No abstract

No abstract

Liquid-phase mass transfer coefficient in bioreactors have been examined. A theoretical model based on the surface renewal concept has been developed. The predicted liquid-phase mass transfer coefficients are compared with the experimental data for a mycelial fermentation broth (Chaetomium cellulolyticum) and model media (carboxymethyl cellulose) in a bench-scale bubble column reactor. The liquid-phase mass transfer coefficient is evaluated by dividing the volumetric mass transfer coefficient obtained experimentally by the specific surface area estimated using the available correlations. The available literature data in bubble column and stirred tank bioreactors is also used to test the validity of the proposed model. A reasonable agreement between the model and the experimental data is found.

Oxygen transfer rates and gas holdups were measured in mycelial fermentation broths of Chaetomium cellulolyticum and Neurospora sitophila, each cultured in a 1300-L pilot-plant-scale airlift fermentor. These cultures exhibited highly non-Newtonian flow behavior coupled with a substantial decrease in oxygen transfer rates. The volumetric mass transfer coefficients in these cultures were found to be 65-70% lower than those in water. The data were compared with the available correlations obtained for simulated fermentation broths. In general, the data for C. cellulolyticum are in satisfactory agreement with the correlations for the model media but the data for N. sitophila are higher than that predicted by the correlations. Model media based correlations are found to be applicable to the fermentation processes if the culture medium does not possess a high yield stress.

The dynamic gassing-out method using nitrogen for gas-liquid K(L)a measurements has been modified so that gassing-out is performed with air and gassing-in with oxygen-enriched air. This new method was proven theoretically valid for use in inert model systems and in actual fermentation systems. The K(L)a values were measured in a 1-m-high bubble column and compared with those obtained from the traditional gassing-in method for three different mixing models in batchwise contacting: mixed gas and liquid (MMB); plug-flow gas and mixed liquid (PMB); and plug-flow gas and plug-flow liquid (PPB). The K(L)a values obtained from the new method were consistent with those of the method of Bartholomew et al. Discrimination between the models appeared to be not important for the 1-m-high column, but the theoretical analysis revealed that it would become necessary for columns about 10 m and higher.