Application of the Shock Layer Theory to the Determination of the Mass Transfer Rate Coefficient and Its Concentration Dependence for Proteins on Anion Exchange Columns

Authors

  • Peter Sajonz,

    1. Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996–1600
    2. Division of Analytical Chemistry, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831–6120
    3. Instrumentelle Analytik, Umweltanalytik, Universität des Saarlandes, D-66123 Saarbrücken, Germany
    Search for more papers by this author
  • Hong Guan-Sajonz,

    1. Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996–1600
    2. Division of Analytical Chemistry, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831–6120
    Search for more papers by this author
  • Guoming Zhong,

    1. Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996–1600
    2. Division of Analytical Chemistry, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831–6120
    Search for more papers by this author
  • Georges Guiochon

    Corresponding author
    1. Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996–1600
    2. Division of Analytical Chemistry, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831–6120
    • Department of Chemistry, University of Tennessee, Knoxville, Tennessee 37996–1600
    Search for more papers by this author

Abstract

The extension of the shock layer theory to systems having a slow mass transfer kinetics and a concentration-dependent rate coefficient is discussed. Experiments were carried out with bovine serum albumin on two anion exchangers, TSK-GEL-DEAE-5PW and Resource-Q. The adsorption isotherm data, determined by single-step frontal analysis, could be fitted to simplified bi-Langmuir equations with very small residuals. A lumped kinetic model (solid film linear driving force model, with rate coefficient kf) was used to account for the mass transfer kinetics. The profile of each breakthrough curve (BC) was fitted to the curve calculated with this transport model and the rate coefficient kf obtained by identification. A linear dependence of kf on the average concentration of the step of the BC was found. The shock layer thicknesses (SLT) calculated for different relative concentrations agreed very well with the experimental results. This justifies the use of the SLT for the direct determination of rate coefficients.

Ancillary