144 JOURNAL OF COSMETIC SCIENCE SELF-EMULSIFICATION OF SURFACTANT-OIL MIXTURES PRODUCED BY DIFFUSION AND CHEMICAL REACTION Clarence A. Miller, Ph.D. Department of Chemical Engineering, Rice University, Houston, Texas 77251-1892 Abstract. Drops having diameters of order 100 gm and containing various combinations of oils, surfactants, and in some cases alcohols were injected into water or aqueous salt or buffer solutions. The resulting dynamic behavior was observed by videomicroscopy. Spontaneous emulsification yielding oil droplets a few microns in diameter was seen in a variety of systems when diffusion and/or chemical reaction caused inversion of the drop from an oil-continuous to a water-continuous phase, leading to local supersaturation in oil. Surfactants used included nonionic (C•2E6) , anionic (Aerosol-OT), and zwitterionic (tetradecyldimethylamine oxide). In some experiments inversion occurred because a lipophilic surfactant was converted to a hydrophilic surfactant, e.g., a double-chain phospholipid to two single-chain surfactants. Introduction. When emulsions are formed by mixing oil and water phases, high shear rates are typically required to generate small drops having diameters of order 1 gin. When it is necessary or desirable to obtain small drops without high shear rates, spontaneous emulsification should be considered. Compositions of the initial oil and water phases are chosen in such a way that small droplets form spontaneously when the phases are brought into contact, i.e., no external energy of agitation is required. "Self-emulsification" refers to situations where a small amount of agitation is supplied to achieve gentle mixing. Typically the droplets form spontaneously, and mixing serves mainly to disperse them throughout a large volume and to bring together portions of the oil and water phases which were not in contact initially. Self-emulsification of oils containing dissolved solutes is not only an intriguing phenomenon but is also of practical interest in delivery of agricultural chemicals and drugs as well as during use of cutting oils, bath oils, etc. A surfactant or a mixture of surfactants is added to the oil, so that it will emulsify spontaneously when contacted with water. The objective of our study was to gain insight into the mechanism of spontaneous emulsification, which has not been well understood, and thereby provide a basis for choosing suitable surfactant/oil mixtures. Results. Rang and Miller • used videomicroscopy to study self-emulsification of n- hexadecane/C•2Es/n-octanol drops some 100 gm in diameter injected into water. Spontaneous emulsification yielding only small oil droplets was observed when the initial ratio of alcohol to hydrocarbon was greater than that in the excess oil phase in equilibrium with a m•croemulsion and water at the temperature of interest, i.e., above the ratio existing at the Phase Inversion Temperature (PIT), and when surfactant

2000 ANNUAL SCIENTIFIC MEETING 145 concentration was sufficiently high. The mechanism of emulsification can be briefly described as follows. Initially water diffused rapidly into the oil phase, converting it to an oil-continuous microemulsion. However, as octanol diffused gradually into the aqueous phase, the ratio of alcohol to surfactant in the films covering the microemulsion droplets decreased, making them more hydrophilic and leading to an increased capability of the microemulsion to solubilize water and a decreased capability to solubilize oil. Eventually, the microemulsion was no longer able to solubilize all the oil present, and oil droplets nucleated. Moreover, the microemulsion itself inverted to become water- continuous and miscible with water, so that the final state was oil droplets dispersed in an aqueous phase, the size distribution of the droplets depending largely on their rate of coalescence. When enough surfactant was present, coalescence was relatively slow, and the droplets remained sinall. No intermediate liquid crystalline phases were formed for this system. However, an intermediate lamellar phase was seen for systems where the surfactant was tetradecyldimethylamine oxide (C14DMAO). Emulsions with small oil droplets were obtained with lower surfactant concentrations in this system, apparently because the lamellar phase reduced the rate of coalescence of the droplets. Nevertheless, both the liquid crystal and the microemulsion eventually became miscible with water as they lost alcohol by diffusion into the aqueous phase. This mechanism of emulsification may be summarized as causing the microemulsion to become supersaturated in oil by shifting it from lipophilic to hydrophilic conditions. However, achieving this behavior by diffusion of a medium- chain alcohol into the aqueous phase may not be desirable for some purposes. Later work showed that emulsification could also be produced in a similar system with octanol replaced by oleyl alcohol, which has a minimal solubility in water 2. In this case equilibrium phase behavior and videomicroscopy observations indicated that, when the system was near its PIT, water diffused into the oil phase, causing first transformation to the lamellar liquid crystalline phase and then nucleation of oil droplets and conversion of the liquid crystal to a water-continuous phase. Use of ionic surfactants instead of nonionic or zwitterionic surfactants may have advantages. For instance, electrical repulsion may help stabilize the emulsion once formed. Accordingly, more recent work utilized the anionic surfactant Aerosol OT (AOT) with no alcohol and effected microemulsion inversion by a reduction in ionic strength s. This reduction was produced by a combination of water diffusion into the injected drop and AOT diffusion into the aqueous phase. Inversion and spontaneous emulsification can also be produced by chemical reactions which convert lipophilic surfactants to more hydrophilic ones. As is well known, one such reaction is conversion of fatty acids to soaps by increasing pH. However, spontaneous emulsification produced by other reactions has been investigated recently in our laboratory, including enzymatic splitting of a double-chain.phospholipid into a lysolecithin and a fatty acid. References. 1. Rang, M.J. and C.A. Miller, Progr. Colloid Polym. Sci. 109, 101 (1998). 2. Rang, M.J. and C.A. Miller, J. Colloid Intel face Sci. 209, 179 (1999). 3. Nishimi, T. and C.A. Miller. Langmuir, in press.