Cautions regarding the physical interpretation of statistically based separation of the ortho substituent effect into inductive, mesomeric, and steric components—II : Acid catalysed esterification of benzoic acids by methanol and cyclohexanol

Abstract

In order to attain a better insight into the composition of Taft E s o, constants the rate data of hydrion catalysed esterification of both m,p-substituted and o-substiluted benzoic acids by cyclohexanol and methanol were submitted to a statistical analysis using inductive mesomeric and steric substituent constants and various dummy variables differently structured. Furthermore a principal component analysis with subsequent identification of the first principal component via multiple regression analysis was applied. It has been demonstrated that in the reactions of m,p substituted compounds some substituents capable of exerting strong mesomeric effects show peculiar characteristics deviating from the general trend. Since the same result was obtained in the correlations of ortho substituted benzoic acids this effect was taken into account using an appropriate dummy variable which in all cases improved the multiple coefficient of determination. It is concluded that the esterification rates of the ortho substituted compounds depend essentially upon inductive and steric effects (taken away OMe OEt and NO 2) as proposed by Taft. While generally the E s o values may be regarded as some measure of a steric effect, this is not true for the methoxy and ethoxy groups.