Model Based Constant Feed Fed-Batch L-Sorbose Production Process for Improvement in L-Sorbose Productivity

R. Giridhar and A. K. Srivastava

Abstract
Batch sorbose productions were done using Gluconobacter oxydans and initial sorbitol concentrations of (γso = 200 kg m–3). The average batch kinetic data was used to develop an unstructured mathematical model to represent the sorbose production system. Model parameters were identified using a non-linear regression technique assisted by a computer program which minimized the deviation between the model predictions and actual batch experimental data. F-test indicated a 99 % confidence on the prediction of model using optimized parameters. Parametric sensitivity analysis indicated maximum specific growth rate (μm) to be the most sensitive model parameter. The batch model was then used to identify nutrient feeding strategies to maintain a constant noninhibitory and non-limiting supply of sorbitol for fed-batch cultivation, in order to improve the sorbose productivity. The adequacy of the fed-batch model was tested by comparing actual fed-batch cultivation data with the model predictions. Excellent agreement was found between experimental data and model simulation. The fed-batch model was found to be a powerful tool for designing feeding strategies for enhancing yield and productivity of sorbose production.

This work is licensed under a Creative Commons Attribution 4.0 International License

Keywords
D-sorbitol, L-sorbose, Gluconobacter oxydans, modeling of kinetics, fed-batch fermentation