402 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS (18) Bolliger, R., and Muenzel, K., Pharm. Acta Helv., 33, 141 (1958). (19) Mantell, C. L., in "Natural Plant Hydrocolloids" (Ref. 5), p. 29. (20) Schon, S. A., and Fuerst, W. J., Dansk. Tids. Farm., 15, 34 (1941). (21) Tillman, W. J., and Kuramoto, R., 7. Am. Pharm. Assoc., $ci. Ed., 46, 211 (1957). (22) Eisman, P. C., Cooper, J., and Jaconica, D.,Ibid.,46, 144 (1957). (23) Taub, A., Meet, W. A., and Clausen, L. W., Ibid., 47, 235 (1958). (24) deNavarre, M. G., Am. Perfumer Aromat., 70, 31 (1957). (25) Meyer, R. J., and Cohen, L., 2 e. $oc. Cosmetic Chemists, 10, 143 (1959). (26) Davies, R. E. M., and Rowson, J. M., 2 e. Pharm. andPharmacol., 10, 30 (1958). (27) Bolliger, R., and Muenzel, K., Pharm. Acta Helv., 34, 79 (1959). (28) Miyawaki, G. M., Patel, N. K., and Kostenbauder, H. B., 7. Am. Pharm. Assoc., Sci. Ed., 4$, 315 (1959). CORROSION TESTING OF AEROSOL PRODUCTS By MoP, P, IS J. ROOT and MERLIN J. MAUP, Y* Presented May 7, 1959, New York City SINCE THE ADVENT of cosmetic aerosols in metal cans, corrosion has taken its place as a major problem in the formulation of these products. Corrosion problems in metal cans have, of course, been dealt with for a good many years, notably in the food industry, as well as others. The classic method of evaluating corrosion in metal containers has been the test pack. The product is placed in the container, sealed and allowed to stand at vari- ous temperatures for a specified period of time and then opened and in- spected. If, after one year's storage, the container and contents were still in good condition, a one-year shelf life could be anticipated. Or, if as was recently the case in England, a can of beef was opened after storage for 134 years and found in good condition, a shelf life of a comparable period could be guaranteed. In this instance, however, it is of little commercial im- portance since the can made at that time was hand wrought and weighed 7 pounds 5 ounces. Although temperature is an important factor in shelf life test, it is still no substitute for time, as elevated temperatures are not always a guarantee of accelerated corrosion. One notable example was the case of the aerosol shampoo which was test packed at room temperature and at 130øF. Since the cans stored at 130øF. for three months were in good condition, no one felt it necessary to examine the cans which were stored at room temperature before the product was marketed. When the cans on the retailers' shelves began perforating, a quick examination of the cans stored in the laboratory at room temperature showed up the same condition. Test pack procedures require an extensive time period and, with today's dynamic marketing practices, especially in our industry, the need for a * G. Bart & Co., Chicago 9, Ill.

CORROSION TESTING OF AEROSOL PRODUCTS 403 direct, less time consuming method is obvious. Nevertheless, the test pack is still indispensible for relating the newer and faster methods with shelf- life experience. In the labora tory, test pack testing is carried out as follows: 1. Shelf-life tests at room temperature. 2. Oven shelf-life tests. 3. Chilling and opening containers. 4. Examination of product and container. In 1955, (1) we reported on a method of analyzing aerosol products for metallic contamination due to corrosion at various periods up to ten weeks. From this analysis it was possible to gain an insight into the electrolytic be- havior of the container metal. Thus, it was possible to demonstrate from the greater iron contamination in an alkyl sulfate shampoo that the iron in the container was anodic to the tin. This method has proven extremely useful for detecting the metallic contamination level at which product dis- coloration or perfume oil deterioration sometimes takes place. It has been noted that in certain instances, product discoloration can occur with as little as 5 p.p.m. of iron or tin in the product from corrosion processes. Corrosion at this level would not be apparent by container examination. Perfume deterioration has been detected with as little as 20 p.p.m. of tin in the product from corrosive action on the tin plate. Progress in corrosion studies took a big stride when, in 1927, corrosion began to be studied from an electrochemical viewpoint. The current flow- ing between the anodes and cathodes that exist on the surface of a corrod- ing metal in accordance with Faraday's Law, is a direct measure of the rate of corrosion. If it were possible to measure this corrosion current as the corrosion reaction proceeds, there would be little need for any other type of corrosion test since, according to Faraday's Law, the amount of current that flows in a given time is a quantitative measure of the amount of metal that has corroded in that time. When two metals are in contact with each other in solution, chemical reaction is converted to electrical energy--as is the case in a battery. The reactions which take place at the metals are as follows: dnodic Reaction M = divalent metal M = M ++ 4- 2e- Cathodic Reaction l/=O= 4- HeO 4- 2e- = 2OH- 2H*4-2e- = He