1 = 0.5103 + 0.0527 – 0.2077 + 0.0789 , Y A B C (1) wh ere Y1 represents the elasticity, A is the hydrogenated lecithin content, B is the PEG- 400 content, and C is the poloxamer 407 content. Using equation 1, coeffi cient was high in the order of PEG-400 B (0.2077), poloxamer 407 C (0.0789), and hydrogenated leci- thin A (0.0527). Main effect was signifi cant in the order of B, C, and A. Figure 11 depicts the 3D response surface graph using a regression equation. Figure 11A is the graph that presents the effect of the amounts of hydrogenated lecithin and PEG- 400 on elasticity when a fi xed amount of poloxamer 407 was mixed in the formulated PLO gel. This graph reveals that elasticity increased with an increase in the content of hydrogenated lecithin and a decrease in the content of PEG-400. Figure 11B represents the effect of the amounts of hydrogenated lecithin and poloxamer 407 on elasticity when a fi xed amount of PEG-400 was mixed. This graph reveals that the content of hydroge- nated lecithin and poloxamer 407 had almost no effect on elasticity. Figure 11C shows the effect of the amounts of poloxamer 407 and PEG-400 when a fi xed amount of hydro- genated lecithin was mixed. This graph demonstrates that elasticity increased with an increase in the content of poloxamer 407 and a decrease in the content of PEG-400. The aforementioned fi ndings suggest that PEG-400 concentration appeared to have the most signifi cant impact on elasticity, whereas the content of hydrogenated lecithin and polox- amer 407 had insignifi cant infl uence on elasticity. FACTORS A FFECTING VISCOSITY (Y2) The resul ts of the ANOVA on viscosity indicated that a quadratic model was appropriate. The p value was 0.0215 ( 0.05) F value was 5.0995, having a signifi cance and the coef- fi cient of determination (R2) was 0.8677. The response surface model for viscosity is as follows: 2 2 2 2 = 0.7035– 0.1748 – 0.2050 + 0.0358 + 0.0916 – 0.1113 – 0.0248 + 0.0186 + 0.0758 – 0.2099 , Y A B C AB AC BC A B C (2) wher e Y 2 re presents the viscosity. In the aforementioned equation, hydrogenated lecithin and PEG-400 were shown to have a reciprocity relation with viscosity, a response vari- able. The squared value of poloxamer 407 content was found to have a reciprocity relation with viscosity. As shown in Table VI, the p value of each coeffi cient revealed that the Table VI ANOVA for RSM Model Run Y1, elasticity Y2, viscosity Y3, DSC EI (1/nm2) MVI (1/nm2) mW/mg p-value 0.0040 0.0215 0.0107 F-value 7.3545 5.0995 5.6237 R2 0.6292 0.8677 0.5648 Model Linear Quadratic Linear PREPARATION AND EVALUATION OF PLURONIC LECITHIN ORGANOGELS 339
content of hydrogenated lecithin and PEG-400 was found to have a signifi cant effect on viscosity ( 0.05), and the squared value of poloxamer 407 content was shown to have a signifi cant effect on viscosity ( 0.05). The 3D response surface graph demonstrates that viscosity decreased with an increase in the content of hydrogenated lecithin and PEG- 400, and decreased in a curved-shape form with an increase in the poloxamer 407 content (Figure 12). Using equa tion 2, coeffi cient was high in the order of PEG-400 B (0.2050), hydroge- nated lecithin A (0.1748), and poloxamer 407 C (0.0358), indicating that the main effect was signifi cant in the order of B, A, and C. Figure 12 depicts the 3D response surface graph using a regression equation. Figure 12A is the graph that shows the effect of the amounts of hydrogenated lecithin and PEG-400 on viscosity when a fi xed amount of poloxamer 407 was used. This graph reveals that viscosity gradually decreased as the amounts of hydrogenated lecithin and PEG-400 increased, but Figure 11. Response surface for elasticity results of PLO gels as a (A) hydrogenated lecithin and PEG-400, (B) hydrogenated lecithin and poloxamer 407, and (C) poloxamer 407 and PEG-400. (A) hydrogenated leci- thin, (B) PEG-400, and (C) poloxamer 407. JOURNAL OF COSMETIC SCIENCE 340
Purchased for the exclusive use of nofirst nolast (unknown) From: SCC Media Library & Resource Center (library.scconline.org)