Spontaneous Deformation of an Oil Droplet Induced by the

Nov 10, 2009 - Department of Physics, Graduate School of Science, Kyoto University, Kyoto, 606-8502 Japan, Department of Physics, Graduate School of S...
0 downloads 9 Views 3MB Size
Download PDF
 Copyright 2009 by the American Chemical Society

VOLUME 113, NUMBER 48, DECEMBER 3, 2009

ARTICLES Spontaneous Deformation of an Oil Droplet Induced by the Cooperative Transport of Cationic and Anionic Surfactants through the Interface Yutaka Sumino,*,⊥,† Hiroyuki Kitahata,‡ Hideki Seto,§ Satoshi Nakata,| and Kenichi Yoshikawa† Department of Physics, Graduate School of Science, Kyoto UniVersity, Kyoto, 606-8502 Japan, Department of Physics, Graduate School of Science, Chiba UniVersity, Chiba, 263-8522 Japan, Institute of Materials Structure Science, High Energy Accelerator Research Organization, Ibaraki, 305-0801 Japan, and Graduate School of Science, Hiroshima UniVersity, Higashi Hiroshima, 739-8526 Japan ReceiVed: April 24, 2009; ReVised Manuscript ReceiVed: October 23, 2009

Spontaneous deformation of a tetradecane droplet with palmitic acid on an aqueous phase with stearyltrimethylammonium chloride is reported. Palmitic acid is transported from the oil droplet to the aqueous phase by the concentration difference between the organic and the aqueous phases. The transport of palmitic acid causes the oil droplet interface to undergo various spontaneous deformations. When the oil droplet is placed on an aqueous surface, its diameter shrinks. Several tens of seconds later, the oil droplet suddenly expands and then shrinks in a second. After such a dramatic deformation, the oil droplet undergoes blebbing on its oil-water interface for over 1 h. We investigated the physicochemical mechanism of these phenomena. We discuss the cause of these deformations in terms of the spatiotemporal variation of the interfacial tension and elucidate that the blebbing deformation is due to the surfactant aggregate generated by cationic and anionic surfactants. 1. Introduction Systems far from equilibrium exhibit rich varieties of spatial and temporal patterns, which are known as dissipative structures.1 An oil-water system is a simple far-from-equilibrium system that involves transport of surfactants. There are two main kinds of research on oil-water systems.2 One considers the temporal fluctuations of the electrical potential at the oil-water * Corresponding author. E-mail: [email protected]. † Kyoto University. ‡ Chiba University. § High Energy Accelerator Research Organization. | Hiroshima University. ⊥ Present address: Department of Applied Physics, The University of Tokyo, Tokyo, 113-8656 Japan.

interface,3 and the other focuses on the spontaneous motion of the oil-water interface.4 The spontaneous motion of oil-water interface has received a lot of interest because it involves isothermal energy transduction from chemical to mechanical energy.5,6 Until now, most studies on spontaneous interfacial motion in an oil-water system have suggested that it is induced by an inhomogeneous distribution of interfacial energy.7–10 On the other hand, it is known that surfactants form a micrometer-scale structure by self-aggregation and can possess elasticity on a millimeter-scale. A system driven by the generation of surfactant aggregates is potentially a simple model of biological motility. In fact, the extension of pseudopods is partially driven by generation of actomyosin gel.11 Furthermore, the intracellular bacterial parasite Listeria (Listeria monocytogenes) propels itself by generating

10.1021/jp9037733 CCC: $40.75  2009 American Chemical Society Published on Web 11/10/2009

15710

J. Phys. Chem. B, Vol. 113, No. 48, 2009

a tail-like actin gel from G-actin in the cytoplasm of its host cell (known as an actin rocket).12–14 These facts show that the generation of aggregates can be a simple and important factor of biological motility, and therefore, the spontaneous motion of oil-water interface caused by the generation of aggregates can be a relevant model. Recently, we reported the blebbing of an oil droplet interface, where an oil droplet and an aqueous phase containing anionic and cationic surfactants met.15 In the study, we theoretically discussed the mechanism of the interfacial blebbing, assuming the generation of an elastic and permeable aggregate on the oil-water interface. Despite the potential importance of this system as a model of biological motility, the physicochemical nature of this system has not yet been clarified. We note that the oil droplet shows spontaneous shrinking and spreading/recoil as precursor events of the interfacial blebbing. In this paper, we aim to elucidate that the blebbing motion is realized through aggregate generation. For this purpose, the following analyses are executed: First, we inspect the manner of these precursor events and the underlying mechanism. Second, the physicochemical properties of the aggregate in relation to the blebbing motion are examined. Finally, variation in the manner of the blebbing is discussed as future problems. 2. Experimental Methods Stearyltrimethylammonium chloride (STAC) and palmitic acid were purchased from Tokyo Chemical Industry Co., Ltd. (S0087 and P1145; assay >97% and >99.5%, respectively). Tetradecane was purchased from Wako Pure Chemical Industries, Ltd. (207-10705; assay >99%). These chemicals were used without further purification. Water was purified by a Millipore Milli-Q system. As the organic phase, palmitic acid was dissolved into tetradecane at concentrations of Cp ) 1-20 mM. We used Cp ) 20 mM unless stated otherwise. STAC was dissolved into water at concentrations of Cs ) 0.1-100 mM. The critical micelle concentration of STAC is about 0.3 mM at room temperature. In this study, the volume of the oil droplet was kept at 100 µL unless otherwise stated. An oil droplet floats on an aqueous phase because the relative density of the organic phase is 0.76 g/cm3, which is lower than that of the aqueous phase (∼1 g/cm3). The droplet deformation was recorded using a digital video camera (NV-GS 100K-K, Panasonic, Japan). The shadowgraph method16 was utilized while the system was illuminated with visible light (Cold spot, Picl-Nex Nippon PI Co., Ltd., Japan). For surface tension measurements, we used a surface tensiometer (GBVP-A3, Kyowa Interface Science Co. Ltd., Japan) with the Wilhelmy method. An oil droplet was placed on a 50 mL aqueous phase with a concentration Cs in a Petri dish with a diameter of 85.5 mm. The oil droplet was pinned using a plastic needle 20 mm from the center of the Petri dish to fix the position of the oil droplet, and a Wilhelmy plate connected to the surface tensiometer was set at the center of the Petri dish. The data of the surface tensiometer were recorded at 100 points/s. For the optical microscope observation, we used a polarization microscope (BX60, Olympus, Japan) equipped with a ×4 objective lens (UPLFLN 4X, N.A. 0.13, W.D. 17 mm, Olympus, Japan). For small-angle X-ray scattering (SAXS) measurements, we used the SAXS apparatus installed at BL15A of the Photon Factory, High Energy Accelerator Research Organization, Tsukuba, Japan. The incident X-rays were monochromatized by Ge(111), and the wavelength was 1.5 Å. All measurements were carried out at room temperature (∼22 °C).

Sumino et al.

Figure 1. (a) Apparent area, A, and (b) shadowgraph images of an oil droplet. τ represents the duration of the shrinking stage. In this plot, τ ) 18.3 s. t ) 0 is the time when the oil droplet was placed on the aqueous surface in a Petri dish. The concentration of the aqueous phase Cs was 1.0 mM. Scale bar corresponds to 10 mm.

3. Results Observation of Droplet Motion. The temporal change in the apparent droplet area, A, and the manner of oil droplet deformation are depicted in Figure 1 and the Supporting Information 1. The concentration of STAC in the aqueous phase was Cs ) 1 mM. The diameter of the Petri dish was 145 mm, and the volume of the aqueous phase was 100 mL. There are three distinct stages in the droplet deformation: shrinking, spreading/recoiling, and blebbing. After an oil droplet was placed on the aqueous surface, it shrunk continuously (shrinking stage); it then started to spread and recoil in a short time span (∼1 s) (spreading/recoiling stage). The oil droplet subsequently underwent continuous interfacial blebbing for over 1 h (blebbing stage). Since the duration of the spreading/recoiling stage was considerably shorter than those of the other stages, the duration of the shrinking stage, τ, was taken to be the induction time for interfacial blebbing to occur. Aggregates were not observed during either the shrinking stage or the spreading/recoiling stage. In contrast, aggregates in the aqueous phase were observed in the blebbing stage. Measurement of Temporal Change in Surface Tension. To investigate the shrinking stage, the surface tension of the air-water interface, γaw, was measured for various Cs’s. The experimental setup is shown in Figure 2a. The time variation of the surface tension, γaw, is shown in Figure 2b. In this measurement, the duration of the shrinking stage, τ, was also recorded (open squares in Figure 2b). When Cs ) 0 mM, the oil droplet just spread on the aqueous surface. The surface tension decreased from 73.4 to 63.0 mN/m. With the presence of STAC in the aqueous phase (Cs * 0), a large reduction in γaw was observed. The final value of γaw ∼ 25 mN/m was independent of Cs. When γaw was about 30 mN/m, the oil droplet showed spreading and recoiling. As it reached a steady value, interfacial blebbing started; thus, the shrinking and the spreading/ recoil stages correspond to the period in which the surface tension of the aqueous phase continuously decreases. The temporal change in γaw was obtained by taking a temporal

Spontaneous Oil Droplet Deformation

J. Phys. Chem. B, Vol. 113, No. 48, 2009 15711

Figure 3. (a) Typical dependence of the induction time τ on various areas of the aqueous surface Aaw. The error bar represents the standard deviation of the measured data. Here the volume of the oil droplet was 100 µL and Cs ) 1 mM. τ was linearly dependent on Aaw. The dotted line is the linear fitting of the experimental data. (b) Dependence of the ratio of the induction time to the area of the aqueous surface τ/Aaw on Cs. The data are obtained from the linear fitting of τ with Aaw from at least 10 observations. The volume of the oil droplet was 100 µL. As Cs increases, τ/Aaw becomes small. Figure 2. Measurement setup (a); time trace of air-water interfacial tension, γaw (b); and its temporal change (c). (a) An oil droplet was placed on an aqueous phase at t ) 0. The oil droplet was pinned with a plastic wire 20 mm from the center of a Petri dish, and the air-water interfacial tension was measured at the center. (b) The concentrations of the aqueous phase, Cs, were 0.2, 1.0, and 20.0 mM, as shown in the plot. Open squares correspond to τ, when the oil droplet underwent spreading. (c) The temporal change in γaw, dγaw/dt, is plotted with respect to t/τ.

difference of obtained data in Figure 2b. As shown in Figure 2c, the decrease rate in γaw, dγaw/dt, reaches its maximum at t ) τ. The temporal fluctuation of the interfacial tension during the blebbing stage is attributable to the dynamical deformation of the air-water interface caused by the droplet blebbing. Measurement of the Duration of τ. The duration of the shrinking stage, τ, was measured for various values of air-water interface area, Aaw, for 100 µL oil droplets. Cs was also varied as a parameter. Before performing the measurement, we confirmed that τ was independent of the volume of the aqueous phase and the depth of the aqueous phase by varying the volume of the aqueous phase while keeping Aaw constant. It was noted that once a droplet was in the blebbing stage and was then removed from the aqueous surface, the shrinking stage did not occur for another droplet placed on the same aqueous surface. To avoid the hysteresis on the composition of aqueous surface, the aqueous phase was changed at each measurement. As seen in Figure 3a as typical results, τ linearly depended on Aaw. By fitting the data, τ for a unit area of the aqueous surface is obtained as shown in Figure 3b. We clearly see that τ became small as Cs increased. Observation of Aggregates. The polarization microscopic measurement was conducted as follows: A 3 µL portion of aqueous phase with Cs ) 50 mM was placed on a glass substrate. A 0.5 µL oil droplet with Cp ) 20 mM was then placed on the aqueous surface. Aggregates were observed immediately after the oil droplet started blebbing. Figure 4a and the Supporting Information 2 show that aggregate with optical anisotropy was generated on the oil-water interface and dispersed into the aqueous phase. After the aggregate was detached from the oil-water interface, the distribution became asymmetric, as in Figure 4a. This is attributable to the entanglement of aggregate, even after its detachment. To elucidate the structure of the aggregate, SAXS measurements were also performed. A 350 µL portion of an organic phase with Cp ) 10 mM was placed on a 700-µL aqueous phase with Cs ) 25 mM. The cell was made of acryl plate and the window

Figure 4. (a) Polarization microscopic image of the aggregate formation at the oil-water interface. Microscopic image of aggregate formation at the oil-water interface. Center: Crossed Nicol image of aggregate formation. Right: Schematic diagram of the situation in the microscopic images. Scale bar corresponds to 200 µm. (b) Setup for SAXS measurement. (c) SAXS profile of the aggregate. In this profile, the first peak is at qm ) 0.0155 Å-1.

for the X-rays was covered with a Kapton film (Figure 4b). In this setup, the shrinking and spreading/recoil stages were not observed due to the absence of an air-water interface in the system because τ ∼ Aaw ) 0. The oil-water interface started to bleb immediately after the organic phase had been transferred to the aqueous phase. In the aqueous phase near the oil-water interface, we obtained the profile of the scattering data shown in Figure 4c. The first Bragg peak was observed at qm ) 0.0155 Å-1, and higher-order peaks up to the sixth order appeared at wave numbers equal to an integral multiple of qm. Phase Diagram of Blebbing. The variation of the manner of the blebbing was examined. To investigate the effect of the concentrations of STAC and palmitic acid, both Cs and Cp were varied, while the volumes of the oil droplet and the aqueous phase were 500 µL and 2000 µL, respectively. Here, we used an aqueous droplet placed on an acrylic plate instead of a plane aqueous phase in order to see the long-time behavior of oil droplet motion. This is because an oil droplet tended to be attracted to a meniscus in the case of a plane aqueous phase, whereas an oil droplet tended to be repelled from the edge of the aqueous droplet. Figure 5 and the Supporting Information 3–5 show the manner of droplet deformation. Blebbing was

15712

J. Phys. Chem. B, Vol. 113, No. 48, 2009

Sumino et al. the extrusion (Figure 5b). The white-turbid aggregate appeared to be different from the aggregate shown in Figure 4. This type of extrusion occurred a few minutes after intensive blebbing, and the oil droplet ceased its motion after the whole droplet was covered with the white-turbid aggregate. The microscopic measurement of the white-turbid aggregate was conducted in the following way: A 100 µL portion of aqueous phase with Cs ) 50 mM was placed on a glass substrate. A 10-µL oil droplet with Cp ) 20 mM was then placed on the aqueous surface. The enlarged image of the oil-water interface covered with the white-turbid aggregate is shown in Figure 5d. With polarization microscopy, it was confirmed that the white-turbid aggregate does not possess optical anisotropy, which is different from the aggregate observed when the droplet is blebbing (Figure 4a). 4. Discussion

Figure 5. Phase diagram of the droplet blebbing and configuration of experimental setup. The letter on the snapshot (a, b, c) represents t minutes after the oil droplet was set on the aqueous surface, and the scale bars represent (a) 20 mm and (b, c) 10 mm. (a) At the open triangles, the droplets exhibited interfacial blebbing. As a result of blebbing, fission of the droplet was induced, and the droplet separated into smaller droplets over time. (b) At the filled circles, an oil droplet remained with a mesh-like structured aggregate at the center of the aqueous droplet. (c) At the open squares, part of the droplet started to be covered by a white-turbid aggregate (4 min), and it finally covered the entire oil-water interface (after 5 min). The interfacial deformation stopped afterward. (d) The microscopic measurement of the whiteturbid aggregate. Left: microscopic image of the oil-water interface covered with white-turbid aggregate. Right: Schematic diagram of the situation in the microscopic images. Scale bar corresponds to 200 µm.

observed in the whole concentration range for Cs > 0.1 mM and Cp > 1 mM. The diameter of the blebs ranged from about 0.1 to 10 mm. As both Cs and Cp decreased, the bleb increased in diameter. In the low concentration range (indicated by the open triangles in Figure 5), the droplet separated into smaller ones (∼10 - 100 µL) as some blebs grew too large to be pinched off from the droplet (Figure 5a). The detailed manner of fission of an oil droplet is exemplified in our previous report.15 In the intermediate concentration range (indicated by the solid circles in Figure 5), the droplet separated into smaller ones with smaller size in the range indicated by the open triangles. However, in this concentration range, a large oil droplet remained at the center of the aqueous droplet: this droplet was partially covered with the aggregate (Figure 5b). It was noted that the blebbing parts of the droplet regularly aligned. In the high concentration range (indicated by the open squares in Figure 5), an extrusion completely covered with a white-turbid aggregate was made, and the whole oil droplet was sucked into

These experiments revealed that an oil-water system with STAC and palmitic acid shows various droplet deformation behaviors due to the spatiotemporal variation of the interfacial tension and aggregate generation. Here, we start by discussing the mechanism of the shrinking and spreading/recoiling stages. Figure 2 shows that there was a decrease in γaw after the oil droplet was set on the aqueous phase. There are two characteristic points in these data. One is that the decrease in γaw is quite large, considering that γaw decreased from 73.4 to 63.0 mN/m without the presence of STAC in the aqueous phase. From this fact, palmitic acid alone cannot account for such a large decrease in γaw. Thus, we propose that the STAC molecule and palmitic acid produce a complex (catanionic complex) that is highly surface-active. The other characteristic of these data is that the final value of γaw ∼ 25 mN/m is the same for different Cs. If the effect of palmitic acid were additional, then the final value of γaw should depend on Cs, and the decrease in γaw from initial value should be the same for all Cs. This is not the case for our experimental results. Therefore, the produced catanionic complex must be the same, independent of the concentration of STAC. The only difference caused by the change in Cs is the speed of the formation of the catanionic complex. This hypothesis is strongly supported by the data shown in Figure 4, stating that τ decreases with an increase in Cs. The decrease in interfacial tension, γaw, seems to play an important role in the oil droplet deformation in the shrinking and spreading/recoil stages, considering the large decrease in γaw in these stages. To interpret the deformation of the oil droplet, it will be helpful to estimate the behavior of γow. We estimate the oil-water interfacial tension, γow, by combining the data in Figures 1a and 2b. For the estimation, we use the following relation for a flat droplet,17,18

-S )

Foil - Foil)e2 (F Fwater water

(1)

where S is a spreading parameter (S ) γaw + γow - γ, and γ is the air-oil interfacial tension) and e is the thickness of an oil droplet. We assume that the data can be rescaled as A(t) ) A(t/τ), and γaw(t) ) γaw(t/τ), as in Figure 6a. The thickness of the oil droplet e(t/τ) was obtained by e(t/τ) ) Voil/A(t/τ). Then, we have

γow(t/τ) ) (γaw(t/τ) - γ) +

( )

Foil Voil Fwater - Foil) ( Fwater A(t/τ)

2

(2) We confirmed that γ was constant (26.1 mN/m), even when there was an aqueous phase beneath it. The volume of the oil

Spontaneous Oil Droplet Deformation

Figure 6. (a) Data of Figures 1a (solid line) and 2b (dashed line, Cs ) 1 mM) used to obtain γow by eq 2. Horizontal axis is rescaled as t/τ. (b) Calculated temporal change in γow. Inset is an enlarged graph close to t ) τ.

droplet, Voil, was constant (100 µL) during the measurement. Putting Foil ) 0.76 g/cm3, Fwater ) 1.0 g/cm3, we obtained the estimated temporal change in γow(t/τ) (Figure 6b). Despite the vivid deformation of the oil droplet at the spreading/recoil stage, the obtained trend of γow is quite similar to that of γaw. In the same way as for γaw, γow reaches a steady value in the blebbing stage. Thus, we confirm that the main factor of the mechanism of interfacial blebbing is not the interfacial tension, γaw and γow. Close inspection of γow in Figure 6b (inset) shows a small fluctuation in γow noted at the spreading/recoil stage. When compared with the large decrease in the temporal change in γow, the fluctuation is small. However, such a small fluctuation should be the cause of the vivid motion observed in the spreading/recoil stage. We are still not sure what causes this small fluctuation. Since the rate of decrease of γaw reaches its maximum at t ) τ (Figure 2c), there might be some type of phase transition of a molecular film composed of a catanionic complex. These two initial stages are, therefore, driven mainly by the spatial variation in the interfacial tension. At these stages, we also noted that there appears a weak Marangoni flow induced by the spatial change in the interfacial tension, γaw and γow. However, as we have discussed above, the interfacial tension should not be the main factor for the interfacial deformation in the blebbing stage since both γaw and γow reach an almost constant value. In the previous study,15 we hypothesized that elastic and permeable aggregate is generated at the oil-water interface. Under that assumption, aggregate is continuously formed at the oil-water interface, and newly formed aggregate pushes out the preexisting one. As a result, the pushed-out aggregate is stretched as compared with the conditions when it is formed. Considering the aggregate is elastic, the stretching of the aggregate induces internal stress in the direction tangential to and normal to the oil-water interface. As a result, the oil droplet is squeezed by the aggregate, and the internal pressure of the oil droplet increases. Furthermore, the stress in the aggregates in the tangential direction results in the collapse of the aggregate at the outermost layers. In this situation, we expect that blebbing of an oil droplet can be driven by the generation/collapse of surfactant aggregate from theoretical discussion. However, up to the present, the aggregate had not yet been experimentally characterized. In this study, we confirmed that the aggregate is actually generated at the oil-water interface. The SAXS data can be interpreted by the formation of a highly ordered lamellar structure, having an interlayer distance, d, of 405 Å, in the aggregate phase, by d ) 2π/qm. The thickness of one layer in such a lamellar structure is several tens of angstroms (i.e., the long-axis length of a surfactant molecule). Therefore, there assumed to be a large amount of water between

J. Phys. Chem. B, Vol. 113, No. 48, 2009 15713 layers in the aggregate. A highly ordered structure and a large amount of water in the aggregate can match with the elasticity and permeability of the aggregate assumed in the previous study. In the present system, the aggregate was not observed in the absence of palmitic acid in tetradecane. Since the concentration is higher than the critical micellar concentration, there should be present micelles in the aqueous phase. This implies that a micelle-to-lamellar transition occurs at the oil-water interface due to the addition of palmitic acid to the aqueous phase. The aggregate of a STAC and water system has been reported,19,20 and an aggregate consisting of a cationic surfactant and a long-chain alcohol has also been reported.21–23 We believe that the aggregate observed in the present system should resemble these aggregates. Further experimental observation is needed to determine the character of the aggregate in the present system and, especially, to elucidate the effect of their generating process. One of the main blebbing patterns observed in this study is the extension and retreat of a circular bleb, which is consistent with that observed in the previous study.15 In addition, various blebbing patterns were observed by varying the initial concentrations of the system, Cs and Cp. The mechanism for the regular mesh pattern, as shown in Figure 5b, observed on the oil droplet in the middle concentration range is not yet fully understood, and we intend to perform a theoretical analysis using a mathematical model in the future. The extrusion behavior of the oil droplet with a white-turbid aggregate is another interesting pattern. The white-turbid aggregate seems to be different from the one observed with a circular bleb. This requires experimental studies of the equilibrium phase diagram of the system. 5. Conclusion In this paper, we constructed an oil-water system in which a fatty acid and cationic surfactant were separately dissolved into oil and water. Three distinct stages in an oil droplet’s behavior were observed after the droplet had been placed at an aqueous surface. Although the shrinking and spreading/recoil stage appear to be different, both stages correspond to the period when the catanionic combination of surfactants fills the entire air-water interface. Furthermore, we concluded that interfacial tension should not be a main factor in the blebbing stage. The character of the aggregate assumed to exist in the previous study15 was confirmed experimentally. Polarized microscopy observation confirmed that there is optically anisotropic aggregate at the oil-water interface in the blebbing stage. The structure of the aggregate formed in the aqueous phase was determined by SAXS measurements. These findings on an oil-water system with catanionic surfactants show that the physicochemical mechanism of blebbing of an oil droplet is actually due to the generation of surfactant aggregate. We also note that analysis of such a chemically far-from-equilibrium system of a molecular film on an aqueous phase is an interesting problem both experimentally and theoretically. Furthermore, the study of the manner of deformation in the blebbing stage induced by the generation and collapse of surfactant aggregate will stimulate the study of biological systems induced by the generation of elastic matter, such as amoeba motion and bacterial motion. Acknowledgment. We thank Dr. Y. Kawabata (Tokyo Metropolitan Univ.) and Dr. Y. Takenaka (Riken) for their helpful comments. This work has been performed under the approval of the Photon Factory Program Advisory Committee.

15714

J. Phys. Chem. B, Vol. 113, No. 48, 2009

(Proposal no. 2007G588). Y.S. is supported by the JSPS (Grant no. 21-3566) Research Fellowships for Young Scientists from the MEXT of Japan. This work was supported in part by a Grant-in-Aid for Creative Scientific Research (Grant no. 18GS0421) to K.Y. from JSPS. This work was also supported in part by a Grant-in-Aid for Scientific Research (no. 20550124) and by a grant from the Asahi Glass Foundation to S.N. Supporting Information Available: A movie is available to illustrate the motion of a droplet. Supporting Information 1 (Movie 1): The motion of a 100 µL oil droplet, that corresponding to Figure 1. The sequence of images was acquired using a video camera with the shadowgraph method. The width and height of the frames in the movie correspond to 40 mm. The movie shown is accelerated by a factor of 3. Supporting Information 2 (Movie 2): The motion of a 0.5 µL oil droplet, corresponding to Figure 4a. The sequence of images was acquired using a video camera with polarized microscopy. The width and height of the frames in the movie correspond to 650 µm. The movie shown is accelerated by a factor of 4. Supporting Information 3 (Movie 3): The motion of a 500 µL oil droplet, corresponding to Figure 5a. The sequence of images was acquired using a video camera and the shadowgraph method. The scale bar is shown in the beginning of the movie. The movie shown is accelerated by a factor of 150. Supporting Information 4 (Movie 4): The motion of a 500 µL oil droplet, corresponding to Figure 5b. The sequence of images was acquired using a video camera and the shadowgraph method. The scale bar is shown in the beginning of the movie. The movie shown is accelerated by a factor of 150. Supporting Information 5 (Movie 5): The motion of a 500 µL oil droplet, corresponding to Figure 5c. The sequence of images was acquired using a video camera and the shadowgraph method. The scale bar is shown in the beginning of the movie. The movie shown is accelerated by a

Sumino et al. factor of 150. These materials are available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Nicolis, G.; Prigogine, I. Self Organization in Nonequilibrium Systems; Wiley: NewYork, 1977. (2) Linde, H.; Schwartz, P.; Wilke, H. Dynamics and Instability of Fluid Interfaces; Sørensen, T. S., Ed.; Springer-Verlag: Berlin, 1979. (3) Yoshikawa, K.; Matsubara, Y. J. Am. Chem. Soc. 1984, 106, 4423. (4) Kai, S.; Mu¨ller, S. C.; Mori, T.; Miki, M. Physica D 1991, 50, 412. (5) Magome, N.; Yoshikawa, K. J. Phys. Chem. 1996, 100, 19102. (6) Sumino, Y.; Magome, N.; Hamada, T.; Yoshikawa, K. Phys. ReV. Lett. 2005, 94, 068301. (7) Sumino, Y.; Yoshikawa, K. Chaos 2008, 18, 026106. (8) Shioi, A.; Kumagai, H.; Sugiura, Y.; Kitayama, Y. Langmuir 2002, 18, 5516. (9) Nakache, E.; Dupeyrat, M.; Vignes-Adler, M. Faraday Discuss. Chem. Soc. 1984, 77, 189. (10) Hanczyc, M. M.; Toyota, T.; Ikegami, T.; Packard, N.; Sugawara, T. J. Am. Chem. Soc. 2007, 129, 9386. (11) Yumura, S.; Fukui, Y. J. Cell Sci. 1998, 111, 2097. (12) Tilney, L. G.; Portnoy, D. A. J. Cell Biol. 1989, 109, 1597. (13) Upadhyaya, A.; van Oudenaarden, A. Curr. Biol. 2003, 13, R734. (14) Brieher, W. M.; Kueh, H. Y.; Ballif, B. A.; Mitchison, T. J. J. Cell Biol. 2006, 175, 315. (15) Sumino, Y.; Kitahata, H.; Seto, H.; Yoshikawa, K. Phys. ReV. E 2007, 76, 055202. (16) Settles, G. S. Schlieren and Shadowgraph Techniques: Visualizing Phenomena in Transparent Media (Experimental Fluid Mechanics); Springer: 2001. (17) de Gennes, P. G.; Brochard-Wyart, F.; Que´re´, D. Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, WaVes; Springer: 2003. (18) Langmuir, I. J. Chem. Phys. 1933, 1, 756. (19) Kodama, M.; Tsujii, K.; Seki, S. J. Phys. Chem. 1990, 94, 815. (20) Kodama, M.; Seki, S. AdV. Colloid Interface Sci. 1991, 35, 1. (21) Yamaguchi, M.; Noda, A. Nihon Kagaku Kaishi 1987, 1632. (22) Yamaguchi, M.; Noda, A. Nihon Kagaku Kaishi 1989, 26. (23) Yamagata, Y.; Senna, M. Langmuir 1999, 15, 7461.

JP9037733