P47 ketoprofen decarboxylation

From: Sandra Monti
Date: 12/12/97
Time: 11:08:27 AM
Remote Name:


Dear Tito, your results are very interesting. You are probably right in invoking an upper triplet state for the reaction, but our results seem to indicate that ISC is faster.

I recall that our view was mainly based on the fact that in neutral aqueous medium, in almost exclusive presence of the carboxylate form, the triplet state was populated and detected by means of a picosecond system. It was shown to absorb at 526 nm and decay with a time constant of c.a. 250 ps, equal to the rise time of the 580 nm transient. These results could be fitted to a model in which the reactive upper triplet of the carboxylate form is in water very close in energy to the lowest triplet, so that the latter fast deactivates via the upper one. In the acid form this state does not exist, thus decarboxylation does not occur; in an environment of different polarity and/or nature this state could be higher in energy and the decarboxylation could proceed more slowly, still having the lowest triplet as precursor. The results we obtained in a cyclodextrin environment (the propionic group located outside the cavity) are consistent with such a view: the triplet carboxylate in the hydrophobic environment of the cavity has a longer lifetime but still seems to be the precursor of decarboxylation. The quantum yield of this process is ca. one half that in water, i.e. 0.42, the lowest triplet of the inclusion complex is populated (picosecond results) with efficiency similar to that observed for the molecule in aqueous environment and similar to that observed in benzophenone, i. e. unitary, and there is an additional, probably photoreductive, deactivation channel of quantum yield similar to that of decarboxylation. The sum of the quantum yields of the two photoprocesses is more than 0.8 (New J. Chem., in press).

In your experimental conditions you are in a non purely aqueous environment: could it be possible that you have a contribution of the undissociated triplet acid form? Moreover suppose you form the carboxylate triplet precursor by two kinetically distinct steps: (i) direct photoexcitation of the ground state carboxylate which converts to the 620 nm transient within the laser pulse and (ii) deprotonation of the triplet state acid with a rate of the same order as the decay rate of the 620 nm transient: could the decay kinetics at 525 nm and 620 nm be reconciled with this model?.

Back to the list of comments