Model-Supported Design of Multicolumn Countercurrent Solvent Gradient Purification Processes for Rapid Implementation

The development of industrial chromatography is typically constrained by timelines and available material. For the design of continuous multicolumn processes like multicolumn countercurrent solvent gradient purification (MCSGP), the required material consumption during development (≈10 to 20 g) is a major issue. In this work we present a method based on mechanistic modeling to design MCSGP operation in a material and time-efficient manner and such as to result in a process that is easy to transfer, and fits within a specific facility. Starting from a viable model, the workflow consists of three major steps: (i) single column simulations are carried out to identify batch chromatography runs that result in a sufficient quality, and to assess recycle parameters expected to result in a fast convergence to cyclic steady state (CSS), (ii) all suitable batch runs are transferred to MCSGP by following the common pragmatic approach and the twin column process is simulated until cyclic steady state; (iii) facility fit is investigated by scaling up and adapting the process to the equipment for commercial production. We report the results of applying this workflow in the context of an industrial case study, where an MCSGP process was designed for the purification of a synthetic peptide. The results highlight how this methodology allows to design MCSGP operating conditions that meet a target purity in a robust manner, and for which the CSS is reached within a few cycles only.