OIL FILMS AND MOISTURE ABSORPTION ON HAIR 137 RES UL TS AND DISCUSSION EFFECTS OF THIN OIL FILMS At each relative humidity stage, moisture regain was calculated from the following formula: Mv R=-x 100% MD (1) Here, Mv is the mass of moisture in the fibers at a given relative humidity stage, M D is the dry mass of the fibers, and R is percent moisture regain. Regain values were then plotted against relative humidity to give the sorption and desorption isotherms. The sorption-desorption isotherms for the untreated and oil-treated samples are displayed in Figure 1. For untreated and oil-treated hair a single run was performed. This was based on the high reproducibility of this measurement, which was established early in the validation study of this instrument, in which three replicates on a single sample gave essentially coincident sorption-desorption isotherms. Based on this observation and also considering the long time taken for a measurement (2.5 days), only one measurement was performed on each sample. The isotherms for the untreated hair behaved as expected, and a maximum regain of 27 .38% occurred at 95% relative humidity. All of the oil-treated hair samples showed lower levels of moisture sorption compared to the untreated sample, with the coconut oil sample having slightly higher regain than the other oil-treated samples at high relative humidity. The maximum regains at 95% RH for the coconut, sunflower, and mineral oil samples were 23.65%, 22.53%, and 22.05%, respectively. The difference between the untreated control and the oil-treated samples is statiscally significant. The differences between the oil-treated samples may not be sig- 30 .-------------------------------------, 25 20 10 5 0 10 --- Untreated -+-coconut --¼- mineral -e-sunflower 20 30 40 50 Target RH (%) 60 70 80 90 Figure 1. Water vapor isotherms at 25 ° C for untreated and oil-treated hair (thin oil coating). 100

138 JOURNAL OF COSMETIC SCIENCE nificant. Treating the hair samples with oil reduced moisture pickup however, a con­ siderable amount of moisture vapor was still found to penetrate into the hair fibers. The thin layer of oil on the surface may act as a barrier reducing the rate of penetration of water vapor. While this is the case for the mineral oil sample, there may be some penetration by the coconut and sunflower oil, which might hinder the absorption of water vapor. Several studies have indeed concluded that due to the long-chain nonpolar hydrocarbon structure, mineral oil does not penetrate into the hair fiber, leaving a layer on the fiber surface, even when exposed to heat (1-3). On the other hand, coconut and sunflower oils, which are polar and have affinity toward keratin, do penetrate into the cortex of the fiber. However, the amounts are expected to be too small to have a large effect on moisture vapor sorption. The diffusion rates for moisture into and out of the fiber at each relative humidity were calculated for all of the samples from the sorption data of Figure 1. These calculations are based on the solutions of Ficks's diffusion equation applied to cylindrical geometry. A simplified version of this solution is given in equation 2: C/C e9 = 4(D thr r2) 112 (2) where Cr is the concentration of the diffusant at time t, Ce 9 is the concentration at equilibrium, D is the diffusion coefficient, r is the fiber radius, and t is time. In a typical DVS sorption-desorption experiment, the humidity and the moisture regain data are obtained in the form shown in Figure 2. Each step in moisture regain is equivalent to a diffusion experiment. Therefore, the numerical data from the sorption (or desorption) experiment can be converted into a plot of (C/Ce 9 ) Vs (t/r2 ) 112 • The initial - � en en :i: C: ·- CL C, C: cu 0 130 125 120 115 110 105 100 0 1000 2000 3000 Time (min.) 100 90 -Am(%) -RH(%) 80 70 60 .-. 50 ::c 40 0:: 30 20 10 0 4000 5000 6000 Figure 2. Target humidities and moisture regains as a function of time in a typical dynamic vapor sorption experiment.