VITAMIN A AND LIPOIC ACID STABILITY 407 the hydrolysis of ascorbic acid but does not guarantee satisfactory stability levels in fi nished products. Instead, the introduction of the phosphoric group in the 2 position (Figure 3) pro- tects the molecule from break-up of the enediol system, confi rming that phophate esters of vitamin C as stable derivatives of vitamin C may be easily used in cosmetic products (14). Although we have included vitamin C derivatives in the original formulation, the stability of vitamin A was enhanced but not the stability of lipoic acid. We supposed there was an incompatibility between lipoic acid and components in the formulation. As a part of the ongoing project on the development of formulations containing lipoic acid, techniques of thermal analysis (DSC), isothermal stress testing (IST), and IR were utilized for drug-excipient compatibility testing (15). In the fi rst phase of the study, DSC was used as a tool to detect any interaction. Use of DSC has been proposed as a rapid method for evaluating the physicochemical interaction between two components. Based on the DSC results alone, methyl p-hydroxybenzoate, propyl p-hydroxybenzoate, butylated hydroxytoluene, non-ionic self-emulsifying wax, propylene glycol, and acetylated lanolin were found to exhibit interac- tion with lipoic acid. Stressed binary mixtures (stored at 50°C for one week) of lipoic acid and excipients were evaluated by HPLC. Binary mixtures were evaluated by IR spectroscopy. The results obtained with the thermal analysis were confi rmed by HPLC and FT-IR studies (15). Based on these results, we proposed some new formulations that were stored in three dif- ferent packaging materials—polyethylene, polypropylene, and glass—in order to evaluate if there is a signifi cative difference in stability in these formulations. There were not sig- nifi cant differences between the stability of vitamin A and lipoic acid in the different packaging materials (16). MATERIALS AND METHODS MATERIALS AND REAGENTS Ascorbyl palmitate was provided by Hoffmann La Roche (Switzerland), magnesium ascorbyl phosphate by Merck (Germany), butylhydroxytoluene by Eastman Chemical Figure 3. (A) Vitamin C. (B) Ascorbyl palmitate. (C) Magnesium ascorbyl phosphate. (D) Sodium ascorbyl phosphate.

JOURNAL OF COSMETIC SCIENCE 408 Company (USA), and vitamin C by Kromberg (Argentina). Vitamin A (as palmitate) was provided by DSM (Netherlands), vitamin E (as acetate) by Merck (Germany), and lipoic acid by Labochim (Laboratorio Chimico Internazionale, Italy). The emulsions consisted of silicone fl uid (Dow Corning, Brazil) mineral oil and petrola- tum (R.A.A.M., Argentina) as oil phase anionic self-emulsifying wax (Flamacer SX, Fla- maquímica, Argentina) as surfactant imidazolidinyl urea (ISP, United Kingdom) as preservative and sorbitol 70% (water solution) (Unión Química Argentina, Argentina) and demineralized water as hydrophilic phase. All chemicals used were of analytical grade. Methanol, acetonitrile, and water were of HPLC grade. Solvents were fi ltered through a 0.45-μm membrane and degassed. PREPARATION OF THE EMULSIONS The anionic emulsifi er was melted in a stainless steel container then silicone fl uid and mineral oil were added. The emulsion was mixed by slow agitation, avoiding the incor- poration of air and keeping the temperature between 72°C and 74°C. Lipoic acid and butylhidroxytoluene were then added. The emulsion was stirred, maintaining the tem- perature until a full dispersion was obtained. Demineralized water, sorbitol 70%, and imidazolidinyl urea were mixed in another stain- less steel container. This mixture was heated to 75°C. Both phases were fi ltered by gravity fi ltration. Then mixture 1 was incorporated into mixture 2 and stirred at 900 rpm for fi ve minutes. Then cooling was started and stirring was slowed down. Vitamins A and E, ascorbyl palmitate, sodium ascorbyl phosphate, magnesium ascorbyl phosphate, and vitamin C diluted in water were incorporated at 45°C. The emulsions were stored for 15 months at room temperature, and were analyzed under the same conditions in all cases. The quantitative compositions of the formulations are shown in Table I. The pHs were corrected, adding 1 N phosphoric acid or 1 N sodium hydroxide if needed. ANALYSIS OF THE ACTIVE INGREDIENTS The analyses of lipoic acid and vitamin A were made by HPLC. Materials and reagents. The working standards employed for lipoic acid and vitamin A were the same as those used in the preparation of the creams. The solvents were HPLC grade. Water (HPLC grade) was obtained by distillation and passed through a 0.45- micron membrane fi lter. Instrumentation. The HPLC system consisted of a dual-piston reciprocating Spectra Phys- ics pump (model ISO Chrom. LC pump), a UV-Vis Hewlett Packard detector (Model 1050), a Hewlett Packard integrator (Series 3395), and a Rheodyne injector (Model 7125). HPLC conditions. The experiment was performed on a LiChroCARTR 125*4 mm HPLC Cartridge LiChrospher® 100 RP-18 (5 μm) (Merck, Darmstadt, Germany) for vitamin A.