ABSORPTION IN HUMAN SKIN 225 Table I Measured and Predicted Log Po•t and Kp (cm/hr) Values for Fragrance Chemicals M log P(a) C log P(b) M Kp(c) Chemical MW (PCMCP) (PCMCP) (P-G) log M Kp (P-G) 3,4-Xylenol 122.17 2.23 2.42 1.3 le-02 3-Cresol 108.14 1.96 1.97 1.03e-02 4-Chloro- 3-xylenol 156.62 3.39 3.48 5.39e-02 4-Cresol 108.14 1.94 1.97 9.95e-03 4-Ethylphenol 122.17 2.58 2.50 2.33e-02 4-Methoxy acetophenone 150.17 1.74 1.80 3.98e-03 4-Methoxybenzyl alcohol 138.17 1.10 1.02 1.65e-03 4-Nitrophenol 139.11 1.91 1.85 6.13e-03 Benzophenone 182.22 3.18 3.18 2.67e-02 Benzyl acetate 150.17 1.96 1.96 5.70e-03 Benzyl alcohol 108.14 1.10 1.10 2.52e-03 Benzyl benzoate 212.25 3.97 3.88 6.37e-02 Cinnamic acid 148.16 2.13 2.09 7.74e-03 Cinnamic alcohol 134.18 1.95 1.41 7.01e-03 Cinnamic aidehyde 132.15 1.90 1.90 6.65e-03 Coumarin 146.15 1.39 1.41 2.37e-03 Cumene 120.20 3.66 3.57 1.40e-01 Heptanoic acid 130.19 2.50 2.41 1.82e-02 Heptanol 116.21 2.72 2.41 3.18e-02 Hexanoic acid 116.16 1.92 1.88 8.60e-03 Hexanol 102.18 2.03 1.88 1.25e-02 Indole 117.15 2.14 2.13 1.22e-02 Isoquinoline 129.16 2.08 1.82 9.3 le-03 Methyl-4-OH benzoate 152.14 1.96 1.98 5.54e-03 Methylsalicylate 152.14 2.34 2.44 1.03e-02 Nitrobenzene 123.11 1.85 1.88 6.96e-03 Octanoic acid 144.17 2.94 3.05 3.08e-02 Octanol 130.23 3.00 2.94 4.13 e-02 Pentanoic acid 102.14 1.39 1.35 4.40e-03 Pentanol 88.15 1.56 1.35 7.08e-03 Phenylethyl alcohol 122.17 1.18 1.36 2.36e-02 Resorcinol 110.11 0.80 0.81 1.50e-03 Thymol 150.22 3.30 3.40 5.09e-02 - 1.88 - 1.99 - 1.27 - 2.00 - 1.63 - 2.40 -2.78 -2.21 - 1.57 -2.24 -2.60 - 1.20 -2.11 -2.15 -2.18 -2.62 -0.85 - 1.74 - 1.50 - 2.07 - 1.90 - 1.92 -2.03 -2.26 - 1.99 -2.16 -1.51 - 1.38 -2.36 -2.15 -2.63 -2.82 - 1.29 C Kp(d) log CKp Chemical (P-G) (P-G) Lit. Kp(e) log lit. Kp Ref. 3,4-Xylenol 1.79e-02 3-Cresol 1.04e-02 4-Chloro-3-xylenol 6.24e-02 4-Cresol 1.04e-02 4-Ethylphenol 2.04e-02 4-Methoxy acetophenone 4.38e-03 4-Methoxybenzyl alcohol 1.45e-03 4-Nitrophenol 5.56e-03 Benzophenone 2.67e-02 Benzyl acetate 5.70e-03 Benzyl alcohol 2.52e-03 Benzyl benzoate 5.50e-02 Cinnamic acid 7.25e-03 Cinnamic alcohol 2.90e-03 - 1.75 - 1.98 - 1.20 - 1.98 - 1.69 -2.36 -2.84 -2.26 - 1.57 -2.24 -2.60 - 1.26 -2.14 -2.54 3.60e-02 - 1.44 1.50e-02 - 1.82 5.30e-02 - 1.28 1.80e-02 - 1.74 3.50e-02 - 1.46 5.60e-03 -2.25 6.00e-03 - 2.22

226 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Table I (continued) C Kp(d) log CKp Chemical (P-G) (P-G) Lit. Kp(e) log lit. Kp Ref. Cinnamic aidehyde 6.65e-03 - 2.18 Coumarin 2.45e-03 - 2.61 9.10e-03 - 2.04 8 Cumene 1.2 le-01 - 0.92 Heptanoic acid 1.57 e-02 - 1.80 2.00e-02 - 1.70 7 Heptanol 1.92e-02 - 1.72 3.20e-02 - 1.49 7 Hexanoic acid 8.06e-03 - 2.09 1.40e-02 - 1.85 7 Hexanol 9.8 le-03 - 2.01 1.30e-02 - 1.89 9 Indole 1.20e-02 - 1.92 Isoquinoline 6.09e-03 - 2.22 1.70e-02 - 1.77 10 Methyl-4-OH benzoate 5.72e-03 - 2.24 9.12e-03 - 2.04 6 Methylsalicylate 1.2 le-02 - 1.92 Nitrobenzene 7• 3 le-03 - 2.14 Octanoic acid 3.68e-02 - 1.43 2.50e-02 - 1.60 7 Octanol 3.74e-02 - 1.43 5.20e-02 - 1.28 7 Pentanoic acid 4.13e-03 - 2.38 2.00e-03 - 2.70 7 Pentanol 5.02e-03 - 2.30 6.02e-03 - 2.22 9 Phenylethyl alcohol 3.16e-03 - 2.50 Resorcinol 1.53e-03 - 2.82 2.40e-04 - 3.62 6 Thymol 5.99e-02 - 1.22 5.30e-02 - 1.28 6 Physicochemical parameters for 33 fragrance chemicals used in the correlation of log Poet and Kp values, experimental Kp values, and respective sources. a) log of measured o/w partition coefficients Poet from PCMCP b) log of calculated Po•t from PCMCP c) permeation constants kP after Potts/Guy based on a) d) Kp after Potts/Guy based on b) e) experimental literature values. conditions, provides an excellent fit (r 2 = 0.86, p 0.001) (Figure 2). When calcu- lated log P values are used instead, the correlation is still highly satisfactory, with an r 2 = 0.83 (p 0.001) (figure not shown). A log/log correlation between predicted and experimental Kps for the same 20 fragrance compounds was plotted for comparison with the original Potts-Guy correlation obtained for Flynn's 93 chemicals (1) (figure not shown). This was done in order to ascertain a possible benefit achieved by bracketing the polarity boundaries more tightly (0.5 log Poet +4 versus -3 log Poet + 6). The resulting r 2 values are 0.72 (p 0.001) for the fragrance compounds, versus 0.67 seen by Potts and Guy for Flynn's set. DISCUSSION Mathematical models predictive of skin permeability mainly focus on the stratum corneum (SC) as the rate-limiting barrier against penetration, and imply that the SC lipids alone can account for the penetration data observed (1). This could certainly be confirmed by the good correlations obtained here between observed and predicted Kps in the polarity range of fragrance compounds. In earlier investigations it was shown that, at high partition coefficients, i.e., at high log P, tissue strata other than the SC assume rate control, as the viable epidermis contributes significant resistance to chemical pen-