HYDROPHILICALLY MODIFIED SILICONE ELASTOMERS 197 Procedure. The ingredients in Part A and Part B were mixed in separate containers until uniform. Part B was then added to Part A while mixing with the same Lightnin mixer and stirrer configuration described for the water-uptake test. The mixer speed was set at 1376 rpm, and after all of Part B was added, mixing was continued for ten minutes. Next, Part C was mixed into the emulsion formed by Parts A and B, and this mixture was then dispersed into Part D while mixing with the same mixer, but at a lower speed (500 rpm) SYNERESIS TEST To determine syneresis for an anhydrous antiperspirant soft solid formula, 30 g of the formula is weighed into a 50-ml polypropylene (disposable) centrifuge tube (Fisher- brand, cat. # 05-539-9). The sample is then spun at 3000 rpm for 30 minutes using a bench top centrifuge (International Equipment Company, model HN-SII). After cen- trifuging, the tube with the sample is placed on an electronic balance and tared. The supernatant fluid is carefully pipetted from the top of the sample and the amount of fluid removed is weighed by difference. Syneresis is reported as a percentage of the original sample weight (i.e., [weight of removed fiuid]/[sample weight] x 100). VISCOSTY MEASUREMENTS Viscosity measurements were made using a Brookfield Model RVDVII+ equipped with a Helipath stand. Various combinations of "T" spindles and speeds were used according to the type of formula tested: PEG-DCP dispersions: Spindle #93 @ 2.5 rpm Antiperspirant gel viscosity: Spindle #93 @ 2.5 rpm Antiperspirant roll-on formula viscosity: Spindle #91 @ 50 rpm (Helipath off) Antiperspirant soft solid formula viscosity: Spindle #93 @ 2.5 rpm RESULTS AND DISCUSSION SYNTHESIS AND EVALUATION OF PEG-DCP VARIANTS In order to evaluate the effects of various composition and process variables on the properties of PEG-DCP, a two-level factorial experimental design was performed. The variables (factors) studied were the level of PEG substitution, the length of the PEG chain, and a process variable that was thought to have the largest effect on the crosslink density. This process variable is referred to as the "crosslink density process parameter." In a designed experiment, a series of runs are made in which the variables under study are systematically changed in order to determine their effects on the properties of interest. For this study, we evaluated the viscosity of the PEG-DCP samples and their emulsification abilities as measured by the water-uptake test. The water-uptake test provides a measure of the emulsification effectiveness. It is used to determine the maximum amount of water that can be incorporated into a w/s emulsion where the dispersion of PEG-DCP in cyclopentasiloxane is used as the continuous phase. The viscosity results for the PEG-DCP samples from the experimental design runs are

198 JOURNAL OF COSMETIC SCIENCE shown in Figures I and 2. Figure 1 shows the results for the set of runs where long-chain PEG was incorporated into the elastomer. All of these PEG-DCP samples had relatively low viscosity, except for the sample where the PEG substitution level and crosslink density were both low. When the PEG-DCP was prepared using short-chain PEG, the results shown in Figure 2 indicate that crosslink density had a larger effect on the viscosity of the dispersion. It should be noted that the central data point shown in Figures 1 and 2 (medium PEG substitution level and medium crosslink density process parameter) corresponds to a PEG-DCP made at the midpoint for all three process variables (including PEG chain length). This is a consequence of the experimental design that calls for multiple runs at the "center point" of the variables being studied. The water-uptake results are shown in Figures 3 and 4. All of the PEG-DCP samples made with long-chain PEG were effective emulsifiers, producing water uptake values of greater than 70%. Water uptake for the PEG-DCP runs with short-chain PEG, shown in Figure 4, were somewhat lower. The sample made with the minimum PEG substi- tution level and high crosslink density had the worst emulsification efficiency, with a water-uptake value of only 25 %. Overall, the PEG-DCP runs where the long-chain PEG was used produced the most effective emulsifiers. When PEG-DCP was made with short-chain PEG, the best emulsifiers were made with the high level of PEG substitu- tion. The water-uptake values were initially measured for PEG-DCP dispersions that con- tained 9% by weight of the elastomer. The water-uptake test was repeated for disper- sions that were diluted (with cyclopentasiloxane) to 7% and 5%. For the PEG-DCP sample made with long-chain PEG, dilution did not have a large effect on water uptake. For two of the PEG-DCP samples made with short-chain PEG, dilution dramatically increased the water uptake, as shown in Figure 5. This dilution effect may be due to the elastomer assuming a more extended configuration at lower concentrations. Emulsions made with the diluted PEG-DCP samples appear to be just as stable as those made with the original PEG-DCP samples (9% elastomer). EVALUATION OF PEG-DCP IN ANTIPERSPIRANT GEL One of the largest commercial applications for w/s emulsions where the continuous phase -- 160000. 140000 120000 100000 Viscosity 80000. • (cP) 60000 40000 20000 Low Medium High Crosslink 0 Density " Process High Medium Parameter Low PEG Substitution Level Figure 1. The effect of PEG substitution level and crosslink density process parameter on the viscosity of PEG-DCP made with long-chain PEG.