Abstract
Suggestions are made for the practical implementation of membrane cascades using diafiltration for the fractionation of solute pairs. Experiments are described that demonstrate the desirability of replacing solvent during the course of each diafiltration, and a parallel modeling development suggests an attractive means for accomplishing this replacement. A batch process is described to achieve such separations by simple assemblies of existing equipment, and suggestions are made for designing continuous processors. Such cascades are attractive for a wide variety of solutes including native proteins, as well as commodity chemicals, and they can be applied to the resolution of enantiomers through simple modifications already described in the public literature. The same techniques can be applied to multicomponent systems using the concept of key components as has long been done in distillation.
The low inherent capital costs and high throughput rates of such membrane cascades strongly suggest that they should compete successfully against a significant number of presently used chromatographic processes, and their simplicity should make then formidable competitors to simulated moving beds as well.
Acknowledgement
The work represented here could not have been done without the generous support from the Wisconsin Alumni Research Foundation and from many donors over the years who contributed to a fund administered for the senior author by the University of Wisconsin Foundation. This fund, carefully husbanded over more than a half century, was greatly increased by the wise financial management of this foundation. The senior author also greatly appreciates being permitted to use laboratories of the Department of Chemical and Biological Engineering a full decade after retirement. Finally, we all thank Prof. T. W. Root for his occasional wise counsel.