Modeling Membrane Formation

Engineering the properties of the membrane system has become a fundamental goal to achieve a membrane with an optimal structure and pore size distribution. Several parameters such as contact phase, composition of phases, polymer solution temperature, and nonsolvent/solvent species must be controlled to achieve this goal. Due to the multiplicity of process variables, the complexity of the interactions, as well as the speed of the process, it is difficult to determine and observe the PI process only through laboratory techniques. Modeling the phase separation process and developing efficient numerical models capable of predicting the final membrane morphology as accurately as possible, depending on the wide range of parameters. In this chapter, two top-down approaches—macroscopic transport models and mesoscopic PF models—as well as bottom-up molecular/particle scale simulations (such as molecular dynamics [MD], Monte Carlo and dissipative particle dynamics [DPD]), are presented to describe the processes of phase inversion, transfer phenomena, and membrane formation. Furthermore, a comprehensive view of the development of models over time and the presentation of numerical results from different models was provided to readers as a key solution for better understanding the limitations and capabilities of the simulations. Additionally, the use of machine learning as a robust tool for predicting membrane performance prior to fabrication is discussed.