Bruce, S. Engelmann, F. Weilnboeck, G.S. Oehrlein, Journal ofVacuum Science & Technology B: Microelectronics and Nanometer Structures26 (2008) 1647.

[90] M. Sumiya, R. Bruce, S. Engelmann, F. Weilnboeck, G.S. Oehrlein, Journal ofVacuum Science & Technology B: Microelectronics and Nanometer Structures26 (2008) 1978.

[91] S. Xu, I. Levchenko, S.Y. Huang, K. Ostrikov, Appl. Phys. Lett. 95 (2009).[92] E. Gogolides, A. Tserepi, V. Constantoudis, N. Vourdas, G. Boulousis, M.E.

Vlachopoulou, K. Tsougeni, D. Kontziampasis, Method for the fabrication ofperiodic structures on polymers using plasma processes, Greek patentnumber 1006618, Publication number: WO/2009/150479 (2008).

[93] D. Kontziampasis, V. Constantoudis, E. Gogolides, Plasma Processes Polym. 9(2012) 866–872.

[94] E. Ueda, P.A. Levkin, Adv. Mater. 25 (2013) 1234–1247.[95] E.K. Sackmann, A.L. Fulton, D.J. Beebe, Nature 507 (2014) 181–189.[96] K.W. Oh, C.H. Ahn, J. Micromech. Microeng. 16 (2006) R13–R39.

http://refhub.elsevier.com/S0167-9317(14)00434-1/h0005

































































































































































[97] M. Zimmermann, P. Hunziker, E. Delamarche, Microfluid. Nanofluid. 5 (2008)395–402.

[98] G.V. Kaigala, R.D. Lovchik, E. Delamarche, Angew. Chem. Int. Ed. 51 (2012)11224–11240.

[99] M. Hitzbleck, L. Gervais, E. Delamarche, Lab Chip 11 (2011) 2680–2685.[100] M. Hitzbleck, L. Avrain, V. Smekens, R.D. Lovchik, P. Mertens, E. Delamarche,

Lab Chip 12 (2012) 1972–1978.[101] L. Gervais, M. Hitzbleck, E. Delamarche, Biosens. Bioelectron. 27 (2011) 64–70.[102] L. Gervais, E. Delamarche, Lab Chip 9 (2009) 3330–3337.[103] L. Gervais, N. De Rooij, E. Delamarche, Adv. Mater. 23 (2011) H151–H176.[104] P. Zhang, Y. Deng, Y. Liu, Y. Wu, Microsyst. Technol. 17 (2011) 1491–1495.[105] K. Tsougeni, D. Papageorgiou, A. Tserepi, E. Gogolides, Lab Chip 10 (2010)

462–469.[106] K. Handique, D.T. Burke, C.H. Mastrangelo, M.A. Burns, Anal. Chem. 72 (2000)

4100–4109.[107] H. Andersson, W. van der Wijngaart, P. Griss, F. Niklaus, G. Stemme, Sens.

Actuators, B 75 (2001) 136–141.[108] J.W. Suk, J.-H. Cho, J. Micromech. Microeng. 17 (2007) N11–N15.[109] G. Londe, A. Chunder, A. Wesser, L. Zhai, H. Cho, Sens. Actuators, B 132 (2008)

431–438.[110] A.P. Washe, P. Lozano-Sanchez, D. Bejarano-Nosas, B. Teixeira-Dias, I. Katakis,

Microelectron. Eng. 111 (2013) 416–420.[111] D.P. Papageorgiou, K. Tsougeni, A. Tserepi, E. Gogolides, Microfluid.

Nanofluid. 14 (2013) 247–255.[112] N. Vourdas, A. Tserepi, A.G. Boudouvis, E. Gogolides, Microelectron. Eng. 85

(2008) 1124–1127.[113] H. He, Y. Yuan, W. Wang, N.-R. Chiou, A.J. Epstein, L.J. Lee, Biomicrofluidics 3

(2009) 022401.[114] N.M. Oliveira, A.I. Neto, W. Song, J.F. Mano, Appl. Phys. Express 3 (2010)

085205.[115] C.G. Obeso, M.P. Sousa, W. Song, M.A. Rodriguez-Pérez, B. Bhushan, J.F. Mano,

Colloids Surf. A 416 (2013) 51–55.[116] A. Ghosh, R. Ganguly, T.M. Schutzius, C.M. Megaridis, Lab Chip 14 (2014)

1538.[117] L. Li, V. Breedveld, D.W. Hess, Colloid Polym. Sci. 291 (2013) 417–426.[118] R. Almeida, J.W. Kwon, Langmuir 29 (2013) 994–998.[119] V.A. Lifton, S. Simon, J. Porous Mater. 18 (2011) 535–544.[120] A.K. Kota, G. Kwon, W. Choi, J.M. Mabry, A. Tuteja, Nat. Commun. 3 (2012)

1025.[121] L. Zhang, Y. Zhong, D. Cha, P. Wang, Sci. Rep. 3 (2013).[122] K. Ellinas, A. Tserepi, E. Gogolides, Microfluid. Nanofluid. (2014) 1–10.[123] C.-H. Choi, C.-J. Kim, Phys. Rev. Lett. 96 (2006) 066001.[124] C. Journet, S. Moulinet, C. Ybert, S.T. Purcell, L. Bocquet, Europhys. Lett. 71

(2005) 104–109.[125] P. Joseph, C. Cottin-Bizonne, J.M. Benoît, C. Ybert, C. Journet, P. Tabeling, L.

Bocquet, Phys. Rev. Lett. 97 (2006) 156104.[126] E. Karatay, A.S. Haase, C.W. Visser, C. Sun, D. Lohse, P.A. Tsai, R.G.H.

Lammertink, Proc. Natl. Acad. Sci. 110 (2013) 8422–8426.[127] C. Lee, C.-J. Kim, Phys. Rev. Lett. 106 (2011).[128] J. Zhou, E.S. Asmolov, F. Schmid, O.I. Vinogradova, J. Chem. Phys. 139 (2013).[129] R. Dey, K. Raj, N. Bhandaru, R. Mukherjee, S. Chakraborty, Soft Matter 10

(2014) 3451.[130] X. Liu, C. Luo, J. Micromech. Microeng. 20 (2010) 025029.[131] A. Chakraborty, M. Xiang, C. Luo, Materials 6 (2013) 3610–3623.[132] A. Accardo, F. Gentile, F. Mecarini, F. De Angelis, M. Burghammer, E. Di

Fabrizio, C. Riekel, Microelectron. Eng. 88 (2011) 1660–1663.[133] F. De Angelis, F. Gentile, F. Mecarini, G. Das, M. Moretti, P. Candeloro, M.L.

Coluccio, G. Cojoc, A. Accardo, C. Liberale, R.P. Zaccaria, G. Perozziello, L.Tirinato, A. Toma, G. Cuda, R. Cingolani, E. Di Fabrizio, Nat. Photonics 5 (2011)682–687.

[134] F. Gentile, M.L. Coluccio, A. Accardo, M. Asande, G. Cojoc, F. Mecarini, G. Das,C. Liberale, F. De Angelis, P. Candeloro, P. Decuzzi, E. Di Fabrizio,Microelectron. Eng. 88 (2011) 1749–1752.

[135] F. Gentile, M.L. Coluccio, A. Accardo, G. Marinaro, E. Rondanina, S. Santoriello,S. Marras, G. Das, L. Tirinato, G. Perozziello, F. De Angelis, C. Dorigoni, P.Candeloro, E. Di Fabrizio, Microelectron. Eng. 97 (2012) 349–352.

[136] F. Gentile, M.L. Coluccio, N. Coppede, F. Mecarini, G. Das, C. Liberale, L.Tirinato, M. Leoncini, G. Perozziello, P. Candeloro, F. De Angelis, E. Di Fabrizio,ACS Appl. Mater. Interfaces 4 (2012) 3213–3224.

[137] F. Gentile, G. Das, M.L. Coluccio, F. Mecarini, A. Accardo, L. Tirinato, R.Tallerico, G. Cojoc, C. Liberale, P. Candeloro, P. Decuzzi, F. De Angelis, E. DiFabrizio, Microelectron. Eng. 87 (2010) 798–801.

[138] G. Simone, G. Perozziello, E. Battista, F. De Angelis, P. Candeloro, F. Gentile, N.Malara, A. Manz, E. Carbone, P. Netti, E. Di Fabrizio, Microelectron. Eng. 98(2012) 668–671.

[139] L. Tirinato, F. Gentile, D. Di Mascolo, M.L. Coluccio, G. Das, C. Liberale, S.A.Pullano, G. Perozziello, M. Francardi, A. Accardo, F. De Angelis, P. Candeloro, E.Di Fabrizio, Microelectron. Eng. 97 (2012) 337–340.

[140] F. Gentile, M.L. Coluccio, E. Rondanina, S. Santoriello, D. Di Mascolo, A.Accardo, M. Francardi, F. De Angelis, P. Candeloro, E. Di Fabrizio,Microelectron. Eng. 111 (2013) 272–276.

[141] A. Ressine, D. Finnskog, G. Marko-Varga, T. Laurell, NanoBiotechnology 4(2008) 18–27.

[142] N.T. Krupenkin, J. Ashley Taylor, N.E. Wang, P. Kolodner, M. Hodes, R.T.Salamon, Langmuir 23 (2007) 9128–9133.

[143] D. ‘T Mannetje, S. Ghosh, R. Lagraauw, S. Otten, A. Pit, C. Berendsen, J. Zeegers,D. van den Ende, F. Mugele, Nat. Commun. 5 (2014), http://dx.doi.org/10.1038/ncomms4559. Article number: 3559, 6 pages.

[144] A.I. Neto, C.R. Correia, C.A. Custódio, J.F. Mano, Adv. Funct. Mater. 24 (32)(2014) 5096–5103.

[145] J. Kim, C.J. Kim, Proceedings of the 15th IEEE Micro Electro MechanicalSystems (MEMS) (2002) 479–482.

[146] K.S. Seo, R. Wi, S.G. Im, D.H. Kim, Polym. Adv. Technol. 24 (2013)1075–1080.

[147] M.C. Draper, C.R. Crick, V. Orlickaite, V.A. Turek, I.P. Parkin, J.B. Edel, Anal.Chem. 85 (2013) 5405–5410.

[148] S. Sharma, M. Srisa-Art, S. Scott, A. Asthana, A. Cass, Methods Mol. Biol. (2013)207–230.

[149] K. Choi, A.H.C. Ng, R. Fobel, A.R. Wheeler, Ann. Rev. Anal. Chem. (2012) 413–440.

[150] P. Bayiati, A. Tserepi, P.S. Petrou, K. Misiakos, S.E. Kakabakos, E. Gogolides, C.Cardinaud, Microelectron. Eng. 84 (2007) 1677–1680.

[151] P. Bayiati, A. Tserepi, P.S. Petrou, S.E. Kakabakos, K. Misiakos, E. Gogolides, J.Appl. Phys. 101 (2007) 103306-1–103306-9.

[152] T.N. Krupenkin, J. Ashley Taylor, T.M. Schneider, S. Yang, Langmuir 20 (2004)3824–3827.

[153] J. Heikenfeld, M. Dhindsa, J. Adhes. Sci. Technol. 22 (2008) 319–334.[154] M.W. Lee, S.S. Latthe, A.L. Yarin, S.S. Yoon, Langmuir 29 (2013) 7758–7767.[155] N. Vourdas, A. Tserepi, V.N. Stathopoulos, Appl. Phys. Lett. 103 (2013).[156] M.E. Vlachopoulou, P.S. Petrou, S.E. Kakabakos, A. Tserepi, E. Gogolides, J. Vac.

Sci. Technol., B 26 (2008) 2543–2548.[157] M.E. Vlachopoulou, P.S. Petrou, S.E. Kakabakos, A. Tserepi, K. Beltsios, E.

Gogolides, Microelectron. Eng. 86 (2009) 1321–1324.[158] V.C. Rucker, K.L. Havenstrite, B.A. Simmons, S.M. Sickafoose, A.E. Herr, R.

Shediac, Langmuir 21 (2005) 7621–7625.[159] K. Tsougeni, P.S. Petrou, A. Tserepi, S.E. Kakabakos, E. Gogolides,

Microelectron. Eng. 86 (2009) 1424–1427.[160] K. Tsougeni, A. Tserepi, V. Constantoudis, E. Gogolides, P.S. Petrou, S.E.

Kakabakos, Langmuir 26 (2010) 13883–13891.[161] K. Tsougeni, G. Koukouvinos, P.S. Petrou, A. Tserepi, S.E. Kakabakos, E.

Gogolides, Anal. Bioanal. Chem. 403 (2012) 2757–2764.[162] H.-X. Ren, X. Chen, X.-J. Huang, M. Im, D.-H. Kim, J.-H. Lee, J.-B. Yoon, N. Gu, J.-

H. Liu, Y.-K. Choi, Lab Chip 9 (2009) 2140.[163] K.M. Evans-Nguyen, S.C. Tao, H. Zhu, R.J. Cotter, Anal. Chem. 80 (2008) 1448–

1458.[164] A.I. Neto, C.A. Custódio, W. Song, J.F. Mano, Soft Matter 7 (2011) 4147–4151.[165] Y. Koc, A.J. de Mello, G. McHale, M.I. Newton, P. Roach, N.J. Shirtcliffe, Lab

Chip 8 (2008) 582–586.[166] K. Tsougeni, P.S. Petrou, D.P. Papageorgiou, S.E. Kakabakos, A. Tserepi, E.

Gogolides, Sens. Actuators, B 161 (2012) 216–222.[167] N.J. Shirtcliffe, P. Roach, Methods Mol. Biol. (2013) 269–281.[168] W. Song, J.F. Mano, Soft Matter 9 (2013) 2985–2999.[169] P. Bayiati, A. Malainou, E. Matrozos, A. Tserepi, P.S. Petrou, S.E. Kakabakos, E.

Gogolides, Biosens. Bioelectron. 24 (2009) 2979–2984.[170] A. Malainou, K. Tsougeni, K. Ellinas, P.S. Petrou, V. Constantoudis, E.

Sarantopoulou, K. Awsiuk, A. Bernasik, A. Budkowski, A. Markou, I.Panagiotopoulos, S.E. Kakabakos, E. Gogolides, A. Tserepi, J. Phys. Chem. A117 (2013) 13743–13751.

[171] N. Andreeva, T. Ishizaki, P. Baroch, N. Saito, Sens. Actuators, B 164 (2012) 15–21.[172] A. Valsesia, P. Colpo, M.M. Silvan, T. Meziani, G. Ceccone, F. Rossi, Nano Lett. 4

(2004) 1047–1050.[173] A. Malainou, P.S. Petrou, S.E. Kakabakos, E. Gogolides, A. Tserepi, Biosens.

Bioelectron. 34 (2012) 273–281.[174] G. Simone, Microelectron. Eng. 111 (2013) 369–373.[175] S.M. Oliveira, N.M. Alves, J.F. Mano, J. Adhes. Sci. Technol. 28 (2012) 843–863.[176] F. Rossi, P. Colpo, J. Phys: Conf. Ser. 252 (2010) 012001.[177] P. Colpo, A. Ruiz, L. Ceriotti, F. Rossi, Adv. Biochem. Eng. Biotechnol. (2009)

109–130.[178] T. Ishizaki, N. Saito, O. Takai, Langmuir 26 (2010) 8147–8154.[179] W. Song, D.D. Veiga, C.A. Custódio, J.F. Mano, Adv. Mater. 21 (2009) 1830–1834.[180] S.M. Oliveira, W. Song, N.M. Alves, J.F. Mano, Soft Matter 7 (2011) 8932–8941.[181] Y. Yang, Y. Lai, Q. Zhang, K. Wu, L. Zhang, C. Lin, P. Tang, Colloids Surf. B 79

(2010) 309–313.[182] Y. Lai, L. Lin, F. Pan, J. Huang, R. Song, Y. Huang, C. Lin, H. Fuchs, L. Chi, Small 9

(2013) 2945–2953.[183] K. Tsougeni, A. Bourkoula, P. Petrou, A. Tserepi, S.E. Kakabakos, E. Gogolides,

Microelectron. Eng. 124 (2014) 47–52.[184] F. Poncin-Epaillard, J.M. Herry, P. Marmey, G. Legeay, D. Debarnot, M.N.

Bellon-Fontaine, Mater. Sci. Eng., C 33 (2013) 1152–1161.[185] J. Ballester-Beltrán, P. Rico, D. Moratal, W. Song, J.F. Mano, M. Salmerón-

Sánchez, Soft Matter 7 (2011) 10803–10811.[186] R. Di Mundo, M. Nardulli, A. Milella, P. Favia, R. Dagostino, R. Gristina,

Langmuir 27 (2011) 4914–4921.[187] K.J. Cha, K.S. Park, S.W. Kang, B.H. Cha, B.K. Lee, I.B. Han, D.A. Shin, D.S. Kim,

S.H. Lee, Macromol. Biosci. 11 (2011) 1357–1363.[188] M.B. Oliveira, C.L. Salgado, W. Song, J.F. Mano, Small 9 (2013) 768–778.[189] X. Liu, L. Yuan, D. Li, Z. Tang, Y. Wang, G. Chen, H. Chen, J.L. Brash, J. Mater.

Chem. B 2 (2014) 5718–5738.[190] M.L. Lopez-Donaire, J.P. Santerre, J. Appl. Polym. Sci. 131 (2014).[191] H. Zhu, Z. Guo, W. Liu, Chem. Commun. 50 (2014) 3900–3913.






















































































http://dx.doi.org/10.1038/ncomms4559

http://dx.doi.org/10.1038/ncomms4559

















































































hierarchical micro and nano structured, hydrophilic, superhydrophobic and superoleophobic surfaces...

Documents