Current experimental, statistical, and mechanistic approaches to optimizing biomolecule separations in aqueous...

Area of Science:

• Biotechnology
• Chemical Engineering
• Separation Science

Background:

• Aqueous two-phase systems (ATPS) are established for biomolecule purification, offering selective, low-shear, high-yield separations.
• ATPS are increasingly considered for continuous biomanufacturing due to cost-effectiveness compared to chromatography.
• Lack of process understanding hinders robust development and industry adoption of ATPS.

Purpose of the Study:

• To review the application of statistical and mechanistic models for ATPS.
• To compare the progress and potential of different modeling approaches in bioseparations.
• To explore strategies for automating ATPS optimization using model-generated data.

Main Methods:

• Survey of literature on statistical models (e.g., response surface methodology, artificial neural networks) for ATPS.
• Evaluation of semi-empirical thermodynamic models for predicting liquid-liquid equilibria and separations.
• Discussion of strategies for automated optimization and data generation using models.

Main Results:

• Models are being developed to predict ATPS behavior, reducing experimental needs.
• Statistical and mechanistic models offer different strengths in process understanding.
• Automated optimization using artificial data is a promising strategy.

Conclusions:

• Modeling is crucial for advancing ATPS development and overcoming adoption barriers.
• Bridging the gap between modeling and process understanding will accelerate ATPS commercialization.
• Enhanced modeling will enable broader adoption of continuous processing in biomanufacturing.