(459f) The challenges of detailed CFDsimulations ofstirred lab-scalebioreactors

Reliable detailed CFD simulations of stirred lab-scale bioreactors are relevant (a) with the view of finding power numbers, as measuring torques at small scales is an issue, (b) because cell cultures usually start in mL vessels, or (c) in the context of designing down-scaled versions of larger-scale bioprocesses. These lab-scale vessels are usually provided with specific impeller types, a single baffle, an aeration device and one or more dip pipes. As a result, a vortex may be formed that interferes with baffle and dip pipes and which turns the flow chaotic, i.e. transient.

In addition, in RANS simulations, the domain is usually divided into two zones: an inner one that ‘rotates’ with the revolving impeller and an outer stationary zone. To this end, we have the disposal of two options: the steady-state Multiple Reference Frames (MRF) model and the transient Sliding Mesh (SM) approach. Defining such two zones in a small vessel with inserts is a challenge. The presentation will report on our findings when using MRF and SM and testing various grid types and grid densities, with a focus on gas hold-up, energy dissipation rates and spurious velocities.