Chemical Engineering Research & Design, Vol.134, 463-475, 2018
Dynamic modeling of the chemo-enzymatic epoxidation of alpha-pinene and prediction of continuous process performance
For the first time, a valid and reliable deterministic model for the description and prediction of the reaction dynamics of the chemo-enzymatic epoxidation in a continuous stirred tank reactor (CSTR) is presented. The model covers the different steps of the reaction cascade and all kinetic constants were found to be uniquely identifiable, which is a prerequisite for trustworthy model predictions. The chemo-enzymatic epoxidation is a multistep reaction, which consists of two consecutive reactions: the lipase-catalyzed peracid formation followed by a Prilezhaev epoxidation, wherein numerous factors govern the product formation. This makes the design of an optimal process particularly challenging and modeling mandatory. To obtain the model structure, the goodness of fit and (practical) parameter identifiability were taken into account. Different mass action kinetics, as well as mechanistic approaches, were investigated. Kinetic constants were fitted to experimental data using a multi-experiment fitting, and the practical identifiability of the estimated constants was shown. The developed CSTR process model, which describes the dynamics of the multistep chemo-enzymatic epoxidation of a-pinene, matched the experimental data very accurately. Moreover, it was successfully verified and validated over a broad range of operating conditions. (C) 2018 Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Keywords:Chemo-enzymatic epoxidation;alpha-Pinene oxide;Practical identifiability;Validation;Dynamic modeling