(297b) Flow inside Bioreactors: Comparing Predictions from Lattice-Boltzmann to Experimental Data

The lattice-Boltzmann method (LBM) provides an inherently time-dependent approach for modeling momentum transport through open and closed bioreactors. At low Reynolds numbers involving laminar and transitional flow, LBM can be used to perform direct numerical simulations that resolve the full spectrum of eddies inside the reactor. In turbulent systems, LBM can be paired with a sub-grid eddy viscosity model to perform large eddy simulations (LES) that resolve those eddies larger than the local grid spacing. These insights into flow turbulence have important implications on the design of bioreactors, as related to cell growth, product homogeneity, and processing time.

In this work, we examine agitated tanks and compare the mean and time-varying flow fields predicted from LBM to those obtained from PIV and LDA measurements. In regions both within and away from the trailing vortex, we find good quantitative agreement between the predictions from LBM and the experimental data. Next, we assess the effects of system resolution on the predicted flow field and the coupling between lattice spacing, energy dissipation, and the effects of the sub-grid filter. We then show how this technique can be used to optimize bioreactor performance, as characterized by blend time and energy dissipation.