Coupling mass transport and chemical equilibrium models for improving the prediction of SWRO permeate boron concentrations

A new simulation approach is presented for predicting boron concentrations in the product water of seawater reverse osmosis operations. The new (numerical) approach links traditional mass-transfer models (the solution–diffusion transport approach and the concentration polarization film-layer model) with full aqueous-phase thermodynamic species characterization, performed by chemical equilibrium software (PHREEQC), based on the Pitzer approach. The new approach results in a more accurate calculation of the boric acid (B(OH)3) molar fraction which develops close to the membrane wall, on the feed side, thereby improving the prediction accuracy of B(OH)3 permeation. Specifically, acknowledging that the pH value of the feed invariably changes as seawater brine progresses through the membranes' train, calculation of this pH change, as performed in the new approach, enables a more physically-accurate and better simulation of the boric acid fraction. The new approach is shown in the paper to result in a prediction that matches better empirical results obtained from the operation of a pilot-scale SWRO plant, as compared to the traditional approach.

► A new reverse-osmosis boron permeation numerical simulation method is presented.
► The method couples classic transport methods with chemical equilibrium models.
► Simulation results are shown to better match measured B permeate concentrations.
► A prominent difference from classic models is in pH calculation throughout RO path.