ARTICLE pubs.acs.org/Langmuir
Multiple Pickering Emulsions Stabilized by Microbowls Yoshimune Nonomura,* Naoto Kobayashi, and Naoki Nakagawa Department of Biochemical Engineering, Graduate School of Science and Engineering, Yamagata University, Yonezawa 992-8510, Japan ABSTRACT: Some researchers have focused on the adsorption of solid particles at fluidfluid interfaces and prepared emulsions and foams called “Pickering emulsions/foams”. However, while several reports exist on simple spherical emulsions, few reports are available on the formation of more complex structures. Here, we show that holes on particle surfaces are a key factor in establishing the variety and complexity of mesoscale structures. Microbowls, which are hollow particles with holes on their surfaces, form multiple emulsions (water-in-oil-in-water and oil-in-water-in-oil emulsions) by simply mixing them with water and oil. Furthermore, stable potato-like or coffee-bean-like emulsions are also obtained, although nonspherical emulsions are usually unstable because of their larger interfacial energies. These findings are useful in designing the building blocks of complex supracolloidal systems for pharmaceutical, food, and cosmetic products.
’ INTRODUCTION Solid particles exhibiting suitable wettability are adsorbed at liquidliquid or airliquid interfaces, and they stabilize emulsions and foams.1,2 This phenomenon has been extensively studied and followed closely by many researchers, because some materials including colloidosomes, dry water, and particle films were prepared by particles adsorbed at interfaces.36 Many of these materials were simple spherical emulsions because surface tension gives liquid droplets a spherical shape, minimizing the surface area for a given volume.7 On the other hand, supracolloidal systems with complex structures have been produced to create new materials such as network structures,8,9 bicontinuous structures,10,11 squeezing air bubbles,12 and anisotropic structures formed by the arrested coalescence of the droplets.13 Such particle behaviors have been discussed on the basis of energy changes with adsorption at interfaces.14 Mbamala et al. showed that there is often a barrier that prevents spontaneous adsorption of particles when the particles have positive or negative charges.15 However, once this barrier is overcome, the particles are trapped at the interface because of capillary forces. The dipoledipole interactions organize the solid particles into a two-dimensional triangular lattice and the above-mentioned supracolloidal systems.16 Facile preparation methods are required for such complex structures in the pharmaceutical, food, and cosmetics industries.17 We focused on the effects of particle shape on interfacial behaviors.1820 As surfactants and amphiphilic polymers form supramolecular structures, solid particles with individual shapes can form unique supracolloidal structures.21 In the present study, we discovered that microbowls, which are silicone particles with holes on their particle surfaces (Figure 1),22,23 were adsorbed at liquidliquid interfaces and showed anomalous emulsification behaviors. For example, multiple emulsions, water-in-oil-inwater (WOW) and oil-in-water-in-oil (OWO) emulsions, were formed by simply mixing microbowls with water and oil. WOW emulsions are dispersions of oil droplets containing fine water r 2011 American Chemical Society
droplets, while OWO emulsions are those of water droplets containing fine oil droplets. In general, stable multiple emulsions are not formed with a single solid particle and standard emulsification methods.2426 We examined the mixed state of ternary systems consisting of microbowls, water, and oil using optical and fluorescence confocal microscopes. These observations revealed the effects of the unique particle shapes on their emulsification behaviors.
’ EXPERIMENTAL SECTION Materials. Microbowls, bowl-shaped silicone resin particles SPT001, and spherical silicone resin particles SPT-002 were obtained from Takemoto Oil and Fat Co. These white fluidic powders had average particle diameters of 2.5 and 2.0 μm, respectively. The wettabilities of the particles were evaluated from the sinking time t into water or n-dodecane. Thirty milligrams of particles was placed carefully and evenly on the surface of 20 cm3 water or n-dodecane contained in a tube with a diameter of 32 mm at 298 K. The time taken for all the particles to disappear from the liquid surface was measured and adopted as the sinking time. The times t for microbowls and spherical particles were as follows: for the mircobowls, t = above 14 days in water and 360 s in n-dodecane; for the spherical particles, t = above 14 days in water and 1 s in n-dodecane. These results predicted that both particles were hydrophobic, but the microbowls had lesser affinity for n-dodecane than did the spherical particles. n-Dodecane (CH3(CH2)10CH3) and 1-dodecanol (CH3(CH2)11OH) were purchased from Kanto Chemical Co. (Japan). Perfluoropolyether (PFPE) and dimethylpolysiloxane (DMS) were commercially available: FOMBLIN HC/04 (CF3[(OCF(CF3)CF2)n (OCF2)m]OCF3, n/m = 20/40, molecular weight = 1500; Ausimont K.K.) and KF96A(1CS) ((CH3)3SiO[Si(CH3)2O]nSi(CH3)3, molecular Received: January 27, 2011 Revised: March 7, 2011 Published: March 18, 2011 4557
dx.doi.org/10.1021/la2003707 | Langmuir 2011, 27, 4557–4562
Langmuir weight = 900; Shinestu Chemical Co. Ltd.). All of the chemicals were used as received. Methods. We prepared 10 g of ternary mixtures consisting of solid particles, water, and oil as follows. The solid particles were dispersed in the oil phase using an AS-ONE AHG-160A homogenizer at 20 000 rpm for 10 min. The resulting dispersion was mixed with the water phase by a vortex mixer (Vortex-genie 2; Scientific Industries Inc.) in a screw-cap test tube. The uniform operation, that is, adding the water phase of 1/5 of the target amount and mixing for 2 min, was repeated five times. We added 0.05 wt % rhodamine 6G (reagent grade; Kanto Chemical Co.) to the water phase. This water-soluble pigment stained the water phase and
ARTICLE
facilitated the identification of the mixed states. From the results of the preliminary studies, we determined that rhodamine 6G did not affect the condition of the ternary systems. The mixed states were characterized 5 min after the preparation using an Olympus BHT optical microscope and a Mettler conductivity meter SG3. Electrical resistance is convenient for identifying mixed states, because this physical property reflects the continuous phase of the mixtures; that is, the mixtures are conductive if the continuous phase is water and insulating if the continuous phase is oil. We visualized water droplets containing 0.05 wt % rhodamine 6G or oil droplets containing 0.5 wt % pyrene on the microscope slide with a fluorescence confocal microscope (Olympus FluoView FV10i confocal laser scanning microscope systems). We used a 10 magnification lens for visualization and image capturing. Interpretation of Symbolism. In this study, S, W, and O denote the powder, water, and oil phases, respectively. The notations O þ WO, OW þ W, O þ OWO, and WOW þ W denote two-phase regions consisting of an emulsion phase and an excess oil/water phase, while the notation O þ WO þ W denotes three-phase regions consisting of an emulsion phase and two oil and water phases. The weight composition of the powder particles in an oil/solid particle mixture is expressed as R = solid particle/(oil þ solid particle), while that of water is expressed as β = water/(water þ oil þ solid particle).
’ RESULTS AND DISCUSSION Figure 1. Scanning electron microscopic image of microbowls. Scale bar is 5 μm.
Effects of the Composition on the Mixed States. Figure 2a shows the mixed state diagram of microbowl/water/n-dodecane
Figure 2. Mixed state diagrams of microbowls/water/oil ternary systems. (a) Solid particles/water/n-dodecane ternary systems. In the solid particles, the composition ratios of microbowls and spherical particles are as follows: 100:0, 75:25, 50:50, and 0:100 (w/w). (b) The ternary systems consisting of microbowls, water, and some oils, that is, n-dodecane, 1-dodecanol, DMS, PFPE, and their mixtures when R = 0.3. 4558
dx.doi.org/10.1021/la2003707 |Langmuir 2011, 27, 4557–4562
Langmuir
Figure 3. Optical microscopic images of microbowls/water/n-dodecane ternary systems: (a) WO emulsion state, 11:43:46 (wt/wt/wt); (b) OW emulsion state, 4:69:27 (wt/wt/wt); (c) OWO emulsion state, 8:32:60 (wt/wt/wt); and (d) WOW emulsion state, 16:48:36 (wt/wt/ wt). Prior to emulsification, rhodamine B was added to the water phase. Scale bar is 100 μm.
ternary systems. Mixed states were classified into four different regions and were changed with the water composition. At low water compositions, mixtures were in the water-in-oil (WO) emulsion state (Figure 3a). The mixtures were fluid while the grain size of the water droplets on which microbowls were adsorbed was on the order of several hundreds of micrometers. Sedimentation of the water droplets and separation of excess oil were observed in the emulsion systems. Most emulsion droplets had an elongated potato-like shape. Such nonspherical emulsion droplets are rarely generated in the absence of external forces.12,27,28 At high water compositions, the mixtures were in the oil-in-water (OW) emulsion state. In the ternary mixture, spherical oil droplets on which microbowls were adsorbed were dispersed as shown in Figure 3b. Phase inversions induced by the addition of the dispersed phase occur when the particles have intermediate hydrophobicities.2931 In ternary systems, multiple emulsion states were found between the OW and WO emulsion states (Figure 2a, bowl: sphere = 100:0). While OWO emulsions formed in mixtures with low solid particles/n-dodecane ratios (R), WOW emulsions formed in those with high R values. Dispersions of fine oil droplets within larger water droplets were dispersed in a continuous oil phase when R = 0.050.20 (Figure 3c), while those of water droplets within larger oil droplets were dispersed in the water phase when R = 0.30.4 (Figure 3d). The multiple structures of these emulsion systems were identified by fluorescence confocal microscopy (Figure 4). Prior to emulsification, water-soluble rhodamine B and oil-soluble pyrene were added to the water and oil phases, respectively. The images revealed the
ARTICLE
compartmentalization of the oil droplets (blue region) in the inner water droplets (red region) of the OWO emulsions and that of the water droplets in the inner oil droplets of the WOW emulsions (Figure 4a and b). The microbowls covered the oilwater interfaces (Figure 4c). The formation of multiple emulsions by a one-step mixing procedure has also been reported in silica nanoparticles/water/ triglyceride oil ternary systems and silica nanoparticles/ionic liquid/toluene ternary systems.30,32 In these previous studies, multiple emulsions were observed under strict conditions. For example, in the ternary system consisting of hydrophobic silica, oil, and water, a phase inversion occurred from WO to WOW multiple emulsions only when the oil phase was triglyceride.30 Such multiple emulsions were not obtained when the oil phase was a common oil such as silicone oils or hydrocarbons. In the ternary system consisting of hydrophobic silica, toluene, and an ionic liquid, stable oil/ionic liquid/oil multiple emulsions were obtained only when the ratio of SiOH on the silica surface was 47%.32 On the other hand, in the ternary systems containing microbowls, multiple emulsions were obtained under various conditions. As mentioned in the next section, multiple emulsions were obtained in systems containing hydrocarbon, silicone oil, and fluorinated oil. In addition, not only pure microbowls but also mixtures with other surface active particles (spherical silicone resin particles, hydrophobic silica, and hydrophobic mica) stabilized multiple emulsions. These results predicted that microbowls were more likely to form multiple emulsions than other surface-active particles. The oil composition is an important factor that determines whether the particles formed multiple emulsions. Figure 2b shows a mixed state diagram of ternary systems consisting of microbowls, water, and some oils when R = 0.3. The states of the ternary systems of the water compositions β = 00.8 are shown for 1-dodecanol/n-dodecane mixtures, DMS, and PFPE. The phase transition from the WO emulsion state to the OW emulsion state and the formation of WOW emulsions were observed in systems containing n-dodecane, DMS, and PFPE. On the other hand, only WO emulsions were formed in systems containing dodecanol or almost dodecanol/n-dodecane mixtures. The particle shape is also an important factor. Figure 2a shows the mixed state diagrams of ternary systems containing different ratios of microbowls and spherical silicone resin particle mixtures (bowl:sphere = 75:25, 50:50, 0:100). The water composition β at the phase inversion increased with an increase in the spherical particle composition of the powder components. In systems containing only spherical particles, all mixtures were in the WO emulsion state (Figure 2a); the phase inversion to an OW emulsion state and the formation of multiple emulsions were not observed. These emulsion behaviors of spherical silicone particles are caused by their wettabilities. The surface active agents with relatively hydrophobic/lipophilic properties are likely to form WO emulsions.33 As mentioned in the Experimental Section, the sinking times t into water or ndodecane predict that both microbowls and spherical particles were hydrophobic, but the spherical particles had higher affinity for n-dodecane than did the microbowls. Why do microbowls form multiple emulsions only by mixing? In nonionic surfactant/oil/water ternary systems, multiple emulsions were obtained during phase inversion from WO emulsions to OW emulsions when the hydrophile lipophile balance (HLB), viscosity of the oil phase, and stirring intensity were suitable to the system.34 On the other hand, in solid particles/oil/water 4559
dx.doi.org/10.1021/la2003707 |Langmuir 2011, 27, 4557–4562
Langmuir
ARTICLE
Figure 4. Fluorescence confocal microscope images of microbowls/water/n-dodecane ternary systems. Prior to emulsification, rhodamine B and pyrene were added to water and oil phases, respectively. (a) OWO emulsion state, 6:37:57 (wt/wt/wt); and (b,c) WOW emulsion state, 15:50:35 (wt/ wt/wt). Images on the left and middle are fluorescence images at 591 and 461 nm, respectively, while those on the right are their superimpositions. Scale bars are (a,b) 100 μm and (c) 5 μm.
Figure 5. Schematic illustrations on (a) the contact angle hysteresis at the edge of the microbowl surface and (b) imprinting.
ternary systems, multiple emulsions were formed when the particles showed sufficient contact angle hysteresis as well as intermediate hydrophobicity.29,30 In general, the contact angle of solid particles at the oilwater interface is unique and is determined by Young’s law.14 However, at the defects of the solid surfaces, the contact angle is not unique.7,18,19,35 We postulated that microbowls demonstrated large contact angle hysteresis because of their distinctive particle shape. For example, at the edge of the microbowl surface, the contact angle is not unique from θ to θ þ R, as shown in Figure 5a. We hypothesize that the mircobowls performed a double role: a high HLB surfactant and a low HLB surfactant because of the contact angle
hysteresis, while stabilizing both outer and inner drop surfaces of multiple emulsions. Effects of the Mixing Conditions on the Mixed States. In addition to the factors mentioned thus far, the mixing conditions also affect the mixed states of ternary systems. Multiple emulsions were observed only when water phases were added to an oil phase in which the microbowls were already dispersed. When the oil phases were added to the water phases in which the particles were dispersed, all systems were in the OW emulsion state; multiple emulsions as well as WO emulsions were not obtained. A significant finding in the current study was the fact that the surface property of microbowls depended on whether the particles, in their initial states, were in the oil phase or the water phase. Only OW emulsions were formed when the particles were first dispersed in the water phase, although WO, OW, and multiple emulsions were formed when the particles were dispersed in the oil phase. We refer to this phenomenon as “imprinting” after the well-known habit of birds wherein a young bird learns from its parents’ characteristics. One possible explanation for the influence of the particle location on the emulsion type might also be associated with the contact angle hysteresis. The same particles exhibited a different contact angle with the oilwater interface because of the liquid that first established contact with the particles.30 The wettability of solid particles changes drastically with liquid in the holes; they are hydrophilic when the holes are suffused by water and lipophilic when they are suffused by oil or air.36 We have some evidence on the immersion of water or oil into the holes on the microbowls. Figures 5b and 6a are schematic illustrations of 4560
dx.doi.org/10.1021/la2003707 |Langmuir 2011, 27, 4557–4562
Langmuir
ARTICLE
Figure 6. Fluorescence confocal microscope images of (a) microbowls in water and (b) those in n-dodecane. Prior to dispersion in water (n-dodecane), the microbowls were filtered through filter paper after dispersion in n-dodecane (water). Rhodamine B and pyrene were added to the water and oil phases, respectively. Images on the left and middle are fluorescence images at 591 and 461 nm, respectively, while those on the right are their superimpositions. Scale bar is 5 μm.
Figure 7. Optical microscope images of microbowls/water/n-dodecane (8:20:72, wt/wt/wt). Scale bar is 100 μm.
microbowls in water and fluorescence confocal microscopic images, respectively. Prior to dispersion in water, the microbowls were filtered through filter paper after dispersion in n-dodecane. Water-soluble rhodamine B and oil-soluble pyrene were added to water and oil phases, respectively. Fluorescence due to pyrene molecules at a fluorescence wavelength of 461 nm was observed from some holes on microbowl surfaces and indicated that n-dodecane was retained in the holes (blue regions in the middle and right images). Interestingly, black circles were observed in similar images. These results predicted that the microbowls containing oil in the holes and those containing air were mixed in the dispersion. On the other hand, the right illustration in Figure 5b is an image of microbowls in n-dodecane. Prior to dispersion in n-dodecane, the microbowls were filtered after dispersion in water. Fluorescence due to rhodamine molecules at 591 nm indicated that water was retained in the hole or the pores on the microbowl surface. As mentioned, the liquid in the holes or pores on the microbowl surfaces did not exchange easily. This retention ability of microbowls induced the imprinting phenomenon. Change of the Emulsion Droplets during the Drying Process. We found an interesting state during the drying process of the WO emulsions. Figure 7 outlines a change of 5 mg of the WO emulsion stabilized by microbowls on a glass plate at 294 K and 41% relative humidity. A deficit of the emulsion droplets began several minutes after its application on a glass plate. A coffee-bean-like form with a large cavity was subsequently
observed after 1800 s. Wrinkles were formed on the surface of the emulsion droplets in the case of spherical silicone resin particles, while the formation of a large cavity or wrinkles was not observed in the case of the emulsions stabilized by surfactant molecules. The formations of the elongated potato-like and coffee-beanlike emulsions were significant because they contradicted the common assumption that emulsions and bubbles exist in the spherical form with minimum energy.7,12 When particles adsorb at an oilwater interface at sufficiently high concentrations, the interface loses mobility and displays solid-like characteristics. This phenomenon is called “interfacial jamming”.10 Jamming can arrest interfacial tension-driven morphological coarsening in oil/ water systems and can, therefore, stabilize morphologies with unusual interfacial shapes, for example, nonspherical drops and bubbles. Under shear flow in a mixing process, deformed emulsion droplets return to spherical droplets upon relaxation.37 However, the deformed shape is maintained when interfacial jamming is formed at the oilwater interface.11,27,28 Effects of interfacial jamming on the shape of Pickering emulsions or foams have been reported in previous papers.12,13 Subramaniam et al. have shown that gas bubbles can exist in nonspherical shapes if the bubbles are compressed between two glass plates.12 Studurt et al. demonstrated that partially coated droplets generated in a microcapillary device can undergo spontaneous coalescence into stable nonspherical structures.13 4561
dx.doi.org/10.1021/la2003707 |Langmuir 2011, 27, 4557–4562
Langmuir
’ CONCLUSIONS In the present study, we found that microbowls exhibited anomalous emulsification behaviors such as the formation of multiple/nonspherical emulsions and drastic hysteresis that is associated with imprinting. These findings suggested that microbowls are useful in achieving unique formulations for drug delivery systems, functional cosmetics, and food. The hole of the microbowl was decisive in determining the formation of such complex structures and will be a critical element in designing building blocks of supracolloidal systems as well as particle wettabilities. ’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT This study was supported by a Grant-in-Aid for Young Scientists (Start-up) 19810001 from the Ministry of Education, Culture, Sports, Science, and Technology, Japan, and a grant from The Cosmetology Research Foundation. We thank Professor Kaoru Tsujii and Professor Shigeyuki Komura for their helpful suggestions and Takemoto Oil and Fat Co. for donating the microbowls. ’ REFERENCES
ARTICLE
(21) Cheng, L.; Zhang, G.; Zhu, L.; Chen, D.; Jiang, M. Angew. Chem., Int. Ed. 2008, 47, 10171–10174. (22) Noda, I.; Kamoto, T.; Sasaki, Y.; Yamada, M. Chem. Mater. 1999, 11, 3693–3701. (23) Nonomura, Y.; Ikeda, T.; Sugimoto, M.; Taniguchi, T. Chem. Lett. 2011, 40, 136–137. (24) Garti, N.; Aserin, A.; Tiunova, I.; Binyamin, H. J. Am. Oil Chem. Soc. 1999, 76, 383–389. (25) Sekine, T.; Yoshida, K.; Matsuzaki, F.; Yanaki, T.; Yamaguchi, M. J. Surf. Deterg. 1999, 2, 309–315. (26) Arditty, S.; Schmitt, V.; Giermanska-Kahn, J.; Leal-Calderon, F. J. Colloid Interface Sci. 2004, 275, 659–664. (27) Subramaniam, A. B.; Mejean, C.; Abkarian, M.; Stone, H. A. Langmuir 2006, 22, 5986–5990. (28) Subramaniam, A. B.; Abkarian, M.; Mahadevan, L.; Stone, H. A. Langmuir 2006, 22, 10204–10208. (29) Binks, B. P.; Lumsdon, S. O. Langmuir 2000, 16, 2539–2547. (30) Binks, B. P.; Rodrigues, J. A. Langmuir 2003, 19, 4905–4912. (31) Nonomura, Y.; Kobayashi, N. J. Colloid Interface Sci. 2009, 330, 463–466. (32) Binks, B. P.; Dyab, A. K. F.; Fletcher, P. D. I. Phys. Chem. Chem. Phys. 2007, 9, 6391–6397. (33) Bancroft, W. D. J. Phys. Chem. 1913, 17, 501–019. (34) Salager, J. L.; Forgiarini, A.; MaRquez, L.; Pe~ na, A.; Pizzino, A.; Rodriguez, M. P.; Rondon-Gonzalez, M. Adv. Colloid Interface Sci. 2004, 108109, 259–272. (35) Mayama, H.; Nonomura Y. Langmuir, DOI: 10.1021/ la104600x. (36) Nonomura, Y.; Komura, S. J. Colloid Interface Sci. 2008, 317, 501–506. (37) Becu, L.; Benyahia, L. Langmuir 2009, 25, 6678–6682.
(1) Pickering, S. U. J. Chem. Soc. 1907, 91, 2001–2021. (2) Binks, B. P. Curr. Opin. Colloid Interface Sci. 2002, 7, 21–41. (3) Dinsmore, A. D.; Hsu, M. F.; Nikolaides, M. G.; Marquez, M.; Bausch, A. R.; Weitz, D. A. Science 2002, 298, 1006–1009. (4) Noble, P.; Cayre, O. J.; Alargova, R. G.; Velev, O. D.; Paunov, V. N. J. Am. Chem. Soc. 2004, 126, 8092–8093. (5) Aussilous, P.; Quere, D. Nature 2001, 411, 924–927. (6) Binks, B. P.; Murakami, R. Nat. Mater. 2006, 5, 865–869. (7) de Gennes, P.-G.; Brochard-Wyart, F.; Quere, D. Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves; Springer: Berlin, 2003; Chapter 1, p 9. (8) Nonomura, Y.; Sugawara, T.; Kashimoto, A.; Fukuda, K.; Hotta, H.; Tsujii, K. Langmuir 2002, 18, 10163–10167. (9) Nonomura, Y.; Fukuda, K.; Komura, S.; Tsujii, K. Langmuir 2003, 19, 10152–10156. (10) Stratford, K.; Adhikari, R.; Pagonabarraga, I.; Desplat, J.-C.; Cates, M. E. Science 2005, 309, 2198–2201. (11) Herzig, E. M.; White, K. A.; Schofield, A. B.; Poon, W. C. K.; Clegg, P. S. Nat. Mater. 2007, 6, 966–971. (12) Subramaniam, A. B.; Abkarian, M.; Mahadevan, L.; Stone, H. A. Nature 2005, 438, 930. (13) Studart, A. R.; Shum, H. C.; Weitz, D. A. J. Phys. Chem. B 2009, 113, 3914–3919. (14) Levine, S.; Bowen, B. D.; Partridge, S. J. Colloids Surf. 1989, 38, 325–343. (15) Mbamala, E. C.; von Gr€unberg, H. H. J. Phys.: Condens. Matter 2002, 14, 4881–4900. (16) Pieranski, P. Phys. Rev. Lett. 1980, 45, 569–572. (17) Utada, A. S.; Lorenceau, E.; Link, D. R.; Kaplan, P. D.; Stone, H. A.; Weitz, D. A. Science 2005, 308, 537–541. (18) Nonomura, Y.; Komura, S.; Tsujii, K. Langmuir 2004, 20, 11821–11823. (19) Nonomura, Y.; Komura, S.; Tsujii, K. J. Phys. Chem. B 2006, 110, 13124–13129. (20) Nonomura, Y.; Komura, S.; Tsujii, K. J. Oleo Sci. 2004, 53, 607–610. 4562
dx.doi.org/10.1021/la2003707 |Langmuir 2011, 27, 4557–4562
