J. Soc. Cosmetic Chemists, 19,395-410 (May 27, 1968) Mechanical Hysteresis Modified Hair of Chemically DONALD E. DEEM, B.S., and MARTIN M. RIEGER, Ph.D.* Synopsis--A new parameter for the mechanical testing of hair fibers is described. The dimensionless "hysteresis ratio" is defined as the ratio between the work of unloading and loading the fiber. It is shown that this ratio and its temperature dependence are useful for detecting chemical modifications of fibers, such as introduction of bulky groups, deamina- tion, crosslinking, or reduction. An attempt is made to interpret the meaning of this param- eter and of the shape of the load extension curves on the basis of the generally accepted structure of a-keratin. INTRODUCTION A variety of methods for the determination of damage to keratin fibers, especially wool, has been described in the literature (1-4). Ever since Speakman (5) reported the existence of three distinct regions in the stress-strain curve of wool fibers, much work has been published relating the mechanical properties of wool to damage. Mechanical parameters of wool and hair have been examined not only in tension (6) but also in bending (7) and torsional (8) modes. Still more recently Feughelman and Watt (9) performed torsional measurements on chemically modified wool. Wool or hair fibers which had been bleached (10), waved (11), hydrolyzed (12), reduced and alkylated (12), cross-linked (13, 14), and ninhydrin-reacted (15) have been subjected to different mechanical and chemical testing procedures. The classical method (6) for studying the mechanical properties of hair in tension is the determination of the 20 or 30% index. In this pro- cedure the work necessary to extend the fiber 20 or 30% of its original length before and after chemical modifications is determined. The * Warner-Lambert Research Institute, 170 Tabor Road, Morris Plains, N.J. 07950. 895

396 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS index is calculated as the ratio of the work after modification to work before modification. Experimental procedures for this type of study have been facilitated by the availability of recording tensile testers, since the work is proportional to the area under the recorded stress- strain curve. The extensions are carried out on the same fiber before and after modification, and consequently the average deviations are normally within + 5%. When single hairs are handled this way, the procedures used in modify- ing the fibers do not approach normal cosmetic procedures. Fibers must be soaked in chemical solutions, and the ratio of hair to, e.g., wav- ing lotion, is much smaller in the case of single hair treatment than in normal hair waving. A technique which more closely approaches actual practice, i.e., use of hair bundles for chemically modifying the fibers, requires that load-extension curves be determined on one set of un- treated fibers and that a final run be made using a different assembly of modified hairs. This method raises the average deviations to about + 15 to 30% on samples of 30 hairs. This deviation is due, in part, to great differences found in the cross section of neighboring human hair, which cannot be grown to the uniformity of some wools. A second difficulty with the 20 or 30% index is the fact that different chemical modifications of the fiber may yield stress-strain curves having the same shape and comparable areas, even though the structural changes within the fiber are quite different. An ideal parameter for measuring mechanical changes in keratin fibers would be one which (a) has low average deviation regardless of the absolute magnitudes of the stress-strain curves, (b) is sensitive to damage to the hair, and (c) gives an indication of the structural changes taking place in the hair. It is the purpose of this study to determine whether a new parameter, the ratio between the work of unloading and loading of hair fibers, is a meaningful measure of hair damage or chemical modifications. This ratio will be called the hysteresis ratio. Hysteresis of keratin fibers un- der conditions of loading and unloading has been mentioned in several publications (16, 17) but has apparently not been extensively studied. In addition, a hysteresis ratio has been used in the rubber industry to predict breaking energy (18). Experiments in this laboratory have shown that the hysteresis ratio is the same for fibers from the same lot and is not affected by the degree of elongation, at least between 20 and 30%. The experimental pro- cedure is simple, requiring neither the exact length nor the cross-sec-