Langmuir 2005, 21, 7083-7089
7083
Articles Spontaneously Formed trans-Anethol/Water/Alcohol Emulsions: Mechanism of Formation and Stability Natalia L. Sitnikova, Rudolf Sprik,* and Gerard Wegdam van der Waals-Zeeman Instituut, Universiteit van Amsterdam, Valckenierstraat 65-67, 1018 XE Amsterdam, The Netherlands
Erika Eiser van’t Hoff Instituut voor Moleculaire Scheikunde, Universiteit van Amsterdam, Nieuwe Achtergracht 166, 1018 TV Amsterdam, The Netherlands Received December 22, 2004. In Final Form: May 27, 2005 We studied the spontaneous emulsification and droplet growth mechanism in trans-anethol/water/ ethanol solutions, also known as the beverage ouzo, using dynamic light scattering spectroscopy. This simple ternary mixture is a generic example of a system that forms microemulsions spontaneously when brought into the two-phase region. The volume fraction of the dispersed phase was found to profoundly affect the growth rates of the droplets, which is a new finding that has not been predicted by the LifshitzSlyozov-Wagner theory. Time-dependent measurements show that the droplet growth is governed by Ostwald ripening (OR), and no coalescence was observed. Furthermore, the OR rates increase with increasing oil concentration at low alcohol content. We attribute this behavior to enhanced droplet-droplet interactions. At high ethanol concentrations, we found that the measured rates decreased as the oil concentration increased. The OR growth mechanism completely correlates with changes in droplet size. The kinetics of droplet growth shows that the ripening has a saturation limit at a droplet radius of about 1.5 µm. Thus, formed emulsions remain stable for months.
1. Introduction Spontaneously formed emulsions are created by simply bringing two immiscible liquids into contact and waiting until they emulsify without having to apply any external thermal or mechanical energy source.1 The motivation for improving the understanding of the spontaneous emulsification process is its extensive use in industrial applications. For instance, many products used in the food industry,2 detergents,3 body-care products,4 agricultural applications,5 drug-delivery systems,6 and many others come in the form of emulsions that need to stay stable over a long period of time. Fifty years ago, Davies and Rideal discussed three mechanisms by which spontaneous emulsification can be produced, namely, by transient negative interfacial tension, interfacial turbulence, or “diffusion and stranding”.7 The first two involve a mechanical breakup of the interface, and the third one involves a chemical instability mechanism. A good example of the diffusion and stranding * Corresponding author. E-mail:
[email protected]. (1) Miller, C. A. Colloids Surf. 1988, 29, 89-102. (2) Wakerly, M. G.; Pouton, C. W.; Meakin, B. J.; Morton, F. S. Am. Chem. Soc. Symp. Ser. 1986, 311, 242-255. (3) Salager, J. L. In Handbook of Detergents - Part A: Properties; Broze, G., Ed.; Marcel Dekker: New York, 1999; p 253. (4) Simion, F. A.; Starch, M. S.; Witt, P. S.; Woodford, J. D.; Edgett, K. L. In Textbook of Cosmetic Dermatology, 2nd ed.; Baran, R., Maibach, H. I., Eds.; Martin Dunitz: London, 1998; p 309. (5) Groves, M. J.; Mustafa, R. M. A. J. Pharm. Pharmacol. 1974, 26, 671-681. (6) Pouton, C. W. Eur. J. Pharm. Sci. 2000, 11, S93-S98. (7) Davies, J. T.; Rideal, E. K. Interfacial Phenomena; Academic Press: New York, 1963; p 359.
mechanism is the toluene/alcohol/water system. Because of the condensation of one liquid upon diffusive separation of the second one, the alcohol diffuses into the water phase, depleting the toluene/alcohol mixture. Consequently, this dilution of alcohol inside the toluene/alcohol mixture leaves the oil stranded in the water phase, forming micrometersized droplets. Such microemulsions appear cloudy at the interfacial region separating the bulk toluene/alcohol from the water phase. Despite its relevance to industrial processes and everyday life, the mechanism and the dynamics of spontaneous emulsification are still poorly understood, and experiments are scarce. Only few recent experimental studies deal with the equilibrium properties of truly spontaneously formed emulsions.8,9 For instance, Vitale and Katz showed that the formation of emulsions in the divinyl benzene (DVB)/ethanol/water system is a result of the so-called “ouzo” effect (relating to the Greek alcoholic beverage ouzo). It occurs in the metastable spinodal decomposition region of the phase diagram because of oil supersaturation upon sudden dilution with water. The size of the emulsion droplets is on the order of micrometers, varying with the proportions of the three components and the temperature. It was also shown that spontaneous emulsification can occur only in the case of a suitable solventshere and in ouzo it is the alcoholsthat is miscible in both oil and water in all proportions. The dispersions that Vitale and Katz investigated also showed that the emulsions formed because of the ouzo effect can be stable (8) Vitale, S. A.; Katz, J. L. Langmuir 2003, 19, 4105-4110. (9) Grillo, I. Colloid Surf., A 2003, 225, 153-160.
10.1021/la046816l CCC: $30.25 © 2005 American Chemical Society Published on Web 06/30/2005
7084
Langmuir, Vol. 21, No. 16, 2005
over a period of a few days only, when the oil density was equal or close to that of the continuous aqueous phase, and the oil had to be insoluble in water. Recently, the commercial alcoholic beverage pastis, which is composed of anethol (with some aromatic molecules), water, and ethanol, and a model pseudoternary system composed of the anethole/ethanol/D2O + H2O system were studied by means of small-angle neutron scattering.9 trans-Anethol is the essential ingredient of pastis, ouzo, and other anise-based beverages that are consumed in the form of emulsions.8,9 The initial size of the droplets was found to be less than 1 µm in diameter and was observed to be growing with time and temperature. However, no systematic studies on the growth mechanism or stability of the droplets were made. Theoretical studies closely related to the ouzo effect have been described by Miller and Ruschak.10 They developed a diffusion-path theory that is able to predict spontaneous emulsification and proposed the solutions to the diffusion equations for two semi-infinite nonequilibrium phases in contact. These solutions allow us to make predictions about the initial evolution of the droplet formation. They confirmed this experimentally for several ternary systems, in particular, for the toluene/water/ ethanol and the toluene/water/propanol systems in which spontaneous emulsification develops locally. Hence, near the initial interface diffusion produces a region of local supersaturation whenever a composition within the twophase region of the phase diagram is chosen. Miller and Ruschak were also able to predict the phase in which this emulsification should occur and showed that convection due to interfacial turbulence does not noticeably influence the emulsification process itself but only the thickness of the supersaturated region near the interface. There are theoretical studies11-13 on the dynamics of the spontaneous emulsification process in the presence of surfactants. For example, Cates and co-workers13 described a model for spontaneous emulsification resulting from transiently negative interfacial tension between water and oil regions that were formed under conditions of strong adsorption of surfactant molecules to the interface. The main result of this work is that the diffusion of the surfactant molecules to the interface determines the growth and finite size of the microemulsion droplets. However, authors described another scenario of spontaneous emulsification (irrelevant in the context of the ouzo effect) when the oil/water negative interfacial tension gives rise to hydrodynamic instability in the system. Here we have studied the mechanism and kinetics of droplet formation in the ternary trans-anethol/ethanol/ water system using mainly dynamic light scattering (DLS). This system forms spontaneously stable emulsions without the need for mechanical stirring or adding surfactants and represents a restricted class of systems for the study of self-emulsifying processes. Another crucial aspect of the anethol/ethanol/water emulsions is their stability, which can be rather significant (a few months), unlike all other self-emulsifying ternary systems studied so far.7-10 This is of great interest because this stability is achieved in the absence of any added surfactant, making it unique among ternary emulsions. A better understanding of the (10) Ruschak, K. J.; Miller, C. A. Ind. Eng. Chem. Fundam. 1972, 11, 534-540. (11) Dynamics and Instability of Fluid Interfaces; Sorensen, T. S., Ed.; Lecture Notes in Physics; Springer: Berlin, 1979. (12) Szleifer, I.; Kramer, D.; Ben-Shaul, A.; Gelbart, W. M.; Safran, S. A. J. Chem. Phys. 1990, 92, 6800-6804. (13) Granek, R.; Ball, R. C.; Cates, M. E. J. Phys. II 1993, 3, 829849.
Sitnikova et al.
dynamics of such spontaneously formed particles is useful both for the design of new separation or transport processes and the development of products for various commercial applications. Here we report on the initial growth rate of the droplets and their final size. 2. Materials and Methods 2.1. Materials. All emulsions were prepared using purified Millipore Direct-Q water with a resistivity of 18.25 Ω/sm2. Ethanol (>99,7% pure) and trans-anethol with Mw ) 148.2. (>99% pure) were obtained from Aldrich and used as received. We checked the trans-anethol oil (molecular formula C10H12O) for traces of cis-anethol by gas chromatography and NMR spectroscopy. Only a very low concentration of cis-anethol was found, excluding the possibility of its influence on the stability of our emulsions. Gas chromatography (GC) was used to confirm the purity of trans-anethol and to make sure that surfactants were absent. GC analysis was carried out on an Interscience GC-8000 gas chromatograph with a 100% dimethylpolysiloxane capillary column (DB-1, 30 m, 0.325 mm). The carrier gas was nitrogen, and the solvent used to dilute the sample was ethanol. The measurements were performed starting on an isotherm at 110 °C (2 min). Then the temperature was increased by 30 °C min-1 to 280 °C and kept there for 5 min. The anethol peak in the GC chromatogram was found to be coexistent with a minor fraction of anisaldehyde or anisic acid, which was less than 1% and could not act as a surfactant or cosurfactant. No other contaminants have been found. 2.2. Emulsion Preparation. All emulsions were prepared by the same procedure: to create the emulsions, first small amounts of trans-anethol were completely dissolved in ethanol. Subsequently, water was added to the homogeneous single-phase solution to dilute the ethanol content to the desired weight fraction. The presence or absence of emulsification was registered visually. To minimize mixing effects due to external forces, the oil/alcohol mixtures were prepared in tubes of 1 cm diameter, and the water was subsequently added slowly to avoid intense turbulent mixing at the interface. Because the initial area of the interface was small, this process was most successful in minimizing disturbance during contacting. The emulsions were formed without any vigorous mechanical agitation, and immediate whitening was observed after the mixtures were brought into contact with water. Emulsions were prepared at several oil and alcohol concentrations. Immediately after the preparation, the droplet growth was followed with DLS. 2.3. Experimental Techniques. Dynamic Light Scattering. The droplet sizes were analyzed using photon correlation spectroscopy with an ALV-5000 digital multiple τ-correlator (ALV-laser GmbH, Langen, Germany) using a He-Ne laser with λ ) 632.8 nm as the light source. In a typical experiment, the initial average radius of the droplets was around 500 nm, and their growth rate was followed in time until they reached a final average radius of 1-1.5 µm. Because of the low weight fraction of anethol (2 × 10-5-2 × 10-3) used in the experiments, no further dilution was required to perform the size analysis. This allowed for sequential particle size measurements directly in the cuvette first at close intervals of 1 min and later for the duration of a few minutes. Most of the measurements were performed by measuring the scattered light in the single light scattering homodyne regime. Under these conditions, multiple light scattering may influence the results of the analysis. The negligible role of multiple scattering in the experiments was verified by measurements using cross-correlation dynamic light scattering14 as well as by the angular dependence. The incoherent scattering coming from multiple scattered light within the center of the beam was so small that we did not detect any significant multiple scattering contributions. In fact, all of our measurements showed a coherence factor close to unity, even for turbid samples, indicating that our data were nearly perfectly coherent (Figure 1). In addition, angular dependence measurements were performed to ensure that the droplets behave purely diffusively. (14) Schhroder, J.-M.; Wiegand, S. Phys. Chem. Chem. Phys. 2000, 2, 1493-1495.
trans-Anethol/Water/Alcohol Emulsions
Figure 1. Auto-correlation functions for the same ouzo sample with Coil ) 0.2 mg/mL in a water/ethanol mixture (0.90/0.10 wt %) 60 s (|), 2 days (b), and 45 days (O) after the preparation. Turbidity. In addition to DLS experiments, some spectroscopic turbidity measurements were also performed using a Varian Cary 50 (Australia) photospectrometer. From the turbidity spectra obtained for wavelengths varying between 250 and 1000 nm, a rapid characterization of the droplet size was made. The analysis of the spectral turbidity is based on Mie-scattering calculations of the droplets in the solvent environment.15 The results are identical to the DLS results reported here. Some additional direct visual observations of the evolving emulsions were performed using a basic optical transmission microscope (Bausch and Lomb) with a 40× objective directly immersed into the solution. pH. The pH of emulsion solutions was measured using an Orion microprocessor analyzer/410A with a fine control of about 1%. All experiments were performed at 20 ( 1 °C. Analysis of Size Distribution Measurements. Most of the light scattering experiments were performed in the homodyne regime. In such an experiment, the thermal fluctuations of the colloidal particles or emulsion droplets in the solution generated a fluctuating scattered intensity signal on a photon counter that was placed at a fixed scattering angle θ. The fluctuations are related to the Brownian movement of the scattered particles and can be analyzed in terms of time-dependent autocorrelation functions g2(τ) for the intensity and g1(τ) for the field that are related by the Bloch-Siegert relation g2(τ) ) 1 + β[g1(τ)]2, where β is the coherence factor depending on the experimental conditions.16,17 For monodisperse particles in solution, the field-correlation function decays exponentially, g1(τ) ) exp(-Γτ), with a decay rate of Γ ) Dq2, where D is the diffusion coefficient of the particles and q is the scattering wave vector, q ) (4πn/λ)sin(θ/2), with n being the refractive index of the solvent, λ being the wavelength of the laser in vacuum, and θ being the scattering angle. The Stokes-Einstein relation, D ) kBT/6πηRh, relates the diffusion coefficient to the hydrodynamic radius Rh; kB is Boltzmann’s constant, T is the temperature, and η is the dynamic viscosity of the solvent. The viscosity data for water/ethanol mixtures were taken from the literature.18 Typical results of a measurement made over a period of time on one and the same sample are illustrated in Figure 1. Good single-exponential behavior of g2(τ) -1 (or β[g1(τ)]2) is found in each case, reflecting the fact that multiple scattering is negligible and the droplets in the emulsions are quite monodisperse. Testing the diffusivity of the emulsion droplets, we measured and analyzed the spectra of two representative samples for different scattering angles (Figure 2). Plotting the correlation function versus q2 for samples low and high in water concentration showed that all curves fall on the same master curve, reassuring us that the droplet motion in these samples is purely diffusive. Hence, the absence of any angle dependence allowed us to make all of (15) Mie, G. Ann. Phys. 1908, 25, 377. (16) Koppel, D. E. J. Chem. Phys. 1972, 57, 4814-4820. (17) Berne, B. J.; Pecora, R. Dynamic Light Scattering; Kreiger: Malabar, FL, 1990. (18) Transport Phenomena Data Companion; Janssen, L. P. B. M., Warmoeskerken, M. M. C. G., Eds.; Delftse Uitgevers MaatschappijI11: Delft, The Netherlands, 1982; p 144.
Langmuir, Vol. 21, No. 16, 2005 7085
Figure 2. Angular dependence of the mean droplet radius (from cumulant analysis) for emulsions with Coil ) 0.0023 (b) and 0.01 g/mL (O) in a water/ethanol mixture (0.80/0.20 wt %).
Figure 3. Phase diagram of the ternary trans-anethol/water/ ethanol system. The solid dots present the miscibility limit. our measurements at fixed scattering angle θ ) 90° (Figure 2). The correlation data were processed using the cumulant method.16,17 As was pointed out by de Smet et al.,19 corrections due to changes in the size distribution may lead to a difference between the intensity and number-weighted cumulant analysis. To accurately analyze the effect of changes of the particle size distribution as function of time and the influence on the observed average radius, we also used nonlinear fitting techniques that are part of the standard analysis methods of the ALV setup.
3. Results and Discussions Phase Diagram. Figure 3 shows our measured ternary phase diagram for the three-component trans-anethol/ ethanol/water mixture. The phase diagram was constructed by titrating various oil/ethanol mixtures with distilled water. The solid circles in Figure 3 represent the experimentally determined miscibility limit of the oil component in the system. All solution compositions above these points are in the transparent single phase. The onset of the ouzo effect or rather the two-phase region was determined by visual observations from the change of transparent to turbid solutions, which always occurred immediately after the components were brought into contact. In most instances, the emulsion was easily detectable. It should be mentioned that the far right side of the phase diagram at high concentrations of oil represents the region of reverse microemulsions consisting of small water droplets dispersed in the continuous anethol-rich phase. Here we concentrate only on the more common oil-in-water emulsion. When spontaneous emulsification commences, the solution rapidly moves to the metastable region between the binodal and spinodal lines. It occurs because the (19) De Smet, Y.; Deriemaeker, L.; Parloo, E.; Finsy, R. Langmuir 1998, 15, 2327-2332.
7086
Langmuir, Vol. 21, No. 16, 2005
Figure 4. Partial phase diagram determined for T ) 20 °C: to the right of the triangles begins the region where the ouzo effect becomes observable (2 - one-phase region, O - stable ouzo region, b - unstable ouzo and two-phase separated region).
alcohol, as it diffuses from the oil into the water, carries with it some of the anethol molecules that still have a finite solubility in alcohol and alcohol/water mixtures. As the alcohol diffuses further into the water, the associating oil molecules become expelled from the water-rich solution and are stranded in the form of fine emulsion droplets. Thus, adding water to a solution of homogeneous oil/ ethanol mixtures leads to an abrupt decrease of the solubility of oil in the water-rich continuous phase. This effect causes the strong local concentration fluctuations of solute molecules, and homogeneous nucleation can start.8 Here we also should stress that the ouzo effect is observed only when the water content of the initially homogeneously mixed ternary solution was increased suddenly, but other approaches to the same metastable state (the same composition of the ternary system) were not successful. Whether we added the oil directly to the water/ethanol mixture or alcohol to the oil/water solution, no spontaneous emulsification was observed. This indicates that the emulsification is strongly related to the kinetics of the system. Our experiments also confirm the conclusion of previous investigators7,10 that interfacial tension and mechanical turbulence do not significantly influence the conditions for spontaneous emulsification. For example, when pure water was added to the solutions containing 0.08 wt % anethol and 10 wt % alcohol, no emulsification occurred, although interfacial turbulence is present. When the composition of anethol was changed from 0.08 to 0.1 wt % while keeping the fraction of alcohol constant, emulsification immediately was observed after these solutions were brought into contact. Our observations at different temperatures showed that spontaneous emulsification can be also produced by temperature changes. For example, when a single-phase mixture near the phase-separation line was cooled by 3 °C, emulsification was also produced. Thus, both of these results give evidence that a local phase transition can produce emulsions immediately when passing the binodal. Figure 4 shows the experimentally determined metastable ouzo region of our ternary system at room temperature. This partial phase diagram shows that the ouzo region (open circles) is very narrow with respect to anise oil concentrations and not well represented by a ternary phase diagram. Therefore, we present only the magnified left part of triangular phase diagram using a logarithmic scale for the oil fraction. Here it should be noted that we actually observed two ouzo regimessa stable and an
Sitnikova et al.
Figure 5. Semilogarithmic plot of the time dependence of R (b, left axis) and the scattering intensity (O, right axis) for emulsion droplets of 0.04 wt % oil in an ethanol/water mixture (0.05/0.95 wt %).
unstable one. Of course, thermodynamically the two-phase region always consist of “nucleation and growth” and “spinodal decomposition” regions. Hence, the unstable ouzo regime lies in the vicinity of the spinodal curve where the milky ouzo phase coexist with macroscopically sized oil droplets that float on top of the sample container (the relative density of anethol being about 0.983 g/cm3) and initially a few drops with unknown composition on the bottom of the mixture. Finally, after a few days, these samples coalesce as a result of the spinodal decomposition mechanism.20 This mechanism is relatively fast because already small density fluctuations lead to a spontaneous phase separation into oil- and water-rich regions that become unstable and grow because of the coalescence of neighboring droplets.21 Initial Growth Rate. The stable ouzo region appears when extremely small droplets are formed. These initial droplets grow solely because of diffusion and therefore very slowly in time. This process creates long-lived micrometer-sized droplets. Our particular laboratory made ouzo samples that remained stable for 2 to 3 months after their preparation, depending on mixture composition. Figure 5 shows a typical plot of the droplets’ average radius as a function of time. The change in droplet radius has two distinct time stages: (1) a spontaneous emulsification and initial growth regime accompanied by an increase in scattering intensity and (2) an equilibrium regime where the droplets do not grow anymore. Furthermore, the latter also shows a slight decrease in scattering intensity. It is useful to describe the earlier and later stages separately. Our data suggest that the initial growth is due to molecular diffusion, known as Ostwald ripening,22 rather then aggregation-diffusion23 or coalescence.24 Coalescence and Ostwald ripening can both be monitored by following changes in the mean droplet size over time. A distinction between them is the difference in the growth rates: exponential versus linear. Coalescence is the formation of a large droplet from the merging of smaller ones with subsequent thinning and disruption of the liquid film that covers them. This requires that the small droplets come into contact. According to the Van der Tempel theory,24 coalescence occurs when the volume of the particles increases exponentially with time (20) Vrij, A. Discuss. Faraday Soc. 1996, 42, 23-33. (21) Kabalnov, A. S.; Wennerstrom, H. Langmuir 1996, 12, 276292. (22) Kabalnov, A. S.; Pertzov, A. V.; Shchukin, E. D. J. Colloid. Interface Sci. 1987, 118, 590-597. (23) Atkins, P. W. Physical Chemistry, 6th ed.; Oxford University Press: Oxford, U.K., 1999; Chapter 27. (24) Van den Tempel M. Recl. Trav. Chim. Pays-Bas 1953, 72, 433442.
trans-Anethol/Water/Alcohol Emulsions
R3 ) R03 exp(Kt)
Langmuir, Vol. 21, No. 16, 2005 7087
(1)
where R0 and R are the initial and actual particle radii, respectively, K is the system specific coalescence constant, and t is the time. The coarsening due to Ostwald ripening (OR), however, does not depend on droplet coalescence but on the diffusive transport of dissolved matter through the dispersion medium.25 The driving force in this mechanism is the difference in Laplace pressure between the droplets having different radii, favoring the growth of the large droplets at the expense of smaller ones.26 According to the meanfield description developed by Lifshitz, Slyozov, and Wagner (LSW),27-29 the rate of Ostwald ripening, υ, is characterized by a size growth that is linear in time (a3 ≈ t)
υ)
da3 4 ) RDC∞ dt 9
(2)
where a is the average radius of the particles, D is the dispersed-phase molecular diffusion coefficient in the continuous phase, C∞ is the aqueous solubility of the dispersed phase, and R is a material-dependent constant called the capillary length. It is defined as
R)
2φγ RT
(3)
where φ is the molar volume of the dispersed phase, γ is the interfacial tension between the dispersed and the continuous phases, and R and T are the gas constant and the absolute temperature. For this growth mechanism, the increase in volume is proportional to the solubility of the oil-rich phase dispersed in the continuous solvent phase. To identify the ripening mechanism for our spontaneously formed emulsions, we measure the time evolution of the mean radius of the droplets immediately after mixing the components in short intervals. Figure 6 (a and b) shows the plots of the droplets’ volume, R3, as a function of time at various oil concentrations in fixed alcohol/water (0.05/0.95 and 0.30/0.70 wt. %) mixtures. As can be seen from Figure 6, all of the curves are linear within experimental error. The exponential growth predicted for coagulation (eq 1) does not describe the observed behavior over the full observation time shown in Figure 6. The observed linear behavior indicates that the primary mechanism governing aging is Ostwald ripening. We also followed the evolution of the solution under an optical microscope with a 40× objective. Although the size of the particles cannot be measured accurately with the microscope, the position and motion can be followed easily. The observations showed that the droplets move by Brownian motion and meet frequently but do not coagulate or merge. These qualitative optical observations support the Ostwald ripening growth mechanism in the emulsion. The radius of the oil-rich droplets strongly depends on the composition of the ternary mixture and hence how deeply the ternary mixture is quenched into the nucleation and growth region and grows faster at higher oil concentrations. To our knowledge, we are the first to analyze (25) Taylor, P. Colloids Surf. 1995, 99, 175-185. (26) Kabalnov, A. S. Langmuir 1994, 10, 680-684. (27) Lifshitz, I. M.; Sloyzov, V. V. J. Phys. Chem. Solids 1961, 19, 35-42. (28) Kabalnov, A. S.; Shchukin, E. D. Adv. Colloid Interface Sci. 1992, 38, 69-97. (29) Taylor, P. Adv. Colloid Interface Sci. 1998, 75, 107-163.
Figure 6. Plot of the volume of the emulsion droplets as a function of time at different concentrations of trans-anethol in ethanol/water mixtures of (a) 0.05/0.95 wt % and (b) 0.30/0.70 wt %. This is shown to clarify the fact that the slopes are calculated by making the assumption that the emulsification process follows a straight line in the initial growth regime and they have a different direction depending on the ethanol content in the mixtures.
the oil droplets’ growth in surfactant-free alcohol/water mixtures. In some recent publications,26,30-33 the influence of surfactant or surfactants plus cosurfactants on the Ostwald ripening in oil/water emulsions has been reported. There it was shown that the volume fraction of the dispersed phase has a negligible effect on the ripening rates at low dispersed-phase volume fractions.34-35 Taylor and co-workers36 found only a slight increase in the rates with decreasing volume fraction of the dispersed phase. However, we observed a rather profound effect of the volume fraction of the dispersed phase on the Ostwald ripening and in particular on the final droplet size of the ouzo emulsions. At ethanol concentrations below 0.3 wt % in the water/ethanol mixture, the OR rate is seen to increase with an increase of oil concentration, whereas at alcohol fractions above 0.3 wt % the rates were found to decrease as the trans-anethol concentration increased (Table 1). The increase in OR with increasing oil fraction in the mixtures, for ethanol concentrations of
