Purdue E-Pubs

Purdue University Purdue e-Pubs

Birck and NCN Publications Birck Nanotechnology Center

7-6-2011 Single-Step Fabrication and Characterization of Ultrahydrophobic Surfaces with Hierarchical Roughness Susmita Dash Birck Nanotechnology Center, Purdue University, [email protected]

Niru Kumari Birck Nanotechnology Center, Purdue University, [email protected]

Mercy Dicuangco Birck Nanotechnology Center, Purdue University, [email protected]

Suresh V. Garimella Birck Nanotechnology Center, Purdue University, [email protected]

Follow this and additional works at: http://docs.lib.purdue.edu/nanopub Part of the Nanoscience and Nanotechnology Commons

Dash, Susmita; Kumari, Niru; Dicuangco, Mercy; and Garimella, Suresh V., "Single-Step Fabrication and Characterization of Ultrahydrophobic Surfaces with Hierarchical Roughness" (2011). Birck and NCN Publications. Paper 1507. http://dx.doi.org/10.1115/IPACK2011-52046

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information. Proceedings of the ASME 2011 Pacific Rim Technical Conference & Exposition on Packaging and Integration of Electronic and Photonic Systems InterPACK2011 July 6-8, 2011, Portland, Oregon, USA

InterPACKIPACK202011-11-5205204646

SINGLE-STEP FABRICATION AND CHARACTERIZATION OF ULTRAHYDROPHOBIC SURFACES WITH HIERARCHICAL ROUGHNESS

Susmita Dash, Niru Kumari, Mercy Dicuangco and Suresh V. Garimella Cooling Technologies Research Center, an NSF IUCRC School of Mechanical Engineering and Birck Nanotechnology Center Purdue University West Lafayette, Indiana 47907-2088 USA [email protected]

ABSTRACT   , thereby controlling the extent of the liquid-air interface. Hydrophobic surfaces with microscale roughness can be Sub-micron scale roughness coupled with micro-level rendered ultrahydrophobic by the addition of sub-micron scale roughness can render a surface ultrahydrophobic and impart roughness. A simple yet highly effective concept of fabricating improved non-wetting properties in comparison to single-tier hierarchical structured surfaces using a single-step deep roughness. Ultrahyrophobic surfaces are non-wetting surfaces reactive ion etch process is proposed. Using this method the characterized by high contact angles (>150⁰), a low sliding complexities generally associated with fabrication of two-tier angle, and low contact angle hysteresis [2-4]. Hierarchical roughness structures are eliminated. Experiments are roughness is commonly encountered in nature; the extreme conducted on two double-roughness surfaces with different water-repellent characteristic of lotus leaves arises from a surface roughness, achieved by varying the size of the double-roughness structure consisting of nanoscale waxes on microscale roughness features. The surfaces are characterized microscale bumps [5]. in terms of static contact angle and roll-off angle and compared with surfaces consisting of only single-tier microscale Surfaces with such extreme hydrophobicity have important roughness. The robustness of the new hierarchical roughness applications in the development of artificial self-cleaning surfaces is verified through droplet impingement tests. The surfaces and developing water-proof clothing [6] and offer a hierarchical surfaces are more resistant to wetting than the wide range of promising applications including their use in single roughness surfaces and show higher coefficients of microfluidic-based technologies such as lab-on-chip devices, restitution for droplets bouncing off the surface. The droplet microelectromechanical systems (MEMS), and microarray dynamics upon impingement are explored. biochips. An important heat transfer application consists of developing surfaces for dropwise condensation. Dropwise 1. INTRODUCTION condensation is desirable in many heat transfer applications since the heat transfer coefficient associated with dropwise The contact angle of a droplet when placed on a surface is condensation is an order of magnitude higher than that by determined by surface energy as well as surface morphology filmwise condensation. Dropwise condensation is, however, [1]. The morphology of the surface determines whether a not readily achieved on single-tier roughness structures [7]. It droplet will remain in a Cassie (non-wetting) or a Wenzel was recently demonstrated that condensation on hierarchical (wetting) state (Figure 1a and Figure 1b respectively). roughness structures leads to condensation in the form of drops Superhydrophobicity may be imparted to a surface by carefully [8]. Since hydrophobic surfaces resist the formation of a liquid engineering the surface topology and controlling the ratio of film, surface corrosion is also mitigated. areas of the top surface of the pillars to the total base surface

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms single-roughness substrates. Droplet impingement experiments are then carried out for the two double-roughness surfaces to test the robustness of their hydrophobicity. Droplet dynamics on the double-roughness surfaces are explored and compared with droplet behavior on surface with single-roughness elements.

2. SAMPLE PREPARATION AND EXPERIMENTAL SETUP (a) (b) The fabrication procedure developed in the present work eliminates the typical two-step process to create double Figure 1. Schematic illustration of droplet wetting states: roughness structures. The surfaces fabricated consist of silicon (a) Cassie, and b) Wenzel. pillars as the larger roughness element. Photoresist residue is formed on the pillars during the deep reactive ion etch process; The wide range of applications of hierarchical hydrophobic this residue is in the form of ~ 1 µm strands stacked on top of surfaces has encouraged active research in this field. Different the silicon pillars and provides the second-tier roughness. The methods of fabricating such hierarchical surfaces [2, 9] to attain advantage of this method is that a double-roughness surface is ultrahydrophobicity have been demonstrated. The fabrication obtained after a single deep reactive ion etch step. All of double-roughness structures typically involves the fabrication for this work was carried out at the Birck fabrication of the larger-sized features on a substrate (by Nanotechnology Center at Purdue University. standard lithography methods) followed by the deposition of smaller-sized roughness elements on these larger features [2, 9, Silicon wafers with 1 µm thermally grown oxide layers 10]. Efforts at fabricating and testing robust superhydrophobic were used as the substrates. A layer of positive photoresist AZ surfaces which can be easily fabricated and commercially used 1518 was spin-coated and lithographically patterned on the continue to be reported. wafer. A wet-etch process is used to selectively etch the oxide layer from the surface. The oxide layer along with the Analytical and experimental research has corroborated the photoresist acts as the etch mask for the deep reactive ion etch strong effect of surface morphology on the impact behavior of a process. The etching process results in the creation of silicon water droplet and its ability to bounce off the surface [6, 11- pillars. At the same time, the high temperature and ions 14]. Jung and Bhushan [6] demonstrated better water produced in the plasma interact with the photoresist and cause repellency on hierarchical surfaces as compared to single it to distort and form small roughness elements on the pillars roughness elements. They formulated an expression for the which lead to a second-tier roughness. The etch rate for silicon critical velocity of the droplet (based on the capillary pressure was observed to be approximately 4 µm per minute. A and Bernoulli pressure) beyond which it transitions to a Wenzel minimum of 4 minutes of etch time was required for the state on textured surfaces. Varanasi et al. [15] developed a formation of the second-tier roughness structures. The surfaces pressure-balance model to arrive at a condition for droplet are then spin coated with 0.1% solution of Teflon-AF 1600 infiltration into the air gap between the surface structures. (DuPont, Wilmington, DE) in FC-77 (3M, St. Paul, MN) to Denser textured surfaces were expected to provide greater impart hydrophobicity. The thickness of the Teflon layer is capillary pressure and superior resistance to Wenzel wetting of approximately 50 nm and hence the overall roughness of the impacting droplets. primary roughness as well as the sub-micron roughness is not affected by the Teflon coating. The single-roughness surfaces 1 The purpose of the present work is to explore a one-step and 2 used for comparison against the results from double- fabrication methodology for double-roughness surfaces and roughness surfaces are fabricated with the primary geometrical eliminate the complexity involved in producing the second parameters same as those for the double-roughness surfaces 1 layer of sub-micron roughness. Pillars of square cross section and 2 respectively using the negative photoresist SU-8 with are carefully engineered so that the single-roughness features standard lithography. The pillars are subsequently coated with inherently maintain the droplets in a stable Cassie state. The Teflon to impart superhydrophobicity. The pillar geometry of double-roughness surfaces are fabricated with the same primary the two double-roughness surfaces fabricated is provided in roughness as the single-roughness pillars using a single step Table 1. The table also shows the two parameters utilized to DRIE method. The properties of the double-roughness surfaces quantify the primary surface roughness, namely, phi fabricated in this work are compared to those of single- roughness features to estimate the hydrophobicity enhancement a2 4ah ,2 and rm ,1 2 , where a is the size of the imparted by the second layer of roughness elements. The static p p contact angle of a milliliter-sized sessile droplet and the roll-off square pillars, p is the pitch, and b is the size of the air gap angle of the droplet on such surfaces are experimentally between the pillars such that p a b . These parameters also determined and compared to the results from the corresponding

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms represent the roughness of the single-roughness surfaces. The angles measured at different locations of the substrate. The dimensions of the pillars are chosen such that a droplet deviation in static contact angle is within ± 2⁰. The roll-off inherently assumes a Cassie state when placed on these results are repeatable to within ± 3⁰ and representative values surfaces. Figure 2 shows SEM images of the two double- from one case each are presented. The roll-off experiments are roughness surfaces 1and 2. The roughness values (Ra) of the carried out on the double-roughness as well as the single- sub-micron roughness on the double-roughness surfaces 1 and roughness surfaces. 2 are on the order of 0.33 μm and 0.27 μm, respectively. Droplet impingement tests are carried out on the double Table 1. Parameters of the hierarchical surfaces. and the single-roughness surfaces to test the resistance of the surfaces to wetting under impact. Droplet impingement was

, ,

also tested with a smooth hydrophobic silicon substrate coated

h ) ) )

, ize )

itch with Teflon to provide a baseline for comparison. A single

p m μm

ϕ μm μm ( r micron ( (

b/a droplet of volume 3 μl is released from a height of 1 cm by μm -

a ( Pillar eight a p Surface R

h means of a high-precision automated dispensing system fitted Pillar s Roughness, Pillar Sub with a micro-syringe as shown in Figure 3. The impact dynamics of the droplet are visualized with a high-speed 1 27 42 32 0.41 0.56 2.96 0.33 camera (1024 Photron PCI) at 2000 frames per second. The images are subsequently analyzed using MATLAB [16] and 2 13 25 32 0.27 0.92 3.66 0.27 Image J, an image processing program available from the National Institutes of Health [17]. The advancing and the receding interface of the droplet, as well as the velocity at which the droplet leaves the substrate are tracked.

(a) Reservoir

Data Acquisition

Microsyringe

(b) Light Source

High Speed Substrate Diffuser Camera

Figure 2. SEM images of hierarchical surfaces a) surface 1, and Figure 3. Experimental setup for droplet impingement test. b) surface 2. The images to the right show the static shape of a 3. water droplet placed on the respective surface. RESULTS AND DISCUSSION 3.1. Deionized (DI) water droplets of volume 3 μl ± 0.1 μl are Static Contact Angle and Roll-off Angle used for all the characterization experiments. The static contact When a droplet gently placed on a substrate is in its Cassie angle of the droplets on the surfaces is measured using a state (Figure 1a), the static contact angle θc can be goniometer (Rame Hart). The goniometer is equipped with an approximated using the Cassie equation [1] as automatic tilt stage. For the droplet roll-off experiments, the stage is tilted at 0.8⁰ per sec with the droplet on top and images 1 (1) are simultaneously captured to analyze the advancing and the c cos  1  1  cos 0  receding contact angles and the roll-off angle (the tilt angle at which droplet motion is initiated) of the droplet on the where θ0 is the Young’s contact angle on the smooth surface substrate. Three sets of tests are carried out on each substrate. (measured to be 120⁰ for water on a smooth surface coated with The reported static contact angle is the average of the contact Teflon). The contact angle of the droplet on the single-tier

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms roughness alone of surface 1 and surface 2 as calculated from provides an estimate of the energy loss due to the Cassie equation would be 142⁰ and 149⁰, respectively. The impact/interaction with the structured surface. The double- actual contact angles on the single-roughness surfaces 1 and 2 roughness surfaces show a lower contact angle hysteresis in are measured to be 142⁰ ± 3⁰ and 147⁰ ± 3⁰, respectively. The comparison to the single-roughness surfaces with similar static contact angles with a water droplet on both double- surface parameters. The reduction in the contact angle roughness surfaces were measured to be 161⁰ ± 2⁰, showing a hysteresis is 16⁰ ± 2⁰ on the double roughness surfaces (Figure significant enhancement of hydrophobicity due to the presence 4b). The temporal evolution of the advancing and the receding of the second layer of roughness. contact angles on the double-roughness surfaces prior to roll off illustrates that the advancing contact angle remains almost The roll-off angle decreases with an increase in droplet fixed at its static contact angle value while the receding contact volume due to the increase in the gravitational force (mg) angle decreases till the gravitational force exceeds the net acting on the droplet. Droplet roll-off experiments are carried surface tension force so that the droplet rolls off (Figure 5). For out both on the double-roughness and single-roughness surfaces the single-roughness surface 2, the advancing contact angle to determine the reduction in the contact angle hysteresis as increases while the receding contact angle decreases before the well as enhancement of the roll-off characteristics of the droplet starts rolling. surface due to the second layer of roughness. The capillary 1/2 length of a water droplet defined as  is equal to 2.7 g  mm. The characteristic length scale (diameter) of the water droplet of volume 3 μl used in the experiments is approximately equal to 1.79 mm and is less than the capillary length. This implies that the effect of gravity can be considered negligible and the droplet assumed to be of spherical-cap shape. Also, this results in the surface forces being more dominant in comparison to the gravitational forces in determining the rolling tendency of the droplets.

On the single-roughness surfaces 1 and 2, the roll-off angle is observed to be very high. For surface 2 (b/a = 0.92), the roll- off angle is 51⁰ ± 3⁰, and for surface 1 (b/a = 0.56), the droplet did not roll off even at a very high inclination angle of 90⁰. For the very small droplets employed in the experiments, the (a) gravitational force is unable to overcome the surface tension force which holds the droplet on the surface. This is consistent with the observations of Varanasi et al. [15] who reported that for a b/a ratio less than 1, a 1 μl droplet did not roll off. The test was repeated for single-roughness surface 1 using a larger droplet volume of 5 μl. In this case the droplet did roll off, but again, at a very high roll-off angle of 37⁰. For single roughness surface 2, the roll-off angle reduced to 32⁰ when a droplet volume of 5 μl is used.

Experiments on the double-roughness surfaces showed lower roll-off angles. A 3 μl droplet rolled off at an inclination angle of 15⁰ from surface 1 and an angle of 9⁰ from surface 2. Thus the presence of the secondary roughness layer reduces the roll-off angle drastically for both the surfaces. Figure 4a shows a comparison between the roll-off angle of water droplets on (b) the single roughness and double roughness surfaces. It is noted that 3 μl droplets were used in all experiments, with the exception that a 5 μl droplet was used for the single-roughness Figure 4. (a) The roll-off angle, and (b) contact angle hysteresis surface 1. for the single-roughness surface (surface 1 and surface 2) and the hierarchical roughness surface (surface 1 and surface 2). Contact angle hysteresis refers to the difference between the advancing and the receding contact angles of a droplet, which depends upon the surface roughness/irregularities. It

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms retracting back on the substrates, in the contact time, and the droplet height attained after it bounces back.

Figure 7 shows the temporal variation of the wetted diameter of the droplet when the droplet is in contact with the surface. The droplet takes approximately 3.5 ms to reach its maximum wetted diameter for all the surfaces considered. The maximum wetted diameter is almost the same, and is approximately equal to 1.2 times the droplet diameter, irrespective of the substrate used as shown in Figure 7. However, the rate at which the interface of the droplet retracts before bouncing off the surface varies depending on the nature of the surface. There is a direct correlation between the contact angle hysteresis and the residence time of the droplet on the surface during impact. The lower the hysteresis, the lesser is the time that the droplet stays on the surface. The droplet stays for a much longer time on single-roughness surface than on the double-roughness surfaces, a resultant of higher energy loss on the single-tier rough surface. For tests with the 1 cm drop height, the droplet takes 14 ms to bounce off the single- roughness surface whereas the time is reduced to 11 ± 1 ms for Figure 5. Temporal evolution of advancing and receding the double-roughness surfaces (both 1 and 2). This difference contact angles during roll-off on double-roughness surfaces 1 is mainly attributed to the difference in energy loss as a result and 2 and single-roughness surface 1. of impact on the surfaces. As stated earlier, the hysteresis is greater when only one tier of roughness elements is present, 3.2. Droplet Impingement Dynamics with a correspondingly higher loss of energy. The contact time varies between surfaces and is slightly different from the A droplet impingement test is the most demanding test of characteristic time scale (based on the balance between inertia the water repellency of a surface under dynamic conditions. 1/2 R3 Droplet impingement tests are conducted on both single and and capillarity) given as (2.6 0.1) by Okumura double roughness surfaces with a fixed drop height of 1 cm.  The relative importance of the kinetic energy of the impinging et al. [18]. This is because of the energy loss due to contact droplet and the surface tension force may be compared using angle hysteresis is neglected in the derivation of the contact 2 time. A more sparse distribution of pillars than those the Weber number defined as We  VD where D is the  considered in the present work would yield contact times closer droplet diameter, V is the impinging velocity and γ is the to the characteristic time scale due to the corresponding surface tension. For the droplet impingement height of 1 cm, decrease in the contact angle hysteresis, as has also been considered in our experiments, We = 4. observed by Li et al. [19].

The behavior of the droplet upon impact can be understood The other difference between the single and double- based on two main stages. In the first stage, the droplet roughness surfaces is in the contact angle that the droplet interface advances to attain the maximum wetted diameter. interface makes with the substrate while retracting and the During this phase the kinetic energy of the droplet is stored as wetted diameter of the droplet just prior to detachment from the deformation energy in the droplet. In the second stage, the surface. The wetted diameter of the droplet for single droplet retracts and the stored energy helps it rebound off the roughness surfaces just before bounce-off is much smaller than surface. The first stage (spreading of the droplet) is an inertia that in case of the double-roughness surfaces. The wetted driven phenomenon. Subsequent retraction and bouncing of the diameter of the droplet prior to detachment from the double- droplet off the surface is the basic test for the water repellency roughness surfaces 1 and 2 and the single-roughness surface 1 of the surface. Figure 6 shows images of the droplet at are on the order of 0.4 mm and 0.1 mm, respectively. The different instants when the droplet impingement height is 1 cm. lower wetted diameter indicates the pinch-off phenomenon on Bouncing-off phenomenon was observed for both kinds of the single roughness surface (Figure 6a) which is not seen in surfaces (single and double roughness), but there is a difference case of the double-roughness surfaces. The droplet makes a observed in terms of the contact angle of the droplet while it is very high contact angle on the double-roughness surfaces 1 and 2 while retracting.

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms t = 0 ms t = 2.5 ms t = 4.5 ms t = 6 ms t = 11 ms t = 13 ms t =14 ms t = 14.5 ms Single-roughness surface 1

Pinch off Double-roughness surface 1

Double-roughness surface 2

Figure 6. Images of the droplet profile at different time instants upon impingement from height 1 cm on the (a) single-roughness surface 1, (b) double-roughness surface 1, and (c) double-roughness surface 2.

Figure 7. Variation of the wetted diameter of the droplet on the Figure 8. The maximum height attained by the droplet after single-roughness surface 1 and double-roughness surfaces 1 recoil from the surfaces when the droplet impingement height is and 2 corresponding to droplet impingement height of 1 cm. 1 cm.

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms h The coefficient of restitution is defined as COR  2 h REFERENCES 1 where h2 is the height to which the droplet bounces up and h1 is [1] He B., Patankar N. A., and Lee J., 2003, “Multiple the initial height from which the droplet is dropped. It is an equilibrium droplet shapes and design criterion for rough estimator of the energy retained by the droplet after impact with hydrophobic surfaces,” Langmuir, 19(12), pp. 4999- the substrate. The coefficient of restitution of the surfaces is 5003. used to quantitatively compare the reduction in energy loss of [2] Bhushan B., Koch K., and Jung Y. C., 2009, “Fabrication the droplet on the double-roughness surface. Figure 8 shows and characterization of the hierarchical structure for the maximum height attained by the droplet when it is dropped superhydrophobicity and self-cleaning,” from a height of 1 cm on to the double-roughness surfaces, Ultramicroscopy, 109(8), pp. 1029-1034. single-roughness and smooth surface (silicon wafer coated with [3] Ma M., and Hill R. M., 2006, “Superhydrophobic Teflon). The droplet height follows a parabolic profile with surfaces,” Current Opinion in Colloid and Interface respect to time after rebounding from the surface. The viscous Science, 11(4), pp. 193-202. dissipation is believed to be significant based on the observed [4] Verplanck N., Coffinier Y., Thomy V., and Boukherroub oscillation of the droplet in air after rebounding [14]. The COR R., 2007, “Wettability switching techniques on for double-roughness surfaces 1 and 2 is determined to be 0.5 superhydrophobic surfaces,” Nanoscale Research Letters, and 0.6 respectively; i.e., the droplet retains 50 percent and 60 2(12), pp. 577-596. percent of the energy through the impact on the two surfaces. [5] Neinhuis C., and Barthlott W., 1997, “Characterization As compared to the double-roughness surfaces, the single- and distribution of water-repellent, self-cleaning plant roughness surface 1 exhibits a COR equal to 0.4. The droplet surfaces,” Annals of Botany, 79(6), pp. 667-677. did bounce off from the smooth surface as well for a 1 cm [6] Jung Y. C., and Bhushan B., 2009, “Dynamic effects impingement height, but with a COR of 0.35. The higher induced transition of droplets on biomimetic coefficient of restitution on the double-roughness surfaces in superhydrophobic surfaces,” Langmuir, 25(16), pp. 9208- comparison to the single-roughness and the smooth surfaces is 9218. due to the reduction in the viscous dissipation of the droplet due [7] Wier K. A., and McCarthy T. J., 2006, “Condensation on to the high contact angle (lower contact angle hysteresis) during ultrahydrophobic surfaces and its effect on droplet droplet retraction. mobility: Ultrahydrophobic surfaces are not always water repellant,” Langmuir, 22(6), pp. 2433-2436. 4. CONCLUSION [8] Boreyko J. B., and Chen C. H., 2009, “Self-Propelled A simple one-step fabrication methodology for developing Dropwise Condensate on Superhydrophobic Surfaces,” ultrahydrophobic double-roughness surfaces is demonstrated. Physical Review Letters, 103(18), p. 184501. A single-step deep reactive ion etch method is presented. The [9] Ming W., Wu D., and Van Benthem R., 2005, second layer of roughness caused by the photoresist residue is “Superhydrophobic films from raspberry-like particles,” stable and is an easy way to enhance hydrophobicity of the Nano Letters, 5(11), pp. 2298-2301. surface. Significant improvement of the surface [10] Northen M. T., and Turner K. L., 2005, “A batch hydrophobicity is observed in terms of the non-wetting fabricated biomimetic dry adhesive,” Nanotechnology, characteristics when the second layer of roughness elements is 16, pp. 1159-1166. present as compared to the single-roughness surfaces. The [11] Wang Z., Lopez C., Hirsa A., and Koratkar N., 2007, static contact angle of a water droplet on the surface increases “Impact dynamics and rebound of water droplets on and there is a reduction in the contact angle hysteresis. The superhydrophobic carbon nanotube arrays,” Applied surfaces also withstand better, the impact pressure from an Physics Letters, 91, p. 023105. impinging water droplet (We = 4). The surfaces are further [12] Bergeron V., and Quéré D., 2001, “Water droplets make characterized in terms of the coefficient of restitution, which is an impact,” Physics World, May, pp. 27-31. approximately 0.6 for the double-roughness surfaces when the [13] Malouin Jr B. A., Koratkar N. A., Hirsa A. H., and Wang droplet impingement height is 1 cm. In ongoing work, the heat Z., 2010, “Directed rebounding of droplets by microscale transfer characteristics of such hierarchical surfaces are being surface roughness gradients,” Applied Physics Letters, experimentally determined. 96, p. 234103. [14] Richard D., and Quéré D., 2000, “Bouncing water ACKNOWLEDGEMENT drops,” Europhysics Letters, 50, pp. 769-775. [15] Varanasi K. K., Deng T., Hsu M. F., and Bhate N., 2008, Funding from the National Science Foundation for this “Design of superhydrophobic surfaces for optimum roll- work as a Fundamental Research Supplement to the Cooling off and droplet impact resistance,” Proceedings of the Technologies Research Center is gratefully acknowledged. The ASME International Mechanical Engineering Congress authors would like to thank John McHale for his assistance and Exposition, p. 67808. with the SEM images.

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms

[16] MATLAB, 2007, Reference Manual (Natick, MA: The Mathworks, Inc.). [17] Abramoff M. D., Magelhaes P. J., and Ram S. J., 2004, “Image processing with ImageJ,” Biophotonics International, 11(7), pp. 36-42. [18] Okumura K., Chevy F., Richard D., Quéré D., and Clanet C., 2003, “Water spring: A model for bouncing drops,” Europhysics Letters, 62, pp. 237-243. [19] Li X., Ma X., and Lan Z., 2010, “Dynamic Behavior of the Water Droplet Impact on a Textured Hydrophobic/Superhydrophobic Surface: The Effect of the Remaining Liquid Film Arising on the Pillars' Tops on the Contact Time,” Langmuir, 26(7), pp. 4831-4838.

Downloaded From: http://proceedings.asmedigitalcollection.asme.org/ on 12/17/2013 Terms of Use: http://asme.org/terms