CHOCHOCH2(OH)CH(OH)
│││║
HCOHHOCHCOCOH
││││
HOCHHOCHHOCHHOCH
││││
HCOHHCOHHCOHHCOH
││││
HCOHHCOHHCOHHCOH
││││
CH2(OH) CH2(OH) CH2(OH) CH2(OH)
GlucoseMannoseFructoseEnolic
form

This enolic form is capable of giving rise to all three hexoses, and the change by which the enolic form is produced and converted into an equilibrium mixture of the three corresponding hexoses is catalytically accelerated by alkalis, or rather by hydroxyl ions. In neutral solution the change is so slow that it has never been experimentally observed; in the presence of decinormal caustic soda solution at 70° the conversion is complete in three hours. Precisely similar effects are produced with galactose, which yields an equilibrium mixture containing talose and tagatose, sugars which appear not to be fermentable.

The continued action even of dilute alkaline solutions carries the change much further and brings about a complex decomposition which is much more rapidly effected by more concentrated alkalis and at higher temperatures. This change has been the subject of very numerous investigations [for an account of these see E. v. Lippmann, [1904], pp. 328, 713, 835], but for the present purpose the results recently obtained by Meisenheimer [[1908]] may be quoted as typical. Using normal solutions of caustic soda and concentrations of from 2 to 5 grams of hexose per 100 c.c., it was found that at air temperature in 27 to 139 days from 30 to 54 per cent. of the hexose was converted into inactive lactic acid, C3H6O3, from 0·5 to 2 per cent. into formic acid, CH2O2, and about 40 per cent. into a complex mixture of hydroxy-acids, containing six and four carbon atoms in the molecule. Usually only about 74 to 90 per cent. of the sugar which had disappeared was accounted for, but in one case the products amounted to 97 per cent. of the sugar. About 1 per cent. of the sugar was probably converted into alcohol and carbon dioxide. No glycollic acid, oxalic acid, glycol, or glycerol was produced.

The fact that alcohol is actually formed by the action of alkalis on sugar was established by Buchner and Meisenheimer [[1905]], who obtained small quantities of alcohol (1·8 to 2·8 grams from 3 kilos. of cane sugar) by acting on cane sugar with boiling concentrated caustic soda [p098] solution. It is evident that under these conditions an extremely complex series of reactions occurs, but the formation of alcohol and carbon dioxide and of a large proportion of lactic acid deserves more particular attention.

The direct formation of alcohol from sugar by the action of alkalis appears first to have been observed by Duclaux [[1886]], who exposed a solution of glucose and caustic potash to sunlight and obtained both alcohol and carbon dioxide. As much as 2·6 per cent. of the sugar was converted into alcohol in a similar experiment made by Buchner and Meisenheimer [[1904]]. When the weaker alkalis, lime water or baryta water, were employed instead of caustic potash, however, no alcohol was formed, but 50 per cent. of the sugar was converted into inactive lactic acid [Duclaux, [1893], [1896]]. Duclaux therefore regarded the alcohol and carbon dioxide as secondary products of the action of a comparatively strong alkali on preformed lactic acid. Ethyl alcohol can, in fact, be produced from lactic acid both by the action of bacteria [Fitz, [1880]] and of moulds [Mazé, [1902]], and also by chemical means. Thus Duclaux [[1886]] found that calcium lactate solution exposed to sunlight underwent decomposition, yielding alcohol and calcium carbonate and acetate, whilst Hanriot [[1885], [1886]], by heating calcium lactate with slaked lime obtained a considerable quantity of a liquid which he regarded as ethyl alcohol, but which was shown by Buchner and Meisenheimer [[1905]] to be a mixture of ethyl alcohol with isopropyl alcohol.

It appears, therefore, that inactive lactic acid can be quite readily obtained in large proportion from the sugars by the action of alkalis, whilst alcohol can only be prepared in comparatively small amount and probably only as a secondary product of the decomposition of lactic acid.

The study of the action of alkalis on sugar has, however, yielded still further information as regards the mechanism of the reaction by which lactic acid is formed. A considerable body of evidence has accumulated, tending to show that some intermediate product of the nature of an aldehyde or ketone containing three carbon atoms is first formed.

Thus Pinkus [[1898]] and subsequently Nef [[1904], [1907]], by acting on glucose with alkali in presence of phenylhydrazine obtained the osazone of methylglyoxal, CH3·CO·CHO. This osazone may be formed either from methylglyoxal itself, from acetol, CH3·CO·CH2·OH, or from lactic aldehyde, CH3·CH(OH)·CHO [Wohl, [1908]]. Methylglyoxal itself may also be regarded as a secondary [p099] product derived from glyceraldehyde, CH2(OH)·CH(OH)·CHO, or dihydroxyacetone, CH2(OH)·CO·CH2(OH), by a process of intramolecular dehydration, so that the osazone might also be derived indirectly from either of these compounds [see also Neuberg and Oertel, [1913]]. Methylglyoxal itself readily passes into lactic acid when it is treated with alkalis, a molecule of water being taken up:—

CH3·CO·CHO + H2O = CH3·CH(OH)·COOH.

Further evidence in the same direction is afforded by the interesting discovery of Windaus and Knoop [[1905]], that glucose is converted by ammonia in presence of zinc hydroxide into methyliminoazole,