Solvent-Dependent Transition States for Decarboxylations

The rate constants and kinetic isotope effects for decarboxylation of 4-pyridylacetic acid depend strongly on whether the solvent is water or dioxane, and the present paper interprets this finding. We calculate the solvent dependence of the free energy barrier and of the <sup>13</sup>C and <sup>18</sup>O kinetic isotope effects using a quantum mechanical solvation model based on class IV charges and semiempirical atomic surface tensions. The calculations provide a consistent interpretation of the experimental results, which provides a striking confirmation of the soundness of the solvation modeling. Even more significantly, the agreement of theory and experiment gives us confidence in the physical picture of the reaction provided by the model. This indicates that the location of the transition state, as measured by the length of the breaking C−C bond, is 0.24 Å later than the gas phase in dioxane and 0.37 Å later than the gas phase in water. Charge development at the transition state also depends strongly on the solvent; in particular the CO<sub>2</sub> moiety is 0.07 electronic charge units more negative at the transition state in dioxane than in water.