Design of continuous chromatographic processes under reduced purity requirements

Continuous chromatographic separations (i. e. SMB processes) are usually designed under the assumption that a binary feed mixture should be separated into two pure fractions. A useful tool in this context is the well-known “triangle theory”, which was derived from equilibrium theory and provides simple algebraic expressions for optimal flow rates that are necessary for complete separation (see, e. g., [1]). However, compete separation is not always required; for example, if chromatography is combined with a second separation unit. In this case, the lowered purity required from chromatography can allow for significantly improved overall performance in terms of productivity, solvent consumption, or amount of stationary phase (e. g., [2]). While the case of complete separation is well-studied, there is a lack of investigations on design for reduced purity requirements and, consequently, there is no tool available like the aforementioned “triangle theory”. In this work, we present a systematic study on optimal design of continuous chromatography under reduced purity requirements. For this purpose, optimisations of the rather simple mathematical model for the True Moving Bed (TMB) process are performed using stochastic and gradient-based optimisation techniques. On this basis, characteristic properties of an optimal design are established. In particular, it is demonstrated that the requirement of complete regeneration of adsorbent and mobile phase within the unit may be abandoned, which in turn allows for further improvement of performance. The impact of such optimised design on process robustness is investigated by a sensitivity analysis. Finally, the application of equilibrium theory for process design under reduced purity requirements is discussed. [1] M. Mazzotti et al., J. Chromatogr. A 769 (1997) 3-24 [2] M. Kaspereit et al., J. Chromatogr. A 1092 (2005) 43-5