RISK-BEARING SUBSTANCES IN COSMETICS 185 Confirmation of the identity of the components by glc of the important aromatic diamines and separately of the polyphenol groups was then obtained. The aromatic diamines are separated on an alkaline column (Apiezon L-KOH) •. The aminophenols and the diphenols do not give a response on this column, so that a good confirmation can be made of the identified aromatic diamines by injecting a freshly made crude extract. After several injections the column should be cleaned at a higher temperature, in order to avoid the appear- ance of 'ghost' peaks which generally belong to former injections. The following injection-sequence scheme during the day is highly recommended. Quantitative work is also possible with this column, which however requires precleaning of the extract. The internal standard used is 2 chloro.p.phenylene diamine. The pre- cision of the method, however, is not very high (19, 20, 21). The glc column is selective for the aromatic diamines. Aminophenols and the polyphenols do not give any response. The same extraction as for tlc was used, except that ethyl-acetate only was used as solvent. Freshly made extracts should be used. Experimental Glc conditions: Column 150 cm, 0.25 inch cuts. diam., stainless steel. Filling: Apiezon L 6•o and KOH 10•o on Chromosorb WHP 80/100, which is prepared as follows: Dissolve 420 mg Apiezon L in some toluene. Dissolve 700 mg KOH in a small amount of water and much ethanol. Mix both solutions and add to 5880 mg Chromosorb WHP 80/100. Dry the mixture in a rotating vacuum evaporator, keeping the mixture as granular as possible. Column temperature: 150øC iso- therm. Carrier gas: Nitrogen. Detector FID. Glc reference mixture: a solution in ethylacetate, containing per 100 ml: 45 mg PFD--50 mg MFD--80 mg 25TDA--70 mg 24TDA--190 mg 24DAA. A chromatogram of this mixture (0.7 pl) is shown in Fig. 5. As fouling of the column occurs, it must be cleaned periodically. Otherwise 'ghost peaks' appear after several injections. It is therefore necessary to work exactly according to the following scheme: Injection of the glc reference mixture: 10 min (and not longer). Injection of extract of the first sample: 10 min (and not longer). Injection of the second sample extract: 10 min (and not longer). Injection of the third sample extract: 10 min (and not longer). Cleaning the column at 250øC from 30 min. The quantitative determination of 25TDA is described, but the method can also be modified for the other aromatic diamines. An internal standard (2.Chloro. p.phenylenediamine) is used.

186 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS I I J I0 20 .30 0'7HL gl meysll I I I I I I 40 50 60 70 150øC 50mm/min % 80 90