This is a beautiful, brief, 1910 International Sugar Journal article by Karl Micko who was the earliest user of the birectifier and the eight fraction method of analyzing spirits. Micko uses a delicate saponifying method to isolate the non-saponfiable (non ester) fraction of rum oil which is very likely a rose ketone like damascenone and orginally derived from carontene in the cane but probably also bound in glycosides during processing as molasses.
I am going to try and duplicate his experiment. The rum oil of Rafael Arroyo was called Bauer oil by others (as noted by Kervegant). During birectifier distillation this oil exists as an emulsion that can be quite significant if full flavored spirits across categories (tequila can be quite interesting). The bulk of this emulsion is saponifiable, while the other part is quite small but extremely odor active with pretty extreme threshold of perception. This can lead to confusion when interpreting GCMS data because the most interesting molecules will not have the highest peaks. Many GCMS examinations have used special olfactometry instruments but those do not appear to be in use at any distilleries. The birectifier may be the most practical tool to add olfactometry and examine what you are missing when you use ordinary GCMS analysis.
THE CHEMICAL EXAMINATION OF VARIOUS RUMS.
By KARL MICKO.
In a previous article (this Journal, 1909, 225, 410, and 446) we have dealt with the examination of rum, and have attributed the great difference between Jamaica and artificial rums to the presence of a typical aromatic constituent in the former. We have now extended our studies to other kinds of rum, especially Cuba and Demerara rums, and to different arracks. The special object in view was to determine whether Cuba and Demerara rums contain the same typical aromatic constituent, and to find out in what way these two kinds differ from one another.
Our previous work has shown (loc. cit.) that the typical aromatic constituents of Jamaica rum is more resistent towards alkali than the esters; for whilst the esters are completely saponified by an alcoholic N/10 caustic soda solution in 24 hours at ordinary room temperature the typical aromatic odour persists to finally give place to another more resinous smell.
It was inferred that by carefully saponifying the esters it would be possible to separate the peculiar aromatic constituent, and this was found to be actually so. First the ester content of a portion of distillate was determined using an excess of alkali, then to a second portion of distillate less alkali was added than sufficed for the saponification of the esters. The amount of caustic soda added was less than that actually necessary to effect saponification on account of the fact that the aldehydes readily decompose in alkaline solution forming acid products. In the case of this happening the odour would be affected; but our main object in working with a small amount of alkali as possible was to avoid attacking the non-ester constituent.
The distillate thus treated was alkaline, but the excess of alkali was slight as the esters were almost completely saponified. It was then acidified with tartaric acid, and submitted to fractional distillation.
Working in this way we examined not only the above-mentioned tropical spirits but also some European products; and our conclusions may be summarized as follows:—
All the spirits examined possess a peculiar aromatic constituent which does not belong to the esters. It is possible by means of the above-described method to readily differentiate between artificial spirits flavoured with esters, etherial oils, and other substances and the genuine product. This peculiar aromatic constituent is of great value in judging the purity of spirits, and is in this connection of greater significance than the eaters. It imparts most of the specific aroma to the spirit. It is a general criterion, and has not been imitated up to the present time. It can be separated by carefully saponifying the esters, and has an extremely delicate fruity odour.
The tropical spirits which were examined, Cuba, Demerara, and Jamaica rums, and Batavia arracks, contain in addition to other flavouring constituents the same, or a very similar aromatic constituent, as the one we obtained from Jamaica, rum (this Jl., 1910, 225, et seq.), but in Jamaica rum it is present in very much greater amount.
In Jamaica rum it can be detected by fractional distillation of the original rum, or even by strongly breaking down; whilst in the other spirits, since it occurs in much smaller amount and in admixture with other specific aromatic bodies and eaters, it can only be recognised by carefully saponifying the esters and fractionating the ester-free liquid thus obtained.
Besides the fragrant-smelling bodies another body of a characteristic odour which generally appeared in the last fractions of the fractional distillation was found in the tropical spirits.
As to the origin of the typical aromatic constituent, it may either be formed during fermentation, or may be present as such in the primary material. In the case of Jamaica rum the first supposition is probably true. Perhaps there are present in the sugar cane certain bodies which during fermentation give rise to the aromatic substances; or again possibly the aldehydes, ketones, &c., react during the production of the spirit with the formation of the aromatic bodies.
The results of our analytical examination of the various rums are given in the following table:—