Static Mixing Spacers for Spiral Wound Modules

Monday, June 11, 2012
La Salle A/B (InterContinental New Orleans)
Jiuqing Liu1, Ashkan Iranshahi2 and Glenn Lipscomb2, (1)College of Metallurgical Science and Engineering, Central South University, Changsha, China, (2)Department of Chemical and Environmental Engineering, University of Toledo, Toledo, OH

Membranes in the form of flat sheets are used to perform a variety of separations ranging from seawater desalination to carbon dioxide capture. Two membranes are sealed together to form an envelope or leaf which is wrapped around a central tube to form a compact cylindrical unit for contacting the feed, and potentially a permeate sweep stream, with the membrane. A permeate spacer separates the two sealed membranes to create a flow channel for the permeate that communicates with the central tube. A feed spacer separates the overlapping sections of the leaf as it is wound around the central tube.

The primary purpose of the permeate spacer is to create a flow channel for the permeate. While the feed spacer also creates a flow channel, it often can improve mass transfer rates by mixing the fluid in the feed channel to reduce concentration polarization and fouling – this is critical in seawater desalination applications. Commercial feed spacers most commonly consist of a layered or woven fiber structure that creates recirculation vortices or turbulence which mix the fluid in the feed channel. This mixing comes at the expense of greater pressure drop and associated pumping costs.

We have developed a novel spacer design. The spacer mixes fluid without vorticity or turbulence acting like a static mixer for planar flow channels. Notably the spacer moves fluid adjacent to the bounding surfaces of the flow channel to the center of the flow channel unlike static mixers used in pipes.

Experimental measurements of spacer performance for filtration of dextran solutions are presented. The static mixing spacer offers greatly enhanced performance relative to both an empty channel and commercially available spacers in terms of mass transfer coefficient as a function of fluid velocity. Theoretical predictions of mass transfer coefficient are in reasonable agreement with experiment but deviations exist which are attributed to the strong dependence of fluid material properties on dextran concentration that were not included in the simulations.

Pressure drop measurements also are presented as a function of fluid velocity. The static mixing spacer results in greater pressure drop than commercial spacers. However, a comparison of spacer performance in terms of mass transfer rate as a function of power input indicates the static mixing spacer offers superior performance.


Extended Abstract: File Not Uploaded
See more of this Session: Poster Session
See more of this Group/Topical: Poster Submissions