Optimization of the operational parameters for desalination with response surface methodology during a capacitive deionization process

• Optimization of experimental parameters was carried out by BBD with RSM.
• The equivalent circuit for the CDI system was analyzed by EIS from Nyquist plot.
• Cell voltage, salt concentration and flow rate are key parameters of optimization.
• The response surface model with R2 = 0.9945 and p-value < 0.0001 is highly significant.
• The maximum electrosorption capacity of 10.53 mg/g was obtained after optimization.

The performance of capacitive deionization (CDI) and the effects of various operational parameters in CDI experiments were investigated for desalination on activated carbon electrodes. The pore size distribution determined using density functional theory showed that the carbon electrode materials used in this study contained some mesopores that could reduce the electric double-layer overlapping effect. From a Nyquist plot derived from electrochemical impedance spectroscopy, the equivalent circuit and impedance for the CDI system were analyzed, and the specific capacitance of the electrodes was approximately 71.05 F/g. Optimization of the operational conditions under the main experimental parameters, including the cell voltage (1.2–1.6 V), initial NaCl concentration (200–1000 mg/l) and flow rate (10–40 ml/min), was performed by Box–Behnken design with response surface methodology. A quadratic polynomial model was properly fit to the experimental data with a determination coefficient (R2) of 0.9945 and a p-value < 0.0001. The analysis of variance for the quadratic model demonstrated that the model was highly significant and reliable. The predicted maximum electrosorption capacity (10.67 mg/g) was given under the optimal conditions consisting of a cell voltage, initial NaCl concentration and flow rate of 1.57 V, 1000 mg/l and 25 ml/min, respectively, in accordance with the results of the validation experiments.