A conventional anion exchange column packed with porous particles (BioScale
Q2), and a novel continuous-bed column (UNO(TM) Q1) were compared for disp
lacement separation of dairy whey proteins with polyacrylic acid as displac
er. The steric mass action model was investigated as a means to aid and acc
elerate this development. Characteristic charges and steric factors were me
asured for the proteins and the displacer according to the model, and used
together with the affinity constant derived from the adsorption isotherms f
or simulations, as well as for the construction of the affinity and operati
ng regime plots. If possible. the latter two were used to select conditions
for the actual experiments. In the case of the particle-based column, expe
rimental results and simulations did not agree. In addition, the operating
regime plot could not be constructed. The affinity plot did predict the ord
er in the displacement train correctly, but gave misleading information con
cerning the possible effect of a change in displacer concentration. This is
taken to be a result of the porous nature of the particles, which handicap
s, to some extent, the interaction of the proteins and the displacer molecu
les with the adsorptive surface. Results were considerably better in case o
f the continuous-bed column, where there is no intraparticulate surface. (C
) 2000 Elsevier Science B.V. All rights reserved.