DSC ANALYSIS OF HAIR IN WATER 223 Table II Denaturation Temperatures T o and Enthalpies AH o for Hair Samples Perm-Waved Multiple Times Number of treatments Time of treatment (min) T o ñ s (ø) AH o ñ s (,J/g) Start -- 155.2 + 1.03 15.3 + 0.32 Bleach 152.5 + 0.57 14.9 ñ 0.40 1 10 152.9 ñ 0.58 11.3 ñ 0.36 20 150.3 ñ 0.36 11.5 + 1.05 30 148.7 ñ 1.08 10.2 ñ 0.56 3 10 146.4 ñ 1.07 8.5 ñ 0.97 20 144.0 ñ 0.86 7.4 + 0.68 30 143.5 + 0.57 7.5 ñ 1.04 5 10 136.7 ñ 0.71 3.2 ñ 0.97 20 134.6 ñ 0.74 1.9 + 0.36 30 121.3 ñ 2.01 1.0 ñ 0.33 s: standard deviation for threefold measurements. Start: untreated reference hair material. Bleach: once-bleached reference material, on which the perm-wave treatments were performed for different time periods. The variability of the locations and shapes of the peaks for individual samples can be attributed to general, inherent, natural inhomogeneities of the samples as well as to variations induced by the chemical and physical history of the samples beyond the applied treatments. BLEACHING EFFECTS Figure 2 summarizes the results for peak temperatures and denaturation enthalpies, respectively, for the bleached samples, in the form of a multiple "box & whisker" plot. The results show that the first four bleaches lead to a roughly linear decrease for both parameters. Beyond, T D levels off at a temperature that is 20øC lower than for the untreated material. In parallel, AH o decreases by 40% from 19 to 12 J/g. The plot of denaturation enthalpy vs temperature (Figure 3) reveals a strong linear relationship, showing the similarity in the course with which both parameters decrease. Taking their relationship to specific morphological components into account, as dis- cussed above, it can be concluded that bleaching leads to largely homogeneous damage in IFs and IFAPs. Assuming first-order kinetics and consequently plotting ln(HX•a) vs the number of bleachings, as realized in Figure 4, yields a well-defined straight line from which the apparent reaction rate constant (kn = 0.069 ñ 0.0103 95% confidence limits) is de- duced. Leroy et al. (13) investigated virgin, bleached, and permwaved hair by DSC in the dry state. For the virgin hair they observed a strong bimodality of the curves, which is related to fractions of cortical cells differing in cystine content (10), rather than to a filament/matrix structure (12,15).

224 JOURNAL OF COSMETIC SCIENCE 160 155 ß ,-, 150 E 145 o 140 • 135 130 0 1 2 3 4 5 6 Number of Bleachings 20 Figure 2. Denaturation temperatures T O (left y-axis) and enthalpies AHr• (right y-axis) of the bleached samples as multiple "box & whisker" plot, characterized by the arithmetic means (symbol), the standard errors (box), and the expectation ranges for the 95% confidence limits (whisker). 0 refers to the untreated material. lO 135 y = -33.6 + 0.33 x r = 0.96 140 145 150 155 160 Denaturation Temperature in øC Figure 3. Individual values for denaturation enthalpy AH D plotted versus temperature TD for the bleached hair samples. The linear regression line through the data and its description are given. r is the correlation coefficient. They observed that with bleaching the DSC peaks for dry fibers shift to higher tem- peratures. This is in contrast to our results for wet fibers. Bleaching leads to increased concentrations of cysteic acid and thus of ionic interactions. It can be assumed that these will be effective in the dry fiber but will be broken in the wet state, thus leading to different shift directions of the denaturation peaks through a bleaching treatment.