STATISTICAL APPROACH TO COMMON VARIABLES 439 experimental design was applied to each combination of the following four nonaqueous liquids and thirteen surface-active agents. • Oils Surfactants Paraffinic mineral oil (Marcol GX) Cottonseed oil (Wesson oil) Oleic acid (Emery 233LL) Methylphenyl silicone (Dow Corning 710 Fluid) Cetylethylmorpholinium ethosulfate (G-263) Isopropylammoniumdodecyl benzenesulfonate (IPADBS) Polyoxyethylene (100) stearate (G-2159) Polyoxyethylene (8) stearate (Myrj 45) Polyoxyethylene (23) lauryl alcohol (Brij 35) Polyoxyethylene (15) nonyl phenol (Renex 678) Polyoxyethylene (20) sorbitan monostearate (Tween 60) Polyoxyethylene (20) sorbitan monooleate (Tween 80) Polyoxyethylene (5) sorbitan monooleate (Tween 81) Sorbitan monolaurate (Span 20) Sorbitan monostearate (Span 60) Sorbitan monooleate (Span 80) Sorbitan trioleate (Span 85) A trademark or other designation is stated in parentheses following the name of each oil and surfactant. Thus, with 12 methods for each emulsifier-substrate combination, 624 emulsion preparations were made. Their emulsion type and stability are the data reported in this study. It must be emphasized that the object of this investigation was not to determine the optimum methods for emul- sifying each oil under consideration but, rather, to determine the impor- tance of the preparative variables studied in influencing emulsion type and stability. EXPERIMENTAL In addition to the variables specified in the elements of the experimental design, the following general procedure was followed in the preparation of the emulsions. Distilled water and oils were measured by volume and emulsifier by weight. Relatively less vigorous agitation was obtained with a propeller, while more vigorous agitation was obtained with an Eppenbach Homo-Mixer which is an enclosed turbine type of stirrer. Glass apparatus was used throughout, except for the propeller shaft and blade and the Homo-Mixer. All of one phase was added to the other at one time, with agitation. Agitation, whether by propeller or by means of the Homo-Mixer, was continued for five minutes after addition was completed. In maki.ng emulsions at 70øC., both the water and the oil were previously brought to this temperature. Emulsion type was determined by the dilution tech- nique. Emulsion stability was observed in an Atlab viewer (6). Stability was evaluated initially approximately ten minutes after emulsion prepara- tion, after one day and after one week. RESULTS The stability of the emulsions formed was exlSressed on a numerical scale of 9 to 0, excellent stability being represented by 9. This qualitative

440 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS scale includes the more commonly observed gradations of emulsion separa- tion, as given in the following descriptions of the stabilities corresponding to each rating. Numerical Emulsion Description Rating No visible separation 9 Emulsion is nearly homogeneous without definite separation 8 Emulsion shows faint cream line 6 Creaming with separation of 1% or less oil or water 5 Emulsion has creamed with separation of significant volumes of oil and aque- ous phases 4 Almost complete separation of emulsion 2 Complete separation into components with essentially their original volume 1 Typical results for each of the twelve preparative methods for one oil- emulsifier pair are illustrated by those for paraflinic mineral oil and Tween 60 emulsifier in Table 2. In some cases, as in method 10 in this set, it was not possible to measure the emulsion type because of the poor stability of the resulting emulsion. TABLE 2--EMULSION TYPE AND STABILITY--MINEKAL OIL-TwEEN 60 EMULSIFIER Method Water, External • 'Stability ß No. % Phase Initial 24 Hr. 1 Wk. 1 90 w 4 4 4 2 90 w 4 4 4 11 90 w 5 4 4 12 90 w 4 4 4 5 50 w 4 4 4 8 50 w 4 4 4 9 50 w 5 4 4 10 50 .. 2 2 1 3 30 o 2 2 1 4 30 o 2 2 2 6 30 o 2 2 1 7 30 w 9 5 4 Emulsion Type The results of the statistically designed experiment may be analyzed for both emulsion type and for stability. Some emulsifier-oil combinations, indicated in Table 3, produced predominantly o/w emulsions. Such com- binations, leading mainly to o/w emulsions, ,nay be divided into three groups: (1) those forming only o/w emulsions with all 12 methods, desig- nated in Table 3 by o/w those forming o/w emulsions or emulsions of nonmeasurable type, designated in Table 3 by X and those forming one w/o emulsion together with o/w emulsions and nonmeasurable types, designated in Table 3 by the method number forming a w/o emulsion.