Evaluation of Nanoporous Electrode Materials for Ion Removal and Energy Recovery in Water Treatment by Capacitive Deionization.

摘要

Capacitive deionization (CDI) is a water treatment technology that utilizes electrically polarized electrodes to remove ions from solution. It has been proposed for a variety of environmental applications, including industrial wastewater treatment and desalination. When an external electric potential is applied between electrodes, ions migrate towards the oppositely charged electrodes under electrostatic force and accumulate in the interfacial region by forming an electric double layer (EDL). Novel, ultra-high surface area electrode materials, such as nanoporous carbon are enabling more efficient CDI designs. Though electrosorption on well-defined planar surfaces can be predicted by classic EDL models, the electrosorption in nanoporous electrodes is different. Surface properties of porous electrodes including surface chemical properties and pore size distribution can change EDL and change the ion electrosorption behaviors. In this study, the nanoporous electrodes materials were evaluated for ion removal, energy consumption and recovery during capacitive deionization.;Ion removal processes during CDI involves a combination of electrosorption, physical sorption, chemical sorption and physico-chemical sorption processes. Previous approaches to model ion removal were based on electrosorption alone and were not adequate and accurate enough to interpret observations in CDI processes. The surface chemical properties (i.e. degree of protonation) of the electrode material affect ion removal during CDI. We used an asymmetric CDI cell constructed with alumina and silica nanoparticle (NP) coated electrodes and KCI as a probe electrolyte to gain insights into electrosorption behavior and elucidate underlying process mechanisms. The presence of NPs increased the charge efficiency by shifting the applied potential to a high efficiency range due to protonation/deprotonation occurring on metal oxide surfaces. Our results suggest that the presence of metal oxide NPs can effectively modify the isoelectric points and an increase in planar charge efficiency of up to 20% could be achieved.;In addition to electrode surface modification, EDLs become distorted when the pore size is narrowed down to nano- or sub-nanoscale ranges. EDLs will overlap resulting in reduced electrosorption when the pore size is close to EDL thickness. The electrosorption capacity increases anomalously contradicting traditional EDL predictions when the pore size is in the nano-scale range. We used three types of activated carbon cloth (ACC) with different pore-size distributions to study the impact of pore characteristics on electrosorption during CDI. Results showed that while both mesoand micro- pores contribute to sorption, the underlying sorption mechanisms were different. Sorption capacity, normalized by pore volume, decreased as the mesoporosity-to-microporosity ratio increased. Both ion hydration radius and pore size distribution determined the ion selectivity.;With an improved understanding of the importance of pore size distribution, especially the impact of microporosity, during electrosorption, a microporosity-based EDL theory was applied to analyze the energy consumption and recovery in our flow-through cell under various operational conditions including charge/discharge current, influent ion concentration and flow rate. The results indicated that 30-45% of the energy consumed during charging could be recovered during discharging depending on conditions. The energy consumption for reducing the salinity (NaCI) of brackish water from 32.7 to 5.5 mM by our device was as low as 0.85 kWh/m3 under optimized conditions. However, cycle analysis revealed that the thermodynamic desalination energy was less than 2% of the net energy consumption while the remaining 98% of the net energy consumption was irreversible (cannot be recovered during discharge) energy due to overpotential and charge loss. The results implied that energy consumption could be dramatically reduced by employing more electron-conductive and Faradaic-resistant electrode materials.;*******Abstract Shortened*******.