Premium
The reactive extraction of phenylalanine with aliquat 336: Buffer co‐extraction equilibrium and mass transfer kinetics
Author(s) -
Scarpello Justin T.,
Stuckey David C.
Publication year - 2000
Publication title -
biotechnology and bioengineering
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.136
H-Index - 189
eISSN - 1097-0290
pISSN - 0006-3592
DOI - 10.1002/1097-0290(20000905)69:5<469::aid-bit1>3.0.co;2-m
Subject(s) - aliquat 336 , chemistry , extraction (chemistry) , mass transfer , equilibrium constant , mass transfer coefficient , aqueous solution , chromatography , aqueous two phase system , analytical chemistry (journal) , inorganic chemistry , solvent extraction
The occurrence of significant co‐extraction of buffer anions by the ion exchanger Aliquat 336 is unavoidable when high levels of system buffering is required. The co‐extraction will result in inaccurate equilibrium and mass‐transfer characterization of such a system unless its occurrence is taken into account, making process design and control difficult. A study of the equilibrium of phenylalanine extraction using Aliquat 336, a system where high levels of hydroxyl co‐extraction occurs, was used as a model case to develop a method of accounting for co‐extraction in mass‐transfer modeling. Analysis of the equilibrium between bulk‐aqueous‐phase chloride and phenylalanine concentrations during mass transfer in a stirred‐transfer cell showed there to be linear equilibrium relationships between the two parameters for a given extraction system of the form C Cl,t = α( C A,t − C A, 0 ) for forward extraction and C Cl,t = ϵ C A,t + C Cl, 0 for backward extraction. The constants of proportionality of these relationships, or the “co‐extraction constants,” α and ϵ, were shown to be related to the selectivity of Aliquat 336 for the phenylalanine anion by the relationships α = −(1/ S + 1) and ϵ; = −(1/ S −1 + 1). The linear equilibrium relationships were used to develop two‐film theory mass‐transfer models for both forward and backward extraction that account for co‐extraction. These showed much higher accuracy in modeling stirred‐transfer‐cell data than the equivalent models which ignored co‐extraction. © 2000 John Wiley & Sons, Inc. Biotechnol Bioeng 69: 469–477, 2000.