System-scale modeling and membrane structure parameter optimization for solar-powered sweeping gas membrane distillation desalination system

Abstract

Sweeping Gas Membrane Distillation (SGMD) is a promising desalination method which achieves a desired balance between reducing conductive heat loss and improving mass transport coefficient by using flowing gas instead of condensing water and static air-gap in the permeate side. However, previous studies mostly focused on the effects of operation conditions on the permeate flux by using a module-scale model. The role of membrane properties in a pragmatic system-scale desalination platform is seldom examined. This study newly established a system-scale mathematical model for a solar-powered SGMD desalination system in real outdoor weather conditions and investigated the interaction between membrane resistance and driving force to moisture transfer. Three hollow fiber membrane humidifiers made of membranes with different membrane structure parameters (pore size, porosity and tortuosity of the membrane pore) were tested experimentally to validate the new model. The verified model can be employed to optimize membrane structure parameters in terms of mean pore diameter, porosity and tortuosity. The results show that the controller of membrane resistance gradually transforms from Knudsen diffusion to ordinary molecular diffusion at a normal operating temperature range of 50–80 °C when the pore size increases from 140 nm to 160 nm. It is no longer beneficial to improve system performance by increasing the pore diameters above 150 nm where ordinary molecular diffusion begins to dominate membrane resistance. The optimal porosity is about 0.75, and further increase in porosity does not yield significant improvements in terms of freshwater yield and system Coefficient of Performance (COP). The approximate linear variation of the system performance indices with the tortuosity indicates that smaller tortuosity is favorable for the system performance. The system-scale modeling can be successfully employed for predicting the practical SGMD system performance and optimizing the membrane structure parameters.