(250b) Enzymatic Reactive Extraction of Fermentation Products for the Production of Short-Chain Esters

Esters such as ethyl acetate and ethyl lactate are biodegradable solvents with applications across multiple industries including adhesives, food, and pharmaceuticals. However, the conventional process for producing esters is not eco-friendly because they are largely derived from crude oil. More recently, metabolic engineering has been explored as an alternative route for the production of esters from renewable feedstocks, but the resulting ester titers were fairly low. A major factor for this low titer is that the esterification reaction is not thermodynamically favorable in the aqueous phase as water is a byproduct of the reaction. We propose that by performing reactive extraction this limitation can be overcome as the reaction would mainly happen in the interface and the product would be continuously extracted to the organic phase. In this work, we have identified the optimal conditions required for the production of esters and modeled the enzymatic kinetics.

Our results showed that, for production of ethyl acetate through this reactive extraction approach, Novozyme 435 exhibited significant improvements in esterification with chloroform as the extractant, resulting in an almost 85% conversion efficiency. Further optimizations with phase ratios, pH, and incubation time were performed. It was demonstrated that the pH 6.0 was the most optimum where ethyl acetate titer was found to improve 10 times than that at pH 7.0 with the phase ratio of 1:1. Kinetic studies further demonstrated that reactive extraction at 37oC provides maximum productivity for ethyl acetate synthesis. After initial optimization studies, reactive extraction was performed on cell broth resulting from fermentation that accumulated ethanol and acetate. However, we observed a ~7.5X decrease in ethyl acetate production in the cell broth versus simulated samples that had the same concentration of reactants.

As a last step, a mathematical model was developed for the reactive extraction which will be useful for scaling up studies. The mathematical model employed assumes that enzyme kinetics rather than diffusion was the rate limiting step and that the concentrations of reactants at the interface are equivalent to the initial concentration of reactants. Vmax was found to be 18.5 mmol min-1g-1 of catalyst used, and the Michaelis constants(Km) were 0.957 M and 0.00557 M for acetic acid and ethanol respectively. Overall, we hope that by optimizing the reactive extraction process for ester production, biorefineries could become more competitive and economically feasible for numerous applications.