THERMAL TREATMENTS WITH A CURLING IRON 15 chosen to simulate various tensions that may be applied to hair fibers during styling with a curling iron. After curling for ten minutes and cooling, the hair fiber was gently uncurled and the hair/metal contact zone was examined. ( 3) Repeated, cyclical short-term curling/cooling of wet hair. For this experiment, individual hair fibers were immersed in water for 15 minutes, blotted between paper towels, and then subjected to 20, 50, and 100 cycles of curling with the curling iron. Each cycle consisted of ten seconds of thermal treatment with the hot curling iron and 30 seconds of immersion in water, followed by blotting. The fibers were held at the root ends, and the middle/tip sections were curled, simulating the way the hair (attached to the head) would be curled during normal styling practices. After the respective number of cycles of wetting and thermal exposure, the fibers were allowed to cool. Each hair fiber was gently uncurled, and the fiber/metal (f/m) contact zone was examined for the nature of cuticular damage. ( 4) Prolonged, high-load curling of wet hair. Individual hair fibers were shaped into large loops by raping together the root and tip ends. The looped fibers were immersed in lukewarm water for 15 minutes, blotted between paper rowels, and strung over the heated curling iron for ten minutes under loads of O g, 10 g, 20 g, and 30 g. After cooling, each hair fiber was gently uncurled and the cuticula in the heated zone were examined for damage. FIBER SURF ACE Ex.AMINA TION The f/m contact zone was examined by mounting the inside of the curl "face-up" on double-sided tape and coating with 100 A of platinum. We used a Hitachi S-4500 digital cold field emission scanning electron microscope (FESEM) for this study. RES UL TS AND DISCUSSION For ease of comparison, the typical topography of untreated hair fibers without thermal treatment is shown in Figure 1. The untreated controls show good differentiation of the Figure I. (a, b) Typical topography of untreated hair.

16 JOURNAL OF COSMETIC SCIENCE cuticle cell. There may be some debris on the scale faces and jagged scale edges, which is normal and indicative of chipping damage from standard grooming practices. Twenty hair fibers were prepared for each individual experimental category tested. At least ten of these were examined in the SEM. Damage phenomena or special features observed in more than 7 5 % of the hair specimens of a specific category are described as "typical" or "representative" in this discussion. EFFECTS OF REPEATED, CYCLICAL CURLING/COOLING OF DRY HAIR It should be noted that these fibers had been treated individually in a parallel arrange­ ment. Individually treated fibers will be much more affected by the hot metal surface than a hair fiber assembly, which tends to absorb and distribute the heat and moderate temperature within the hair strand. Hair fibers exposed to 50, 100, 150, and 200 sequential ten-second-long thermal treatments with the curling iron showed an increase in various types of surface damage as a function of progressive thermal exposure. Even though the level of damage increases with the increasing number of cycles of heating/cooling, the same types of damage phenomena are inherent in all exposed fibers. The thermally-induced damage phenom­ ena are shown in Figure 2. There is increased matting and compacting of the surface cuticle cell. While radial cuticular cracking is still observed, fine axial cracking of the exposed surface cuticle cell has become a frequent damage phenomenon (Figure 2a-d). Increased levels of fusion of the scale edges are displayed (Figure 2e, f). After 200 thermal curling treatments, fusion of the scale edges has become especially severe, as can be seen in Figure 2f. In extreme cases of fusion, pores or holes are frequently seen in the fused scale edges. These pores may possibly be areas where moisture and heat escaped via the CMC and endocuticle during the dry thermal treatment (Figure 2£). PROLONGED, HIGH-LOAD CURLING OF DRY HAIR Our SEM observations revealed that prolonged (ten-minute) curling of the "dry" hair fibers under O g, 10 g, 20 g, and 30 g of tension resulted in the following types of damage: (a) cuticular compression and moderate disintegration of the cuticula at the fiber/metal interface, (b) fusion at the scale edges, and (c) moderate-to-severe levels of radial cracking of the cuticula. We also showed that the damage became more severe with increasing load during curling. Topographical examination of the side of the curl of hair fibers that had been in direct contact with the heated metal of the curling iron for ten minutes while freely suspended without tension (0-g load) produced several interesting results that are indicative of moderate levels of cuticular modification: (a) While the inside of some curls showed rather well-maintained surface cuticles and good cell differentiation, other curls dis­ played matting and compacting of the surface cuticle cell. (b) Most fibers showed various levels of radial cracking of the cuticula during gentle uncurling of the hair fiber in the dry state (Figure 3). The above observations seem to suggest that the contact time of ten minutes for