Fractionation of Extracellular Polymeric Substances by Aqueous Three-Phase Partitioning Systems.
Evelyn C AntunesBruna CintraMatthieu BredelHardy TemminkBoelo SchuurPublished in: Industrial & engineering chemistry research (2024)
Extracellular polymeric substances (EPS) are natural polymers secreted by microorganisms and represent a key chemical for the development of a range of circular economy applications. The production of EPS comes with notable challenges such as downstream processing. In this work, a three-phase partitioning (TPP) system was investigated as a fractionation technique for the separation of polysaccharides and proteins, both present in the EPS culture broth. The effect of the type of phase-forming compounds (alcohol, polymer, or ionic liquid, in combination with salt) and its concentration were evaluated and compared to the results previously obtained with model systems. The recyclability of phase-forming compounds used to form the fractionation platform was assessed by ultrafiltration. The best fractionation of EPS was achieved using a TPP system composed of 23 wt % ethanol and 25% K 3 C 6 H 5 O 7 as 82% EPS-PS partitioned to the salt-rich/bottom phase, and 76% EPS-PN was recovered as an interfacial precipitate, which could be readily resolubilized in water. This represented an increase of 1.24 and 2.83-fold in the purity of EPS-PS and EPS-PN, respectively, in relation to the initial feed concentration. Finally, high recovery yields of phase-forming compounds (>99%) and fractionated EPS (>80%) were obtained using ultrafiltration/diafiltration (UF/DF) as the regeneration technique. The substantial fractionation yields, selectivity, and recyclability of the phase-forming compounds confirm the potential of TPP systems in combination with UF/DF as the separation method for real biopolymer mixtures. Key contributions of this study include the demonstration of the applicability of a readily scalable and cost-effective separation technique for the fractionation of EPS from real EPS-containing broths, while also the limitations of prescreening with model systems became clear through the observed deviating trends between model system studies and real broth studies.