324 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS able exceptions are Harrold and Pethica (4) and Blank and Gould (5). However, both of these papers describe only long times (18 to 24 hr) and low concentrations. Another related study deserves to be cited, namely the work of Garrett (6) on the sorption of surfactants by hide powder. Only low concentrations were employed, but the funda- mental X/t dependence was shown'at short times. It is believed that the present study is the first to examine short times and high surfactant concentrations, corresponding in some degree to normal-use conditions. EXPERIMENTAL Undamaged and bleached hair samples were obtained from DeMeo Bros., New York City, and were used as received. The stratum corneum from neonatal rats was used as a model for human skin. The details of preparation of these membranes have been given in a previous publication (7). This stratum corneum has well developed barrier properties, at least in respect to the transmission of water vapor, as has been shown by Singer and coworkers (8) and confirmed by our own determinations (7). Sodium lauryl sulfate was obtained as a pure white crystalline powder from BDH Chemicals, Ltd., Poole, England. Tagged material was purchased from Amersham/Searle Corp., Arlington Hts., Illinois, in the form of small, individual ampoules. Each ampoule contained 2.47 mg of SLS with an activity of 110 microcuries. The "tag" is present as the S-35 isotope and is thus in the anion of the surfactant: Standapol ES-2 (Henkel Co.)--sodium lauryl ether sulfate with 2 mol of ethylene oxide. Standapol ES-3 (Henkel Co.)--sodium lauryl ether sulfate with 3 mol of ethylene oxide. Standapol 130E (Henkel Co.)--sodium lauryl ether sulfate with 12 mol of ethylene oxide. Tergitol 15-S-9 (Union Carbide Corp.)--the 9 mol ethoxylate of secondary C•a to C •5 alcohol. Solutions of desired concentration were made up of the nonradioactive powder and one ampoule of tagged material was added with stirring. Hair samples of about 100 mg each were placed in 20 ml of the solution for times which varied from a few minutes to 8 hr. They were then removed and rinsed twice for a few seconds to remove entrained solution. For stratum corneum the sample size was approximately 2 mg and the solu- tion was 10 ml. In either case the exposed substrate was dissolved with Unisol (Isolab Inc.) and Unisol-Complement was added. The resulting clear solution was counted by the scintillation method on a Packard 3255 Tri-Carb Spectrometer to determine the amount of SLS sorbed. Triplicate experiments were run and averaged for each experi- mental point on the figures. A GRAVIMETRIC PROCEDURE The availability of accurate sorption data from the radiotracer experiments with SLS described below furnished a benchmark from which a simpler gravimetric technique

SORPTION OF KERATINOUS SUBSTRATES 325 has been worked out for materials not available in tracer form. In studies of hair care products such as proteins, polymers and conditioners it is often of interest to de- termine the amount of material which is sorbed by the hair. The high cost of human hair and its difficulty in handling dictate that laboratory samples be of small size, that is of the order of 100 to 200 mg in weight. If the sorbed quantity is in the range 1 to 10% (often the case), then the incremental weight is ! to 20 mg, an amount that is easily de- tected with an analytical balance. For best results the hair should always be weighed directly, not in a container. A suitable configuration is achieved by winding the 100 mg strand around one's finger and stuffing it into a ! oz vial. Enough distilled water is added to wet the hair and the vial is left open overnight in a 50øC oven. By the next day, after drying, a well shaped curl has formed which retains its configuration and can easily be removed by forceps and placed on the balance pan. Hair is very hygroscopic, a fact which creates a problem in gravimetric work. Thoroughly dried hair will gain several percent of its weight in moisture when it is exposed to the atmosphere for only a few minutes. It is thus extremely important to make sure that the sample being weighed is always at the same reference state relative to water vapor. For example, this could be a "bone dry" condition such as is achieved by drying over P2Os or at high temperatures. However, drying by desiccants is time- consuming, while high-temperature drying tends to alter the hair protein irreversibly. On the other hand, simple equilibration for some time at ambient atmosphere is quite unsatisfactory, owing to changes of relative humidity. As a compromise, oven drying at 50øC was adopted. This is low enough so that no damage seems to occur to the hair. The amount of time required to reach equilibrium water content is of the order of 8 to ! 2 hr. Thus samples can conveniently be left overnight in the oven before weighing. (A 10 min period in a desiccator to cool the sample is advisable between the oven and balance.) The reproducibility achieved in this way has generally been very satisfactory. For example, two samples of bleached hair were conditioned in a 50øC oven for a day. Their weights were recorded as 104.8 and 123.4 mg. The samples were then placed in distilled water for three days, removed and dried in the oven overnight. The following day their weights were, respectively, ! 05.0 and 123.6 mg. For accuracy, the hair samples should be dried and weighed several times, both initially and after sorption, until constant values are obtained. In actual practice it was usually found that the weight after the first drying does not change appreciably upon further drying. It is always advisable to run several "controls" for each sorption experiment these are hair samples which are exposed to distilled water for the same time as the du- ration of the sorption. Normally the controls will return to their initial weight but in some cases systematic (i.e., more than one sample) deviations of 0.5 to ! mg can occur. This is possibly related to large changes of ambient laboratory humidity. Such devia- tions should be taken into account for the sorption samples. Some discussion about rinsing procedure is in order here. To a certain extent the method followed must be adapted to the material which is sorbed. Thus, for a cationic polymer, which is very tenaciously sorbed, the hair should be thoroughly rinsed several times. The polymer sorbed will not come off easily even with vigorous washings and more important, the viscous polymer solution which is easily entrapped in the hair must be removed or erratic results will be recorded. At the other end of the spectrum are substances like salts (see below for results on NaBr). In this case sorption is very weak and thorough rinsing completely removes the salt a different procedure must be