EVAPORATION AND THE ODOR QUALITY OF PERFUMES 173 Question: Which smells more of benzyl acetate ? Evaporalion Time, minutes Results 1-6 100% T 34-40 80% T 100-105 100% U 170-180 100% U Ioo•'l" ,% $o \ toO \ \\ \ \ ". \\ \ \ o,lr (01"}•/l•'l.,•N OF THE' FlYATIVE A(-.TIOH OF B[NZY/_ •ENZOATE AN/) oF TglJ)FC'fL ALCOHOL ON R NIXTORE' OF' '•EId7. YL ACETATE' AND HEPTT'L ,4'LC0/"/0 L MIN. Figure 2.--... = % of panel indicating H smells most strongly of benzyl acetate (2a) ..... % of panel indicating T smells most strongly of benzyl acetate (2b). We obtained results of the same type when using linalool and ethyl benzoate as the odorants and nerolidol and diethyl phthalate as the fixa- tives (Exp. 3) and again using methyl nonyl ketone and p-methyl ace- tophenone as odorants and mineral oil and diethyl phthalate as fixatives (Exp. 4) also using benzyl acetate and nonyl acetate as odorants and diethyl phthalate and mineral oil as fixatives (Exp. 5). Solution 0 Solution P 4 g. Linalool 4 g. Linalool 2 g. Ethyl benzoate 2 g. Ethyl benzoate 16 g. Diethyl phthalate 16 g. Nerolidol Question: Which smells more of ethyl benzoate? Evaporation Time, minutes Results 3-8 86% P 3040 67% P 60-70 64% P 120-130 57% P 200-210 93% O Solution D 2 g. Methyl nonyl ketone 1 g. p-Methyl acetophenone 7 g. Diethyl phthalate Solution E 2 g. Methyl nonyl ketone 1 g. p-Methyl acetophenone 7 g. Mineral oil Question: Which smells more like methyl nonyl ketone ?

174 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Evaporation Time, minutes Results 3-10 84% D 3O-4O 67% D 90-100 50% D 240-250 75% E 300-310 100% E Solution L Solution M 2 g. Benzyl acetate 2 g. Benzyl acetate 2 g. n-Nonyl acetate 2 g. n-Nonyl acetate 6 g. Mineral oil 6 g. Diethyl phthalate Question: Which smells more like benzyl acetate ? Evaporation Time, minutes 0-15 30-40 10O-110 250-260 Results 79% L 67% L 84% M 75% M Loss of this clear-cut changeover pattern occurred (a) when the concen- tration of fixative was too low, (b) when there was a distinct difference in lasting power of the two odorants on the blotter, (c) when the two odorants were too similar in character or (d) when the chemical correspondence be- tween odorants and fixatives was not sufficiently close. G^s C•oM^ToG•^P•¾ The foregoing experiments were carried out some time ago. Recently an opportunity presented itself to check our results in such a manner as to eliminate the subjective element of odor evaluations. The acquisition of a Pye Argon Chromatograph enabled us to determine the change in vapor composition of a mixture of odorants without having recourse to the human nose. Collecting 4-10 ml. of the vapor over the mixture and passing it through the instrument, we obtained curves from which we could estimate the relative concentrations of the volatile components in the vapor. We first prepared chromatograms of both the liquid and the vapor phase of a solution containing equal weights of heptyl alcohol and benzyl acetate (Fig. 3a and b). Then we injected into the instrument the vapors of solution T (Fig. 3c) and U (Fig. 3d). The results clearly confirm our earlier findings, viz.: that the addition of tridecyl alcohol selectively decreases the vapor pressure of heptyl al- cohol (it may also decrease it for benzyl acetate, but to a less extent), while benzyl benzoate is a more effective fixative for benzyl acetate than it is for heptyl alcohol. BLENDING The foregoing experiments have indicated that the rate of evaporation of perfume materials is influenced by electrostatic interactions with fixative molecules. The same type of interaction should take place also with other