(316e) Computing Dissolved Oxygen Profiles in Aerobic Fermenters | AIChE

(316e) Computing Dissolved Oxygen Profiles in Aerobic Fermenters

Authors 

Vedapuri, D. - Presenter, Tridiagonal Solutions Inc.
Kasat, G., Tridiagonal Solutions Inc.
Benz, G., Benz Technology International, Inc.



Aerobic fermenters have been a staple method of bioprocessing for many years. Oxygen is necessary for all aerobic fermentation. Maintaining the appropriate and uniform concentration of dissolved oxygen (DO) is important for the successful growth of culture. Large DO gradients can lead to operating problems. Low or High DO levels in some part of reactors may lead to bi-product formation. A priori information on DO distribution can help in proper design and scale-up of industrial fermentation process.

Until now, there has been no way to reliably predict DO distribution. Indeed, it has not been possible to even measure it. A typical aerobic fermenter has only one DO probe; a few rare fermenters have two DO probes: one near the top of the fermenter, and one near the bottom. A substantial DO gradient has been observed in some fermenters: up to 100:1 ratio of the bottom DO to top DO with multiple radial turbines and up to 2:1 with axial/radial combinations.

Computational Fluid Dynamics (CFD) models have been successfully used to predict mixing, gas hold-up distribution, mass transfer & heat transfer rates in stirred gas-liquid fermenters. In this study the CFD models were used to predict DO distribution in a fermenter. Eulerian-Eulerian multiphase model along with standard k-e turbulence model has been used to predict the turbulent gas liquid flow in the vessel. The results of the CFD model were used to calculate oxygen transfer rates and DO distribution in the vessel. While calculating DO it has been assumed that the oxygen consumption rate (OUR) is uniform throughout the vessel.

Multiple simulations were carried to understand the impact of agitator design and operating condition on the DO distribution. Since the actual DO distribution cannot be experimentally verified, the absolute magnitudes of the results may not be accurate. However, it can be said that CFD models can be successfully used to predict DO distribution profiles in aerobic fermenters.