Date of Award
Doctor of Philosophy (PhD)
Environmental Engineering and Earth Science
David A. Ladner
Lawrence C. Murdoch
My research goal is to discover ways to improve the hydrodynamics of reverse osmosis (RO) membrane systems through creative membrane surface patterning and spacer designs. Since concentration polarization (CP) usually promotes membrane fouling, improving hydrodynamics would result in reduced fouling and better membrane performance. With computational fluid dynamics (CFD), we can explore dozens or even hundreds of models with different geometries and boundary conditions. Through plotting their velocity profile, streamlines, shear stress, pressure profile, concentration profile, and so on, we can determine which design would lead to the best performance.
At first, patterned membranes were evaluated and compared with flat membranes to see how the enhanced mixing and distorted fluid flow would affect CP. Geometries including line-and-groove patterns, pillars, and pyramids were investigated. It was found that patterns did not decrease CP, but rather increased CP because of the added roughness. After discussing the changes involving mass transfer and permeate flux through patterned membranes, discrete-object spacers were studied with circular, elliptical, and airfoil plan-view shapes. Unlike conventional mesh spacers that connect like a net, these spacers are printed directly onto the membrane surface and they have much less obstruction to the tangential flow over the membranes. Due to decreased pressure drop, these membrane modules can be built more compactly and therefore the packing capacity can be enhanced.
As the study of these spacers moved forward, one innovative idea was to design them in a way that when reversing the flow direction the areas of high shear would shift. The specially designed spacers can cause anisotropic flow so that the concentration buildup and fouling accumulation would be “scraped off”. In other words, when water enters in one direction, spacers create areas with high velocity and low velocity, which lead to very different mass transfer speed and hydrodynamics, and when the flow switches direction, the areas with high velocity now has low velocity, and vice versa. The studies focused on arranging spacers into layouts such as linear (V shape) and sigmoid shapes. It was found that designing spacers as ellipse shapes with a length to width ratio of 2.4:1 can give the lowest pressure drop. The arrangement of the spacers did not greatly affect the hydrodynamics in these models.
My research approach is using CFD to run multi-scale studies and design modeling systems that could incorporate all potential situations. By coupling mass transfer and permeate flux equations, the aim is to represent reality in the best way possible and the models are verified either by consistency with theoretical models or through experimental studies in the literature. By scrutinizing each model, we gain confidence in choosing the most ideal designs to achieve membrane systems with improved performance.
The biggest challenge in this research is discovering new designs that can reduce CP or fouling without adding extra energy consumption resulting from pressure drop increases. The typical tradeoff in these designs seems to be hard to break; better mixing and higher shear stress can help reduce fouling, but it also usually leads to higher pressure drop. The effort in this line of research continues toward discovering the best designs for both patterned membranes and discrete-object spacers that might deviate from this rule and allow CP reduction even at low pressure drop.
Zhou, Zuo, "Computational Modeling of Patterned Membranes and Spacers for Improved Hydrodynamics and Fouling Reduction in Reverse Osmosis Water Treatment Processes" (2023). All Dissertations. 3346.
Author ORCID Identifier