Hydrophobicity and Freezing of a Water Droplet on Fluoroalkylsilane

Jul 18, 2007 - Department of Metallurgy and Ceramic Science, Graduate School of Science and Engineering, Tokyo Institute of Technology, 2-12-1 O-okaya...
1 downloads 26 Views 182KB Size
8674

Langmuir 2007, 23, 8674-8677

Hydrophobicity and Freezing of a Water Droplet on Fluoroalkylsilane Coatings with Different Roughnesses Shunsuke Suzuki,†,‡ Akira Nakajima,*,†,‡ Naoya Yoshida,‡,§ Munetoshi Sakai,‡ Ayako Hashimoto,‡ Yoshikazu Kameshima,†,‡ and Kiyoshi Okada† Department of Metallurgy and Ceramic Science, Graduate School of Science and Engineering, Tokyo Institute of Technology, 2-12-1 O-okayama, Meguro-ku, Tokyo 152-8552, Japan, Kanagawa Academy of Science and Technology, 308 East, Kanagawa Science Park, 3-2-1 Sakado, Takatsu-ku, Kawasaki-shi, Kanagawa 213-0012, Japan, and Center of CollaboratiVe Research, The UniVersity of Tokyo, 4-6-1 Komaba, Meguro-ku, Tokyo 153-8904, Japan ReceiVed April 13, 2007. In Final Form: June 12, 2007 Smooth (Ra ≈ 0.1 nm) and rough (Ra ≈ 20 nm) coatings of 1H,1H,2H,2H-perfluorodecyltrimethoxysilane (FAS17) were prepared by controlling process conditions. The water contact angles for the smooth and rough coatings were similar (107° and 110°, respectively), but their sliding angles differed considerably (10° and 27°, respectively). The surface potential on the smooth coating, assessed using Kelvin force microscopy, showed a sharp distribution, but that on the rough coating ranged widely, implying large chemical heterogeneity including residual SiOH groups. The freezing temperature of a supercooled water droplet on the rough coating was higher than that on the smooth coating.

I. Introduction Hydrophobic coatings have attracted much attention as an indispensable technology for the production of various industrial items.1 The contact angle has often been used as a criterion for evaluating the static hydrophobicity of solid surfaces. For assessing the dynamic hydrophobicity of solid surfaces, the sliding angle (the critical angle at which a water droplet of a certain weight begins sliding downward) remains a widely used criterion.2-5 Although the contact angle is known to depend on both surface roughness and surface energy,6,7 the sliding angle is also sensitive to those factors, especially on hydrophobic silane coatings. Morimoto et al. reported that even nanometer-level roughness drastically increases the water droplets’ sliding angles.8 Song et al. demonstrated that micrometer-scale in-plane hydrophilic defects increase the sliding angle of the hydrophobic silane coating.9 Increased surface roughness caused by the silane coating occurs because of the silane molecules’ self-condensation. Two reasons explain the increase of the sliding angle by the surface roughness of the silane coating: (i) the increased practical retention force against the droplet motion, such as slipping or rolling attributable to surface physical roughness, and (ii) the pinning of a droplet caused by the exposure of small hydrophilic surface regions such as those of SiOH. Water, which is easily supercooled to less than 0 °C, can be supercooled to approximately -45 °C under an almost zero * To whom correspondence should be addressed. E-mail: anakajim@ ceram.titech.ac.jp. † Tokyo Institute of Technology. ‡ Kanagawa Academy of Science and Technology. § The University of Tokyo. (1) Nakajima, A. J. Ceram. Soc. Jpn. 2004, 112, 533. (2) Carre, A.; Shanahan, M. E. R. J. Adhes. 1995, 49, 177. (3) Wolfram, E.; Faust, R. Wetting, Spreading, and Adhesion; Padday, J. F., Ed.; Academic Press Inc.: London, 1978; Chapter 10. (4) Youngblood, J. P.; McCarthy, T. J. Macromolecules 1999, 32, 6800. (5) Murase, H.; Nanishi, K.; Kogure, H.; Fujibayashi, T.; Tamura, K.; Haruta, N. J. Appl. Polym. Sci. 1994, 54, 2051. (6) Wenzel, R. N. J. Phys. Colloid Chem. 1949, 53, 1466. (7) Cassie, A. B. D. Discuss. Faraday Soc. 1948, 3, 11. (8) Morimoto, T.; Sanada, Y.; Tomonaga, H. Thin Solid Films 2001, 392, 214. (9) Song, J.-H.; Sakai, M.; Yoshida, N.; Suzuki, S.; Kameshima, Y.; Nakajima, A. Surf. Sci. 2006, 600, 2711.

gravity situation.10 This temperature is assumed as the freezing temperature of supercooled water by homogeneous ice nucleation. For a water droplet on a solid surface, ice nucleation should occur by a heterogeneous nucleation mechanism at the solidwater interface.11 Surface physical roughness and its chemical homogeneity are assumed to play important roles in ice nucleation behavior. However, the freezing of supercooled water droplets on hydrophobic surfaces of different roughnesses by silane coatings has not been well investigated so far. Very recently, we developed processing conditions for the roughness control of self-assembled monolayer coatings on a silicon surface using various silanes.12 For the present study, we prepared fluoroalkylsilane coatings with different surface roughnesses by choosing two typical sample preparation conditions. We then examined the relationship among the freezing temperatures of a supercooled water droplet, the solid surface structure, and its hydrophobicity. II. Experimental Section II.1. Sample Preparation. The surface of a Si(100) wafer was cleaned in acetone using ultrasonication and vacuum ultraviolet illumination (172 nm wavelength, UER-20; Ushio Inc., Tokyo, Japan) for 10 min in air at room temperature.13 The precleaned wafer was immersed into 1H,1H,2H,2H-perfluorodecyltrimethoxysilane (FAS17, TSL8233; GE Toshiba Silicones, Tokyo, Japan) solutions. A smooth coating was obtained through immersion into bis(trifluoromethyl)benzene (F6Xy; Wako Pure Chemical Industries Ltd.) with 130 µM FAS-17 for 1 day. The rough coating was prepared by immersion of the wafer into xylene (Wako) with 25 mM FAS-17 for 1 day. After soaking, the sample surface was rinsed using methylene chloride, acetone, and water and then dried at 80 °C. II.2. Evaluation. The surface topography and surface potential were observed using atomic force microscopy (AFM) and Kelvin (10) Conrad, P.; Ewing, E. G.; Karlinsey, L. R.; Sadtchenko, V. J. Chem. Phys. 2005, 122, 064709. (11) Popovitz-Biro, R.; Wang, L. J.; Majewski, J.; Shavi, E.; Leiserowitz, L.; Lahav, M. J. Am. Chem. Soc. 1994, 116, 1179. (12) Nakajima, A. Final proceedings of the research of super hydro-philic/ phobic surface project; Kanagawa Academy of Science and Technology: Kanagawa, Japan, 2007 (in Japanese). (13) Sugimura, H.; Ushiyama, K.; Hozumi, A.; Takai, O. Langmuir 2000, 16, 885.

10.1021/la701077p CCC: $37.00 © 2007 American Chemical Society Published on Web 07/18/2007

Letters

Langmuir, Vol. 23, No. 17, 2007 8675

Figure 2. Histogram numbers of CPD evaluated from the ultrasmooth and rough FAS-17 coatings on Si from KFM images.

Figure 1. (a and b) AFM image and KFM image for the same area (2 µm square) of the Si surface treated with FAS-17 in a F6Xy solution for 1 day. (c and d) AFM image and KFM image of the Si surface treated in a xylene solution for 1 day. force microscopy (KFM) (JSPM-4200; JEOL, Tokyo, Japan) using a Pt-Ti coated Si probe (NSC36-b, 1.75 N and NSC36-a, 0.95 N/m for silane coating, and NSC35-b, 14.0 N/m for Si substrate; all probes were provided by µ-mash Ltd., Narva mmt. Estonia). A modulation bias voltage (25 kHz, 1 V) was applied between the probe and the sample during KFM measurements. AFM operation during the KFM imaging process was conducted in amplitude detection mode using the noncontact mode. The water contact angle and sliding angle were measured using commercial measurement systems (Dropmaster 500 or SA-11; Kyowa Interface Science Co. Ltd.). Five different contact angles were measured for a 4 mg water droplet using the sessile drop method. Sliding angles were measured at three different points for a 30 mg droplet and then averaged. The standard deviation of these angle measurements was within 1 deg. The freezing temperature of a supercooled water droplet on the sample surface was investigated using differential scanning calorimetry (DSC, Q100; TA Instruments, New Castle, DE, USA). A 2.5 mg water droplet was placed on the coating in the sample cell, which consisted of an aluminum cup and lid. The sample was then cooled from room temperature to -35 °C at 0.5 °C/min under flowing N2, and the onset temperature of freezing was measured. Temperature tracking of the sample cell interior was confirmed using the Hg melting point.

III. Results and Discussion Figure 1a and b shows AFM and KFM images (2 µm square) of the sample surface treated by FAS-17 in a F6Xy solution for 1 day. A smooth surface was obtained with average surface roughness (Ra) of 0.1 nm. Figure 1c and d shows similar images on the sample surface treated with FAS-17 in a xylene solution for 1 day. A rough granular structure comprising silane clusters was obtained, with heights ranging widely from 10 to 150 nm. The average surface roughness of this heterogeneous surface was revealed as 20 nm. The water contact angles (WCA) for the smooth and rough coatings were 107° ( 1° and 110° ( 1°, respectively. Further experiments revealed that the apparent water contact angle was almost saturated in 1 day in both solvent cases (F6Xy and xylene). The rough coating had a slightly higher contact angle than the smooth coating. The water contact angles on the smooth surface are commonly described using Young’s equation:

cos θ ) (γSV - γSL)/γLV where γSL, γSV, and γLV respectively denote the interfacial free

energies per unit area of the solid-liquid, solid-gas, and liquidgas interfaces. Wenzel modified Young’s equation to describe the contact angle θ′ on a rough surface.6

cos θ′ ) r(γSV - γSL)/γLV ) r cos θ where r is the roughness factor, defined as the ratio of the actual area of a rough surface to the geometric projected area. The practical roughness factor on this rough coating was obtained as 1.1 using AFM software (WinSPM Data Processing 2.14; JEOL, Japan). Cassie proposed an equation describing the contact angle θ′ at a heterogeneous surface comprising two materials. The contact angle on the surface is calculated using the following equation7 when a unit area of the surface has a surface area fraction f1 with a contact angle θ1 and an area fraction f2 ( ) 1 - f1) with a contact angle θ2:

cos θ′ ) f1 cos θ1 + f2 cos θ2 In a xylene solution, self-condensation of FAS-17 molecules occurs at a higher rate than that of F6Xy.14 We confirmed that surface roughness increases even after the saturation of the contact angle during coating. This behavior is remarkable when coating is performed using a xylene solution. These results imply that the FAS-17 molecules form larger clusters, leaving the residual SiOH groups on the coating surface in xylene. This part is hydrophilic, and the contact angle obtained from the rough coating includes Cassie’s effect. However, such hydrophilic SiOH is expected to be exposed on the surface of the rough coating in the minor part; the major area of the surface might be covered by hydrophobic moieties (CF groups). The respective sliding angles for the smooth and rough coatings are 10° ( 1° and 27° ( 1°. Reproducibility of the coatings’ properties was confirmed. As expected, the rough surface exhibited a higher sliding angle; this difference is more pronounced than that for WCA. Decker et al. demonstrated experimentally that in-plane defects contort the three-phase (solid-liquid-air) contact line even when their size is of the micrometer level.15 Even when the defects’ sizes and their in-plane fractions are small, they effectively pin the three-phase contact line and decrease its mobility. In-plane heterogeneity is more sensitive to the sliding angle than WCA. The distribution of the contact potential difference (CPD) from each KFM image is shown in Figure 2. This figure was obtained by counting the histogram number range for ∼0.002 V for the Kelvin probe image (2.0 × 2.0 µm2). The CPD values for the smooth FAS-17 coating show a sharp peak around -230 mV; (14) Yoshida, N.; Suzuki, S.; Song, J.-H.; Sakai, M.; Kameshima, Y.; Nakajima, A. Proceedings of the Annual Meeting of the Ceramic Society of Japan, Okayama, Japan, 2005; p 154 (in Japanese). (15) Decker, E.; Garoff, L. Langmuir 1997, 13, 6321.

8676 Langmuir, Vol. 23, No. 17, 2007

Letters

Figure 4. Onset points of freezing of a 2.5 mg supercooled water droplet on a plain Si surface and on smooth and rough FAS-17 coatings from DSC measurements.

Figure 3. (a) AFM image (1 µm square) of the Si surface treated with FAS-17 in a F6Xy solution for 10 days. (b) Height (z) profile along with the vector line shown in (a). (c) KFM image obtained at same time as (a). (d) CPD profile along with the vector line shown in (c) (same position as (a)).

the rough coating peak is broad. The sharp CPD peak on the smooth coating implies that the distribution of the surface chemical composition is homogeneous. Once plain (no coating) Si was irradiated by vacuum ultraviolet (λ ) 172 nm) in air for 10 min, the surface potential came to be around 840 mV (the peak sharpness was equivalent for smooth FAS-17), suggesting that the generated SiOH had a larger surface potential. A CPD histogram of the rough coating suggests that both the chemical composition and its distribution are less homogeneous than those of the smooth coating. Under a lower concentration of FAS-17 (130 µM), the molecules react with the surface rather than self-condense, yielding a smooth coating. However, for a long soaking time, for example, 10 days, FAS-17 molecules self-condense gradually and form clusters on the coating surface. Such silane clusters are visible by the granules in Figure 3. The position of C in Figure 3 has a higher CPD than the other granules, indicating

a slight amount of residual SiOH exposed at the cluster surface. The CPD at C was negative because the KFM image and CPD value were obtained by averaging the surface potential values at each position (the resolution was 256 × 256 counts for a 1 µm square). The observed CPD should be negative if a silane cluster has a large amount of fluorocarbon groups in that body. Such an exposed SiOH should be hydrophilic; the sample surface’s chemical heterogeneity would thereby increase. Details related to the dependence of surface roughness and chemical homogeneity on processing conditions will be described elsewhere. Figure 4 shows the heat flow and freezing behavior of the DSC measurements of a supercooled water droplet on the surface of the smooth and rough FAS-17 coatings (Figure 1 shows the morphologies) and on a plain Si surface. The freezing temperatures of a supercooled water droplet on these surfaces were -22.7 °C (smooth), -21.5 °C (rough), and -16.3 °C (plain Si). The DSC measurement accuracy in this study is within 0.1 °C; therefore, the difference in freezing temperature of the smooth and rough coatings is considerable. These temperatures were higher than -45 °C, suggesting that freezing occurred by a heterogeneous nucleation mechanism. Additionally, we observed the freezing behavior of a supercooled water droplet on the surface of Si coated with FAS-17 using a high-speed camera system, thereby confirming that ice nucleation occurred from the solid-liquid interface and that the freezing temperature order was the same. Using a plain smooth Al plate, we examined the droplet mass dependence of the freezing temperature by an increase of the droplet mass to 30 mg, which corresponds to a greater than 4 times increase of the solid-liquid interface. We confirmed that the freezing temperature increased by only 0.8 °C. The KFM results show that the surface roughness and chemical heterogeneity are not independent of the coating of FAS-17. Probably, some residual SiOH groups exist in the rough coating because of selfcondensation or micelle formation. The practical freezing temperature difference between the rough coating and smooth one was 1.2 °C, although the roughness factor of the rough coating was 1.1. These results imply that the difference in freezing temperature between the smooth and rough coatings is not mainly attributable to different numbers of nucleation sites (apparent surface roughness); it is caused by the nucleation behavior difference (surface chemical heterogeneity). For rough coatings, the solid-water interface must be covered with a mixture of CF groups and SiOH groups. Both groups exhibit different interactions with water molecules;16-18 this heterogeneity decreases (16) Umeyama, H.; Morokuma, K. J. Am. Chem. Soc. 1977, 99, 1316. (17) Claesson, P. M.; Christenson, H. K. J. Phys. Chem. 1988, 92, 1650.

Letters

Langmuir, Vol. 23, No. 17, 2007 8677

the stability of water molecules at the interface and induces ice nucleation at higher temperatures than those on the smooth coating.

IV. Conclusion For this study, we prepared smooth and rough fluoroalkylsilane (FAS-17) coatings by controlling the respective process conditions. The water contact angles for the smooth and rough coatings were nearly identical. Nevertheless, the difference of the sliding (18) Saengsawang, O.; Remsungnen, T.; Fritzsche, S.; Haberlandt, R.; Hannongbua, S. J. Phys. Chem. B 2005, 109, 5684.

angle was considerable. The distribution of surface potential on the smooth coating was sharp, but that on the rough coating ranged widely, suggesting great chemical heterogeneity with the residual SiOH groups. The freezing temperature of a supercooled water droplet on the rough coating was higher than that on the smooth coating because of the lower stability of the solid-water interface caused by chemical heterogeneity. Acknowledgment. This work was supported partly by JSPS Research Fellowship No. H17-08586. LA701077P