Forward osmosis (FO) is considered as an energy-efficient process for numerous applications. Although its performance is determined by the spatially varied operation factors and the length of the channel, most of the reported simulation studies rely on length-averaged lumped models. Here, we introduce a one-D model based on heat and mass transfer and transport behavior for both bulk draw and feed channel flows. We find prediction results to be in good agreement with two different experimental results at inlet feed temperatures below 25 °C. However, the difference of water flux (Jw) and reverse salt flux (RSF) between measured and predicted data increases when both feed and draw temperatures also increase. Our theoretical simulation study first reveals that the feed temperature near the membrane active layer surface is the main factor for improving water and salt permeabilities. We find that, with a channel width of 0.3 m and a channel length of 2.5 m, Jw and RSF calculated using the length-averaged based lumped model are overestimated by 13.01% and 13.12%, respectively, compared to those obtained using our new spatial variation model. Our study demonstrates that the length-averaged based lumped model is not an appropriate simulation model to predict the performance of large-scale FO modules at lower inlet velocities.
Bibliographical noteKAUST Repository Item: Exported on 2020-10-01
Acknowledgements: The research reported in this paper was supported by funding from King Abdullah University of Science and Technology (KAUST), Saudi Arabia. The authors acknowledge help, assistance and support from the Water Desalination and Reuse Center (WDRC) staff, and Mrs. Elisabeth M. Lutanie, Science Editor, Writer, Publication Services and Researcher Support at KAUST for editing the paper.