Evaluation of Brackish Water Desalination by Membrane Capacitive Deionization Systems: Performance Tradeoffs in Salt Removal, Hydraulic Recovery, Charge Efficiency, and Specific Energy Consumption

As potable water demand continues to exceed availability in several regions of the world, alternative water sources such as desalination and direct potable reuse have become of high importance in water sustainability discussions. The conventional desalination technologies include reverse osmosis (RO) and thermal distillation. Major barriers to the applicability of these technologies for desalination include inorganic scaling (RO) and high energy requirements respectively. This research provides an evaluation for the feasibility of Membrane Capacitive Deionization (MCDI), which is a technology gaining interest due to its theoretically low energy requirements for brackish water (less than 3 g/L) desalination. The motivation for this research is a deficiency in the literature of the parametric dependencies in MCDI, especially the effect of the operating hydraulic residence time (also known as the system detention time). The detention time is a parameter that allows for comparison between different MCDI devices across a range of sizes and flow rates. The first objective of this research was to perform a sensitivity analysis for different operating parameters on key desalination performance metrics in MCDI, as well as to analyze tradeoffs in operation at high hydraulic recovery. Furthermore, this research covered the development of a model based on equivalent electrical circuit to predict the performance of MCDI for desalination. The third objective was to evaluate the scale-up of high capacity aqueous processed electrodes for use in MCDI. The experimental parametric analysis as discussed in this dissertation was performed with a commercial MCDI device and a laboratory fabricated MCDI unit. It was observed that operating the MCDI device at higher detention times showed lower charge efficiency and resulted in higher specific energy consumption. The total cumulative salt removal was observed to increase with increasing detention time up to 60 s, beyond which there was marginal difference in salt removal but steady decline in charge efficiency. An increase in feed (sodium chloride) salinity from 1,302 mg/L to 5,271 mg/L for the same detention time also decreased the charge efficiency of the system from 74% to 50%, respectively. The Simplified Randles electrical circuit model that was developed in this research was able to accurately simulate the electrical current, charge efficiency, and product conductivity performance for detention time less than 60 s and cell voltage less than 1.2 V. The least root mean squared error (RMSE) analysis used for the model development was able to generate charge efficiency predictions within 3% of measured values for seventeen different experiments. The aqueous processed electrodes fabricated and tested in a laboratory-built reactor had a salt adsorption capacity of 0.41 meq/g, exceeding performance observed by collaborators at initial small-scale development. Overall, the findings of this research will help industries and municipalities identify the applicability and tradeoffs for the use of MCDI for desalination of brackish water