Transesterification of substituted ethanols - modelling studies

Kinetic and molecular modeling of transesterification of ethyl acetate and substituted ethanols with porcine pancreatic lipase (PPL) and Candida cylindracea lipase (YL) is studied. Competitive product inhibition model is able to explain the observed behaviour well. The molecular dimensions of the substrates and the products, thermodynamic parameters and charges were estimated by molecular modelling based on minimum energy conformation and semi empirical quantum mechanical calculations. A strong correlation exists in the case of YL between conversion and (a) the cross sectional area of the alcohols, (b) difference between minimum energies of ester and alcohol, and (c) difference between the electrostatic potential volume of ester and alcohol. Only a weak correlation exists between these parameters, in the case of PPL whereas a strong correlation exists for PPL between conversion and charge on the alcohol oxygen of the substrate. Linear regression equations were developed for predicting the conversion based on these factors. Artificial neural networks (ANN) were employed to bring out the dependence of the catalytic activity on the molecular structure of the substituted ethanol. ANN model with charge on oxygen and molecular weight seems to fit the maximum reaction velocity in the case of PPL very closely. In the case of YL, ANN model with electrostatic potential volume difference and van der Waals volume fits the data satisfactorily. The variation of the maximum velocity with these parameters was also predicted using artificial neural network analysis. These studies indicate that both the lipases behave in different manner for the transesterification reaction. (C) 2004 Elsevier B.V. All rights reserved.