Genuine Champaca Flower Oil1).
The genuine champaca tree, Michelia Champaca, L. (family Magnoliaceae), which is characterized by its yellow flowers, occurs throughout tropical Asia, particularly in Java and the Philippines. However, it does not occur so abundantly that the perfume can be obtained on a large scale from the fragrant flowers. Attempts to prepare the oil by steam distillation failed, on the one hand because of the small yield, and on the other hand because the oil thus obtained resembles the flowers but little as to odor.
R.F.Bacon-), however, has pointed out that the maceration of the flowers with paraffin oil, in which the flowers are allowed to remain 24 hours and which is used nine times over for the extraction of fresh flowers, and subsequent extraction of the hydrocarbon oil with strong alcohol yields an oil with a fine and strong odor. Later the same investigator3) describes a product, obtained with a yield of 0,2 p. c. (the method of preparation is not given) which upon standing separates a considerable amount of crystals. The oil, separated by filtration, congealed to a semi-solid mass upon standing, due to the separation of an amorphous mass (presumably resinified oil) which upon the addition of 70 p.c. alcohol was isolated as a brown odorless substance. The filtrate was evaporated in a vacuum at 40° until a brown oil had separated which possessed a fine cham1) The oil of guaiac wood distilled from Bulnesia Sarmienti (which see), which enters commerce as champaca wood oil, has nothing in common with the genuine champaca oil. Report of Schimmel & Co. April 1893, 42.
2) Philippine Journ. of Sc. 4 (1909), A. 131.
3) Ibidem 5 (1910), 262.
paca odor. In this way oils were obtained that possessed the following properties: d30/30o0,9543 to 1,020; nD30o1,4550 to 1,4830; S.V. 160 to 180. The optical rotation could not be determined because of the dark color of the oil. It was soluble in 70 p.c. and stronger alcohol, the solution having a neutral reaction.
Upon saponification the oil of champaca flowers loses its characteristic odor almost completely. From 50 g. of oil 1,5 g. = 3 p. c. of phenols, principally isoeugenol (m. p. of benzoyl-derivative 103°) were obtained. In addition the oil contained 15 g. = 30 p.c. of acids (none of which boiled below 140° under 40 mm. pressure, indicating the absence of methylethylacetic acid) and 23 g. = 46 p. c. of neutral substances the odor of which resembles that of bay oil.
Oils likewise distilled in Manila by B. T. Brooks1) had the following properties: d30/30o0,904 to 0,9107; nD30o1,4640 to 1,4688; E.V. 124 to 146; E. V. after acetylation 199. They could not be fractionated in vacuum because they resinified. Brooks identified cineol (m. p. of the iodol compound 112 to 114°). Possibly p-cresol methylether is also present but it could not be identified. When the oil was shaken with sodium acid sulphite solution benzaldehyde could be isolated (m. p. of phenylhydra-zone 149 to 151°). Presumably another aldehyde is present, but it was not further investigated. Of alcohols Brooks found benzyl alcohol which he characterized by oxidation to benzaldehyde, also phenylethyl alcohol. In the alkaline saponification liquid he found benzoic acid. The benzaldehyde content of the oil amounts to about 6 p.c. Of benzoic acid it contains but 0,5 p.c. Brooks is of the opinion that this may have resulted from the oxidation of the benzaldehyde or benzyl alcohol.
In the oil of champaca flowers there occurs a crystalline substance, which melts at 165 to 166°, to which Bacon assigns the empirical formula C10H20O5. Contrary to the observations of Bacon, Brooks found that this substance reacts quantitatively with bisulphite solution. However, it cannot be regenerated from the bisulphite compound. It is a ketone the phenylhydra-zone of which melts at 161°, the semicarbazone at 205 to 206°.
1) Philippine Journ. of Sc. 6 (1911), A, 333. - Journ. Americ. chem. Soc. 33 (1911), 1763.
From Bacon's investigations, to which reference has been made above, it becomes apparent that formerly the so-called champaca oils of commerce1) were either distillates of a mixture of ylang-ylang blossoms and champaca blossoms or of a mixture of the genuine yellow blossoms and of the white blossoms of Michelia longifolia, the latter being used rather freely.
An oil examined for its constituents by Schimmel & Co.2) which, according to the manufacturer in Java, had been distilled mainly from the white flowers to which few of the yellow ones had been added, had the following properties3): d15o 0,8861; aD - 11° 10'; A.V. 10,0; E.V.21,6; E.V. after acetyla-tion 150,1; soluble in 2 vol. of 70 p. c. alcohol, the addition of 4 vol. and more, however, renders the solution decidedly turbid; soluble in 1 vol. and more of 80 p. c. alcohol, but the addition of more than 7 vol. produces opalescence due to the separation of paraffin. The oil was light brown in color and revealed a slight bluish fluorescence, particularly in alcoholic solution. It contained about 60 p. c. of \-linalool (m. p. of phenylurethane 65 to 66o) and small amounts of geraniol (m. p. of diphenyl-urethane 80 to 81°). Fraction 248 to 255° consisted of eugenol methyl ether as became evident from its oxidation to veratric acid (m. p. 179 to 180°; analysis of silver salt). Of terpenes mere traces only were present. The first fraction contained esters of methylethylacetic acid (b. p. of the free acid 176 to 177"; aD - 16° 40'), apparently as esters of methyl and ethyl alcohols. It is not improbable that the higher boiling alcohols, found in the oil (linalool and geraniol), are in part combined with methyl-ethylacetic acid. Furthermore, the presence of the acid in the free state was established, for when 250 g. of oil were shaken with 5 p.c. potassium hydroxide solution, methylethylacetic acid was obtained as well as a small amount of a phenol that was not further investigated.
1) Schimmel's Bericht April 1882, 7; Report of Schimmel & Co. April 1894, 59; October 1891, 15; April 1897, 11. -) Ibidem October 1907, 32. 3) Ibidem October 1906, 23.
Benzoic acid, which had previously been found in champaca oils, due presumably to the ylang ylang blossoms used in the distillation, could not be found in this oil.
According to Elze1), champaca oil also contains nerol (m. p. of the diphenyl urethane 50 to 50,5°). In what kind of oil this alcohol has been found cannot clearly be learned from the publication.