[elFittich]

Hi everyone,

at the moment I am trying to track massless particles in a stirred tank to analyze the mixing process (poincare-section, stretching etc.). To reduce the calculation time, I am using the frozen flow field (based on MRF/SM).
However, the particles will always stick to the wall of the impeller after a certain time. This is reasonable, as the velocity on the impeller wall is zero.

So, is there a way to perform massless particle tracking on a frozen flow field in stirred tanks anyway?

Any comments and suggestions are appreciated!

Best regards!

[CeesH]

Yes, you can cheat your way around it, but I would not recommend it if you want accurate distributions. I did something similar for Lagrangian kinetic models for biomass, but there a more or less homogeneous distribution sufficed - no need for 100% accurate distributions.

What I did was:
- model the impeller as a 2D sheet body, not as a full 3D model!
- First solve the flow with the impeller as a wall, no particles
- once you freeze the flowfield, switch the impeller b.c. to interior.

See e.g. Haringa et al. 2016 Engineering in life science or 2017 chem. eng. sci for more details

The particles will now flow through. Do the same for the baffles.
Be wary when using turbulence, the discrete random walk model may lead to particle clustering in low-turbulence zones (corners at the reactor top for example). If you can implement continuous random walk or so, that would be better in the long run (never had time to do so unfortunately)

[elFittich]

Thank you very much, I will try this approach. Have you tried to patch the velocity in the boundary layers to cause a slipping of the particles?