273868 Minimum-Column-Volume Design of Capture Chromatography for Mo Recovery

Monday, October 29, 2012: 5:00 PM
405 (Convention Center )
Lei Ling1, Pei-Lun Chung1, Amanda Youker2, Dominique C. Stepinski2, George F. Vandegrift2 and N.-H. Linda Wang1, (1)School of Chemical Engineering, Purdue University, West Lafayette, IN, (2)Argonne National Laboratory, Argonne, IL

Capture chromatography is widely used for many different applications. An adsorption column is used to recover a target component from a finite batch of feed solution. In designing this process, one needs to specify the column size and feasible operating velocities that can satisfy a column pressure drop limit and a dimensionless breakthrough concentration, which is the ratio of breakthrough concentration to feed concentration at the end of loading. A low dimensionless breakthrough concentration (0.01) gives a high capture yield (>99%). A systematic design method based on key dimensionless groups has been developed for Langmuir adsorption isotherm systems. For a fixed loading time, the dimensionless column length is inversely proportional to the loading factor, which is calculated from the feed concentration, feed volume, column volume, and the Langmuir isotherm parameters. This hyperbolic curve is defined as the operating line. Along an operating line, the loading factor and dimensionless breakthrough concentration increase when dimensionless column length decreases. On the other hand, the minimum dimensionless column length needed for a fixed dimensionless breakthrough concentration was found to increase with increasing loading factor. The curve that shows the relation between the two dimensionless groups is defined as the constant breakthrough limit curve, which can be obtained either from experiments or rate model simulations based on intrinsic adsorption and mass transfer parameters. The minimum column volume for capture was found from the intersection point of the operating line with the constant breakthrough limit curve. If intra-particle diffusion controls wave spreading, and feed concentration is within a linear isotherm region, the constant breakthrough limit curve is only a function of the linear isotherm parameter of a given system. The maximum linear velocity and the corresponding column length that satisfy the desired pressure limit and breakthrough limit can be found from the minimum column volume.

This design method was verified with Mo recovery from an aqueous uranyl sulfate solution. A Titania sorbent, which was found to have high affinity for Mo, was tested in this study. The adsorption isotherm was obtained from batch equilibrium tests at 60 ºC. The isotherm data were closely correlated using the Langmuir isotherm equation. Mo diffusivity was estimated by matching the experimental frontal curves with simulated frontal curves obtained from rate model simulations. The minimum column volume for capture was verified with breakthrough data from lab scale columns. A sensitivity analysis was carried out to find a safety factor in column volume to take into account potential variations of material properties and errors in the estimated intrinsic parameters. The columns designed based on the estimated safety factor were tested for Mo recovery. The results showed that Mo capture yield was nearly 100% during loading, and average stripping yield was higher than 96%.

This design method requires a small number of experiments, and can achieve the maximum sorbent productivity at a high capture yield. If bed void, particle size, feed concentration, feed volume, loading time, or loading velocity is varied, the design can be readily modified without further simulations or experiments. This method can also be applied to the design of capture columns for other applications.

Extended Abstract: File Uploaded