Alkali, action saccharinic acids

This benzilic acid type of rearrangement is the result of the action of alkali on the dicarbonyl compound, and is accelerated by calcium ions. The formation of saccharinic acids by the action of aqueous alkali on sugars is very well known 82,84,92 however, if ammonia is present, very little8 or no production of saccharinic acid has been reported. The reaction of the intermediate carbonyl compounds with ammonia is faster than the benzilic acid type of rearrangement to give saccharinic acid, and, hence, substituted imidazoles are formed, as illustrated in Scheme 9. 

Komoto detected lactic acid in the mixture from reaction of D-glucose with ammonia,4 and presumed that it was produced from pyruvaldehyde formed by decomposition of D-glucose. Lactic acid has, indeed, been found as a product of the action of alkali (lime-water) on substituted D-glucose and substituted D-fructose,81,83,96 and the mechanism of its formation involves the reversible aldol reaction, followed by formation of pyruvaldehyde, and the benzilic acid rearrangement already described for saccharinic acid this is illustrated83,96 in Scheme 11. 

Shortly after the discovery of Peligot s a -D-glucosaccharin, Dubrun-faut reported that the calcium salt of a monobasic acid resulted from the action of lime-water on maltose. Cuisinier named the acid isosaccharinic acid, after he had prepared from it a crystalline lactone (CeHioOt) isomeric with Peligot s a -D-glucosaccharin. The name was expanded to a -D-iso-saccharinic acid after Nef obtained evidence of the concurrent formation of its epimer, /3 -D-isosaccharinic acid, in the hexose-alkali reaction. 

An experimental demonstration that the Nef mechanism for the initial conversion of a sugar by alkali to the a-dicarbonyl structure is not acceptable was provided by a study of the action of alkali on 2 hydroxy-3-meth-oxy-3-phenylpropiophenone. Nicolet observed that the products in this case are 2,3-diphenyllactic acid and methanol. This conversion, which is completely analogous to the formation of a saccharinic acid from a reducing sugar, demonstrated clearly that carbon-oxygen cleavage occurs at the /3-carbon atom, rather than at the a-carbon atom, with respect to the carbonyl group. 

The preceding explanation of the failure of a 2-0-substituted aldose to form saccharinic acids (on treatment with alkali) finds substantiation in the results of a study of the action of lime-water, at room temperature, on 2,3-di-O-methyl-D-glucose. The presence, in this latter reaction mixture, of an fl[,/3-unsaturated aldehyde(IV, R = CH3) was established by its further, facile conversion to 5-(hydroxymethyl)-2-furaldehyde upon acidification. 

Dihydroxybutanoic acid, a 3-deoxytetronic acid, is the only theoretically possible four-carbon metasaccharinic acid it was isolated by Nef in the dl form (Ilab) in the course of his work on the action of sodium hydroxide on D-arabinose. It therefore constitutes the only four-carbon saccharinic acid isolated to date from a sugar-alkali reaction. Resolution of this on acid (Ilab) was accomplished by Nef, who showed that the dextrorotatory acid (Ila) upon oxidation gave rise to 2-deoxy-L-f/Z2/cero-tetraric acid [(-)-)-d-malic acid XV]. During the course of this work, several derivatives of the L and DL saccharinic acids (lib and Ilab) were prepared, but only the brucine salt of the d acid (Ila) was reported. 

Lactic acid occurs amongst the products of the action of alkali on hexoses. From 1 mole of D-glucose, treated at 25°C. with benzyltrimethylammonium hydroxide, Evans reported the production of 1.2 moles of racemic lactic acid (60 % of a theory of 2 moles per mole hexose). The substance apparently can be obtained by the action of alkali on any sugar (inclusive of trioses, pentoses, disaccharides, etc.) (99). Its preparation from sucrose under conditions of high temperature and pressure has been extensively studied from the viewpoint of potential industrial application. Lactic acid may be considered as the saccharinic acid (see below) related to glyceraldehyde and may arise from the rearrangement of a triose fragment. 

---