This study focused on optimizing a selective demineralization process for seawater purification using ion-exchange technology. Experiments were conducted in three semi-batch reactors containing cation, anion, and mixed resins. Key process parameters included temperature (25 °C–50 °C), resin depth (23–82 cm), and pH (2–12). Statistical modeling and optimization were performed using Response Surface Methodology (RSM) with a central composite design, addressing both single and multi-objective criteria. A desirability function was used to assess process performance based on multiple response variables, such as the removal of trace metals (Ca2+, Mg2+, Mn2+, Zn2+, Fe2+, Cu2+, Ba2+, Cd2+), conductivity reduction, and total dissolved solids (TDS) elimination. Ten quadratic regression models were developed to describe the relationships between input parameters and responses, achieving high R2 values (≥0.7) for most responses except Cu2+, Mn2+, and Ba2+. Multi-objective optimization highlighted TDS, conductivity, and the removal of Ca2+, Mn2+, and Mg2+ as critical targets due to their significant impact on water hardness. The optimal conditions (temperature of 43.9 °C, resin depth of 75.45 cm, and pH of 5.9) yielded a composite desirability score of 0.77. Under these conditions, the process achieved over 99% removal efficiency for key cations (Ca2+, Mg2+), significant conductivity reduction, and near-complete TDS elimination. However, Mn2+ removal efficiency reached approximately 85%, likely due to its lower diffusion coefficient and higher hydration enthalpy. The results, particularly from the multicriteria optimization combined with desirability function approaches, highlight the effectiveness of ion-exchange resins in seawater demineralization and offer a robust framework for enhancing process performance.
Read full abstract