Temperature-dependent boron permeability through reverse-osmosis membranes: Implications for full-scale simulations

Boron removal from desalinated seawater is essential for obtaining high quality water, suitable for irrigation. Removal of boron by RO (reverse osmosis), the leading desalination technology, is significantly affected by variations in feed water temperature. Nevertheless, a widely agreed quantitative method for describing the temperature effect on the boric acid permeability constant, an important parameter in RO process modeling, was thus far unavailable. In this paper, different methods for describing permeability constants as a function of temperature were systematically evaluated against empirical results. It was demonstrated that non-specific temperature correlations, which are based on a single permeability measurement at a reference temperature, result in increased deviations from the measured permeability as the temperatures shift away from the reference value. A more accurate approach is to determine membrane-specific temperature correlations based on measured permeabilities at the relevant temperature range. Subsequently, the influence of accurate temperature correction on process modeling was assessed by comparing experimental boron rejections at practical conditions to simulation results. It was found that a reliable boric acid permeability coefficient is particularly significant at warm temperature, where boron rejection is less effective. Finally, implications to process design are discussed in light of accurate temperature dependent boron removal simulations.