(657c) Constant-Pattern Design of Displacement Chromatography | AIChE

(657c) Constant-Pattern Design of Displacement Chromatography


Harvey, D. M. - Presenter, Purdue University
Choi, H., Purdue University
Wang, N. H. L., Purdue University
Displacement chromatography has been used in a variety of applications, especially those related to the biopharmaceutical industry. Compared to traditional elution chromatography, displacement chromatography can deliver products at higher concentrations, higher purity, and higher yield, which makes it an attractive option in many separation systems. However, the design of displacement chromatography systems is a complex problem that requires the specification of many design parameters. While previous work has been done to determine the column length, and loading required to form an isotachic train in ideal systems, it is still unknown when the mixed band regions between components reach a constant pattern. Additionally, the design and optimization of displacement systems is done using trial and error, requiring many simulations or experiments.

In this study, a general method was developed to design displacement systems to obtain high purity products with high yield and high productivity. The number of design parameters were greatly reduced through dimensionless groups analysis. Then, rate model simulations were used to examine the effects of the dimensionless groups to determine the conditions which are required to reach the constant pattern state. An expression for yield as a function of dimensionless variables was derived as well. The yield and productivity are greatly affected by the selectivity of the components, overall mass transfer coefficient, feed composition, loading fraction, volume of feed loaded, and the displacer concentration. Given a target yield, displacer concentration, feed volume and composition, as well as column properties, this method can be used to determine the necessary column length and operating flow rate to reach the target yield. Because the constant-pattern design method utilizes dimensionless groups, the results are easily scalable while maintaining high purity, high yield, and high productivity.