j. Soc. Cosmet. Chem., 34, 273-284 (September/October 1983) Analysis of subsidiary colors in D&C Red No. 9 and its purification: Development of non-allergenic D&C Red No. 9 MASAYUKI NAGANUMA, YUTAKA OHTSU, YOSHIO KATSUMURA, MASAHIRO MATSUOKA, YOSHIHIRO MORIKAWA, MUNEO TANAKA, and TAKEO MITSUI, $hiseido Laboratories, 1050 Nippa- cho, Kohoku- ku, Yokohama- shi, 223Japan. Received February 15, 1983. Presented at the 12th Congress of the International Federation of Societies of Cosmetic Chemists, Paris, September 1982. Synopsis Analysis of commercial D&C Red No. 9 revealed the presence of eleven aromatic azo compounds (subsidiary colors) including Sudan I (1-phenylazo-2-naphthol), which has previously been reported as a strong sensitizer (1-5). Results of the modified guinea pig maximization test (4) indicated that each of the subsidiary colors was a contact sensitizer, while D&C Red No. 9 itself was not. It was found that aromatic amine impurities contained in Red Lake C Amine (2-amino-4-methyl-5- chlorobenzenesulfonic acid) were the source of formation of the subsidiary colors in D&C Red No. 9. It was also confirmed that D&C Red No. 9 synthesized with Red Lake C Amine from which these aromatic amines were removed as completely as possible, contained no subsidiary colors. Based on these results, a practical manufacturing process for D&C Red No. 9, free from contact sensitizers, was developed. INTRODUCTION D&C Red No. 9 has been widely used for makeup products including lipstick. It is an extremely important red pigment, but it has been reported that D&C Red No. 9 is a contact sensitizer (5). In order to determine whether the contact sensitivity is due to the color itself or the impurities contained therein, an analytical investigation and modified guinea pig maximization tests were carried out. Many impurities were identified through the analysis of D&C Red No. 9. The modified guinea pig maximization test showed that these impurities were contact sensitizers, while the color itself was not. As a result of the investigation, a manufacturing process to produce D&C Red No. 9 free from contact sensitizers was developed. This report deals only with D&C Red No. 9. However, the same approach has been taken for other colors to identify the causative 273

274 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Chloroform Extract ] I [HPLC Separation (Condition l)[ I l Fraction 11 • I IHPLC Separation (Condition 1I) I I I Subsidiary Colors I Condition I Column : TSK LS-410', 8ram i.d.x 25cm, Temp.: R.T. Eluent : Methanol, 4ml/min. Detector: UV (254nm) * TOYO SODA (JAPAN) Condition 1I Column : Lichro$orb S1-100', 8mm i.d. x25cm, Temp.: R.T. Eluent : 2.5•25% chloroform/n-hexane, 4ml/min. Detector: UV (254nm) * MERCK (GER) Figure 1. Flow chart for isolation of subsidiary colors in D&C Red No. 9. compounds of contact sensitivity and to establish the purification process. These purified colors have been successfully formulated into cosmetic products. Recently, there are some indications that D&C Red No. 9 is to be delisted because the color is suspected of being carcinogenic during long term feeding studies. Even if D&C Red No. 9 is delisted, our study can be applied to other colors. MATERIALS Eight samples of D&C Red No. 9 produced by three manufacturers in Japan and the U.S. were used. Four of them were straight colors, and the remaining four were diluted with barium sulfate. Six samples of Red Lake C Amine produced by four Japanese manufacturers and meeting the standard set forth by the Japanese Industrial Standard (jIS) were used. Two samples of •-naphthol produced by two Japanese manufacturers were employed, which met the standard of the JIS. I 9 115 143 ..• I ,I ':i ß , , 0 50 1(•0 150 200 OH 250 300 (M/Z) 350 I 115 171 I 0 ,q 0 1(•0 1 ,•:)0 20O 2.62 CH3 OH 250 3iJ0 (M/Z) 350 143 •). 115 11•3 1•)0 2i•0 2,•0 300 (M/Z) 350 Figure 2. Mass spectra of isolated components.