Predictive modeling and experimental implementation of organic acids in stream recovery by reactive extraction in membrane contactors
Abstract
Membrane-based reactive extraction as an in situ product recovery technique is a promising strategy for process intensification, in particular in the case of the bioproduction of organic acids. Reactive extraction allows a high selectivity for the extraction of the targeted acid and the microporous membrane keeps biocatalysts in the aqueous broth while implementing a large liquid–liquid surface area and ensuring a dispersion-free contact, without problems of emulsion formation. This paper deals specifically with the extraction of biobased 3-hydroxypropionic acid using tri-n-octylamine in n-decanol. In order to maintain an effective driving force for 3-HP transfer into the organic phase, this latter was continuously regenerated by recovering the acid in a back-extraction aqueous phase, giving a complete pertraction process. A mass transfer model for this process was developed. It is based on the boundary layer theory and takes into account chemical and physical equilibria of complexation/dissociation and partitioning, species diffusion in the membrane pores and viscosity variations in the organic phase. Viscosity highly depends on acid concentration, increasing up to 50% when 3-HP concentration reaches 28 g.L-1. Thus, it was possible to predict different experimental results with R2 ≥ 0.99, totally neglecting chemical kinetics and interfacial resistance for both extraction and back-extraction steps. The model allows the prediction of extraction kinetics with (1) fixed initial concentrations and (2) with gradual 3-HP feed (mimicking a bioconversion) in transient and steady states coupled with back-extraction (globally also called pertraction). Model based analysis of mass transfer mechanisms led to the construction of a nomogram giving 3-HP stationary concentration in the case of a typical production rate, enabling for example a rapid organic phase selection or membrane sizing.
Fichier principal
Postprint Chemarin Athes 2022 CEJ (Modelling and experimental acid recovery).pdf (1.9 Mo)
Télécharger le fichier
Origin : Files produced by the author(s)