Reverse micellar mass-transfer processes: Spray column extraction of lysozyme

Authors

  • Gary J. Lye,

    1. Biotechnology and Biochemical Engineering Group, Dept. of Food Science & Technology, University of Reading, Reading, RG6 6AP, U.K.
    Current affiliation:
    1. Department of Chemical Engineering, The University of Edinburgh, Kings Buildings, Mayfield Rd., Edinburgh EH9 3JL, U.K
    Search for more papers by this author
  • Juan A. Asenjo,

    1. Biotechnology and Biochemical Engineering Group, Dept. of Food Science & Technology, University of Reading, Reading, RG6 6AP, U.K.
    Search for more papers by this author
  • D. Leo Pyle

    Corresponding author
    1. Biotechnology and Biochemical Engineering Group, Dept. of Food Science & Technology, University of Reading, Reading, RG6 6AP, U.K.
    • Biotechnology and Biochemical Engineering Group, Dept. of Food Science & Technology, University of Reading, Reading, RG6 6AP, U.K.
    Search for more papers by this author

Abstract

Protein partitioning kinetics was measured for the semibatch extraction of lysozyme in a laboratory-scale, liquid–liquid spray column. The organic, isooctane phase contained reverse micelles formed from the anionic surfactant, sodium di-2-ethylhexyl sulfosuccinate. For the extraction of protein from aqueous to reverse micellar phases, experiments were performed over a range of dispersed-phase flow rates for cases of the organic- or aqueous-phase dispersion. The influence of aqueous-phase pH and ionic strength, which influence electrostatic interactions between protein and reverse micelles, was also investigated. Results were interpreted in terms of a two-film model of mass transfer. The nature of the dispersed pahse could significantly influence the partitioning kinetics, while study of the droplet hydrodynamics suggested that stagnant drops were formed regardless of which phase was dispersed. Literature correlations for describing the droplet-formation process and droplet hydrodynamics predicted measured values satisfactorily. Attempts wer also made to predict overall mass-transfer coefficients based on existing correlations describing mass transfer during droplet formation, free rise (or fall), and coalescene. Predicted values of KL were 2–10 times greater than measured values, probably because of large concentrations of surfactant used to formulate the reverse micelle phases. This approach did, however, provide detailed information on the quantity of protein transferred during the successive processes of droplet formation, free rise (or fall) and coalescence.

Ancillary