Dynamics of nonlinear multicomponent chromatography: Interplay of mass transfer, intrinsic sorption kinetics, and reaction
Abstract
A comprehensive framework has been developed for characterizing and scaling up complex biochemical separations. This framework consists of two facets: (1) a VErsatile Reaction-SEparation (VERSE) model which considers detailed mass transfer effects, equilibrium and non-equilibrium sorption, and solution and solid phase reactions and (2) a concise use of ratios of rates of mass transfer, sorption, and reaction to determine the controlling phenomena for a given separation. Detailed convergence and numerical stability studies have been conducted on the VERSE model. The VERSE model and associated dimensionless group approach has been successfully used to characterize a wide range of complex behavior in biochemical chromatography, including detailed mass transfer, slow sorption kinetics, and aggregation. Aggregation and slow sorption kinetics simulations were validated with experimental data from immobilized metal ion affinity chromatography of proteins and $\beta$-lactoglobulin A data from literature. Salt gradient elution of bovine serum albumin was well predicted by the VERSE model using a salt modulator isotherm. Procedures for estimating isotherm and mass transfer parameters from column pulse experiments are developed using a stage model with a linear driving force approximation for mass transfer. The estimated isotherm parameters are verified by comparison to those determined by batch equilibrium experiments. Several general principles may be inferred from the results. Changes in sorbent affinity can alter the amounts of each mass transfer resistance. The highest mass transfer resistance controls peak and front broadening. Once the slowest mass transfer rate (highest mass transfer resistance) has been determined, other rates, such as those of sorption and reaction, should be compared to that mass transfer rate. The slowest overall rate controls the characteristics of the system, such as peak splitting and merging. Once the controlling rates have been optimized to give the desired separation, scale up of the system can be guided by keeping those controlling dimensionless groups constant.
Degree
Ph.D.
Advisors
Wang, Purdue University.
Subject Area
Chemical engineering|Pharmacology|Biochemistry
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.