Removal of Trace Uranium from Groundwaters Using Membrane Capacitive Deionization Desalination for Potable Supply in Remote Communities: Bench, Pilot, and Field Scale Investigations.

The performance of membrane capacitive deionization (MCDI) desalination was investigated at bench, pilot, and field scales for the removal of uranium from groundwater. It was found that up to 98.9% of the uranium can be removed using MCDI from a groundwater source containing 50 µg/L uranium, with the majority (94.5%) being retained on the anode. Uranium was found to physiochemically adsorb to the electrode without the application of a potential by displacing chloride ions, with 16.6% uranium removal at the bench scale via this non-electrochemical process. This displacement of chloride did not occur during the MCDI adsorption phase with the adsorption of all ions remaining constant during a time series analysis on the pilot unit. For the scenarios tested on the pilot unit, the flowrate of the product water ranged from 0.15 to 0.23 m3/h, electrode energy consumption from 0.28 to 0.51 kW h/m3, and water recovery from 69 to 86%. A portion (13-53% on the pilot unit) of the uranium was found to remain on the electrodes after the brine discharge phase with conventional cleaning techniques unable to release this retained uranium. MCDI was found to be a suitable means to remove uranium from groundwater systems though with the need to manage the accumulation of uranium on the electrodes over time.

Analysis of Ozonation Processes Using Coupled Modeling of Fluid Dynamics, Mass Transfer, and Chemical Reaction Kinetics.

The efficacy of oxidation of recalcitrant organic contaminants in municipal and industrial wastewaters by ozonation is influenced by chemical reaction kinetics and hydrodynamics within a reactor. A 3D computational fluid dynamics (CFD) model incorporating both multiphase flow and reaction kinetics describing ozone decay, hydroxyl radical (•OH) generation, and organic oxidation was developed to simulate the performance of continuous flow ozonation reactors. Formate was selected as the target organic in this study due to its well-understood oxidation pathway. Simulation results revealed that the dissolved ozone concentration in the reactor is controlled by rates of O3(g) interphase transfer and ozone self-decay. Simulations of the effect of various operating conditions showed that the reaction stoichiometric constraints between formate and ozone were reached; however, complete utilization of gas phase ozone was hard to achieve due to the low ozone interphase mass transfer rate. Increasing the O3(g) concentration leads to an increase in the formate removal rate by ∼5% due to an enhancement in the rate of O3(g) interphase mass transfer. The CFD model adequately describes the mass transfer occurring in the two-phase flow system and confirms that O3(g) interphase mass transfer is the rate-limiting step in contaminant degradation. The model can be used to optimize the ozone reactor design for improved contaminant degradation and ozonation efficiency.

Brackish groundwater desalination by constant current membrane capacitive deionization (MCDI): Results of a long-term field trial in Central Australia.

A long-term field trial of membrane capacitive deionization (MCDI) was conducted in a remote community in the Northern Territory of Australia, with the aim of producing safe palatable drinking water from groundwater that contains high concentrations of salt and hardness ions and other contaminants. This trial lasted for 1.5 years, which, to our knowledge, is one of the longest reported studies of pilot-scale MCDI field trials. The 8-module MCDI pilot unit reduced salt concentration to below the Australian Drinking Water Guideline value of 600 mg/L total dissolved solids (TDS) concentration with a relatively high water recovery of 71.6 ± 8.7 %. During continuous constant current operation and electrode discharging at near zero volts, a rapid performance deterioration occurred that was primarily attributed to insufficient desorption of multivalent ions from the porous carbon electrodes. Performance could be temporarily recovered using chemical cleaning and modified operating procedures however these approaches could not fundamentally resolve the issue of insufficient electrode performance regeneration. Constant current discharging of the electrodes to a negative cell cut-off voltage was hence employed to enhance the stability and overall performance of the MCDI unit during the continuous operation. An increase in selectivity of monovalent ions over divalent ions was also attained by implementing negative voltage discharging. The energy consumption of an MCDI system with a capacity of 1000 m3/day was projected to be 0.40∼0.53 kWh/m3, which is comparable to the energy consumption of electrodialysis reversal (EDR) and brackish water reverse osmosis (BWRO) systems of the same capacity. The relatively low maintenance requirements of the MCDI system rendered it the most cost-efficient water treatment technology for deployment in remote locations. The LCOW of an MCDI system with a capacity of 1000 m3/day was projected to be AU$1.059/m3 and AU$1.146/m3 under two operational modes, respectively. Further investigation of particular water-energy trade-offs amongst MCDI performance metrics is required to facilitate broader application of this promising water treatment technology.

Insufficient desorption of ions in constant-current membrane capacitive deionization (MCDI): Problems and solutions.

Membrane capacitive deionization (MCDI) is a water desalination technology that involves the removal of charged ions from water under an electric field. While constant-current MCDI coupled with stopped-flow during ion discharge is expected to exhibit high water recovery and good performance stability, previous studies have typically been undertaken using NaCl solutions only with limited investigation of MCDI performance using multi-electrolyte solutions. In the present work, the desalination performance of MCDI was evaluated using feed solutions with different levels of hardness. The increase of hardness resulted in the degradation of desalination performance with the desalination time (Δtd), total removed charge, water recovery (WR) and productivity decreasing by 20.5%, 21.8%, 3.8% and 3.2%, respectively. A more serious degradation of WR and productivity would be caused if Δtd decreases further. Analysis of the voltage profiles and effluent ion concentrations reveal that the insufficient desorption of divalent ions at constant-current discharge to 0 V was the principal reason for the degradation of performance. The Δtd and WR can be improved by discharging the cell using a lower current but the productivity decreased by 15.7% on decreasing the discharging current from 161 to 107 mA. Discharging the cell to a negative potential was shown to be a better option with the Δtd, total removed charge, WR and productivity increasing by 27.4%, 23.9%, 3.6% and 5.3%, respectively, when the cell was discharged to a minimum voltage of - 0.3 V. Use of such a method should be feasible for operation of full scale MCDI plants and would be expected to lead to better regeneration of the electrode, improved desalination performance and, potentially, a significant reduction in the need for use of clean-in-place procedures.

Machine learning modelling of a membrane capacitive deionization (MCDI) system for prediction of long-term system performance and optimization of process control parameters in remote brackish water desalination.

Membrane Capacitive Deionization (MCDI) is a promising electrochemical technique for water desalination. Previous studies have confirrmed the effectiveness of MCDI in removing contaminants from brackish groundwaters, especially in remote areas where electricity is scarce. However, as with other water treatment technologies, performance deterioration of the MCDI system still occurs, hindering the stability of long-term operation. Herein, a machine learning (ML) modelling framework and various ML models were developed to (i) investigate the performance deterioration due particularly to insufficient charging/discharging of the electrode caused by accumulation of ions and electrode scaling and (ii) optimise MCDI operating parameters such that the impacts of these deleterious effects on unit performance were minimized. The ML models developed in this work exhibited a prediction accuracy of cycle time with average mean absolute percentage error (MAPE) values of 16.82% and 16.09% after 30-fold cross validation for Random Forest (RF) and Multilayer Perceptron (MLP) models respectively. The pre-trained ML model predicted different declining trends of water production for two different operating conditions and provided corresponding recommendations on frequencies of chemical cleaning. A case study on the adjustment of operating parameters using the results suggested by the optimization ML model was conducted. The model validation results showed that the overall water production and water recovery of the system using the cycle-based optimized process control parameters (SCN 1) exceeds the MCDI system performance under three fixed parameter settings that were used at each stage of SCN 1 by 1.78% to 4.48% and 2.95% to 9.46%, respectively. Permutation-based and Shapley additive explanation (SHAP) coefficients were also employed for variable importance (VIMP) analysis to uncover the "black-box" nature of the ML models and to better understand the various features' contributions to overall MCDI system performance.

Electrochemical Ni-EDTA degradation and Ni removal from electroless plating wastewaters using an innovative Ni-doped PbO2 anode: Optimization and mechanism.

In this work, a novel Ni-doped PbO2 anode (Ni-PbO2) was prepared via a co-electrodeposition method and used to remove Ni-ethylenediaminetetraacetic acid (Ni-EDTA) from solutions typical of electroless nickel plating wastewater. Compared with a pure PbO2 electrode, Ni doping increased the oxygen evolution potential as well as the reactive surface area and reactive site concentration and reduced the electron transfer resistance thereby resulting in superior Ni-EDTA degradation performance. The 1% Ni-doped PbO2 electrode exhibited the best electrochemical oxidation activity with a Ni-EDTA removal efficiency of 96.5 ± 1.2%, a Ni removal efficiency of 52.1 ± 1.4% and an energy consumption of 2.6 kWh m-3. Further investigations revealed that 1% Ni doping enhanced both direct oxidation and hydroxyl radical mediated oxidation processes involved in Ni-EDTA degradation. A mechanism for Ni-EDTA degradation is proposed based on the identified products. The free nickel ion concentration initially increased as a result of the degradation of Ni-EDTA complexes and subsequently decreased as a consequence of nickel electrodeposition on the cathode surface. Further characterization of the cathode deposits by X-ray diffraction and X-ray photoelectron spectra indicated that the deposition products were a mixture of Ni0, NiO and Ni(OH)2 with elemental Ni accounting for roughly 80% of the deposited nickel. Results of this study pave the way for the application of anodic oxidation processes for efficient degradation of Ni-containing complexes and recovery of Ni from nickel-containing wastewaters.

Optimization of constant-current operation in membrane capacitive deionization (MCDI) using variable discharging operations.

Membrane capacitive deionization (MCDI) is an emerging electric field-driven technology for brackish water desalination involving the removal of charged ions from saline source waters. While the desalination performance of MCDI under different operational modes has been widely investigated, most studies have concentrated on different charging conditions without considering discharging conditions. In this study, we investigate the effects of different discharging conditions on the desalination performance of MCDI electrode. Our study demonstrates that low-current discharge (1.0 mA/cm2) can increase salt removal by 20% and decrease volumetric energy consumption by 40% by improving electrode regeneration and increasing energy recovery, respectively, while high-current discharge (3.0 mA/cm2) can improve productivity by 70% at the expense of electrode regeneration and energy recovery. Whether discharging electrodes at the low current or high current is optimal depends on a trade-off between productivity and energy consumption. We also reveal that stopped flow discharge (85%) can achieve higher water recovery than continuous flow discharge (35-59%). However, stopped flow discharge caused a 20-30% decrease in concentration reduction and a 25-50% increase in molar energy consumption, possibly due to the higher ion concentration in the macropores at the end of discharging step. These results reveal that an optimal discharging operation should be obtained from achieving a balance among productivity, water recovery and energy consumption by varying discharging current and flow rate.

Scale-up and Modelling of Flow-electrode CDI Using Tubular Electrodes.

A novel design for a flow-electrode capacitive deionization (FCDI) system consisting of tubular electrodes in a shell and tube heat exchanger configuration is proposed. Each electrode consists of a metallic mesh current collector along the inner circumference of a tubular ion-exchange membrane. This tubular FCDI design is suitable for scale-up as it consists of easily manufactured components which can be assembled in an array. An apparatus with 4 tubular electrodes with a large effective area (202.3 cm2) was constructed and shown to provide a high net salt (NaCl) removal rate (0.15 mg s-1 at 1.2 V applied voltage and ∼2000 mg L-1 influent total dissolved solids concentration). A computational fluid dynamics (CFD) model incorporating ion migration and transport mechanisms was developed to simulate the ion concentration and electrical potential profiles in the water channel. The results of CFD modelling highlighted the need to maximize regions of both high potential gradient and high hydraulic flow in order to achieve optimal salt removal. In brief, this study presents a new design approach for FCDI scale-up and provides a computational tool for optimization of this design and future innovative FCDI designs.

Impact of FO Operating Pressure and Membrane Tensile Strength on Draw-Channel Geometry and Resulting Hydrodynamics.

In an effort to improve performances of forward osmosis (FO) systems, several innovative draw spacers have been proposed. However, the small pressure generally applied on the feed side of the process is expected to result in the membrane bending towards the draw side, and in the gradual occlusion of the channel. This phenomenon potentially presents detrimental effects on process performance, including pressure drop and external concentration polarization (ECP) in the draw channel. A flat sheet FO system with a dot-spacer draw channel geometry was characterized to determine the degree of draw channel occlusion resulting from feed pressurization, and the resulting implications on flow performance. First, tensile testing was performed on the FO membrane to derive a Young's modulus, used to assess the membrane stretching, and the resulting draw channel characteristics under a range of moderate feed pressures. Membrane apex reached up to 67% of the membrane channel height when transmembrane pressure (TMP) of 1.4 bar was applied. The new FO channels considerations were then processed by computational fluid dynamics model (computational fluid dynamics (CFD) by ANSYS Fluent v19.1) and validated against previously obtained experimental data. Further simulations were conducted to better assess velocity profiles, Reynolds number and shear rate. Reynolds number on the membrane surface (draw side) increased by 20% and shear rate increased by 90% when occlusion changed from 0 to 70%, impacting concentration polarisation (CP) on the membrane surface and therefore FO performance. This paper shows that FO draw channel occlusion is expected to have a significant impact on fluid hydrodynamics when the membrane is not appropriately supported in the draw side.

Development of a mobile groundwater desalination system for communities in rural India.

The consumption of saline groundwater has contributed to a growing incidence of renal diseases, particularly in coastal communities of India. Although reverse osmosis (RO) is routinely used to remove salt from groundwater, conventional RO systems (i.e. centralized systems using spiral wound RO elements) have limited utility in these communities due to high capital and maintenances costs, and lack of infrastructure to distribute the water. Consequently, there is a need to develop an appropriate solution for groundwater treatment based on small-scale, mobile and community-led systems. In this work, we designed a mobile desalination system to provide a simple platform for water treatment and delivery of goods to rural communities. The system employs tubular RO membranes packed in a single, low-profile vessel which fits below the cargo space. The low-profile enables minimal intrusion on the space available for the transportation of goods. Pressure is delivered by a belt driven clutch pump, powered by the engine. Water is treated locally by connecting the intake to the village well while the vehicle idles. A combined numerical and experimental approach was used to optimise the module/system design, resulting in ∼20% permeate flux enhancement. Experimental results revealed that the system can produce 16 L per square meter of membrane area per hour (LMH) at a salinity level of 80 ppm from a ∼2000 ppm groundwater when it is feed at 1 m3/h at 8 bars. This indicates that a vehicle equipped with 12 m2 of tubular RO membranes can deliver 1 m3 of drinkable water by using ∼0.9 L of diesel. Assuming eight such systems could be implemented in a community to fulfil the water demands for a village with 2000 residents, a social business study revealed that a payback time of 2.5 years is achievable, even if the sale price of the water is relatively low, USD 0.18 (Rs 12, which is half of the lowest market price) per 20 L, including providing a goods transportation service at price of USD 5.25 (Rs 350) per 100 km.