INTERNAL LIPIDS IN HAIR HEALTH 353 Figur e 4. Oxidation of unsaturated lipids. Figure 5. LPO levels in single source ponytails before and after coloring treatment. signifi cant drop in cycles to break for hair, where the hair had lipids removed via extrac- tion (66,314 for alpha value versus 18,542, where alpha value is 63.2% of fi bers broken), demonstrating the importance of maintaining CMC integrity on hair strength. This importance of lipids on strength was also demonstrated in a publication by Camacho- Bragado et al. (9), where the authors suggest that breakage can occur via formation and propagation of fl aws at the cell membrane locations starting at the cuticle–cortex interface. This insight sets up the question as to whether lipids can be added back into hair and specifi cally into the CMC to build back lipid structures and thus improve hair strength. It is ideal to deliver lipids back into hair from every wash and thus from a daily-use product such as a shampoo. Shampoos contain high levels of anionic surfactant and are designed to remove lipid materials, i.e. sebum, so delivery of lipids from these products is challenging. A method of delivery has been developed, however, which uses a gel network structure of lipid and surfactant to deposit on hair and then on dilution release lipid into the CMC. There are several lipid classes that can be delivered by this ap- proach and one of these is FaOHs. A gel network is made by mixing a combination of C16 FaOH (cetyl alcohol) and C18 FaOH (stearyl alcohol) together with a surfactant, SLE1S and then this gel network is added into a standard shampoo chassis. FaOHs are preferred over fatty acids for formulation stability, as at shampoo pH fatty acids will deprotonate (pKa ~ 4.8) and destabilize the gel network structure. These two classes of lipids have very similar physiochemical properties (Table II), and C16 and C18 chain lengths were specifi cally chosen to best match internal lipid structures. SAXS and WAXS x-ray data confi rmed that an Lβ lamellar structure is formed with a d-spacing of approximately 90Å in a shampoo and that this gel network remains intact over time (10). Penetration of FaOH was measured in a similar way to measuring internal structural

JOURNAL OF COSMETIC SCIENCE 354 lipids an initial extraction with hexane-quantifi ed external FaOHs and a second ex- traction with chloroform:methanol–quantifi ed internal lipids. Figure 8 illustrates the importance of gel network to aid FaOH penetration comparing equal FaOH levels in a shampoo either added straight into the shampoo or as part of a gel network. Very low levels penetrated from the product with FaOH just added, but substantial penetration is measured from the gel network containing product. Table III shows fatigue data when hair treated with an ammonia/hydrogen peroxide oxidant system was washed 16 cycles in the same gel network shampoo versus an identical shampoo with no added gel network. The alpha value shown is obtained from the Weibell distribution function and is the characteristic lifetime when 63.2% of fi bers have broken. The beta value is the Weibell shape parameter and provides characterization of the distribution function. The 2.3% gel network containing product showed a signifi cantly higher alpha value than the control shampoo, i.e. more cycles are needed to break the hair after treatment with the gel network product. Weibell and Kaplan–Meier statistics using alpha values were used to determine signifi cance between products and using both methods show that the 2.3% GN product was signifi cantly different from the control and no gel net- work product. The log-rank prob ChiSq numbers were 0.0955 (90.4%) for 2.3% gel network versus no gel network and 0.0037 (99.6%) for 2.3% gel network versus control. Figure 6. Unsaturated fatty acid levels before and after coloring. Figure 7 . Change in LPO levels after UV exposure.