Designing Isothermal Batch Deracemization Processes With Optimal Productivity: I. Parametric Analysis Using A Population Balance Equation Model

Thomas Vetter

Research output: Contribution to journalArticlepeer-review

89 Downloads (Pure)

Abstract

Isothermal batch deracemization (Viedma ripening) is an attractive process variant to separate conglomerate forming enantiomers. These processes rely on a complex interplay between a racemization reaction in the liquid phase and crystal growth/dissolution, agglomeration and breakage in the solid phase. While reports in the literature have shown the applicability of Viedma ripening to a variety of substances, a comprehensive investigation on how to obtain maximum productivity from such processes is so far missing. This contribution introduces a novel operating protocol based on a series of batches, wherein part of the product of one batch is used to generate an initial solid phase enantiomeric excess in the next batch. It is shown that initial enantiomeric excess leading to maximum productivity depends on the kinetics involved in Viedma ripening, as well as process parameters, such as the suspension density. This relationship is explored using a parametric sensitivity analysis carried out on a process model based on dimensionless population balance equations. The general trends identified from the parametric analysis highlight that processes with maximum productivity should be carried out at i) high breakage intensities, ii) low agglomeration intensities, and iii) high suspension densities. The initial enantiomeric excess necessary to reach maximum productivity varies strongly with the kinetics of the different phenomena involved and falls within a wide range of 25–80%.
Original languageEnglish
JournalCrystal Growth & Design
Early online date7 Apr 2020
DOIs
Publication statusPublished - 2020

Fingerprint

Dive into the research topics of 'Designing Isothermal Batch Deracemization Processes With Optimal Productivity: I. Parametric Analysis Using A Population Balance Equation Model'. Together they form a unique fingerprint.

Cite this