- Email: [email protected]
Some Aspects of Stability of Multiple Emulsions in Personal Cleansing Systems
Journal of Colloid and Interface Science 256, 208–215 (2002) doi:10.1006/jcis.2002.8326
Some Aspects of Stability of Multiple Emulsions in Personal Cleansing Systems Tirucherai V. Vasudevan1 and Mark S. Naser Unilever Research U.S., 45 River Road, Edgewater, New Jersey 07020 Received September 14, 2001; accepted February 27, 2002; published online April 23, 2002
implement in shampoo and body wash applications due to the presence of high concentration of high HLB surfactants. Multiple emulsion has found applications in pharmaceuticals and cosmetics (3–5). However, the systems employed in these applications contained only a low level (<5 wt%) of high HLB surfactant in the external aqueous phase. Since high HLB surfactants greatly influence the interfacial properties of the low HLB emulsifier, and hence the stability of the multiple emulsion, these studies have only a limited value in providing guidance for formulating stable multiple emulsions in personal cleansing systems. In this paper, the effect of different low HLB emulsifiers, as well as oils, on the stability of multiple emulsion has been presented and the mechanism of breakdown of multiple emulsions in the presence of high HLB surfactants has been elucidated. Only polymeric low HLB emulsifiers were used in this study since the literature information indicates that monomeric emulsifiers cannot provide long term stability in multiple emulsion systems (6, 7). Also, all the experiments were carried out using sodium chloride solution as the internal aqueous phase to modulate the osmotic pressure imbalance as well as to minimize Ostwald ripening. These two factors have been shown to play a critical role in determining multiple emulsion stability (8). Three types of oil, namely mineral oil, vegetable oil, and silicone oil, were used in the study since multiple emulsion yields have been reported to depend on the oil type with mineral oils providing higher yields than vegetable oils (9). The higher yield obtained with the mineral based oils compared to the vegetable based oils was reported to be due to the greater interaction of emulsifiers with mineral oils than vegetable oils.
The two dominant factors that were found to affect the stability of multiple emulsions in high HLB surfactant systems are the osmotic pressure imbalance between the internal aqueous phase and the external aqueous phase, and the adsorption/desorption characteristics of the emulsifier/surfactant film at the oil/water interface. Synergistic interaction between the low HLB emulsifier and the high HLB surfactant that produces very low interfacial tension of the order of 10−2 mN/m at the oil/water interface was found to occur in some of the systems investigated. Long term stability was observed in multiple emulsion containing these systems. However, no synergy was observed in systems in which either the oil or the emulsifier, or both, contained unsaturated chains. In fact, desorption of the adsorbed surfactant film was observed in systems containing unsaturated chains. The observed desorption from the interface of the emulsifier in these systems was attributed mainly to the inability of the unsaturated chains to form a close packed, condensed interfacial film. Presence of closely packed, condensed interfacial film is necessary to prevent solubilization of the adsorbed low HLB emulsifier by the high HLB surfactant. Multiple emulsions prepared using systems containing unsaturated hydrocarbons were highly unstable. C 2002 Elsevier Science (USA) Key Words: multiple emulsions; cleansing systems; high surfactant systems; unsaturated oils; unsaturated emulsifiers.
INTRODUCTION
In the past several years, major efforts have been devoted to obtaining enhanced deposition of oils from shampoos and body washes. Several new technologies have emerged from this endeavor. The important technologies among these are cationic polymer assisted deposition for colloidal sized particles (1) and direct transfer of large oil droplets (2). However, deposition of water soluble benefit agents has remained a daunting task. Deposition of water soluble benefit agents onto skin from wash-off systems can potentially be achieved by incorporating them within either an oil (multiple emulsion technology) or in liposomes/niosomes and depositing the oil or liposome using existing methods such as polymer assisted deposition and direct deposition. Liposome/niosome technology is rather difficult to
EXPERIMENTAL
Materials The materials used to prepare the multiple emulsions are listed in Table 1. Equipment Homogenizer—Mixer
1
To whom correspondence should be addressed. Current address: Unilever HPC U.S.A., 3100 Golf Road, Rolling Meadows, IL 60008. E-mail: vasu. [email protected]. 0021-9797/02 $35.00
C 2002 Elsevier Science (USA)
All rights reserved.
Primary emulsions (W1 /O) were prepared using a Janke & Kunkel RE 162P mixer equipped with a variable speed 208
209
STABILITY OF MULTIPLE EMULSIONS
TABLE 1 Materials Material
Supplier
Sodium chloride
Fisher
Trade name
Light mineral oil Octyl palmitate Silicone oil Sunflower oil
Fisher Goldschmidt Dow Corning Welch, Clarke & Holme Inc.
Tegosoft OP DC200 fluid 50 cst
Low HLB emulsifier
Cetyl dimethicone copolyol PEO polyhydroxy stearic acid Polyglyceryl polyricinoleic acid
Goldschmidt ICI Quest
Abil EM90 Arlacel P135 Admul WOL
saturated saturated unsaturated
External aqueous phase surfactant
Sodium laureth sulfate Cocoamidopropyl betaine
Stepan Goldschmidt
Steol CS-330L Tegobetain F50
saturated saturated
Internal aqueous phase solute Oil
sidescraper blade, a variable speed Ultra Turrex T25 homogenizer, and an addition funnel. Secondary emulsions (O/W2 ) were prepared with the same equipment but without homogenization. Tensiometers Interfacial tension in the low HLB emulsifier–water system was evaluated by sessile drop method using a contact angle goniometer from Material Associates Inc. Interfacial tension in mixed high HLB surfactant/low HLB emulsifier systems was determined using a spinning drop interfacial tensiometer from Kruss. All the measurements were carried out at 25 ± 0.2◦ C. Microscope The droplet size of the emulsions (W1 /O as well as W1 /O/W2 ) was measured using a Zeiss Axioplan microscope by diluting the emulsions in oil (in the case of W1 /O) or in surfactant solution (in the case of W1 /O/W2 ) on a glass slide. The dilution was kept to a minimum, enough to observe distinct droplets.
Type saturated saturated unsaturated
(same) drop was equilibrated with the new solution. Desorption measurements were carried out by removing a known amount of oil phase containing a known concentration of emulsifier and replacing it with oil without dissolved emulsifier. Interfacial tension was calculated from the fluid drop profile following the method described by Anastasiadis et al. (10). Oil (low HLB emulsifier)/water (high HLB emulsifier) system. A drop of oil with dissolved emulsifier was injected into a capillary tube, filled with aqueous phase, and rotated at a preset speed so that length of the drop is greater than 4 times the elongated drop diameter. The diameter of the drop was then measured as a function of time. The densities of aqueous phase and the oil phase were calculated by measuring the weights of known volumes of liquid and the difference in densities was determined. The interfacial tension was then calculated from the elongated drop diameter using Vonnegut’s equation (11). Multiple Emulsion Preparation Primary Emulsions
Conductivity Meter Aqueous phase conductivity was measured using a CDM83 Radiometer Copenhagen conductivity meter equipped with a CDC304 conductivity probe. The meter was calibrated monthly with KCl solutions in order to ensure its accuracy between 1 and 30 mS/cm. Methods Adsorption/Desorption Study Oil (low HLB emulsifier)/water system. Adsorption study was carried out by determining and analyzing, using video image processing, the axisymmetric sessile fluid drop profile of a drop of water immersed in oil containing a known concentration of emulsifier. After completion of the measurement known amount of the oil phase was removed without disturbing the drop and replaced with oil containing a higher concentration of emulsifier to achieve the desired concentration level in the final solution. The axisymmetric sessile drop profile was determined after the
Oil phase. The oil and the low HLB emulsifier were mixed together at room temperature using overhead stirring, in a known weight ratio. Internal aqueous phase (W1 ). Sodium chloride solutions of appropriate concentrations were prepared and the pH was adjusted to 7.0 using 0.01 M sodium hydroxide. W1 /O preparation. 210 grams of the oil phase (O) was charged into the J & K vessel. Four hundred ninety grams of the salt solution (W1 ) was added to the oil phase steadily through the addition funnel over the course of 10–12 min while mixing with the side-scraper (60–80 rpm). Slow addition of aqueous phase under low shear mixing is essential to obtain a stable primary emulsion. Upon complete addition of the aqueous phase, the mixing speed of the side-scraper was reduced to 40 rpm and the homogenizer was turned on. Homogenization was carried out for 3 min each at 8000, 9500, and 13,500 rpm at room temperature. The heat generated during the homogenization process raised the temperature of the batch to ∼35◦ C.
210
VASUDEVAN AND NASER
TABLE 2 Schematic, Simplified Combination of Ingredients to Form Multiple Emulsions
60 Parts W1/O Emulsion + 40 Parts External (Aqueous)
Ingredient
%
Ingredient
%
Salt Solution (pH 7) Oil/Emollient Low HLB Surf
70 27 3
Oil/Emollient Low HLB Surf
16.2 1.8
= Surfactant (pH 7)
34-37.5
Surfactant Water (pH 7)
To ensure that there was no phase inversion of W1 /O emulsion, a few big drops (millimeters in diameter) of the emulsion was mixed with salt solution containing sodium chloride at the same concentration as in W1 (to avoid inversion from osmotic pressure imbalance). In the case of inverted W1 /O emulsion, the drops immediately broke into fine drops of oil (micrometer size). Otherwise, the large drop stayed in tact upon dispersion into sodium chloride solution. Multiple Emulsion External aqueous phase (W2 ). External aqueous phase was prepared in a jacketed vessel by mixing the minority surfactant into the majority surfactant at 60◦ C with intense mixing. SLE2 S to betaine in the external aqueous phase was maintained at 6 : 4 (w/w), based on the active ingredient level, and the total surfactant concentration was maintained at ∼37.5 weight percent. The pH of each surfactant solution was pre-adjusted to 7.0 using 0.1 M sodium hydroxide. The external aqueous phase, which had gel-like consistency, was centrifuged at 7000 rpm for 20 min to remove the entrapped air.
13.6-15 Approx 68
W1 /O/W2 preparation. 90 grams of the primary emulsion (W1 /O) was charged into the smaller scale J&K vessel followed by addition of 60 g of the external aqueous phase (W2 ). The mixture was hand-mixed gently for 15 s using a spatula. This was followed by mixing with the side-scraper for 8 min at 70– 105 rpm (higher speed was used when mixing, as observed visually, was found to be inadequate at lower speed). A small spatula was inserted into the vessel to serve as a baffle to ensure good mixing around the center spindle of the side scraper. The resulting double emulsions were transferred into 250-ml separatory funnels to await separation of the aqueous phase for emulsion stability analysis. A schematic illustrating the simplified combination of ingredients to form the multiple emulsion is shown in Table 2. Stability measurements. After W1 /O/W2 preparation, the samples were transferred to a separatory funnel. Two types of stability measurements were made. The first type of stability measurements involved separation, after one day, of bottom clear layer of aqueous phase and measurement of conductivity of this phase (the results from these tests are reported in Fig. 1). This
FIG. 1. Effect of internal aqueous phase solute (sodium chloride) concentration on the conductivity of external aqueous phase—LMO + Abil EM90/SLES + Betaine system.
STABILITY OF MULTIPLE EMULSIONS
211
type of stability measurement was made only on samples with sodium chloride concentration of 6 wt% and lower and 14 wt% and higher in the internal aqueous phase (W1 ). This is because no clear phase separation occurred in samples with sodium chloride concentration of 8 to 12 wt% in the internal aqueous phase after one day. The second type of stability measurements involved separation, at different time intervals ranging from 14 to 84 days, of the bottom clear layer and measurement of the volume of the separated layer. The volume of the separated layer divided by the total volume of the multiple emulsion is reported as a percentage of clear layer separation in Figs. 2 and 4.
RESULTS AND DISCUSSION
Effect of Sodium Chloride Concentration in the Internal Aqueous Phase on Multiple Emulsion Stability Light Mineral Oil (LMO) + Abil EM90/SLES + Betaine system. The effect of sodium chloride concentration in the internal aqueous phase (W1 ) on the conductivity of the separated aqueous phase measured one day after preparation is shown in Fig. 1. The initial (W2 ) conductivity of the external aqueous phase (40.9 mS/cm) is shown as a horizontal line. In the low sodium chloride concentration range, 2 to 6 wt%, the conductivity of the separated aqueous phase is less than the initial conductivity of the W2 phase. The conductivity decreases with decreasing sodium chloride concentration. Also, in this range, visual thinning and phase separation of the multiple emulsion was seen. Lower separated aqueous phase conductivity relative to the initial W2 conductivity suggests that water is diffusing out from the W1 phase to the W2 phase due to osmotic pressure imbalance. As the W1 phase loses water, the oil droplets shrink causing a decrease in phase volume of the primary emulsion. As a result, the emulsion thins and phase separates as observed. The lower the sodium chloride concentration in W1 phase the greater is the amount of water that migrates to the W2 phase (most of W2 phase separates into a bottom clear layer after one day, at sodium chloride concentration of 6 wt% and lower). The increased dilution of the W2 phase with the water from the W1 phase results in the conductivity of the separated phase being lower than the initial conductivity of the W2 phase. In the high range of W1 salt concentration of 14 to 16 wt%, the conductivity of the separated aqueous phase a day after preparation is higher than the initial W2 conductivity. Also, in this range the conductivity of the separated aqueous phase increases with an increase in salt concentration. Visually, rapid thickening of the emulsion was observed followed by thinning and phase separation. Higher conductivity of the separated aqueous phase a day after preparation compared to the initial W2 conductivity can occur as a result of migration of water from the W2 phase to the W1 phase. In this case the oil droplet would swell resulting in increased phase volume and thickening. Initial thickening of the emulsion can, therefore, be attributed to the migration of water from the W2 phase to the W1 phase. However, subsequent
FIG. 2. Effect of internal aqueous phase solute (sodium chloride) concentration on physical stability of multiple emulsions—LMO + Abil EM90/SLES + Betaine system.
thinning followed by phase separation can only imply breaking of the internal emulsion. This happens due to swelling of the oil droplets, caused by water diffusing from the W2 phase to the W1 phase and consequent thinning of the oil layer leading to eventual break-up of the emulsion (12). The separated aqueous phase in this case constitutes both W1 and W2 contents. In the mid-range of sodium chloride concentration, 8 to 11 wt%, the osmotic pressure imbalance is moderate and hence the emulsions are relatively stable and no phase separation occurs a day after preparation. Since there was no phase separation in this range after a day, no conductivity measurements were made. The long term stability in this range was examined and the results, expressed as bottom clear layer separation as a function of sodium chloride concentration, are shown in Fig. 2. The results show the highest stability to occur around a sodium chloride concentration of about 9 wt%. At this concentration, there is less than a 1.0 volume percent bottom clear layer separation after 84 days. Optical micrographs at 200X of the multiple emulsion taken after 84 days (Fig. 3) corroborate the phase separation data presented in Fig. 2. The systems containing 8 and 9 wt% sodium chloride in the internal aqueous phase show predominantly multiple (W/O) drops whereas that at 10 wt% salt concentration shows only a small fraction of multiple drops. Only empty oil droplets are seen in the system containing 11 wt% sodium chloride. Sunflower oil (SFO) + Admul WOL/SLES + Betaine system. The effect of sodium chloride concentration, in the 6 to 9 wt% range, on the bottom clear layer separation in the multiple emulsion containing SFO + Admul WOL as the oil phase is shown in Fig. 4 for different storage times. Two striking differences are seen between these results and those shown in Fig. 2 for the LMO + Abil EM90 system: (i) the effect of salt on the bottom clear layer separation is minimal and (ii) the magnitude of the bottom clear layer separation is much higher in the SFO + Admul WOL system than in the LMO + Abil EM90 system. These results indicate that SFO + Admul WOL system is highly unstable and the observed instability is predominantly caused by effect/s other than the osmotic pressure imbalance.
212
VASUDEVAN AND NASER
FIG. 3.
Photomicrographs (200X magnification) of multiple emulsions at different salt concentrations: LMO + Abil EM90/SLES + Betaine system.
Interfacial Tension Measurements in LMO/Abil EM90 and SFO/Admul WOL Systems with and without High HLB Surfactants
examine the effect of high HLB surfactant mixture used in the study, SLES + betaine, on the adsorption/desorption characteristics of the low HLB emulsifier at the O/W interface.
A major factor causing instability in multiple emulsion systems has been proposed to be the interaction of the high HLB surfactant with the low HLB emulsifier film at the O/W2 interface (8). Interfacial tension measurements were conducted to
Effect of concentration of low HLB emulsifier on interfacial tension in the absence of high HLB surfactants. The effect of emulsifier concentration on the interfacial tension of water/LMO + Abil EM90 interface as well as water/SF oil + Admul WOL interface is shown in Fig. 5. Abil EM90 shows a higher affinity for the LMO/water interface than that exhibited by Admul WOL toward SFO/water interface. However, this shows a difference only of the weight effectiveness of the two emulsifiers and does not provide irrefutable evidence to explain the instability observed with the SFO + Admul WOL containing multiple emulsion system. The desorption data, shown in Fig. 6, indicates that both the emulsifiers are strongly adsorbed at the interface and do not desorb, exhibiting behavior typical of polymers adsorbed at interfaces.
FIG. 4. Effect of internal aqueous phase solute (sodium chloride) concentration on physical stability of multiple emulsions—SF oil + Admul WOL/SLES + Betaine system.
Effect of concentration of low HLB emulsifier on interfacial tension in the presence of high HLB surfactants. Interfacial tension of LMO + Abil EM90 and SFO + Admul WOL mixtures against 3.5 wt% SLES + Betaine (2 : 3 w/w) solution, measured as a function of the low HLB emulsifier concentration
213
STABILITY OF MULTIPLE EMULSIONS
FIG. 5. Effect of low HLB emulsifier concentration on interfacial tension of LMO + Abil EM90 and SF oil + Admul WOL at the oil/water interface.
is shown in Fig. 7. There is a distinct difference between the curves obtained for LMO + Abil EM90 and SFO + Admul WOL mixtures. In the case of LMO + Abil EM90 mixture, an inverted s-shaped curve is obtained with the interfacial tension decreasing sharply at an emulsifier concentration of about 0.5 wt% showing a synergistic interaction between the low HLB emulsifier and the high HLB emulsifier at the O/W interface. The area per molecule based on the Gibbs adsorption equation decreases by about five times in the sharp transition region indicating surface aggregation/condensation of the low and high HLB surfactant molecules. In the case of SFO + Admul WOL mixture the interfacial tension curve shows no sharp decrease indicating the absence of synergistic interaction between the low and high HLB surfactants. Furthermore, at an emulsifier concentration of about 10 wt% the interfacial tension obtained with LMO + Abil EM90
FIG. 7. Effect of low HLB emulsifier concentration on interfacial tension: LMO + Abil EM90 and SF oil + Admul WOL at the oil/SLES + betaine solution interface.
mixture (∼0.02 mN/m) is more than an order of magnitude less than that obtained with SFO + Admul WOL mixture. This suggests the presence of much higher concentration of the surfactants at the interface in LMO + Abil EM90/SLES + betaine system compared to SFO + Admul WOL/SLES + betaine system. The concentration of the high HLB surfactant mixture in the external phase of the multiple emulsions tested in this study is about 35 wt%, which is 10 times higher than the concentration used in the interfacial tension measurements. It was not possible to carry out interfacial tension measurements at high surfactant concentrations due to high viscosity of the surfactant solution and the inadequacy of the spinning drop tensiometer to handle viscous solutions. Table 3 shows that the interfacial tension values obtained using 3.5 and 10 wt% solutions of SLES + Betaine mixture are about the same indicating that at least above 3.5 wt% level, there is no effect of high HLB surfactant concentration on interfacial tension. Therefore, conclusions drawn based on measurements conducted using 3.5 wt% SLES + betaine solutions will be valid at higher concentrations as well.
TABLE 3 Effect of Surfactant Concentration on Interfacial Tension
FIG. 6. Adsorption/desorption characteristics of LMO + Abil EM90 and SF oil + Admul WOL at the oil/water interface.
Aqueous phase = SLE2 S − Betaine
Organic phase = 2.5wt% Abil EM90 in LMO
Surfactant concentration Wt%
Interfacial tension mN/m
3.5 10.0
0.016 0.018
214
VASUDEVAN AND NASER
TABLE 4 Effect of Emulsifier and Oil Type on Desorption of Adsorbed Emulsifier Emulsifier → Oil Light mineral oil Octyl palmitate Sunflower oil ∗
Abil EM90
Arlacel P135
Admul WOL
No desorption/stable multiple emulsion No desorption/stable multiple emulsion *
No desorption/stable multiple emulsion *
*
Desorption/unstable multiple emulsion
Desorption/unstable multiple emulsion Desorption/unstable multiple emulsion
Multiple emulsion could not be prepared using these systems since the primary emulsion itself was unstable and phase separated within two days.
Effect of time on interfacial tension in systems containing low and high HLB surfactants. Interfacial tension measurements were made as a function of time at an emulsifier concentration of 10 wt% in the oil phase, at a fixed SLES + betaine concentration of 3.5 wt% in the aqueous phase. In the case of LMO + Abil EM90 mixture the interfacial tension decreases steadily till about 200 min beyond which it remains constant. In the case of SFO + Admul mixture, the interfacial tension decreases initially indicating adsorption of the emulsifier at the interface (Fig. 8). At about 3 min an interfacial tension value of ∼0.04 mN/m is reached which is very close to the value of 0.01 mN/m obtained with the LMO + Abil EM90 system. After about 3 min, however, the interfacial tension starts increasing suggesting desorption of the adsorbed surfactants. The observed desorption is due to the presence of high HLB surfactant since in its absence no desorption occurs (see Fig. 6).
periments are summarized in Table 4 along with the multiple emulsion stability obtained in these systems. A careful analysis of the result shows that desorption is seen only in systems in which either the oil or the emulsifier, or both, contain unsaturated hydrocarbon chains. Also, no stable multiple emulsion was obtained with systems that showed desorption. The above results indicate that stable multiple emulsion can be prepared only with emulsifier/surfactant systems that form a well packed condensed layer at the O/W interface. The surfactant systems that do not pack well at the interface will be micellized by the high HLB surfactant in the external aqueous phase. Since the presence of unsaturated chains disrupt the packing at the interface, stable multiple emulsions containing high concentration of high HLB surfactants cannot be prepared with oils or emulsifiers that contain unsaturated hydrocarbon chains.
Interfacial tension measurements with different oil/low HLB emulsifier systems. Interfacial tension measurements were carried out with different oil/low HLB emulsifier systems to determine the role of oil as well as that of low HLB emulsifier on the adsorption/desorption characteristics in the presence of high HLB surfactant, SLES/Betaine. The results of this set of ex-
SUMMARY AND CONCLUSIONS
The two dominant factors that affect the stability of multiple emulsions in high HLB surfactant systems are the osmotic pressure imbalance between the internal and the external aqueous phase and the integrity of the emulsifier/surfactant film formed at the O/W interface. Relatively stable multiple emulsions were formed using a silicone based polymeric emulsifier in cleansing systems when light mineral oil was used as the oil phase. Interfacial measurements showed that the above system owes its stability to the presence of stable interfacial film formed by the interaction between the low HLB emulsifier and the high HLB surfactant. Multiple emulsions prepared using sunflower oil as the oil phase and Admul WOL (polyglyceryl polyricinoleic acid) as the low HLB emulsifier were unstable due to desorption of the emulsifier/surfactant adsorbed at the interface. The results also showed that desorption of the adsorbed surfactant + emulsifier film occurred in systems in which either the oil or the emulsifier or both contained unsaturated chains. REFERENCES
FIG. 8. Effect of time on interfacial tension: LMO + Abil EM90 and SF oil + Admul WOL at the oil/SLES + betaine solution interface.
1. Reid, E. S., and Murray, A. M., Conditioning shampoo comprising a surfactant, a non-volatile silicone oil and guar hydroxypropyltrimonium chloride as a cationic conditioning polymer, U.S. Patent 5085857.
STABILITY OF MULTIPLE EMULSIONS 2. Rushton, H., Gummer, C. L., and Flasch, H., 2-in-1 shampoo technology: State-of-the-art shampoo and conditioner in one, Skin Pharmacol. 7, 78 (1994). 3. Takahashi, Y., Yoshida, T., and Takahashi, T., Process for producing a W/O/W type multiple emulsion for medicines, cosmetics etc., U.S. Patent 4,931,210. 4. Hameyer, P., and Jenni, K. R., Emulsifiers for multiple emulsions, Cosmetics and Toiletries 111, 39 (1996). 5. Garti, N., Fenkel, M., and Schwartz, R., J. Disp. Sci. Technol. 4, 237 (1983). 6. Garti, N., Fenkel, M., and Schwartz, R., J. Disp. Sci. Technol. 4 , 287 (1983).
215
7. Garti, N., Colloids Surf. A. 123–124, 233 (1997). 8. Rosano, H. L., Gandolfo, F. G., and Hidrot, J.-D. P., Colloids Surf. A. 138, 109 (1998). 9. Florence, A. T., and Whitehill, D., J. Colloid Interface Sci. 57, 353 (1996). 10. Anastasiadis, S. H., Chen, J. K., Koberstein, J. T., Siegel, A. F., Sohn, J. E., and Emerson, J. A., J. Colloid Interface Sci. 119, 55 (1987). 11. Vonnegut, B., Rev. Sci. Inst. 13, 6 (1947). 12. Jager-Lezer, N., Denine, R., Grossiord, J. L., Wepierre, J., Rault, S., and Seiller, M., Cosmetics and Toiletries 111, 53 (1996).
Some Aspects of Stability of Multiple Emulsions in Personal Cleansing Systems Tirucherai V. Vasudevan1 and Mark S. Naser Unilever Research U.S., 45 River Road, Edgewater, New Jersey 07020 Received September 14, 2001; accepted February 27, 2002; published online April 23, 2002
implement in shampoo and body wash applications due to the presence of high concentration of high HLB surfactants. Multiple emulsion has found applications in pharmaceuticals and cosmetics (3–5). However, the systems employed in these applications contained only a low level (<5 wt%) of high HLB surfactant in the external aqueous phase. Since high HLB surfactants greatly influence the interfacial properties of the low HLB emulsifier, and hence the stability of the multiple emulsion, these studies have only a limited value in providing guidance for formulating stable multiple emulsions in personal cleansing systems. In this paper, the effect of different low HLB emulsifiers, as well as oils, on the stability of multiple emulsion has been presented and the mechanism of breakdown of multiple emulsions in the presence of high HLB surfactants has been elucidated. Only polymeric low HLB emulsifiers were used in this study since the literature information indicates that monomeric emulsifiers cannot provide long term stability in multiple emulsion systems (6, 7). Also, all the experiments were carried out using sodium chloride solution as the internal aqueous phase to modulate the osmotic pressure imbalance as well as to minimize Ostwald ripening. These two factors have been shown to play a critical role in determining multiple emulsion stability (8). Three types of oil, namely mineral oil, vegetable oil, and silicone oil, were used in the study since multiple emulsion yields have been reported to depend on the oil type with mineral oils providing higher yields than vegetable oils (9). The higher yield obtained with the mineral based oils compared to the vegetable based oils was reported to be due to the greater interaction of emulsifiers with mineral oils than vegetable oils.
The two dominant factors that were found to affect the stability of multiple emulsions in high HLB surfactant systems are the osmotic pressure imbalance between the internal aqueous phase and the external aqueous phase, and the adsorption/desorption characteristics of the emulsifier/surfactant film at the oil/water interface. Synergistic interaction between the low HLB emulsifier and the high HLB surfactant that produces very low interfacial tension of the order of 10−2 mN/m at the oil/water interface was found to occur in some of the systems investigated. Long term stability was observed in multiple emulsion containing these systems. However, no synergy was observed in systems in which either the oil or the emulsifier, or both, contained unsaturated chains. In fact, desorption of the adsorbed surfactant film was observed in systems containing unsaturated chains. The observed desorption from the interface of the emulsifier in these systems was attributed mainly to the inability of the unsaturated chains to form a close packed, condensed interfacial film. Presence of closely packed, condensed interfacial film is necessary to prevent solubilization of the adsorbed low HLB emulsifier by the high HLB surfactant. Multiple emulsions prepared using systems containing unsaturated hydrocarbons were highly unstable. C 2002 Elsevier Science (USA) Key Words: multiple emulsions; cleansing systems; high surfactant systems; unsaturated oils; unsaturated emulsifiers.
INTRODUCTION
In the past several years, major efforts have been devoted to obtaining enhanced deposition of oils from shampoos and body washes. Several new technologies have emerged from this endeavor. The important technologies among these are cationic polymer assisted deposition for colloidal sized particles (1) and direct transfer of large oil droplets (2). However, deposition of water soluble benefit agents has remained a daunting task. Deposition of water soluble benefit agents onto skin from wash-off systems can potentially be achieved by incorporating them within either an oil (multiple emulsion technology) or in liposomes/niosomes and depositing the oil or liposome using existing methods such as polymer assisted deposition and direct deposition. Liposome/niosome technology is rather difficult to
EXPERIMENTAL
Materials The materials used to prepare the multiple emulsions are listed in Table 1. Equipment Homogenizer—Mixer
1
To whom correspondence should be addressed. Current address: Unilever HPC U.S.A., 3100 Golf Road, Rolling Meadows, IL 60008. E-mail: vasu. [email protected]. 0021-9797/02 $35.00
C 2002 Elsevier Science (USA)
All rights reserved.
Primary emulsions (W1 /O) were prepared using a Janke & Kunkel RE 162P mixer equipped with a variable speed 208
209
STABILITY OF MULTIPLE EMULSIONS
TABLE 1 Materials Material
Supplier
Sodium chloride
Fisher
Trade name
Light mineral oil Octyl palmitate Silicone oil Sunflower oil
Fisher Goldschmidt Dow Corning Welch, Clarke & Holme Inc.
Tegosoft OP DC200 fluid 50 cst
Low HLB emulsifier
Cetyl dimethicone copolyol PEO polyhydroxy stearic acid Polyglyceryl polyricinoleic acid
Goldschmidt ICI Quest
Abil EM90 Arlacel P135 Admul WOL
saturated saturated unsaturated
External aqueous phase surfactant
Sodium laureth sulfate Cocoamidopropyl betaine
Stepan Goldschmidt
Steol CS-330L Tegobetain F50
saturated saturated
Internal aqueous phase solute Oil
sidescraper blade, a variable speed Ultra Turrex T25 homogenizer, and an addition funnel. Secondary emulsions (O/W2 ) were prepared with the same equipment but without homogenization. Tensiometers Interfacial tension in the low HLB emulsifier–water system was evaluated by sessile drop method using a contact angle goniometer from Material Associates Inc. Interfacial tension in mixed high HLB surfactant/low HLB emulsifier systems was determined using a spinning drop interfacial tensiometer from Kruss. All the measurements were carried out at 25 ± 0.2◦ C. Microscope The droplet size of the emulsions (W1 /O as well as W1 /O/W2 ) was measured using a Zeiss Axioplan microscope by diluting the emulsions in oil (in the case of W1 /O) or in surfactant solution (in the case of W1 /O/W2 ) on a glass slide. The dilution was kept to a minimum, enough to observe distinct droplets.
Type saturated saturated unsaturated
(same) drop was equilibrated with the new solution. Desorption measurements were carried out by removing a known amount of oil phase containing a known concentration of emulsifier and replacing it with oil without dissolved emulsifier. Interfacial tension was calculated from the fluid drop profile following the method described by Anastasiadis et al. (10). Oil (low HLB emulsifier)/water (high HLB emulsifier) system. A drop of oil with dissolved emulsifier was injected into a capillary tube, filled with aqueous phase, and rotated at a preset speed so that length of the drop is greater than 4 times the elongated drop diameter. The diameter of the drop was then measured as a function of time. The densities of aqueous phase and the oil phase were calculated by measuring the weights of known volumes of liquid and the difference in densities was determined. The interfacial tension was then calculated from the elongated drop diameter using Vonnegut’s equation (11). Multiple Emulsion Preparation Primary Emulsions
Conductivity Meter Aqueous phase conductivity was measured using a CDM83 Radiometer Copenhagen conductivity meter equipped with a CDC304 conductivity probe. The meter was calibrated monthly with KCl solutions in order to ensure its accuracy between 1 and 30 mS/cm. Methods Adsorption/Desorption Study Oil (low HLB emulsifier)/water system. Adsorption study was carried out by determining and analyzing, using video image processing, the axisymmetric sessile fluid drop profile of a drop of water immersed in oil containing a known concentration of emulsifier. After completion of the measurement known amount of the oil phase was removed without disturbing the drop and replaced with oil containing a higher concentration of emulsifier to achieve the desired concentration level in the final solution. The axisymmetric sessile drop profile was determined after the
Oil phase. The oil and the low HLB emulsifier were mixed together at room temperature using overhead stirring, in a known weight ratio. Internal aqueous phase (W1 ). Sodium chloride solutions of appropriate concentrations were prepared and the pH was adjusted to 7.0 using 0.01 M sodium hydroxide. W1 /O preparation. 210 grams of the oil phase (O) was charged into the J & K vessel. Four hundred ninety grams of the salt solution (W1 ) was added to the oil phase steadily through the addition funnel over the course of 10–12 min while mixing with the side-scraper (60–80 rpm). Slow addition of aqueous phase under low shear mixing is essential to obtain a stable primary emulsion. Upon complete addition of the aqueous phase, the mixing speed of the side-scraper was reduced to 40 rpm and the homogenizer was turned on. Homogenization was carried out for 3 min each at 8000, 9500, and 13,500 rpm at room temperature. The heat generated during the homogenization process raised the temperature of the batch to ∼35◦ C.
210
VASUDEVAN AND NASER
TABLE 2 Schematic, Simplified Combination of Ingredients to Form Multiple Emulsions
60 Parts W1/O Emulsion + 40 Parts External (Aqueous)
Ingredient
%
Ingredient
%
Salt Solution (pH 7) Oil/Emollient Low HLB Surf
70 27 3
Oil/Emollient Low HLB Surf
16.2 1.8
= Surfactant (pH 7)
34-37.5
Surfactant Water (pH 7)
To ensure that there was no phase inversion of W1 /O emulsion, a few big drops (millimeters in diameter) of the emulsion was mixed with salt solution containing sodium chloride at the same concentration as in W1 (to avoid inversion from osmotic pressure imbalance). In the case of inverted W1 /O emulsion, the drops immediately broke into fine drops of oil (micrometer size). Otherwise, the large drop stayed in tact upon dispersion into sodium chloride solution. Multiple Emulsion External aqueous phase (W2 ). External aqueous phase was prepared in a jacketed vessel by mixing the minority surfactant into the majority surfactant at 60◦ C with intense mixing. SLE2 S to betaine in the external aqueous phase was maintained at 6 : 4 (w/w), based on the active ingredient level, and the total surfactant concentration was maintained at ∼37.5 weight percent. The pH of each surfactant solution was pre-adjusted to 7.0 using 0.1 M sodium hydroxide. The external aqueous phase, which had gel-like consistency, was centrifuged at 7000 rpm for 20 min to remove the entrapped air.
13.6-15 Approx 68
W1 /O/W2 preparation. 90 grams of the primary emulsion (W1 /O) was charged into the smaller scale J&K vessel followed by addition of 60 g of the external aqueous phase (W2 ). The mixture was hand-mixed gently for 15 s using a spatula. This was followed by mixing with the side-scraper for 8 min at 70– 105 rpm (higher speed was used when mixing, as observed visually, was found to be inadequate at lower speed). A small spatula was inserted into the vessel to serve as a baffle to ensure good mixing around the center spindle of the side scraper. The resulting double emulsions were transferred into 250-ml separatory funnels to await separation of the aqueous phase for emulsion stability analysis. A schematic illustrating the simplified combination of ingredients to form the multiple emulsion is shown in Table 2. Stability measurements. After W1 /O/W2 preparation, the samples were transferred to a separatory funnel. Two types of stability measurements were made. The first type of stability measurements involved separation, after one day, of bottom clear layer of aqueous phase and measurement of conductivity of this phase (the results from these tests are reported in Fig. 1). This
FIG. 1. Effect of internal aqueous phase solute (sodium chloride) concentration on the conductivity of external aqueous phase—LMO + Abil EM90/SLES + Betaine system.
STABILITY OF MULTIPLE EMULSIONS
211
type of stability measurement was made only on samples with sodium chloride concentration of 6 wt% and lower and 14 wt% and higher in the internal aqueous phase (W1 ). This is because no clear phase separation occurred in samples with sodium chloride concentration of 8 to 12 wt% in the internal aqueous phase after one day. The second type of stability measurements involved separation, at different time intervals ranging from 14 to 84 days, of the bottom clear layer and measurement of the volume of the separated layer. The volume of the separated layer divided by the total volume of the multiple emulsion is reported as a percentage of clear layer separation in Figs. 2 and 4.
RESULTS AND DISCUSSION
Effect of Sodium Chloride Concentration in the Internal Aqueous Phase on Multiple Emulsion Stability Light Mineral Oil (LMO) + Abil EM90/SLES + Betaine system. The effect of sodium chloride concentration in the internal aqueous phase (W1 ) on the conductivity of the separated aqueous phase measured one day after preparation is shown in Fig. 1. The initial (W2 ) conductivity of the external aqueous phase (40.9 mS/cm) is shown as a horizontal line. In the low sodium chloride concentration range, 2 to 6 wt%, the conductivity of the separated aqueous phase is less than the initial conductivity of the W2 phase. The conductivity decreases with decreasing sodium chloride concentration. Also, in this range, visual thinning and phase separation of the multiple emulsion was seen. Lower separated aqueous phase conductivity relative to the initial W2 conductivity suggests that water is diffusing out from the W1 phase to the W2 phase due to osmotic pressure imbalance. As the W1 phase loses water, the oil droplets shrink causing a decrease in phase volume of the primary emulsion. As a result, the emulsion thins and phase separates as observed. The lower the sodium chloride concentration in W1 phase the greater is the amount of water that migrates to the W2 phase (most of W2 phase separates into a bottom clear layer after one day, at sodium chloride concentration of 6 wt% and lower). The increased dilution of the W2 phase with the water from the W1 phase results in the conductivity of the separated phase being lower than the initial conductivity of the W2 phase. In the high range of W1 salt concentration of 14 to 16 wt%, the conductivity of the separated aqueous phase a day after preparation is higher than the initial W2 conductivity. Also, in this range the conductivity of the separated aqueous phase increases with an increase in salt concentration. Visually, rapid thickening of the emulsion was observed followed by thinning and phase separation. Higher conductivity of the separated aqueous phase a day after preparation compared to the initial W2 conductivity can occur as a result of migration of water from the W2 phase to the W1 phase. In this case the oil droplet would swell resulting in increased phase volume and thickening. Initial thickening of the emulsion can, therefore, be attributed to the migration of water from the W2 phase to the W1 phase. However, subsequent
FIG. 2. Effect of internal aqueous phase solute (sodium chloride) concentration on physical stability of multiple emulsions—LMO + Abil EM90/SLES + Betaine system.
thinning followed by phase separation can only imply breaking of the internal emulsion. This happens due to swelling of the oil droplets, caused by water diffusing from the W2 phase to the W1 phase and consequent thinning of the oil layer leading to eventual break-up of the emulsion (12). The separated aqueous phase in this case constitutes both W1 and W2 contents. In the mid-range of sodium chloride concentration, 8 to 11 wt%, the osmotic pressure imbalance is moderate and hence the emulsions are relatively stable and no phase separation occurs a day after preparation. Since there was no phase separation in this range after a day, no conductivity measurements were made. The long term stability in this range was examined and the results, expressed as bottom clear layer separation as a function of sodium chloride concentration, are shown in Fig. 2. The results show the highest stability to occur around a sodium chloride concentration of about 9 wt%. At this concentration, there is less than a 1.0 volume percent bottom clear layer separation after 84 days. Optical micrographs at 200X of the multiple emulsion taken after 84 days (Fig. 3) corroborate the phase separation data presented in Fig. 2. The systems containing 8 and 9 wt% sodium chloride in the internal aqueous phase show predominantly multiple (W/O) drops whereas that at 10 wt% salt concentration shows only a small fraction of multiple drops. Only empty oil droplets are seen in the system containing 11 wt% sodium chloride. Sunflower oil (SFO) + Admul WOL/SLES + Betaine system. The effect of sodium chloride concentration, in the 6 to 9 wt% range, on the bottom clear layer separation in the multiple emulsion containing SFO + Admul WOL as the oil phase is shown in Fig. 4 for different storage times. Two striking differences are seen between these results and those shown in Fig. 2 for the LMO + Abil EM90 system: (i) the effect of salt on the bottom clear layer separation is minimal and (ii) the magnitude of the bottom clear layer separation is much higher in the SFO + Admul WOL system than in the LMO + Abil EM90 system. These results indicate that SFO + Admul WOL system is highly unstable and the observed instability is predominantly caused by effect/s other than the osmotic pressure imbalance.
212
VASUDEVAN AND NASER
FIG. 3.
Photomicrographs (200X magnification) of multiple emulsions at different salt concentrations: LMO + Abil EM90/SLES + Betaine system.
Interfacial Tension Measurements in LMO/Abil EM90 and SFO/Admul WOL Systems with and without High HLB Surfactants
examine the effect of high HLB surfactant mixture used in the study, SLES + betaine, on the adsorption/desorption characteristics of the low HLB emulsifier at the O/W interface.
A major factor causing instability in multiple emulsion systems has been proposed to be the interaction of the high HLB surfactant with the low HLB emulsifier film at the O/W2 interface (8). Interfacial tension measurements were conducted to
Effect of concentration of low HLB emulsifier on interfacial tension in the absence of high HLB surfactants. The effect of emulsifier concentration on the interfacial tension of water/LMO + Abil EM90 interface as well as water/SF oil + Admul WOL interface is shown in Fig. 5. Abil EM90 shows a higher affinity for the LMO/water interface than that exhibited by Admul WOL toward SFO/water interface. However, this shows a difference only of the weight effectiveness of the two emulsifiers and does not provide irrefutable evidence to explain the instability observed with the SFO + Admul WOL containing multiple emulsion system. The desorption data, shown in Fig. 6, indicates that both the emulsifiers are strongly adsorbed at the interface and do not desorb, exhibiting behavior typical of polymers adsorbed at interfaces.
FIG. 4. Effect of internal aqueous phase solute (sodium chloride) concentration on physical stability of multiple emulsions—SF oil + Admul WOL/SLES + Betaine system.
Effect of concentration of low HLB emulsifier on interfacial tension in the presence of high HLB surfactants. Interfacial tension of LMO + Abil EM90 and SFO + Admul WOL mixtures against 3.5 wt% SLES + Betaine (2 : 3 w/w) solution, measured as a function of the low HLB emulsifier concentration
213
STABILITY OF MULTIPLE EMULSIONS
FIG. 5. Effect of low HLB emulsifier concentration on interfacial tension of LMO + Abil EM90 and SF oil + Admul WOL at the oil/water interface.
is shown in Fig. 7. There is a distinct difference between the curves obtained for LMO + Abil EM90 and SFO + Admul WOL mixtures. In the case of LMO + Abil EM90 mixture, an inverted s-shaped curve is obtained with the interfacial tension decreasing sharply at an emulsifier concentration of about 0.5 wt% showing a synergistic interaction between the low HLB emulsifier and the high HLB emulsifier at the O/W interface. The area per molecule based on the Gibbs adsorption equation decreases by about five times in the sharp transition region indicating surface aggregation/condensation of the low and high HLB surfactant molecules. In the case of SFO + Admul WOL mixture the interfacial tension curve shows no sharp decrease indicating the absence of synergistic interaction between the low and high HLB surfactants. Furthermore, at an emulsifier concentration of about 10 wt% the interfacial tension obtained with LMO + Abil EM90
FIG. 7. Effect of low HLB emulsifier concentration on interfacial tension: LMO + Abil EM90 and SF oil + Admul WOL at the oil/SLES + betaine solution interface.
mixture (∼0.02 mN/m) is more than an order of magnitude less than that obtained with SFO + Admul WOL mixture. This suggests the presence of much higher concentration of the surfactants at the interface in LMO + Abil EM90/SLES + betaine system compared to SFO + Admul WOL/SLES + betaine system. The concentration of the high HLB surfactant mixture in the external phase of the multiple emulsions tested in this study is about 35 wt%, which is 10 times higher than the concentration used in the interfacial tension measurements. It was not possible to carry out interfacial tension measurements at high surfactant concentrations due to high viscosity of the surfactant solution and the inadequacy of the spinning drop tensiometer to handle viscous solutions. Table 3 shows that the interfacial tension values obtained using 3.5 and 10 wt% solutions of SLES + Betaine mixture are about the same indicating that at least above 3.5 wt% level, there is no effect of high HLB surfactant concentration on interfacial tension. Therefore, conclusions drawn based on measurements conducted using 3.5 wt% SLES + betaine solutions will be valid at higher concentrations as well.
TABLE 3 Effect of Surfactant Concentration on Interfacial Tension
FIG. 6. Adsorption/desorption characteristics of LMO + Abil EM90 and SF oil + Admul WOL at the oil/water interface.
Aqueous phase = SLE2 S − Betaine
Organic phase = 2.5wt% Abil EM90 in LMO
Surfactant concentration Wt%
Interfacial tension mN/m
3.5 10.0
0.016 0.018
214
VASUDEVAN AND NASER
TABLE 4 Effect of Emulsifier and Oil Type on Desorption of Adsorbed Emulsifier Emulsifier → Oil Light mineral oil Octyl palmitate Sunflower oil ∗
Abil EM90
Arlacel P135
Admul WOL
No desorption/stable multiple emulsion No desorption/stable multiple emulsion *
No desorption/stable multiple emulsion *
*
Desorption/unstable multiple emulsion
Desorption/unstable multiple emulsion Desorption/unstable multiple emulsion
Multiple emulsion could not be prepared using these systems since the primary emulsion itself was unstable and phase separated within two days.
Effect of time on interfacial tension in systems containing low and high HLB surfactants. Interfacial tension measurements were made as a function of time at an emulsifier concentration of 10 wt% in the oil phase, at a fixed SLES + betaine concentration of 3.5 wt% in the aqueous phase. In the case of LMO + Abil EM90 mixture the interfacial tension decreases steadily till about 200 min beyond which it remains constant. In the case of SFO + Admul mixture, the interfacial tension decreases initially indicating adsorption of the emulsifier at the interface (Fig. 8). At about 3 min an interfacial tension value of ∼0.04 mN/m is reached which is very close to the value of 0.01 mN/m obtained with the LMO + Abil EM90 system. After about 3 min, however, the interfacial tension starts increasing suggesting desorption of the adsorbed surfactants. The observed desorption is due to the presence of high HLB surfactant since in its absence no desorption occurs (see Fig. 6).
periments are summarized in Table 4 along with the multiple emulsion stability obtained in these systems. A careful analysis of the result shows that desorption is seen only in systems in which either the oil or the emulsifier, or both, contain unsaturated hydrocarbon chains. Also, no stable multiple emulsion was obtained with systems that showed desorption. The above results indicate that stable multiple emulsion can be prepared only with emulsifier/surfactant systems that form a well packed condensed layer at the O/W interface. The surfactant systems that do not pack well at the interface will be micellized by the high HLB surfactant in the external aqueous phase. Since the presence of unsaturated chains disrupt the packing at the interface, stable multiple emulsions containing high concentration of high HLB surfactants cannot be prepared with oils or emulsifiers that contain unsaturated hydrocarbon chains.
Interfacial tension measurements with different oil/low HLB emulsifier systems. Interfacial tension measurements were carried out with different oil/low HLB emulsifier systems to determine the role of oil as well as that of low HLB emulsifier on the adsorption/desorption characteristics in the presence of high HLB surfactant, SLES/Betaine. The results of this set of ex-
SUMMARY AND CONCLUSIONS
The two dominant factors that affect the stability of multiple emulsions in high HLB surfactant systems are the osmotic pressure imbalance between the internal and the external aqueous phase and the integrity of the emulsifier/surfactant film formed at the O/W interface. Relatively stable multiple emulsions were formed using a silicone based polymeric emulsifier in cleansing systems when light mineral oil was used as the oil phase. Interfacial measurements showed that the above system owes its stability to the presence of stable interfacial film formed by the interaction between the low HLB emulsifier and the high HLB surfactant. Multiple emulsions prepared using sunflower oil as the oil phase and Admul WOL (polyglyceryl polyricinoleic acid) as the low HLB emulsifier were unstable due to desorption of the emulsifier/surfactant adsorbed at the interface. The results also showed that desorption of the adsorbed surfactant + emulsifier film occurred in systems in which either the oil or the emulsifier or both contained unsaturated chains. REFERENCES
FIG. 8. Effect of time on interfacial tension: LMO + Abil EM90 and SF oil + Admul WOL at the oil/SLES + betaine solution interface.
1. Reid, E. S., and Murray, A. M., Conditioning shampoo comprising a surfactant, a non-volatile silicone oil and guar hydroxypropyltrimonium chloride as a cationic conditioning polymer, U.S. Patent 5085857.
STABILITY OF MULTIPLE EMULSIONS 2. Rushton, H., Gummer, C. L., and Flasch, H., 2-in-1 shampoo technology: State-of-the-art shampoo and conditioner in one, Skin Pharmacol. 7, 78 (1994). 3. Takahashi, Y., Yoshida, T., and Takahashi, T., Process for producing a W/O/W type multiple emulsion for medicines, cosmetics etc., U.S. Patent 4,931,210. 4. Hameyer, P., and Jenni, K. R., Emulsifiers for multiple emulsions, Cosmetics and Toiletries 111, 39 (1996). 5. Garti, N., Fenkel, M., and Schwartz, R., J. Disp. Sci. Technol. 4, 237 (1983). 6. Garti, N., Fenkel, M., and Schwartz, R., J. Disp. Sci. Technol. 4 , 287 (1983).
215
7. Garti, N., Colloids Surf. A. 123–124, 233 (1997). 8. Rosano, H. L., Gandolfo, F. G., and Hidrot, J.-D. P., Colloids Surf. A. 138, 109 (1998). 9. Florence, A. T., and Whitehill, D., J. Colloid Interface Sci. 57, 353 (1996). 10. Anastasiadis, S. H., Chen, J. K., Koberstein, J. T., Siegel, A. F., Sohn, J. E., and Emerson, J. A., J. Colloid Interface Sci. 119, 55 (1987). 11. Vonnegut, B., Rev. Sci. Inst. 13, 6 (1947). 12. Jager-Lezer, N., Denine, R., Grossiord, J. L., Wepierre, J., Rault, S., and Seiller, M., Cosmetics and Toiletries 111, 53 (1996).