Introduction
Oligodextran is a dextran derivative with a relatively low molecular weight (Mw) (< 70 kDa), which has been widely used as a prebiotic in food and as an antiphlogistic and antioxidant in pharmaceutical applications 1,2. Production of oligodextran through enzymatic hydrolysis of industrial grade dextran is a safe and efficient process that minimizes secondary pollution (e.g. the generation of large amounts of organic wastewater in traditional acid hydrolysis) 3,4. The combination of ultrafiltration (UF) with an enzymatic reactor, however, could purify the resulting oligodextran and narrow its Mw distribution which in turn would enhance oligodextran bioactivity 5. Compared with the traditional enzymatic reactor and membrane separation hybrid process, which are performed separately, an enzymatic membrane reactor (EMR) that simultaneously performs enzymatic reaction and membrane separation enables enzyme reuse, reduces product inhibition, and avoids excessive hydrolysis of target products 6. Several researchers have investigated the feasibility of EMR for oligodextran production 7-10. For example, Torras et al.7 successfully fabricated a UF membrane loaded with dextranase for manufacturing oligodextrans. The Mw of oligodextran can be controlled by the loading amount of enzyme 7 and the molecular weight cut-off (MWCO) of the membrane 8. In our own previous study 9,10 the membrane fouling that resulted from the enzyme was found to enable narrowing of membrane pore size distribution while oligodextran products of uniform Mw distribution were obtained.
In spite of extensive experimental investigation of EMR performance, few efforts have been devoted to understanding the interaction between enzymatic reaction and membrane filtration, yet these are critical for the performance of the overall EMR process. Establishment of a reliable model, which describes the transport phenomena of oligodextran and enzymatic hydrolysis reactions, enables us to systematically investigate the influences of operating conditions and membrane properties on the performance of EMR and to optimize reactor design and membrane fabrication. However, most of previous studies dedicated to modelling EMR have focused on kinetic modeling and have considered only mass-transfer mechanisms 11-13 in which the membrane only serves as a barrier or carrier of enzymes. In these studies the effect of membrane sieving on the transport mechanisms of the product has therefore not been taken into account. This is a critical knowledge gap that needs to be systematically addressed to guide the optimization of EMR for oligodextran production because the membrane also functions as a selective barrier in tailoring the molecular weight of the product. To the best of our knowledge no mathematical model regarding the EMR for oligodextran production has so far been reported.
The aim of the current study was therefore to develop a mathematical model to characterize the transport phenomena in EMR and predict EMR oligodextran production performance on the basis of the kinetics of dextran hydrolysis, the force balance model, and the stagnant film model as an approach for achieving process optimization. Our model evaluates membrane properties (pore size and porosity) and operating conditions (operating pressure and agitation speed) in relation to filtration performance. Furthermore, feeding modes (water feeding vs. substrate feeding) were also explored. Our work elaborates the inherent relationship between the operating parameters of EMR and its production efficiency, which can provide important insights into fundamental understanding and designing of EMR systems.