541 CHARACTERIZATION OF BLEACHED HAIR E a is the activation energy for denaturation of α-keratin. By assuming that the denaturation mechanism is invariable at different heating rates, the peak in T D may be used to approximate α = 0.5 (8,27,28). Figure 18 demonstrates the correlation of the assessed E a against the EDF determined by FTIR imaging. The results indicate that the energy barrier for denaturation decayed linearly for the first hour of bleaching (R2=0.99) — after which, the activation energy plateaued at approximately 180 kJ/mol. The range of determined E a values agrees with the literature however, perhaps due to inherently complex multistep keratin denaturation processes, or to the source of our virgin European dark brown tresses, we observed a decrease in E a , rather than an increase, with increased bleaching times (13,14). DRY DSC OF BLEACHED HAIR FIBER SNIPPETS In dry DSC experiments, no water is added to the DSC pans and denaturation and pyrolysis events are difficult to distinguish. Hence, we accepted that denaturation and pyrolysis were not Figure 18. HPDSC Ozawa denaturation activation energy versus FTIR imaging EDF band area ratio. The Ea plateau at 180 kJ/mol is associated with bleaching times ≥120 min. The standard deviation is ± 13 kJ/mol. Results from the Kissinger method (using TD) are overlaid with results from the OFW Ea analysis (27). The thermal measurements were performed in distilled water. Figure 19. Dry DSC critical matrix mobility and denaturation/pyrolysis enthalpy versus Raman EDF−1. The critical mobility temperature was assessed by the peak in Cp as a function of increasing temperature.

542 JOURNAL OF COSMETIC SCIENCE unique events and instead decided to use our sapphire-calibrated DSC to monitor continuous changes in C p as a function of increasing temperature. Heat capacity changes are related to changes in mobility, where increases in sample mobility at lower temperatures are connected to decreased physical and covalent matrix cross-link density. Figure 19 shows that conversion of -S—S- to -SO 3 − appears to have increased the temperature of the apparent C p inflection, meaning that ionic and hydrogen-bonded networks stabilized the dry-fiber cortex during applied heating. The peak in C p linearly correlates with increasing cysteic acid (R2=0.96), where the C p maxima ranged from 223°C for the unbleached control to 234°C for the 240 min bleached sample. Figure 19 additionally graphs the dry DSC pyrolysis enthalpy against trends in the normalized Raman 509 cm−1 band and clearly demonstrates that more thermal energy is required to vaporize matrix components with higher levels of cystine (R2=0.95). Hence, the formation of two moles cysteic acid from the scission of one mole of disulfide influences the disruption mechanism for cortical denaturation and pyrolysis. One plausible explanation is that after bleaching strongly acidic cysteic acid salts (pKa 2) rearrange and form strong ionic cross-links with basic ammonium moieties in keratin. Consequently, modified denaturation/ pyrolysis pathways are required to initiate cortical flow and pyrolysis of matrix protein (3,14). Figure 20 pictorially demonstrates the deleterious effects of excessive heating, which involved flow (left image) and the synchronous vaporization of cortical components (right image). To produce flow, scission of disulfide bonds in the matrix and interphase reduced the cross- link density and lowered the matrix viscosity to critical values. TGA-FTIR analysis of the gaseous pyrolysis effluent confirmed that flow was then activated by liberation of carbon dioxide gas (CO 2 ) from the degrading IFKPs, which pushed the molten components out of the durable microtubule. Along with other unpublished supporting micrographs, the right image suggests that vaporization occurred from the medulla outward, which is a result that had been previously reported in the literature (29). MTGA OF BLEACHED HAIR FIBER SNIPPETS MTGA combines nonlinear heating with a sinusoidal temperature program to obtain pyrolysis events and kinetic parameters in a single experiment. Furthermore, by combining modulation with dynamic high-resolution thermogravimetry, overlapping weight loss events may be more easily resolved. MTGA is a dry fibers testing technique, where hair fiber snippets are pyrolyzed at high temperatures in a 0% RH nitrogen environment. Figure 21 plots the activation energy for pyrolysis against cysteic acid concentration, where the bleaching times increase from left to right on the abscissa. Next to the E a data are the MTGA pyrolysis onset temperatures, where the extrapolated onsets correlate positively and Figure 20. View of the cortex of a bleached fiber at temperatures 250°C. In the left image, note the cortical remnants that extruded from the interior of the fiber. In the right image, observe the reduction in cortical material and the persistence of the cuticle. At higher temperatures, the cortex was torn asunder and only the cuticle remained.