This study aims to optimize ammonium removal from NH4Cl-enriched groundwater at different concentrations using an electrodialysis (ED) process. A customized design (CD) based on response surface methodology (RSM) was employed to develop predictive models and improve the performance of the demineralization system. Ion removal efficiency was evaluated in 32 unique experimental configurations, taking into account variations in three input parameters: voltage (A), initial ammonium concentration (B) and demineralization rate (C). These parameters were selected for their impact on two response variables: electric conductivity (Y1) and final ammonium concentration (Y2). An in-depth analysis of variance (ANOVA) was performed to examine the variables and their interactions. The results indicated that Y1 was significantly influenced by C, while Y2 was influenced by B. In addition, the predictive models demonstrated strong correlations, with a coefficient of determination (R2) greater than 0.88 for both response variables. The RSM approach applied to optimize the parameters studied identified the following optimum values: 14.17 V for A, 1 mg/L for B and 70 % for C, giving Y1 of 215.377 μS/cm and Y2 of 0.279 mg/L.
Read full abstract