Cage, radical

The mechanism shown in Scheme 6 is, for the most part, consistent with points (1) to (9). Thus, initially formed is a o--complex that is stable only at low temperatures. Upon matrix warm-up, a caged, radical pair forms and, if the R portion possesses a sufficient excess of vibrational energy, decomposition processes may occur. The radicals combine to form RPdX, which may, or may not, be isolated. 

The factor of seven variation between k2 for the ordinary and benzylic tertiary hydrogen is too small to be associated with carbonium ion formation. The observed degrees of retention and 0 transfer imply a caged radical rapidly reducing Mn(VI) to Mn(V) by accepting oxygen... 

A subsequent detailed analysis of the permanganate oxidation of the tertiary hydrogen atom of 4-phenylvaleric acid in 2.5 M potassium hydroxide solution supports the caged radical mechanism. The reaction order is two overall, A h/ d is ca. 11.5, ring substitution has little elfect on the rate (p 0) and the oxidation proceeds with a net 30-40 % retention of optical configuration. 

All these quantities are shown in figure 3.2, where Ais the the activation enthalpy for the cage radical pair recombination and Ais the activation enthalpy for diffusive cage escape. 

However, solvent viscosity, rather than polarity, has been a useful tool for mechanistic purposes. Although the quantum yield of the ortho-rearranged product of 4-methylphenyl acetate (20) does not change with viscosity of the medium, the formation of 4-methylphenol (22) is highly sensitive to this factor. Thus, its quantum yield is 0.45 in ethanol (1.00 cP) but only 0.02 in Carbowax 600 (109 cP) (Scheme 8 Table 3) [13], This clearly supports the mechanism involving caged radical pairs. A related aspect is the intramolecular nature of the process confirmed by the lack of cross-coupling products in crossover experiments with mixtures of different esters [10]. 

Clear-cut evidence on the intermolecular nature of the excitation step and the need for in-cage radical annihilation has been gained from a study on the solvent-cage effect in the peroxyoxalate reaction. The viscosity effect on the singlet quantum yields was verified using the binary solvent system toluene-diphenyhnethane. A rise in solvent viscosity from 0.50 cP (toluene) to 2.67 cP (diphenylmethane) leads to an up to tenfold increase in the quantum yields, demonstrating the intermolecular nature of the excitation step. 

The separation of 3.1-3.5 A is close to the sum (3.3 A) of the van der Waals radii of S and N. One interpretative difficulty now becomes apparent What keeps the caged radicals, which on this basis are essentially in physical contact, from recombining by formation of a covalent bond Such close contacts of partners have been reported in cages in the crystalline state but apparently not in glassy media. 

---