The increasing demand for optically pure compounds (enantiomers) stimulates the development of new enantiomer separation processes on an industrial scale. The separation of enantiomers by ultrafiltration of enantioselective micelles was studied in a cascaded system. The feasibility of this separation concept was proven by a cascade of five lab-scale units and a bench-scale system using an industrial membrane module. Model calculations show that the separation is not improved greatly at enantioselectivities (ratio of affinity constants) above 10 and the affinity of the enantioselective micelles for the substrate (enantiomers) plays a crucial role in the performance of the separation process. Modeling the separation as a function of the dimensionless affinity number (affinity constant (.) enantiomer feed concentration) shows that there is an operation window in which the separation satisfies its criterion of high yield and high enantiomeric excess. For countercurrent affinity systems, this implies that any research effort should focus on developing (microheterogeneous) media characterized by weak interactions with the substrate to allow for operation at high feed concentrations, not on high affinity systems.