Reducing Chloride Ion Permeation during Seawater Electrolysis Using Double-Polyamide Thin-Film Composite Membranes.

Low-cost polyamide thin-film composite membranes are being explored as alternatives to expensive cation exchange membranes for seawater electrolysis. However, transport of chloride from seawater to the anode chamber must be reduced to minimize the production of chlorine gas. A double-polyamide composite structure was created that reduced the level of chloride transport. Adding five polyamide layers on the back of a conventional polyamide composite membrane reduced the chloride ion transport by 53% and did not increase the applied voltage. Decreased chloride permeation was attributed to enhanced electrostatic and steric repulsion created by the new polyamide layers. Charge was balanced through increased sodium ion transport (52%) from the anolyte to the catholyte rather than through a change in the transport of protons and hydroxides. As a result, the Nernstian loss arising from the pH difference between the anolyte and catholyte remained relatively constant during electrolysis despite membrane modifications. This lack of a change in pH showed that transport of protons and hydroxides during electrolysis was independent of salt ion transport. Therefore, only sodium ion transport could compensate for the reduction of chloride flux to maintain the set current. Overall, these results prove the feasibility of using a double-polyamide structure to control chloride permeation during seawater electrolysis without sacrificing energy consumption.

Modeling Ion Transport across Thin-Film Composite Membranes During Saltwater Electrolysis.

Affordable thin-film composite (TFC) membranes are a potential alternative to more expensive ion exchange membranes in saltwater electrolyzers used for hydrogen gas production. We used a solution-friction transport model to study how the induced potential gradient controls ion transport across the polyamide (PA) active layer and support layers of TFC membranes during electrolysis. The set of parameters was simplified by assigning the same size-related partition and friction coefficients for all salt ions through the membrane active layer. The model was fit to experimental ion transport data from saltwater electrolysis with 600 mM electrolytes at a current density of 10 mA cm-2. When the electrolyte concentration and current density were increased, the transport of major charge carriers was successfully predicted by the model. Ion transport calculated using the model only minimally changed when the negative active layer charge density was varied from 0 to 600 mM, indicating active layer charge was not largely responsible for controlling ion crossover during electrolysis. Based on model simulations, a sharp pH gradient was predicted to occur within the supporting layer of the membrane. These results can help guide membrane design and operation conditions in water electrolyzers using TFC membranes.

Thin-Film Composite Membranes for Hydrogen Evolution with a Saline Catholyte Water Feed.

Hydrogen gas evolution using an impure or saline water feed is a promising strategy to reduce overall energy consumption and investment costs for on-site, large-scale production using renewable energy sources. The chlorine evolution reaction is one of the biggest concerns in hydrogen evolution with impure water feeds. The "alkaline design criterion" in impure water electrolysis was examined here because water oxidation catalysts can exhibit a larger kinetic overpotential without interfering chlorine chemistry under alkaline conditions. Here, we demonstrated that relatively inexpensive thin-film composite (TFC) membranes, currently used for high-pressure reverse osmosis (RO) desalination applications, can have much higher rejection of Cl- (total crossover of 2.9 ± 0.9 mmol) than an anion-exchange membrane (AEM) (51.8 ± 2.3 mmol) with electrolytes of 0.5 M KOH for the anolyte and 0.5 M NaCl for the catholyte with a constant current (100 mA/cm2 for 20 h). The membrane resistances, which were similar for the TFC membrane and the AEM based on electrochemical impedance spectroscopy (EIS) and Ohm's law methods, could be further reduced by increasing the electrolyte concentration or removal of the structural polyester supporting layer (TFC-no PET). TFC membranes could enable pressurized gas production, as this membrane was demonstrated to be mechanically stable with no change in permeate flux at 35 bar. These results show that TFC membranes provide a novel pathway for producing green hydrogen with a saline water feed at elevated pressures compared to systems using AEMs or porous diaphragms.

Relative Insignificance of Polyamide Layer Selectivity for Seawater Electrolysis Applications.

Low-cost polyamide thin-film composite (TFC) membranes are being explored as alternatives to cation exchange membranes for seawater electrolysis. An optimal membrane should have a low electrical resistance to minimize applied potentials needed for water electrolysis and be able to block chloride ions present in a seawater catholyte from reaching the anode. The largest energy loss associated with a TFC membrane was the Nernstian overpotential of 0.74 V (equivalent to 37 Ω cm2 at 20 mA cm-2), derived from the pH difference between the anolyte and catholyte and not the membrane ohmic overpotential. Based on analysis using electrochemical impedance spectroscopy, the pristine TFC membrane contributed only 5.00 Ω cm2 to the ohmic resistance. Removing the polyester support layer reduced the resistance by 79% to only 1.04 Ω cm2, without altering the salt ion transport between the electrolytes. Enlarging the pore size (∼5 times) in the polyamide active layer minimally impacted counterion transport across the membrane during electrolysis, but it increased the total concentration of chloride transported by 60%. Overall, this study suggests that TFC membranes with thinner but mechanically strong supporting layers and size-selective active layers should reduce energy consumption and the potential for chlorine generation for seawater electrolyzers.

Vapor-Fed Cathode Microbial Electrolysis Cells with Closely Spaced Electrodes Enables Greatly Improved Performance.

Hydrogen can be electrochemically produced in microbial electrolysis cells (MECs) by current generated from bacterial anodes with a small added voltage. MECs typically use a liquid catholyte containing a buffer or salts. However, anions in these catholytes result in charge being balanced predominantly by ions other than hydroxide or protons, leading to anode acidification. To enhance only hydroxide ion transport to the anode, we developed a novel vapor-fed MEC configuration lacking a catholyte with closely spaced electrodes and an anion exchange membrane to limit the acidification. This MEC design produced a record-high sustained current density of 43.1 ± 0.6 A/m2 and a H2 production rate of 72 ± 2 LH2/L-d (cell voltage of 0.79 ± 0.00 V). There was minimal impact on MEC performance of increased acetate concentrations, solution conductivity, or anolyte buffer capacity at applied voltages up to 1.1 V, as shown by a nearly constant internal resistance of only 6.8 ± 0.3 mΩ m2. At applied external voltages >1.1 V, the buffer capacity impacted performance, with current densities increasing from 28.5 ± 0.6 A/m2 (20 mM phosphate buffer solution (PBS)) to 51 ± 1 A/m2 (100 mM PBS). These results show that a vapor-fed MEC can produce higher and more stable performance than liquid-fed cathodes by enhancing transport of hydroxide ions to the anode.

Thermodynamic and Kinetic Analyses of Ion Intercalation/Deintercalation Using Different Temperatures on NiHCF Electrodes for Battery Electrode Deionization.

Prussian blue analogues are used in electrochemical deionization due to their cation sorption capabilities and ion selectivity properties. Elucidating the fundamental mechanisms underlying intercalation/deintercalation is important for the development of ion-selective electrodes. We examined the thermodynamic and kinetic properties of nickel hexacyanoferrate electrodes by studying different temperatures effects on intercalation/deintercalation with monovalent ions (Li+, Na+, K+, and NH4+) relevant to battery electrode deionization applications. Higher temperatures reduced the interfacial charge transfer resistance and increased the diffusion coefficient of cations in the solid material. Ion transport in the solid material, rather than interfacial charge transfer, was found to be the rate-controlling step, as shown by higher activation energies for ion transport (e.g., 31 ± 3 kJ/mol for K+) than for interfacial charge transfer (5 ± 1 kJ/mol for K+). The largest increase in cation adsorption capacity with temperature was observed for NH4+ (28.1% from 15 to 75 °C) due to its smallest activation energy. These results indicate that ion hydration energy determines the intercalation potential and activation energies of ion transport in solid material control intercalation/deintercalation rate. Together with the endothermic behavior of deintercalation and exothermic behavior of intercalation, the higher operating temperature results in improvement of ion adsorption capacity depending on specific cations.

Metal-Ion Depletion Impacts the Stability and Performance of Battery Electrode Deionization over Multiple Cycles.

Prussian blue hexacyanoferrate (HCF) materials, such as copper hexacyanoferrate (CuHCF) and nickel hexacyanoferrate (NiHCF), can produce higher salt removal capacities than purely capacitive materials when used as electrode materials during electrochemical water deionization due to cation intercalation into the HCF structure. One factor limiting the application of HCF materials is their decay in deionization performance over multiple cycles. By examining the performance of CuHCF and NiHCF electrodes at three different pH values (2.5, 6.3, and 10.2) in multiple-cycle deionization tests, losses in capacity (up to 73% for CuHCF and 39% for NiHCF) were shown to be tied to different redox-active centers through analysis of dissolution of electrode metals. Both copper and iron functioned as active centers for Na+ removal in CuHCF, while iron was mainly the active center in NiHCF. This interaction of Na+ and active centers was demonstrated by correlating the decrease in performance to the concentration of these metal ions in the effluent solutions collected over multiple cycles at different pHs (up to 0.86 ± 0.14 mg/L for iron and 0.42 ± 0.17 mg/L for copper in CuHCF and 0.38 ± 0.05 mg/L for iron in NiHCF). Both materials were more stable (<11% decay for CuHCF and no decay for NiHCF) when the appropriate metal salt (copper or nickel) was added to the feed solutions to inhibit electrode dissolution. At a pH of 2.5, there was an increased competition between protons and Na+ ions, which decreased the Na+ removal amount and lowered the thermodynamic energy efficiency for deionization for both electrode materials. Therefore, while an acidic pH provided the most stable performance, a circumneutral pH would be useful to produce a better balance between performance and longevity.

Long-Term Succession Shows Interspecies Competition of Geobacter in Exoelectrogenic Biofilms.

Geobacter spp. are well-known exoelectrogenic microorganisms that often predominate acetate-fed biofilms in microbial fuel cells (MFCs) and other bioelectrochemical systems (BESs). By using an amplicon sequence variance analysis (at one nucleotide resolution), we observed a succession between two closely related species (98% similarity in 16S RNA), Geobacter sulfurreducens and Geobacter anodireducens, in the long-term studies (20 months) of MFC biofilms. Geobacter spp. predominated in the near-electrode portion of the biofilm, while the outer layer contained an abundance of aerobes, which may have helped to consume oxygen but reduced the relative abundance of Geobacter. Removal of the outer aerobes by norspermidine washing of biofilms revealed a transition from G. sulfurreducens to G. anodireducens. This succession was also found to occur rapidly in co-cultures in BES tests even in the absence of oxygen, suggesting that oxygen was not a critical factor. G. sulfurreducens likely dominated in early biofilms by its relatively larger cell size and production of extracellular polymeric substances (individual advantages), while G. anodireducens later predominated due to greater cell numbers (quantitative advantage). Our findings revealed the interspecies competition in the long-term evolution of Geobacter genus, providing microscopic insights into Geobacter's niche and competitiveness in complex electroactive microbial consortia.

Improving the Thermodynamic Energy Efficiency of Battery Electrode Deionization Using Flow-Through Electrodes.

Ion intercalation electrodes are being investigated for use in mixed capacitive deionization (CDI) and battery electrode deionization (BDI) systems because they can achieve selective ion removal and low energy deionization. To improve the thermodynamic energy efficiency (TEE) of these systems, flow-through electrodes were developed by coating porous carbon felt electrodes with a copper hexacyanoferrate composite mixture. The TEE for ion separation using flow-through electrodes was compared to a system using flow-by electrodes with the same materials. The flow-through BDI system increased the recoverable energy nearly 3-fold (0.009 kWh m-3, compared to a 0.003 kWh m-3), which increased the TEE from ∼6% to 8% (NaCl concentration reduction from 50 to 42 mM; 10 A m-2, 50% water recovery, and 0.5 mL min-1). The TEE was further increased to 12% by decreasing the flow rate from 0.50 to 0.25 mL min-1. These findings suggest that, under similar operational conditions and materials, flow-through battery electrodes could achieve better energy recovery and TEE for desalination than flow-by electrodes.

Surveying Manganese Oxides as Electrode Materials for Harnessing Salinity Gradient Energy.

The potential energy contained in the controlled mixing of waters with different salt concentrations (i.e., salinity gradient energy) can theoretically provide a substantial fraction of the global electrical demand. One method for generating electricity from salinity gradients is to use electrode-based reactions in electrochemical cells. Here, we examined the relationship between the electrical power densities generated from synthetic NaCl solutions and the crystal structures and morphologies of manganese oxides, which undergo redox reactions coupled to sodium ion uptake and release. Our aim was to make progress toward developing rational frameworks for selecting electrode materials used to harvest salinity gradient energy. We synthesized 12 manganese oxides having different crystal structures and particle sizes and measured the power densities they produced in a concentration flow cell fed with 0.02 and 0.5 M NaCl solutions. Power production varied considerably among the oxides, ranging from no power produced (ß-MnO2) to 1.18 ± 0.01 W/m2 (sodium manganese oxide). Power production correlated with the materials' specific capacities, suggesting that cyclic voltammetry may be a simple method to screen possible materials. The highest power densities were achieved with manganese oxides capable of intercalating sodium ions when their potentials were prepoised prior to power production.

Evaluation of Electrode and Solution Area-Based Resistances Enables Quantitative Comparisons of Factors Impacting Microbial Fuel Cell Performance.

Direct comparisons of microbial fuel cells based on maximum power densities are hindered by different reactor and electrode sizes, solution conductivities, and materials. We propose an alternative method here, the electrode potential slope (EPS) analysis, to enable quantitative comparisons based on anode and cathode area-based resistances and operating potentials. Using EPS analysis, the brush anode resistance ( RAn = 10.6 ± 0.5 mΩ m2) was shown to be 28% lower than the resistance of a 70% porosity diffusion layer (70% DL) cathode ( RCat = 14.8 ± 0.9 mΩ m2) and 24% lower than the solution resistance ( RΩ = 14 mΩ m2) (acetate in a 50 mM phosphate buffer solution). Using a less porous cathode (30% DL) did not impact the cathode resistance but did reduce the cathode performance due to a lower operating potential. With low-conductivity domestic wastewater ( RΩ = 87 mΩ m2), both electrodes had higher resistances [ RAn = 75 ± 9 mΩ m2, and RCat = 54 ± 7 mΩ m2 (70% DL)]. Our analysis of the literature using EPS analysis shows how electrode resistances can easily be quantified to compare system performance when the electrode distances are changed or the sizes of the electrodes are different.

Electro-Forward Osmosis.

The impact of ion migration induced by an electrical field on water flux in a forward osmosis (FO) process was examined using a thin-film composite (TFC) membrane, held between two cation exchange membranes. An applied fixed current of 100 mA (1.7 mA cm-2) was sustained by the proton flux through the TFC-BW membrane using a feed of 34 mM NaCl, and a 257 mM NaCl draw solution. Protons generated at the anode were transported through the cation exchange membrane and into the draw solution, lowering the pH of the draw solution. Additional proton transport through the TFC-BW membrane also lowered the pH of the feed solution. The localized accumulation of the protons on the draw side of the TFC-BW membrane resulted in high concentration polarization modulus of 1.41 × 105, which enhanced the water flux into the draw solution (5.56 LMH at 100 mA), compared to the control (1.10 LMH with no current). These results using this electro-forward osmosis (EFO) process demonstrated that enhanced water flux into the draw solution could be achieved using ion accumulation induced by an electrical field. The EFO system could be used for FO applications where a limited use of draw solute is necessary.

Balancing Water Dissociation and Current Densities To Enable Sustainable Hydrogen Production with Bipolar Membranes in Microbial Electrolysis Cells.

Hydrogen production using two-chamber microbial electrolysis cells (MECs) is usually adversely impacted by a rapid rise in catholyte pH because of proton consumption for the hydrogen evolution reaction. While using a bipolar membrane (BPM) will maintain a more constant electrolyte pH, the large voltage loss across this membrane reduces performance. To overcome these limitations, we used an acidic catholyte to compensate for the potential loss incurred by using a BPM. A hydrogen production rate of 1.2 ± 0.7 L-H2/L/d (jmax = 10 ± 0.4 A/m2) was obtained using a Pt cathode and BPM with a pH difference (ΔpH = 6.1) between the two chambers. This production rate was 2.8 times greater than that of a conventional MEC with an anion exchange membrane (AEM, 0.43 ± 0.1 L-H2/L/d, jmax = 6.5 ± 0.3 A/m2). The catholyte pH gradually increased to 11 ± 0.3 over 9 days using the BPM and Pt/C, which decreased current production (jmax = 2.5 ± 0.3 A/m2). However, this performance was much better than that obtained using an AEM as the catholyte pH increased to 10 ± 0.4 after just one day. The use of an activated carbon cathode with the BPM enabled stable performance over a longer period of 12 days, although it reduced the hydrogen production rate (0.45 ± 0.1 L-H2/L/d).

Regenerable Nickel-Functionalized Activated Carbon Cathodes Enhanced by Metal Adsorption to Improve Hydrogen Production in Microbial Electrolysis Cells.

While nickel is a good alternative to platinum as a catalyst for the hydrogen evolution reaction, it is desirable to reduce the amount of nickel needed for cathodes in microbial electrolysis cells (MECs). Activated carbon (AC) was investigated as a cathode base structure for Ni as it is inexpensive and an excellent adsorbent for Ni, and it has a high specific surface area. AC nickel-functionalized electrodes (AC-Ni) were prepared by incorporating Ni salts into AC by adsorption, followed by cathode fabrication using a phase inversion process using a poly(vinylidene fluoride) (PVDF) binder. The AC-Ni cathodes had significantly higher (∼50%) hydrogen production rates than controls (plain AC) in smaller MECs (static flow conditions) over 30 days of operation, with no performance decrease over time. In larger MECs with catholyte recirculation, the AC-Ni cathode produced a slightly higher hydrogen production rate (1.1 ± 0.1 L-H2/Lreactor/day) than MECs with Ni foam (1.0 ± 0.1 L-H2/Lreactor/day). Ni dissolution tests showed that negligible amounts of Ni were lost into the electrolyte at pHs of 7 or 12, and the catalytic activity was restored by simple readsorption using a Ni salt solution when Ni was partially removed by an acid wash.

Impact of Ohmic Resistance on Measured Electrode Potentials and Maximum Power Production in Microbial Fuel Cells.

Low solution conductivity is known to adversely impact power generation in microbial fuel cells (MFCs), but its impact on measured electrode potentials has often been neglected in the reporting of electrode potentials. While errors in the working electrode (typically the anode) are usually small, larger errors can result in reported counter electrode potentials (typically the cathode) due to large distances between the reference and working electrodes or the use of whole cell voltages to calculate counter electrode potentials. As shown here, inaccurate electrode potentials impact conclusions concerning factors limiting power production in MFCs at higher current densities. To demonstrate how the electrochemical measurements should be adjusted using the solution conductivity, electrode potentials were estimated in MFCs with brush anodes placed close to the cathode (1 cm) or with flat felt anodes placed further from the cathode (3 cm) to avoid oxygen crossover to the anodes. The errors in the cathode potential for MFCs with brush anodes reached 94 mV using acetate in a 50 mM phosphate buffer solution. With a felt anode and acetate, cathode potential errors increased to 394 mV. While brush anode MFCs produced much higher power densities than flat anode MFCs under these conditions, this better performance was shown primarily to result from electrode spacing following correction of electrode potentials. Brush anode potentials corrected for solution conductivity were the same for brushes set 1 or 3 cm from the cathode, although the range of current produced was different due to ohmic losses with the larger distance. These results demonstrate the critical importance of using corrected electrode potentials to understand factors limiting power production in MFCs.

Membrane-based processes for sustainable power generation using water.

Water has always been crucial to combustion and hydroelectric processes, but it could become the source of power in membrane-based systems that capture energy from natural and waste waters. Two processes are emerging as sustainable methods for capturing energy from sea water: pressure-retarded osmosis and reverse electrodialysis. These processes can also capture energy from waste heat by generating artificial salinity gradients using synthetic solutions, such as thermolytic salts. A further source of energy comes from organic matter in waste waters, which can be harnessed using microbial fuel-cell technology, allowing both wastewater treatment and power production.

Advanced Materials, Technologies, and Complex Systems Analyses: Emerging Opportunities to Enhance Urban Water Security.

Innovation in urban water systems is required to address the increasing demand for clean water due to population growth and aggravated water stress caused by water pollution, aging infrastructure, and climate change. Advances in materials science, modular water treatment technologies, and complex systems analyses, coupled with the drive to minimize the energy and environmental footprints of cities, provide new opportunities to ensure a resilient and safe water supply. We present a vision for enhancing efficiency and resiliency of urban water systems and discuss approaches and research needs for overcoming associated implementation challenges.

Substantial Humic Acid Adsorption to Activated Carbon Air Cathodes Produces a Small Reduction in Catalytic Activity.

Long-term operation of microbial fuel cells (MFCs) can result in substantial degradation of activated carbon (AC) air-cathode performance. To examine a possible role in fouling from organic matter in water, cathodes were exposed to high concentrations of humic acids (HA). Cathodes treated with 100 mg L(-1) HA exhibited no significant change in performance. Exposure to 1000 mg L(-1) HA decreased the maximum power density by 14% (from 1310 ± 30 mW m(-2) to 1130 ± 30 mW m(-2)). Pore blocking was the main mechanism as the total surface area of the AC decreased by 12%. Minimization of external mass transfer resistances using a rotating disk electrode exhibited only a 5% reduction in current, indicating about half the impact of HA adsorption was associated with external mass transfer resistance and the remainder was due to internal resistances. Rinsing the cathodes with deionized water did not restore cathode performance. These results demonstrated that HA could contribute to cathode fouling, but the extent of power reduction was relatively small in comparison to large mass of humics adsorbed. Other factors, such as biopolymer attachment, or salt precipitation, are therefore likely more important contributors to long-term fouling of MFC cathodes.

Harvesting Energy from Salinity Differences Using Battery Electrodes in a Concentration Flow Cell.

Salinity-gradient energy (SGE) technologies produce carbon-neutral and renewable electricity from salinity differences between seawater and freshwater. Capacitive mixing (CapMix) is a promising class of SGE technologies that captures energy using capacitive or battery electrodes, but CapMix devices have produced relatively low power densities and often require expensive materials. Here, we combined existing CapMix approaches to develop a concentration flow cell that can overcome these limitations. In this system, two identical battery (i.e., faradaic) electrodes composed of copper hexacyanoferrate (CuHCF) were simultaneously exposed to either high (0.513 M) or low (0.017 M) concentration NaCl solutions in channels separated by a filtration membrane. The average power density produced was 411 ± 14 mW m(-2) (normalized to membrane area), which was twice as high as previously reported values for CapMix devices. Power production was continuous (i.e., it did not require a charging period and did not vary during each step of a cycle) and was stable for 20 cycles of switching the solutions in each channel. The concentration flow cell only used inexpensive materials and did not require ion-selective membranes or precious metals. The results demonstrate that the concentration flow cell is a promising approach for efficiently harvesting energy from salinity differences.

Graphene-Coated Hollow Fiber Membrane as the Cathode in Anaerobic Electrochemical Membrane Bioreactors--Effect of Configuration and Applied Voltage on Performance and Membrane Fouling.

Electrically conductive, graphene-coated, hollow-fiber porous membranes were used as cathodes in anaerobic electrochemical membrane bioreactors (AnEMBRs) operated at different applied voltages (0.7 and 0.9 V) using a new rectangular reactor configuration compared to a previous tubular design (0.7 V). The onset of biofouling was delayed and minimized in rectangular reactors operated at 0.9 V compared to those at 0.7 V due to higher rates of hydrogen production. Maximum transmembrane pressures for the rectangular reactor were only 0.10 bar (0.7 V) or 0.05 bar (0.9 V) after 56 days of operation compared to 0.46 bar (0.7 V) for the tubular reactor after 52 days. The thickness of the membrane biofouling layer was approximately 0.4 µm for rectangular reactors and 4 µm for the tubular reactor. Higher permeate quality (TSS = 0.05 mg/L) was achieved in the rectangular AnEMBR than that in the tubular AnEMBR (TSS = 17 mg/L), likely due to higher current densities that minimized the accumulation of cells in suspension. These results show that the new rectangular reactor design, which had increased rates of hydrogen production, successfully delayed the onset of cathode biofouling and improved reactor performance.