Integrated operation of biotransformation and simulated moving bed (SMB) separation is an attractive option for high-yield manufacturing of commercially relevant compounds such as rare sugars and sialic acids from equilibrium-limited isomerase- or aldolase-catalyzed reactions. Here, we present the first lab-scale implementation of such a process using the production of d-psicose, which is currently under consideration as low calorie sweetener, by d-tagatose epimerase-catalyzed epimerization from d-fructose as a model system. While a typical batchwise eprimerization of d-fructose would stop at 25%, a yield of 97% was obtained when operating the fully integrated process consisting of SMB, enzyme membrane reactor (EMR) and nanofiltration (NF) for a number of days with absolute product purities. Next to the proof of principle, important process characteristics such as startup time, stability and robustness were investigated. By pre-equilibrating the NF unit to the projected conditions, startup times could be reduced to the contributions from EMR and SMB (in this case below 5 h) which was perfectly in line with the projected range of operation time of a few days. Robustness was probed by introduction of a perturbation, specifically a 2-fold increase in process feed concentration, which did not compromise any of the set specifications. Next, long-term operation of the respective units indicated a potential process time of at least 5 days, which could be easily extended in the future by engineering a more stable enzyme variant and implementing a cleaning-in-place approach for SMB column regeneration. In summary, the principle feasibility of such process integration for fine chemical synthesis could be successfully demonstrated.

Integration of enantioseparation by simulated moving bed (SMB) and mild enzymatic racemization enables the production of single enantiomers from a racemic mixture in theoretically 100 % yield and hence overcomes the 50 % yield limitation of conventional SMB processes. We implemented such a process consisting of a Chirobiotic TAG column-SMB, an amino acid racemase-containing enzyme membrane reactor, and a nanofiltration unit for concentration of the distomer-enriched SMB raffinate prior to racemization on lab-scale for the production of enantiopure D-methionine. The integrated process scheme was operated continuously for over 30 hours without significant variations in product concentration and purity and with a yield of 93.5 %, demonstrating the feasibility of this integrated process concept. Furthermore, a rational analysis of the integrated process on the basis of a shortcut model was conducted. The process model consists of a true moving bed equilibrium stage model to represent the SMB, a continuous stirred tank reactor model with reversible Michaelis-Menten kinetics to represent the enzyme membrane reactor, a nanofiltration model and feed node mass balances, and enabled the identification of optimal operating points (flow rate ratios, enzyme concentration) at a variety of process specifications and objectives. Optimal operating points were calculated for different cost distributions between the applied materials such as stationary phase, enzyme, solvent, and nanofiltration membrane. By assigning plausible pricing data and lifetimes to the respective materials, variable costs for the specific process considered in this work were estimated.

No abstract

No abstract

No abstract

No abstract

No abstract

No abstract