In situ product recovery: Submerged membranes vs. external loop membranes

Abstract

The proper membrane bioreactor design is crucial for ensuring efficient in situ product recovery in bioprocesses. Conventional membrane processes using external loops to separate liquid products from liquid media are problematic, because the cells/enzymes are unnecessarily subjected to shear stress, O2-deficiencies and fluctuations of the reaction process. Thus, this study reviews membrane processes for liquid–liquid separation based on submerged membranes compared to external loop membranes. Various designs of whole-cell bioreactors and of enzyme reactors are introduced, and relevant membrane process parameters are discussed such as anti-fouling and sterilization measures. For whole-cell systems, flux values of 1–70Lh−1m−2 and 30–80Lh−1m−2 were found for internal and external membrane processes, respectively. For enzyme systems, both the internal and external flux values ranged from 1 to 40Lh−1m−2. In conclusion, submerged membranes in whole-cell systems are inherently very good alternatives for sensitive mammalian cell systems. Having good flux values and requiring no extra pumps, these internal membranes rule out additional shear stress and allow tightly controlled reaction conditions. Regarding internal membrane design, ceramic rods are the most robust membranes in terms of material stability and filtration performance. Such a design as well as hollow-fiber polymer membranes have good potential for scale-up. For enzyme reactors, submerged membranes represent the best option for liquid–liquid separation. Since it is essential to control pH and temperature, these internal membranes inherently preserve enzyme activity. Ultimately, submerged membranes are very promising alternatives for product recovery in many whole-cell and enzyme systems.