Font Size: a A A

Monitoring, modeling, and control of expanded-bed adsorption

Posted on:1999-02-25Degree:Ph.DType:Dissertation
University:University of Colorado at BoulderCandidate:Thelen, Travis VernonFull Text:PDF
GTID:1468390014973854Subject:Chemical Engineering
Abstract/Summary:
Expanded-bed adsorption (EBA) is a downstream processing method for the purification of proteins from cell-laden feedstocks. An expanded-bed presents the possibility of protein recovery in a single step, eliminating the often costly clarification processing steps such as centrifugation, ultrafiltration, and precipitation. However, an obstacle to the successful commercialization of EBA technology is the inability to accurately monitor and control the degree of expansion in these systems. A level measurement system and monitoring technique based on ultrasonics is presented. The use of this technique is demonstrated for bed-height measurement in EBA systems with unclarified and clarified feedstocks. The use of the ultrasonic technique for monitoring for the problematic conditions of plugging and bubbling is also demonstrated. The derivation and application of two-phase flow equations for solid-liquid fluidization in the Stokes flow regime is presented. Two specific flows are studied. The first considers changes in the fluidization velocity; the second looks at changes in the fluid properties. Simulation results are in excellent agreement with experimental results for the bed-expansion studies. Furthermore, the model provides an explanation for the observed time-delay phenomenon in bed expansion for step changes in fluid properties. Lastly, the monitoring and modeling strategies developed were used in the implementation of two model-based control strategies for bed-height regulation in EBA system, Internal Model Control (IMC) theory for tuning of a proportional-integral-derivative (PID) controller and linear quadratic regulator (LQR) theory with Kalman filtering for state estimation. The IMC-based PID controller and the LQR-Kalman filter controller tuned using the infinite-dispersion limit of the two-phase model were implemented on an experimental system. Both controllers were found to perform well, even in the presence of high protein loading and high cellular concentration.
Keywords/Search Tags:EBA, Monitoring, Model
Related items