116 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS whereas c•-tocopheryl acetate is a feature ingredient in many cosmetics, and, as such, is usually present in relatively higher concentrations. The recent popularity of vitamin E preparations for topical application, together with reports of adverse dermatological reactions linked to cosmetics containing this vitamin (1,2), led to concern about the desirability of including these compounds in such products. Therefore, a sensitive gas-liquid chromatographic (GLC) method was developed for their determination. Vitamin E has been determined in other commercial preparations by GLC. Pillsburyet al. (3) measured vitamin E in pharmaceuticals by GLC on an 8 ft glass column packed with 5 per cent SE-30 on Gas-Chrom P. Their method involved external standardiza- tion, which necessitates daily construction of standard curves. Mahn et al. (4) de- termined vitamin E in multivitamin preparations by GLC on a 4 ft copper column packed with 10 per cent SE-30 on Aeropak 30. They selected dotriacontane (a C-32 hydrocarbon) as the internal standard and thus eliminated the need for calibration curves. They observed no differences between results obtained with metal or pyrex GLC columns. However, they worked with relatively large amounts (10-15/a,g), and losses of c•-tocopherol due to the metal column may not have been significant. Their analysis of multivitamin creams included a preliminary separation on alumina. Subse- quently, Rudy and others (5) conducted a collaborative study on a method similar to the one developed by Mahn and coworkers. The study, although showing the method to be precise, did not establish accuracy because collaborators' results could not be compared to actual amounts present. The procedure reported here incorporates the best features of these previous methods and includes modifications to decrease the time of analysis and to accommodate the small amounts of vitamin often present in complex cosmetic mixtures. It consists of preliminary extraction of the vitamin from cosmetics with organic solvents, adsorption chromatography on alumina with hexane and ether as the eluants, and determination by GLC on a 10 ft x 4 mm i.d. glass column packed with 5 per cent SE-30 on Gas- Chrom Q. Results were calculated by the method of internal standardization in which dotriacontane served as the internal standard. EXPERIMENTAL APPARATUS AND REAGENTS Adsorption column.' An aluminum oxide chromatographic column was prepared as follows: aluminum oxide,* 80-100 mesh, neutral certified Brockmann activity I, was.::? brought to Brockmann activity III by thoroughly mixing 1 ml of water with each 8 g of:• aluminum oxide and allowing this mixture to equilibrate overnight in a sealed con- tainer. Eight g of the equilibrated aluminum oxide was then slurried in hexane and.• poured into a 1.5 x 30 cm glass chromatographic tube containing a glass wool plug and 25 to 30 ml of hexane. Hexane was drained to the level of the adsorbent. The level of liquid in the tube was not permitted to fall below the top of the alumina before or dur• ing the analysis. *Fisher Scientific Co., 711 Forbes Ave., Pittsburgh, PA 15219.

DETERMINATION OF VITAMIN E 117 Gas chromatograph.' An F and M* model 810 gas chromatograph equipped with a flame ionization detector was used with the following operating conditions: temperatures: detector 348øC, injection port 300øC, oven 285øC carrier gas flow (helium) 66 ml/min maximum range and attenuation settings, 10 x 32 or such that 8 to 10 •g ofd-•-toco- pheryl acetate gave 50 per cent or greater recorder response. GLC Column: The GLC column was prepared as follows: 80-100 mesh Gas-Chrom Q•- was coated with approximately 5 per cent SE-30• by the funnel coating method (6). The support was packed in a 10 ftx 4 mm i.d. glass column using line vacuum and gentle vibration. The packed column was conditioned overnight at 285øC with no car- rier flow and then at 285øC for 2 days with a carrier flow of 10 to 20 ml/min. The flow was raised to 66 ml/min, oven temperature was lowered to 200øC, and 50/xl of Silyl-8 II were injected onto the column. Stock solutions: Individual stock solutions ofd-g-tocopherol,# d•-tocopheryl acetate,# and dotriacontane** were prepared weekly by accurately weighing approximately 50 mg of each into separate 10 ml volumetric flasks and diluting to volume with hexane. Because d-•-tocopherol is easily air oxidized and is light sensitive, its stock solution was stored under nitrogen. When not in use, the stock solutions were refrigerated. Ex- posure to light can be minimized by using amber volumetric flasks or by wrapping the flasks with aluminum foil. Solvents.' Solvents were American Chemical Society (ACS) reagent grade, or equivalent. PREPARATION OF COSMETICS Known amounts ofd•-tocopherol or d-a-tocopheryl acetate in hexane were added to samples of the following cosmetics: a commercial bath oil a bath oil containing Tween 61,-•-ñ isopropyl myristate, talc, and perfume oils a vanishing cream containing stearic acid, cetyl alcohol, water, glycerol, KOH, NaOH, and perfume oil a vanishing cream containing stearic acid, lanolin, Carbopo1934,*** methylparaben, ethylparaben, propyo lene glycol, water, triethanolamine, and perfume oil and two commercial shampoos. The final preparations contained 0.1, 1, or 7 per centd-a-tocopherol or d-a-tocopheryl acetate. Because d-•-tocopherol is light-sensitive, it was added to the sample just prior to analysis and exposure to light during the analysis was minimized. *F and M Scientific Corp., Avondale, PA 19311. 'l'Applied Science Laboratories, Inc., P.O. Box 440, State College, PA 16801. 1:General Electric Co., Schenectady, NY. ]1 Pierce Chemical Co., P.O. Box 117, Rockford, IL 61105. i '•: •Eastman Kodak Co., 343 State St., Rochester NY 14650. **Arialabs, Inc., 80 Republic Dr., North Haven, CT 06473 ß •'•'•'Atlas Chemical Industry, Wilmington, DE. **'B.F. Goodrich Chemical Co., 6100 Oak Tree Blvd., Cleveland, OH 44131.