UV DAMAGE ON GRAY HAIR 107 DYNAMIC CONTACT ANGLE The dynamic advancing contact angle at the interface between a hair fiber and de-ionized water was measured automatically using a Cahn DCA-315 dynamic contact angle ana- lyzer at room temperature. The average dynamic advancing contact angle value was obtained from measurements of ten fibers randomly collected from each hair tress. TRANSVERSE SWELLING The change in the central cross-sectional areas of hair fibers immersed in 0.1 N NaOH solution was automatically measured using the Mitutoyo LSM-5000 for five minutes at room temperature. The average percent increase in the cross-sectional area of ten fibers randomly collected from each hair tress was calculated and used as the average extent of transverse swelling. RESULTS AND DISCUSSION In the following section, the data will be presented to showcase the effects that a quaternized UV absorber, CATC, had on various properties of gray hair under UV irradiation. The performance of CATC is compared to the performance of a conventional neutral sunscreen agent, OMC. The substantivity of CATC or OMC on a hair surface, as delivered from a simple shampoo test system--aqueous 10% SLS solution containing 2.0% actives of UV filter--was determined for comparison. It was found that CATC was quite substantive on gray hair, its substantivity being 4.50 mg per 100 grams of hair. On the other hand, only a trace amount of OMC was detected on the hair surface treated with the same surfactant system. This significant difference in substantivity likely explains why CATC is superior as a UV protector on hair in comparison to the con- ventional UV filter, OMC, as delivered from the tested shampoo system. DIFFERENCES BETWEEN GRAY AND BLACK HAIR Four pairs of hair samples were obtained from four panelists, respectively, including three men and one woman. Each pair of hair samples consisted of 20 black and 20 gray hair fibers from one individual head. The fifth pair of hair samples consisted of 20 black and 20 gray hair fibers, which were randomly collected from a commercial 90% gray hair tress. Data on maximum center diameter, Dmax, minimum center diameter, Drain, center cross-sectional area, A, center ellipticity, E, strain-to-break, stress-to-break, and swelling are summarized in Table I. A t-test was performed for statistical analysis of the data using Microsoft Excel. Based on t-test results, no statistically significant differences were found between these black and gray hair fibers in terms of their center maximum diameter, center ellipticity, center cross-sectional area, stress-to-break, strain-to-break, and transverse swelling. The only difference observed was the center minimum diameter: the black fibers had slightly larger center minimum diameters than those of gray fibers.

108 JOURNAL OF COSMETIC SCIENCE Table I Physicochemical Properties of Gray and Black Hairs Statistical Hair type Gray Black difference D .... (mm) 0.0832 ñ 0.0145 0.0846 ñ 0.0122 No Drm n (mm) 0.0582 ñ 0.0064 0.0616 ñ 0.0062 Yes A (mm 2) 0.003835 ñ 0.000951 0.004124 ñ 0.000886 No E = D .... /D .... 1.433 ñ 0.218 1.377 ñ 0.169 No Stress-to-break (Gpa) (3.43 ñ 0.58) x 10 • (3.31 ñ 0.65) x 108 No Strain-to-break (%) 50.39 +_ 6.94 49.32 ñ 7.22 No Swelling (%), 5 min 11.78 ñ 3.04 12.43 ñ 2.31 No CHANGES IN HAIR COLOR The differences in color between the four gray hair tresses after 15 days of UV irradiation are very apparent by visual inspection. The hair tresses were put on the fiat glass of a Hewlett Packard ScanJet 6200C scanner and scanned. The computer images of the four hair tresses are presented in Figure 1. It can be seen that tress 2 retained its original color and had virtually the same color as the control, tress 1, but that tresses 3 and 4 developed a yellow tint compared with tress 1. The determined changes in light index, DL, in blue-yellow index, Db, in total color difference, DE, and in yellowing index, DYI, for each tress before and after UV irra- diation are presented in Figure 2. It was found that all of the changes in color index values of tress 2 were minimal, meaning that very little color change took place in tress 2. It was also observed that the light index values (L) for tresses 3 and 4 decreased (DL 0) after UV irradiation. This indicates that tresses 3 and 4 turned slightly darker than their original color. Their blue-yellow index values, b, and the corresponding yellowing index values, YI, of tresses 3 and 4 increased (Db and DYI 0) after UV irradiation. This means that tresses 3 and 4 turned yellower in comparison to their original color. All of these changes in color index values are consistent with our visual observations. The hair-yellowing effect suggests that a UV-induced degradation of hair was taking place. As noted by Roper and Finnimore, a light yellow residual color, or a newly formed yellow photochemical product, remains on hair following UV exposure (12). To compare the effect of UV irradiation on hair color changes of different types of hairs, we studied color changes in dark brown hair according to the procedures described above. Data on DL, Db, and DE of brown hair exposed to 20 days of UV irradiation are presented in Figure 3. Inspection of Figures 2 and 3 shows that the value of the total color difference (DE), 4.45, for the unprotected brown hair after 20 days of UV-B irradiation is smaller than the corresponding DE value of 7.42 for unprotected gray hair after 15 days of UV-B irradiation. This suggests that natural gray hair is more sensitive to color change than natural dark brown hair. It is also interesting to note that brown tresses 3 and 4 were somewhat lightened in color after UV exposure. Their light index values, L, were larger than their original values (DL 0). This suggests that melanin in unprotected brown hair underwent photobleaching during UV irradiation. From Figures 2 and 3, it is seen that both gray and brown hair tresses treated with CATC had small changes in hair