CMA: integration of fluid dynamics and microbial kinetics in modelling of large-scale fermentations

• Published in: Chemical engineering journal (Lausanne, Switzerland : 1996) Vol. 84; no. 3; pp. 463 - 474
• Main Authors: Vrábel, Peter, van der Lans, Rob G.J.M, van der Schot, Freija N, Luyben, Karel Ch.A.M, Xu, Bo, Enfors, Sven-Olof
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 15.12.2001; Elsevier
• Subjects: batch; Biological and medical sciences; Biomass; Bioreactors; Biotechnology; Compartment model; Computer simulation; escherichia-coli; Fluid dynamics; Fundamental and applied biological sciences. Psychology; Integration; Metabolism; Methods. Procedures. Technologies; Microbial engineering. Fermentation and microbial culture technology; Microbial kinetics; Multiple impellers; overflow metabolism; reactors; Scale-down; Substrates; yield; Compartment model; Scale-down; Multiple impellers; Fluid dynamics; Microbial kinetics; Escherichia coli; Bacteria; Kinetics; Fermentation; Modeling; Enterobacteriaceae
• ISSN: 1385-8947; 1873-3212; 1873-3212
• DOI: 10.1016/S1385-8947(00)00271-0

Transport limitation is regarded as one of the major phenomena leading to process yield reduction in large-scale fermentations. Knowledge of both the fluid dynamics and the microbial kinetics is needed for understanding and describing situations in large-scale production bioreactors. Microbial kinetics of Escherichia coli including flow metabolism was determined in lab-scale batch and fed-batch experiments. The effect of high substrate fluctuations on metabolism was quantified in scale-down experiments. This knowledge was incorporated into a flow model based on the compartment model approach (CMA). The flow model was verified by mixing time experiments on aerated reactors mixed with multiple impellers at different regimes with liquid volumes 8–22 m 3. The integral model, predicting local glucose, acetate and biomass concentrations in different parts of the reactor, was compared to three large-scale fermentations performed in two different reactors. If lab-scale kinetics was used, the biomass prediction overestimated the biomass concentration. Lab-scale kinetics modified by the results of scale-down experiments incorporating the effect of substrate fluctuations gave a rather satisfying description of biomass concentration. Glucose gradients in different parts of the reactor and acetate produced as a result of overflow metabolism were predicted on a qualitative level. The simulations show that at present the decisive factor for a successful integration of fluid dynamics and microbial kinetics is the kinetics.