539 CHARACTERIZATION OF BLEACHED HAIR suggesting that the intrinsic chemical environment of the amorphous and crystalline keratins had been degraded to a continuum of residual conformational states. For example, in the control sample, the onset of degradation started at temperatures above 135°C however, in the 240 min bleached sample, denaturation began at approximately 120°C. Table IV summarizes the HPDSC analyses performed on the 2–3 mm snippets of bleached hair exhibited in Figures 2 and 14. As surmised by gradual changes in T D and little change in ΔH D , the data in Table IV demonstrate that bleaching initially influenced the highly cross-linked amorphous IFAPs and disulfide-based IFKP–IFAP linkages rather than denaturing the crystalline structure of the IFKPs. However, dramatic changes in the crystalline and amorphous contents were noted at longer bleaching times (≥120 min). Hence, the progression of cystine to cysteic acid was steady, yet the crystalline IFKPs in hair remained resilient until approximately the 2 h bleaching mark. The wet DSC T D parameter correlates well with EDF results from FTIR imaging, FTIR-ATR, and EDF/EDF−1 from Raman spectroscopy cross-sectional work, where a decrease in T D is proportional (R2=0.95) to an increase or decrease in EDF or EDF−1, respectively (see Figures 12 and 15). Hence, as reported in the literature, T D directly relates to decreases in matrix cross-link density (7,8,13). Figure 15. Correlation of wet DSC denaturation temperature with Raman EDF−1 (band height ratio). Figure 16. Correlation of normalized denaturation temperature with the square root of bleaching time.

540 JOURNAL OF COSMETIC SCIENCE Regarding published associations between cystine degradation and mechanical testing, Robbins adapted data from Edman and Marti (5) to demonstrate that up to 45% of intrinsic cystine bonds are cleaved by aggressive hydrogen peroxide bleaching protocols. He concluded that decreases in the mechanical properties of peroxide-treated fibers were diffusion- controlled processes, where the calculated diffusion coefficient from tensile properties was assessed at 1.8 × 10−9 cm2min−1. Hence, since decreasing EDF−1 and T D are proportional to losses in matrix cross-linking, and EDF−1 changes proportionally with T D (Figure 15), by the transitive property of equality we expected T D to obey Fick’s law (1,24). Indeed, Figure 16 provides evidence of this correlation (R2=0.99), where the diffusion constant was estimated at 1.1 × 10−8 cm2min−1, which closely agrees with D determined from Raman scattering results (Figure 13) and magnitudes of D presented in the literature (1,3,24,25,26). Furthermore, Edman and Marti were among the first researchers to theorize that the hair fiber cortex, and more specifically, the conversion of disulfide bonds to cysteic acid in the cortex are responsible for the loss in wet tensile properties as a function of bleaching time (1,5). They used the 20% Index test to report oxidative damage results. In the 20% Index test, the ratio of work required to stretch the fiber 20% after treatment is compared to the work required to stretch the fiber 20% before treatment. By chance, we used similar bleaching conditions and treatment times as Edman, where only the 90 min data point is not included in Edman’s findings. Figure 17 contains the Edman 20% Index data and T D plotted as a function of the 1040/1080 cm−1 FTIR imaging band ratio. The data indicate that decreased tensile work correlates with increasing bleaching time (R2=0.95) and associated decreases in cortical cysteic acid. Further, with increased bleaching times, trends in T D (wet fibers) positively correlate with wet tensile work results, where diminished cystine is the common element. Raman scattering experiments yielded similar results, but with a positive slope, where Figure 12B demonstrates the complementary FTIR versus Raman correlation. By varying heating rates in HPDSC and assessing T D , the activation energy for denaturation of bleached fibers may be assessed. ASTM E698 describes the Flynn–Wall–Ozawa model- free method for evaluating the activation energy by measuring temperatures corresponding to fixed conversion values (α) from experiments performed at different DSC heating rates. By plotting ln(α) against 1/T, the slope gives -E a /R, where R is the molar gas constant, and Figure 17. Correlation of Edman 20% Index (R2=0.95) and TD (R2=0.99) data (from this work) with the FTIR imaging 1040/1080 cm−1 EDF results. Edman’s 20% Index and TD decrease proportionally with increasing bleaching time. Note that the magnitudes of bleaching time and cysteic acid increase with increasing 1040/1080 cm−1 ratio.