pubs.acs.org/Langmuir © 2010 American Chemical Society
Ionic Liquid Microdroplets as Versatile Lithographic Molds for Sculpting Curved Topographies on Soft Materials Surfaces J. Perera-Nun~ez, A. Mendez-Vilas, L. Labajos-Broncano, and M. L. Gonzalez-Martı´ n* Department of Applied Physics, University of Extremadura, Avda. Elvas s/n, 06071 Badajoz, Spain, and Networking Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN), Badajoz, Spain Received July 13, 2010. Revised Manuscript Received September 13, 2010 Soft lithography comprises a set of approaches for shaping the surface of soft materials such as PDMS on the microscopic scales. These procedures usually begin with the development of templates/masters normally generated by electron or photolithography techniques. However, the richness in available shapes is limited, usually producing shapes containing sharp parts. Innovation is called for to develop reliable approaches capable of imparting well-defined 3D curved shapes to these solids, a topology that is somehow unnatural for solid surfaces. Here we report on the use of tiny drops of room-temperature ionic liquid, organic liquids that have attracted increasing amounts of attention in recent years because of their unique chemical properties) as a versatile platform for imprinting PDMS with tunable 3D curved geometry, which is out of reach of conventional lithographic techniques and ranges from almost flat depressions to almost closed cavities on the millimeter to micrometer scale. The concept exploits a peculiar combination of physical properties displayed by ionic liquids as their null volatility and their polarity, together with some unique properties of liquid surfaces as their virtually null surface roughness. Proof-of-concept experiments show their application as chemical microreactors and ultrasmooth optical lenses. This all-liquid method is simple, low-cost, versatile, maskless, tension-free, and easily scalable, so we envision a community-wide application in numerous modern physical, chemical, biological, and engineering settings.
Introduction The ability to sculpt soft materials such as poly(dimethylsiloxane) (PDMS) on small scales is the basis of currently important scientific and technological achievements such as optical microlenses, chemical microreactors, and lab-on-a-chip and cell culture/sorting devices, to name a few. The process usually begins with electrongenerated or photolithographically generated solid templates/ masters, which limit the richness of accessible shapes, usually producing geometries containing corners or edges (cylinders, square-sectioned structures, pyramids, etc.).1,2 Innovation is called for in the development of reliable approaches to provide solid surfaces with well-defined 3D curved shapes, a surface topology that is somewhat unnatural for solids. PDMS is the polymer of choice in many of the above-mentioned applications because of its favorable chemical, mechanical, optical, and processing properties and its biocompatibility.3 It is considered to be chemically inert and highly nonpolar, with bulk properties that are tunable by varying the level of cross linking of its monomers. Liquid is the type of condensed matter whose surface displays curved shapes as a result of surface tension/energy. In particular, small droplets deposited on a solid surface (sessile droplets) adopt a spherical-cap geometry, with the contact angle being dependent on the three interfacial tensions. Given that liquid microdispensing technologies capable of producing small liquid droplets of different natures are being rapidly developed, especially in the context of microfluidic and lab-on-a-chip technologies, it is appealing to explore the feasibility of using ultrasmall liquid *To whom correspondence should be addressed. E-mail:
[email protected]. Phone: 34-924-289532. Fax: 34-924-289651. (1) Xia, Y. N.; Whitesides, G. M. Angew. Chem., Int. Ed. 1998, 37, 550. (2) Quake, S. R.; Scherer, A. Science 2000, 290, 1536–1540. (3) Silicones in Medical Applications, Dow Corning Company technical note. http://www.dowcorning.com/content/publishedlit/Chapter17.pdf (accessed October 2009).
17712 DOI: 10.1021/la102799x
droplets as lithographic molds. They could overcome the design restrictions and high costs associated with conventional lithography methods because they could provide a way to impart curved topography to malleable materials such as PDMS. In the past, however, the exploitation of ultrasmall drops for this purpose has been challenging because of the fact that such ultrasmall liquid volumes evaporate very quickly,4-9 even in saturated atmospheres.10 Attempts to tackle these issues have provided complex routes or limited control over accessible contact angles.11-13 Room-temperature ionic liquids (RTILs, hereafter ILs) are organic liquids that have attracted increasing amounts of attention in recent years as a result of their unique chemical properties. They are entirely ionic, highly polar in general, nonvolatile,14 and liquid over a wide range of temperature. Although most of the expectations raised by ILs are related to their unique and tunable physicochemical properties and their potential as green solvents (they have been named designer solvents), striking applications (4) Villarroya, M.; Abadal, G.; Verd, J.; Teva, J.; Perez-Murano, F.; Figueras, E.; Montserrat, J.; Uranga, A.; Esteve, J.; Barniol, N. IEEE Trans. Nanotechnol. 2007, 6, 509–512. (5) NanoDispense(R) Contact Angle Measurements, FirstTenAngstroms technical note, http://www.firsttenangstroms.com (accessed October 2009). (6) Ondarcuhu, T.; Arcamone, J.; Fang, A.; Durou, H.; Dujardin, E.; Rius, G.; Perez-Murano, F. Eur. Phys. J. Spec. Top. 2009, 166, 15–20. (7) Saya, D.; Leı¨ chle, T.; Pourciel, J. B.; Bergaud, C.; Nicu, L. J. Micromech. Microeng. 2007, 17, N1–N5. (8) Arcamone, J.; Dujardin, E.; Rius, G.; Perez-Murano, F.; Ondarcuchu, T. J. Phys. Chem. B 2007, 111, 13020–13027. (9) Jung, Y. C.; Bhushan, B. J. Vac. Sci. Technol., A. 2008, 26, 777–782. (10) Butt, H. J.; Golovko, D. S.; Bonaccurso, E. J. Phys. Chem. B 2007, 111, 5277–5283. (11) Giang, U. T.; Lee, D.; King, M. R.; DeLouise, L. A. Lab Chip 2007, 7, 1660–1662. (12) Leopoldes, J.; Damman, P. Nat. Mater. 2006, 5, 957–961. (13) Park, J. Y.; Hwang, C. M.; Lee, S.-H. Biomed. Microdevices 2009, 11, 129–133. (14) Marsh, K. N.; Boxall, J. A.; Lichtenthaler, R. Fluid Phase Equilib. 2004, 219, 93–98.
Published on Web 09/30/2010
Langmuir 2010, 26(22), 17712–17719
Perera-N un~ez et al.
Article
have already been found outside the pure chemical arena. Borra et al.15 have exploited the intrinsically smooth surface topography of an ionic liquid, jointly with their null volatility in vacuum, to construct a liquid primary telescope mirror on the earth’s moon. Rutten et al.16 have reported the use of IL surfaces as substrata for high-density data storage. These are just two recent examples of how a peculiar combination of physical properties of ILs can be exploited technologically. In this work, we report on the use of tiny drops of ILs deposited on a solid surface as versatile templates for imprinting the PDMS surface with tunable 3D curved geometry ranging from almost flat depressions to almost closed cavities from the millimeter to the micrometer scale. The concept exploits a peculiar combination of some physical properties displayed by ILs, such as their almost null vapor pressure, which permits total control of the volume of the ILs droplets, their high polarity, and their tunable physicochemical properties, together with some unique properties of liquid surfaces such as their virtually null surface roughness.
Experimental Section Materials: Ionic Liquids, Solid Substrata, and PDMS. Hydrophilic imidazolium-based ionic liquids, sharing the same anion ([BF4]-) and 1-alkyl-3-methylimidazolium cations ([CnMIm]þ) and purchased from Sigma-Aldrich, was used in this study as received. Concretely, 1-ethyl-3-methylimidazolium tetrafluoroborate ([EMin][BF4]), 1-butyl-3-methylimidazolium tetrafluoroborate ([BMim][BF4]), 1-hexyl-3-methylimidazolium tetrafluoroborate ([HMim][BF4]), and 1-decyl-3-methylimidazolium tetrafluoroborate ([DecMim][BF4]) were used. Solid surfaces of high (glass), intermediate (PVC, PS), and low (PDMS, PTFE) polarity were also employed in this study. Polymeric substrata (PTFE, PDMS, PS, and PVC) were obtained from Goodfellow (England) and used as received. Previous works had revealed the high contact angles formed by imidazoliumbased ionic liquids on rough fluoropolymers.3,17 A glass substrate (Menxel-Glaser, microscope slide, Germany) was vigorously cleaned with a smooth cotton cloth using distilled water with light soap and sonicated in distilled water (10 min). Then, the substrate was immersed in a chromic acid solution for up to 1 h, extensively rinsed with water, and again sonicated in distilled water (10 min). Finally, the surface was dried in an oven at 40 °C for 1 h. Sylgard 184 silicone, a curable two-part PDMS elastomer, was purchased from DOW Corning. Liquid PDMS, with a very low surface tension (20.4 mN/m),18 was prepared by mixing well 10 parts of the prepolymer base with 1 part of the curing agent (by weight) and degassing for 1 h under vacuum in order to remove the bubbles formed during the mixing procedure. The nonpolar and noncharged character of the PDMS is a relevant feature to this work because it minimizes the interaction with the highly charged (ionic) and polar ILs. PDMS is known not to interact strongly with polar substances such as water and electrolytes, whereas it suffers from swelling in organic solvents. Because of this, it could happen that ILs dissolve PDMS because of the dual electrolyte/organicsolvent character of these liquids. However, it has been reported, first on the macroscopic and later on the molecular level, that noncharged polymers such as PDMS and poly(vinyl alcohol) were not penetrated by imidazolium-based ILs.19,20 Finally, another (15) Borra, E. F.; Seddiki, O.; Angel, R.; Eisenstein, D.; Hickson, P.; Seddon, K. R.; Worden, S. P. Nature 2007, 447, 979–981. (16) Rutten, F. J.; Tadesse, H.; Licence, P. Angew. Chem., Int. Ed. 2007, 46, 4163–4165. (17) Gao, L.; McCarthy, T. J. J. Am. Chem. Soc. 2007, 129, 3804–3805. € (18) Chen, W.; Fadeev, A. Y.; Hsieh, M. C.; Oner, D.; Youngblood, J.; McCarthy, T. J. Langmuir 1999, 15, 3395–3399. (19) Sch€afer, T.; Rodrigues, C. M.; Alfonso, C. A. M.; Crespo, J. G. Chem. Commun. 2001, 17, 1622–1623. (20) Sch€afer, T.; Di Paolo, R. E.; Franco, R.; Crespo, J. G. Chem. Commun. 2005, 20, 2594–2596.
Langmuir 2010, 26(22), 17712–17719
good point is that the ILs employed in this work are denser than liquid PDMS (e.g., [BMim][BF4] and liquid PDMS densities are 1.120 and 0.965 g/cm3, respectively), which avoids flotation problems. For the replication of already microstructured PDMS surfaces, prior to replication they were first covered with an antistiction silane agent ((heptadecafluoro-1,1,2,2-tetrahydrodecyl)dimethylchlorosilane (C12H8ClF17Si), Gelest). On the basis of the measurement of the water contact angle of the silanized surface at different exposure times to silane vapor, we determined that 15 min was enough for complete coverage. Droplet Replication Procedure. A schematic representation of PDMS micromolding using IL microdroplets as a template is depicted in Figure 1 (under Spanish patent application number 201030795). First, a macroscopic IL droplet was deposited onto the solid substrate using a micropipet. Then, numerous droplets were generated on the solid substrate by gently dragging the first IL droplet along the surface with the aid of thin fibers. Of course, droplets can be formed or deposited on the solid surface by more sophisticated techniques (e.g., drop-on-demand dispensing21 and flow focusing22 technologies), but our simple method is an easy way to generate randomly distributed droplets, from microscopic to macroscopic sizes, on the substrata with a high spatial density. Once the IL droplets were formed on the substrate surfaces, the degassed liquid PDMS was gently poured over the solid surfaces. A magnified optical view of a [BMim][BF4] droplet immersed in liquid PDMS and deposited onto PTFE is included in Figure 1. The system displayed stability over time. Finally, the PDMS solution was cured at 60 °C for 2 h, and the resulting solid PDMS replica was removed from the liquid master and sonicated in acetone for 5 min in order to remove the IL remaining on the PDMS surface. Experimental Techniques. The [BMim][BF4] ionic liquidliquid PDMS interfacial tension was measured using the pendant drop technique, giving a value of 21.25 mN/m. For that, a DSA10 goniometer from Kruss (Germany) was used. Prior to determining the interfacial tension, the surface tension of the [BMim][BF4] ionic liquid was measured, giving a value of 44 mN/m at room temperature (26 °C), which is within the range of values available in the literature.23 X-ray photoelectron spectroscopy (XPS) has been used to investigate the elemental composition of the PDMS surface after the micromolding and washing procedure. To this end, a K-alpha model (Thermo Scientific, U.K.) with a pass energy of 200 eV, an Al KR excitation source (1486.6 eV), and an electron collector placed at a normal takeoff angle were used under a chamber pressure of 10-9 Torr. To analyze the imprinted topographies on the cured PDMS surface, the micromolded PDMS was cut into several sections and side-view images were taken using a Nikon Epiphot 300 inverted metallographic microscope with a Micro Publisher 5.0 digital camera (Qimaging, Canada). The quantification of the geometry of the cavities was accomplished using Image J (NIH), a public domain Java image-processing program, and a plug-in developed by Stalder et al. for contact angle measurement using a model derived from a first-order perturbation solution of the Laplace equation for axisymmetric drops.24 The surface roughness of liquid-replicated PDMS surfaces was quantified using a PicoLE AFM from Agilent Technologies operated in tapping mode. (21) Ulmke, H.; Mitschke, M.; Bauckhage, K. Chem. Eng. Technol. 2001, 24, 69–70. (22) Ga~nan-Calvo, M.; Gonzalez-Prieto, R.; Herrada-Gutierrez, M. A.; RiescoChueca, P.; Flores-Mosquera, M. Nat. Phys. 2007, 3, 737–742. (23) Rilo-Pico, E.; Garcı´ a-Garabal, S.; Varela, L. M.; Rodrı´ guez, J.; Cabeza, O. Proceedings of Termo-2008, XI Encuentro Inter-Bienal del Grupo Especializado de Termodin amica (GET) de las Reales Sociedades Espa~ nolas de Fı´sica y Quı´mica, Jaca, Spain, Sept 7-10 2008; pp 169-177. (24) Stalder, A. F.; Kulik, G.; Sage, D.; Barbieri, L.; Hoffmann, P. Colloids Surf., A 2006, 286, 92–100.
DOI: 10.1021/la102799x
17713
Perera-N un~ez et al.
Article
Figure 1. Schematic representation of the process used to replicate liquid IL structures from the macroscale to the microscale. (a) Distribution of IL droplets, from macroscopic to microscopic, onto the substratum. (b) PDMS coating and curing process. (c) Removal and washing of the substrate. (d) Final micromolded PDMS surface. An optical side view of an IL droplet over PTFE immersed in liquid PDMS is highlighted with an arrow.
Figure 2. Cross-sectional views of the produced spherical-cap surface topographies, with varying curvature (quantified by the cavity contact angle, upper row), at different dimensional domains (left column). The material used as the substratum for the [BMim][BF4] droplet is also indicated in the upper row. Environmental scanning electron microscopy (ESEM) was used to image the microcrystal growth inside the imprinted cavities on the cured PDMS surface. ESEM imaging was performed at the MNCN-CSIC (Madrid, Spain) using an INSPECT model from FEI Company with a 30 kV acceleration voltage. Noise arising in almost closed cavities is likely a result of the very high local curvature and/or capillary condensation of water vapor from the ESEM chamber.
Results and Discussion Sculpting Curved Topographies. Droplets of the 1-butyl-3methylimidazolium tetrafluoroborate ([BMIm][BF4]) IL were generated with diameters that spanned 3 orders of magnitude (from a few micrometers, 100 μm, to macroscopic droplets, 103μm). The process, described in the Experimental Section, is graphically depicted in Figure 1. As mentioned there, the contact angle formed by the IL drop was changed using substrata of different polarities. Figure 2 shows cross-sectional views of the imprinted topographies taken by optical microscopy for each of the IL-substratum combinations on different dimensional scales. To avoid optical distortions in the side views, only cavities close to the PDMS cutting plane, which could be sharply imaged, were 17714 DOI: 10.1021/la102799x
considered. Also, simultaneous focusing of the cavity and the striations observed in the PDMS cutting plane ensures that the cavity is very close to the cutting plane. As seen in Figure 2, lens effects are produced specially in large-angle cavities, where bright spots corresponding to the microscope lamp appear at the center. From these results, it is readily seen that commercially available materials can be used as substrata for sessile IL droplets in order to produce curved topographies on the PDMS surface with a very wide range of curvature, ranging from almost flat cavities using glass to almost closed cavities using PTFE and PDMS, with intermediate-curvature cavities using PVC and PS. Eventual deviations from spherical-cap geometry toward ellipsoidal geometries as a result of hydrostatic influences were explored by mathematical fitting. Profiles of all of the cavities could be excellently well fitted to spherical caps, which rules out any gravity effect on the droplet shape (Figure 3). This finding can be rationalized by considering the interplay between the Laplace and hydrostatic pressures exerted over a liquid droplet immersed in another fluid, which is quantified by the capillary length, lcap= (γ/gΔF)1/2, with γ being the liquid-liquid interfacial tension, ΔF being the density difference in the two fluids, and g being Langmuir 2010, 26(22), 17712–17719
Perera-N un~ez et al.
Article
Figure 3. Fitting of created topographical modifications to spherical-cap geometry.
gravitational acceleration. This length provides an upper limit for the droplet size when neglecting the influence of gravity. By taking into account the interfacial tension between liquid PDMS and the [BMIm][BF4] IL, 21.25 mN/m, as well as the densities of this IL and PDMS, 1.120 and 0.965 g/cm3, respectively, an lcap value of 3.7 mm is obtained. This result justifies why all of the cavities aroused on the PDMS surface via the sculpting method have spherical-cap geometry because all of the droplets considered in this study and those of interest in lithography have values that are well below this critical value. For each IL-solid substratum combination, cavity contact angles were measured over the whole range of size achieved in order to explore any potential size effect. Indeed, the phenomenon of contact angle size dependence is usually studied in terms of a phenomenological line tension that acts at the three-phase line, which is expected to play a significant role only on the microscopic scale.25 Some works have explored line tension values for IL droplets.26,27 In our case, with droplet diameters varying over 3 orders of magnitude (the highest ever reported) and without influence of evaporation, no trend in the contact angle, only statistical fluctuations, appeared, which rules out any significant influence of the line tension (Figure 4). This invariance ensures no change in curvature due to the size of the ILs droplets, which is definitely a good point with a view to applications. The mean (scale-independent) contact angles of the different topographical cavities on the PDMS surface are given in Figure 2. (25) Mendez-Vilas, A.; Jodar-Reyes, A. B.; Gonzalez-Martı´ n, M. L. Small 2009, 5, 1366–1390. (26) Furuta, T.; Nakajima, A.; Sakai, M.; Isobe, T.; Kameshimaa, Y.; Okada, K. Chem. Lett. 2009, 38, 580. (27) Halka, V.; Freyland, W. Phys. Chem. 2007, 222, 117–121.
Langmuir 2010, 26(22), 17712–17719
Figure 4. Cavity contact angle vs diameter of the spherical cavities produced by the replication of [BMim][BF4] droplets deposited on glass, PTFE, PDMS, PVC, and PS.
Another parameter that might have an impact on the cavity shape and on the contact angle is the thickness of the liquid PDMS layer. To this end, the topographies produced on the PDMS surface by our procedure using three thickness values of the PDMS layer, all within a realistic working range (0.5 mm, 3 mm, and 1 cm), were analyzed. The results are summarized in Figure 5. As seen, within the experimental error, no significant effect of the layer thickness on the contact angles was found. To rationalize this result, let us consider the additional hydrostatic pressure generated by the PDMS layer on an IL drop because this overpressure causes a hydrostatic work of compression on the IL drops that could reduce their size and change their contact angle DOI: 10.1021/la102799x
17715
Article
Figure 5. Average contact angle of the cavities produced by [BMim][BF4] droplet replication as a function of the PDMS layer thickness.
while maintaining their spherical shape because the droplet size is always lower than the capillary length. For example, for the 1-cmthick PDMS layer, this overpressure is about 88.2 Pa, which is much lower than atmospheric pressure. However, its effect is not appreciable because of the relatively incompressible character of ILs with isothermal compressibilities lower than those of organic solvents and similar to that of water.28 This insensitivity of the IL microdroplets to the PDMS layer thickness is definitely another good point, ensuring higher reproducibility for inter- and intralaboratory experiences. This is in contrast with using gas microbubbles as templates,11 where the bubble volume varies with the thickness of the PDMS layer as a result of the high gas permeability of PDMS. Fortunately, problems of this nature are not present in our approach because of the noncompatibility between PDMS and the IL. An XPS surface chemical analysis was performed inside and outside the microcavities in order to detect eventual IL residues in the micromolded PDMS surface after a simple washing step involving ultrasonication in an acetone bath for 5 min. A representative XPS survey spectrum is shown in Figure 6 for the PDMS surface (inside and outside a cavity) and for a [BMIm][BF4] droplet. As seen, C and O are present both in the IL and PDMS. In the case of the IL, O arises from water adsorption from the environment. Si peaks from PDMS are clearly visible. There was no evidence of the presence of B, F, or N, the IL characteristic elements, either inside or outside the microcavity. This is consistent with previous findings on the nonsolvent behavior of ILs towards PDMS.29,30 This ease of cleaning should pave the way for the community-wide lithographic use of IL microdroplets. Other solvents such as acetonitrile, which is almost as polar as ILs and does not swell PDMS,31 might also be used. The topography of the PDMS surface resulting from replication of the IL [BMIm][BF4] droplet was quantified using atomic force microscopy (AFM). As seen in Figure 7, an extremely smooth surface is obtained, with topographic fluctuations on the order of a few nanometers over a 3 μm region. Ra and Rrms roughness were 0.60 ( 0.1 and 0.80 ( 0.1 nm, respectively. This is in contrast to other methods using solidified droplets (for example, melt metal or polymer droplets), which may roughen upon solidification.32,33 (28) Zhiyong, G.; Brennecke, J. F. J. Chem. Eng. Data 2002, 47, 339–345. (29) Sch€afer, T.; Rodrigues, C. M.; Alfonso, C. A. M.; Crespo, J. G. Chem. Commun. 2001, 17, 1622–1623. (30) Sch€afer, T.; Di Paolo, R. E.; Franco, R.; Crespo, J. G. Chem. Commun. 2005, 20, 2594–2596. (31) Lee, J. N.; Park, C.; Whitesides, G. M. Anal. Chem. 2003, 75, 6544–6554. (32) Li, R.; Ashgriz, N.; Chandra, S.; Andrews, J. R. J. Mater. Sci. 2007, 42, 9511–9523. (33) Chatain, D.; Carter, W. C. Nat. Mater. 2004, 3, 843–845.
17716 DOI: 10.1021/la102799x
Perera-N un~ez et al.
Proof of Concepts. We envisage a wide range of application fields for microstructuring such as that reported in this work, including microfluidics, micro-optics (microlenses or optical microcavities), microchemistry, and biology/biotechnology (microchips, microarrays, cell culture surfaces, and the effect of physical confinement on cell adhesion and differentiation). Optical Microlenses. Plastic mini/microlenses are currently playing a key role in reducing the size and cost of modern optical communication and data storage systems. The smoothness of the curved topography is a critical point in the design of optical microlenses. In contrast to the production of glass microlenses, which have to be meticulously ground and polished in an extremely expensive and tricky process,34 our all-liquid approach naturally produces ultrasmooth lenses with no surface processing, which is a direct consequence of using liquid droplets as a template (a liquid droplet is expected to have a surface roughness on the order of the thermal fluctuations of its molecules).35,36 Figure 8 shows the capability of the topographically modified PDMS (low-curvature modification) to act as optical mini- and microlenses. As reported elsewhere,37 microlenses based on fluid-fluid interfaces are optically smooth if the roughness is below 1/10 of the wavelengths of most of the light spectrum (i.e., 400 nm < λ < 700 nm). In our case, lenses shown in Figure 8 are characterized by an Rrms roughness of ∼1 nm, a 400-700 fraction of λ, which causes their liquidlike appearance. Also, because no mechanical or thermal process is involved, residual stresses are avoided. The whole study featuring the optical performance of these curved topographies will be reported elsewhere. Finally, we mention that the direct-printing (maskless) nature of the sculpting procedure that we proposed here will allow printing over nonplanar substrata. This should find immediate applications in optics and photonics, such as in the development of new artificial compound eye concepts, which are of importance in the design of novel vision devices.38-40 Besides, good spherical fitting is also crucial for optical microresonators, which are based on the circulation of light within a single microcavity, and which can be exploited, for example, for the label-free ultrasensitive detection of pathogens.41 Chemical Micro(Femtoliter) Reactors. In Figure 9, we show a potential application of a large-curvature cavity on the PDMS surface as a chemical microreactor. By locally altering the surface topography, we can confine liquids and chemical reactions in ultrasmall volumes, enabling high-throughput screenings to be performed. In this way, for example, biomolecules can be studied at a functionally relevant concentration. The growth of micro/ nanocrystals inside these microreactors is another attractive application because they are required in many scientific and technological areas for their unique properties. Because confined liquids can be evaporated much more slowly that freely exposed liquids,42 good crystals can be produced inside these microvessels. This is especially appealing in the case of large biomolecules, such as proteins, which are often available in small quantities and are not easily crystallized. Crystallization is a prerequisite for their 3D (34) http://www.smartalix.com/liquidlens.html (accessed October 2009). (35) Braslau, A.; Pershan, P. S.; Swislow, G.; Ocko, B. M.; Als-Nielsen, J. Phys. Rev. A 1988, 38, 2457–2470. (36) Yang, C. Y.; Ho, F. H.; Wang, P. J.; Yeh, J. A. Langmuir 2010, 26, 6314–6319. (37) Zeng, X.; Jiang, H. J. Microelectromech. Syst. 2008, 17, 1210–1217. (38) Duparre, J. W.; Wippermann, F. C. Bioinspiration Biomimetics 2006, 1, R1–R16. (39) Jeong, K. H.; Kim, J.; Lee, L. P. Science 2006, 312, 557–561. (40) Lee, L. P.; Szema, R. Science 2005, 310, 1148–1150. (41) Ren, H. C.; Vollmer, F.; Arnold, S.; Libchaber, A. Opt. Express 2007, 15, 17410–17423. (42) Boedicker, J. Q.; Vincent, M. E.; Ismagilov, R. F. Angew. Chem., Int. Ed. 2009, 48, 5908–5911.
Langmuir 2010, 26(22), 17712–17719
Perera-N un~ez et al.
Article
Figure 6. XPS survey spectrum taken inside and outside a cavity (using PS as the substratum for the [BMim][BF4] droplet) after the washing step, jointly with another taken over a [BMim][BF4] droplet.
Figure 7. Tapping-mode AFM image and height profile of a replicated PDMS surface. Figure 9. High-curvature microcavity (cavity contact angle larger than 90°) on a PDMS surface used as a chemical microreactor (top view): micrometer-sized NaCl single crystal (highlighted by an arrow) grown inside the microcavity (ESEM image).
Figure 8. Well-defined spherical-cap geometry (low curvature), jointly with polymer transparency and ultrasmooth surface topography, produces quality plane-concave optical lenses. In the left image, a minilens that is focused on text is shown. The right image contains a microlens, out of focus and in focus, showing a university logo. The inset shows the printed logo and the image formed through the minilens.
structure determination, which enables rational drug design. As a proof of concept, NaCl crystals have been grown inside our Langmuir 2010, 26(22), 17712–17719
high-curvature microvessels. To this end, the micromolded PDMS surface was immersed in a 2 M NaCl aqueous solution and then slowly retracted in a vertical position under ambient conditions. Penetration can be spontaneous, induced by a small pressure,43,44 or can be due to capillarity suction cause by the Laplace pressure in the final stage of evaporation.45 As can be seen in Figure 9, NaCl microcrystals grew within the cavities. Eventually, penetration of the liquid into the microcavities could even be facilitated by hydrophilizing the PDMS surface using UV (43) Liu, B.; Lange, F. F. J. Colloid Interface Sci. 2006, 298, 899–909. (44) Lafuma, A.; Quere, D. Nat. Mater. 2003, 2, 457–460. (45) Bormashenko, E.; Bormashenko, Y.; Stein, T.; Whyman, G.; Pogreb, R.; Barkay, Z. Langmuir 2007, 23, 4378–4382.
DOI: 10.1021/la102799x
17717
Perera-N un~ez et al.
Article
Figure 11. Contact angles of cavities formed using a single sub-
Figure 10. Replication of plane-concave microlenses producing plane-convex lenses (with image inversion). Transmission of the colored university logo is shown.
light or by the transient opening of microcavities by stretching the (elastic) PDMS substrate. If needed, evaporation can even be slowed by increasing the environmental humidity or by transiently closing the upper cavity holes by compressing the PDMS substratum, which could be reopened at the end of the process. In this field, the smoothness of the cavity is desired because irregularities can provide nucleation sites and interfere with crystal growth.46 Finally, the spherical shape of the cavities can be advantageous over other geometries such as cylinders, which are more readily available using conventional lithographic methods but whose corners/edges could provoke heterogeneity in reactions proceeding inside. Versatility of the Presented Micromolding Approach. Expanding Topologies and Patterns. Once the desired topographies have been imparted to the solid surface, they might be replicated, giving rise to protruding topographical features with a wide range of potential applications. In the case of low-curvature cavities, their replication would give rise to plane-convex microlenses. An example can be seen in Figure 10. In the biomedical field, macroscopic features of this kind have been shown to inhibit cell adhesion and proliferation.47 Our approach will allow these topographies to be imprinted from the macro- to the microscale (with the latter being more comparable to cell sizes). In the case of high-curvature cavities, replication could produce very novel topographical structures with potential, for example, in the fabrication of optical high-Q resonators.48 Regarding the spatial arrangement, because microdispensing technology already allows the printing of ultrasmall amounts of liquid with a high spatial density over macroscopic (∼square centimeter) areas (e.g., microarray spotters), our lithographic method based on IL microdroplets is easily scalable when using it to produce IL droplets. Also, this technology could be exploited to write complex patterns (e.g., liquid lines/channels), enabling the printing of complex round-sectioned microfluidic circuits, with this application resting on the interfacial physics of cylindrical droplets.49 Finally, although our method is intrinsically maskless, it could be used in conjunction with masks (e.g., TEM grids or μCP masks) to produce ordered arrays. Expanding the Range of ILs, Substrata, and Polymers. As far as ILs, substrata, and curable polymers are concerned, the method presented is also versatile. In this work, we have presented (46) Dalton Research Group (University of Washington) research note. http:// depts.washington.edu/eooptic/linkfiles/Crystallisation_Techniques.doc (accessed October 2009). (47) Park, J. Y.; Lee, D. H.; Lee, E. J.; Lee, S. H. Lab Chip 2009, 9, 2043–2049. (48) Armani, A. M.; Srinivasan, A.; Vahala, K. J. Nano Lett. 2007, 7, 1823–1826. (49) Gau, H.; Herminghaus, S.; Lenz, P.; Lipowsky, R. Science 1999, 283, 46–49.
17718 DOI: 10.1021/la102799x
stratum (PS) and different ILs sharing the same anion ([BF4]-) and the 1-alkyl-3-methylimidazolium cation ([CnMIm]þ).
the details for one well-characterized, highly polar IL. However, the method is general and other ILs might be used. Indeed, one could use a single substratum and change the ionic liquid in order to modify the polarity/surface tension with the aim of varying the drop contact angle and therefore the curvature of the imprinted topography. Specifically, one might choose to work with a homologous series of ILs sharing the same anion and differing in the alkyl chain length of the organic cation. In Figure 11, we present some preliminary results obtained from this approach, employing PS as the substratum and some ILs belong to the 1-alkyl-3-methylimidazolium family, [CnMIm]þ, with all of them sharing the [BF4]- anion, namely, ionic liquids [EMin][BF4], [BMim][BF4], [HMim][BF4], and [DecMim][BF4]. As seen, the cavity contact angle can also be widely varied in this way. However, we opted to study the concept fully using a single highly polar IL and different substrata to change the contact angle because the use of ILs portraying larger alkyl chains could enhance the chemical (hydrophobic) interactions between the IL and PDMS.50 Many other solid polymers have well-described physicochemical properties, which might be explored as substrata.51 Also, simple treatments such as UV-light irradiation of the polymer surface or changes in the surface roughness may also be used to change the IL droplet contact angle. Other liquid curable polymers might be explored for specific applications if their properties are known. A method to synthesize curable liquid Teflon has been described in the literature.52 On the basis of the nonaffinity between Teflon (PTFE) and ILs, our approach will likely be extensible to this highly solvent-resistant polymer. Finally, because the surface of solidified PDMS can easily be covered with firmly attached oxide layers such as SiO253 and TiO2,54 to which a rich variety of chemical species (e.g., organosilanes) can be bound, our microstructuring approach could be transferred to areas where other surface chemistries are advantageous.
Summary Ionic liquids can be used as a general platform to provide soft materials with well-defined curved topographies, a capability that is out of reach of conventional lithographic methods. Liquid templates can be designed under ambient conditions. Our all-liquid (50) Restolho, J.; Mata, J. L.; Saramago, B. J. Colloid Interface Sci. 2009, 340, 82–86. (51) Kim, M. S.; Khang, G.; Lee, H. B. Prog. Polym. Sci. 2008, 33, 138–164. (52) Rolland, J. P.; Van Dam, R. M.; Schorzman, D. A.; Quake, S. R.; DeSimone, J. M. J. Am. Chem. Soc. 2004, 126, 2322–2323. (53) Abate, A. R.; Lee, D.; Do, T.; Holtze, C.; Weitz, D. A. Lab Chip 2008, 8, 516–518. (54) Girshevitz, O.; Nitzan, Y.; Sukenik, C. N. Chem. Mater. 2008, 20, 1390–1396.
Langmuir 2010, 26(22), 17712–17719
Perera-N un~ez et al.
method is simple, low-cost, versatile, maskless, tension-free, and easily scalable, so we envision a community-wide application of this method because of the enormous importance of microsculpted soft materials in micro-optics, microchemistry, microfluidics, biotechnology, and lab-on-a-chip concepts.
Langmuir 2010, 26(22), 17712–17719
Article
Acknowledgment. The Spanish Ministry for Science and Technology (MAT2009-14695-C04-01) and Junta de Extremadura (GRU09126) are gratefully acknowledged for providing funding for this research. J.P.-N. acknowledges Junta de Extremadura for a predoctoral grant.
DOI: 10.1021/la102799x
17719
