J. Soc. Cosmet. Chem. 28 615-620 (1977)¸ 1977 Society of Cosmetic Chemists of Great Britain A 0olorimetrio method for determining hydrogen peroxide in emulsions K. C. JAMES and E. TSIRIVAS The Welsh School of Pharmacy, University of Wales Institute of Science and Technology, Cathays Park, Cardiff, Great Britain, CF1 3NU and P. D. Papoytsanis S.A., Kato Kifissia, Athens, Greece Received 10 June 1976 Synopsis A method has been developed for the colorimetric determination of hydrogen •roxide, and is dependent on the intensity of the blue colour obtained with dichromate. The effects of a limited range of surfactants on the reproducibility of the process have been investigated, and the method used to determine hydrogen peroxide in two cosmetic emulsions. INTRODUCTION There are numerous methods available for determining hydrogen peroxide, but none can be applied with confidence to its estimation in emulsions. Potassium permanganate gives a fading end point in the presence of organic matter, and iodimetric methods (1), while less sensitive to organic materials, are vulnerable to surfactants (2, 3). Titration with ceric ammonium sulphate is known to be satisfactory in the presence of ethanol and diethyl ether (4) and some organic peroxides, but has not been tested in the presence of surface active materials. Many colorimetric methods have been described, for example (5-7), but most depend on general oxidation and are not specific to peroxide. Some, for example (7) depend on complexation. The oxidation of dichromate ion to the blue, ether-soluble peroxochromic acid (CrOs) is a well known qualitative test for hydrogen peroxide, and is claimed to be specific. The reactions involved follow equations 1 and 2 (8). Cr•O? •- + H•O = HCrOc (1) H + + HCrOl- + 2H•O• = CrO5 + 3HzO (2) Colorimetric methods based on these reactions, and involving extraction with either ethyl acetate (9) or tri-n-butyl phosphate in benzene (10) have been described for the estimation of chromium. The method is now investigated as a possible means of dete•- mining hydrogen peroxide in cosmetic preparations. 615

616 K. C. James and E. Tsirivas EXPERIMENTAL APPARATUS AND MATERIALS Absorbance measurements were carried out using a Unicam Spectrophotometer H 1620, Mk. 2 Digital. Hydrogen peroxide, benzene, ethyl acetate, potassium dichromate, potassium permanganate and sulphuric acid were analytical grade (Analar) reagents, (British Drug Houses Ltd). Tri-n-butyl phosphate was laboratory grade, from the same source. It was freed from pyrophosphates before use (10). Solan E (an ethoxylated lanolin) and Polawax (a non-ionic emulsifying wax) were supplied by Croda Chemicals Ltd, while Texapon N25 (a solution of sodium alkylether sulphate) and Emulgrade F (a mixtree of cetostemyl alcohol, alcohol sulphate and non-ionic emulsifiers) were supplied by Henkel and Cie GmbH. ANALYTICAL METHOD Basic experimental conditions were taken from Sastri and Sundar (10) and modified in accord with our own observations, described below. The procedure finally adopted was as follows: 1 to 2 ml of sample were acidified with 1 ml ol'0.5 N sulphuric acid and a two to three fold excess of potassium dichromate added in the form of a 0.01 M solution. The mixture was cooled to about 8øC and adjusted to 10 ml with water at the same tempera- ture, followed by 10 ml of a 25•o solution of tri-n-butyl phosphate solution in benzene (TBP/benzene). The two phases were mixed well for about 30 sec and then allowed to separate, whereapon an aliquot of the benzene solution was transferred to the spectro- photometer. The absorption against pure solvent is shown in Fig. 1, and indicates a maximum at 584 nm, which was used for all subsequent measurements. 0.8 0.4 I • I • I 500 600 700 Wavelength (nm) Figure 1. Absorption curve of peroxochromic acid. [HaOa] = 2'82 X 10-3M. Successive extractions with small volumes of TBP/benzene were impracticable, because of the known instability of peroxochromic acid in water. Single extractions with