NANO LETTERS
Inherently Superoleophobic Nanocomposite Coatings by Spray Atomization
2009 Vol. 9, No. 1 501-505
Adam Steele,* Ilker Bayer, and Eric Loth Aerospace Engineering Department, The UniVersity of Illinois at Urbana-Champaign, 104 South Wright Street, 306 Talbot Laboratory, Urbana, Illinois 61801 Received December 10, 2008
ABSTRACT We describe a technique to fabricate, for the first time, superoleophobic coatings by spray casting nanoparticle-polymer suspensions. The method involves the use of ZnO nanoparticles blended with a waterborne perfluoroacrylic polymer emulsion using cosolvents. Acetone is shown to be an effective compatibilizing cosolvent to produce self-assembling nanocomposite slurries that form hierarchical nanotextured morphology upon curing. Fabricated coating surface morphology is investigated with an environmental scanning electron microscope (ESEM), and surface wettability is characterized by static and dynamic contact angle measurements. The coatings can be applied to large and/or flexible substrates by spray coating with ease and require no additional surface treatments of commonly used hydrophobic molecules such as fluorosilanes; i.e., the nanocomposites are inherently superoleophobic. The superoleophobic nature of the coatings is also discussed within the framework of Cassie-Baxter and Wenzel wetting theories.
Superoleophobic coatings have many potential applications including fluid transfer, fluid power systems, stain resistant and antifouling materials, and microfluidics among others. However, creating synthetic surfaces with superoleophobic properties, i.e., with apparent oil contact angles greater than 150° and contact angle hysteresis less than 10°, has proven to be much more difficult than creating superhydrophobic surfaces. It is well-known that the degree to which a solid surface repels a liquid mainly depends upon two factors: surface energy and surface morphology. The surface energy affects the liquid-solid surface interface by influencing the attractive forces between the liquid and solid at the molecular scale. When oil contacts a surface, the oil molecules have a stronger attraction to the molecules of the solid surface than to each other due to its low surface tension (i.e., the adhesive forces are stronger than the cohesive forces). Compared to water, oils thus exhibit a much higher solid surface attraction due to this lower surface tension. Surface morphology alteration, on the other hand, at the micro- and/or nanoscale can allow for an air layer to be maintained in the space between the asperities during liquid contact. Known as the lotus effect,1 this result can significantly reduce droplet and fluid flow resistance and it has been successfully employed in creating an array of superhydrophobic surfaces.2 The addition of surface texture will either reduce or increase the solid-liquid contact area depending on whether or not the lotus effect is attained and, thus, will lead to an * Corresponding author,
[email protected]. 10.1021/nl8037272 CCC: $40.75 Published on Web 12/19/2008
2009 American Chemical Society
apparent contact angle, θ*, for a droplet on a surface. This result can be explained by two independently developed models: the Wenzel model3 given in eq 1 and the Cassie-Baxter model4 given in eq 2 cos θ * ) r cos θ cos θ * ) -1 + φs(1 + cos θ)
(1) (2)
where θ is the equilibrium contact angle, r is the ratio of the actual over the apparent surface area of the substrate (a number larger than unity), and φs is the fraction of solid-liquid contact. An oil droplet that completely wets a textured surface is in the “Wenzel state” and tends to leave an oily trail or stain as it slides and spreads.5 Conversely, an oil droplet with a composite interface on a textured surface is in the “Cassie state” and can have far less droplet adhesion. Superoleophobicity has been achieved by only a handful of researchers and, until now, has required complicated and costly fabrication methods which are frequently limited to a particular substrate.2 The extreme difficulty in creating superoleophobic surfaces stems from the fact that oils and alkanes (such as hexadecane and octane) have an equilibrium contact angle less than 90° on all currently known natural and artificial surfaces.6 The lowest surface energy end groups in monolayer films that are currently known are -CH2 > -CH3 > -CF2 > -CF2H > -CF3 in decreasing order;7 i.e., -CF3 has the lowest surface energy.8 To date, the techniques that have been used to create superoleophobic surfaces include silicon etching with fluorosilane functionalization6,9 and anodically oxidized aluminum with fluori-
Figure 1. ESEM images of nanocomposite coatings showing (a) microroughness and self-similarity at 100 µm scale bar and (b) nanoroughness for hierarchical surface structure at 1 µm scale bar.
nated monoalkyl phosphate functionalization.10 A variety of other surfaces have been created that have shown high repellency to other liquids with mild surface tensions such as diiodomethane. Fabrication methods for these surfaces include plasma modification of benzoxazine films,11 electrodeposition processes,12 silicone nanofilament growth with fluorosilane functionalization,13 and plasma polymer layers deposited on microrough PTFE substrates.14 Finally, only a few polymer and nanocomposite coating methods have shown some oleophobicity, but no superoleophobicity has been reported to date.15-18 Superoleophobic nanocomposite coating techniques could perhaps show the most promise for the widest array of applications as they are not specific to a particular substrate, they can be easily applied to large surface areas, and they do not typically require complicated and costly application methods. Nanocomposites have been spray coated in a variety of other applications including solar cells,19 dielectric materials,20 and cylinder bore wall coatings21 to exploit some of these advantages. In the current study, we describe a technique to fabricate, for the first time, superoleophobic coatings by spray casting nanoparticle-polymer suspensions to produce hierarchical nanotextured surface morphology (i.e., multiscale surface roughness) which is self-similar throughout the surface area (Figure 1). The oleophobicity and hydrophobicity of the nanocomposite surfaces are quantified by measuring the apparent contact angle (static, θ*stat; advancing, θ*adv; and receding, θ*rec) of 10 µL oil droplets (DTE 11M, Mobil) and distilled water droplets on each surface. A goniometer (model CAM 200, KSV Instruments) is used to measure the static contact angle, and a high-speed digital camera (Motion Pro X, Red Lake) is used for dynamic advancing and receding contact angle measurement. The surface morphology is also characterized using an environmental scanning electron microscope (XL30, Philips). Nanocomposites are fabricated using solution-processable mixtures of ZnO nanoparticles (Nanoguard, 50 nm diameter, Alfa Aesar) to create nanoroughness and waterborne per502
Table 1. Solution Compositions and Corresponding Acronyms for Textured Surfaces name
slurry components
NC1 (most superhydrophobic)
4% wt ZnO, 88% wt acetone, 2.4% wt perfluoroalkyl methacrylic copolymer, 5.6% wt distilled water 8% wt ZnO, 84% wt acetone, 2.4% wt perfluoroalkyl methacrylic copolymer, 5.6% wt distilled water
NC2 (most superoleophobic)
fluoroalkyl methacrylic copolymer (PMC) (30% wt polymer, 70% wt water; Dupont) as a low surface energy binder. Thin coatings are obtained by spray casting the solutions onto clean microscope glass slides (note that it is also applicable to wide variety of other substrate materials) using an internal mix, double-action airbrush atomizer (model VL-SET, Paasche). The substrates are coated with a single spray application from a distance of approximately 30 cm above the substrate and then air-dried for approximately 12 h. The specific slurry compositions that yielded the most hydrophobic and oleophobic surfaces are described in Table 1. In order to obtain a “dry” coating, a cosolvent with a boiling point much lower than water is required to ensure that a substantial portion of the solvent within the spray mist evaporates before impacting the substrate. If this is not the case, as with water-based solutions, the solvent evaporates mainly on the substrate leading to nonuniform coverage through mechanisms described as the “coffee stain” effect.22 The coffee stain effect causes the water solvent contact line to pin on the substrate during evaporation, transporting nanocomposite material to the edges and forming multiple rings on the coating surface. Acetone is used as a cosolvent in the current work for a dual purpose: (1) to counteract the coffee stain effect and allow uniform curing and (2) to produce nanocomposite slurries that self-assemble to form hierarchical surface morphology upon curing. As shown in Figure 2, with an increasing concentration of acetone cosolvent, the surface structure transitions from a low degree of micromorphology to a high degree of 5-10 µm “microspheres” composed of nanocomposite material, generating Nano Lett., Vol. 9, No. 1, 2009
Figure 2. Apparent contact angle of 10 µL oil and water droplets as a function of acetone cosolvent concentration using a 3.3:1 ZnO: PMC mass fraction.
hierarchical surface roughness. When this roughness structure is achieved, it can be observed that the static contact angle of oil increases significantly from 120° to above 150°. This transition can be explained by considering the solvent evaporation rate during flight of the atomized slurry toward the substrate. If one considers individual droplets, a fast solvent evaporation rate during flight (due to higher acetone concentration) can significantly alter the dispersion concentration of the polymer and nanoparticles within the spray droplet. If most of the solvent evaporates before the droplet reaches the substrate, a droplet with a high concentration of nanoparticles and polymer impacts the surface. This allows for a “dry” coating as compared to a “wet” coating using solvents with lower evaporation rates. Thus, adjusting the solvent evaporation rate of the slurry droplets using cosolvent(s) allows the final coating morphology and its degree of oleophobicity and hydrophobicity to be tuned (see Figure 2). Since the fluropolymer is inherently hydrophobic and nearly oleophobic and ZnO nanoparticles are a highly oleophilic and hydrophilic ceramic, the nanoparticle to polymer ratio within the solution is also very critical. One of the functions of the ZnO nanoparticles in the present composite coatings is to generate nanoscale roughness. Therefore, its surface should remain coated with a thin layer of the polymer and its concentration relative to the PMC dispersed in the solution must be carefully adjusted. It is found that the composite NC2 has the ideal nanoparticle to polymer ratio for creating superoleophobic surfaces (Figure 3). Detailed ESEM analysis of the surface morphology of this coating reveals the formation of a solid polymer microfoam-like structure coated with ZnO nanoparticles. The average size of the foamlike cells is on the order of 1 µm as shown in Figure 3. As shown in the ESEM insets in Figure 4 (2-5 µm scale bars), an increase in the nanoparticle mass fraction causes the surface roughness to become less dictated by the morphology of the polymer and more dictated by the ZnO nanoparticle filler. At very high nanoparticle loadings, the polymer cells are observed to be completely coated with nanoparticles, forming a self-similar nanoscale texture. On Nano Lett., Vol. 9, No. 1, 2009
Figure 3. ESEM images of NC2 at 10 µm scale bar showing the composite surface of both ZnO nanoparticles and PMC cells.
Figure 4. Apparent static contact angle of 10 µL droplets as a function of nanoparticle/polymer mass fraction for performance measurement using ideal acetone cosolvent concentration.
the basis of the data in Figure 4 and considering only contact angle for the moment, the coating morphology can be tailored to obtain both superoleophobic or superhydrophobic states for any ZnO:PMC mass fraction between about 0.8 and 3.3. As noted in Table 1, a mass fraction of 3.3 is ideal for superoleophobicity (NC2) and a mass fraction of 1.7 is ideal for superhydrophobicity (NC1) which results in static contact angles of 157° and 168°, respectively. The abrupt reduction in contact angle when the nanoparticle mass fraction is zero is explained by a loss of appreciable surface roughness due to the lack of compatibilizing ZnO nanoparticles as shown in Figure 5a. PMC is an amorphous polymer, and when the waterborne PMC solution is exposed to a high concentration of acetone, the polymer dispersion is disrupted and partial phase separation occurs (Figure 5b). This separation is explained by the block copolymer segment comprised of quaternary amine functionality, which allows the polymer to remain suspended in water-based solutions as a complex aggregate, reorienting and causing a partial loss of water solubility. The phaseseparated PMC then agglomerates, becoming much too large and viscous for spray atomization and is therefore removed (Figure 5b). However, even with a highly acetone-based solution remaining, the main fluoroacrylic backbone of the 503
Figure 5. ESEM images with 2 µm scale bar of (a) spray-coated acetone-PMC solution showing a lamellar structure with comparatively low surface roughness and (b) phase-separated PMC lamellae from acetone-PMC reaction (removed before spray coating).
copolymer, 2-(perfluoroalkyl)ethyl methacrylate), still remains partially in solution as it is easily dissolvable in acetone. The resulting spray coating then cures devoid of substantial surface roughness (Figure 5a) and remains hydrophobic as shown in Figure 4. Compatibilization of the entire PMC suspension with acetone cosolvent is achieved by dispersing ZnO nanoparticles in acetone prior to the addition of PMC solution. ZnO nanoparticles disrupt the phase separation of the PMC in solution which is a very similar process observed in the phase separation of polyurethane.23 Individual nanoparticles can constrain the polymer chain mobility and, thus, force the thin lamellae to separate and form cellular-like structures within the ZnO nanoparticle network as shown in Figure 3. As a result, a dispersion of ∼10 µm agglomerations of ZnO:PMC particles is formed which can be sprayed though the atomizing nozzle to produce nanocomposites with hierarchical surface morphology upon curing (see Figure 1). The contact angle hysteresis, i.e., the difference between the advancing and receding contact angle, is another critical metric that is used to gauge wettability performance. It is clear in Figure 6 that the oil droplet hysteresis reaches a minimum at a ZnO:PMC mass fraction of 3.3 (NC2, Table 1). This minimum hysteresis value of 4° indicates the ability of an oil droplet to roll almost freely on the surface at low tilt angles without leaving an oily trailing stain, very similar to how water droplets roll on a superhydrophobic lotus leaf. Oil droplets on oleophobic surfaces will, conversely, either remain pinned or slide at much higher tilt angles and usually leave an oily stain behind, greatly reducing oil repellency. The observed superoleophobicity is a rare property among synthetic liquid-repellent surfaces and a first for nanocomposites. Correspondingly, water droplets show low hysteresis values at all positive nanoparticle loadings, signifying superhydrophobicity for the entire range. Hexadecane, an alkane hydrocarbon liquid with low surface tension commonly used in the literature, is also tested on NC2 for comparison and indeed the nanocomposite is also superoleophobic to this fluid (θ* ) 154° and 6° hysteresis). 504
Figure 6. Contact angle hysteresis of 10 µL droplets for mass fraction performance measurement using ideal acetone cosolvent concentration.
The wettability performance is also considered in terms of the Cassie-Baxter model (eq 2) for a contact area fraction φs ) 0.07 as well as the Wenzel model (eq 1) for a high and low roughness factor of r ) 2.96 and r ) 1.07 in Figure 7. It is observed that oil droplets on NC1 and NC2 correlate well to the Cassie-Baxter prediction for a 7% solid-liquid contact area. It is less clear based on this figure whether or not water droplets are in the Cassie state or the Wenzel state on the nanocomposite surface. However, because of the low hysteresis and lack of droplet pinning, both uncharacteristic of droplets in the Wenzel state, it is reasonable to assume that water droplets are also in the Cassie state. Another interesting observation that can be made from Figure 7 is that the oil droplet data points are located in the second quadrant of the figure. This result indicates that untextured PMC is oleophilic (θ < 90° on the x axis) and is then made oleophobic (θ*adv > 90° on the y axis) with surface texturing or, in this case, superoleophobic. Conversely, the water data points are located in the first quadrant, indicating that untextured PMC is hydrophobic and is then made more Nano Lett., Vol. 9, No. 1, 2009
Figure 7. Apparent advancing contact angle as a function of equilibrium contact angle for 10 µL oil and water droplets, compared to the Cassie-Baxter and Wenzel models.
hydrophobic (in this case superhydrophobic) with surface texturing. In conclusion, we have created, for the first time, a superoleophobic (and superhydrophobic) nanocomposite coating that can be easily applied to almost any surface, in addition to the glass substrates used in the current work, similar to spraying paint. The resulting surface attains high contact angles and low contact angle hysteresis values using a precisely controlled slurry composition of a low boiling point cosolvent, a low surface energy polymer, and compatibilizing nanoparticles. The composition can also be varied to tailor the final coating morphology and its degree of oleophobicity and hydrophobicity, while maintaining a Cassie state for both oil and water droplets. Acknowledgment. The authors acknowledge the support of the National Science Foundation and the Center for Compact and Efficient Fluid Power. We thank Scott Robinson and Cate Wallace for ESEM imaging assistance, Glennys Mensing for goniometer assistance, Stephen Moran for wettability measurement work, Dr. John Lambros for high-speed camera assistance, and Dr. Andrew Alleyne for support. References (1) Cheng, Y.; Rodak, D. E. Is the lotus leaf superhydrophobic? Appl. Phys. Lett. 2005, 86, 144101.1-144101.3. (2) Feng, I.; Jiang, L. Design and creation of superwetting/antiwetting surfaces. AdV. Mater. 2006, 18, 3063–3078. (3) Wenzel, R. W. Resistance of solid surfaces to wetting by water. Ind. Eng. Chem. 1936, 28, 988–994. (4) Cassie, A. B. D.; Baxter, S. Wettability of Porous Surfaces. Trans. Faraday Soc. 1944, 40, 546–551.
Nano Lett., Vol. 9, No. 1, 2009
(5) Kiuru, M.; Alakoski, E. Low sliding angles in hydrophobic and oleophobic coatings prepared with plasma discharge method. Mater. Lett. 2004, 58, 2213–2216. (6) Tuteja, A.; Choi, W.; Ma, M.; Mabry, J.; Mazzella, S.; Rutledge, G.; McKinley, G.; Cohen, R. Designing superoleophobic surfaces. Science 2007, 318, 1618–1622. (7) Zisman, W. A. Relation of the equilibrium contact angle to liquid and solid construction. In Contact Angle, Wettability, and Adhesion; Fowkes, F. M., Ed.; ACS Advances in Chemistry Series; American Chemical Society: Washington, DC, 1964; Vol. 43, pp 1-51. (8) Nishino, T.; Meguro, M.; Nakamae, K.; Matsushita, M.; Ueda, Y. The lowest surface free energy based on -CF3 alignment. Langmuir 1999, 15, 4321–4323. (9) Cao, L.; Price, T.; Weiss, M.; Gao, D. Super water- and oil-repellent surfaces on intrinsically hydrophilic and oleophilic porous silicon films. Langmuir 2008, 24, 1640–1643. (10) Shibuichi, S.; Yamamoto, T.; Onda, T.; Tsujii, K. Super water- and oil-repellent surfaces resulting from fractal structure. J. Colloid Interface Sci. 1998, 208, 287–294. (11) Wang, C.; Chiou, S.; Ko, F.; Chou, C.; Lin, H.; Huang, C.; Chang, F. Fabrication of biomimetic super-amphiphobic surfaces through plasma modification of benzoxazine films. Macromol. Rapid Commun. 2006, 27, 333–337. (12) Xi, J.; Feng, L.; Jiang, L. A general approach for fabrication of superhydrophobic and superamphiphobic surfaces. Appl. Phys. Lett. 2008, 92, 053102.1-053102.3. (13) Zimmermann, J.; Rabe, M.; Artus, G. R. J.; Seeger, S. Patterned superfunctional surfaces based on a silicone nanofilament coating. Soft Matter 2008, 4, 450–452. (14) Coulson, S. R.; Woodward, I.; Badyal, J. P. S.; Brewer, S. A.; Willis, C. Super-repellent composite fluoropolymer surfaces. J. Phys. Chem. 2000, 104, 8836–8840. (15) Coulson, S. R.; Woodward, I.; Badyal, J. P. S.; Brewer, S. A.; Willis, C. Ultralow surface energy plasma polymer films. Chem. Mater. 2000, 12, 2031–2038. (16) Xie, Q.; Xu, J.; Feng, L.; Jiang, L.; Tang, W.; Luo, X.; Han, C. Facile creation of a super-amphiphobic coating surface with bionic microstructure. AdV. Mater. 2004, 16, 302–305. (17) Hsieh, C.; Chen, J.; Kuo, R.; Lin, T.; Wu, C. Influence of surface roughness on water- and oil-repellent surfaces coated with nanoparticles. Appl. Surf. Sci. 2005, 240, 318–326. (18) Yabu, H.; Takebayashi, M.; Tanaka, M.; Shimomura, M. Superhydrophobic and lipophobic properties of self-organized honeycomb and pincushion structures. Langmuir 2005, 21, 3235–3237. (19) Nanu, M.; Schoonman, J.; Goossens, A. Nanocomposite threedimensional solar cells obtained by chemical spray deposition. Nano Lett. 2005, 5, 1716–1719. (20) Zhao, X.; Hinchliffel, C.; Johnston, C.; Dobson, P.; Grant, P. Spray deposition of polymer nanocomposite films for dielectric applications. Mater. Sci. Eng. B 2008, 151, 140–145. (21) Bobzin, K.; Ernst, F.; Zwick, J.; Schlaefer, T.; Cook, D.; Nassenstein, K.; Schwenk, A.; Schreiber, F.; Wenz, T.; Flores, G.; Hahn, M. Coating bores of light metal engine blocks with a nanocomposite material using the plasma transferred wire arc thermal spray process. J. Therm. Spray Technol. 2008, 17, 344–351. (22) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S. R.; Witten, T. A. Capillary flow as the cause of ring stains from dried liquid drops. Nature 1997, 389, 827–829. (23) Zheng, J.; Ozisik, R.; Siegel, R. Disruption of self-assembly and altered mechanical behavior in polyurethane/zinc oxide nanocomposites. Polymer 2005, 46, 10873–10882.
NL8037272
505
