(730e) Design of a Pervaporation Assembly Using Static Mixers

Chemically inert membrane-based separation assemblies have been widely employed in numerous industries during processes of phase separation, purification, extraction, and many more. Pervaporation is a powerful and efficient membrane-based process to separate the minor components from a liquid mixture solution by partial vaporization and permeation. In addition, pervaporation can be implemented in a continuous synthesis process that allows for smaller dead volumes, intensified heat and mass transfer, better process control, safer operation, and the potential handling of reactions that are too dangerous or difficult to perform in batch.

In general, pervaporation process suffers from mass transfer limitations due to poor mass transport in the feed side (liquid side). A common model to describe the mass transfer mechanism in a pervaporation device is the resistance-in-series model, in which the overall mass transfer coefficient (k_all) is linked with each of k_liquid, k_membrane, and k_gas (1/k_all = 1/k_liquid + 1/k_membrane + 1/k_gas, and 1/k_gas can be safely ignored in most scenarios). In our design, we employ static mixers to create complex flow dynamics thus enhance mass transfer in the feed side (increasing k_liquid). The static mixers can be either commercial ones or self-designed and fabricated using 3D printing technologies.

In this work, we introduce a combination of simulations and experiments approach for pervaporation module design with static mixers as key components in continuous flow process. We design the pervaporation module with static mixers using computer-aided design (CAD) programs, and evaluate its performance by coupling and simulating the hydrodynamics and mass transfer performance. Moreover, the device is tested with ethanol – water and diethyl ether – water systems experimentally. From both approaches, we confirm that static mixers are beneficial to higher separation rates. In addition, we are able to fit the membrane mass transfer coefficient (k_membrane) to show that our simulations are capable of reproducing the experimental data. Then we can use simulations to show whether the devices could fulfill the desired functionality. Therefore, fundamental understanding is gained using the proposed methodology and applied to otherwise more challenging and more complicated membrane processes.