At first, we attempted the decarboxylation according to the peroxide method using cyclohexanol as the solvent. However, it was difficult to establish
a general method because both the reaction time and amount of catalyst required varied to complete the decarboxylation even under the same scale
reaction. For instance, when 4-hydroxy-L-proline was treated with peroxide catalyst (tert-butyl peroxide or tetralin hydroperoxide) in cyclohexanol
under reflux, the peroxide consumed ranged between 3% and 10% and decarboxylation took between 7 h and 28 h to complete. This variation was found to
greatly depend on the quality of the cyclohexanol used. Strangely, the decarboxylation rate gradually decreased in proportion to increase in the
purity of the cyclohexanol. For example, in the case of 99.3% purity cyclohexanol, it took 28 h reaction time and the addition of 10% peroxide.
However, with 98.1% purity cyclohexanol, the reaction was completed in 7 h by the addition of only 3% peroxide. These results suggest that there was
some effective substance contained in the cyclohexanol. The most effective way to determine this substance was by the use of its UV spectrum.
cyclohexanol of 99% purity has no peak between 200 to 320 nm, but 98% purity cyclohexanol showed a strong peak at 225 nm. Only 2-cyclohexen-1-one has
a peak at 225 nm within some impurities contained in cyclohexanol. Quantitative analysis by gas chromatography showed that it makes up less than 0.1%
in 99% purity cyclohexanol but more than 0.3% in 98% purity cyclohexanol.
|