The reaction between alcohol and acetic acid anhydride is quantitative in the case of borneol, isoborneol, geraniol,1) menthol2) and santalol, hence enables an accurate determination of these substances. Less fortunate are the conditions with the tertiary alcohols linalool and terpineol, as these, on boiling with acetic acid anhydride, are partly decomposed by splitting off water with the formation of terpenes. Comparative figures, however, can also be obtained with these alcohols if the same amount of acetic acid anhydride is always used and the boiling continued for the same length of time. With linalool there was found as a favorable result after 2 hours of boiling, a deficiency of 15 percent of alcohol.1) Terpineol behaves toward acetic acid anhydride as follows:

1) Bertram and Gildemeister, Journ. f. prakt. Chem. II. 49 (1894), 189. 2) Power and Kleber, Pharm. Rundsch. (New York) 12 (1894), 162; Arch, der Pharm. 232 (1894), 653.

Time of boiling:

Terpinyl acetate formed:

10 minutes

51,2 percent

30 „

75,5 „

45 „

84,4 „

2 hours

77,9 „ .

With terpineol, therefore, heating longer than 45 minutes has a detrimental effect.

Better results are obtained if the suggestion of Boulez2) is followed and a diluent in the ratio of 1:5 is employed. For this purpose either turpentine oil or xylene may be used. Turpentine oil, however, yields a small acetylation number for which a correction should be made, whereas xylene remains totally unchanged. At the same time the period of acetylation must be prolonged. Schimmel & Co.3) report that the maximum of ester determination is reached after boiling 5 to 7 hours. Longer boiling again reduces the amount of ester formed as shown by the following table.

Even with this modification, satisfactory results are obtained only with terpineol, whereas the accuracy of the results obtained with linalool leaves much to be desired. The quantitative acetylation reported to have been observed by Boulez depended on wrong computation as pointed out by Schimmel & Co.3)

I. 20 parts linalool + 80 parts xylene.

Duration of acetylation........

3 hrs.

5 hrs.

7 hrs.

12 hrs.

20 hrs.

Ester v

53,3

60,4

63,0

63,3

51,6

°/o C10H18hO...............

15,3

17,4

18,2

18,3

14,8

Found with reference to 100 parts of the alcohol............

76,5

87,0

91,0

91,5

73,8

1) Report of Schimmel & Co. April 1893, 45.

2) Corps gras industriels 33 (1907), 178; Bull. Soc. chim. IV. 1 (1907), 117.

3) Report of Schimmel & Co. April 1907, 121.

II. 20 parts terpineol +80 parts xylene.

Duration of acetylation........

3 hrs.

4 hrs.

5 hrs.

7 hrs.

12 hrs.

Ester v

55,2

67,7

68,8

68,1

65,1

°/o C10H18hO.........

15,8

19,6

20,0

19,8

18,8

Found with reference to 100 parts of the alcohol.............

94,6

98,0

99,8

98,8

94,0

As in the case of the saponification, the presence of phenols and aldehydes exerts an unfavorable influence on the acetylation. If larger amounts of these substances are present they must be removed before the alcohol content can be determined. Citronellal, however, constitutes an exception. During the process of acetylation this is converted quantitatively into isopulegyl acetate. Hence it can be assayed like an alcohol.1) In this case, however, two hours are necessary for the acetylation as well as for the saponification. It is also necessary to add the right amount of sodium acetate, namely about 2 gr. for every 10 cc. of oil and acetic acid anhydride. Unless these conditions are observed, incorrect values are obtained as becomes apparent from the experiments made in the laboratory of Schimmel & Co.2)

Grams of sodium acetate for every 10 cc.

of oil and acetic acid anhydride

lg.

2g.

2,5 g.

Duration of acetylation

1 hr.

2 hrs.

1 hr.

2 hrs.

1 hr.

2 hrs.

Citronellal pure

Duration of saponification

lhr.

79,7%

94,2%

82,1%

96,9%

91,0%

103,8%

2 hrs.

77,3°/o

94,8°/o

82,4°/o

99,8°/o

95,9%

104,4%

Mixture of Citronellal . . 40,0 Geraniol . . . 40,0 Pinene .... 20,0

Duration of saponification

lhr.

70,6%

76,2%

73,7%

78,8%

74,0%

79,0%

2 hrs.

70,8 %

76,9%

75,0%

80,3%

74,6%

79,9%

1) Report of Schimmel & Co. October 1896, 43. According to more recent investigations by Semmler it would seem probable that all aldehydes, the CHO group of which is next to a CH or CH2 group, react with acetic acid anhydride with the formation of mono esters. The aldehydes apparently change to the corresponding enol forms. For several aldehydes of this class Semmler has supplied the experimental proof. In the case of the change from citronellal to isopulegyl acetate, the closing of the cycle is preceeded, according to Semmler, by the formation of an enol acetate. Berl. Berichte 42 (1909), 584, 963, 1161, 2014.

2) Comp. also Report of Schimmel & Co. April 1910, 156.

Verley and Bolsing1) have tried to simplify the acetylation process. They esterify the oil with a definite amount of acetic acid anhydride in the presence of pyridine, titrate the acetic acid not combined with the alcohol, and compute the alcohol content. The method has not proven satisfactory\vith the terpene alcohols, but yields better results with the phenols. Hence it is applied, if at all, in connection with the latter (comp. p. 593).

In like manner, the attempts to apply Schryver's method for the determination of phenols (p. 592) to alcohols has proven a failure.-)