Quantitative predictive modelling of ultrafiltration processes: colloidal science approaches

Area of Science:

• Chemical Engineering
• Colloid Science
• Separation Processes

Background:

• Membrane ultrafiltration (UF) is crucial for industrial separations.
• Accurate predictive models are needed for UF process optimization.
• Existing models require enhancement with detailed surface interaction physics.

Purpose of the Study:

• To develop predictive methods for membrane ultrafiltration.
• To incorporate fundamental colloidal interactions and hydrodynamics into UF models.
• To improve the accuracy of ultrafiltration process simulations.

Main Methods:

• Utilized cell-model for electrostatic interactions, London-van der Waals forces, and entropic pressure.
• Calculated osmotic pressures and gradient diffusion coefficients.
• Developed methods for local solution viscosity calculation.
• Integrated calculated properties into mathematical models for frontal and cross-flow UF.

Main Results:

• Models show excellent agreement with experimental ultrafiltration data.
• Predictive calculations correlate well with observed ultrafiltration rates.
• The approach allows predictions from fundamental properties like protein sequence.

Conclusions:

• Developed robust predictive models for membrane ultrafiltration.
• Fundamental colloidal and hydrodynamic interactions are key to accurate UF modeling.
• These methods offer versatile tools for UF process design and optimization.