Gaurav Agrawal

Comparison of various ternary simulated moving bed separation schemes by multi-objective optimization

Over the past decade, many modifications have been proposed in simulated moving bed (SMB) chromatography in order to effectively separate a binary mixture. However, the separation of multi-component mixtures using SMB is still one of the major challenges. In addition, the performance of SMB system highly depends on its operating conditions. Our study address this issue by formulating a multi-objective optimization problem that maximizes the productivity and purity of intermediate eluting component at the same time. A number of optimized isocractic ternary SMB operating schemes are compared both in terms of productivity and amount of desorbent to feed ratio. Furthermore, we propose a generalized full cycle (GFC) formulation based on superstructure formulation encompassing numerous operating schemes proposed in the literature. We also demonstrate that this approach has a potential to find the best ternary separation strategy among various alternatives.

Systematic optimization and experimental validation of ternary simulated moving bed chromatography systems.

Over the past several decades, many modifications have been proposed in SMB chromatography in order to effectively separate a binary mixture. However, the separation of a multi-component mixture using SMB is still one of the major challenges. Recently, a computational study was performed which compared various existing isocratic ternary separation operating schemes (including the JO process) in terms of the maximum throughput attained, and Generalized Full Cycle strategy was proposed based on a systematic design, which was found to have significant improvement over existing strategies [Agrawal and Kawajiri (2012)]. Nevertheless, the operating strategies were not experimentally validated. In this study, we validate both JO and Generalized Full Cycle SMB systems experimentally. A simultaneous optimization and model correction scheme has been implemented to arrive at the optimal operating condition which satisfies the optimal productivity as well as the desired purity and recovery of products experimentally.

Optimization of reactive simulated moving bed systems with modulation of feed concentration for production of glycol ether ester

In this article, we extend the simulated moving bed reactor (SMBR) mode of operation to the production of propylene glycol methyl ether acetate (DOWANOL™ PMA glycol ether) through the esterification of 1-methoxy-2-propanol (DOWANOL™ PM glycol ether) and acetic acid using AMBERLYST™ 15 as a catalyst and adsorbent. In addition, for the first time, we integrate the concept of modulation of the feed concentration (ModiCon) to SMBR operation. The performance of the conventional (constant feed) and ModiCon operation modes of SMBR are analyzed and compared. The SMBR processes are designed using a model based on a multi-objective optimization approach, where a transport dispersive model with a linear driving force for the adsorption rate has been used for modeling the SMBR system. The adsorption equilibrium and kinetics parameters are estimated from the batch and single column injection experiments by the inverse method. The multiple objectives are to maximize the production rate of DOWANOL™ PMA glycol ether, maximize the conversion of the esterification reaction and minimize the consumption of DOWANOL™ PM glycol ether which also acts as the desorbent in the chromatographic separation. It is shown that ModiCon achieves a higher productivity by 12-36% over the conventional operation with higher product purity and recovery.