Activity-Based Models to Predict Kinetics of Levulinic Acid Esterification.

The solvent is of prime importance in biomass conversion as it influences dissolution, reaction kinetics, catalyst activity and thermodynamic equilibrium of the reaction system. So far, activity-based models were developed to predict kinetics and equilibria, but the influence of the catalyst on kinetics has not been succesfully predicted by thermodynamic models. In this work, the thermodynamic model ePC-SAFT advanced was used to predict the activities of the reactants and of the catalyst at various conditions (temperature, reactant concentrations, γ-valerolactone GVL cosolvent addition, catalyst concentration) for the homogeneously acid-catalyzed esterification of levulinic acid (LA) with ethanol. Different kinetic models were applied, and it was found that the catalyst influence on kinetics could be predicted correctly by simultaneously solving the dissociation equilibrium of H SO catalyst along the reaction coordinate and by relating reaction kinetics to proton activity. ePC-SAFT advanced model parameters were only fitted to reaction-independent phase equilibrium data. The key reaction properties were determined by applying ePC-SAFT advanced to one experimental kinetic curve for a set of temperatures, yielding the reaction enthalpy at standard state , activation energy and the intrinsic reaction rate constant k=0.011 s at 323 K, which is independent of catalyst concentration. The new procedure allowed an a-priori identification of the effects of catalyst, solvent and reactant concentration on LA esterification.