Back to overview

Model-based cost optimization of a reaction-separation integrated process for the enzymatic production of the rare sugar D-psicose at elevated temperatures

Type of publication Peer-reviewed
Publikationsform Original article (peer-reviewed)
Publication date 2015
Author Wagner Nina, Bosshart Andreas, Wahler Sonja, Failmezger Jurek, Panke Sven, Bechtold Matthias,
Project Process intensification with integrated multi-enzyme systems
Show all

Original article (peer-reviewed)

Journal Chemical Engineering Science
Volume (Issue) 137
Page(s) 423 - 435
Title of proceedings Chemical Engineering Science


The integration of biocatalysis with continuous separation of product and substrate constitutes an attractive option for overcoming the intrinsic yield limitation in enzymatic reactions that suffer from an unfavourable position of the equilibrium. One example constitutes the recently established continuous process combining enzymatic epimerization in an enzyme membrane reactor (EMR), simulated moving bed (SMB) chromatography and nanofiltration for the high-yield production of the rare sugar D-psicose from its bulk epimer D-fructose with recycling of the unused epimer to the EMR. Rational optimization of the process concept requires the analysis of the overall process economics on the basis of a process model, accounting for the cost contributions of each involved unit operation. Therefore, we constructed a process model consisting of a continuous stirred tank reactor model with reversible Michaelis-Menten kinetics for representation of EMR operation, a transport-dispersive true moving bed model for representation of SMB operation and an ideal filtration model. As biocatalyst costs are usually strongly depending on the reactor temperature we further established a model-based procedure for characterizing the biocatalyst in terms of operational stability and activation as a function of the reactor operating temperature. In order to confirm the design procedure a highly productive run was implemented on our existing lab-scale plant allowing for 2.3 kg of D-psicose per L of SMB column volume and day and 4.5 kg D-psicose per g enzyme and day producing D-psicose at high purity (≈ 98%) and in remarkable yield (96 %). Next, we performed a comprehensive analysis of the process in terms of suitable reactor temperatures based on a multi-objective function addressing the cost contributions of the biocatalyst, the SMB stationary phase, desorbent and the substrate. As expected, the enzyme biocatalyst cost were found to have a major impact on the overall process costs and therefore were minimal at the lowest investigated EMR temperatures, at which almost no enzyme degradation occurred. However, the corresponding slow reaction kinetic translated into very large EMR residence times and strongly reduced purity (84%) in the recycling stream of the SMB. In contrast, at high temperatures in the EMR the main costs could be attributed to the biocatalyst. Overall, this resulted in an optimum operating point at which the reactor is running highly productive (i.e. short residence times and almost no reflux of unconverted substrate from the SMB). The simulated process yield was generally very high (> 94%).