Predicting gas-liquid mass transfer rates in reactors using a bubble parcel model

Abstract

A parcel-based approach for predicting mass transfer in gassed bioreactors is presented. Predictions from this parcel model, which handles bubble break-up and coalescence implicitly, are compared directly to measured data and predictions from full-fidelity discrete bubble models. Guidance for determining when this coarse-graining approach is acceptable, as a function of the vessel operating conditions, is presented. When the approach is appropriate, the parcel-based model can present order-of-magnitude reductions in computational runtime with minimal loss in physical fidelity.