J. Cosmet. Sci., 51,289-302 (September/October 2000) Measurement of hair luster by diffuse reflectance spectrophotometry CARLA SCANAVEZ, MARCO ZOEGA, ANDR]•IA BARBOSA, and IN]•S JOEKES, Instituto de Qu/mica, Universida& Estadual de Campinas, UNICAMP, Caixa Postal 6154, CEP 13083-970, Campinas-SP, Brazil. Accepted for publication August 31, 2000. Synopsis We studied the application of the diffuse reflectance spectrophotometry technique in order to obtain consistent hair-luster measurements. The influence on data quality of such experimental conditions as sample color, texture, geometry, and position, and of such instrumental operating conditions as viewing angle, viewing aperture, and inclusion or exclusion of the specular component, were established. The color- difference parameter (DE) appears as the best parameter to measure hair luster in CIELAB and FMCII systems of equations. Hair tresses submitted to several treatments, including shampooing and conditioning, were used to choose the best reference (zero-luster value) and hair-luster saturation (maximum-luster value). A luster scale was assembled from DE values, pointing out that diffuse reflectance spectrophotometry data allows measuring and quantifying hair luster. INTRODUCTION Luster is an essential quality for hair beauty. However, it is difficult to measure it by physical methods. The literature describes a few light-scattering measurements using goniophotometry (1-5), but usually visual evaluations are applied for describing hair luster. In fact, luster or gloss is an optical phenomenon that results from the specular reflection of light (referring to the mirrorlike reflection) from a smooth surface. As the surface becomes rougher, the luster is reduced and the diffuse reflection from the surface increases. A completely nonlustrous rough surface is a diffuse reflector (6,7). Therefore, diffuse reflectance spectrophotometry should be able to measure luster. This is a well known method for color measurements in opaque substances and surfaces, widely ap- plied in the paint and paper industry (6) however, it has not been applied to luster measurements. We have investigated the capacity of simple diffuse reflectance equipment to measure hair luster. From the spectra, an ordinary software calculated color parameters as defined by two color systems, CIELAB (Commission International on Illumination L*, a*, b*) and FMCII (Friele-MacAdam-Chickering), both based on just-perceptible differences of color. Lightness difference (DL), chromaticity difference (DC), and color difference 289

290 JOURNAL OF COSMETIC SCIENCE (DE) were used in order to characterize the hair color/luster alterations after cosmetic and other treatments. EXPERIMENTAL MATERIALS Hair samples. Several tresses of virgin dark-brown and black hair were obtained from Ind. Com. de Cosm6ticos Natura Ltda, Itapecerica da Serra/SP, Brazil. Hair samples were cleaned by 8 h extraction in ethyl ether in a soxhlet and rinsed in distilled water prior to use. The tresses were oriented from root to tip end, and tied near the root end, combed, and stored in a desiccator. The tresses were kept in the instrument room 24 h before the measurements in order to attain humidity equilibrium. Table I shows some features of the tresses used. Cosmetic products and formulations. The following commercial products and formulations were used in this work: Rexona © soap, Tabu © commercial brilliantine, Revlon Aqua- marine © shampoo for dry hair, Neutrox © conditioner, L6real Els•ve Multivitaminas © shampoo, and standard and PQ7 formulations. The manufacturers and composition of these products are shown in Appendix A. Instrument. The diffuse reflectance measurements were made using a diffuse reflectance spectrophotometer, Macbeth Color-Eye 2020. The diffuse reflectance spectrophotometer viewing conditions involve an integrating sphere, a hollow metal sphere inches in diameter and painted white inside. An integrating sphere collects all light reflected from the surface of a sample placed against an opening in its side. Provision is usually made for including and excluding that part of the light reflected in a specular direction from a sample. The measurements were done while keeping the same tress region and turning the hair tress sample in the instrument sample holder. The instrument operation con- ditions were (a) configuration CRIIS (C: ceramic calibration, R: reflectance, I: ultraviolet waves, I: specular component, S: short viewing aperture) and (b) D65 illuminant. Spectra rendered values of L* (lightness of the color), a* (redness if positive coordinate or greenness if negative coordinate), b* (yellowness if positive coordinate or blueness if negative coordinate) from the CIELAB system and X (coordinate x), Y (coordinate y), Z (coordinate z) from the FMCII system. From these, the calculated parameters were DL*, DL (lightness difference), DC*, DC (chromaticity difference), and DE*, DE (color difference), for the CIELAB and FMCII systems of equations, respectively. Table I Virgin Hair Sample Characteristics as Visually Observed/Tress Sample Denomination and Correlated Experiment Hair characteristics Tress sample Experiment Less damaged dark-brown hair (25 cm) More damaged dark-brown hair (20 cm) Less damaged black hair (15 cm) D-B1, D-B2, D-B3 DD-B1, DD-B2, DD-B3 B1, B2, B3, B4, B5 Experimental optimization, color-parameters evaluation Cumulative treatments: standard and PQ7 formulations Internal and external references, statistical analyses