246 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS pressure and the kinetics of vapor sorption was monitored continuously until equilib- rium was achieved. At this point, desorption was started by replacing the organic solvent system with activated charcoal or Drierite. Both drying systems were found to be equally effective. Experiments were carried out at 23 and 32øC. Measurements at the higher temperature were carried out using the technique described in (6). Stratum comeurn was prepared according to the procedure suggested by Kligman and Christopher (13) and described in detail in (14). The treatment of the stratum comeurn in organic solvents was conducted as follows: a piece of guinea pig stratum comeurn was placed between 2 small pieces of saran gauze held in a specially constructed Teflon ©* frame. The frame was placed in a beaker containing the chloroform-methanol mixture (2:1), maintained at 40øC, for 30 rain with gentle agitation. This was followed by immersing the frame in a beaker containing distilled water maintained at the same temperature for another 30 min. The comeurn was then taken out and dried at room temperature before storing it in the refrigerator. Although, no quantitative estimate was made, it was observed that the above described treatment brought about a substantial decrease in the dry weight of the comeurn (- 30 to 50). The treated cor- neum was examined for its water vapor sorption and diffusion properties, which were compared to data on intact (untreated) guinea pig corneum, obtained earlier. In order to acquire some information on possible structural changes in the comeurn samples, a preliminary SEM examination was conducted. The sample preparation for the SEM was as follows. Pieces of the stratum comeurn were mounted on stubs using transfer tape. The stubs were then placed into a vacuum evaporator and coated with 5 nm of carbon followed by 30 nm of 80:20 gold/palladium from two angles. DISCUSSION Quantitative analysis of equilibrium sorption isotherms of water vapor in keratin has been described in detail (6). Briefly, a number of theories and equations were em- ployed, including the BET and D'Arcy-Watt equations, the Flory-Huggins polymer solution theory, and Zimm's clustering function. In general, the water vapor sorption isotherms on excised skin and hair were found to fir the D'Arcy-Watt eq. (16). ' A,B,(P/P0) + C(P/P0) + DE(P/P0) (1) W = • 1 + B,(P/Po) ! E(V/Po) i O -- where W is the weight of sorbate adsorbed by 1 g of sorbent Ai = mni/N is the number of primary sites of type i, multiplied by the molecular weight of sorbate and divided by Avogadro's number N B• is a constant which is a measure of the attraction of the sites for the sorbate •v- is the number of different types of sorption sites for primary adsorption described by a Langmuir isotherm C is a constant for the linear ap- *E. I. DuPont De Nemours Co., Wilmington, DE.

THE STRATUM CORNEUM 247 proximation to Langmuir adsorption on specific sites and D and E are constants describing secondary adsorption processes. P/P0 is the relative vapor pressure (relative humidity) of the sotbate. The results of the analysis indicate that in the low relative humidity range (0 to 30 per cent RH) sorption of water vapor occurs on reactive sites. Multilayer formation, ac- companied by extensive clustering, occurs at the high relative humidity end of the sorption isotherm. Clustering was evaluated according to Zimm's Eq. (17): 6^^ = _ ,B [0(a^/o^)] (2) v^ j PT where G^^ is the cluster integral for water molecules and V^ and a^ denote, respec- tively, the partial molar volume and the activity (relative humidity) of water. 0^ and 0B are the volume fractions of components water and keratin. Examination of the sorption/desorption isotherms showed that hysteresis is not a general phenomena in keratins, and was observed in a number of samples only. In general, the sotpriori isotherm was shown to exhibit the equilibrium properties re- quired for thermodynamic treatment. A background review of the mathematics of diffusion, and of the application of the principles developed by Crank and coworkers to the problem of water vapor diffusion in swelling keratins (for example, stratum corneum and human hair), has been given in (7). Determination of the diffusion coefficient is based on accurate measurements of the kinetics of vapor sorption or desorption in a small sheet of stratum-corneum ac- cording to the following equation: Mt M• .=0 (2n + 1) 2'•2 e -D(2n + l) 2•r2t/412 (3) where Mt denotes the total amount of diffusing substance which has entered the sheet at time t, and M• the corresponding quantity after infinite time. Equation (3) is the exact solution of Fick's basic diffusion equation, assuming a constant D, for the boundary conditions c=c0 0 (x (1 t=0 C = C• X = 0andX = 1 t ) 0 where Co is the initial concentration of water in the sheet when the surfaces (X = 0 and X = t0 are exposed to a constant concentration C• of vapor. The boundary conditions describe sorption or desorption depending on the values of Co and C•. -- A simple method based on eq. (3) for the determination of a mean value D (when D is not a constant) has been suggested by Crank (18). The value oft/12 for which Mt/M,:• =