Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://nanofluidics.phys.msu.ru/paper/Mendeleev-2012.pdf Äàòà èçìåíåíèÿ: Sat Oct 20 13:02:07 2012 Äàòà èíäåêñèðîâàíèÿ: Sat Feb 2 22:02:22 2013 Êîäèðîâêà:

Mendeleev Communications
Focus Article, Mendeleev Commun., 2012, 22, 229­236

Superhydrophobic textures for microfluidics
Olga I. Vinogradova*a,b and Alexander L. Dubova
a

b

A. N. Frumkin Institute of Physical Chemistry and Electrochemistry, Russian Academy of Sciences, 119991 Moscow, Russian Federation. Fax: +7 495 952 5308; e-mail: oivinograd@yahoo.com Department of Physics, M. V. Lomonosov Moscow State University, 119991 Moscow, Russian Federation
DOI: 10.1016/j.mencom.2012.09.001

Superhydrophobic surfaces have opened a completely new field of investigation with both fundamental and practical perspectives. Research on these materials has mostly focused on their extreme non-wettability, which has large-scale implications in the context of self-cleaning and impact processes. However, the implications of superhydrophobicity for transport phenomena, which are especially important at micro- and nanoscales, remain largely unexplored. Here, we summarize recent advances in this field, including the physical causes of water repellency, the origins of superhydrophobicity and a current switch in focus from wetting to related areas such as the remarkable drag-reducing ability of superhydrophobic materials. In particular, we show that superhydrophobic surfaces induce novel hydrodynamic properties such as giant effective slip, superfluidity and mixing, and affect electrokinetic phenomena. We also discuss developments and strategies in the fabrications of superhydrophobic materials for relevant applications, including microfluidic lab-on-a-chip devices. We finally suggest several remaining challenges in the field. Introduction Controlling the wettability of solid materials is a key issue in surface engineering, and wetting phenomena have been studied for at least two centuries. If a solid surface is flat, smooth and chemically homogeneous, the equilibrium contact angle is deter mined by Young's equation
cos q = ( g
SV

socalled superhydrophobicity. For these surfaces, the effective water contact angle can exceed 150° (see Figure 1).
(a) (b)

­ gSL) / gLV,

(1)
20 µm (c) (d)

where gSV , gSL and gLV are the surface tensions of solid­vapor, solid­liquid and liquid­vapor interfaces, respectively. This contact angle is fixed by the chemical nature of the three phases. In many common situations, this angle lies between 0° and 90° (i.e., a hydrophilic case), but some solids can have a contact angle greater than 90° (a hydrophobic case). However, even on the most hydrophobic solids (fluorinated materials), the contact angle never exceeds 120°. This implies, for instance, that they cannot generate some special properties of obvious practical interest, based on water repellency. However, only very few solids are molecularly smooth. Con versely, most of them, including surfaces of many plants1 and animals, are rough being naturally decorated with special micro and nanotextures. Such a roughness may also be induced by fabrication or coating processes . If these materials are chemically hydrophobic, the presence of a microtexture can dramatically lower the ability of drops to stick due to water repellency or

5 µm

1 µm

Figure 1 Scanning electron microscopy images of superhydrophobic mate rials with different textures and a water drop deposited on a superhydro phobic substrate: (a) hierarchical structure of the Lady's Mantle (Alchemilla diplophylla) leaf (reprinted with permission from ref.1, © 1997, by Annals of Botany Company); (b) a water drop on a fractal superhydrophobic surface (reprinted with permission from ref. 2, © 1996, by American Chemical Society); (c) patterned array of silica micropillars obtained by nanoimprint lithography (adapted from ref. 3) and (d) randomly rough superhydrophobic surface obtained via aerosol assisted chemical vapor deposition (reprinted with per mission from ref. 4, © 2009, by The Royal Society of Chemistry).

Olga I. Vinogradova is the head of the Laboratory of Physical Chemistry Physical Chemistry and Electrochemistry (Russian Academy of Sciences) Lomonosov Moscow State University. Her main research interests are in the wetting, colloidal interactions, and electrochemical systems motivated by science, energy, biophysics etc.

of Modified Surfaces at the A. N. Frumkin Institute of and a professor of the Physics Department at the M. V. field of `soft condensed matter' including fluid dynamics, problems in modern micro and nanofluidics, materials

Alexander L. Dubov is a PhD student at the Laboratory of Physical Chemistry of Modified Surfaces at the A. N. Frumkin Institute of Physical Chemistry and Electrochemistry (Russian Academy of Sciences). His PhD project combines the experimental studies and computer (latticeBoltzmann) simulations of the wetting and hydrodynamic properties of superhydrophobic materials decorated with wellcontrolled micropillars and microgrooves.

© 2012 Mendeleev Communications. All rights reserved.

­ 229 ­


Focus Article, Mendeleev Commun., 2012, 22, 229­236

The issue of large contact angle of superhydrophobic mate rials was first addressed more than 50 years ago being strongly motivated by industrial interest in textile coatings.5,6 However, only in the last decade the field is rapidly advanced, not the least due to potential macroscopic applications (that mostly rely on the formation of a stable air­liquid interface) such as selfcleaning materials,7,8 nonwetting and antifog glass coatings,9­12 bio medical devices, optics13 etc. Thanks to a rapid development of fabrication methods,14­16 first of all techniques coming from microelectronics, we are now able to elaborate substrates whose surfaces are patterned (at micro and nanometer scales) in a very wellcontrolled way that provided a system to test theoretical ideas. In parallel, a significant progress has been made in the characterization of properties of superhydrophobic surfaces, such as wettability and stability of air trapped within the texture. How ever, despite a decade of intense research, superhydrophobicity studies were mostly focused on the surface design of new materials and the characterization of their wetting properties. Moreover, these surfaces even designed for wetting purposes still plagued with problems that restrict their practical applications: contact angle hysteresis, fragility under pressure and difficult and/or expensive fabrication. It is of course important to address remaining challenges in the area of wetting, but today the main research questions become slightly different. Over past few years, the cutting edge in the development of synthetic superhydrophobic surfaces has shifted from wetting towards their greatly enhanced properties that can impact the dynamics of liquids.17­19 For instance, due to a very low degree of sticking to superhydrophobic surfaces, the water drop slides20 or rolls with amazingly large velocity, and a drop hitting such a material just bounces off, which is of obvious interest for such applications as waterproof clothes or wind shields.9 These macroscopic dynamic studies raised a question of a possible large effective slip of water past superhydrophobic materials with trapped gas and its dragreducing ability, which could be extremely important in microfluidics17,18,21,23 ­ a recently emerged distinct field that requires manipulation of fluids in very thin channels and has the potential to influence subject areas from chemical synthesis and biological analysis to optics and information technology. These new opportunities generated new challenges in the design of superhydrophobic materials and also posed many new fundamental issues to address. For example, the effective contact angle cannot be used for characterizing the dynamic properties of superhydrophobic surfaces, and other physical parameters should be employed to optimize materials for given dynamic applications.10,24 Here, we summarize recent advances in the design of super hydrophobic materials by focusing mostly on new challenges from the past several years. We start with a very brief description of the two states of superhydrophobicity. Then follow the results highlighting the new opportunities related to enhanced transport phenomena and their optimization. In the following section we discuss challenges in the materials design for wetting properties and microfluidic applications with a brief consideration of fabrica tion methods. We close with some suggestions concerning experi mental and theoretical studies in this expanding area of research. Wetting: the two states of superhydrophobicity Equation (1) is violated due to irregularities of the superhydro phobic surface. There are two main regimes of wetting of such a surface: either the liquid follows the solid surface, or it leaves air inside the texture. In the first regime (Wenzel scenario5), an increase in the surface area (due to the texture) amplifies the natural hydro phobicity of the material. The key parameter controlling the effective contact angle, cos qW, on this surface is the solid rough

(a) Wenzel state

Liquid

Gas

qW w d

Effective interface

Solid (b) Cassie state Liquid Gas

qC L h

Effective interface

Solid Figure 2 Displacing of the contact line in the (a) Wenzel (homogeneous), and (b) Cassie (heterogeneous) regimes.

ness r, defined as the ratio between the actual and apparent surface areas [Figure 2(a)]:
cos qW = r cos q. (2)

In what follows that to get a very large effective (Wenzel) contact angle one has to increase the roughness of hydrophobic materials. In the second state (first described by Cassie and Baxter6), the liquid only contacts the solid on the top of asperities, on fraction of surface we denote as jS [Figure 2(b)], owing to a trapped air. Assuming that these air pockets have a flat interface with the liquid, we derive the following formula for the effective contact angle:
cos qC = jS ­ 1 + jS cos q, (3)

whence the large effective (Cassie) angle is possible only at a very small fraction of the solid­gas interface. Since most of the `super' properties of superhydrophobic sur faces are due to the air trapped within the texture, the Cassie state is often more desirable than the Wenzel state.25­28 During the last decade, much insight has been gained into the relative stability of the Cassie state, the CassietoWenzel transition,9,22,25­27,29­31 which can occur for a number of reasons, such as an increase in pressure, evaporation, external forces or the presence of surface defects.32­34 This characterization of the wetting properties has been addressed with experiments, theory27 and computer simulations.35 Most of approaches were based on thermodynamic (energy) arguments. A comparison between the surface energies of the Wenzel and Cassie states (by assuming the liquid­air interface to be flat) suggests that stable air pockets are only possible when hydrophobicity is large enough, i.e., at q > qcr , where
cos qcr = (1 ­ jS) /(r ­ jS). (4)

The more advanced energy approach takes into account the curvature of meniscus:36
( g + hDp) cos qcr ­ g cos q > 0, (5)

where h is the texture height, and Dp is the capillary pressure. One of the startling implications of these studies is that the stable Cassie state is possible even at some hydrophilic textures (e.g., mushroomshaped pillars etc.30,37).

­ 230 ­


Focus Article, Mendeleev Commun., 2012, 22, 229­236

For very dilute textures, the energy conditions cannot be ful filled owing to the limitations on chemical hydrophobicity. In this situation, the Cassie state should generally be metastable, and the transition to the Wenzel state happens when the bottom interface of the drop contacts the ground level of the texture.28,38 Note that the actual fractions of solids in contact with the liquid are unknown for arbitrary rough or fractal surfaces, so that one can normally get only a qualitative agreement with theoretical models. Over a decade to go further surfaces with the controlled design of periodic roughness have been intensively used.39 Note that r or jS are no longer adjustable parameters, but they can be precisely calculated from known texture parameters, such as height h, diameter d and distance w (see Figure 2). For example, for a surface decorated by an array of rectangular pillars, one can easily evaluate r = 1 + 4dh / L2 and jS = d /L, where L = w + d is the period of the texture. Correspondingly, for a grooved surface, we get r = 1 + h/L and jS = d/L. Thus, they can vary largely for a given texture. In efforts to better understand the connection between the texture parameters and the stability of the Cassie state, the CassietoWenzel transition has been studied by many groups.28,38,40 Attempts to understand these systems were mostly focused on the wetting of a low density isotropic array of pillars since for such a texture the advancing and even receding angles are outstandingly large.39 The surface with holes does not have this property because of a much higher solid fraction, which inevitably makes the contact angle smaller. The same remark concerns the case of stripes, which is also different because it introduces some anisotropy in wetting and in the hysteresis of the contact angle.41,42 However, despite an increasing body of publications on wetting of anisotropic,30 complex43 and hierarchical surfaces, the quantitative understanding of these systems is still at its infancy. The important property of superhydrophobic surfaces is a contact angle hysteresis,44 which is due to the roughness or hetero geneity of the interface and the elasticity of a contact line.3,45 The hysteresis and pinning are huge at the Wenzel state, which causes a strong adhesion of the drop to such a surface. However, the hysteresis is normally low at the Cassie state, and this allows one a remarkable mobility of the drop, which renders surfaces selfcleaning, and causes droplets to roll (rather than slide) under gravity and rebound (rather than spread) upon impact.9,46,47 The reduction in contact angle hysteresis can dramatically improve droplet mobility and mixing following droplet coalescence in dropletbased, socalled digital microfluidics.48 However, a moderate hysteresis of the contact angles at the directional Cassie surfaces could also be employed in microfluidic devices since it allows one to deflect, sort and capture drops.49 Finally, note that even when the CassietoWenzel transition takes place, it is not clear how to induce the opposite Wenzelto Cassie route, and the transition between the two states is strongly hysteretic. Some preliminary ideas, such as heating or surface vibration50 and electrowetting51 have been proposed. Recent ideas include a gasrestoration mechanism on surfaces with multiple length scales.52 However, more work in this area will be needed. Shift of focus: transport phenomena During the last few years it was well recognized that the unique properties of superhydrophobic materials can significantly impact transport phenomena so that the field has expanded beyond wetting. As we mentioned above, already the observation of an unusual mobility of a drop at superhydrophobic surfaces suggested some microfluidic applications,49 and it also raised the stimulating question of a possible effective hydrodynamic slip.10 A liquid flowing on a hydrophilic solid does not slip at the interface between both phases, but a slip length b (the distance within the solid at which the flow profile extrapolates to zero) might exist if the solid is hydrophobic.53­55 For hydrophobic smooth and

homogeneous surfaces, b can be of the order of tens of nanometers,56­58 but not much more. This effect can be drama tically amplified if the hydrophobic solid is rough, provided that air is trapped in the textures.59,60 Therefore, one could potentially benefit from such a slip in various labonachip applications,21 fuel transport, cooling of electronic chips, etc. This related to reduced drag ability, new topic of interest appears to be a rewarding one for the field as a whole. Below we describe these new opportunities. In response to the challenge of building superhydrophobic surfaces for microfluidics, research efforts have mainly focused on the design of new types of patterned materials to manipulate flows and on the characterization of their friction properties. New investigations required new methods to quantify the effective pro perties of superhydrophobic surfaces since the traditional approach based on characterizing superhydrophobic materials by using solely their contact angle cannot provide a physical insight into transport phenomena. Instead, the drug reduction can be quantified by an effective slip length of the heterogeneous Cassie surface, which refers to a situation where slippage at a complex heterogeneous surface is evaluated by flow averaging over the length scale of the experimental configuration18,21,61,62 (see Figure 3). Straight forward calculations53,63 suggest that a lubricating gas film of a thickness h of viscosity hg (which is about 50 times smaller than the viscosity of a liquid, h) leads to the apparent local slip
h h b = h c - 1m . h . hg hg (6)

Therefore, slip lengths up to hundreds of mm may be then obtained over a gas layer stabilized with a rough texture. However, the composite nature of the interface requires regions of lower slip (or no slip) in direct contact with the liquid, so the effective slip length of the superhydrophobic surface is reduced. A significant progress has been made during past years in the quantitative understanding of an effective slippage past superhydro phobic surfaces. In contrast to wetting studies, the major focus was on surfaces with directional patterns, such as the arrays of parallel superhydrophobic grooves that generate anisotropic effective slip in the Cassie regime. The hydrodynamic slip is different along and perpendicular to the stripes. Axial motion is preferred, and such designs are appropriate when liquid must be
(a) uinter b No-slip plane (b) Effective interface u u
bulk

u

bulk

Liquid

face average

Effective interface Gas

eff

Solid Liquid

interface average

Gas b No-slip plane
eff

Solid

Figure 3 Schematic diagram of velocity profiles and effective slip lengths, beff , near (a) randomly rough and (b) periodic superhydropic surfaces. Dis ordered surfaces display a broad spectrum of scales. Instead, periodic surfaces are typically characterized by a characteristic length of a texture, L, local slip at the gas area, b, and a fraction of the liquid­solid interface, jS.

­ 231 ­


Focus Article, Mendeleev Commun., 2012, 22, 229­236

z

H y

b

|| eff

a u
Figure 4 Sketch of a flat tions along the plates. One texture (adapted from ref. gradient, ­ÿp, produces an
S

­p x

b

^ eff

channel of thickness H with notation for direc wall represents an anisotropic superhydrophobic 18). The driving force, here a uniform pressure effective slip velocity, uus, in a different direction. u

guided. The flow anisotropy is characterized by the secondrank e effective slip tensor, beff º {bijff}, and is represented by a sym metric, positive definite 2 â 2 matrix diagonalized by a rotation
cos a sin a b || 0 beff = Sa e eff = o S-a, Sa = e o. 0 b eff -sin a cos a (7)

Therefore, the flow along any direction of the anisotropic surface can be determined, once the eigenvalues of the effective slip tensor are found from the known spatially nonuniform scalar || ^ slip. For all anisotropic surfaces, the eigenvalues beff and beff of the sliplength tensor correspond to the fastest (greatest forward slip) and slowest (least forward slip) orthogonal directions61 (see Figure 4). In the general case of any direction a, this means that the flow past such surfaces becomes misaligned with the driving force, which is due to the generation of a secondary (transverse) flow.18,64 The quantitative understanding of liquid slippage even in case of a single anisotropic superhydrophobic surface (where H >> L) is still challenging. The problem was tackled theoretically,36,65­67 and several numerical approaches have also been proposed either at the molecular scale, using molecular dynamics,68 or at larger meso scopic scales using finite element methods,69 lattice Boltzmann70,71 or dissipative particle dynamics72 simulations. Recent work sug gested that for any anisotropic surface with arbitrary scalar slip b(y) varying in only one direction, the transverse component of the effective sliplength tensor is equal to a half of the longitudinal one with a two times larger local slip73
= b eff [ b( y)/L] = || b eff [ b( 2y)/L] . 2

(8)

This result opened the possibility of solving a broad class of hydrodynamic problems for onedimensional single textured surfaces. For patterns composed of noslip and partial slip stripes || (i.e., flat liquid­gas interface), beff is65
b
|| eff

.

ln { sec [ p( 1 - jS) /2]} L . (9) p 1 + ( L/pb) ln { sec [ p( 1 - jS)/2] + tan [ p( 1 - jS) /2]}

^ For a flow in a transverse configuration, beff , the result65 can be easily found from equation (9) by applying equation (8). The largest anisotropy of the flow is expected at the limit of b >> L (which was shown to be mathematically equivalent to a perfect slip, b ® ¥),66 i.e., when the eigenvalues of the sliplength tensor are maximal. However, the flow becomes isotropic in the opposite limit of b << L, suggesting that the interplay of different length scales can be used to impact flow properties even qualitatively. For example, in || ^ the situation of a large beff ­ beff one can generate a secondary flow transverse to the direction of an applied force in the vicinity of a super hydrophobic wall by making the velocity profile twisted.18 There fore, superhydrophobic surfaces may find applications in passive microfluidic mixing74 or separation of particles and/or biomolecules.

However, most solids are isotropic, that is, without a preferred direction. This situation is more complicated than that considered above. Nevertheless, the simplest scaling expressions55,75 have been proposed for the geometry of pillars (in the idealized case of perfect slip, b ® ¥, at the gas sectors) by predicting beff µ L /(p j1/2), S which suggests that in the limit of jS ® 0 the array of pillars should give a larger effective slip than longitudinal stripes. Recent approximate analysis76 and semianalytical results77 justified this scaling dependence. Note that, recently, attempts have also been made to evaluate effective slips for isotropic meshlike78 and fractal79 surfaces. However, our fundamental understanding of flow past isotropic surfaces has just begun. A conclusion from the above analysis is that the effective slip is maximized by reducing the solid­liquid area fraction jS and increasing the local slip at the gas sectors, and the role of the texture geometry is essential. Several experimental studies have been conducted to verify these theoretical predictions for both random isotropic80 and periodic surfaces.81­83 In most of the studies, there has been qualitative (or nearly quantitative) agree ment with the theoretical expectations, but the measured slip was only about a few micrometers. This could indicate that the current analytical models of superhydrophobic slip neglect the meniscus curvature, which introduces an additional dissipation mecha nism.67,75,84 These relatively low values of an effective slip are also partly due to a relatively large liquid­solid area fraction. Since at low jS the Cassie state is metastable, it is difficult to obtain a giant effective slip with simple patterns. To increase the stability of the gas phase, it has been suggested to use hierarchical patterns.2,79,85 Such complex hierarchical patterns, which combine roughness on micro and nanoscales, are very common in nature.16 It is reasonable to expect that, in synthetic biomimetic surfaces, the microtexture would provide a large slip, while the presence of a secondary nanotexture could improve the stability of the Cassie state. Indeed, the presence of a secondary texture was shown to decrease the slip length.86,87 We remark that a slip in the hundreds of micron range was reported for such surfaces by using the rheological method. Note that for all types of textures the results obtained with the rheometry technique are far larger than measured by a velocimetry flow profiling, so that would require further verification and theoretical understanding. We remind that the above results apply only for a single surface. If the flow is conducted in a channel with two confining surfaces separated by a distance H ¸ L, the eigenvalues of the sliplength tensor become dependent on H,62,70 which reflects the fact that an effective slip is not a characteristic of the super hydrophobic interface solely, but it may depend on the flow configuration.18,55 In the general case of an arbitrary channel thickness, the problem can only be treated semianalytically or numerically.70,72 However, some general principles to maximize or minimize the effective slip can be suggested in the lubrication limit, H << L (see Figure 5).62,88 Among all possible textures, parallel stripes attain the largest (or smallest) possible slip in a thin channel for a parallel (or perpendicular) orientation with respect to the mean flow. Bounds for isotropic textures are tighter, and the upper (lowest) bounds for effective slip can be attained for a special HashinStrikman pattern. However, the fractal geometry is not necessary since periodic honeycomblike structures can also attend the bounds. Finally, in the particular case of a medium that is invariant by a rotation followed by a p / 2 phase interchange (chessboards, etc.) the effective slip is18
beff = 3H - H. 4 - 1 + 3b/( H + b) (10)

Note that in the limit H >> b for all textures the effective slip coincides with surface average. If H << b, beff µ (1 ­ jS)/ jS for all patterns. This suggests that the key parameter determining effective

­ 232 ­


Focus Article, Mendeleev Commun., 2012, 22, 229­236 (a) (b) (c)

(d)

(e)

(f)

An alternative method to generate a fluid flow in micro fluidics is electroosmosis, i.e., flow generation by an electric field (Figure 6). The fluid outside of the electric double layer (of a thickness of the order of the Debye length, lD) moves as a plug, which has many advantages as compared with conven tional pressuredriven flows.21,92 As discussed earlier,93­95 electro osmotic flow can be significantly amplified by an effective super hydrophobic slip provided the gas­liquid interface is charged, which is not an unrealistic assumption.96 For example, for uniformly charged (qg = qs = q0) anisotropic superhydrophobic surfaces (in the limit b ® ¥), the electroosmotic velocity reads
u =Eq0 lD beff I , h c + lD m
(11)

Figure 5 Special textures arising in theory for a thin channel:62,64,88 (a) stripes, which attain the Wiener bounds of the maximal and minimal effective slip, if oriented parallel or perpendicular to the pressure gradient, respectively; (b) the Hashin­Shtrikman fractal pattern of nested circles, which attains the maximal/minimal slip among all isotropic textures (patches should fill up the whole space, but their number is limited here for clarity); (c) the chessboard texture, whose effective slip follows from the phaseinterchange theorem; (d) circular pillars; (e) the parquet texture for mixing/separation and (f) honeycomb texture.

where I is the unit tensor. The flow is truly anisotropic, and it can exhibit a large, possibly one or two order of magnitude enhance ment even at relatively large jS. An interesting scenario is expected for oppositely charged gas and solid sectors (qg = ­qs = ­q0), which leads to
u =Eq0 lD beff I , h c - lD m
(12)

slip in a thin channel is the solid area fraction, which should be minimized. If it is very small, jS ® 0, such a thin channel can even produce a kind of superfluidity with a pluglike flow.62 A straightforward implication of a superhydrophobic slip in a thin channel would be the generation of a very strong transverse flow, which is much more efficient than that in a thick channel.18 A thin channel situation is also more appropriate since the largest transverse flow is generated at intermediate values of solid fractions, namely, at jS = 0.5, where the effective (forward) slip is relatively small, but the Cassie state is typically stable.64,72 An important conclu sion from this result is that the surface texture which optimizes a transverse flow can significantly differ from the optimizing slip. Hydrodynamic drag in a thin gap have been recently measured with the surface force apparatus,89 and atomic force microscopy.90 The extracted slip length from the fitting of experimental data was of the order of 100 nm, which is only slightly larger than at smooth hydrophobic solid and much smaller than in case of a single super hydrophobic interface. This is qualitatively consistent with the theoretical predictions on a dependency of slip on the height of the channel.70 However, in their analysis authors make assumptions of a constant isotropic, homogeneous, and independent of distance slip that are not generally valid for a heterogeneous superhydrophobic surface as discussed above. We suggest that further analysis of these measurements or similar measurements with other textures should employ recent analysis that takes these effects into account.91
( a) Liquid

by suggesting a very reach behaviour. For example, an inhomo geneous surface charge can induce a flow along and opposite to the field, depending on jS. Another striking result is that an electroneutral superhydrophobic surface can generate an extremely large electroosmotic slip.94 Strong enhancement of transport on superhydrophobic surfaces was also predicted for diffusio osmosis ­ the motion of a fluid induced by a solute gradient.97 These effects remain to be proven experimentally, and they could be very important in the context of energy conversion devices and many other applications. Fabrication of superhydrophobic surfaces: design, materials and methods There are now enough methods of fabrication of superhydro phobic surfaces for various, mostly nonwetting applications. Below we briefly summarize some of them. Here, we do not intend to provide a comprehensive report on available methods for the fabrication of superhydrophobic surfaces, which can be found in reviews,14­16,98 but rather summarize important recent developments relevant for microfluidics applications. Most experimental methods allow one to produce randomly rough superhydrophobic surfaces. They include solgel based methods, chemical deposition, sublimation, controlled polymeri zation, selfassembly of nanorods and nanotubes on the surfaces, electrochemical methods etc. (Figure 7).7,11,14 Since the control of jS is impossible with randomly rough surfaces, these surfaces are often (but not always) wetted in the Wenzel regime. Electrochemical methods include electrochemical deposi tion,100,101 electrochemical polymerization,102,103 electrospin ning,30,104 reactions in galvanic cells,105 etc. These methods are usually used for the fabrication of rough metal surfaces. The rough ness of the surface is controlled by the duration of treatment and the applied voltage. The advantage of the electrochemical methods is weak dependence on the properties of substrates. These can be oxides, some metals or different types of polymer matrices, for instance, based on polyelectrolytes.100 The methods allow one to fabricate the surfaces with different morphology: they can consist of spheres, clusters, whiskers or rods of different size and shape.30,104 The solgel methods99,106 based on a solgel transition are often used for the preparation of oxide surfaces or their organic derivatives. The hydrolysis and condensation steps allow one to control the foam formations to get the surface of the required roughness. The alcoholates of various metals (Zn, Al and Ti) or silicon are used as precursors.99,107­109

l

D

q ( b) l q

s

Charged solid Liquid

D g

q
s

Gas b
eff

No-slip plane

Charged solid

Figure 6 Electroosmotic flow on (a) hydrophilic and (b) superhydro phobic surfaces. The tangential electric field E is directed to the left.

­ 233 ­


Focus Article, Mendeleev Commun., 2012, 22, 229­236 (a) (b)

10 µm (c) (d)

20 µm

10 µm

2 µm

Figure 7 Randomly rough superhydrophobic surfaces: (a) and (b) polymer based coatings obtained by electrospinning (reprinted with permission from ref. 30, © 2008, by The National Academy of Sciences of the USA); (c) gold coated foams obtained via the solgel processing of methyltriethoxysilane (reprinted with permission from ref. 99, © 2003, by American Chemical Society); and (d) carbon nanotubes forest after functionalization with thiols (reprinted with permission from ref. 80, © 2006, by The American Physical Society).

Chemical vapor deposition is one of the most efficient ways to produce surfaces with random roughness or partially ordered nanotube/nanorod arrays. Surfaces are made of different poly mers, such as polymethylsilsesquioxane,110 silicon elastomers4 or inorganic compounds (carbon nanotubes,80 metals or oxides111­113). Surface properties are controlled by the gas pressure and tem perature of substrates. To improve the properties of surfaces, the method is often used in combination with plasma treatment (so called plasma assisted chemical vapor deposition),112 or aerosol producing (aerosol assisted chemical vapor deposition).4 This posttreatment allows one to control the texture parameters. The related method of sublimation is used for the fabrication of randomly rough surfaces prepared by the sublimation of the volatile precursors (usually acetylacetonates) of metal oxides.100 Randomly rough surfaces can also be obtained by hydro thermal methods,114 layerbylayer deposition,101 femtosecond laser irradiation115 etc. Recently the graphenebased materials have also been shown116 to possess superhydrophobic properties. Methods of fabrication of textured surfaces with welldefined topology are based on etching the surface through the mask preset digitally (in chemical and electron etching methods), or on replication of the surface with existing texture (lithography, 2D photonic crystals and anodized aluminum oxide). In templatebased methods, membranes with an ordered array of onedimensional pores with a welldefined size are normally used. As a matrix, one can use anodized aluminum oxide.117 Poly
Photolithography UV light

mers (such as PVA and PAN) can be used as precursors, which crystallize in the pores. These methods have some disadvantages, so that we focus on more advanced lithography techniques below. Lithography methods (Figure 8) are based on the use of well defined templates, so that the texture is accurately set in advance (in the program controlling the electron etching process, printed mask, rigid template replicable onto another material etc.). Such types of lithography as electron and ion etching, soft lithography and photolithography are already very welldeveloped and widely used. Beside there are some novel methods of lithography based on imprint processes and a combination of several types of lithography as we discuss below. The electronbeam etching lithography uses an electron beam, which is scanning the surface of the electron resist creating the template texture set digitally (with a nanometer resolution). Usually, such (expensive!) surfaces are used as masters to pro duce other surfaces (by using other types of lithography). The deep reactive ion etching uses reactive plasma, which etches the chemically unprotected areas of a material. The protecting chemical coating is added through the mask and is not sensitive to gas plasma. The type and amount of gas depends on the mate rial to etch. For example, silicon surfaces require SF6 plasma. This method is cheaper and easier, but it is also used to fabricate masters for other types of lithography. Photolithography exploits photosensitive polymers that cover substrates and are irradiated (usually by UV light) through the previously printed mask. The UVopaque masks could be pro duced with a submicron resolution. The photoresist can cross link under radiation (socalled positive photoresist) or, oppositely, it can dissolve during the exposure (negative photoresist). The not crosslinked part of the polymer can be removed after the exposure. Soft lithography is currently the most popular method since it is simple, cheap and flexible.118,119 The method uses masters fabricated by photolithography, electron or reactive ion litho graphy. These masters are replicated on an elastic PDMS polymer. Crosslinking leads to the formation of an elastic material, which replicates the texture of the master. PDMS with its excellent optical transparency, low toxicity and high permeability to oxygen and carbon dioxide is a material suitable for the fabrication of microchannels to manipulate with biological cells, and micro fluidic PDMS systems have already found applications in cell biology.23 However, PDMS has some disadvantages when micro fluidic systems are used for chemical synthesis (especially in organic and medicinal chemistry) since it dissolves, or swells in common organic solvents.120 The use of polymers other than PDMS can solve this problem and allows reactions to proceed at high tem perature and pressure, but the fabrication could be more difficult. The same remarks concern the use of glass materials, which can be obtained by nanoimprint lithography. This technique is
Imprint lithography Pressure Liquid PDMS Hard master Elastic or hard mold Imprintable surface Thermal or photo hardening under pressure Patterned surface

Soft lithography

Mask Photoresist Exposure after-developing Hardened photoresist patterned surface

Thermal and catalyst curing Elastic PDMS replica

Figure 8 Schematics of lithographic techniques as a sequence of stages: the optical lithography on photoresist; the soft lithography using the photo lithographed pattern as a master; imprint lithography using PDMS replica as a mold.

­ 234 ­


Focus Article, Mendeleev Commun., 2012, 22, 229­236

(a)

100 µm

(b)

q < 90° Flat surface Hydrophobization

1 µm (c) (d) 500 nm

q 110° Flat hydrophobized surface

q > 150° Patterned hydrophobized surface

50 µm
Figure 9 Different types of superhydrophobic textured surfaces for micro fluidic applications obtained by lithographic methods: (a) concentric circular striped texture (reprinted with permission from ref. 41, © 2009, by Elsevier Inc.) so that the grates are parallel to a liquid flow in the rheometer system; (b) circular holes used in the surface force apparatus experiment (reprinted with permission from ref. 89, © 2008, by The American Physical Society); (c) the array of pillars with gradient of jS, (reprinted with permission from ref. 127, © 2009, by EPL Association) which can be used to generate a loco motion of the drop in a dropletbased microfluidics; (d) array of parallel nanometric stripes (reprinted with permission from ref. 128, © 2005, by American Chemical Society).

Figure 10 Contact angle evolution after the hydrophobization of flat and patterned surfaces. Hydrophobizator: SiCl3­CnF2n ­ 1; fatty acids etc.

between theory and experiment, which has to be understood. There are many opportunities to design new experiments and to develop improved theoretical models for electroosmosis near superhydrophobic surfaces. Very promising directions are certainly diffusioosmosis and thermoosmosis, which could lead to a giant amplification of flow in microchannels, even if the liquid­gas interface is uncharged. The solutions to these problems will require new theoretical and experimental approaches, and we expect a significant expansion into this very interesting area of research. We are grateful to E. S. Asmolov, E. Barthel, M. Z. Bazant, A. V. Belyaev, F. Feuillebois, J. Teisseire and D. Quere for many stimulating discussions. This work was supported by the Russian Academy of Sciences (priority programme `Assembly and Inves tigation of Macromolecular Structures of New Generations'). References
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 C. Neinhuis and W. Barthlott, Ann. Bot., 1997, 79, 667. T. Onda, S. Shibuichi, N. Satoh and K. Tsujii, Langmuir, 1996, 12, 2125. A. L. Dubov, J. Teisseire and E. Barthel, EPL, 2012, 97, 26003. C. R. Crick and I. P. Parkin, J. Mater. Chem., 2009, 19, 1074. R. N. Wenzel, Ind. Eng. Chem., 1936, 28, 988. A. B. D. Cassie and S. Baxter, Trans. Faraday Soc., 1944, 40, 546. N. J. Shirtcliffe, G. McHale, S. Atherton and M. I. Newton, Adv. Colloid Interface Sci., 2010, 161, 124. M. Nosonovsky and B. Bhushan, Curr. Opin. Colloid Interface Sci., 2009, 14, 270. D. Quere, Rep. Prog. Phys., 2005, 68, 2495. M. Callies and D. Quere, Soft Matter, 2005, 1, 55. B. Bhushan and Y. C. Jung, Prog. Mater. Sci., 2011, 56, 1. S. Herminghaus, Europhys. Lett., 2000, 52, 165. O. Sato, S. Kubo and Z. Z. Gu, Acc. Chem. Res., 2009, 42, 1. X. M. Li, D. Reinhoudt and M. CregoCalama, Chem. Soc. Rev., 2007, 36, 1350. F. Xia and L. Jiang, Adv. Mater., 2008, 20, 2842. Z. Guo, W. Liu and B. L. Su, J. Colloid Interface Sci., 2011, 353, 333. L. Bocquet and E. Lauga, Nat. Mater., 2011, 10, 334. O. I. Vinogradova and A. V. Belyaev, J. Phys. Condens. Matter, 2011, 23, 184104. G. McHale, M. I. Newton and N. J. Shirtcliffe, Soft Matter, 2010, 6, 714. M. Reyssat, D. Richard, C. Clanet and D. Quere, Faraday Discuss., 2010, 146, 19 H. A. Stone, A. D. Stroock and A. Ajdari, Annu. Rev. Fluid Mechanics, 2004, 36, 381. D. Quere, Annu. Rev. Mater. Res., 2008, 38, 71. G. M. Whitesides, Nature, 2006, 442, 368. A. V. Belyaev and O. I. Vinogradova, Soft Matter, 2010, 6, 4563. N. A. Patankar, Langmuir, 2004, 20, 7097. A. Lafuma and D. Quere, Nat. Mater., 2003, 2, 457. A. Marmur, Langmuir, 2003, 19, 8343. M. Reyssat, J. M. Yeomans and D. Quere, EPL, 2008, 81, 26006. C. Ishino, K. Okumura and D. Quere, Europhys. Lett., 2004, 68, 419. A. Tuteja, W. Choi, J. M. Mabry, G. H. McKinley and R. E. Cohen, Proc. Natl. Acad. Sci. USA, 2008, 105, 18200. C. W. Extrand, Langmuir, 2002, 18, 7991. N. Verplanck, Y. Coffinier, V. Thomy and R. Boukherroub, Nanoscale Res. Lett., 2007, 2, 577. J. Heikenfeld and M. Dhindsa, J. Adhesion Sci. Technol., 2008, 22, 319.

already exploited in industry for the production of compact discs, diffraction gratings etc., and the resolution of this technique has already reached 10 nm. Resist materials are normally thermo plastic polymers121­123 or polymeric UV resists.123 Another type of resists is the socalled solgel resist3,124­126 that makes it pos sible to fabricate oxide and oxideorganic surfaces. This method represents an important step forward, and it can be used to texture large areas, which could be important for highvalue applications of microfluidic systems. In general, the modern lithography methods allow one to fabricate various complex textures (see Figure 9), which could be useful in superhydrophobic microfluidics. Finally, note that most of the textures produced by modern fabrication methods are not hydrophobic, and they require some chemical treatment (hydrophobization) to become superhydro phobic. The hydrophobization can be performed by physical or chemical adsorption (Figure 10). The latter can be carried out in a water or gas phase using (depending on the material of the texture) alkanethiols derivatives,129 silanebased organic com pounds, fatty acids130 etc. The bounding group of the hydro phobizator should be chosen according to the material of the patterned surface: usually it is thiol for the metallic surfaces and silane chloride (or other easily hydrolyzed silane derivatives) for silica and silica based polymers (as PDMS) and metal oxides. Conclusions and future directions In this short review we have tried to discuss the new opportunities in research on superhydrophobic materials, related to greatly enhanced transport properties, which could be very important for manipulations of liquids in microfluidic devices. In particular, advances in lithography to pattern substrates have raised several questions in the modeling of liquid motions over these surfaces and led to the concept of the effective tensorial slip past super hydrophobic surfaces with trapped gas. Such surfaces have the potential to influence microfluidics (or to extend microfluidic systems to nanofluidics), by generating very fast and well con trolled forward and transverse flows in smaller devices. But the field is still at an early stage of development, and would require further development, both at the level of basic science and technologies. Many problems must still be addressed, and many challenges remain. For example, there is quantitative discrepancy

­ 235 ­


Focus Article, Mendeleev Commun., 2012, 22, 229­236 34 N. T. Nguyen, G. P. Zhu, Y. C. Chua, V. N. Phan and S. H. Tan, Langmuir, 2010, 26, 12553. 35 A. Dupuis and J. M. Yeomans, Langmuir, 2005, 21, 2624. 36 C. CottinBizonne, C. Barentin, E. Charlaix, L. Bocquet and J. L. Barrat, Eur. Phys. J. E, 2004, 15, 427. 37 A. Tuteja, W. Choi, M. L. Ma, J. M. Mabry, S. A. Mazzella, G. C. Rutledge, G. H. McKinley and R. E. Cohen, Science, 2007, 318, 1618. 38 S. Moulinet and D. Bartolo, Eur. Phys. J. E, 2007, 24, 251. 39 J. Bico, C. Marzolin and D. Quere, Epl, 1999, 47, 220. 40 C. Pirat, M. Sbragaglia, A. M. Peters, B. M. Borkent, R. G. H. Lammertink, M. Wessling and D. Lohse, Epl, 2008, 81, 66002. 41 W. Choi, A. Tuteja, J. M. Mabry, R. E. Cohen and G. H. McKinley, J. Colloid Interface Sci., 2009, 339, 208. 42 O. Bliznyuk, V. Veligura, E. S. Kooij, H. J. W. Zandvliet and B. Poelsma, Phys. Rev. E, 2011, 83, 041607. 43 A. M. Peters, C. Pirat, M. Sbragaglia, B. M. Borkent, M. Wessling, D. Lohse and R. G. H. Lammertink, Eur. Phys. J. E, 2009, 29, 391. 44 J. F. Joanny and P. G. de Gennes, J. Chem. Phys., 1984, 81, 552. 45 M. Reyssat and D. Quere, J. Phys. Chem. B, 2009, 113, 3906. 46 G. McHale, N. J. Shirtcliffe and M. I. Newton, Langmuir, 2004, 20, 10146. 47 D. Oner and T. J. McCarthy, Langmuir, 2000, 16, 7777. 48 J. P. Rothstein, Annu. Rev. Fluid Mechanics, 2010, 42, 89. 49 M. A. Nilsson and J. P. Rothstein, Phys. Fluids, 2012, 24, 062001. 50 E. Bormashenko, R. Pogreb, G. Whyman, Y. Bormashenko and M. Erlich, Appl. Phys. Lett., 2007, 90, 201917. 51 G. Manukyan, J. M. Oh, D. van den Ende, R. G. H. Lammertink and F. Mugele, Phys. Rev. Lett., 2011, 106, 014501. 52 C. Lee and C. J. Kim, Langmuir, 2009, 25, 12812. 53 O. I. Vinogradova, Langmuir, 1995, 11, 2213. 54 O. I. Vinogradova, Int. J. Miner. Process., 1999, 56, 31. 55 L. Bocquet and J. L. Barrat, Soft Matter, 2007, 3, 685. 56 O. I. Vinogradova and G. E. Yakubov, Langmuir, 2003, 19, 1227. 57 O. I. Vinogradova, K. Koynov, A. Best and F. Feuillebois, Phys. Rev. Lett., 2009, 102, 118302. 58 C. CottinBizonne, B. Cross, A. Steinberger and E. Charlaix, Phys. Rev. Lett., 2005, 94, 056102. 59 O. I. Vinogradova, N. F. Bunkin, N. V. Churaev, O. A. Kiseleva, A. V. Lobeyev and B. W. Ninham, J. Colloid Interface Sci., 1995, 173, 443. 60 C. CottinBizonne, J. L. Barrat, L. Bocquet and E. Charlaix, Nat. Mater., 2003, 2, 237. 61 M. Z. Bazant and O. I. Vinogradova, J. Fluid Mechanics, 2008, 613, 125. 62 F. Feuillebois, M. Z. Bazant and O. I. Vinogradova, Phys. Rev. Lett., 2009, 102, 026001. 63 D. Andrienko, B. Dunweg and O. I. Vinogradova, J. Chem. Phys., 2003, 119, 13106. 64 F. Feuillebois, M. Z. Bazant and O. I. Vinogradova, Phys. Rev. E, 2010, 82, 055301(R). 65 A. V. Belyaev and O. I. Vinogradova, J. Fluid Mechanics, 2010, 652, 489. 66 E. Lauga and H. A. Stone, J. Fluid Mechanics, 2003, 489, 55. 67 M. Sbragaglia and A. Prosperetti, Phys. Fluids, 2007, 19, 043603. 68 N. V. Priezjev, J. Chem. Phys., 2011, 135, 204704. 69 N. V. Priezjev, A. A. Darhuber and S. M. Troian, Phys. Rev. E, 2005, 71, 041608. 70 S. Schmieschek, A. V. Belyaev, J. Harting and O. I. Vinogradova, Phys. Rev. E, 2012, 85, 016324. 71 J. HyvÄluoma, C. Kunert and J. Harting, J. Phys. Condens. Matter, 2011, 23, 184106. 72 J. J. Zhou, A. V. Belyaev, F. Schmid and O. I. Vinogradova, J. Chem. Phys., 2012, 136, 194706. 73 E. S. Asmolov and O. I. Vinogradova, J. Fluid Mechanics, 2012, 706, 108. 74 A. D. Stroock, S. K. W. Dertinger, A. Ajdari, I. Mezic, H. A. Stone and G. M. Whitesides, Science, 2002, 295, 647. 75 C. Ybert, C. Barentin, C. CottinBizonne, P. Joseph and L. Bocquet, Phys. Fluids, 2007, 19, 123601. 76 A. M. J. Davis and E. Lauga, J. Fluid Mechanics, 2010, 661, 402. 77 C. O. Ng and C. Y. Wang, Microfluid. Nanofluid., 2010, 8, 361. 78 A. M. J. Davis and E. Lauga, Phys. Fluids, 2009, 21, 113101. 79 C. CottinBizonne, C. Barentin and L. Bocquet, Phys. Fluids, 2012, 24, 012001. 80 P. Joseph, C. CottinBizonne, J. M. Benoit, C. Ybert, C. Journet, P. Tabeling and L. Bocquet, Phys. Rev. Lett., 2006, 97, 156104. 81 J. Ou, B. Perot and J. P. Rothstein, Phys. Fluids, 2004, 16, 4635. 82 P. Tsai, A. M. Peters, C. Pirat, M. Wessling, R. G. H. Lammertink and D. Lohse, Phys. Fluids, 2009, 21, 112002. 83 C. H. Choi, U. Ulmanella, J. Kim, S. M. Ho and C. J. Kim, Phys. Fluids, 2006, 18, 097105. 84 J. Hyvaluoma and J. Harting, Phys. Rev. Lett., 2008, 100, 246001. 85 N. Gao, Y. Y. Yan, X. Y. Chen and D. J. Mee, Mater. Lett., 2011, 65, 2902. 86 C. Lee and C. J. Kim, Phys. Rev. Lett., 2011, 106, 014502. 87 C. Lee and C. J. Kim, Langmuir, 2011, 27, 4243. 88 F. Feuillebois, M. Z. Bazant and O. I. Vinogradova, Phys. Rev. Lett., 2010, 104, 159902. 89 A. Steinberger, C. CottinBizonne, P. Kleimann and E. Charlaix, Phys. Rev. Lett., 2008, 100, 134501. 90 A. Maali, Y. Pan, B. Bhushan and E. Charlaix, Phys. Rev. E, 2012, 85, 066310. 91 E. S. Asmolov, A. V. Belyaev and O. I. Vinogradova, Phys. Rev. E, 2011, 84, 026330. 92 T. M. Squires and S. R. Quake, Rev. Modern Phys., 2005, 77, 977. 93 S. S. Bahga, O. I. Vinogradova and M. Z. Bazant, J. Fluid Mechanics, 2010, 644, 245. 94 A. V. Belyaev and O. I. Vinogradova, Phys. Rev. Lett., 2011, 107, 098301. 95 T. M. Squires, Phys. Fluids, 2008, 20, 092105. 96 B. J. Kirby and E. F. Hasselbrink, Electrophoresis, 2004, 25, 187. 97 D. M. Huang, C. CottinBizonne, C. Ybert and L. Bocquet, Phys. Rev. Lett., 2008, 101, 064503. 98 P. Roach, N. J. Shirtcliffe and M. I. Newton, Soft Matter, 2008, 4, 224. 99 N. J. Shirtcliffe, G. McHale, M. I. Newton and C. C. Perry, Langmuir, 2003, 19, 5626. 100 A. Nakajima, A. Fujishima, K. Hashimoto and T. Watanabe, Adv. Mater., 1999, 11, 1365. 101 N. Zhao, F. Shi, Z. Q. Wang and X. Zhang, Langmuir, 2005, 21, 4713. 102 M. Nicolas, F. Guittard and S. Geribaldi, J. Polym. Sci., Part A: Polym. Chem., 2007, 45, 4707. 103 H. Yan, K. Kurogi, H. Mayama and K. Tsujii, Angew. Chem. Int. Ed., 2005, 44, 3453. 104 I. Sas, R. E. Gorga, J. A. Joines and K. A. Thoney, J. Polym. Sci., Part B: Polym. Phys., 2012, 50, 824. 105 F. Shi, Y. Y. Song, H. Niu, X. H. Xia, Z. Q. Wang and X. Zhang, Chem. Mater., 2006, 18, 1365. 106 S. S. Latthe, H. Imai, V. Ganesan and A. V. Rao, Microporous Mesoporous Mater., 2010, 130, 115. 107 N. J. Shirtcliffe, G. McHale, M. I. Newton, C. C. Perry and P. Roach, Chem. Commun., 2005, 3135. 108 K. Tadanaga, N. Katata and T. Minami, J. Am. Ceram. Soc., 1997, 80, 3213. 109 N. J. Shirtcliffe, G. McHale, M. I. Newton and C. C. Perry, Langmuir, 2005, 21, 937. 110 J. Zimmermann, F. A. Reifler, G. Fortunato, L. C. Gerhardt and S. Seeger, Adv. Funct. Mater., 2008, 18, 3662. 111 Y. Horiuchi, Y. Shimizu, T. Kamegawa, K. Mori and H. Yamashita, Phys. Chem. Chem. Phys., 2011, 13, 6309. 112 T. Ishizaki, J. Hieda, N. Saito and O. Takai, Electrochim. Acta, 2010, 55, 7094. 113 M. Miwa, A. Nakajima, A. Fujishima, K. Hashimoto and T. Watanabe, Langmuir, 2000, 16, 5754. 114 F. Shi, X. X. Chen, L. Y. Wang, J. Niu, J. H. Yu, Z. Q. Wang and X. Zhang, Chem. Mater., 2005, 17, 6177. 115 T. Baldacchini, J. E. Carey, M. Zhou and E. Mazur, Langmuir, 2006, 22, 4917. 116 J. Rafiee, M. A. Rafiee, Z. Z. Yu and N. Koratkar, Adv. Mater., 2010, 22, 2151. 117 L. Feng, S. H. Li, H. J. Li, J. Zhai, Y. L. Song, L. Jiang and D. B. Zhu, Angew. Chem. Int. Ed., 2002, 41, 1221. 118 J. C. McDonald, D. C. Duffy, J. R. Anderson, D. T. Chiu, H. K. Wu, O. J. A. Schueller and G. M. Whitesides, Electrophoresis, 2000, 21, 27. 119 J. C. McDonald and G. M. Whitesides, Acc. Chem. Res., 2002, 35, 491. 120 J. N. Lee, C. Park and G. M. Whitesides, Anal. Chem., 2003, 75, 6544. 121 A. Pozzato, S. Dal Zilio, G. Fois, D. Vendramin, G. Mistura, M. Belotti, Y. Chen and M. Natali, Microelectron. Eng., 2006, 83, 884. 122 R. Pogreb, G. Whyman, R. Barayev, E. Bormashenko and D. Aurbach, Appl. Phys. Lett., 2009, 94, 221902. 123 B. D. Gates, Q. B. Xu, M. Stewart, D. Ryan, C. G. Willson and G. M. Whitesides, Chem. Rev., 2005, 105, 1171. 124 C. Peroz, C. Heitz, E. Barthel, E. Sondergard and V. Goletto, J. Vacuum Sci. Technol. B, 2007, 25, L27. 125 C. Peroz, V. Chauveau, E. Barthel and E. Sondergard, Adv. Mater., 2009, 21, 555. 126 A. Letailleur, J. Teisseire, N. Chemin, E. Barthel and E. Sondergard, Chem. Mater., 2010, 22, 3143. 127 M. Reyssat, F. Pardo and D. Quere, EPL, 2009, 87, 5. 128 D. G. Choi, J. H. Jeong, Y. S. Sim, E. S. Lee, W. S. Kim and B. S. Bae, Langmuir, 2005, 21, 9390. 129 J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo and G. M. Whitesides, Chem. Rev., 2005, 105, 1103. 130 E. Hosono, S. Fujihara, I. Honma and H. S. Zhou, J. Am. Chem. Soc., 2005, 127, 13458.

Received: 24th July 2012; Com. 12/3959

­ 236 ­