Effect of Microtextured Surface Topography on the Wetting Behavior of

Article pubs.acs.org/Langmuir

Effect of Microtextured Surface Topography on the Wetting Behavior of Eutectic Gallium−Indium Alloys Rebecca K. Kramer,*,† J. William Boley,† Howard A. Stone,‡ James C. Weaver,§ and Robert J. Wood§,∥ †

School of Mechanical Engineering, Purdue University, West Lafayette, Indiana 47907, United States Department of Mechanical and Aerospace Engineering, Princeton University, Princeton, New Jersey 08544, United States § Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, Massachusetts 02115, United States ∥ School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts 02138, United States ‡

ABSTRACT: Liquid-embedded elastomer electronics have recently attracted much attention as key elements of highly deformable and “soft” electromechanical systems. Many of these fluid−elastomer composites utilize liquid metal alloys because of their high conductivities and inherent compliance. Understanding how these alloys interface with surfaces of various composition and texture is critical to the development of parallel processing technology, which is needed to create more complex and low-cost systems. In this work, we explore the wetting behaviors between droplets of gallium−indium alloys and thin metal films, with an emphasis on tin and indium substrates. We find that metallic droplets reactively wet thin metal foils, but the wettability of the foils may be tuned by the surface texture (produced by sputtering). The effects of both composition and texture of the substrate on wetting dynamics are quantified by measuring contact angle and droplet contact diameter as a function of time. Finally, we apply the Cassie−Baxter model to the sputtered and native substrates to gain insight into the behavior of liquid metals and the role of the oxide formation during interfacial processes.

1. INTRODUCTION Recent interest in highly deformable electronic components, such as flexible and stretchable sensors and circuits,1,2 has motivated new applications for Ga−In alloys. While semiconductor-based stretchable circuits exist,3 they are not appropriate for high current density or very high strain (>10%) applications. Ga−In alloys remain in the liquid phase for temperatures well below room temperature, are nontoxic, low viscosity, and highly conductive. When embedded in an elastomer film, Ga−In alloys function as highly stretchable electrical wiring4,5 or sensors that respond to deformation with changes in resistance.6−9 In particular, liquid-embedded elastomer conductors are currently fabricated via casting a low-modulus elastomer into a mold, sealing the molded features, and filling the embedded channels with Ga−In alloy using a needle and syringe. This method of prototyping is labor intensive and limits geometry, requiring an inlet and outlet to fill channels and eliminating the possibility of closed loops or islands. Continued development of liquid-embedded elastomer electronics requires a more automated and readily scalable fabrication process. Inspired by the ease and practicality of wave-soldering processes in printed circuit board production, we present experiments conducted with the goal of achieving selective wetting of Ga−In alloys on an elastomeric surface without sacrificing the inherent elasticity of the elastomer. Understanding this wetting behavior will equip researchers in the field © XXXX American Chemical Society

of liquid-embedded elastomer electronics with an additional way to control component feature sizes. Ga−In sessile droplets are tested on a silicone-based elastomer substrate coated with thin metal films (tin or indium; rolled or sputtered), from which both composition and texture are found to affect the wetting condition. This work complements several new investigations regarding manufacturing with Ga−In alloys, including vacuum-induced patterning,10,11 contact printing,12,13 and three-dimensional printing of free-standing columns and spheres.14 The wetting behavior observed in this work also complements previous work on wetting dynamics of molten droplets,15 as well as wettability and contact interaction of gallium-containing alloys with nonmetallic solids.16

2. EXPERIMENTAL SECTION In this study, two liquid metals are investigated: eutectic gallium− indium (eGaIn) and galinstan (GaInSn; see Table 1 for composition and properties). These eutectic liquids were chosen for their low melting temperatures and low resistivities17−19 in comparison with other conductive liquids, such as ionic liquids. For example, 1-ethyl-3methylimidazolium ethyl sulfate has an electrical resistivity of 2.51 Ωm,20 which is 7 orders of magnitude greater than that of eGaIn or galinstan. Received: November 11, 2013 Revised: December 16, 2013

A

dx.doi.org/10.1021/la404356r | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Table 1. Summary of Gallium−Indium Alloy Properties eutectic gallium− indium

galinstan

composition

75% Ga, 25% In

melting point bulk viscosity electrical resistivity surface tension

15.5 °C 1.99 × 10−3 Pa·s 2.94 × 10−7 Ωm

68.5% Ga, 21.5% In, 10% Sn, trace amounts of Cu −19 °C 2.4 × 10−3 Pa·s 2.89 × 10−7 Ωm

624 mN/m

718 mN/m

We tested the wettability of Ga−In droplets on thin metal films of indium and tin by measuring both contact angle and droplet contact diameter as a function of time using the sessile drop method. Contact angle values were obtained through experiments on an Attension Theta Lite Tensiometer (Biolin Scientific; Linthicum Heights, MD) and image analysis with Attension software set to “polynomial” mode. We chose to experiment with indium and tin substrates because of anticipated compatibility. All wettability experiments were conducted at room temperature. The sputtering of indium and tin occurred in an argon environment. Samples were allowed to fully cool to room temperature before removal from the sputtering chamber. Sputter targets were obtained from American Elements, noted to be 99.999% pure. The sputtering chamber was used exclusively for indium and tin deposition during the time of these experiments and was thoroughly cleaned between each use. Furthermore, energy-dispersive spectroscopy (EDS) confirmed no detectable surface contaminant. Therefore, we have no reason to believe that any surface contaminant would account for more than 0.0001% of the total composition, below the amount detectable via EDS. The sputtering condition (chamber pressure and current) and the melting temperature of the sputter target play a large role in the roughness of the sputtered surface. In general, sputtering at higher currents will allow clustering and increase surface roughness of the deposited film. Low-melting-temperature metals are especially sensitive to current adjustments. Therefore, rough surface morphology is expected when sputtering a low-melting-temperature metal (such as In or Sn) at a relatively high current. In our case, we applied 100 mA in 30 s intervals so as not to melt the sputter target. Indeed, as illustrated in Figure 1, the sputtering conditions in this study increased the surface roughness of the native substrates employed. Successful roughening of surfaces via sputtering of low-melting temperature metals has also been demonstrated in other studies.21,22

Figure 1. SEM images of indium and tin surfaces under various manufacturing conditions. (a) Indium foil, 100 μm thick. (b) Indium foil (100 μm thick) sputtered with indium for 10 min (RRMS = 190 nm). (c) Tin foil, 20 μm thick. (d) A silicon wafer sputtered with indium for 10 min (RRMS = 190 nm).

repeated at least three times, and contact angles were evaluated for at least 50 images per experiment. The wetting response was highly variable across metallic surfaces as a function of both material and texture. As seen in Figure 2a, sputtering a surface with indium renders the surface repellent to eGaIn (sputter target: American Elements, 99.999% pure). However, exposure to bare indium foil results in a high level of wetting, as seen in Figure 2b (indium foil: Sigma Aldrich, no. 357308, 99.999% pure). Similarly, galinstan wets reactively to a smooth tin surface (tin foil: Sigma Aldrich, no. 14511) (Figure 2c) but is repelled by sputtered films of tin (sputter target: American Elements, 99.999% pure). 3.1. Wetting Behavior of eGaIn. As shown in Table 2, eGaIn highly wets bare indium foil but only partially wets tin foil. Additionally, eGaIn is repelled entirely from both indium and tin sputtered surfaces. Furthermore, the equilibrium state of eGaIn on indium foil was found to embrittle the foil via grain boundary penetration, a phenomenon similar to the penetration of liquid gallium into the grain boundaries of aluminum,23,24 thus indicating reactive wetting. Given the reaction between eGaIn and foils of indium and tin, it is not intuitive that thin sputtered films should repel the alloy. As seen in Figure 1, deposition by sputtering yields a film with relatively high surface roughness, and this effect is exacerbated with longer sputter times (e.g., the substrates sputtered with indium for 2 and 10 min exhibited root-meansquare surface roughness (RRMS) values of 50 and 190 nm, respectively). This result is analogous to studies showing that hydrophobicity of a substrate may be controlled by microtexturing the surface.25 It was briefly considered that a surface oxide may be preventing rapid or reactive wetting with sputtered layers; however, it is unlikely that indium will oxidize at low temperatures (the material was only heated during sputtering, which occurred in an argon environment), and sputtered samples did not show the characteristic yellow hue of indium oxide. In addition, we conducted several experiments wherein any potential oxide was removed with a flux cleaning

3. RESULTS Throughout the experimental phase of this work, it became clear that wetting between the Ga−In alloys and indium or tin substrates demonstrates one of three distinct outcomes: (1) a high contact angle with no wetting, (2) conventional partial wetting, or (3) initial conventional partial wetting, followed by reactive wetting. To clarify, we will define conventional wetting as a dynamic response that leaves the solid−liquid interface materially and morphologically unchanged. However, during reactive wetting, the solid−liquid interface undergoes complex material and morphological changes. Examples of outcomes 1 and 3 are shown in Figure 2. As evident from the t = 0 s images in Figure 2a−c, the sessile drops maintain a conical shape at the top, similar to solder droplets. This shape is inherent to the formed oxide layer on the surface of the drop, as discussed previously.19 Table 2 provides a summary of the characteristic contact angle between Ga−In alloys and various surfaces tested in this study. For each solid−liquid combination, the experiment was B

dx.doi.org/10.1021/la404356r | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 2. (a) An eGaIn droplet on a surface of indium foil (100 μm thick) that has been sputter coated with indium for 2 min (RRMS = 50 nm). The droplet does not wet the surface. (b) An eGaIn droplet on a surface of indium foil (100 μm thick). The droplet rapidly wets on the order of seconds. (c) A galinstan droplet on a surface of tin foil (20 μm thick). The droplet wets the surface slowly over a period of days, demonstrating reactive wetting. Scale bar in each image is 1 mm in length.

Table 2. Summary of Characteristic Equilibrium Contact Angles between Ga−In Alloys and Indium/Tin Surfacesa drop material eGaIn

galinstan

substrate

sputter material

sputter time (min)

contact angle (deg) 158b 162b 140 → 10c

Si

In

In foil (100 μm thick)

none

2 10 −

In none

2 −

160b 135

Sn none

2 10 −

159b 160b 145

none

−

145 → 20c

Sn

2

160b

Sn foil (20 μm thick) Si In foil (100 μm thick) Sn foil (20 μm thick)

Figure 3. Liquid galinstan that has been in contact with a tin foil surface (20 μm thick) for over 4 days. Reactive wetting can be inferred not only by the slow spreading of the liquid on the surface but also the signature of the surface material changes occurring around the perimeter of the droplet. Scale bar is 1 mm in length.

a

All sputtering treatments were administered with 100 mA in 30 s intervals so as not to melt the sputter target. bNo wetting, substrate is “metallophobic”. cThe value to the left of the arrow is the conventional equilibrium contact angle achieved prior to reactive wetting. The value to the right of the arrow is reactive wetting equilibrium contact angle.

this can be seen not only by the slow spreading of the liquid droplet on the surface but also by the signature of the surface material changes occurring around the perimeter of the droplet (Figure 3). The newly formed alloy was analyzed using EDS in a scanning electron microscope over five samples, with the average composition (percentage by weight) being 57.63% ± 0.24% gallium, 22.69% ± 0.09% indium, and 19.68% ± 0.20% tin. This increased percentage of tin (from the original composition of galinstan) is consistent with the assumption that the underlying tin foil is absorbed into the galinstan. 3.3. Wetting Dynamics of eGaIn and Galinstan. Dynamic contact angles of eGaIn droplets on an indium foil are shown as a function of time in Figure 4a, where both the left and right contact angles of four independent droplets are analyzed. This plot gives one representation of the wetting that takes place between liquid eGaIn and a smooth indium surface; however, several important notes must be made about these observations. First, in several experiments a time delay was observed (on the order of seconds) before spreading was initiated. In some cases, mechanical agitation of the substrate was required to initiate the reactive wetting. This feature can be

agent, which was found to have no effect on eGaIn’s wettability to the sputtered surfaces. 3.2. Wetting Behavior of Galinstan. Unlike eGaIn, galinstan exhibits conventional partial wetting with indium foil. We hypothesize that this difference in outcomes is due to a lesser percentage of indium relative to the composition of the eutectic (eGaIn is 25% In, while galinstan is only 21.5% In), where indium-to-indium contact is assumed to be the catalyst for dynamic wetting. On the other hand, galinstan reactively wets tin foil because of the presence of tin (galinstan is 10% Sn). The complete composition of both eGaIn and galinstan is given in Table 1. An important result in this study is the equilibrium state of galinstan in contact with tin foil. The materials react together to become an alloy of a new composition. During reactive wetting, C

dx.doi.org/10.1021/la404356r | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 4. (a) Dynamic contact angles of eGaIn droplets wetting on an indium foil surface. Both left and right contact angles are plotted. Different symbols represent contact angles measured on different sides of different drops under the same experimental conditions. (b) Baseline length of eGaIn droplets wetting on indium and corresponding to the contact angles plotted in part a. Different symbols represent baseline lengths of different drops under the same experimental conditions. (c) Average baseline length of eGaIn droplets (four) wetting on indium and corresponding to the data in part b. Inset: Logarithmic plot and power-law fit, with β ≈ 1/4 (inertial-like wetting) and β ≈ 1/10 (viscous wetting). (d) Average baseline length of galinstan droplets (three) wetting on a tin foil substrate. Inset: Logarithmic plot and power-law fit, with β ≈ 1/7. All droplets were measured to be approximately 3 μL in volume. Error bars represent one standard deviation in each direction.

continuous spreading of the droplet; hence, contact angle and droplet spreading are not well-correlated. A clearer representation of the droplet spreading is seen by looking at the baseline length of the droplet as a function of time, where the baseline length is defined as the droplet diameter at the contact interface. Data showing the baseline of four wetting droplets is shown in Figure 4b. Similarly, Figure 4c depicts the average baseline value of these droplets. Figure 4d shows the average baseline length of three droplets of galinstan on a tin foil substrate. Here, it is seen that wetting follows the same trends as in the previous case, yet the timescales are very different. In the slow reactive wetting case, the droplets appear to wet similarly at the start with nearly no variation in the data. For t > 12 h, the data begin to show significant variations. This result is likely due to imperfections in the foil that cause the droplet to preferentially spread in certain directions as it covers increasing surface area. Finally, it is interesting to compare the power laws associated with both the fast and slow wetting processes. In Figure 4, parts c and d include insets of the averaged data on a logarithmic scale. We assume that the contact radius of the droplets follows a power law and therefore scales with time as R = αtβ. From this, a linear fit on a logarithmic scale can be employed to estimate the value β. For the slow reactive spreading (galinstan

seen in Figure 2b, where the right side of the droplet begins spreading before the left. In Figure 4a, the data was manually aligned such that t = 0 when either the left or right contact angle begins to change. This delayed reaction is a result of two potential factors: (1) The presence of the gallium-oxide film that forms around the droplet during exposure to air, which will often break as the droplet impacts the substrate; however, when this does not occur, slight agitation or eventual embrittlement of the underlying grains will “break” the oxide film and allow the encased gallium−indium alloy to spread rapidly on the surface. (2) The amount of indium present at the liquid−solid interface is close to the minimum amount necessary for reactive wetting to occur. This further supports the different wetting behaviors of eGaIn and galinstan on indium foil and implies that liquid motion along the contact interface will spontaneously react when an indium presence threshold is surpassed. Another important point is that in Figure 4a complete spreading appears to occur almost instantaneously in several cases, while the visual representation in Figure 2b shows the droplet spreading over a time period of approximately 10 s. Referring to Figure 2b, it appears that during wetting, liquid begins to escape out from beneath the oxide layer. The contact angle when this first begins does not appear to change during D

dx.doi.org/10.1021/la404356r | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

on tin), we find β ≈ 1/7. This value for β is in agreement with previous reports on dissipation occurring near the contact line when spreading is dominated by gravity in a two-dimensional scenario.26 For the faster reactive spreading (eGaIn on indium), the data appear to have two trends: at short time scales there is a region with a relatively steep slope on the logarithmic plot, while at longer time scales the slope of the data decreases. This initial steep slope (β ≈ 1/4) is in agreement with the values previously reported for viscous dominant spreading of oil and water droplets on silicon micropillar arrays.27 Over time, the less steep slope (β ≈ 1/10) is in agreement with Tanner’s law of spreading on a smooth surface.28 This suggests that the reactive wetting induces a smoothing effect of the substrate (i.e., the eGaIn reacts with the indium foil to cause a morphological transition of the surface to a smoother configuration). Other two-part wetting dynamics have also been observed in previous studies.29,30

metallaphobic wettability of sputtered indium observed within this study, is indicative of a Cassie state. Consider a liquid droplet of eGaIn on a surface sputtered with indium. The sputtering process coats the surface with small “clumps” of indium, as seen in Figure 1d. The Cassie−Baxter relationship is given as: cos θ* = rφ cos θ1 + (1 − φ)cos θ2

(1)

where θ1 is the contact angle between eGaIn and a solid (flat) indium surface, θ2 is the contact angle between eGaIn and air (180°), and θ* is the contact angle between eGaIn and a microtextured (sputtered) indium surface. In addition, rφ and (1 − φ) are the areal ratios of the eGaIn−indium interface and eGaIn−air interface compared to the total droplet contact area, respectively. Confocal microscopy was employed to investigate the applicability of the Cassie−Baxter relationship to the observed enhanced wetting effect provided by sputtering. Figure 6 shows a summary of these results applied to a silicon wafer sputtered with indium for 10 min. Image analysis was conducted on the intensity image shown in Figure 6a to find the areal ratio of the eGaIn−air interface compared to the total droplet contact area, (1 − ϕ). To achieve this, the intensity image was converted to a binary image via MATLAB’s image processing toolbox. The threshold level was adjusted manually until all clusters were present in the binary image. This led to the binary image shown in Figure 6b with a corresponding threshold level of 60/255. From the binary image, (1 − φ) can be approximated as the ratio of the total number of black pixels to the total number of pixels in the image, resulting in (1 − φ) ≈ 0.79. Assuming negligible penetration of the liquid droplet into the air gaps formed by the sputtered indium, the areal ratio of the eGaIn− indium interface compared to the total droplet contact area can be approximated by rφ ≈ φ ≈ 0.21. Via substitution, we find that

4. DISCUSSION Table 2 shows us that the wetting properties between liquid gallium−indium and a metallic solid can be tuned by roughness. Contrary to typical hydrophobic surfaces modulated by surface texture (i.e., the lotus leaf effect), this metallophobic surface is enabled by the gallium-oxide skin around the metallic droplet, which prevents liquid penetration of the texture and any subsequent reactive wetting (as shown in Figure 5). This

cos θ1 = Figure 5. Schematic of the liquid droplet on an unstructured rough surface. Top gray line in detailed view indicates drop surface.

cos θ* + (1 − φ) rφ

(2)

and θ1 ≈ 140.0°. Interestingly, this is the same as the measured contact angle between eGaIn and a solid indium surface just prior to reactive wetting (Table 2). Therefore, the Cassie− Baxter relationship appears to predict the contact angle between eGaIn and solid indium prior to reactive wetting but not the equilibrium contact angle at the end of reactive wetting. In summary, we have shown that eGaIn droplets have an affinity for indium substrates, but this property may be altered

gallium-oxide layer has been observed to resist surface stresses up to ∼0.5 N/m,18 which explains why a gallium−indium droplet would be unable to penetrate surface texture without considerable impact force or agitation. Furthermore, the structurally stable gallium-oxide layer formed on the surface of the droplet, along with the

Figure 6. (a) Intensity image of sputtered indium (sputter time = 10 min/RRMS = 190 nm) on a silicon wafer. (b) Binary representation of intensity image (threshold level of 60/255). (c) Surface reconstruction obtained with a 3D laser confocal microscope. Scale bars = 20 μm. E

dx.doi.org/10.1021/la404356r | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(6) Park, Y. L.; Majidi, C.; Kramer, R.; Bérard, P.; Wood, R. J. Hyperelastic pressure sensing with a liquid-embedded elastomer. J. Micromech. Microeng. 2010, 20, 125029. (7) Kramer, R. K.; Majidi, C.; Wood, R. J. Wearable tactile keypad with stretchable artificial skin. In 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS); IEEE: Piscataway, NJ, 2011; pp 1103−1107. (8) Kramer, R. K.; Majidi, C.; Sahai, R.; Wood, R. J. Soft curvature sensors for joint angle proprioception. In 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS); IEEE: Piscataway, NJ, 2011; pp 1919−1926. (9) Majidi, C.; Kramer, R.; Wood., R. J. A non-differential elastomer curvature sensor for softer-than-skin electronics. Smart Mater. Struct. 2011, 20, 105017. (10) Cumby, B. L.; Hayes, G. J.; Dickey, M. D.; Justice, R. S.; Tabor, C. E.; Heikenfeld, J. C. Reconfigurable liquid metal circuits by laplace pressure shaping. Appl. Phys. Lett. 2012, 101 (17), 174102−174102. (11) Park, J.; Wang, S.; Li, M.; Ahn, C.; Hyun, J. K.; Kim, D. S.; Rogers, J. A.; Huang, Y.; Jeon, S.; et al. Three-dimensional nanonetworks for giant stretchability in dielectrics and conductors. Nat. Commun. 2012, 3, 916. (12) Jeong, S. H.; Hagman, A.; Hjort, K.; Sundqvist, J.; Wu, Z. Liquid alloy printing of microfluidic stretchable electronics. Lab Chip 2012, 12 (22), 4657−4664. (13) Tabatabai, A.; Fassler, A.; Usiak, C.; Majidi, C. Liquid-phase galliumindium alloy electronics with microcontact printing. Langmuir 2013, 29 (20), 6194−6200. (14) Ladd, C.; So, J.-H.; Muth, J.; Dickey, M. D. 3D printing of free standing liquid metal microstructures. Adv. Mater. 2013, 25 (36), 5081−5085. (15) Schiaffino, S.; Sonin, A. A. Molten droplet deposition and solidification at low Weber numbers. Phys. Fluids 1997, 9, 3172−3187. (16) Naidich, J. V.; Chuvashov, J. N. Wettability and contact interaction of gallium-containing melts with non-metallic solids. J. Mater. Sci. 1983, 18 (7), 2071−2080. (17) Karcher, C.; Kocourek, V.; Schulze, D. Experimental investigations of electromagnetic instabilities of free surfaces in a liquid metal drop. Presented at International Scientific Colloquium Modelling for Electromagnetic Processing, Hannover, March 24−26, 2003. (18) Dickey, M. D.; Chiechi, R. C.; Larsen, R. J.; Weiss, E. A.; Weitz, D. A.; Whitesides, G. M. Eutectic gallium-indium (egain): A liquid metal alloy for the formation of stable structures in microchannels at room temperature. Adv. Funct. Mater. 2008, 18 (7), 1097−1104. (19) Chiechi, R. C.; Weiss, E. A.; Dickey, M. D.; Whitesides, G. M. Eutectic gallium-indium (egain): A moldable liquid metal for electrical characterization of self-assembled monolayers. Angew. Chem. 2008, 120 (1), 148−150. (20) Noda, K.; Iwase, E.; Matsumoto, K.; Shimoyama, I. Stretchable liquid tactile sensor for robot-joints. In 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS); IEEE: Piscataway, NJ, 2010; pp 4212−4217. (21) Song, H. K.; Jeong, J. K.; Kim, H. J.; Kim, S. K.; Yoon, K. H. Fabrication of CuIn-Ga-Se thin film solar cells by sputtering and selenization process. Thin Solid Films 2003, 435 (1), 186−192. (22) Staudt, C.; Wucher, A. Generation of large indium clusters by sputtering. Phys. Rev. B 2002, 66, Article 075419. (23) Ludwig, W.; Pereiro-Lopez, E.; Bellet, D. In situ investigation of liquid Ga penetration in Al bicrystal grain boundaries: Grain boundary wetting or liquid metal embrittlement? Acta Mater. 2005, 53 (1), 151− 162. (24) Trenikhin, M. V.; Bubnov, A. V.; Kozlov, A. G.; Nizovskii, A. I.; Duplyakin., V. K. The penetration of indium-gallium melt components into aluminum. Russian J. Phys. Chem. A 2006, 80 (7), 1110−1114. (25) Bico, J.; Thiele, U.; Quéré, D. Wetting of textured surfaces. Colloids Surf., A 2002, 206 (1), 41−46. (26) Ehrhard, P. Experiments on isothermal and non-isothermal spreading. J. Fluid Mech. 1993, 257, 463−483.

by surface roughness. Introducing micrometer-scale texture to the surface renders the surface “metallophobic”. While this nonwetting state is likely an initial result of the surface roughness, we believe that it is stabilized over longer times from oxidation of the droplet where it is exposed to air. The model employed above helps us to estimate how much of the droplet is in contact with the rough substrate and how much is interfaced with air (and thus susceptible to oxide formation) but is limited to predicting the contact angle of the drop prior to reactive wetting.

5. CONCLUSION Wetting behaviors between droplets of Ga−In alloys and thin metal films of indium and tin have been presented. It was found that both composition and texture affect the wettability of the substrate. The dynamics of rapid and slow reactive wetting were analyzed, yielding power laws for cases of Ga−In spreading on indium and tin foils, respectively. For eGaIn on indium foil, the droplet rapidly spreads with a radius varying in two phases: (1) similar to viscous wetting on micropillar arrays and (2) similar to spreading on a smooth surface, where the associated power law exponents are 1/4 and 1/10, respectively. For galinstan on tin foil, the droplet radius reactively spreads proportionally to t1/7, indicative of gravity-dominant dissipation occurring at the contact line. In contrast, microtextured (sputtered) films were found to repel Ga−In droplets. The surface texture of sputtered indium has been characterized to validate the Cassie−Baxter model for nonwetting surfaces, showing good agreement with experiments prior to reactive wetting but poor agreement after reactive wetting. Wetting and nonwetting surfaces for Ga−In alloys will impact manufacturing processes for liquid-embedded elastomer electronics, as well as other applications that require patterning of Ga−In alloys as liquid conductors.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by the National Science Foundation (award number DMR 08-20484) and the Wyss Institute for Biologically Inspired Engineering.

■

REFERENCES

(1) Rogers, J. A.; Someya, T.; Huang, Y. Materials and mechanics for stretchable electronics. Science 2010, 327 (5973), 1603. (2) Cheng, S.; Wu, Z. Microfluidic stretchable RF electronics. Lab Chip 2010, 10 (23), 3227−3234. (3) Kim, D. H.; Rogers, J. A. Stretchable electronics: Materials strategies and devices. Adv. Mater. 2008, 20 (24), 4887−4892. (4) So, J. H.; Thelen, J.; Qusba, A.; Hayes, G. J.; Lazzi, G.; Dickey, M. D. Reversibly deformable and mechanically tunable fluidic antennas. Adv. Funct. Mater. 2009, 19 (22), 3632−3637. (5) Paik, J. K.; Kramer, R. K.; Wood, R. J. Stretchable circuits and sensors for robotic origami. In 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS); IEEE: Piscataway, NJ, 2011; pp 414−420. F

dx.doi.org/10.1021/la404356r | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(27) Kim, S. J.; Moon, M.; Lee, K.; Chang, Y. S.; Kim, H. Liquid spreading on superhydrophilic micropillar arrays. J. Fluid Mech. 2011, 680, 477−487. (28) Tanner, L. H. The spreading of silicone oil drops on horizontal surfaces. J. Phys. D: Appl. Phys. 1979, 12 (9), 1473. (29) Bird, J. C.; Mandre, S.; Stone, H. A. Short-time dynamics of partial wetting. Phys. Rev. Lett. 2008, 100 (23), 234501. (30) Courbin, L.; Bird, J. C.; Reyssat, M.; Stone, H. A. Dynamics of wetting: from inertial spreading to viscous imbibition. J. Phys.: Condens. Matter 2009, 21 (46), 464127.

G

dx.doi.org/10.1021/la404356r | Langmuir XXXX, XXX, XXX−XXX