(459b) Using CFD to Predict Mixing Coefficients for Scale up

Scaling a process or operation is an important concept in industry. Operating conditions such as agitation rate and gas sparging rate do not produce the same output at different scales. Mixing profiles can vary depending on the scale, which can have a significant impact on process phenomena such as heat, mass transfer, crystallization, and reaction kinetics. Understanding mixing is therefore imperative to understand the process. Mixing can be characterized on three different scales, macro, meso, and micromixing, which correspond to the bulk liquid, droplet, and molecular scales, respectively. The equations describing mixing at these scales can be predicted from mixing time engineering correlations and specific scale-up parameters.

Computational fluid dynamics (CFD) is a useful tool that can be used to provide process insight that experimental analysis cannot. The degree of detail obtained includes velocity distributions, energy dissipation, and macro mixing times which can then be used to predict methods of enhancing mixing efficiency. Vessel-specific mixing coefficients can be regressed for macro, meso, and micro mixing times using the velocity and energy dissipation data from the simulations. Mixing timescales can therefore be predicted for different operating conditions whilst reducing physical experimentation.

Literature values for the mixing coefficients exist however they are for specific reactor-impeller configurations. To date, there are a lack of studies conducted to generate a list of these coefficients and how they vary based on the different reactor variables such as vessel size, impeller type, and the number of baffles. The researcher is therefore forced to choose a coefficient generated in laboratory studies that most closely resemble their system. This introduces inherent error into the calculation and can drastically reduce the accuracy of any mixing time correlation output. For example, the meso-mixing time correlation requires a coefficient that relates the local turbulent eddy dissipation to the power per unit mass. The use of CFD provides an exact measurement of the required coefficient which can be obtained for the bespoke reactor or process under investigation. This improves accuracy and can lead to less experimentation, improved process understanding, and better scale up.