Process modeling and optimization of continuous chiral resolution by integration of membrane and crystallization technologies

Chiral resolution will increase its importance in the pharmaceutical, pesticide, and food industries since future chiral products will be structurally more complex while their quality will need to meet stringent demands to accomplish with health, crop, and environmental purposes. Currently, obtaining enantiopure crystals from racemic compound-forming systems (90–95% of chiral compounds) remains a challenge. The combination of membrane ultrafiltration and crystallization in a continuous process is a potential solution to this problem. This work aims to optimize this continuous chiral resolution process using process modeling. First, a model for the membrane ultrafiltration step assumes a one-site competition in which the chiral selector binds preferentially to one of the enantiomers. The amino acids phenylalanine, alanine, and valine in aqueous media along with the chiral selector BSA are the model systems studied. Then, three process performance parameters (purity, recovery, and a combination of the previous two) are assessed to optimize the conditions for the continuous membrane-based step. Recoveries up to 70% are achieved while fulfilling purity levels beyond eutectic point compositions, a requirement for the enantiopure crystallization. Additionally, the properties of an ideal process chiral selector are explored. Finally, the crystallization step (with recoveries up to 65%) is coupled to evaluate the maximum recovery of the overall process. Simulations show that the membrane-crystallization process performance is governed by a trade-off between purity and recovery. We predict that an ideal BSA-similar chiral selector should bear an unprotonated binding site for a wide pH range and a difference between enantiomer complexation constants of at least 2–3 pK units to improve the process performance.