Oxygen transfer in small scale animal cell culture reactors: comparison of two reactors by experimental and numerical methods.
Integration of life sciences & engineering
Integration of Life Sciences & Engineering - Poster (T5-P)
Keywords: bioengineering, animal cell culture, oxygen transfer, CFD, VOF
Nowadays, animal cells such as Chinese Hamster Ovary cells are widely used as biocatalysts for the production of therapeutic molecules (monoclonal antibodies, recombinant proteins). Large-scale culture of these cells is performed in bioreactors whose volume keeps increasing. Shear stress distributions and volumetric oxygen transfer coefficient (kLa) are two key parameters for a successful scale-up of these reactors. This scale-up is complex as these parameters are strongly coupled (an increase in agitation rate promotes oxygen transfer but may be lethal to the cells). Moreover only few studies dedicated to true cell culture reactors and none to oxygen transfer in spinner flasks are found in the literature. One of our previous studies (Barbouche et al., 2007) has shown the positive effects of hydrodynamic stress on the quantity of cells produced, as long as these ones remain lower than a critical value. Nevertheless, it is necessary to quantify the contribution of the simultaneous improvement of the oxygen transfer to clearly establish a correlation between cell behaviour and reactor hydrodynamics. To do this, classic experimental methods of kLa measurements have been combined with numerical simulations for the prediction of the kLa evolution with the agitation rate and with the culture system used.
Two reactors have been studied. A 250 mL spinner flask and a 1.4 L agitated and sparged bioreactor. For the first one, a surface aeration is encountered; for the second system, both surface aeration and gas sparging coexist. The liquids studied are PBS and PF-BDM culture medium. No oxygen probe can be placed in the spinner flask so the kLa is measured by the sulphites method. In the reactor, a Mettler-Toledo oxygen probe allows the use of the dynamic method. The agitation rates respectively vary from 40 to 300 rpm and from 80 to 1000 rpm. When only surface aeration is encountered, Computational Fluid Dynamics (CFD) consisting in a Volume Of Fluid approach (VOF) are used to calculate precisely the evolution of the free surface shape when the agitation rates are increased. As no gas sparging is considered in the calculations, the Euler-Euler approach is not provided. Shape deformations, and consequently the interfacial area a, are compared and linked to kLa experimental results. Moreover, the simulations provide the volumetric and surface turbulent energy dissipation rate, which are helpful for the knowledge of surface and bulk turbulence.
Concerning the spinner flask, our results show that, under a critical agitation rate of 160 rpm approximately, no changes of kLa are observed. Beyond this value, the free surface becomes wavy and a central vortex exists. The oxygen transfer is then appreciably improved. This is confirmed by our VOF numerical simulations which precisely predict this transition of the free surface from a quiet to a wavier aspect. For the reactor, gas sparging efficiency can rapidly be surpassed by surface aeration if agitation rates become higher than 150 rpm. Indeed, with these conditions, oxygen bubbles are transferred to the liquid bulk by surface turbulence. Once again, our VOF simulations allow the predictions of this transition from a non-interpenetrating two-phase flow to a bubbly gas-liquid flow.
At last, our study establishes an original and generalized correlation for surface aeration which links the volumetric oxygen transfer coefficient to the turbulent energy dissipation rate whatever the reactor design. This correlation will be useful for a better knowledge and a better description of the complex interactions between cell physiology and reactor hydrodynamics.
* Barbouche et al. (2007). Coupling between cell kinetics and CFD to establish a physio-hydrodynamic correlation in various stirred culture systems. 20th meeting of the European Society of Animal Cell Technology, June 17-20, Dresden, Germany.
Presented Wednesday 19, 13:30 to 15:00, in session Integration of Life Sciences & Engineering - Poster (T5-P).
