A framework to predict and experimentally evaluate polymer-solute thermodynamic affinity for two-phase partitioning bioreactor (TPPB) applications

Abstract

BACKGROUND: Selection of a polymer for two-phase partitioning bioreactor (TPPB) applications has previously been limited to heuristic approaches. However, recent interest has focused on first principles' selection methods based on polymer crystallinity, glass transition temperature and polymer-solute thermodynamic affinity. In this work, a framework is proposed to evaluate and predict polymer-solute thermodynamic affinity via the polymer-phase activity coefficient. RESULTS: Polymer screening via thermodynamic affinity was shown to be most effective at very dilute concentrations, where partition coefficients can be estimated using infinite dilution activity coefficients. In the absence of published values, UNIFAC-vdW-FV or Flory-Huggins based activity models can provide very good predictions for the polymer-phase activity coefficient, significantly improving upon previous approaches using Hildebrand and Hansen solubility parameter differences. For non-dilute systems, however, the activity models failed to consider the full effects of concentration on partition coefficient. Additionally, a reduction in polymer molecular weight resulted in improved partition coefficients, a phenomena well described by the activity models. CONCLUSION: Predicting and experimentally quantifying polymer-solute thermodynamic affinity at very dilute concentrations will aid future attempts at TPPB polymer selection. Furthermore, experimental partition coefficient data at a range of operational concentrations will indicate how TPPB effectiveness will change throughout the fermentation course. Finally, reduction of polymer molecular weight to improve solute partitioning should be investigated further for a range of polymers. © 2014 Society of Chemical Industry.