Cell-culture bioreactors lie at the heart of the processes used to produce large-molecule, protein-based therapeutics. In cell culture, mammalian cells are grown outside the human/plant body. These cells produce therapeutic proteins and antibodies. This is much easier said than done. In fact, cells do not cooperate much when they are grown outside the (human or plant) body. The question then is: Why is it so difficult for cells in culture to have the same physiological function in laboratory as in our body?
Actually in the lab, you can be providing cells a perfectly sterile environment, utilizing high end laboratory cell culture equipment and have supreme cell culture reagents, but your cells are still not happy? The answer is simple: working environment and oxygen levels. Cells become highly selective with respect to the working environment when grown outside the body. Oxygen availability is the key parameter.
Following that comes balancing the pH, stripping off CO2, removing the by-products and keeping an amiable temperature — all this needs to be done with perfect precision. Another dimension: These environment needs are not identical for all kind of cell-cultures — each type has their own unique requirement. Keeping several parameters to a specified level in a large bio-reactor is the challenge of process engineers. And, they indeed need to put a great deal of effort to keep the cells happy.
In this video, you can see how ANSYS Computational Fluid Dynamics (CFD) is used to predict the blending time and exposure time for mixing tanks and bioreactors.
Computational fluid dynamics helps co-relating the aspects of a perfect environment for cells with design parameters like impeller type, sparge rate, agitation rate, tank configuration, placement of equipment’s etc. It’s like a “virtual lab” where one can carry out many numerical experiments to complement real-life experiments. This combination of numerical and real-life experiments helps in reducing the cycle of iterations in arriving at the final operating condition for a given production bioreactor.
There are several stages of simulation in bioreactor modeling — primarily studying flow and mixing time or blend time. Blend time knowledge helps in estimating the response time of the pH controller. The next level of simulation is to calculate the gas distribution inside tank. An advanced analysis would include oxygen dissolution analysis and estimating exact mass transfer co-efficient.
In closing, for more information, my colleague Gilles Eggenspieler created a CFDTechTip about customized software called “mixing template”. Mixing template makes it easy for design engineer without CFD background to carry out all primary stage simulations. Advanced levels of analysis can be added manually to include calculation of mass-transfer co-efficient. All these simulations greatly enhance the understanding of the workings of a bio-reactor. Results from the simulation and real-life experiment greatly greatly accelerates the search for ‘pleasant operating conditions’ to keep the micro-organisms happy.