382 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS gradual change in weight, with significant loss occurring at 20 cycles and above, while the samples chlorinated in alkaline solutions show no significant change. These results do not agree fully with the results found in the literature on the chlorina- tion of wool (13,18,19) in which the minimum weight loss occurred in the neutral (not the alkaline) and the highest occurred in the acidic region. Valk (18) attributed the differences in weight loss of wool with chlorination at different pH levels to differences in the extent to which the peptide bond cleavage and the disulfide bond oxidation takes place. In the acid region, both the cleavage (of peptide bonds) and oxidation (of cystine) are extensive in the neutral region, cleavage is low but oxidation is high and, in the alkaline region, cleavage is high but oxidation is limited. Extensive cleavage of peptide chains and their subsequent removal through the epicuticle results in weight loss de- spite the weight addition caused by binding oxygen and chlorine to the hair. A major difference in the chlorination studies of wool and hair is the concentration of the chlorine in solution, the hair studies generally involving much lower concentra- tions. In view of the results found here on human hair and the explanation given by Valk on wool, one could easily conjecture that in the present study in which hair was exposed to dilute concentrations of chlorine (10 ppm), peptide bond cleavage would be limited but disulfide bond oxidation would be predominant. The main difference in the results of chlorination at different pH would arise from differences in the extent of oxidation of disulfide bonds. The oxidation being extensive in the acidic region and minimal in the alkaline region, one could expect weight loss also to be highest in the former condition and lowest in the latter. The reduction in force and in work to extend wet fibers 20 percent after chlorination are shown in Table III. The two parameters showed very similar trends. The most signifi- cant change in these parameters with number of cycles of treatment took place in the pH 2 samples. The change at pH was more gradual (still significant over a difference of 20 cycles of treatment), and that at pH 10 essentially absent. The mechanical properties of wet keratin fibers are typically explained in terms of a model in which crystalline microfibrils are embedded in a matrix of noncrystalline material. As the fiber is ex- Table III Effect of pH and Cycles of Chlorination on the Reduction in Force and in Work Required to Extend Blond Hair Fibers 20% % Reduction % Reduction Cycles pH force s* n** work s 10 10 0.42 1.45 12 2.03 4.30 9 7 4.03 1.34 11 8.00 3.03 8 2 9.74 2.01 12 12.81 1.97 9 20 10 - 1.10 1.10 12 -0.23 3.79 9 7 7.76 1.61 11 13.35 3.78 8 2 20.18 6.76 11 25.48 3.18 8 30 10 1.24 1.26 12 2.60 3.53 9 7 19.19 1.75 11 21.62 3.58 8 2 40.77 6.58 12 48.22 9.17 9 * s = standard deviation. **n = number of observations.

CHLORINATION OF HAIR AND pH 383 Table IV Effect of pH and Cycles of Chlorination on the Ratio of Knot Breaking-to-Fiber Tenacity (K/T) of Blond Hair Cycles pH K/T s* 0 10 0.575 0.055 7 O.6OO 0. 134 2 0.591 0.155 15 10 0.594 0.063 7 O.924 O.O77 2 O.796 0.07! 30 10 0.648 0.084 7 O.955 0. 148 2 0.850 0.122 * s = standard deviation for eight observations. tended, the crosslinked matrix offers resistance to the unfolding of the microfibrils (10b). Treatment with chlorine reduces the number of disulfide crosslinks in the matrix material, thus leading to a •lecrease in the force necessary to extend the fibers. The values of the ratio of knot breaking-to-fiber tenacity (K/T) of fibers tested under standard laboratory conditions are shown in Table IV. The values of K/T were generally affected by both pH and cycles of treatment. The effects of cycles at pH I0 and the effect of pH on control samples are not significant. At pH 2 and 7, increase in cycles from 0 to 15 resulted in a highly significant increase in the value of K/T. Increase in cycles beyond I5 produced little further change. Decrease in pH from I0 to 7 resulted in a significant increase in the"value of K/T, but further decrease in pH to 2 did not result in a significant change in K/T. The above results of the knot strength in chlori- nated fibers give no evidence of brittleness. Rather, the results indicate an increase in flexibility with chlorination at pH 7 and 2. Consequently, the harsh feel commonly associated with chlorinated hair may be attributed more appropriately to altered surface morphology and increased interfiber friction than to increased rigidity of the fibers. CONCLUSIONS This study has demonstrated that the effects of mild chlorination on various properties of human hair are dependent on the pH of the reaction medium. The two hair types, the natural blond and the dark brown, reacted similarly to the chlorination treatment. Acid chlorination produced the greatest and most rapid change in interfiber friction, surface morphology, weight loss, and the reduction in force required to extend wet fibers 20%. Chlorination at pH 7 produced a more gradual change in these properties, and alkaline chlorination produced little significant change. The results of this study generally agree well with the effects found from the chlorina- tion of wool in more highly concentrated solutions. Both systems produce similar changes in frictional properties, surface morphology, and wet tensile properties under acidic and neutral conditions. Under alkaline conditions, however, some dissimilarities exist. There was little apparent effect of chlorination in dilute alkaline solutions on the various fiber properties studied. This lack of effect could be the result of limited pene- tration of hypochlorite ions into the hair.