STRUCTURAL ANALYSIS OF PERMANENT WAVED HAIR BY SAXS 123 Permed hair. Bleached hairs were wrapped around a 15-mm-diameter spiral rod, and were treated for 15 min at 30°C using 6.0 wt% thioglycolic acid aq. and 0.8 wt% ammonia aq. (pH 9.4). After washing with water, the hairs were treated for 15 min at 30°C using 7.0 wt% sodium bromate aq. (pH 6.9) (⇒ permed hairs). Stretched hairs. The permed hairs were combed through fi ve times. This treatment was repeated 30 times. SLS-immersed hair. The permed hairs were immersed in 5.0 wt% SLS aq. (pH 8.5) for 3 min. This subsequent treatment was repeated 30 times. Stretched SLS-immersed hair. The permed hairs were immersed in 5.0 wt% SLS aq. (pH 8.5) for 3 min. After immersing the hairs in water for 2 min, they were combed through fi ve times. This subsequent treatment was repeated 30 times. SAXS MEASUREMENTS Asian untreated hairs, with a diameter of 70–90 μm, from a single head were used. In addition, naturally curly hairs were excluded and only straight hairs ware selected. Con- cerning naturally curly hair, it has been commonly believed that the biased distribution of the paracortex-like structure on the concave side and the orthocortex-like structure on the convex side was related to curl (13). On the other hand, it is known that straight hair possesses the homogenous distribution of two types of cortical cells (13). The hair samples (15 hairs per specimen) described in the following text were measured using a scanning microbeam SAXS (beamline BL40XU) at the large synchrotron radiation facility SPring-8 (Proposal No. 2012B1388). The temperature and humidity in the facility were 26°C and 30% RH, respectively. The measurement conditions were as follows: X-ray wavelength: 0.083 nm (X-ray energy: 15 keV), camera length: approx. 1612 mm, beam size diameter: approx. Five micrometre (1st pinhole: 5 μm, 2nd pinhole: 200 μm), beam stop diameter: 6 mm, exposure time: 3 s. An imaging intensifi er (4 inch) and CCD camera (ORCA-II-ER Hamamatsu Photonics Co., Ltd., Hamamatsu, Japan) were used as detec- tors. Calibration was carried out using silver behenate (d = 5.838 nm). X-rays were radiated perpendicularly to the fi ber axis. The sample position with respect to the X-ray microbeam was moved by steps of 5 μm from the fi ber periphery to the center of the fi ber. The measure- ment of the perm-treated hairs started from the convex side of the curl. We then extracted a scattering intensity profi le from the two-dimensional SAXS pattern, and simulated the equatorial intensity profi le around the scattering vector S = 0.118 nm-1 to obtain the dis- tance between IFs by using the method proposed by Briki et al. (14). Based on the scattering intensity profi le in an azimuthal direction around S = 0.118 nm-1, the full width at half maximum (FWHM) of the peak derived from IFs was obtained and used as an index of the IF orientation with respect to the hair longitudinal axis (expressed in degrees). The t-test was used to evaluate signifi cant differences in all the statistical analyses (bilateral distribution, α = 0.05, *p 0.05, **p 0.01, ***p 0.001). CHEMICAL TREATMENTS FOR SAXS MEASUREMENTS Bleached hair. Bleached hairs were prepared in the same manner as described previously. Permed hair. Permed hairs were prepared in the same manner as described previously.

JOURNAL OF COSMETIC SCIENCE 124 Stretched hairs. The permed hairs were repeatedly stretched 50 times with a force of 0.1 N using a SV-201N type tension/compression testing machine (IMADA SEISAKUSHO Co., Ltd., Toyohashi, Japan) to restore their original state. SLS immersed hair. The permed hairs were immersed in 5.0 wt% SLS aq. (pH 8.5) for 3 min, 30 times. Stretched SLS immersed hair. After immersing permed hairs in 5.0 wt% SLS aq. (pH 8.5), the hairs were repeatedly stretched 50 times with a force of 0.1 N using a SV-201N type tension/compression testing machine to restore their original state. RESULTS CURL DIAMETER EFFICIENCY As shown in Figure 1, the phenomenon of curl fallout was validated in the three types of hair samples. Among them, the stretched SLS immersed hair had the lowest value of curl diameter effi ciency (82.6%). There was a signifi cant difference between the permed hair and the stretched SLS-immersed hair (p = 0.007). The stretched hair (90.8%) or SLS-immersed hair (95.3%) also had lower values of curl diameter effi ciency than the permed hair (97.1%). SAXS MEASUREMENTS Figure 2 shows a typical 2-D profi le measured using SAXS. A strong scattering peak derived from the IFs was observed on the equatorial axis. To examine if the IF’s organization Figure 1. Rate of change of curl diameter effi ciency because of each treated hair. t-test was used for statistical analysis with *p 0.05, **p 0.01, ***p 0.001.