Influence of Liquid-liquid Equilibria of Oleic Acid/ethanol/water on the Modelling of Solvent-free Ethyl Oleate Synthesis

M. BRIOZZO; M.L. FORESTI; G. TRUBIANO; M. L. FERREIRA; V. BUCALÁ; S. BOTTINI

Village Rio das Piedras

Congreso; ENPROMER 2005. 2nd MERCOSUR CONGRESS ON CHEMICAL ENGINEERING; 2005

. Enzymes are powerful catalytic tools in a wide variety of chemical processes. In this work, experimental data from the direct esterification of oleic acid and ethanol catalyzed by immobilized lipase from Candida antarctica B are presented. Reaction was performed in a solvent free system. According to the amount of water present in the reaction medium (either initially added or produced by the chemical reaction) one or two liquid phases were initially detected. However at equilibrium, for all experiments performed, the generation of water during reaction assured the presence of a second aqueous phase. Therefore, at equilibrium conditions the liquid-liquid and the chemical reaction equilibria have to be study simultaneously. The reaction equilibrium constant for the formation of ethyloleate by a lipase-catalyzed reaction should only depend on temperature, if the standard states are properly selected. However, several authors have shown that the equilibrium constant varies with the initial substrate ratio an also with the water percentage added to reaction media. The scarce published reports dealing with biphasic systems introduce the concept of a biphasic equilibrium apparent constant that depends on the organic to aqueous phases volume ratio. Although this method has been reported to be a valuable tool to predict experimental data, it is far from the classical approach used to handle reversible chemical reactions. The aim of this work is to consider the biphasic nature of the reactive mixture not by means of the biphasic equilibrium constant (which is based on total concentrations) but through a complete mathematical model, that takes into account the liquid-liquid and reaction equilibria. This model allows fitting the thermodynamic equilibrium constant to well predict the experimental equilibrium conversions. The liquid-liquid equilibrium data required by the model are predicted by UNIFAC.Enzymes are powerful catalytic tools in a wide variety of chemical processes. In this work, experimental data from the direct esterification of oleic acid and ethanol catalyzed by immobilized lipase from Candida antarctica B are presented. Reaction was performed in a solvent free system. According to the amount of water present in the reaction medium (either initially added or produced by the chemical reaction) one or two liquid phases were initially detected. However at equilibrium, for all experiments performed, the generation of water during reaction assured the presence of a second aqueous phase. Therefore, at equilibrium conditions the liquid-liquid and the chemical reaction equilibria have to be study simultaneously. The reaction equilibrium constant for the formation of ethyloleate by a lipase-catalyzed reaction should only depend on temperature, if the standard states are properly selected. However, several authors have shown that the equilibrium constant varies with the initial substrate ratio an also with the water percentage added to reaction media. The scarce published reports dealing with biphasic systems introduce the concept of a biphasic equilibrium apparent constant that depends on the organic to aqueous phases volume ratio. Although this method has been reported to be a valuable tool to predict experimental data, it is far from the classical approach used to handle reversible chemical reactions. The aim of this work is to consider the biphasic nature of the reactive mixture not by means of the biphasic equilibrium constant (which is based on total concentrations) but through a complete mathematical model, that takes into account the liquid-liquid and reaction equilibria. This model allows fitting the thermodynamic equilibrium constant to well predict the experimental equilibrium conversions. The liquid-liquid equilibrium data required by the model are predicted by UNIFAC.