Reductones periodate oxidation

On the basis of experiments with myoinositol (50), it has been suggested that triose reductone should be oxidized by periodate to yield two molar equivalents of formic acid and one molar equivalent of carbon dioxide. However, it has been reported by two groups (1,29) that crystalline triose reductone is oxidized by two moles of periodic acid to give formic acid and glyoxylic acid, free iodine being liberated during the... 

The appearance of free iodine during the periodate oxidation of compounds having an active hydrogen atom (27) or an ene-diol structure (1,39) has frequently been observed, and this implies that further reduction of iodate, formed from periodate during the main reaction, takes place. It has, in fact, been shown that, in acid solution, iodate is fairly readily reduced by such compounds as triose reductone (27), dihydfoxy-fumaric (39), and tartronic (32) acids. 

We therefore carried out periodate oxidation of triose reductone in dilute solutions using sodium metaperiodate as the oxidizing agent (55,56), Triose reductone could react with periodate according to the following reaction sequence ... 

Mesoxalic dialdehyde can be reasonably expected (16,28,50) to undergo normal glycol cleavage and give one mole of formic acid and one mole of glyoxylic acid in fact, when a second molar equivalent of periodate was added to the above solution, two molar equivalents of titratable acid were formed. If an excess of periodate is now added, two molar equivalents of titratable acid remain, but in addition, one molar equivalent of carbon dioxide can be expelled from the solution. Thus, in the overall reaction, one mole of triose reductone is oxidized by three moles of periodate to give two moles of formic acid and one mole of carbon dioxide ... 

Thus, if triose reductone is, in fact, the first intermediate in the periodate oxidation of malonaldehyde, the total consumption of periodate per mole of malonaldehyde should be four molar equivalents two moles of formic acid and one mole of carbon dioxide should be formed, in accordance with the sequence proposed by Fleury and his collaborators (22). As in the case of the periodate oxidation of malonic acid (32) the rate determining step should be the hydroxylation step. 

Crystalline triose reductone has been shown (56) by titration with strong base and with iodine, to exist in solution, for practical purposes, entirely as the enol form. In addition, the fact that it reduces exactly three molar equivalents of periodate to give quantitative yields of formic acid and of carbon dioxide indicates that it is also oxidized entirely in this form. However, nothing is known of the rate of enolization of tartronic dialdehyde and the possibility therefore remains that part of it may be oxidized in the dialdehydo form. If this were the case, the results of periodate oxidations would be dependent on the ratio of the rate of enolization of tartronic dialdehyde to the rate of its oxidation by periodate, since the oxidation of triose reductone is, again, for practical purposes, instantaneous. 

A similar route has been suggested to account for the formation of glyoxylic acid (18, R = H) and methyl glyoxylate (18, R = Me) during the periodate oxidation of inositols and (9-methylinositols, respectively. Formation of these carboxy derivatives was depicted by sequences involving enolization to such reductones as 16, which are hydroxylated to 17 and then cleaved oxidatively. 

We examined the reaction of triose reductone with both periodate and iodate (55,56), and found that, whereas iodine was invariably set free from both sodium periodate and sodium iodate if the concentration of the reductone were greater than 10 3M, no iodine was liberated at lower concentrations (e.g. 6 x 10 4M) of substrate, even in the presence of relatively large amounts of the oxidants. 

Hesse and Mix (29) oxidized a relatively concentrated solution of triose reductone using limited amounts of free periodic acid. In these conditions, the iodic acid formed by the initial reduction of periodic acid could be further reduced and the reduction product could then, in turn, react with the remaining periodic acid and liberate iodine. Thus glyoxylic acid could be isolated from the oxidation mixture, as no periodate was available for its oxidation. 

However, when we oxidized malonaldehyde (56) in the conditions just described for triose reductone, although formic acid and carbon dioxide were produced in high yields, the periodate consumption was erratic. Similar results were obtained with deoxy sugars. This discrepancy may be caused by the incomplete enolization of the first intermediate, hydroxy malonaldehyde —i.e. tartronic dialdehyde (5,22,32), to triose reductone, or may concern the hydroxylation step itself. 

---