Computational fluid dynamics simulation of the roll-to-roll coating process for the production of thin film composite membranes including validation

On the way toward a carbon-neutral economy, a rise in the demand for separation technologies can be expected. This holds especially for membranes, which are energy efficient and thus very promising technologies. However, this challenges membrane researchers to consider the sustainability and scalability of their membrane fabrication processes. At our institute, we employ a roll-to-roll coating process for the production of thin film composite membranes. This procedure is relatively easy to scale up and can be adapted for different polymers/solvents and thus applications. To increase the production efficiency and optimize the process for any polymer of interest, it is necessary to develop a solid understanding of the physics of this production system. Therefore, we would like to present a numerical model based on computational fluid dynamics that can predict the film thickness of a polydimethylsiloxane-based polymer coated in a roll-to-roll setup. In the future, this model should improve the production efficiency and fine-tuning of process parameters. We verify the numerical procedure with a mesh refinement study and validate the predicted film thicknesses with experimental results for different roll speeds and polymer concentrations. The predicted variability of the thickness is assessed by a design of experiments study and compares relatively well to the measured variations of the coated membrane thickness.