3D laminar flow simulations were conducted using OpenFOAM to resolve the temperature, concentration, velocity, and pressure field for two hollow fiber vacuum membrane distillation configurations with feed solution flowing either inside or outside a single hollow fiber. The fiber has a circular cross-section, a fixed length of 120mm, and an inner diameter of 2 mm. The wall thickness was varied from 150 to 450μm, and the pore diameter was varied from 0.1 to 0.3μm based on commercial fibers. The feed solution is an aqueous solution of water and NaCl. The feed flow was simulated at a Reynolds number of 500 and vacuum pressure of 5,000 Pascals. It was found that there was a 75% increase in flux, from 9.58 to 41.41 kg/m2h, between the worst and the best case in membrane properties. Increasing the pore diameter or wall thickness while the other value was fixed resulted in a 45–57% flux increase depending on the fixed value. The module with the feed solution flowing outside a hollow fiber yields 24% higher flux than the module with the feed solution flowing inside the hollow fiber at the same conditions.

No abstract

2D Large Eddy Simulations (LES) were conducted to study the effect of channel geometry on the Vacuum Membrane Distillation process. The geometry was altered by imposing a sinusoidal (“wiggly”) wall profile. The results of the study show at a critical Reynolds number between 750 and 1,000, transient vortices appear in the channel. As the Reynolds number is increased from 1,000 to 1,500, the origin of the vortices moves further upstream, and the frequency and intensity of the vortex activity increase. Areas, where vortex shedding provides mixing, serve to enhance the performance of membrane distillation profoundly. The mixing reduces temperature and concentration polarization along the membrane surface. With better mixing provide at Reynolds number 1,500 versus 1,000, the difference in performance versus the corresponding flat sheet membrane case is increased. The vapor flux in the wiggly channel module increases 6% at Reynolds number 1,000 and 22% at Reynolds number 1,500 compared to the flat channel module. The change in flux is from 61.6 to 65.7 kg/m2/h at Reynolds number 1,000 and 66.5 to 81.4 kg/m2/h at Reynolds number 1,500. Temperature polarization was also mitigated at Reynolds number 1,500 even though the flux was increased.

CFD simulations were run to determine the impact of buoyancy-driven natural convection on Membrane Distillation, a thermally driven filtration process. A computational domain was created with a high Grashoff number, and vertical and horizontal modules were considered. Convection cells were dominant in the channel at low Reynolds numbers and effectively increased the performance of Membrane Distillation modules by up to 66%. This was done by generating convection cells that reduced the temperature/concentration polarization inside that module, typically forming due to the low Reynolds numbers and correspondingly high thermal and concentration boundary layers. It is shown here that natural convection enhances the flux performance of membrane distillation immensely.

The open-source C++ toolbox OpenFOAM is used to perform the Computational Fluid Dynamics (CFD) simulations in two-dimensional microfluidic devices to characterize the viscoelastic flow. The Oldroyd-B constitutive equation is coupled with continuity and momentum equations. Multiple stabilization methods are applied to the numerical simulation to simulate High Weissenberg Number Problem (HWNP) in the microchannel. We applied the Log Conformation Reformulation (LCR) method to guarantee the positive definiteness of the stress tensor. The CUBISTA scheme and the improved Both Side Diffusion (iBSD) method are applied to predict the flow behavior at high elasticity regions without numerical oscillations. Various microstructures, including circles and flat plates, are placed in the center of the channel as the confinement. Our previous work demonstrated that the polyhedral mesh with hexahedral inflation layers effectively meshes complex microstructures in microchannels. A viscoelastic fluid is injected from the inlet at varying flow rates, corresponding to the local Weissenberg number up to 25. A parametric study is conducted on the first normal stress difference (N1) in specific regions with an accurate prediction of the viscoelastic flow field near the microstructures.