THE GREEN HAIR PROBLEM 3 The various tresses were entered into their respective treatment baths and the treatment was given for a period of i hr at 22øC, with continuous stirring of the solution. The tresses were then rinsed thoroughly in distilled water and blow-dried at room temperature. METHODS OF EVALUATION The actual chlorine contents in the various treatment baths were measured spectropho- tometrically using the ortho-tolidine method, standardized against iodine-thiosulfate titration (10). The actual copper contents in the treatment baths were determined by atomic absorption using a standard procedure (11). The various hair samples generated in the experiment were evaluated for imparted color, copper content and physical location of copper absorption. The imparted color was examined visually. Copper content was determined by atomic absorption (11) for this purpose, the sample was prepared by dissolving 27 mg of hair in 6 ml of conc. HNO 3, adding i ml of conc. HC104 to the solution, heating the solution to 200øC for 2 hr and making the volume up to i0 ml by addition of distilled water. The physical location of copper absorption was determined by energy dispersive X-ray analysis (EDXA) with the aid of a scanning electron microscope (SEM). For this purpose, the sample was prepared by flat-embedding the hair fibers in an epoxy resin block and cutting it across with a microtome so that the cross sections of the hair fibers can be mapped for copper. The resin block was mounted on a SEM stub and the surface to be examined was coated omnidirectionally with several hundred angstroms of carbon by vacuum evaporation to reduce the SEM charging artifact. The surface was then examined in a JOEL JSM-U3 scanning electron microscope equipped with an EDXA 707A energy dispersive X-ray analyzer. Maps for copper were made of the various hair cross sections using 100,000 X-ray counts, using a counting time varying between 15 and 75 min. RESULTS The contents of the four treatment baths before and after the treatments are shown in Table II. The results of the evaluation of the various hair samples are described in Table III. The EDXA-SEM photographs of cross sections of hair fibers from the tresses treated at the 6.0-pH condition are shown in Figure i. DISCUSSION OF RESULTS GENERAL OBSERVATIONS The copper depletion data (Table II) and the copper contents and imparted color of the hair samples (Table III) clearly indicate that hair absorbs the metal and consequently turns green when treated with a solution of a weak complex of the metal. This is the case with the experiment conducted at the 6-pH level. Little copper absorption and no color development occurred on treatment with the solution of the more stable copper citrate complex used at the 8.5-pH level. This lack of copper absorption cannot be attributed with certainty either to the stability of the citrate form

4 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Table II Initial and Final Contents of the Treatment Baths Treatment Bath//1 Treatment Bath//2 Initial Final Initial Final C• 2 ppm 2.7 0.24 2.7 0.15 Cu + + ppm 10.2 8.2 10.2 9.8 Actual pH 5.9 5.2 5.9 5.3 Treatment Bath//3 Treatment Bath//4 Initial Final Initial Final C•2 ppm 3.1 0.24 3.1 0.28 Cu + + ppm 10.6 10.5 10.6 -- Actual pH 8.6 7.5 8.6 7.4 or to the high pH without additional experiments. The EDXA-SEM photographs (Figure 1) show that the absorbed copper is primarily located in the peripheral region of the hair fiber. The results also show that even mild oxidation enhances copper absorption, while pretreatment with a quaternary ammonium compound formulation inhibits the process. The depletion data also register small reductions in the pH of the solutions and almost complete consumption of the chlorine from each of the treatment solutions. The high affinity of chlorine in hair and the resultant effects have been investigated in detail in another set of experiments in our laboratory. These factors fall beyond the scope of the present paper. As will be noted later, chlorine does not seem to play a role in green coloration of hair. The mechanism of hair-copper interaction cannot be deduced from these results alone. Published literature in reference to interaction between copper and insoluble keratins (12-15), however, speculatively attribute copper-keratin complex formation primarily to side-chain carboxyl groups and, to a small degree, to amino groups, disulfide groups and their derivatives. EFFECT OF MILD OXIDATION The larger amounts of copper absorbed by the oxidized hair are in agreement with the observations made by Edman and Marti (9), but there results cannot be readily Table III Evaluation of the Hair Samples Pretreatment Water Conditioner Hair Unoxidized Oxidized Unoxidized Oxidized Imparted color light green green none none pH = 6.0 Cu ++ content 1661 4545 128 366 /ag/g of hair Imparted color none none none none pH = 8.5 Cu ++ content 18 43 0 0 /zg/g of hair