JOURNAL OF COSMETIC SCIENCE 432 The colorimetric data for each coupler, even if in different way, show that by increasing the pH of the medium, in the pH range of 8.5–9.0 a considerable color variation takes place. In order to achieve an estimate of the presumably different distributions of colored pigment formed by these reactions above and below this pH range, the reaction solutions of PPD both alone and with each coupler at pH 7.5 and 9.5 were analyzed by preparative thin-layer chromatography. In the absence of a coupler and at pH 7.5, the best conditions for the Bandrowski’s base formation (2), the TLC from PPD alone shows the presence of several colored bands. The main band contains an intense dark brown component (Rf = 0) that, according to the literature, is a high-molecular-weight material, and an intense dark blue band (Rf = 0.15) that the literature data assign to Bandrowski’s base (10). The L, a*, and b* values of the color formed on wool by PPD under the same conditions are 15.98, 1.24, and −4.04, respectively, and may be considered in agreement with the TLC data. The reaction between resorcinol (RES) and PPD, according to Corbett (6), gives a very unstable magenta hydroxyindoaniline 3 that may lead to a brown polymer (polymeric hydroxyindoaniline) and to a dark green molecule 5 that, furthermore, may be trans- formed to low-molecular-weight brownish intermediates (8). From the data reported in Table I, in the pH range of 7.5–8.5, the pigment formed on the wool shows a small L value, with values of a* and b* very near to zero (very dark brown color). For pH values 9.0 a strong increase of b* takes place, while the value of a* even if positive, is always small. The pigment formed in the range of pH 9.0–10.5 is a brown that moves to light brown. TLC of the system PPD-RES at pH 7.5 shows many other minor products having Rf 0.5, a strong dark brown component with Rf = 0, a trace of a narrow dark blue band (Bandrowski’s base), and a large moderate brown/light greenish band with Rf = 0.2–0.3. At pH 9.5 the brown/light greenish band is more intense, while the component with Rf = 0 has almost disappeared. The colorimetric data appear consistent with the Table I L, a*, b* of the Colors Due to the Reaction of p-Phenylendiamine with Different Couplers as a Function of pH Colorimetric Parameters pH 7.5 8 8.5 9 9.5 10 10.5 RES L 17.31 16.08 17.04 28.61 36.78 41.4 45.13 a* 1.24 1.03 1.43 3.64 3.49 2.52 2.12 b* 2.72 1.96 2.99 11.91 15.44 18.04 17.56 MeRES L 17.51 18.18 20.51 30.28 43.15 49.01 48.41 a* 5.21 5.5 6.8 9.24 8.98 8.8 9.1 b* 5.39 5.97 8.02 12.7 13.36 16.85 18.49 AP L 17.22 20.47 20.72 21.74 28.05 28.08 27.05 a* 1.24 1.22 1.32 2.36 5.39 7.77 8.48 b* 3.74 5.3 5.39 5.77 4.44 1.67 0.78 AHA L 24.77 23.09 19.63 18.39 13.36 13.8 15.86 a* 5.43 5.44 5.39 6.61 4.46 3.95 3.82 b* −11.65 −11.58 −11.36 −14.38 −9.13 −9.98 −11

REACTION OF PPD AND BENZENE DERIVATIVES 433 chro matographic data. Indeed, at lower pH, the recognized L, a*, and b* values agree with the strong prevalence of a dark brown pigment formed by reaction between PPD and RES or by PPD with itself. When the pH is increased, the colorimetric values show that the formation of the dark brown pigment falls in the pH range of 8.5–9.0. At pH 9.0 the increase in b* value is consistent with the prevalence of moderate brown components of probably low molecular weight, while the moderate increase in a* could be due to the presence of the dark green molecule 5 (in the CIELab space the green color is characterized by positive b* and negative a* values, while for the brown color a* and b* are always positive). As regards 2-methylresorcinol (MeRES), the pigment formed is always a brown color (from dark to medium brown) that is different from RES and is characterized by some reddish shade when the pH value is increased (see Table I). The TLC of the mixture of the reaction carried out at pH 7.5 shows a moderate-intensity dark brown band with Rf = 0, a dark blue small band with Rf identical to that of the Bandrowski’s base, and a moderate brown band followed by a large reddish one (Rf = 0.2–0.3). At pH 9.5 the component with Rf = 0 is at a low concentration and the brown/reddish band is very intense, indicat- ing that the formation of this pigment is favored by the higher pH. This reddish band could be due to the presence of the indoaniline 3. In the case of RES this species is not detectable and, according to Corbett, is highly unstable. After prolonged exposure to the daylight of the TLC plate, the reddish band turns to brown. In order to obtain further information about the features of the colored pigments derived from these two couplers, weathering experiments were performed as described in the Methods section. The only signifi cant data were those obtained from direct sunlight exposure of the colored wool samples. The pigments obtained at pH ≤ 8.5 from both couplers faded moderately after sunlight exposure, and the relative L, a*, and b* values are not reported. The data related to the weathering of the pigments obtained at pH ≥ 9.0 appear particularly interesting and are reported in Table II and III. It is evident that the brown colors from resorcinol fade moderately while the colors from MeRES fade with a progressive reduction of the redness converging to a brown color. These results suggest that the products deriving from RES and MeRES by sunlight may be similar. The colorimetric data related to the couple 3-aminophenol (AP) and PPD (see Table I) provide a dark brown color that moves to a dark red-purple at pH 9.5. Starting from pH 9.0 the value of b* rapidly decreases while a* increases. These data are in agree- ment with TLC results. In fact, TLC of the reaction performed at pH 7.5 shows a grey- brown pigment with Rf = 0, and a large brown band followed by a red-purple band. At pH 9.5 the compounds with Rf = 0 are identical but the red-purple band is much more intense. On exposure to daylight, the red-purple band turns to brown. According to Cor- bett (5), with ferricyanide the magenta aminoindoaniline 3 may be considered the only product, while with hydrogen peroxide the formation of the brown molecule 5 should be preferred. Our results, on the contrary, show that hydrogen peroxide also supports the formation of aminoindoaniline 3. The progressive lowering of the b* value when the pH is rising agrees with an increase of the red-purple pigment. Weathering experiments carried out on wool strips colored at pH 10.5 (see Table IV), the best approach to obtaining the compound 3, show that also without the direct sunlight, four days are enough to realize a strong variation from a dark red to a brown color. These results point out that the aminoindoaniline 3 is very unstable, leading, like RES and MeRES, to a brown color.