404 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS stored at various temperatures for 11 days. An initial fraction of oil is separated as the ultracentrifuge is brought up to speed. The per cent of oil separated at 10,000 rpm remained constant for a long period of time. Increased rpm increased the per cent of oil separated. Sequential accelerating and decelerating rates of separation were observed. There appears to be a maximum amount of oil that can be spun out at a given speed of centrifugation. Eventually an apparently constant rate of ap- pearance of clear oil is reached at the maximum centrifugal speed pos- sible. When the emulsion was continuously ultracentrifuged at 40,740 rpm, an instantaneous separation of oil was observed and the rate of oil separation decreased. A total separation of 60-70% of the available oil was observed before the oil separated by the same apparent constant rate observed in the terminal phase of Fig. 2. Microscopic inspection of the resilient cream that resulted from ultracentrifugation in the preparative ultracentrifuge showed that it consisted of small particles of practically constant size which were closely packed. This close packing had caused the particles to appear polygonal. In contrast, the fresh, uncentrifuged emulsion consisted of a spectrum of particle sizes. Particle size studied by the Coulter Counter* demonstrated that the emulsion had a diameter mode of 3 v with a range between 1.0 and 2.0 v. As the emulsions aged, a new mode became apparent at 4.0 •. The specific surface areas were calculated from Coulter Counter plots at different aging times and temperatures. At 17 days after preparation, derived values (10%m2/g) were 29.1 at 26øC, 28.4 at 35øC, and 26.8 at 55øC. At 24 days, these values were 27.7 at 26øC, 26.9 at 35øC, and 26.8 at 55øC. Equilibrium surfactant concentrations (absorbance at 222 m• by the analytical procedure described) were at 26øC: 0.625, 0.600, 0.610, and 0.580 at 0, 6, 11, and 24 days after preparation at 35øC: 0.603, 0.597, and 0.581 at 6, 11, and 24 days after preparation. At 45 ø and 55øC the values were 0.599, 0.582, and 0.589, 0.577 for 6 and 11 days, respec- tively. As much as 70-95% of the oil in the emulsion could be ac- counted for by the counted particles. Flotation Rate Studies o/ Tohtene-7% G-2151 Sur/actant-Water Emulsions The 50/50 toluene-water emulsion, 7% G-2151, prepared in the Waring Blendor was photographically monitored at several rpm vah•es Coulter Electronics, Hialeah, Fla.

EMULSION STABILITY 405 • 21 2.0 1.9 1.7 1.5 1.0 0.9 .,. \ e•e• e• e•e• 500 1,000 1,500 :•,000 2,500 •,000 •,500 Time in Seconds Figure 3. Example of flotation of oil particles from the bottom of 50% toluene-watcr emul- sion prepared with 7% G-2151 surfactant. The logarithm of the distance of the emulsion- water boundary from the center of the rotor is plotted against the time in seconds of ultra- ccntrifugation at 24,630 rpm in the analytical ultracentrifuge. Figure 3 is a typical plot of the loga- rithm of the distance of the emulsion-water boundary from the center of the rotor, i.e., log x vs. time in accordance with the expected Svedberg relation (1): log x = kt + log Xo = S(rpm)2t + log Xo (1) where xo is the distance of the lower boundary of the cell from the center of the rotor (i.e., the position of the particle at time t: 0) and S is defined as the sedimentation (or flotation) constant. It follows that the flotation rates k = S(rpm) 2 (2) and a plot of the h values or slopes of plots in accordance with Eq. 1 rs. various (rpm)2should be linear, of slope S, and pass through the origin. The high surfactant concentration (7%) emulsions showed no separa- tion of clear toluene under centrifugation forces of 258,000 g's for a period of three days. Microscopic investigation of the opaque portion of the emulsion demonstrated toluene oil particles entrapped in pre- cipitated surfactant. It should be noted that the flotation of low sur- factant concentration (1%) emulsions prepared by the spray techniques was too rapid for study by this technique.