Ultrahydrophobic Surfaces Effects of Topography Length

Ultrahydrophobic Surfaces Effects of Topography Length

(Parte 1 de 2)

Ultrahydrophobic Surfaces. Effects of Topography Length Scales on Wettability

Didem O‹ner and Thomas J. McCarthy*

Polymer Science and Engineering Department, University of Massachusetts, Amherst, Massachusetts 01003

Received April 20, 2000. In Final Form: June 23, 2000

We discuss dynamic hydrophobicity from the perspective of the force required to move a water droplet on a surface and argue that the structure of the three-phase contact line is important. We studied the wettability of a series of silicon surfaces that were prepared by photolithography and hydrophobized using silanizationreagents.Hydrocarbon,siloxane,andfluorocarbonsurfaceswereprepared.Thesurfacescontain posts of different sizes, shapes, and separations. Surfaces containing square posts with X-Y dimensions of 32 ím and less exhibited ultrahydrophobic behavior with high advancing and receding water contact angles.Waterdropletsmovedveryeasilyonthesesurfacesandrolledoffofslightlytiltedsurfaces.Contact angleswereindependentofthepostheightfrom20to140ímandindependentofsurfacechemistry.Water droplets were pinned on surfaces containing square posts with larger dimensions. Increasing the distance between posts and changing the shape of the posts from square to staggered rhombus, star, or indented square caused increases in receding contact angles. We ascribe these contact angle increases to decreases inthecontactlengthandincreasesintortuosityofthethree-phasecontactline.Themaximumlengthscale of roughness that imparts ultrahydrophobicity is 32 ím.


Thewettabilityofsolidsurfacesisanimportantproperty that is manipulated in numerous practical applications and, as well, is used as an analytical technique (contact angleanalysis)tocharacterizesurfacesinbasicmaterials research.Thereisanenormousliterature1thatdescribes research over the past 60 years that has been directed at studying and/or controlling the interaction between fluids(forthemostpartwater)andsolids.Inmanysenses this mature field of science is well understood and is not controversial.

Inthispaperweareconcernedwithhydrophobicityand begin with a general discussion of this effect. Figure 1 shows a schematic representation of droplets of water on two different surfaces and asks the question, “Which surface is more hydrophobic?” Almost everyone would agree that the surface on the left is more hydrophobic than the one on the right; the contact angle is higher in theleftcase.Thesituationmay,however,bemorecomplex. What if when the surface on the right is tilted 1° from the horizontal, the droplet slides off, and when the surface on the left is tilted to any angleseven upside downsthe dropletstayspinnedtothesurface?Withthesefactsknown we must say that the surface on the right is more hydrophobic. This is not just semantics; this issue has practical importance. In a static situation, clearly the left surface is more hydrophobic, but if you want a waterrepellant surface, the static contact angle is irrelevant; you want the drop to move with very little applied force. Thus, dynamic wettability, which is a function of contact angle hysteresis and not the contact angle, is important.

The importance of contact angle hysteresis to hydrophobicity was first addressed decades ago, and the relationship shown in eq 1 was derived and reported.2

The equation predicts the minimum angle of tilt (R)a t which a droplet (with surface tension çLV) will spontane- ously move, where ıA and ıR are the advancing and receding contact angles, g is the force due to gravity, and mandwarethemassandwidth(horizontaltothedirection of drop movement) of the droplet. It is clear from this equation that the difference between advancing and receding contact angles (hysteresis) and not the absolute valuesofthecontactanglesisimportanttohydrophobicity. Thatthisisindeedthecaseisintuitivefromtheargument depicted in Figure 2. Consider three droplets of water on threesurfaces.Thesurfacesincasesaandbexhibitwater contactanglesofıA/ıR)120°/80°,andthesurfaceincase * To whom correspondence should be addressed. E-mail:


(1) Mittal, K. L., Ed. Contact Angle, Wettability and Adhesion; VSP: Utrecht, The Netherlands, 1993. (2) Furmidge, C. G. L. J. Colloid Sci. 1962, 17, 309.

Figure 1. A schematic representation of water droplets on two surfaces. Which surface is more hydrophobic?

Figure2. Waterdropletsbehavedifferentlyontiltedsurfaces because of the contact angle and contact angle hysteresis.

mg(sin R)/w ) çLV(cos ıR - cos ıA) (1)

10.1021/la00598o C: $19.0 © 200 American Chemical Society Published on Web 08/26/2000

c exhibits ıA/ıR ) 70°/70°. In case a, a 5 íL droplet was increasedinvolumeto10íL;thus,itintersectsthesurface at its advancing angle. In case b, a 15 íL droplet was decreasedinvolumeto10íL;thus,itintersectsthesurface at its receding angle. In case c, the 10 íL droplet is in equilibrium with ıA/ıR ) 70°/70°. When the surfaces are tilted, different events happen with each droplet. On

surface a, the downhill side of the droplet can advance, but the uphill side stays pinned until the receding angle is reached.This 2-dimensionalrepresentationoversimplifies the situation. There is a barrier to advancing due to the change in shape of the droplet; the volume stays constantbuttheair-waterinterfacialareaincreases.On surfaceb,theuphillsideofthedropletcanrecede,butthe downhill side stays pinned until the advancing angle is reached. Again there is a barrier to receding due to the change in shape of the droplet. On surface c, the droplet can advance and recede simultaneously with no change in the droplet shape.

This discussion, thus far, offers little new insight into wettability or hydrophobicity as these issues were discussed (in some form) in the literature decades ago,2,3-9 but it is apparently necessary as indicated by numerous recent reports10-19 of surfaces described as “super water repellant”,“ultrahydrophobic”,and“ultrawaterrepellant” that report only one contact angle.

Another issue that is of practical importance on real surfacesisthethree-phase(solid-liquid-air)contactline structure (shape, length, continuity of contact, amount of contact; these are both topological and topographical concerns). When liquid moves in contact with a surface, for instance water moving through a pipe, the velocity at the surface is 0. This is the no-slip boundary condition of fluid mechanics.20 Therefore, when a droplet moves on a surface,theonlysolid-liquidinterfacialwatermolecules thatmovearethoseonthecontactline.Onmostmaterials, a droplet placed on the surface will come to rest at a local energy minimum (due to either chemical structure or topography), the contact line will be fixed, and there will be energy barriers for advancing and receding; these are the causes of hysteresis. We recently21,2 described two approachestodestabilizingthecontactline.Oneinvolves preparing randomly rough hydrophobic surfaces by varioustechniquesthatformcontorteddiscontinuouscontact lines with water droplets. The other involves preparing smooth surfaces with flexible, liquidlike covalently at-

(3) Shuttleworth,R.;Bailey,G.L.Discuss.FaradaySoc.1948,3,16. (4) Bikerman, J. J. J. Phys. Chem, 1950, 54, 653. (5) Good, R. J. J. Am. Chem. Soc. 1952, 74, 504. (6) Schwartz, A. M.; Minor, F. W. J. Colloid Sci. 1959, 14, 584. (7) Johnson, R. E., Jr.; Dettre, R. H. Adv. Chem. Ser. 1963, 43, 112. (8) Dettre, R. H.; Johnson, R. E. Wetting; SCI Monograph No. 25;

Society of Chemical Industry; London, 1967; p 144. (9) Wolfram, E.; Faust, R. In Wetting, Spreading and Adhesion,

Padday, J. F., Ed.; Academic Press: London, 1978; p 213. (10) Onda, T. Shibuichi, S.; Satoh, N.; Tsujii, K. Langmuir 1996, 12, 2125. (1) Shibuichi, S.; Onda, T.; Satoh, N.; Tsujii, K. J. Phys. Chem. 1996, 100, 19512. (12) Tadanaga, K.; Katata, N.; Minami, T. J. Am. Ceram. Soc. 1997, 80, 1040. (13) Tadanaga, K.; Katata, N.; Minami, T. J. Am. Ceram. Soc. 1997, 80, 3213. (14) Veeramasuneni,S.;Drelich,J.;Miller,J.D.;Yamauchi,G.Prog.

Org. Coat. 1997, 31, 265. (15) Ogawa, K.; Soga, M.; Takada, Y.; Nakayama, I. Jpn. J. Appl.

Phys., Part 2 1993, 32, L614. (16) Kunugi, Y.; Nonaku, T.; Chong, Y. B.; Watanabe, N. J. Electroanal. Chem. 1993, 353, 209. (17) Schakenraad, J. M.; Stokroos, I.; Bartels, H.; Busscher, H. J.

Cells Mater. 1992, 2, 193. (18) Hozumi, A.; Takai, O. Thin Solid Films 1997, 303, 2. (19) Miller, J. D.; Veeramasuneni, S.; Drelich, J.; Yalamanchili, M. R.; Yamauchi, Y. Polym. Eng. Sci. 1996, 36, 1849.

(20) Bird,R.B.;Stewart,W.E.;Lightfoot,E.N.TransportPhenomena;

Wiley: New York, 1960; p 37. (21) Chen, W.; Fadeev, A. Y.; Hsieh, M. C.; O‹ ner, D.; Youngblood, J.

P.; McCarthy, T. J. Langmuir 1999, 15, 3395. (2) Youngblood, J. P.; McCarthy, T. J. Macromolecules 1999, 32, 6800.

Figure3. 2-Dimensional(X-Y)representationsoftwosurfaces with the same f1 and f2 values, but very different contact line structures. The dark lines are meant to represent possible contact lines.

Figure 5. Representative SEM image of a surface described inFigure4.Thissurfacecontains2ím 2ím 100ímsquare posts.

7778 Langmuir, Vol. 16, No. 20, 2000 O‹ner and McCarthy

tached monolayers. Rotating groups on these surfaces move the contact line. In both cases the contact line is unstable; thus, the droplet is constantly advancing and receding at different contact line points. Droplets do not come to rest on these surfaces, or they move very easily.

One of the objectives of the work reported in this paper istoresolvetheissueofthesizescaleofroughnessneeded to impart ultrahydrophobicity. The more recent reports of anomalously high contact angles10-19 have indicated that micrometer-, submicrometer, and nanometer-scale roughness imparts this property. Older reports8,23,24 from the 1940s, 1950s, and 1960s and wettability data on naturalwater-repellantsurfaces,25however,suggestthat much larger features (tens to hundreds of micrometers) can also function in this manner. It is not clear from the literature whether the surfaces with high length scale roughness also have lower length scale topographies superimposed on the larger features. We have prepared a series of silicon oxide surfaces by photolithography that haveroughnessondifferentlengthscales,hydrophobized these surfaces using reactive silane chemistry, and measured the wettability using dynamic contact angle analysis. A recent report by Bico et al.26 describes the wettability of three surfaces, one of which is similar to some of those reported here.

Experimental Section

General Procedures. Ethanol, toluene, sulfuric acid, and hydrogen peroxide (30%) were used as received from Fisher. Organosilanes were obtained from Gelest and used as received. House-purifiedwater(reverseosmosis)wasfurtherpurifiedusing a Millipore Milli-Q system that involves reverse osmosis, ion exchange, and filtration steps (1018 ¿/cm). Contact angle measurements were made using a Rame-Hart telescopic goniometer with a 24-gauge flat-tipped needle; dynamic advancing and receding angles were recorded as the probe fluid, water, purified as described above, was added to and withdrawn from thedrop,respectively.Thevaluesreportedareaveragesofgreater thaneightmeasurementsmadeondifferentareasofthesample surface. The modified surfaces exhibited very homogeneous surfacesasevidencedbythecontactangle,andallmeasurements forallsurfaceswerewithin(2°oftheaverages.Scanningelectron micrographswereobtainedusinganAmray1803TCinstrument with an accelerating voltage of 25 kV.

Preparation of Silicon Substrates. Silicon wafers (4 and 3 in.) were obtained from the International Wafer Service (〈100〉 orientation,P/B-doped,resistivityfrom20to40 ¿cm,thickness from 450 to 575 ím). A contact lithographic mask (with hexagonallyarrayedsquarepostsof16,32,64,and128ímlength and width) was constructed by Photronics Inc. The other masks were designed using a CAD program and prepared with a GCA PG3600F optical pattern generator. Photolithography was used to transfer the patterns of the masks onto the silicon wafers. Afterirradiation,thewaferswereetchedusingaPlasmaTherm SLR-770 for different durations. When the etching process was complete, the wafers were cleaned using a Branson IPC P2000 Barrel Etcher. Wafers were then placed in a solution of ammonium hydroxide, hydrogen peroxide, and water (4:1:1) for 15 min, rinsed with copious amounts of water, and spin dried. The wafers were then cut into 1.5 1.5 cm pieces, placed in a custom-designed (slotted hollow glass cylinder) sample holder, andcleanedbysubmersionintoamixtureofconcentratedsulfuric acid and hydrogen peroxide (30%) (7:3) overnight. The wafers were then rinsed with copious amounts of purified water and dried in a clean oven at 120 °C for 2 h immediately prior to silanization reactions.

ReactionofSiliconSubstrateswithOrganosilanes.The silicon substrates were placed in a custom-designed (slotted hollow glass cylinder) sample holder, which was then placed in a flask containing 0.5 mL of organosilane reagent: dimethyldichlorosilane(DMDCS),n-octyldimethylchlorosilane(ODMCS), orheptadecafluoro-1,1,2,2-tetrahydrodecyldimethylchlorosilane (FDDCS).Thewaferswerenotincontactwiththeliquidsilanes. The vapor-phase reactions were carried out for 3 days at 65-70 °C. The hydrophobized wafers were rinsed with toluene (two aliquots), ethanol (three aliquots), 1:1 ethanol/water (two aliquots), water (two aliquots), ethanol (two aliquots), and then water(threealiquots)andwerethendriedinacleanovenat120 °C for 30 min.

(23) Cassie, A. B. D.; Baxter, S. Trans. Faraday Soc. 1944, 40, 546. (24) Bartell, F. E.; Shepard, J. W. J. Phys. Chem. 1953, 57, 211. (25) Meinhuis, C.; Barthlott, W. Ann. Bot. (London) 1997, 79, 667. (26) Bico, J.; Marzolin, C.; Quere, D. Europhys. Lett. 1999, 47, 220.

Table 1. Water Contact Angle Data for Silane-Modified Hexagonally Arrayed 40 ím High Square Post Surfaces Described in Figures 4 and 5

DMDCS-modified ODMCS-modified FDDCS-modifiedsilicon surface ı (deg) ı (deg) ı (deg) ı (deg) ı (deg) ı (deg)

Table 2. Water Contact Angle Data for

ODMCS-Modified Surfaces Containing 16 ím 16 ím and 32 ím 32 ím Square Posts of Different Heights siliconsurface post height (ím) ıA(deg) ıR (deg)

Figure 6. 2-Dimensional (X-Y) representations of surfaces containing different geometry posts: RP, staggered rhombus posts; StP, four-armed star-shaped posts; IP, indented square posts.

Ultrahydrophobic Surfaces Langmuir, Vol. 16, No. 20, 2000 7779

Results and Discussion

Wehavediscussedintwopublications,21,22whysurfaces containing hydrophobic posts should form discontinuous and unstable three-phase contact lines with water droplets. The posts must be close enough together and hydrophobicenoughthatwaterdoesnotintrudebetween them. Surfaces such as these are referred to as composite surfaces as the intersection of the water droplet with the surface consists of a composite mixture of water-solid interface area and water-air interface area. The wettability of such surfaces was first addressed by Cassie,2 and eq 2 was derived that relates the contact angle of a liquid on a composite (air-solid mixture) surface (ıC)t o the contact angle on a smooth surface of the same solid

(ı) and the fraction of water-solid interface area (f1) and water-air interface area (f2), f1 + f2 ) 1. We stress that this equation is derived for a drop at equilibrium on a surface;thus,thecontactanglespredictedare“equilibrium contact angles”. We argue above that the structure of the three-phase contact line is important to dynamic wetta- bility (ıA, ıR, and hysteresis), and Cassie’s analysis does not take the three-phase contact line structure into account. Surfaces with many different topographies can

havethesamef1andf2values,butcanhaveverydifferent contact line structures. Figure 3 shows representations of two surfaces with the same f1 and f2 values. Cassie’s equation predicts the same equilibrium contact angles, but the advancing and receding contact angles will be verydifferentonthesesurfacesbecause,incasea,anearly continuous contact line can form (pinning the drop), but in case b, the contact line is discontinuous and unstable.

Figure 4 is a 2-dimensional schematic representation of one series of silicon oxide surfaces that was prepared using photolithographic techniques. These surfaces are hexagonally arrayed square posts varying in size from 2 ím 2 ímt o1 28 ím 128 ím. Different post heights (20,40,60,80,100,and140ím)werepreparedbyvarying theetchingtimefortwodifferentsizeposts.Thepostsare spaced as indicated in the figure. Figure 5 is a representative scanning electron microscopy (SEM) image of one of the surfaces (2 ím 2 ím 100 ím posts). These surfaces were oxidatively cleaned (to remove any of the lithography mask and/or etching chemicals) and chemically modified by reaction with silanization reagents in the vapor phase. Reaction conditions were chosen to give dense monolayers.27

(Parte 1 de 2)