Nitrate reduction to ammonia via coupling electrofiltration with a cooperative nitrite-enriching component

Abstract

A functionalized metal / carbon nanotube-based electrified nanoporous membrane with a cooperative nitrite-enriching component to achieve energy-efficient ammonia production from low-concentration nitrate. The use of flow-through electrofiltration reduces energy consumption and treatment time. Integrating an ionophore as a cooperative component into the conductive filtration membrane enriches nitrite within the catalyst microenvironment, enabling conversion of unreacted NO2- to NH3 to enhance overall selectivity without directly modifying the catalytic active sites.

Description

NITRATE REDUCTION TO AMMONIA VIA COUPLING ELECTROFILTRATION WITH A COOPERATIVE NITRITE-ENRICHING COMPONENTSTATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT

[0001] This invention was made with government support under EEC-1449500, awarded by the NSF Nanotechnology-Enabled Water Treatment (NEWT) Engineering Research Center (ERC). The U.S. Government has certain rights in the invention.FIELD OF THE INVENTION

[0002] The present invention relates generally to an electrochemical water treatment membrane system and method for the electrochemical conversion of nitrate in a water stream.BACKGROUND OF THE INVENTION

[0003] Nitrate (NO3) and nitrite (NOT) are both compounds that can be toxic and that can also be safe in certain amounts. NO?’ is widely used in food processing, particularly in curing meats. NO3- helps prevent the growth of harmful bacteria, and its use is regulated to ensure safety.Although small amounts of NO?’ are not considered to be harmful, NO?’ can react with amines to form nitrosamines, which are carcinogenic. This means that NO?’ may pose health risks depending on the conditions (e.g., exposure to certain compounds and heat, or acidic conditions), especially in the presence of secondary amines. In addition, NO?’ can interfere with oxygen transport in the blood (methemoglobinemia), which is particularly dangerous for infants. Since NO?’ is more reactive than nitrate and can more easily form nitrosamines, NO?’ is generally more toxic than nitrate in water.

[0004] The different toxicity levels of NO?’ and NO3’ are also reflected in the drinking water limits set by organizations such as the World Health Organization (WHO) and the U.S.Environmental Protection Agency (EPA). For example, the EPA’s Maximum Contaminant Level (MCL) for NO3’ is 10 mg / L as nitrogen (NO3-N), while for NO?’, it is only 1 mg / L for nitrite as nitrogen (NO?-N). This difference in toxicity level illustrates that NO?’ is considered to be more dangerous in terms of its impact on human health related to blood oxygen transport, its potential to form carcinogenic compounds, and its faster absorption and reactivity in the body.

[0005] In addition to minimizing the environmental risks of generating NO?’, reducing NO?’ formation is also important for optimizing the yield of ammonia (NH3). While producing N? is areasonable alternative, it would be desirable to provide an environmentally friendly water treatment process through simultaneous NH3 production, providing a potential sustainable pathway for the production of fertilizer or carbon-free fuel. That is, it would be desirable to provide a system and process for generating NH3 for removal from water and recovery as a useful product. If NH3 is targeted for recovery, then the remaining water may be discharged to the environment after NH3 removal.

[0006] However, the NH3 separation process that follows the NO?’ reduction step (e.g., electrochemical stripping or heating) likely would not be effective in removing NO?’. Therefore, minimizing residual NO?’ in water when NH3 is the target product is crucial to eliminate generation of a harmful waste stream from this process. Further, NO?’ generation is a common issue for NOs’ reduction to NH3, because NO?’ is the main quasi-stable intermediate during the electrochemical reduction of NO3’ before its complete conversion to NH3, as described, for example, in Applied Catalysis B: Environmental 236 (2018): 546-568 and Electrochim. Acta 50 (2005) 4318- -4326. Therefore, improving the understanding of NO?’ generation represents an important aspect for developing electrochemical processes for generating NH3 from NOs’.

[0007] Artificial NH3 synthesis is currently a major industrial process, as NI h-based fertilizers are used to produce food to feed about half of the world’s population. Traditional NH3 synthesis consumes about 2% of global energy output each year, and due to reliance upon fossil fuels both as feedstock (to make H? used as a reagent for ammonia synthesis) and to power the process, it accounts for about 1 to 2% of global CO? emissions. Additionally, the high temperatures and pressures required for traditional NH3 synthesis make the process only economically feasible on a large scale, severely limiting the number of synthesis plants and exacerbating fertilizer shortages in Africa and parts of Asia. A process to produce NH3 from renewable energy and nonfossil fuel feedstocks under mild conditions could not only address the above challenges, but also provide a means of producing green ammonia as a carbon-free fuel.

[0008] One potential method for obtaining green NH3 is to convert NOs’ from wastewater sources (i.e., contaminated surface water, agricultural runoff, treated sewage, industrial wastewaters) and water using electrochemical reduction. However, many of these water sources contain relatively low concentrations of NOi’, which leads to inefficient conversion due to mass transport limitations. In addition, electrochemical NCh’ reduction can produce undesirableproducts such as NO?’, which can remain as a residual toxic component in treated water and reduce the overall nitrate-to-ammonia efficiency.

[0009] One route for obtaining sustainable NH3 without H2 involves the oxidation of N2 from air to NOXfollowed by absorption in water and then reduction to NH3. Such a scheme requires efficient and selective processes for converting aqueous NOx species such as NOs’ to NH3. Furthermore, the direct utilization of NO3’ in wastewater as an NH3 feedstock could provide an energy-saving approach by eliminating air separation and N2 oxidation, while also providing environmental benefits by abating NCh’ pollution.

[0010] Decarbonized NH3 synthesis is crucial for offsetting the enormous carbon emissions from the Haber-Bosch fertilizer synthesis process, and may also provide a path to NH as an alternative, carbon-free fuel.

[0011] Nitrogen (N?) electrolyzers suffer from extremely low yields or require hazardous and expensive non-aqueous electrolytes, and thermocatalytic processes are challenging to use with current renewable energy sources. Electrochemically reducing NCh’ from wastewaters (e.g., treated municipal wastewater, agricultural runoff) to ammonia has been suggested as a promising approach to low-carbon NH3 synthesis, with the added benefit of mitigating harmful levels of nitrate in drinking water and the environment. Groundwater sources are considered to be contaminated with NCh’ if they are about 3 mg / L and most incidences of EPA limit violations the U.S. are at levels about 10 to about 20 mg / L, with some violations reaching 20 to 50 mg / L, largely in California. However, electrochemical reduction of NCh’ at these low levels requires overcoming mass transport limitations. Additional challenges for achieving practical NCh’ reduction include lowering costs by reducing or eliminating the use of expensive precious metal catalysts, increasing energy efficiency, and reducing treatment time.

[0012] Electrofiltration through nanoporous electrodes or electrified membranes (EMs) has been suggested in order to overcome mass transport limitations by reducing the diffusion boundary layer, leading to highly efficient NO?’ conversion. EMs also provide a scalable and versatile electrolyzer platform, eliminating the need for an ion exchange membrane and enabling modular architecture based on mature water treatment membrane paradigms.

[0013] However, the diminished boundary layer in EMs also leads to increased diffusion of partially-reduced intermediate NO?’ away from the electrode during NO.f reduction. Specifically, significant amounts of NO?’ can desorb from the catalyst surface and transport to the bulkelectrolyte prior to further reduction to NH3, decreasing the overall NH3 selectivity and increasing specific energy consumption for NH3 synthesis as well as decreasing the overall green NH3 production.

[0014] Previous studies have suggested that NCh’ is the main quasi-stable intermediate during the electrochemical reduction of NCh’ before its complete conversion to NH3, thereby serving as the “divergent center” for controlling the overall final product selectivity. The residual NCh' in treated water (for which NCh’ has been converted to NH3 and the NH3 has been separated for commercial use) poses significant environmental and safety concerns, as NCh’ is a toxic carcinogen and harmful to aquatic organisms.

[0015] NCh’ desorption / accumulation is a common issue for NCh’ reduction to NH3 over a variety of metals (e.g., Fe, Cu, Ag). Therefore, improving the understanding of reactions involving NCh' along the NCh’-to-NHs pathway is crucial for developing strategies to enhance the conversion efficiency to desirable products and consequently reduce cost, specific energy consumption, and operational time.

[0016] Moreover, in a flow-through operational mode, the reduced boundary layer significantly enhances the NCh’ conversion rate and energy efficiency, but it also promotes the diffusion of the partially-reduced, toxic intermediate NCh’ away from the electrode. As a result, more significant amounts of NCh’ transport away from the catalyst surface and move into the bulk electrolyte before being further reduced to NH3, thereby diminishing the overall green NH3 production rate and efficiency.

[0017] Despite the drawbacks of NCh’ byproduct formation, significant research efforts continue to focus on enhancing NCh’ conversion without addressing the generation of excessive NCh’ in the product water.

[0018] Therefore, there remains a need in the art for an improved electrocatalytic filtration membrane system for preferentially reducing NCh' to NH3.SUMMARY OF THE INVENTION

[0019] It is an object of the present invention to provide an electrocatalytic membrane system for preferentially reducing nitrate to ammonia in a water stream.

[0020] It is another object of the present to provide an electrochemical membrane system that includes an electrolytic filtration membrane for the electrocatalytic conversion of nitrate.

[0021] It is still another object of the present invention to provide an improved conductive filtration membrane that includes an ionophore that exhibits selective adsorption properties to specific ions.

[0022] It is still another object of the present invention to provide an electrocatalytic membrane system in which the anode comprises a mixed metal oxide mesh.

[0023] It is still another object of the present invention to provide a method of making the conductive filtration membrane for use in the system of the invention.

[0024] To that end, in one embodiment, the present invention relates generally to a conductive filtration membrane, the conductive filtration membrane comprising: a. carbon nanotubes; and b. metal particles preferably wherein the metal particles comprise copper particles, more preferably wherein the copper particles comprise copper nanoparticles; and c. an ionophore that exhibits selective adsorption properties to specific ions, wherein the ionophore is interspersed with the metal particles; and

[0025] In another embodiment, the present invention relates generally to an electrocatalytic membrane filtration system for preferentially reducing nitrate to ammonia in a water stream, the electrochemical membrane filtration system comprising: a. a water source comprising nitrate; b. a pump for pumping the water source into a reactor cell; c. a reactor cell comprising: i. an inlet for receiving the water source into the reactor cell; ii. a conductive filtration membrane as described in further detail herein, wherein the conductive filtration membrane functions as a cathode; iii. an anode electrically connected to the conductive filtration membrane; and iv. an outlet for removing ammonia and water from the reactor cell.

[0026] In still another embodiment, the present invention relates generally to a method of electrochemically reducing nitrate from a water stream into ammonia, the method comprising the steps of: a) providing an electrocatalytic membrane filtration system as described in further detail herein;b) circulating water containing a concentration of nitrate, wherein the water passes through the conductive filtration membrane containing the ionophore, and wherein the ionophore enhances the interaction between generated nitrite and the catalyst surface, thereby minimizing desorption of nitrite and enhancing overall generation of ammonia; and c) removing permeate comprising ammonia and water.

[0027] In still another embodiment, the present invention relates generally to a method of making a conductive filtration membrane, the method comprising the steps of: a. dispersing metal particles and carbon nanotubes in a resin solution to create a metal particle-carbon nanotube suspension; b. combining an ionophore with a plasticizer to prepare an ionophore mixture, wherein the ionophore exhibits selective adsorption properties to specific ions; c. combining the metal particle-carbon nanotube suspension with the ionophore mixture to prepare a suspension comprising the metal particle- carbon nanotube suspension and the ionophore distributed therein; d. vacuum filtering the resulting mixture onto a substrate, optionally wherein the substrate comprises a microporous ceramic substrate, to produce the conductive filtration membrane; e. drying the conductive filtration membrane at an elevated temperature for a period of time; and f separating the conductive filtration membrane from the microporous ceramic substrate.BRIEF DESCRIPTON OF THE FIGURES

[0028] Features and aspects of embodiments are described below with reference to the accompanying figures, in which elements are not necessarily depicted to scale, and in certain views, parts may have been exaggerated or removed for purposes of clarity. Like reference numbers refer to corresponding parts throughout the figures.

[0029] The following figures are included to illustrate certain aspects of the present disclosure, and should not be viewed as exclusive embodiments. The subject matter disclosed is capable ofconsiderable modifications, alterations, combinations, and equivalents in form and function, as will occur to one having ordinary skill in the art and the benefit of the disclosure.

[0030] Exemplary embodiments of the present disclosure are further described with reference to the appended figures. It is to be noted that the various features, steps and combinations of features / steps described below and illustrated in the figures can be arranged and organized differently to result in embodiments which are still within the scope of the present disclosure.

[0031] Fig. l(a)-Fig. 1(g) depict a characterization of carbon nanotube-based electrified membrane integrating tetraaza-tetraoxa macrocycle ionophore with copper nanoparticles (TTM- CuNP / CNT-EM).

[0032] Figs. 2(a)-2(i) depict the electrochemical nitrate reduction performance of the TTM- CuNP / CNT-EM.

[0033] Figs. 3(a)-3(e) depict a mechanism investigation of TTM in nitrite absorption / desorption and its impact on nitrate reduction product selectivity.

[0034] Fig. 4(a)-4(f) depict nitrite binding models on CuNP and TTM.

[0035] Fig. 5 depicts pore diameter size distribution of the TTM-CuNP / CNT-EM, obtained from SEM image analysis.

[0036] Fig. 6 depicts HAADF-STEM and corresponding EDS images of the nano-scale morphology of CuNP / CNT-EM.

[0037] Fig. 7 depicts EDS elemental mapping for C, N, and Cu in the TTM-Cu / CNT-EM.

[0038] Fig. 8 depicts XPS spectra of the TTM-CuNP / CNT-EM, with the main elements in the membrane indicated.

[0039] Fig. 9 depicts FTIR spectrum of the TTM, with selected chemical moieties identified.

[0040] Fig. 10 depicts Nyquist plots of the CuNP / CNT-EM and TTM-CuNP / CNT-EM at an applied frequency range from 106Hz to 0.1 Hz. The inset schematic depicts the corresponding equivalent circuit diagram.

[0041] Fig. 11(a) depicts AFM images of CuNP / CNT-EM and Fig. 11(b) depicts AFM images of TTM-CuNP / CNT-EM surfaces. The arithmetic mean roughness of each membrane is marked above the images.

[0042] Fig. 12 depicts a comparison of energy consumption between flow-by and flow-through operational modes under the flow rate of 2 mL / min and current density range from 0.4 to 2.4 mA / cm2with a 10 mM KNCh and 10 mM K2SO4 electrolyte.

[0043] Fig. 13 depicts turnover frequency for nitrate reduction using CuNP / CNT-EM and TTM- CuNP / CNT-EM under a current density range of 0.4 to 2.4 mA / cm2. These experiments were conducted in flow-through operational mode. The turnover frequency (TOF) was obtained using measured electrochemically active surface area and assuming an active site density p (10‘6mol cm’2).

[0044] Fig. 14 depicts a comparison of partial current density for NFh and NCh" under the current density range from 0.4 to 2.4 mA cm’2with a 10 mM KNO3 and 10 mM K2SO4 electrolyte using the TTM-CuNP / CNT-EM.

[0045] Figs. 15(a) and 15(b) depict a comparison of linear sweep voltammetry curves between (a) CuNP-coated and (b) TTM-CuNP-coated electrode at the scan rate of 10 mV / s.

[0046] Fig. 16 depicts a comparison of turnover frequency for nitrate reduction between CuNP / CNT-EM and TTM-CuNP / CNT-EM under the current density range from 0.4 to 2.4 mA / cm2. The turnover frequency is obtained using the mass of Cu and the measured electrochemically active surface area.

[0047] Fig. 17(a) depicts CV curves in a non-faradaic region (0.61 to 0.81 V vs. RHE) at scan rates of 10 to 80 mV s1for CuNP / CNT-EM. Fig. 17(b) depicts linear regression between the charging current differences in the potential window of the CV vs. scan rate for CuNP / CNT-EM for determination of the electrochemically active surface area (ECSA).

[0048] Fig. 18(a) depicts CV curves in a non-faradaic region (0.61 to 0.81 V vs. RHE) at scan rates of 10 to 80 mV s ' for TTM-CuNP / CNT-EM. Fig. 17(b) depicts linear regression between the charging current differences in the potential window of the CV vs. scan rate for TTM- CuNP / CNT-EM for determination of the ECSA.

[0049] Fig. 19 depicts pH changes of the permeate solution under the current density range from 0.4 to 2.4 mA cm’2with a 10 mM KNO3 and 10 mM K2SO4 electrolyte.

[0050] Fig. 20 depicts a photograph of the experimental setup, including the feed solution, pump, pressure gauge, permeate flow, flow-through reactor, electronic balance, and electrochemical workstation.

[0051] Fig. 21 depicts a photograph of the flow-through reactor.

[0052] Fig. 22 depicts binding energies calculated using 6-31G++(d,p) and 6-311++G(d,p) basis sets.

[0053] Fig. 23 depicts the binding energies calculated using PBE and TPSSh functionals.

[0054] Fig. 24 depicts nitrate conversion results using the TTM-CuNP / CNT-EM with wastewater treatment effluent under a flow rate of 2 mL / min and current density of 2.8 mA / cm2.

[0055] Fig 25 depicts an XRD pattern of CuNPs. The Cu{ 111 } surface was the dominant crystal facet in the CuNPs.

[0056] Fig. 26(a) depicts a schematic illustration of the electrified filtration cell including a 25 ml feed chamber with a RuCh-IrCh / Ti mesh anode and a 25 ml permeate chamber with a membrane cathode, with 1 cm spacing between the electrodes. Fig. 26(b) depicts a schematic illustration of the process for removing oxygen generated by the oxygen evolution reaction at the anode.

[0057] Fig. 27 depicts a schematic illustration of the difference between flow-through and flow- by modes.

[0058] Fig. 28 depicts the influence of CuNP amount (mg) on membrane permeate flux (L h'1m'2bar1) and nitrate conversion (%) under a current density of 2.8 mA / cm2with a 10 mM KNO3 and 10 mM K2SO4 electrolyte.

[0059] Fig. 29 depicts the influence of TTM amount (mg) on membrane permeate flux (L h'1m'2bar1) and NH3 selectivity (%) under a current density of 2.8 mA / cm2with a 10 mM KNO3 and 10 mM K2SO4 electrolyte.

[0060] Figs. 30(a) to 30(c) depict chronopotentiometric data (vs. Ag / AgCl) for Figs. 2(b) to 2(e).

[0061] Figs. 31(a) and 31(b) depict selectivity (%) ofNCh’-N and NH3-N products over CuNP / CNT-EM (Fig. 31 (a)) and TTM-CuNP / CNT-EM (Fig. 31(b)) under different cathodic potentials. The potentials were obtained from the chronopotentiometric data presented in Figs. 30(a) and 30(b).

[0062] Fig. 32 depicts the influence of membrane thickness (pm) on nitrate conversion (%) under a current density of 2.8 mA / cm2with a 10 mM KNO3 and 10 mM K2SO4 electrolyte.

[0063] Fig. 33 depicts NH3 production with different nitrite concentrations in the feed solution under a potential of -2.1 V vs. Ag / AgCl using a CuNP / CNT-EM.

[0064] Fig. 34 depicts a top view of the SEM images of the CuNP / CB.

[0065] Fig. 35 depicts a comparison of the permeate flux between the CuNP / CNT-EM and CuNP / CB.

[0066] Fig. 36 depicts the long-term chronopotentiometry stability tests (n=3) of the TTM- CuNP / CNT-EM using real surface water over 100 hours.

[0067] Fig. 37 depicts NH3 faradaic efficiency of the TTM-CuNP / CNT-EM under different current densities (0.8 to 2.8 mA cm'2) in flow-through mode.

[0068] Fig. 38 depicts a comparison of turnover frequency for nitrate reduction between CuNP / CNT-EM and TTM-CuNP / CNT-EM under the current density range from 0.4 to 2.4 mA cm'2. These experiments were conducted in flow-through operational mode.

[0069] Fig. 39 depicts pH changes of the permeate solution under the current density range from 0.4 to 2.4 mA cm'2with a 10 mM KNO3 and 10 mM K2SO4 electrolyte in flow-through mode.

[0070] Fig. 40 depicts a preliminary techno-economic analysis in accordance to one aspect of the present invention.

[0071] Fig. 41 depicts a comparison of NO ' adsorption and desorption between CuNP and TTM-loaded CuNP over 3000 s using quartz crystal microbalance (QCM) measurements. The solution was changed from ultrapure water to NaNO? solution, and then back to ultrapure water.

[0072] Fig. 42 depicts frequency changes due to NO?' adsorption and desorption between CuNP and TTM-loaded CuNP over 3000 s obtained from quartz crystal microbalance measurements.

[0073] Fig. 43 depicts a comparison of NO?' adsorption and desorption by TTM-loaded CuNP over 3000 s using quartz crystal microbalance (QCM) measurements. Mass changes were calculated from the measured frequency changes. The solution was changed from ultrapure water to NaNC solution, and then back to ultrapure water. Inset: Binding energy between NO?' and TTM.DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0074] The present invention relates generally to a new technology that includes a conductive filtration membrane comprising carbon nanotubes and metal particles, preferably metal nanoparticles and incorporates an ionophore as a key non-catalytic cooperative nitrite-enriching component. The use of the membrane platform enables more efficient electrochemical conversion of nitrate due to elimination of diffusional mass transport limitations during electrofiltration through the nanoporous membrane, resulting in much higher conversion activity compared to traditional flow-by mode.

[0075] Due to the properties of the ionophore for enhancing nitrite adsorption / reducing desorption independent of the applied potential, incorporating the ionophore into the conductive filtration membrane enriches the concentration of nitrite that is desorbed from the catalyst tofacilitate re-adsorption and further conversion to ammonia. As a result, the ammonia selectivity is significantly enhanced. Overall, this system achieves a high energy efficiency for ammonia generation from nitrate electrochemical reduction.

[0076] As used herein, “a,” “an,” and “the” refer to both singular and plural referents unless the context clearly dictates otherwise.

[0077] As used herein, the term “about” refers to a measurable value such as a parameter, an amount, a temporal duration, and the like and is meant to include variations of + / -15% or less, preferably variations of + / -10% or less, more preferably variations of + / -5% or less, even more preferably variations of + / -!% or less, and still more preferably variations of + / -0.1% or less of and from the particularly recited value, in so far as such variations are appropriate to perform in the invention described herein. Furthermore, it is also to be understood that the value to which the modifier “ about” refers is itself specifically disclosed herein.

[0078] As used herein, spatially relative terms, such as “beneath,” “below,” “lower,” “above,” “upper,” “front,” “back,” and the like, are used for ease of description to describe one element or feature's relationship to another element(s) or feature(s). It is further understood that the terms “front” and “back” are not intended to be limiting and are intended to be interchangeable where appropriate.

[0079] As used herein, the terms “comprises” and / or “comprising,” specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0080] In one embodiment, the present invention relates generally to a conductive filtration membrane, the conductive filtration membrane comprising: a. carbon nanotubes; and b. metal particles preferably wherein the metal particles comprise copper particles, more preferably wherein the copper particles comprise copper nanoparticles; and c. an ionophore that exhibits selective adsorption properties to specific ions, wherein the ionophore is interspersed with the metal particles; and

[0081] In one embodiment, the carbon nanotubes, copper nanoparticles and the ionophore are substantially evenly dispersed throughout the conductive filtration membrane. In one embodiment, the ionophore enhances conversion of nitrate to ammonia during nitrateelectrocatalytic reduction by enhancing the interaction between generated nitrite and the catalyst surface, thereby minimizing desorption of nitrite and enhancing overall generation of ammonia.

[0082] The present invention uses a nanoporous conductive filtration membrane to overcome mass transport limitations and enable high activity for nitrate reduction. To address the production of undesirable nitrite, a nitrite-selective ionophore is incorporated into the copper catalyst-functionalized membrane. The ionophore enhances nitrate adsorption, enriching the concentration of nitrite in the electrocatalyst microenvironment to enable re-adsorption to the catalyst for further reduction to ammonia.

[0083] In one embodiment the carbon nanotubes have a diameter within the range of about 1 to about 25 nm, more preferably about 5 to about 15 nm and have a length within a range of about 1 to about 50 pm, more preferably about 2 to about 20 pm. In one embodiment, the carbon nanotubes comprise multiwalled carbon nanotubes.

[0084] It was also discovered that carbon black (CB) cannot form a nanoporous membrane due to the lack of an interwoven structure, and SEM images of the CuNP / CB are depicted in Fig. 34. Additionally, in comparison to CNTs, using carbon black would necessitate a greater amount of electrically insulating binder (0.3 wt% PAN used for the CNT-based membranes), which could lead to an unstable membrane with overall higher electrical resistance. To further evaluate the characteristics of CNT-based and CB-based samples, experiments were conducted to investigate the permeate flux of these two substrates. These results are depicted in Fig. 35 and show that the CNT-based membrane exhibited a high permeability (-265 L h1m2bar1), whereas the CB- based sample was completely impermeable.

[0085] In one embodiment, the metal particles may be selected from the group consisting of copper, iron, nickel, tin, and tungsten metals. Other similar metals and metal alloys may also be used. In one embodiment, the metal particle comprise nanoparticles, which metal nanoparticles may be selected from the group consisting of copper nanoparticles, nickel nanoparticles, tin nanoparticles, tungsten nanoparticles, and combinations of the foregoing. In one embodiment, the metal nanoparticles comprise copper nanoparticles. In one embodiment, the metal nanoparticles have a largest dimension that is within a range of about 1 to about 100 nm, preferably about 5 to about 75 nm, more preferably about 10 to about 50 nm. In one embodiment, the shape of the metal nanoparticles may be selected from substantially spherical particles, platelet, nanotubes and nanowires, among others. Preferably the shape of the metalnanoparticles is at least substantially spherical and the metal nanoparticles have a diameter that is within the range of about 1 to about 100 nm, preferably about 5 to about 75 nm, more preferably about 10 to about 50 nm.

[0086] The loading of CuNP is preferably within the range of about 10 to about 50 mg, more preferably about 25 mg to about 35 mg. In one embodiment, the loading of Cu NP is about 30 mg of CuNPs. It was found that this amount achieves a high nitrate conversion rate (-95%) while also maintaining the highest permeate flux (-260 L h1m2bar1, as shown in Fig. 28). Increasing the CuNP loading beyond 30 mg did not obviously enhance nitrate conversion; rather, it decreased permeate flux, thereby compromising fdtration efficiency. Thus, 30 mg was determined to be the optimal balance between nitrate conversion and permeate flux. Similarly, the loading of the ionophore is preferably within the range of about 5 to about 20 mg, more preferably about 10 to about 15 mg. In one embodiment, a loading of TTM of about 10 mg was chosen because it achieved -95% NH3 selectivity while preserving the same high permeate flux, ensuring an optimal balance between selectivity and filtration efficiency. Further increasing the TTM loading would not result in a significant increase in NH3 selectivity but would instead reduce the permeate flux. Fig. 28 and Fig. 29 depict the influence of CuNP loading and TTM respectively on membrane permeate flux and nitrogen conversion.

[0087] Furthermore, membrane thickness was not found to significantly impact nitrate reduction performance, as shown in Fig. 32. However, in one embodiment, the thickness of the membrane is within the range of about 10 to about 100 pm, more preferably about 25 to about 75 pm. This thickness can change depending on a variety of factors, including, for example, the desired flow rate through the membrane, the desired conversion rate, the initial nitrate concentration, the size of the electrocatalytic filtration system, among others.

[0088] It was determined that residence time is primarily controlled by flow rate rather than membrane pore volume (i.e., membrane thickness) under the conditions tested in the examples below. Additionally, at high conversion rates facilitated by the membrane, altering thickness has no effect, as conversion is no longer limited by residence time.

[0089] In one embodiment, the specific ions comprise nitrite ions and wherein the ionophore enhances the interaction between generated nitrite and the catalyst surface.

[0090] The invention described herein incorporates an ionophore as a non-catalytic cooperative component to influence reaction selectivity without directly modifying the catalyst. In oneembodiment, the ionophore comprises a salen or salophen ionophore. Suitable salen or salophen ionophores may be selected from the group consisting of a tetraaza-tetraoxa macrocycle, crown ethers known for their high affinity for specific cations, and calixarenes effective in selectively binding anions, and combinations of the foregoing, which all could improve electrochemical reaction efficiency in ion-based processes. Similarly, metal-organic frameworks with tunable pore structures could selectively stabilize or capture intermediates in reactions such as CO2 reduction, targeting unstable species such as carboxyl (*COOH) or formate (HCOO ) to enhance selectivity for desired products. In one embodiment, the ionophore comprises a tetraaza-tetraoxa macrocycle.

[0091] In one embodiment, the tetraaza-tetraoxa macrocycle has the following Structure:

[0092] As described herein, the present invention introduces a new approach for achieving ultrafast and highly selective conversion of nitrate to ammonia through the use of a conductive filtration membrane that, in one preferred embodiment, incorporates tetraaza-tetraoxa macrocycle (TTM) ionophore-modified copper nanoparticles (CuNP) into a carbon nanotube (CNT)-based electrified membrane (TTM-CuNP / CNT-EM).

[0093] Because copper is a suitable material for nitrate-to-ammonia, copper was used in the examples herein to demonstrate the function of an ionophore-modified electrofiltration system over existing copper-based systems. However, other metal particles, including other metal nanoparticles, would also be usable in the practice of the invention. In one embodiment, the present invention describes an electrocatalyst in which a cooperative adsorption component (i.e., the TTM ionophore) is incorporated and functionalizing a nanoporous water treatment membrane with these materials, as well as elucidating the mechanisms involved in the performance enhancements with this system.

[0094] However, the mechanisms described herein extend beyond copper-based catalysts. Instead, the invention described herein provides a framework that can be extended to a wide variety of catalysts within the nitrate reduction field and beyond. By manipulating the local reaction microenvironment and mitigating nitrite release from the microenvironment to the bulk electrolyte, the present invention offers a new approach to addressing common challenges associated with nitrate-to-ammonia conversion.

[0095] The ionophore plays a crucial role as a cooperative adsorption component, enriching the concentration of NO ' within the catalyst microenvironment and thereby significantly improving NH3selectivity without directly altering the catalytic active site. This molecular-level manipulation of intermediate NO?' species during the reaction represents a new paradigm for achieving efficient NCh' to NH3 electrochemical conversion.

[0096] In one embodiment, the conductive filtration membrane has an average pore size in the range of about 10 to about 100 nm, more preferably in the range of about 45 to about 80 nm.

[0097] In one embodiment, the conductive filtration membrane is configured to exhibit a stability of at least 100 hours or at least 110 hours or at least 120 hours or at least 130 hours or at least 150 hours or at least 200 hours or at least 300 hours or more as measured by a chronopotentiometry stability test of the conductive filtration membrane when the conductive filtration membrane is placed in a flow-through reactor cell at 0.8 mA / cm2.

[0098] In another embodiment, the present invention also relates generally to an electrocatalytic membrane filtration system for preferentially reducing nitrate to ammonia in a water stream, the electrocatalytic membrane filtration system comprising: a. a water source comprising nitrate; b. a pump for pumping the water source into a reactor cell; c. a reactor cell comprising: i. an inlet for receiving the water source into the reactor cell; ii. a conductive filtration membrane as described in further detail herein, wherein the conductive filtration membrane functions as a cathode; iii. an anode electrically connected to the conductive filtration membrane;; and iv. an outlet for removing ammonia and water from the reactor cell.

[0099] In one embodiment, the electrodes are fixed to maintain an electrolyte-filled gap between the electrodes, and the flow-through mode prevents reduced products from re-oxidation on the upstream anode. The anode and cathode are positioned within the reactor cell so that there is a gap filled with electrolyte between them, which prevents short circuiting.

[0100] Due to the applied pressure, water continuously flows first through the mesh anode and then permeates through the cathodic membrane. The flow-through mode allows for direct transport of electrolyte ions from the anode to the cathode, enabling charge balance in the absence of an ion exchange membrane. Additionally, the flow-through mode prevents back- transport of solution, so any reduced species that have permeated through the cathode exit the cell and cannot transport back to the anode, eliminating re-oxidation of reduced species.

[0101] In one embodiment, the system comprises a feed tank for housing the water. In one embodiment the pump is configured to pump the water stream from the feed tank into the reactor cell. The reactor cell may also be configured to return a portion of the feed stream to the feed tank.

[0102] In one embodiment, the anode is a mesh anode. In general, the anode material may be any material that can be used as the oxidation counter electrode, such as for oxygen evolution. In one embodiment, the mesh anode is a mixed metal oxide mesh. The mixed metal oxides may include, for example, RuCh, IrCh, TiCh, Ta2Os, other similar metal oxide, and combinations of any of the foregoing, on a titanium or other similar substrate.

[0103] In one embodiment, the mixed metal oxide mesh is ruthenium-iridium oxide coated titanium (RuCh-IrCh / Ti) mesh, which is an effective anode material due to its high activity for the oxygen evolution reaction (OER) and high stability. The combination of ruthenium dioxide (RuCh) and iridium dioxide (IrCh) provides high electrochemical stability and excellent conductivity, which are crucial for maintaining structural integrity and facilitating efficient electron transfer during electrochemical reactions. Additionally, these oxides exhibit high catalytic activity, particularly for OER, making them ideal for processes like water splitting.

[0104] The synergy between Ru02 and IrO2 enhances the overall performance of the anode, with RUO2 offering superior conductivity and catalytic activity, while IrO2 contributes to enhanced stability and durability. The titanium (Ti) substrate further adds to the anode’s effectiveness by providing robust corrosion resistance, thanks to its stable passive oxide layer. This combinationensures that the anode can operate in harsh environments without degrading, while also reducing the overpotential for oxygen evolution, thereby improving energy efficiency.

[0105] In another embodiment, the present invention also relates generally to a method of electrochemically reducing nitrate from a water stream into ammonia, the method comprising the steps of a) providing a conductive membrane filtration system as described in further detail herein; b) circulating water containing a concentration of nitrate through the reactor cell, wherein the water passes through the conductive filtration membrane containing the ionophore, and wherein the ionophore enhances the interaction between generated nitrite and the catalyst surface, thereby minimizing desorption of nitrite and enhancing overall generation of ammonia; and c) removing permeate comprising ammonia and water from the reactor cell.

[0106] In one embodiment, the water comprises nitrate at a concentration in the range of about 1 to about 500 millimolar, more preferably about 1 to about 100 millimolar, more preferably about to 10 about 100 millimolar.

[0107] In one embodiment, the permeate flowrate of the water through the electrocatalytic membrane filter system is in the range of about 1 ml / min to about 10 ml / min, more preferably about 1 ml / min to about 5 ml / min. However, this can be easily changed based in part on the applied pressure as well as the membrane thickness and pore volume / permeability.

[0108] In addition, the cross-flow rate over the membrane is around 200 ml / min, but can be much smaller (even 0) or larger. The permeate flow rate (ml / min) is an extrinsic property, where the permeability (volume / time / membrane area / pressure) is intrinsic. The typical permeate flow rate is about 1 to about 5 ml / min, preferably about 1.5 to about 2.5 ml / min, but this may change based on changing the pressure or the membrane area without modifying the membrane materials.

[0109] In one embodiment, the conversion of nitrate to ammonia is greater than about 60% or greater than about 65% or greater than about 70% at a current density within the range of about 0.4 to about 2.4 mA / cm2.

[0110] In one embodiment, the conductive filtration membrane exhibits a Faradaic efficiency of nitrate reduction towards nitrite and ammonia products of greater than 60% of ammonia and less than 25% of nitrite.

[0111] In one embodiment, the conductive filtration membrane exhibits a stability of at least about 100 hours or at least about 110 hours or at least about 120 hours or at least about 130 hours or at least about 150 hours or at least about 200 hours or at least about 300 hours or more as measured by a chronopotentiometry stability test of the conductive filtration membrane in reactor cell at 0.8 mA / cm2

[0112] In another embodiment, the present invention also relates generally to a method of making a conductive filtration membrane, the method comprising the steps of: a. dispersing metal particles and carbon nanotubes in a resin solution to create a metal particle-carbon nanotube suspension; b. combining an ionophore with a plasticizer to prepare an ionophore mixture, wherein the ionophore exhibits selective adsorption properties to specific ions; c. combining the metal particle-carbon nanotube suspension with the ionophore mixture to prepare a suspension comprising the metal particle-carbon nanotube suspension and the ionophore distributed therein; d. vacuum filtering the resulting mixture onto a substrate, optionally wherein the substrate comprises a microporous ceramic substrate to produce the conductive filtration membrane; e. drying the conductive filtration membrane at an elevated temperature for a period of time; and f. separating the conductive filtration membrane from the microporous ceramic substrate.

[0113] In one embodiment, wherein the resin solution comprises a resin, preferably a thermoplastic resin dispersed in an organic solvent. In one embodiment, the thermoplastic resin is selected from the group consisting of acrylonitrile-butadiene-styrene (ABS) resins, polystyrene, acrylonitrile-styrene (AS) resins, acrylic resins and polyacrylonitrile resins. In one embodiment, the resin is a polyacrylonitrile resin. The organic solvent may be selected from the group consisting of N-methylpyrrolidone (NMP), N,N-dimethylacetamide (DMAC), dimethylformamide (DMF), diethylformamide, y-butyrolactone (GBL), dimethylsulfoxide(DMSO), and combinations of the foregoing. In one embodiment, the solvent comprises dimethylformamide.

[0114] In one preferred embodiment, the resin system comprises a polyacrylonitrile resin dispersed in dimethylformamide. In one embodiment, the concentration of the resin in the solvent is in the range of about 0.01 to about 1.0 wt.%, more preferably about 0.05 to about 0.5 wt.%, more preferably about 0.1 to about 0.3 wt. %. In one exemplary embodiment, the resin system comprises about 0.1 to about 0.3 wt.% polyacrylonitrile resin dispersed in dimethylformamide.

[0115] As discussed above, in one embodiment, the ionophore comprises a salen or sal ophen ionophore, preferably a tetraaza-tetraoxa macrocycle. In one embodiment, the ionophore is combined with a plasticizer, and the plasticizer may be selected from the group consisting of di- n-decyl phthalate, polypropylene glycol alkylphenyl ether), dibutyl phthalate, isodecyl pelargonate, cyclohexyl pyrrolidone, dioctyl phthalate, and / or di -n-decyl phthalate, preferably the plasticizer comprises dibutyl phthalate. In one embodiment, the ratio of ionophore to plasticizer is in the range of about 10: 1 to about 1 : 1, more preferably about 3: 1 to about 1 : 1.

[0116] In one embodiment, the membrane contains a total mass loading of about 60 grams, wherein about 30 mg are copper nanoparticles, 20 mg are carbon nanotubes, and 10 mg are the ionophore / plasticizer mixture in a ratio of about 2:1. Other mass loadings, including different amounts of metal nanoparticles, carbon nanotubes, and ionophore / plasticizer mixture are also usable in the instant invention and will depend in part on the particular types of metal nanoparticles, carbon nanotubes, ionophores, and plasticizers used.

[0117] In one embodiment, the substrate is a microporous ceramic membrane that has a pore size of about 0.5pm to about 10 pm, more preferably about 0.7 pm to about 3 pm and a diameter of about 25 mm to about 100 mm, more preferably about 40 to about 75 mm.

[0118] The diameter of the substrate is generally not critical as the membrane is scalable and may be scaled up. Likewise, the pore size of the substrate is also highly variable so long as it is small enough to support the conductive filtration membrane and large enough so that pores are larger than those of the conductive filtration membrane.

[0119] In one embodiment, the conductive filtration membrane is dried for a period of time at an elevated temperature. For example, in one embodiment, the conductive membrane is dried forabout 2 to about 24 hours, preferably about 6 to about 12 hours, more preferably about 8 hours at a temperature of about 50 to about 120°C, more preferably about 80 to about 100°C.

[0120] While ionophores exhibiting strong affinity for specific ions have been employed in ionspecific sensors, these materials have not been frequently explored for controlling ion binding in electrocatalysis. Since ionophores facilitate the diffusion of target ions into sensor matrices, their bonding capabilities could potentially influence the adsorption and desorption of weakly-bound intermediates such as NCh’ during nitrate electrocatalytic reduction. As an example, salen and salophen ionophores have been demonstrated to effectively bond with NCh' ions due to their tetradentate Schiff base properties.

[0121] The NCh' can form H—O bonds with amine groups in these ionophores, resulting in a reduced dipole moment and uniform electrostatic attraction. By incorporating these ionophores into NC ' reduction electrocatalysts, the interaction between generated NCh' and the catalyst surface might be enhanced, effectively minimizing desorption of NCh' and transport away from the surface to enhance the overall NH3 generation efficiency. Since NC ' reduction in EMs suffers from significant NCh' byproduct formation (despite the high NCh' conversion activity achieved in the efficient electrofiltration mass transport mode), employing ionophore-modified electrocatalysts in EMs could potentially achieve highly efficient conversion of NCh' to NH3.

[0122] In one embodiment, the invention described herein introduces a new approach to optimize nitrate-to-ammonia conversion with low energy consumption by incorporating tetraazatetraoxa macrocycle (TTM) ionophore-modified copper nanoparticles (CuNP) into a carbon nanotube (CNT)-based electrified membrane (TTM-CuNP / CNT-EM).

[0123] In one embodiment, the metal particles comprise copper particles, more preferably the copper particles comprise copper nanoparticles, which is a common electrocatalyst for nitrate reduction to ammonia.

[0124] In one embodiment, it was found that the flow-through TTM-CuNP / CNT-EM demonstrated the most competitive energy consumption for nitrate reduction to ammonia of 21.4 kWh kg-NEh'1under low power consumption conditions with stability over at least 115 hours. The TTM-CuNP / CNT-EM exhibited significantly higher NH3 selectivity compared to the ionophore-free CuNP / CNT-EM due to inhibiting the desorption of the stable intermediate NCh’.

[0125] The system achieves high energy efficiency for ammonia generation from electrochemical nitrate reduction (g-NEL per kWh), outperforming systems that use expensiveprecious metals. Due to the mild operating conditions and mechanical and chemical stability of the membrane, the system is promising for practical modular applications and for direct use with intermittent renewable energy sources. The system also operates with very low power consumption (low current).

[0126] In one preferred embodiment, as described herein, the system does not include an ion exchange membrane. Instead, it is designed as an ion exchange membrane-free setup. The operation of this ion exchange membrane-free system is managed by a cross-flow configuration, where the water flow effectively removes any oxygen generated by the oxygen evolution reaction (OER) on the upstream anode, preventing it from impacting the cathode.

[0127] The treatment capacity of the electrofiltration membrane is determined by the membrane permeability, membrane area, and applied pressure. The electrofiltration membrane demonstrates a permeability consistent with micro- or nanofiltration of 263.5 L h’1m'2bar1, making it highly suitable for scaling up to large-scale applications. The treatment capacity can then be adjusted based on the membrane size and applied pressure. Additionally, the mechanical flexibility of the membrane facilitates its scalability, with the potential to be integrated into spiral-wound modules, a commonly used format for industrial membrane technologies (in contrast to typical electrochemical systems). This adaptability could facilitate translating the membrane to larger systems without compromising performance, positioning it as a promising solution for widespread use in water treatment facilities or decentralized modular treatment systems.

[0128] The ammonia generated in this system may need to be concentrated and recovered, enabling the generation of valuable, more concentrated NEE from dilute nitrate streams, providing carbon-neutral or carbon-free ammonia production technologies.

[0129] In addition, ammonia has great potential for sustainable energy storage. While producing N2 is a reasonable alternative (and likely would be preferable for many drinking water treatment applications), achieving highly efficient ammonia generation from dilute nitrate streams (such as in this study) can provide a potentially promising alternative for simultaneous water treatment and production of fossil-free ammonia. By combining nitrate removal with ammonia production, our electrofiltration membrane not only addresses important water contamination issues but also could contribute to sustainable energy storage, thereby providing a fossil-free source of ammonia for fuel as an additional value stream. Selective nitrate reduction to N2 is important for scenarios where ammonia generation is not needed or ammonia handling is too complicated; however,nitrate reduction to NH3 provides potentially exciting opportunities to generate additional economic incentives for nitrate removal while also supplying a means of generating green ammonia. Additionally, highly selective and rapid systems for nitrate reduction to ammonia can also be applied in schemes involving tandem N2 oxidation to NOXfollowed by reduction of the NOx absorbed in water to NH3, further enabling green ammonia technologies.

[0130] For low-concentration nitrate (e.g., 10 mM), the main limiting factor for reaction is generally mass transport rather than intrinsic catalytic activity. This is connected to the term “ultrafast”, which refers to the significantly reduced observed reaction time achieved through the flow-through mode of operation, rather than the intrinsic kinetics of the copper catalyst (most electrochemical nitrate reduction studies measure TOFs that are rate-limited by diffusion). The flow-through mode mitigates mass transport limitations that typically restrict the conversion of low-concentration nitrate, improving local nitrate availability near the membrane surface and facilitating observed reaction kinetics that are similar to intrinsic reaction kinetics. Furthermore, the flow-through mode increases the collision frequency of nitrate molecules with active sites within the membrane due to advection, enhancing the nitrate reaction rate. As a result, integrating the flow-through mode with the membrane described herein resulted in a significantly higher TOF and observed nitrate conversion rate compared to other systems.

[0131] The invention will now be described with reference to the following non-limiting examples.Example 1. Synthesis and characterization of ionophore-modified electrocatalytic membrane:

[0132] Multi-walled carbon nanotubes (MWNTs, CNTs) were purchased from Millipore Sigma (No. 698849). The CNTs were synthesized using chemical vapor deposition (CVD). During this process, a volatile precursor undergoes thermal decomposition at elevated temperatures, resulting in the formation of a solid deposit on a substrate. The CNTs have diameters ranging from 6 to 13 nm, with an average length of 10 pm. The multi-walled CNTs remain stable in an inert atmosphere at temperatures up to 3697°C. Following the CVD process, HC1 demineralization is performed.

[0133] Chemicals and materials used in the Examples below included copper nanoparticles (CuNPs, particle size: 25 nm, Sigma-Aldrich), CNTs (multiwalled carbon nanotubes, >98%carbon basis, outer diameter x length = 6-13 nm x 2.5-20 gm, inner diameter 2-6 nm, Sigma- Aldrich, more details in Supplementary Note 1), N,N-dimethylformamide (DMF, >99.8%), polyacrylonitrile (PAN, average Mw = 150,000), tetraaza-tetraoxa macrocycle ionophore (TTM), dibutyl phthalate (DBP, 99.5%), all of which were available from Sigma Aldrich, microporous ceramic membrane substrate (pore size: 0.8 pm, diameter: 46 mm, Sterlitech, Kent, WA, USA), and RuCh-IrCh / Ti mesh (Thickness: 2.5 mm, UTron Technology Co. Ltd). All chemicals were of analytical grade and were used without further purification. All solutions in the Examples below were prepared using ultrapure water.

[0134] A flow-through carbon nanotube-based electrified membrane integrating tetraaza- tetraoxa macrocycle ionophore with carbon nanoparticles (TTM-CuNP / CNT-EM) was fabricated by combining tetraaza-tetraoxa macrocycle ionophores (TTM) and copper nanoparticles (CuNP) with CNTs and polyacrylonitrile (PAN) in a dimethylformamide (DMF) suspension, followed by vacuum filtering onto a microporous ceramic substrate with a total material loading amount of 60 mg.

[0135] The CuNPs and CNTs were dispersed in a poly-acrylonitrile solution (0.1 wt% in dimethylformamide, DMF) to create a CuNP-CNT suspension, maintaining a mass ratio of CuNP to CNTs at 3:2. This solution was then subjected to ultrasonication for 1 hour. The 9- hexadecyl-1,7,1 l,17-tetraoxa-2,6,12,16-tetraazacycloeicosane, known as tetraaza-tetraoxa macrocycle ionophore (TTM), was combined with the plasticizer dibutyl phthalate (DBP) to prepare an ionophore mixture with a mass ratio of TTM to DBP at 2: 1. This mixture was ultrasonicated for 30 min to ensure homogeneity. DBP served the role of ensuring the homogeneous distribution of ionophores within the membrane. The CuNP-CNT solution and the ionophore mixture were combined and then sonicated for an additional 12 minutes using an ultrasonic probe. The resulting suspension was vacuum filtered onto a microporous ceramic membrane substrate (pore size: 0.8 pm, diameter: 46 mm, Sterlitech, Kent, WA, USA), with a total material loading of 60 mg. Finally, the obtained TTM-CuNP / CNT-EM was rinsed using DI water and dried at 90°C for 8 hours.

[0136] The CuNP / CNT-EM was fabricated through the above method, but without incorporating the ionophore mixture, maintaining the same CuNP mass loading of 30 mg.

[0137] 30 mg of CuNPs achieved high nitrate conversion (-95%) while also maintaining the highest permeate flux (-260 L h'1m’2bar1, as shown in Fig. 28. Similarly, a loading of TTMwas 10 mg, which achieved -95% NHi selectivity while maintaining the same high permeate flux, as shown in Fig. 29, ensuring an optimal balance between selectivity and filtration efficiency.

[0138] Membrane Characterization;

[0139] Scanning Electron Microscope (SEM) (SU8230, Hitachi) coupled with energy-dispersive X-ray spectroscopy (EDS) (XFlash 5060FQ, Bruker) was employed to investigate the surface and cross-section morphologies as well as the elemental distribution of the TTM-CuNP / CNT- EM. The morphology of the CNT-EM surface and the thickness of the CNT-EM were characterized using SEM at an accelerating voltage of 10.0 kV. Specifically, the CNT-EM was submerged in liquid nitrogen for approximately 60 s and then was taken out and immediately cut into semicircle-shaped samples (radius 5 mm) to generate a cross-sectional view field. The average pore size and thickness of the CNT-EM were determined using FIJI software. Water contact angles were measured by the sessile drop method using a contact angle goniometer (Oneattension, Biolin Scientific). Permeate flux was calculated by dividing the permeate flow rate by the effective membrane area. Membrane surface roughness was analyzed by an atomic force microscope (Dimension Fastscan, Bruker) operated in tapping mode.

[0140] Raman spectra were acquired using a Raman microscope (HR-800, Horiba Jobin Yvon) and a 50x magnification objective (W.D.=0.5 mm, N.A.=1). Prior to measurements, the spectrometer was calibrated against a pure Si sample to ensure the variation in the silicon phonon mode (520.7 cm’1) was kept under 1 cm’1. A piece of TTM-CuNP / CNT-EM was cut into the dimensions of -0.5 cm x 0.5 cm and mounted onto a glass slide using double-sided tape for characterization. Corresponding Raman spectra were obtained using 10 mW of power delivered by a 633 nm laser to the sample while the laser spot was maintained at - 1 pm. Each spectrum was obtained within -40 s over a range of 200 to 2200 cm-1 with an exposure time of 8 s and a repetition of 2 to eliminate potential spikes from cosmic rays. LabSpec 6 software was used to identify peaks in obtained Raman spectra, and baseline subtraction was conducted in Origin.

[0141] XPS was carried out using a scanning X-ray photoelectron spectroscopy (XPS) microprobe (PHI VersaProbe II, Physical Electronics) with Al Ka radiation to analyze the surface elements of the TTM-CuNP / CNT-EM. The XPS measurements were taken in ultra-high vacuum conditions to prevent contamination. Before starting the analysis, the sample surface was cleaned using an argon ion beam.

[0142] Transmission electron microscope (TEM) images and EDS mappings were taken by an FEI Titan Themis spherical aberration corrected S / TEM coupled with four silicon Supe-X drift detectors operated at 300 kV. This TEM is configured with an extreme field emission gun (X- FEG) and a DCOR plus probe Cs corrector. In the HAADF STEM mode, the collection angles were 40-200 mrad, with a semi convergence angle of 25.2 mrad. The atomic fraction of different elements was analyzed via Velox software.

[0143] Electrochemical measurements:

[0144] Electrochemical measurements were performed using an electrochemical workstation (CHI 660E potentiostat) in a typical three-electrode electrochemical cell containing the working electrode, a RuCh-IrCh / Ti mesh as the counter electrode, and an Ag / AgCl electrode as the reference electrode. All potentials in this study were measured against the Ag / AgCl and converted to the RHE reference scale by E(V vs. RHE) = E(V vs. Ag / AgCl) + 0.059*pH + 0.197. Linear sweep voltammetry (LSV) measurements were recorded at a scan rate of 10 mV / s. Electrochemically active surface area (EC SA) was determined by measuring a series of cyclic voltammograms within the non-faradaic potential region of 0.61 to 0.81 V (vs. RHE) under different scan rates (20, 30, 40, 50, 60, 70, and 80 mV / s). Electrochemical impedance spectroscopy (EIS) measurements were conducted in the frequency range from 106Hz to 0.1 Hz using an excitation amplitude (AEac) of 10 mV. Data were fitted based on the equivalent circuit models in ZSimpWin software.

[0145] Flow-through experiments:

[0146] Flow-through experiments were conducted using an electrochemical cross-flow membrane filtration system as show in Fig. 20. The flow-through reactor cell used in the Examples included a 25 mL feed chamber with a RuCh-IrCh / Ti mesh anode and a downstream 25 mL permeate chamber with a TTM-CuNP / CNT-EM cathode, as shown in Fig. 21. The system does not include an ion exchange membrane. The electrodes are fixed to maintain an electrolyte- filled gap between the electrodes, and the flow-through mode prevents reduced products from reoxidation on the upstream anode. Additionally, the high cross-flow velocity of the feed over the upstream anode effectively removes oxygen generated by the oxygen evolution reaction (OER), preventing it from impacting the cathode. By connecting with an electrochemical workstation (CHI 660E potentiostat), the RuCh-IrCh / Ti mesh served as the anode and the TTM-CuNP / CNT- EM served as the cathode with an effective area of 12.6 cm2. The flow-through experiments wereconducted at a cross-flow rate of 200 mL / min with a recirculating feed solution (10 mM KNO3 and 10 mM K2SO4) using a peristaltic pump (Masterflex, Cole-Parmer) at room temperature (25 °C). The permeate flux was controlled by adjusting the applied pressure using a valve and a pressure gauge (±0.2 bar). Feed water passed through the membrane pores under the applied pressure to obtain treated permeate water, as shown in Figs. 26(a) and 26(b).

[0147] Fig. 26(a) depicts a schematic illustration of the electrified filtration cell including a 25 ml feed chamber with a RuCh-IrCh / Ti mesh anode and a 25 ml permeate chamber with a membrane cathode, with 1 cm spacing between the electrodes and Fig. 26(b) depicts a schematic illustration of the process for removing oxygen generated by the oxygen evolution reaction at the anode.

[0148] Flow-by experiments used the same setup, but the cross-flow rate was reduced to match the permeate flow rate used in corresponding flow-through tests (e g., 2 mL / min) using the peristaltic pump, and no pressure was applied to prevent permeation through the membrane as shown in Fig. 27. The same current density, controlled by the CHI 660E potentiostat, was applied to the membrane under both flow modes.

[0149] Fig. 37 depicts the NH3 faradaic efficiency of the TTM-CuNP / CNT-EM cathode under different current densities (0.8 to 2.8 mA cm'2) in flow-through mode.

[0150] Quartz crystal microbalance (QCM) tests:

[0151] Partitioning of nitrite in the TTM was investigated using a quartz crystal microbalance with dissipation (QCM-D). The TTM-CuNP and CuNP were coated on clean 5 MHz QCM sensors. The analysis was done in a Q-sense flow module with parallel channels (Biolin Scientific, Linthicum Heights, MD). All experiments were conducted at a flow rate of 0.1 mL / min and a temperature of 25°C. The samples were exposed to the following solutions in this order: (i) ultrapure water, (ii) NaNCh solution at pH 7 ± 0.2 with a concentration of 100 mM, and (iii) ultrapure water. Baseline frequencies were assessed for stability (i.e., a rate of change <0.25 ng cm2min ') before exposing the sample to solutes (i.e., 100 mM NaNCh). The variation in frequency between the baseline signal and the equilibrated signal was computed as an indicator of sensor mass changes.

[0152] Faradaic efficiency, yield rate, and energy consumption calculations:

[0153] The Faradaic efficiency (FE) of nitrate reduction towards NCh' and NH3 products in the flow-through mode were calculated as follows:FE = (CAx F x n x Q) / i where CAis the concentration of product A at a given time, F is the Faraday constant (96485.3 C mol'1), n is moles of electrons transferred required to form a given product (i.e., two for NO?’ and eight for NH3), Q is the permeate flow rate of the product solution, and i is the total current. The yield rate (r) was calculated using the following equation: r = (CAx Q) / mcatwhere mcatis the mass loading of the catalysts.

[0154] The energy consumption per order (ECEO) represents the electrical energy needed to degrade nitrate by 1 order of magnitude, equivalent to 90% removal efficiency, which was calculated as follows:ECEO= (U x i) / (Q x log[Cf / Cp]) where U is the cell potential, and Cf and CPare the nitrate concentration in the feed and permeate, respectively.

[0155] Computational calculations:

[0156] DFT calculations for CuNP binding were performed using the PBE functional. Wave functions were composed of a summation of plane waves with kinetic energies up to 350 eV. Electronic structures and energies were converged and changes in energy fell below 2* 10’3eV. The Brillouin Zone was sampled with a gamma-point centered 6><6x l k-point mesh, which resulted in a 0.004 eV difference from a finer 8x8x 1 mesh. PAW pseudopotentials reduced explicitly calculated electrons to: Is of H; 2s and 2p of O and N; and 3d and 4s of Cu. The DFT- D3 correction accounted for nonlocal van der Waals interactions. The adsorption free energies in the liquid phase (AGbind_Cu) were calculated using thermodynamic Hess cycles with the following equations:where AEads_Cu(g) is the adsorption energy of HNO2, EA_, EH2, EHA, and E, are the calculated energies of NO2, hydrogen, HNO2, and Cu surface, respectively. TAS is the entropy loss ofaqueous species upon adsorption at room temperature (298 K), AGsolvationis the energy difference between gaseous HNO2 and aqueous HNO2, AGprotonationis the protonation energy of HNO2 at a given pH, and CURHE is the number charge multiplied by the applied potential vs. reversible hydrogen electrode (RHE). Specific values for AS and AGsolvationat standard state were obtained from NIST-JANAF Thermodynamical Tables. Full entropy loss was assumed in the calculation, as the different entropy loss hypothesis was tested in previous work.

[0157] DFT calculations for NCh' binding to TTM were performed using the GAMESS program, and optimized structures were visualized with Macmolplt. Molecular structures were optimized at the PBE DFT level with a 6-31 l++(d,p) basis set. There was no significant deviation in binding energies between the 6-3 lG++(d,p) and 6-311++G(d,p) basis sets as shown in Fig. 22; the latter was used for all molecular relaxations and energy calculations due to the efficient computation cost. The TPSSh functional achieved mean absolute deviation in binding energies of only 0.03 eV from those of PBE as shown in Fig. 23, and thus the PBE functional was used for all calculations. Van der Waals dispersion forces were accounted for using the DFT-D3 correction of Grimme. Molecular structures were optimized until the gradient in structural energy was less than 0.0001 Hartree / Bohr. An implicit water solvent was implemented via the polarizable continuum model (PCM) to account for solvation effects with atomic size represented using simplified united atomic radii. PCM has been similarly used to account for solvent effects on oxyanion binding to organic molecules. Binding energies (AEbind_TTM) and free binding energies on TTM for all molecules (AGbind TTM) were calculated using the following equation:AEbind TTM= ETTM,NOZ— ETTM— ENQ2AGbind TTM= AEbind— TAS — 2.303kT(pH — pKa) where AEbind_TTMis the binding energy in of a given NO2 species to the TTM monomer, ETTM*NO2is the total energy of the system with the nitro species bound at a site on the TTM, ETTMis the total energy of the isolated TTM monomer, and ENQ2is the total energy of the isolated NO2 species. AS was taken as the negative of the free ion entropy in solutionSaq(J mol-1K-1), representing a complete loss of solution phase entropy upon binding, and the specific values were obtained from the CRC handbook. R is the gas constant in J mol'1K'1, and the third term accounts for the free energy of deprotonation for the ions. It was assumed that theTTM polymer is nonconductive, so electric potential effects on the binding free energy were neglected.Results:

[0158] Fig. 1(a) to 1(g) depict a characterization of the carbon nanotube-based electrified membrane. Fig. 1(a) is a schematic illustration of the TTM-CuNP / CNT-EM synthesis process. Fig. 1 (b) depicts an optical image of the TTM-CuNP / CNT-EM. Fig. 1(c) depicts top view SEM images and corresponding EDS mapping of the TTM-CuNP / CNT-EM. The elements of EDS mapping included Cu, C, and N. Fig. 1(d) depicts a cross-sectional SEM image of the TTM- CuNP / CNT-EM. Fig. 1(e) depicts TEM, HAADF-STEM, and corresponding EDS images of the nano-scale morphology of the TTM-CuNP / CNT-EM. The inset image of Fig. 1(e) is the HRTEM image of the indicated selected area (dashed-line box) of Fig. 1(e). Fig. 1(f) depicts raman spectra of the TTM-CuNP / CNT-EM and CuNP / CNT-EM. Fig. 1(g) depicts permeate flux and water contact angle of the TTM-CuNP / CNT-EM and CuNP / CNT-EM. Error bars represent the s.d. (n = 3) and data are presented as mean values ± s.d. in (g). The s.d. denotes the standard deviations calculated from three independent measurements using three different membrane samples.

[0019] The permeability of the TTM-CuNP / CNT-EM (diameter: 46 mm, as shown in Fig. 1(b)) was attributed to the interwoven structure of the CNTs forming a three-dimensional porous, conductive electrode, with an average pore size of 42 nm ± 16.6 nm as shown in Fig. 1(c) and in Fig. 5 and a thickness of about 50 pm, as shown in Fig. 1(d).

[0160] Scanning electron microscopy-energy dispersive X-ray spectroscopy (SEM-EDS) mapping of the TTM-CuNP / CNT-EM revealed the uniform distribution of Cu, C, and N as shown in Fig. 1(c) in the membrane, indicating that the CuNP (Cu), CNTs (C), and TTM (N) were evenly dispersed throughout the CNT-based membrane. To further characterize the nanoscale morphology of the membrane, transmission electron microscopy (TEM) and high- angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) as well as EDS mapping were conducted as shown in Fig. 1(e) and Fig. 6. TEM and EDS mapping clearly showed that the CuNP were successfully loaded onto the outer surface of CNTs, and these particles were surrounded by TTM, as shown in Fig. 1(e) and Fig. 7.

[0161] Table 1 depicts the atomic fraction in different areas from EDS maps of the TTM- Cu / CNT-EM in Fig. 7.Table 1.

[0162] X-ray photoelectron spectroscopy (XPS) was conducted to further confirm the presence of Cu, C, and N in the membrane, as shown in Fig. 8. The chemical structure of TTM was analyzed using Fourier-transform infrared spectroscopy (FTIR), as shown in Fig. 9. The peaks at 760 cm'1, 1097 cm'1, and 3286 cm'1were assigned to N-H stretching, C-N stretching, and N-H wagging, respectively, indicating the presence of secondary amines (-NH-) in the TTM.

[0163] The incorporation of TTM into the membrane framework was not found to have a significant effect on its chemical and electrical properties. Raman spectroscopy was used to assess alterations in the chemical structure during functionalization and TTM incorporation as shown in Fig. 1(f). The TTM-CuNP / CNT-EM presented a similar D / G ratio (0.96) to CuNP / CNT-EM (0.93), as shown in Fig. 1(f), suggesting that the CNTs maintained their pristine structure after incorporating TTM. Furthermore, the TTM-coated membrane exhibited excellent electrical conductivity. Electrochemical impedance spectroscopy (EIS) results showed a similarly small bulk resistance of 13.2 compared to the CuNP / CNT-EM (Rs: 12.9 )), as shown in Fig. 10, suggesting that the TTM did not impede electron transfer in the EM.

[0164] Electrochemical impedance spectroscopy (EIS) measurements were performed using a three-electrode configuration in a 10 mM K2SO4 solution at room temperature. The frequency range was from 106Hz to 0.1 Hz, with an excitation amplitude (AEac) of 10 mV. The collected data were analyzed and fitted to equivalent circuit models using ZSimpWin software to enable accurate interpretation of the impedance characteristics.

[0165] The resistivity of the TTM-CuNP / CNT-EM was approximately 1.97 X 10-3Q m. Although this resistivity is not as low as that of pure metal electrodes (e.g., Au, Ag), it is much lower than that of other reported membrane electrodes (e.g., coal-based carbon membraneelectrode (0.3 Q m), graphene hydrogel membrane electrode (0.5 Q m), and nickel-organic gel membrane electrode (1.97 x 107Q m)).

[0166] The resistivity of the membrane was determined by cutting it into a 3x3 cm2piece. The membrane was secured and connected to a multimeter. Electrical resistance was measured across the membrane, and resistivity was calculated using the formula as follows:where R the measured resistance, A is the cross-sectional area, and L is the length. The measurement was repeated three times for accuracy, and the average resistivity value was calculated.

[0167] The influence of TTM on the physical properties of the membrane was also evaluated. The surface roughness of the membrane was ascertained using atomic force microscopy (AFM), as shown in Fig. 11. Based on arithmetic mean roughness (Ra) evaluations, the surfaces of the TTM-CuNP / CNT-EM (Ra: 68.1 nm) and CuNP / CNT-EM (Ra: 71.8 nm) were found to be analogous as shown in Fig. 11. The water contact angle of the TTM-CuNP / CNT-EM (37.9°) was also similar to that of the CuNP / CNT-EM (36.9°), as shown in Fig. 1(g), with the small contact angles arising from the use of hydrophilic PAN as the binder in the porous matrix. This hydrophilic property coupled with the nanoporous structure (average pore diameter of 42 nm), as shown in Fig. 1(b) and Fig. 5, allowed the TTM-CuNP / CNT-EM to maintain a high permeability of 263.5 L h’1m’2bar’1, as shown in Fig. 1(g) for an electrolyte solution of 10 mM KNO3 and 10 mM K2SO4. Therefore, the TTM could be successfully integrated into the conductive interwoven structure of the CuNP-based CNT-EM without adversely impacting its properties, resulting in a new type of ionophore-functionalized nanoporous electrocatalytic membrane with high uniformity, permeability, conductivity, and mechanical stability.Example 2. Energy-efficient nitrate-to-ammonia electrochemical reduction performance:

[0168] The carbon nanotube-based electrified membrane of Example 1 was evaluated to determine the energy efficiency of the nitrate to ammonia electrochemical reduction performance and Figs 2(a) to 2(i) depict the electrochemical nitrate reduction performance of the TTM- CuNP / CNT-EM.

[0169] Fig. 2(a) depicts a schematic illustration of the electrified filtration cell including a 25 mL feed chamber with a RuCh-IrCh / Ti mesh anode and a 25 mL permeate chamber with the EM ascathode. Fig. 2(b) depicts a comparison of NCh’ conversion (%) between flow-by and flow- through operational modes under the current density range from 0.4 to 2.8 mA / cm2and a permeate flow rate of 2 mL / min with a 10 mM KNO3 and 10 mM K2SO4 electrolyte. Fig. 2(c) depicts a comparison of energy consumption per order (the electrical energy consumption required to degrade nitrate by 1 order of magnitude, equivalent to 90% removal efficiency) for nitrate conversion between flow-by and flow-through operational modes under the flow rate of 2 mL / min with a 10 mM K2SO4 and 10 mM KNO3 electrolyte. Figs. 2(d) and 2(e) depicts the s electivity (%) of NCh’-N and NH3-N products over CuNP / CNT-EM (Fig. 2(d)) and TTM- CuNP / CNT-EM (Fig. 2(e)) under different current densities (0.4 to 2.8 mA cm'2). Fig. 2(f) depicts a comparison of NCh’-N conversion and NH3-N selectivity between the examples described herein (in 6 s treatment time) and other reports (in at least 120 min treatment time), such as Mxene-based cathodes and precious metal-modified electrocatalysts. Fig. 2(g) depicts a comparison of linear sweep voltammetry curves between CuNP-coated and TTM-CuNP-coated electrode in batch reactor at the scan rate of 10 mV / s. Fig. 2(h) depicts long-term chronopotentiometry stability test of the TTM-CuNP / CNT-EM in the flow-through cell at 0.8 mA / cm2. Fig. 2(i) depicts loss of Cu from the TTM-CuNP / CNT-EM as a function of filtration cycles. After operating for 1 h, the membrane was rinsed and dried and then used for the next cycle. Error bars represent the s.d. (n = 3) and data are presented as mean values ± s.d. in Figs. 2(b), 2(c), 2(d) and 2(e).

[0170] As shown in Fig. 2(c), the energy consumption per order for the flow-through mode (0.77 kWh m'3) was approximately 35 times lower than that of the flow-by mode (27.13 kWh / m3) under 2.8 mA / cm2, clearly demonstrating that electrofiltration is a promising approach for low- concentration NO3' reduction. Moreover, the 0.77 kWh nr3energy consumption is notably lower than that reported for previous nitrate electrochemical reduction materials, including Pd-Cu electrocatalysts (5.72 kWh m'3), Cu-Pt bimetallic 3D electrocatalysts (13 kWh m'3), and AuiCu alloy-based electrocatalysts (8.22 kWh nr3).

[0171] To evaluate the stability and durability of the TTM-CuNP / CNT-EM for practical applications, a long-term chronopotentiometry test (n=3) was conducted with a continuous flow of real surface water (collected from Lake Wintergreen near Yale’s main campus; dissolved organic carbon (DOC): 11.9 mg / L; added NOs'-N: 10 mg / L) was conducted in the flow-through cell over 100 hours and the results are shown in Fig. 36

[0172] The corresponding chronopotentiograms corresponding to Fig. 2(b)-2(e) are in Fig. 30(a) to Fig. 30(c). Chronopotentiometric data (vs. Ag / AgCl) are depicted in Fig. 30(a) for the CuNP / CNT-EM in flow-by mode, in Fig. 30(b) for the CuNP / CNT-EM in flow-through mode, and in Fig. 30(c) for the TTM-CuNP / CNT-EM in flow-through mode. The relatively high cathodic potentials (reported without IR compensation due to high solution resistance) can be attributed to the low concentration of the electrolyte and large distance between the electrodes (due to design constraints of the non-optimized, bench-scale flow-through reactor).

[0173] No significant potential difference is observed between CuNP / CNT-EM and TTM- CuNP / CNT-EM at the same current density. Note that the higher cathodic potential can be attributed to the low concentration of electrolytes (without IR compensation) and large distance between the reference electrode and the cathode (due to flow-through reactor design constraints that prevent closer placement to avoid reactor damage), which increased the solution resistance during the reaction.

[0174] The electrochemical NCh reduction performance of the EM cathode was evaluated using a cross-flow electrofiltration system under ambient temperature and pressure with a permeate flow rate of 2 mL min'1, as depicted in Fig. 2(a). Using the flow-through operational mode with the CuNP / CNT-EM, a single-pass NCh' conversion of 70.6% was achieved for a feed concentration of 10 mM at a current density of 2.4 mA cm'2. In comparison, the traditional flow- by mode yielded a mere 4.4% conversion at the same current density as shown in Fig. 2(b). The 66.2% increase in the conversion achieved by the flow-through mode can be attributed to the elimination of diffusional mass transport limitations in the nanoporous EM as shown in Fig. 2(c). In addition to enhancing NCh' reduction activity, the more efficient conversion enabled by the flow-through mode also drastically reduced the energy consumption.

[0175] The energy consumption per order (ECEO) represents the electrical energy needed to degrade nitrate by 1 order of magnitude per volume of treated water, equivalent to 90% removal efficiency, and was calculated as follows:ECEO= (U x i) / (Q x log[Cf / Cp]) where U is the cell potential corresponding to the current i, and Cf and CPare the nitrate concentration in the feed and permeate, respectively.

[0176] As shown in Fig. 2(c) and Fig. 12, the energy consumption per order for the flow-by mode (35.5 kWh / m3) was approximately 25 times higher than that of the flow-through mode (1.4 kWh / m3), clearly demonstrating that flow-through is a more promising approach for NCh’ reduction. Additionally, the flow-through operational mode enables high NC ' conversion within a low current density range (0.4 to 2.4 mA / cm2), as shown in Fig. 2(b), presenting an energyefficient method for W reduction to NH3. For example, the low power consumption could enable facile integration with off-grid renewable energy systems to generate ammonia using small amounts of excess power at peak generation times.

[0177] Within the low current density range, though, NCh’ was the main product formed for NCh’ reduction using the CuNP / CNT-EM as the flow-through cathode, as illustrated by the notably higher Faradic efficiency (FE) of NCh’ (> 50%) compared to NH3 (< 30%) shown in Fig. 2(d). Unfortunately, NCh’ is of significant environmental concern in treated water, being more toxic and carcinogenic than NCh’, in addition to reducing the efficiency for NH3 synthesis.

[0178] Comparatively, the TTM-CuNP / CNT-EM significantly enhanced the FE of NH3 (>60%) and decreased the F.E. of NCh’ (<25%) under the same flow-through conditions at a given current density or potential as shown in Fig. 2(d) and Figs. 31(a) and 31(b), indicating the function of the ionophore in inhibiting NCh’ formation during the NC ’ reduction process. Further, the TTM-CuNP / CNT-EM had a much lower NC ’ yield rate as shown in Fig. 13 and higher NEE yield rate as shown in Fig. 2(f) than the CuNP / CNT-EM, achieving 91.9% NH3-N selectivity at the current density of 2.8 mA / cm2. This favorable generation of NH3 compared to NCh’ using the TTM-CuNP / CNT-EM was also evidenced by the higher NH3 partial current density than that of NCh’ as shown in Fig. 14, as well as the reduced reduction onset potential for NCh’ measured using linear sweep voltammetry as shown in Fig. 15.

[0179] To investigate the impact of ionic interferences on nitrate conversion, common ions such as Ca2+and SCh2' typically present in surface water were introduced into the feed solution, following the Nanotechnology-Enabled Water Treatment Engineering Research Center standard as shown below in Table 2. The membrane maintained a nitrate removal efficiency of approximately 94%, suggesting minimal ionic interference during nitrate reduction. The robustness of the high nitrate removal performance in the presence of co-ions may be attributed to the high collision frequency of nitrate molecules with active sites within the membrane due to advection through nanopores during flow-through filtration.Table 2. Constituents of synthetic surface water (Nanotechnology- Enabled Water Treatment Engineering Research Center standard).

[0180] Implementation of the TTM-CuNP / CNT-EM in flow-through mode enabled achievement of an ultra-low energy consumption of 21.4 kWh kg-NFF'1for ammonia synthesis, which is lower than that reported in previous studies including those using expensive Ru-based nanocluster electrocatalysts and a pulse nitrate-to-ammonia synthesis method as shown in Fig. 2(g)-

[0181] To evaluate the stability and durability of the TTM-CuNP / CNT-EM, a long-term chronopotentiometry test with a continuous electrolyte flow was conducted in the flow-through cell as shown in Fig. 2(h). The voltage required to achieve 0.8 mA / cm2on the membrane remained stable over 115 hours, with a stable NFF yield rate of -33.3 pg h'1cm'2as shown in Fig. 2(h). No significant CuNP leaching (less than 1% loss of Cu from the catalysts) was observed in the permeate solution as shown in Fig. 2(i), further demonstrating the durability of the membrane for long-term continuous applications. Overall, the TTM-CuNP / CNT-EM under flow-through operational mode not only achieved highly efficient O.f removal with the lowest energy consumption per kg NH3 reported for electrochemical nitrate reduction, but also showed superior performance for inhibiting NO ' formation and achieving long-term stability for NH3 synthesis.Example 3. Mechanism of ionophore modification of nitrite adsorption / desorption for enhancing NH3 selectivity

[0182] The carbon nanotube-based electrified membrane of Example 1 was also evaluated to determine the mechanism of ionophore modification of nitrite adsorption / desorption for enhancing ammonia selectivity.

[0183] The incorporation of TTM into the CuNP / CNT-EM significantly reduced the formation of toxic NO?' and enhanced the NH3 production rate and energy efficiency during NO.3'reduction. The effect of TTM on the catalyst active sites, directly or serving as an independent cooperative component, was investigated.

[0184] Figs. 3(a)-3(e) depict a mechanism investigation of TTM in nitrite absorption / desorption and its impact on nitrate reduction product selectivity. Fig. 3(a) depicts a comparison of NCh' conversion (%) over different EMs for a range of current densities (0.4 to 2.4 mA cm'2), including CNT-EM, TTM / CNT-EM, CuNP / CNT-EM, and TTM-CuNP / CNT-EM. Fig. 3(b) depicts a comparison of Tafel slope for NO ' to NEE between CuNP-coated and TTM-CuNP- coated electrode. Fig. 3(c) depicts a schematic representation of NO ' reduction (ci) without and (c?) with TTM incorporation, in which ki represents the reaction rate of NOs' to NO?', k? represents the reaction rate of NO?' to NH3, and Di represents the desorption rate of NO?'. Fig. 3(d) depicts a comparison of NO?' adsorption and desorption between CuNP and TTM-loaded CuNP over 3000 s using quartz crystal microbalance measurements. The solution was changed from ultrapure water to NaNO? solution, and then back to ultrapure water. Fig. 3(e) depicts a selectivity test using interfering ions (CF, K+, and SO42) and target ions (NO?').

[0185] As a control, it was demonstrated that TTM in the absence of CuNP does not provide significant activity, evidenced by the considerably higher NO.f conversion in the TTM- CuNP / CNT-EM (29.2%) compared to that in a TTM / CNT-EM without CuNP (1.2%) at the same current density of 0.4 mA cm'2as shown in Fig. 3(a). Furthermore, incorporating TTM into the CuNP / CNT-EM did not influence the NOs' conversion, which remained around 70% at 2.4 mA cm'2with and without TTM as also shown in Fig. 3(a). Moreover, the TTM-CuNP / CNT-EM did not show substantial enhancement in turnover frequency, as shown in Fig. 16 for NO?' conversion compared to the CuNP / CNT-EM, as calculated using their respective electrochemically active surface areas (ECSA) as shown in Figs. 17 and 18.

[0186] Moreover, the TTM-CuNP / CNT-EM did not show substantial enhancement in turnover frequency (TOF) for NOV conversion compared to the CuNP / CNT-EM, as calculated using their respective electrochemically active surface areas (ECSA). This similar TOF further suggested that TTM did not directly modify the active site for NO?' reduction. In addition, the TOF of the TTM-CuNP / CNT-EM exceeded 2000 h which is significantly higher than that reported in previous studies for 10 mM nitrate reduction using electrocatalysts such as copper-based metalorganic framework (323.3 h'1), bifunctional copper-cobalt spinel (43.2 h'1), and nitrogen-doped carbon supported copper nanoparticles (897.3 h'1). This high TOF can primarily be attributed tothe flow-through mode, which mitigates mass transport limitations that restrict the conversion of low-concentration nitrate and increase the collision frequency of nitrate molecules with active sites within the membrane through advection.

[0187] Table 3 below provides an ECSA analysis and calculation. Slopes are calculated based on the scan rate-current density difference (Aj / 2 = (ja— jc) / 2) relationship and their linear fitting in Figs. 17 and 18. The Cdi was determined by the surface area of the electrified membranes and their slopes. The ECSA was determined by: ECSA = Cdi / Cs, where Cdi is the double layer capacitance and Cs is the specific capacitance of the sample. In this study, a general specific capacitance of Cs=0.04 mF cm’2for the CuNP electrode was used based on that reported in other studies.Table 3.

[0188] Since TTM did not directly contribute to electrocatalytic NCh’ reduction activity, it is believed that the NCh' adsorption properties of the TTM might enable it to serve as a cooperative component, modifying the catalytic microenvironment to enhance NH3 selectivity. At low current densities, NCh’ is typically the primary product of NCh’ reduction. This is attributed to the faster conversion rate of NCh’ to NCh’ (denoted as ki), as shown in Fig. 3(ci) compared to NCh’ to NH3 (denoted as k2), as shown in Fig. 3(ci). Following the initial reduction step, a substantial amount of generated NCh’ can desorb from the catalyst surface and transport to the bulk electrolyte solution prior to further reduction to NH3 as shown in Fig. 3(ci). Thus, a decline in NCh’ concentration within the electrolyte is associated with a decreased NC ’ desorption rate (Di) as shown in Fig. 3(ci). Based thereon, it is believed that the increase of the NCh’ adsorption / decreased desorption would enrich the NCh’ concentration within the local environment in the electrocatalytic membrane, thereby promoting the reduction of NCh’ to NH3 as shown in Fig. 3(c2).

[0189] This stability was further validated using real wastewater treatment effluent (collected from the UConn Water Resource Recovery Facility), for which nitrate levels were reduced from ~14 mg / L to 2 mg / L, which is significantly below the EPA standard, under a low current density of 2.8 mA / cm2.

[0190] To quantitatively assess the nitrite adsorption capacity of TTM within the membrane described herein, quartz crystal microbalance (QCM) measurements were performed on the CuNP and TTM-CuNP during exposure to NaNO? solution and ultrapure water as shown in Fig. 3(d). Enhancement in NO?' adsorption would be evidenced by greater changes in mass on the QCM sensors. It was observed that CuNP alone showed a modest mass shift of about ~2 pg cm'2(~10 Hz, where mass is converted from the resonant frequency upon transitioning from ultrapure water exposure to NO?' solution. In contrast, the TTM-loaded CuNP exhibited a more substantial mass shift of ~8.5 pg cm'2(~40 Hz,) over the same 2000 s interval upon exposure to NO?' solution. This pronounced change in mass underscored the pivotal role of TTM in augmenting NO? adsorption. Upon analyzing desorption by transitioning from the NO solution exposure back to ultrapure water, the mass for the CuNP reverted to its baseline of ~0 pg / cm2. Meanwhile, the TTM-loaded CuNP only shifted back to ~4 pg / cm2(Fig. 3(d)). This suggested that NO?' desorbed less readily from the TTM-loaded CuNP, allowing the TTM-CuNP / CNT-EM to retain a higher concentration of NO?' near the electrode surface and potentially facilitating its conversion to NH3.

[0191] Other ions were introduced to the ionophore to test the selectivity of the TTM interaction with various ions which may be present. The concentration of ions, including Cl', K", and SO42', tested for their interaction with the TTM ionophore are 10 mg / L for each ion, which is consistent with the concentration of the targeted pollutant nitrate. The same concentration of each ion was chosen for the selectivity tests in order to obtain comparable response data, as the open circuit potential response is concentration-dependent.

[0192] A selectivity test was performed by drop-casting the TTM solution (5 uL) onto the surface of a carbon-based electrode (radius: 2.5 mm). The carbon-based electrode was prepared using single-walled carbon nanotubes (SWCNTs, >98%) as a solid contact layer. After drying at room temperature over 48 h, the electrode was submerged into an aqueous solution (100 mL) containing 10 mg / L NOi' (as NaNCh) continuously stirred at 100 rpm. Subsequently, an interaction selectivity test was conducted by sequentially adding interference ions commonlyfound in nitrate-containing water sources 10, including Cl’ (as NaCl, 10 mg / L), K+(as KC1, 10 mg / L), and SO42’ (as K2SO4, 10 mg / L), into the nitrate solution. The open circuit potential readings were continuously recorded at intervals of 0.1 seconds using a CHI 660E potentiostat.

[0193] The NCh’ interacts with an amine group on TMM via H--0 bonds to the amide group H atom in TTM with bond lengths of 2.13 A. This H atom, influenced by the electronic effects of neighboring C and O atoms bonded to its adjacent N atom, experiences a reduction in dipole moment, leading to a diminished polarity. This reduced polarity results in a uniform electrostatic attraction between TTM and NCh’. In contrast, for sulfate (SO42), the interaction is different due to the higher negative charge and larger size of this ion. Sulfate forms stronger electrostatic attractions but lacks the optimal hydrogen bonding configuration. The bulkiness and tetrahedral geometry of sulfate create steric hindrance, reducing the efficiency of the H— O bonding and leading to weaker overall interactions with TMM. Consequently, sulfate is less effectively coordinated compared to the more compact and geometrically compatible NO2 .

[0194] The affinity of TTM-CuNPs to nitrate in addition to nitrite was also investigated. CuNPs can adsorb nitrate efficiently due to the similarity between the d-orbital energy levels of CuNPs and the LUMO 71* orbital. QCM was used to investigate the impact of TTM on nitrate adsorption by CuNPs. The results demonstrated that TTM-loaded CuNPs still maintained strong nitrate adsorption, as shown with a mass change of approximately 5 pg cm’2by QCM measurement and a binding energy of -0.26 eV by DFT calculations. This high adsorption affinity confirmed that the presence of TTM does not impair nitrate adsorption. This result is in agreement with the consistent nitrate conversion (%) observed in Fig. 3(a) after the incorporation of TTM into the membrane.

[0195] Moreover, regarding the use of QCM and open circuit potential (OCP) measurements, QCM is used to quantitatively measure the adsorption of specific ions, such as nitrite. In contrast, OCP is used for adsorption selectivity testing because it allows for a more precise understanding of the type of adsorption process. Specifically, OCP offers a distinct advantage in that it helps differentiate between activated and non-activated absorption by preventing electron energy transfer and electrochemical reactions under near-zero current flow, thus enabling a clearer assessment of adsorption mechanisms without the need for external energy input.

[0196] The OCP approach was especially useful for evaluating the non-activated adsorption process. For example, a significant change in the OCP was observed when nitrite (NO2 ) wasintroduced, indicating clear non-activated adsorption of nitrite on TTM. In contrast, when sulfate (SO ) or chloride (Cl’) solutions were introduced, no significant changes were observed, suggesting that these interfering ions do not affect nitrite non-activated adsorption. By using OCP, it was possible to focus solely on the non-activated adsorption driven by affinity to TTM, providing a more accurate understanding of TTM’s selective adsorption behavior towards nitrite. This mechanism also corresponds to the DFT results, which show voltage-independent affinity of nitrite to TTM, in contrast to a bare CuNP surface which has voltage-dependent nitrite binding that diminishes at more negative reduction potentials relevant to the experimental conditions (see e.g., Fig. 4(e)).

[0197] Open circuit potential measurements were employed to eliminate the effects of activated adsorption and isolate the non-activated adsorption properties of TTM. The nontarget ions (O’, K+, and SO42') caused nearly negligible interference to the open circuit potential of a TTM- loaded CuNP electrode. When NO2’ was added to the nitrate solution, the open circuit potential changed substantially, exhibiting the selectivity of TTM for NCh' adsorption, as shown in Fig. 3(c). In comparison, when NCh' was introduced into a solution with only a CuNP electrode, there was no significant change in potential, suggesting that TTM has a strong selectivity for NCh’ over other ions such as CF, K', and SO42’ compared to CuNP, as shown in Fig. 3(e). This strong selective adsorption of NCh' by the TTM can facilitate the enrichment of the electrocatalyst microenvironment with NCh'. This mechanism is further supported by the observed increase in NH3 production when higher concentrations of nitrite were used as the feed solution with the CuNP / CNT-EM, as shown in Fig. 33, suggesting that a higher NCh' concentration at the electrode / membrane surface can enhance NH3 production.

[0198] To quantitatively assess the nitrite adsorption capacity of TTM, quartz crystal microbalance (QCM) measurements were performed on the CuNP and TTM-CuNP during exposure to NaNCh solution and ultrapure water as shown in Fig. 3(d).

[0199] Overall, it was found that the introduction of TTM primarily functions as a cooperative selective adsorption component to enrich NO2' in the local catalyst microenvironment, rather than directly modifying the catalytic active site. The nuanced regulation of the intermediate species, NOz’, revealed a novel paradigm for efficient nitrate-to-ammonia electrochemical conversion. Such insights open avenues for the implementation of cooperative adsorptioncomponents in other electrocatalytic systems, tailored to meet specific electrochemical reaction criteria without directly modifying the electrocatalysts themselves.

[0200] To investigate the impact of TTM on nitrate adsorption by CuNPs, QCM measurements were performed with a nitrate solution. The results demonstrated that TTM-loaded CuNPs maintained strong nitrate adsorption, with a mass change of approximately 5 pg / cm2. Additionally, DFT calculations revealed a binding energy of -0.26 eV for nitrate on TTM-coated CuNPs, further suggesting that the presence of TTM does not impair nitrate adsorption on CuNP. This is consistent with the nitrate conversion (%) observed in Fig. 3(a) before and after the incorporation of TTM into the membrane.Example 4. Evaluation of nitrite binding on CuNP and TTM through density functional theory calculations:

[0201] To provide molecular validation for experimental results, the binding of NO2 to CuNP and TTM was systematically evaluated through density functional theory (DFT) calculations. The Cu{ 111 } surface was selected to represent CuNP, as it was the dominant crystal facet identified by XRD analysis as shown in Fig. 25. Furthermore, the {111} facet is the most stable crystal Cu surface and is associated with high nitrate conversion activity.

[0202] Fig. 4(a)-4(f) depict Nitrite binding models on CuNP and TTM. Fig. 4(a) depicts comparisons of NO?' binding energy on CuNP and TTM at base conditions (pH=7, potential=0 V vs RHE). Fig. 4(b) depicts bader charge analysis of NO2 adsorption on CuNP and TTM. Fig. 4(c) depicts optimized structures of NO?' bound to CuNP and TTM. Fig. 4(d) depicts a Pourbaix diagram of NO?' adsorption on CuNP electrode surface. The adsorption free energy values are shown in the color map. Fig. 4(e) depicts a Pourbaix diagram of NO?' adsorption free energy on TTM-CuNP electrode surface, showing the respective binding energies on CuNP and TTM. For pH regions where both surfaces bind NO?', the surface with stronger binding is shown in the forwardmost layer in the diagram. Fig. 4(f) depicts the variation of NO?' binding free energy on CuNP and TTM at specific points (P1-P4) extracted from Fig. 4(e), with corresponding values (pH, potential in V vs. RHE) of Pl : (5.0, 0.3); P2: (7.0, 0.0); P3: (9.0, -0.3); P4: (11.0, -1.0).

[0203] Calculations revealed that CuNP strongly binds NO?' in the absence of applied potential, where the binding free energy is -20.2 kJ mol'1, as shown in Fig. 4(a) with O-Cu bond lengths of 2.05 A. Bader charge analysis indicated that such strong binding is due to significant Badercharge transfer from Cu to NCh’ (0.64 e), as shown in Fig. 4(b). In comparison, TTM shows relatively weaker binding with NCh’ in the absence of applied potential, with binding energy of - 6.0 kJ mol’1as shown in Fig. 4(a) and Bader charge transfer of 0.16 e as shown in Fig. 4(b). The NCh’ interacts with an amide group on TMM via weak H--0 bonds with bond lengths of 2.13 A, as shown in Figs. 4(a)-4(c). Such weak binding can be attributed to the involvement of the amide group H atom in TTM. This H atom, influenced by the electronic effects of neighboring C and O atoms bonded to its adjacent N atom, experience a reduced dipole moment, diminishing its polarity. As a result, this weakened dipole moment led to a uniform electrostatic attraction between TTM and NCh, contributing to the observed binding affinity through the hydrogen bond.

[0204] However, the binding strength of NCh’ is highly dependent upon the potential fluctuations and pH variations that occur during electrochemical reactions. As shown in Figs. 4(d) and 4(e), a Pourbaix diagram was constructed to comprehensively assess the binding free energy of NCh’ across a wide potential range (-3 to 1 V vs. RHE) and pH range (5 to 12) on both CuNP and TTM. The binding free energy between NCh’ and the negatively charged CuNP surface decreases with more negative applied potential, as shown in Fig. 4(d). This is attributed to the increased electrostatic repulsion of NCh’ by the negatively charged CuNP surface, leading to NCh’ desorption into solution during the cathode reaction and inhibiting NH3 production. In contrast, the affinity of NCh’ remains consistent across variations in the applied potential on the TTM phase, with TTM maintaining exothermicity for NCh’ binding over a wide pH range, as shown in Fig- 4(e).

[0205] To quantitively compare the variation of NCh’ binding energy on CuNP and TTM with pH and potential, four points (Pl -P4) characterized by increased negative potential and pH were analyzed, as shown in Fig. 4(f) The results demonstrated that as the pH rises from 5 to 11 and the potential becomes more negative from 0.3 to -1 V vs. RHE (from Pl to P4), as shown in Figs. 4(e) and 4(f), the binding energy on CuNP decreases from -37.7 at Pl to 0 kJ mol’1at P4. Conversely, TTM exhibits a binding energy of -28.9 kJ mol’1at pH 11 and potential of -1 V vs. RHE (P4) as shown in Fig. 4(f), where the conditions at P4 closely resemble our experimental conditions (pH: ~11), as shown in Fig. 19. This result suggests that the presence of TTM expanded the operational window (the potential and pH) at which NCh can remain near the cathode surface, based on preventing its transport away from the CuNP surface to the solution.As a result of the enriched NO?' concentration near the cathode surface, NO?' can be more easily re-adsorbed to the catalyst active sites for further reduction to NH3. Therefore, these results further validate the role of TTM as a cooperative adsorption component in Cu-based electrocatalytic NO ' reduction based on enriching the concentration of NO?' in the catalyst microenvironment to enable enhanced overall conversion to NH3.

[0206] Thus, electrochemically converting nitrate in contaminated water into ammonia offers an environmentally friendly solution for both water treatment and fossil-free fertilizer or carbon- free fuel generation. By incorporating a nitrite-adsorbing TTM ionophore into a CuNP-CNT- based electrofiltration membrane, highly efficient single-pass removal of low-concentration Nth' (>70%) with high NH3 selectivity (>80%) can be achieved, attaining the lowest currently reported NCh'-to-NFh energy consumption of 21.4 kWh / kg-NHs. Additionally, the TTM- CuNP / CNT-EM demonstrated superior performance in limiting the formation of NO?' (<15%) and maintaining long-term stability (>115 hours).

[0207] Moreover, the mechanical flexibility of the TTM-CuNP / CNT-EM enhances its scalability since it could be integrated into spiral-wound modules, a widely adopted format in industrial membrane technology. This adaptability makes the membrane suitable for both larger systems and modular treatment units without compromising performance, positioning it as a promising solution for broad implementation in water treatment facilities and existing water treatment technology architectures. Additionally, combining the electrofiltration membrane with NH3 separation and capture systems could enable recovery of ammonia as a valuable product, supporting the development of carbon-neutral or carbon-free ammonia production.

[0208] The introduction of TTM plays a key role as a cooperative adsorption component, enhancing the presence of NO?' within the local catalyst microenvironment to significantly improve NH3 selectivity without directly modifying the catalytic active site. This molecular-level manipulation of intermediate NO?' species during reaction reveals a new paradigm for achieving efficient NOs'-to-NEE electrochemical conversion. These discoveries illustrate possibilities for applying cooperative adsorption components in diverse electrocatalytic systems, enabling customization of local reaction environments near active sites to modify overall reaction selectivity without directly altering the electrocatalyst. This approach can open a broader pool of catalysts, including inexpensive materials, that are suitable for achieving active and selectivereactions, while also providing greater flexibility to design cooperative materials ensembles for precise molecular tuning of energy-efficient reactions.

[0209] The TTM-CuNP / CNT-EM system presented herein also offers a promising option for simultaneous point-of-use water treatment and generation of fossil-free fertilizer, featuring a compact and modular device for continuous-flow operation.

[0210] A preliminary techno-economic analysis TEA) was conducted to assess the economic costs, including capital, operational, and disposal costs, as well as the health benefits (NO3- removal) and fertilizer benefits (NH3 recovery) compared to existing nitrate removal technologies. The TEA was based on a model of farm communities with varying populations (ranging from 500 to 10,000 people) and an average of approximately 3.75 acres of cropland per person. Compared to conventional NOi' removal methods recommended by the EP A, such as ion exchange (IX), reverse osmosis (RO), and electrodialysis (ED), the TTM-CuNP / CNT-EM system could save up to $0.78 per m3, resulting in annual savings around $1,500,000 for a community of 10,000 people. This cost reduction can be attributed to -40% reduction in capital cost. The TTM-CuNP / CNT-EM system achieved a shorter hydraulic retention time, with ultrafast nitrate conversion in just 6 seconds, allowing the system to be significantly more compact than other conventional methods. Additionally, the TTM-CuNP / CNT-EM system eliminated the need for costs associated with NO3- concentrate or brine disposal due to its high conversion efficiency of near 95%, whereas the competing technologies separate rather than transform nitrate. Furthermore, the TTM-CuNP / CNT-EM system produces NH3 that can be utilized as fertilizer at a cost of approximately $0.04 / m3. In contrast, conventional methods do not produce any fertilizer or the associated economic benefits. When considering the additional cost of NH3 separation (e.g., using a gas-permeable membrane for acid stripping), the overall cost remains comparable to that of other conventional methods. A summary of the results of the TEA is provided below in Table 4.

[0211] Table 4 presents the parameters for estimating water flow, nitrate contamination, and the capital, operational, and disposal costs, as well as the health and fertilizer benefits for the TTM-CuNP / CNT-EM system reported in this study, electrodialysis (ED), reverse osmosis (RO), and ion exchange (IX).Table 4.

[0212] Figs. 40(a) and 40(b) depict the preliminary TEA in which Fig. 40(a) provides comparison of economic costs, including capital, operational, disposal, and ammonia separation costs for nitrate removal from water, among this study, electrodialysis (ED), reverse osmosis (RO), and ion exchange (IX) and Fig. 40(b) provides a comparison of health and fertilizer benefits for nitrate removal among this study, ED, RO and IX.

Claims

WHAT IS CLAIMED IS:

1. An conductive filtration membrane, the conductive filtration membrane comprising: a. carbon nanotubes; and b. metal particles preferably wherein the metal particles comprise copper particles, more preferably wherein the copper particles comprise copper nanoparticles; and c. an ionophore, wherein the ionophore is interspersed with the metal particles; and wherein ionophore exhibits selective adsorption properties to specific ions.

2. The conductive filtration membrane according to claim 1, wherein the specific ions comprise nitrite ions and wherein the ionophore enhances the interaction between generated nitrite and the catalyst surface.

3. The conductive filtration membrane according to claim 1, wherein the ionophore comprises a salen or sal ophen ionophore.

4. The conductive filtration membrane according to claim 3, wherein the ionophore comprises a tetraaza-tetraoxa macrocycle.

5. The conductive filtration membrane according to any of claims 1 to 4, wherein the conductive filtration membrane has an average pore size in the range of about 10 to about 100 nm, more preferably in the range of about 45 to about 80 nm.

6. An electrochemical membrane filtration system for preferentially reducing nitrate to ammonia in a water stream, the electrochemical membrane filtration system comprising: a. a water source comprising nitrates; b. a pump for pumping the water source into a reactor cell; c. a reactor cell comprising: i. an inlet for receiving the water source into the reactor cell ii. a conductive filtration membrane according to any of claims 1 to 7, wherein the conductive filtration membrane functions as a cathode;iii. an anode electrically connected to the an electrostatic filtration membrane; and and iv. an outlet for removing ammonia and water from the reactor cell.

7. The electrochemical membrane filtration system according to claim 6, wherein the anode comprises a mixed metal oxide mesh, preferably wherein the mixed metal oxide mesh comprises a ruthenium-iridium oxide coated titanium (RuCh-IrCh / Ti) mesh.

8. The electrochemical membrane filtration system according to claim 6 or claim 7, wherein the reactor cell is a flow-through reactor cell.

9. A method of electrochemically reducing nitrates from a water stream into ammonia, the method comprising the steps of: a. providing a conductive membrane filtration system according to any of claims 6 to 8; b. circulating water containing a concentration of nitrate in a reactor cell, wherein the water passes through the conductive filtration membrane containing the ionophore, and wherein the ionophore enhances the interaction between generated nitrite and the catalyst surface, thereby minimizing desorption of nitrite and enhancing overall generation of ammonia; and c. removing permeate comprising ammonia and water from the reactor cell.

10. The method according to claim 9, wherein the water comprises nitrate at a concentration in the range of about 1 to about 500 millimolar, more preferably about 1 to about 100 millimolar, more preferably about to 10 about 100 millimolar.

11. The method according to claim 9, wherein the conversion of nitrate to ammonia is greater than about 60% or greater than about 65% or greater than about 70% at a current density within the range of about 0.4 to about 2.4 mA / cm2.

12. The method according to claim 9, wherein the conductive filtration membrane exhibits a Faradaic efficiency of nitrate reduction towards nitrite and ammonia products of greater than 60% of ammonia and less than 25% of nitrite.

13. The method according to claim 9, wherein the conductive filtration membrane exhibits a stability of at least about 100 hours or at least about 110 hours or at least about 120 hours or at least about 130 hours or at least about 150 hours or at least about 200 hours or at least about 300 hours or more as measured by a chronopotentiometry stability test of the conductive filtration membrane in reactor cell at 0.8 mA / cm2.

14. A method of making a conductive filtration membrane, the method comprising the steps of a. dispersing metal particles and carbon nanotubes in a resin solution to create a metal particle-carbon nanotube suspension; b. combining an ionophore with a plasticizer to prepare an ionophore mixture, wherein the ionophore exhibits selective adsorption properties to specific ions; c. combining the metal particle-carbon nanotube suspension with the ionophore mixture to prepare a suspension comprising the metal particle-carbon nanotube suspension and the ionophore distributed therein; d. vacuum filtering the resulting mixture onto a substrate, optionally wherein the substrate comprises a microporous ceramic substrate to produce the conductive filtration membrane; e. drying the conductive filtration membrane at an elevated temperature for a period of time; and f. separating the conductive filtration membrane from the microporous ceramic substrate.

15. The method according to claim 14, wherein the resin solution comprises a resin dispersed in an organic solvent.

16. The method according to claim 15, wherein the resin is selected from the group consisting of acrylomtriie-butadiene- styrene (ABS) resins, polystyrene, acrylonitrilestyrene (AS) resins, acrylic resins and polyacrylonitrile resins, preferably wherein the resin comprises polyacrylonitrile.

17. The method according to claim 15, wherein the organic solvent is selected from the group consisting of N-methylpyrrolidone (NMP), N,N-dimethylacetamide (DMAC), dimethylformamide (DMF), diethylformamide, " / -butyrolactone (GBL), dimethyl sulfoxide (DMSO), and combinations of the foregoing, preferably wherein the organic solvent comprises dimethylformamide.

18. The method according to any of claims 14 to 17, wherein the ionophore comprises a salen or salophen ionophore.

19. The method according to claim 20, wherein the ionophore comprises a tetraaza-tetraoxa macrocycle.

20. The method according to any of claims 14 to 17, wherein the conductive filtration membrane has an average pore size in the range of about 10 to about 100 nm, more preferably in the range of about 45 to about 80 nm.

21. The method according to any of claims 14 to 17, wherein the carbon nanotubes, metal particles and the ionophore are substantially evenly dispersed throughout the conductive filtration membrane.

22. The method according to any of claims 14 to 17, wherein the ionophore enhances conversion of nitrate to ammonia during nitrate electrocatalytic reduction by enhancing the interaction between generated nitrite and the catalyst surface, thereby minimizing desorption of nitrite and enhancing overall generation of ammonia.

23. The method according to any of claims 14 to 17, wherein the conductive filtration membrane is configured to exhibit a stability of at least 100 hours or at least 110 hours or at least 120 hours or at least 130 hours or at least 150 hours or at least 200 hours or at least 300 hours or more as measured by a chronopotentiometry stability test of the conductive filtration membrane when the conductive filtration membrane is placed in a flow-through reactor cell at 0.8 mA / cm2.

24. The method according to any of claims 14 to 17, wherein the metal particles comprise copper particles, more preferably wherein the copper particles comprise copper nanoparticles.

Images