The oxygen mass transfer coefficient, kLa, is a critical parameter in bioprocess performance, influencing the design of stirred tank reactors. During a fermentation process, the behavior of bubbles and the kLa can be effected by various components present in the broth. This study
...
The oxygen mass transfer coefficient, kLa, is a critical parameter in bioprocess performance, influencing the design of stirred tank reactors. During a fermentation process, the behavior of bubbles and the kLa can be effected by various components present in the broth. This study investigates the effect of organic compounds (ethanol, glycerol and acetic acid), which are produced during a yeast fermentation process (S. cerevisiae). It is concluded that acetic acid appears to decrease the geometric mean bubble diameter, in aqueous solutions. As a surfactant, the acetic acid molecules accumulate in the gas-liquid interface, thereby inhibiting bubble coalescence. This leads to an increase in interfacial area and a 34% rise in kLa has been measured. Glycerol, however, showed no significant impact in water. Ethanol should exhibit a similar trend to acetic acid, as reported in previous literature (Puiman, Elisiário, et al., 2022). Remarkably, when these organic compounds were added to growth medium for a yeast fermentation - which consist of synthetic medium, glucose and vitamins - the effects were negligible. This likely due to the coalescence-enhancing properties of antifoam, which is present in synthetic media. On the other hand, the concentrations of the organic compounds tested were relatively low. In addition, during the fermentation process, the production of organic compounds did not significantly affect the bubble size. A slight increase in kLa was observed, arguably due to a reduction in working volume. Whilst performing the tests, the kLa was measuring with three different experimental determination methods: the dynamic pressure method, the dynamic gassing-out method, and the gaseous oxygen balance method. The dynamic gassing-out method appeared a bit more consistent than the dynamic pressure method, possibly due to pressure stabilization issues. The gaseous oxygen balance method did not provide consistent results throughout the entire fermentation, as the method is sensitive and relies on very accurate gas measurements, especially when the oxygen consumption is low. The dynamic gassing-out method proved to be stable and provided the most results used for comparisons of experiment. However, considering its limitations, it requires validation with a ground truth, such as a validated reliable chemical method. The experimental kLa data was compared with predicted values, using empirical correlations. All predicted values remained within 50% margin to the experimental values (determined with dynamic gassing-out method). To be able to use these prediction models in bioprocess design, further investigation is essential.