435026 Proper Accounting of Mass Transfer Resistances in Forward Osmosis: Improving the Accuracy of Model Predictions of Structural Parameter

Thursday, November 12, 2015: 8:30 AM
155C (Salt Palace Convention Center)
Ngoc Bui, Physical and Life Sciences, Lawrence Livermore National Laboratory, Livermore, CA, Jeffrey McCutcheon, Chemical Engineering, University of Connecticut, Storrs, CT and Jason T. Arena, Department of Chemical and Biomolecular Engineering, University of Connecticut, Storrs, CT

Proper Accounting of Mass Transfer Resistances in Forward Osmosis: 

Improving the Accuracy of Model Predictions of Structural Parameter

Ngoc Bui, Lawrence Livermore National Laboratory, Livermore, CA

Jason Arena, Jeffrey R. McCutcheon, University of Connecticut, Chemical and Biomolecular Engineering Department, Storrs, CT

Forward Osmosis (FO) and pressure retarded osmosis (PRO) have recently been revitalized as a sustainable and versatile membrane-based separation technology platform for water and power production, respectively. To design membranes for various FO processes, it is important to understand critical structure-performance relationships, especially with respect to mass transfer. This work demonstrates a more accurate method for calculating structural parameter (S) of asymmetric osmotic membranes using experimental data and a theoretical flux model which encapsulates all significant boundary layer phenomena. External boundary layer effects on the porous side of the membrane have been neglected in many current models. In these models, external concentration polarization (ECP) effects get combined with the internal concentration polarization (ICP), resulting in inflated S values. In this study, we proposed a new flux model in which ECP effects are accounted for so that S can be more accurately measured. This model considered the in-series resistances for solute transport based on intrinsic properties of the membrane as well as boundary layers at membrane surfaces and within the support layer. The results indicate that ICP is less severe than previously predicted and that cross-flow velocity, temperature and concentration of the draw and the feed solutions impact both external and internal concentration polarization. Our calculations also surprisingly show that changes in cross-flow velocity impact internal concentration polarization due to induced mixing within the support layer. Furthermore, new definitions of membrane reflection coefficient and total resistance to solute transport emerge from this work. Also, we suggest that it is critical to consider the “residence time” of solutes in the vicinity of the selective layer in determining the membrane selectivity.

 


Extended Abstract: File Not Uploaded
See more of this Session: Modeling Transport in Membrane Processes
See more of this Group/Topical: Separations Division