Separation of enantiomers of a chiral epoxide by simulated moving bed chromatography

Abstract

The paper deals with chiral separation by simulated moving bed (SMB) chromatography. The separation of chiral epoxide enantiomers in microcrystalline cellulose triacetate using methanol as eluent is considered as illustrative example. The use of microcrystalline cellulose triacetate for the chromatographic separation of enantiomers is first reviewed and a methodology for obtaining basic data (adsorption equilibrium isotherms, axial dispersion and mass transfer coefficient) is discussed. A model for the prediction of the cyclic steady-state performance of the SMB, based on the analogy with the true moving bed, is developed assuming multicomponent adsorption equilibria, axial dispersion flow and linear driving force approximation to describe the intraparticle mass transfer rate. The simulation package is used to predict the effect of operating variables on the process performance and to define the regions for enantiomer separation. A simple optimization procedure is proposed for choosing the best SMB operating conditions. This procedure is extensively tested for the separation of chiral epoxide enantiomers. The experimental operation of a SMB pilot unit was carried out for this system. Purities and recoveries higher than 90% were obtained for both extract and raffinate, using a 420 ml inventory of stationary phase. The SMB pilot allows the continuous resolution of 52 g of racemic mixture per day and per liter of bed, with a solvent consumption of 0.4 l of mobile phase per gram of racemic mixture processed. The simulation package is also used to predict the steady-state internal concentration profiles for the SMB operation with reasonable agreement with experimental results.

Chiral separation by SMB chromatography

“…Experimentally, a raffinate purity close to 100% (PUR =99.6%) was obtained, but the extract purity was a little lower (PUX =97.5%). This is a new experimental result which improves rather interestingly the purity performance reported for the same system in a previous work [17]. This improvement was achieved by using the right operating conditions following the simulation results obtained with the TMB model.…”

“…2000, and k = 0.4 s − 1 [17]. Experimentally, a raffinate purity close to 100% (PUR =99.6%) was obtained, but the extract purity was a little lower (PUX =97.5%).…”

“…Open squares show the path of equal purities for extract and raffinate. Since the objective of the SMB operation is to obtain the two enantiomers, the path of equal purities is the optimum trajectory that must be followed [17]. Fig.…”

“…The separation used microcrystalline cellulose triacetate (particle diameter of 45 mm) as stationary phase and pure methanol as eluent at 25°C. A linear+Langmuir model was proposed to describe the adsorption equilibrium isotherms [17]: Table 7 presents the experimental conditions for the SMB operation. Using the simulation package presented in this work, a complete separation of both enantiomers was expected.…”

Design of SMB Chiral Separations Using the Concept of Separation Volume

“…[18 -20,31 -37] Consequently, if mass transfer effects are important, the operating conditions obtained by the equilibrium theory design must be modified, i.e., the region for complete separation defined by the equilibrium approach is reduced under nonideal conditions. [18,19,22,23,35,36] As we will show in the present work, mass transfer resistance can modify the better SMB operating conditions, leading to significant differences in the performance parameters such as productivity and solvent consumption. As it was already noticed for linear systems, [22] the constraints for Sections 1 and 4 will also be changed and will no longer depend solely on equilibrium data.…”