Euler-Lagrange analysis towards representative down-scaling of a 22 m3 aerobic S. cerevisiae fermentation

Cees Haringa,Amit T Deshmukh,Robert F Mudde,Henk J Noorman

doi:10.1016/j.ces.2017.01.014

Cees Haringa, Amit T Deshmukh + Show 2 more

Open Access

https://doi.org/10.1016/j.ces.2017.01.014

Copy DOI

Abstract

With reaction timescales equal to or shorter than the circulation time, the ideal mixing assumption typically does not hold for large scale bioreactors. As a consequence large scale gradients in extra-cellular conditions such as the substrate concentration exist, which may significantly impact the metabolism of micro-organisms and thereby the process performance. The influence of extra-cellular variations on the organism can be tested using so-called scale-down simulators, laboratory scale setups where deliberate, controlled fluctuations are imposed in the extra-cellular environment.The major challenge associated with this scale-down philosophy is to design a scale-down simulator that resembles the extra-cellular environment of the industrial process. Previously, Euler-Lagrange CFD was explored to investigate the large scale environment from the microbial point of view (Haringa et al., 2016a), collecting statistics of the frequency and magnitude of environmental fluctuations that can serve as a basis for scale-down design. In this work, we apply this methodology to a validated CFD simulation of a 22m3 aerated fermentation of S. cerevisiae, and devise possible scale-down strategies based on this CFD data, both with fluctuating feed profiles and multiple compartments. All designs are deemed feasible within the limitations of current scale-down equipment.

Highlights

We propose the scale-down simulator designs based directly on Computational Fluid Dynamics (CFD) data
We show, for the first time, scale-down simulator design proposals that are directly based on CFD simulations of an industrial scale fermentor
Good agreement was observed in terms of power number and gas holdup, and reasonable agreement was obtained for kla and the mixing time

Summary

Introduction

H kL a Ks Y sx M ms Nc Np Ns P qp qs qs;max qref QM substrate concentration (in broth) [mol/kg] substrate concentration (mass based) [mg/kg] biomass concentration (in broth) [g/kg] feed concentration [mol/L] off-bottom clearance impeller [m] Impeller diameter [m] sauter mean diameter [mm] diffusion coefficient [m2/s] substrate feed rate [mol/s] feed flowrate [L/L/s] gravitational acceleration [m/s2] tank height [m] mass transfer coeff. Biomass yield on substrate [Cmolx=mols] Nm [Impeller moment] maintenance coefficient [mols=Cmolx=h] total number grid cells [–] total number particles [–] impeller revolutions [1/s] power input [W] specific production rate of product [molp=Cmolx=h] specific uptake rate of substrate [mols=Cmolx=s] max. H kL a Ks Y sx M ms Nc Np Ns P qp qs qs;max qref QM substrate concentration (in broth) [mol/kg] substrate concentration (mass based) [mg/kg] biomass concentration (in broth) [g/kg] feed concentration [mol/L] off-bottom clearance impeller [m] Impeller diameter [m] sauter mean diameter [mm] diffusion coefficient [m2/s] substrate feed rate [mol/s] feed flowrate [L/L/s] gravitational acceleration [m/s2] tank height [m] mass transfer coeff. [1/h] affinity constant for substrate [mol/kg] max. biomass yield on substrate [Cmolx=mols] Nm [Impeller moment] maintenance coefficient [mols=Cmolx=h] total number grid cells [–] total number particles [–] impeller revolutions [1/s] power input [W] specific production rate of product [molp=Cmolx=h] specific uptake rate of substrate [mols=Cmolx=s] max. specific uptake rate of substrate [mols=Cmolx=s] reference qs=qs;max [–] Maximum qs=qs;max obtained in SD simulator [–] Qg T t Ug V a Dt

Methods

Results

Conclusion