| The efficient scale-up of industrial fermentation process is always a challenge.To increase product yields and ensure consistent product quality,key issues of industrial fermentations,including process optimization and scale up are aimed at maintaining optimum and homogenous reaction conditions.Thus researches focusing on the interaction of heterogeneous flow field in bioreactors and physiological response of microorganisms are emerging recently,and computational fluid dynamics(CFD)has become an efficient tool for the investigation of engineering parameters in bioreactors.Based on the founding of scaling effect in Aspergillus niger fermentation process,the issues existed in scale-up for different fermentation systems were studied by CFD methods,and scale-up strategies were proposed for shear-control,mass-transfer-control and mixing-control fermentations respectively.Finally,a method based on CFD and time-constant analysis was raised to solve scale-up issues systematically.To address the dynamic regulatory mechanism of A.niger being exposed to inhomogeneous glucose concentrations,glucose perturbation experiment were carried out based on the steady state of A.niger chemostat culture,and dynamic profiles of the intracellular metabolites in central carbon metabolism were tracked in a time scale of seconds.The upper glycolysis and pentose phosphate pathway showed sharp variations after glucose perturbation,while the lower glycolysis,TCA cycle and amino acid pools represented a moderate and prolonged response due to the allosteric regulation of enzymes and buffering function of metabolites with large pool sizes.Improved glucose-6-phosphate enhanced the metabolic flux to PP pathway remarkably,which provided not only more redox cofactors(NADPH)for protein synthesis but also more precursors(phosphoribosyl pyrophosphate and ribose-5-phosphate)for cell growth.Moreover,reduction of the total adenine nucleotides and major precursor amino acids indicated the upregulated RNA synthesis was required to produce stress proteins,and partially explained the drop of glucoamylase production when A.niger experienced a fluctuated glucose concentration environment.To analysis the key factors in industrial scale-up process,a lab-scale bioreactor was taken as research object to establish the CFD methods for quantification of engineering parameters,including mass transfer,mixing and shear.The CFD methods were verified by the measurement of kLa using gas balance method.The established CFD methods for simulation of engineering parameters provided a methodological foundation for investigation of engineering aspects and scale-up in industrial-scale bioreactors.To achieve the scale-up of shear sensitive cells,this study raised a new scale-up strategy based on 3D shear space for large-scale animal cell culture.Firstly,the shear environments of bioreactors ranging from lab-scale to industrial-scale were quantitatively analyzed by CFD methods and successfully validated by particle image velocimetry(PIV)experiments.Moreover,the quantitative relationships between shear parameters(including shear rates in impeller and tank zone,overall averaged shear rate and maximum shear rate)and impeller tip velocity were established.Besides,correlation analysis between shear-related parameters and viable cell density indicated that shear rates in impeller and tank zone,and the overall averaged shear rate were identified as the three key shear indices for scale-up of Spodoptera frugiperda Sf9.Further,an optimized 3D operation space for shear rate was established according to the three key shear parameters obtained under preferable operation conditions in lab-scale bioreactors.Based on that,stirring speed in large-scale bioreactors were determined using the correlation proposed.Ultimately,we achieved successful scale-up of Spodoptera frugiperda Sf9 in industrial bioreactors up to 1000 L using this strategy.The scale-up strategy proposed here are supposed to be applicable to guide the scale-up process of shear-sensitive cells.For the scale-up of highly efficient aerobic strains,the flow field comparison between pilot scale and industrial scale bioreactors,and physiological performance of E.coli in high-cell-density fermentation were investigated.The results showed that the lower cell density in industrial scale bioreactor was caused by limited oxygen supply in lower part of the bioreactor.Thus four optimization strategies were raised and validated by CFD method.By changes of impeller type,impeller diameter and sparger direction,the best result suggested that the volumetric oxygen transfer rate(kLa)was increased by 16%,reaching the level in pilot scale bioreactor.Additionally,based on oxygen transfer capacity of the industrial scale bioreactor obtained by CFD methods,as well as the relationship between oxygen supply and productivity during B.subtilis fermentation,a fed-batch process with C/N source was proposed and optimized at different oxygen supply levels.The results showed that the utilization of C/N sources during the fermentation was more balanced under limited oxygen supply condition(OUR=50-60 mmol/L/h),which led a higher producing rate and yield than that under adequate oxygen supply condition.Further,the industrial scale bioreactor was improved to satisfy the scale-up of the optimized fed-batch process,resulting in titer of 6000 U,nearly quadrupled.For the needs of high efficient mixing in limited substrate fed-batch process,the flow fields and engineering parameters in 2 t pilot scale bioreactor and 80 t industrial scale bioreactor were compared by CFD simulation.The four Rushton turbine impeller combinations in 80 t bioreactor resulted in long mixing time(t95=67.2 s)and apparent substrate concentration gradient,and thus reduced the fermentation titer.To ensure adequate oxygen supply and improve mixing,a new impeller combination was applied to the 80 t bioreactor and analyzed by CFD methods.Results showed that the optimized impeller combination decreased the mixing time from 67.2 s to 34.9 s.What's more,the mixing time was heavily dependent on the flow field at the feeding point.By feeding from middle part of the fermenter,the minimal mixing time was shortened to 17.7 s.However,feeding from multiple points could only reduce the substrate concentration gradient,but had no significant effect on mixing time.Based on the above foundings,a methodology for scale-up based on CFD and time-constant analysis was proposed.First,according to the physiology performance of inducible E.coli in high-cell-density fermentation,the time constants about oxygen consumption and substrate consumption were determined in growth phase and inducing phase respectively.By comparison of time constants in consumption type(for the strain)and that in supplying type(for the bioreactor),we verified that oxygen supply was the key factor in growth phase,while mixing in inducing phase.Specifically,the oxygen transfer time constant(tmt)should be longer than 4.2 s(or kLa>0.236s-1)during cell growth,and the mixing time should be less than 36 s during inducing,Therefore,a 20 m3 bioreactor was designed based on reactor design theories,and its performances were verified by CFD methods.It showed that the newly designed bioreactor possessed a kLa>0.236 s-1 and a mixing time less than 36 s,which met the scale-up requirements. |
PDF Full Text Request