a Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, No. 266, Fangzheng Avenue, Beibei District, Chongqing 400714, China.
b School of Science, STEM College, RMIT University, 124 La Trobe Street, Melbourne, VIC 3000, Australia.
The electrochemical nitrate reduction reaction (NO3RR) offers a highly promising, carbon-neutral pathway to simultaneously remediate nitrate-contaminated wastewater and synthesize valuable ammonia. However, while significant progress has been made in catalyst design, the industrial scale-up of NO3RR remains fundamentally bottlenecked by severe mass transport limitations, concentration polarization, and detrimental localized microenvironmental shifts. This review demonstrates the research focus from material science to chemical engineering, providing a comprehensive analysis of electrochemical reactor design for NO3RR. We track the structural evolution from conventional batch reactors to continuous flow-by and flow-through architectures that actively manipulate fluid dynamics to overcome traditional diffusion limits. Furthermore, we critically evaluate state-of-the-art membrane electrode assemblies, highlighting the crucial mechanistic roles and structural limitations of cation, anion, and bipolar exchange membranes in managing ionic crossover and localized interfacial pH. We also investigate the scalable precision manufacturing techniques required to fabricate these core components, to component microstructures and comprehensive NO3RR performance. Furthermore, the review also demonstrates the integration of downstream in situ ammonia recovery systems and explores advanced operational strategies, such as pulsed electrolysis, to mitigate catastrophic mineral scaling in complex, real-world aquatic environments. Ultimately, this review outlines a systematic engineering framework essential for mitigating operational complexities and facilitating the viable industrialization of NO3RR systems.
Ammonia (NH3) is a foundational chemical for modern agriculture and a highly promising carbon-free energy carrier [1,2]. For over a century, global NH3 production has relied heavily on the Haber-Bosch process, an energy-intensive industrial process that consumes roughly 1-2% of global energy and contributes to nearly 2% of global CO2 emissions [3-5]. Simultaneously, the excessive use of nitrogenous fertilizers has severely disrupted the global anthropogenic nitrogen cycle[6]. Large quantities of unutilized nitrogen runoff into water bodies primarily as nitrate (NO3-), leading to severe environmental crises such as eutrophication, harmful algal blooms, and critical human health risks[7]. In this context, the electrochemical nitrate reduction reaction (NO3RR) has emerged as a highly compelling technology. Driven by renewable electricity, NO3RR offers a mutually beneficial approach: it remediates toxic NO3- from wastewater streams while simultaneously achieving the decentralized, carbon-neutral synthesis of high-value NH3 [8,9].
From a fundamental perspective, the electrochemical conversion of NO3- to NH3 is a highly complex multi-step process involving the transfer of eight electrons and nine protons (NO3-+9H++8e-→NH3+3H2O). This intricate pathway generates various nitrogen-containing intermediates and requires highly active electrocatalysts to lower the activation energy barriers [10,11]. Furthermore, the thermodynamic potential of NO3RR is closely situated near that of the hydrogen evolution reaction (HER) [12,13]. Consequently, vigorous water splitting often competes with nitrate reduction, severely degrading the Faradaic efficiency (FE) and overall energy efficiency of the target reaction. Although numerous studies in the last decade focus on rational catalyst design via single-atom dispersion, defect engineering and alloying strategies to suppress HER and improve intrinsic NH3 selectivity [14-16], merely optimizing catalysts fails to render this technology commercially applicable.
As NO3RR research pivots toward industrial application, a significant research shift from material science to chemical engineering is urgently required. The fundamental bottleneck preventing the scale-up of NO3RR lies in reactor design and system engineering [17,18]. In practical scenarios, particularly during the treatment of low-concentration wastewater, the reaction rate is strictly governed by mass transport rather than inherent catalytic kinetics. The sluggish diffusion of NO3- from the bulk electrolyte to the electrode surface leads to severe concentration polarization, capping the achievable current densities [19.20]. Furthermore, the rapid consumption of protons during high-rate NO3RR drastically alters the localized microenvironment, creating extreme interfacial pH gradients that can destabilize catalysts and shift reaction pathways [21]. Traditional batch-type reactors, such as H-cells, which dominate current NO3RR literature, are inherently flawed for practical applications [22,23]. They suffer from severe mass transfer limitations, large inter-electrode ohmic drops, and are largely incapable of sustaining the industrial-level current densities ( > 100 mA cm-2 to >1 A cm-2) required for economic viability. To overcome these fundamental limitations, available reactors that optimize fluid dynamics, manage gas-liquid-solid interfaces, and minimize ohmic resistance are paramount.
To move laboratory-validated catalysts from bench-scale testing to practical industrial ammonia electrosynthesis deployment, a comprehensive review of both reactor engineering and scalable precision manufacturing is indispensable. Therefore, this review aims to shift the spotlight toward the physical and engineering aspects of NO3RR. We begin by elucidating the fundamental principles of mass transport, concentration polarization, and the evolution of the localized interfacial microenvironment. Subsequently, we systematically track the evolution of reactor structures (Figure 1), analyzing the transition from conventional H-cells to continuous flow-by and flow-through configurations that break traditional diffusion limits. Special emphasis is placed on state-of-the-art membrane electrode assembly (MEA) and zero-gap (ZG) architectures, which represent the frontier of scalable, high-current-density operation. Finally, we investigate the scalable precision manufacturing techniques required for NO3RR components, and discuss the integration of downstream ammonia recovery systems and scaling up reactors to process real-world wastewater, providing a roadmap for the future industrialization of NO3RR.
2. Mass transport and local microenvironment in NO3RR
While catalyst design focuses on lowering intrinsic activation energy barriers, the practical performance of an electrochemical reactor is dictated by how efficiently reactants are delivered to, and products are removed from, the solid-liquid interface. In the NO3RR, the complex eight-electron transfer process makes the system acutely sensitive to interfacial mass transfer and localized microenvironmental shifts [24,15]. Understanding these phenomena from an engineering perspective is the prerequisite for designing scalable reactors.
2.1 Principles of mass transfer
In an electrochemical reactor, the movement of species (such as NO3-, H+, and OH-) to and from the electrode surface is governed by the Nernst-Planck equation, which delineates the total flux (Ji) into three distinct mechanisms: diffusion, migration, and convection[26].
Where Di is the diffusion coefficient, $\nabla {\mathit{C}}_{\mathit{i}}$ is the concentration gradient, zi is the charge of the ion, $\nabla \mathit{\varphi }$ is the electric potential gradient, and $\mathit{v}$ is the bulk fluid velocity.
Diffusion: Driven by the concentration gradient across the Nernst diffusion layer. As NO3- is consumed at the cathode, its surface concentration drops, driving bulk NO3- toward the electrode. In stagnant or batch systems like H-cells, diffusion is often the sole transport mechanism near the boundary, severely limiting the reaction rate.
Migration: The movement of charged species under the influence of an electric field. Critically for NO3RR, the reactant NO3- is an anion. At the negatively charged cathode, electrostatic repulsion forces NO3- away from the electrode surface, fundamentally hindering the reaction. To mitigate this detrimental migration flux, highly concentrated supporting electrolytes (e.g., 0.1 to 1.0 M Na2SO4) are typically added to suppress the electric field gradient ($\nabla \mathit{\varphi }$).
Convection: The transport of species via bulk fluid motion. From a reactor design standpoint, convection is the most highly controllable variable. Crucially, the bulk fluid velocity ($\mathit{v}$) in the Nernst-Planck equation is inherently linked to the hydrodynamic residence time (HRT) of the reactor, where a lower HRT translates to a higher bulk fluid velocity. By transitioning from stagnant batch systems to continuous flow configurations, engineers intentionally minimize the HRT to maximize $\mathit{v}$. As will be discussed in Section 4, state-of-the-art flow-by systems can operate with an HRT of less than a minute, while advanced flow-through architectures can compress the residence time to merely 10 seconds. By engineering such short HRTs, the high bulk fluid velocity massively amplifies the convective flux (${\mathit{C}}_{\mathit{i}}\mathit{v}$). Through the modulation of fluid dynamics, including the adoption of continuous flow and refinement of flow channel geometries, engineers are able to drastically reduce diffusion layer thickness, enabling a transition of the system from a diffusion-limited regime to a kinetically controlled regime.
2.2 Local pH effects
One of the most profound engineering challenges in NO3RR is the management of the localized microenvironment, specifically the dramatic fluctuations in interfacial pH [27,18]. Regardless of the bulk electrolyte pH, the conversion of NO3- to NH3 requires a massive supply of protons or the generation of hydroxide ions. In neutral or alkaline wastewater, the reaction proceeds as:
NO3-+6H2O+8e-→NH3+9OH-
The generation of nine OH- ions per molecule of NH3 produced causes a severe localized pH surge at the electrode-electrolyte interface. Even if the bulk solution is neutral (pH 7), the local pH within the hydrodynamic boundary layer can rapidly exceed 12 or 13 at high current densities.
This alkaline microenvironment has several detrimental consequences. First, it alters the thermodynamic potentials of intermediate pathways, often promoting the parasitic HER or shifting selectivity toward unvalued byproducts like NO2-. Second, in real-world wastewater treatment, a high localized pH induces the rapid precipitation of scale-forming cations (e.g., Ca2+ and Mg2+) present in the water, which passivates the catalyst surface and blocks active sites.
While chemical buffers are often used in laboratory settings to suppress ΔpH, they are economically unviable for industrial scale-up [12]. Therefore, fluid dynamic intervention via reactor design is essential. High-shear flow-by reactors or 3D porous flow-through devices enhance forced convection, rapidly sweeping the generated OH- away from the boundary layer and drawing in fresh bulk electrolyte, thereby dynamically stabilizing the interfacial pH without chemical additives.
2.3 Concentration polarization
When NO3RR is applied to environmental remediation, reactors must process groundwater or industrial effluents with relatively low NO3- concentrations (typically under 100 ppm). In these dilute streams, the reaction rate is inevitably bottlenecked by mass transport rather than intrinsic catalytic activity, leading to severe concentration polarization [29-31].
As the applied overpotential increases, the rate of NO3- consumption outpaces its diffusion rate until the surface concentration effectively reaches zero. At this point, the system hits its mass-transfer limiting current density (iL), which mathematically bounds the maximum processing capacity of the reactor:
Where n=8 for NH3 synthesis, F is Faraday’s constant, ${\mathit{D}}_{\mathit{N}{\mathit{O}}_{3}^{-}}$ is the diffusion coefficient of nitrate, and Cbulk is the bulk nitrate concentration.
Equation 2 reveals that for a given wastewater stream (fixed Cbulk), the only engineering pathway to elevate ${\mathit{i}}_{\mathit{L}}$ is to minimize δ. Conventional H-cells suffer from thick diffusion layers (δ>50μm), leading to nominal limiting currents often below 10 mA cm-2 for dilute feeds. To break this mass transport limit, modern continuous flow reactors are engineered to minimize δ（down to <10μm） through turbulent flow fields or by forcing the electrolyte through hierarchical 3D porous electrode matrices. Decoupling the bulk hydrodynamics from the boundary layer thickness is the foundational premise behind advanced flow cells and membrane electrode assemblies (MEAs), marking the critical transition from laboratory testing to industrial piloting.
3. Conventional batch reactors
Historically, the vast majority of literature on the NO3RR has relied on conventional batch reactors [32,33]. Depending on the presence or absence of an ion-combining separator, these batch systems are broadly categorized into undivided (single-chamber) and divided (dual-chamber) configurations. While indispensable for early-stage material science, both devices present distinct and insurmountable engineering bottlenecks for practical scale-up.
3.1 Undivided single-chamber reactors
In an undivided single-chamber reactor, the working electrode, counter electrode, and reference electrode are all submerged in a single, continuous pool of electrolyte without any physical barrier or membrane separating them. From a purely operational standpoint, single-chamber cells are the most rudimentary and inexpensive to assemble. Because they eliminate the need for an ion-exchange membrane, they fundamentally bypass the substantial ionic resistance associated with trans-membrane ion transport. Consequently, for a given applied current, single-chamber reactors typically exhibit a lower overall cell voltage compared to their divided counterparts, theoretically improving the voltage efficiency of the system [34,35].
Despite the lowered ohmic resistance, undivided reactors are fundamentally unviable for practical NH3 electrosynthesis due to severe cross-contamination and parasitic redox cycling. Without a separating membrane, the ammonia (or ammonium, NH4+) generated at the cathode is free to diffuse through the bulk electrolyte toward the anode. Under the highly oxidizing potentials required to drive the anodic oxygen evolution reaction (OER), the newly synthesized NH3 is rapidly re-oxidized back into N2, NO2-, or NO3-. This internal short-circuiting of the nitrogen cycle creates a futile loop that severely depresses the measurable Faradaic efficiency (FE) and overall yield of NH3. Furthermore, the mixing of anodic oxygen gas and cathodic H2 from the competing HER poses significant safety hazards, particularly explosive risks at larger scales. Therefore, while single-chamber batch cells may be used for preliminary, low-current catalyst screening, they are virtually useless for evaluating actual NH3 production metrics.
3.2 Divided dual-chamber reactors
To resolve the catastrophic re-oxidation issues inherent to undivided cells, researchers universally turn to the divided dual-chamber reactor, most commonly known as the H-type cell [36-38]. Structurally, it consists of two distinct glass compartments connected by a narrow channel and strictly separated by an ion-exchange membrane.
The H-cell remains the workhorse of fundamental electrocatalysis. By compartmentalizing the half-reactions, the membrane effectively prevents the anodic re-oxidation of NH3 and allows for accurate product quantification. Furthermore, the H-cell’s static geometry and easily sealed headspace make it highly amenable to in-situ or operando analytical tools (e.g., DEMS, FTIR), enabling researchers to dynamically track volatile intermediates and elucidate complex 8-electron reaction pathways.
Despite its diagnostic utility, the H-cell presents severe engineering flaws that inherently preclude its industrial scale-up. Primarily, its reliance on macroscopic magnetic stirring fails to disrupt the thick, stagnant hydrodynamic boundary layer, creating profound mass transport limitations that restrict operational current densities to typically <100 mA cm-2, an order of magnitude short of the ≥1 A cm-2 commercial threshold. Compounding this mass transfer bottleneck is the massive ohmic (IR) resistance induced by the large inter-electrode distance (several centimeters) and the central ion-exchange membrane, which leads to severe voltage efficiency losses and unmanageable Joule heating at higher applied currents. Furthermore, the inherent transient nature of batch operation creates a non-steady-state environment. Together, these profound diffusion limits, severe thermodynamic energy losses, and the inability to sustain continuous processing fundamentally restrict the H-cell’s viability for real-world wastewater remediation or large-scale electrosynthesis.
4. Continuous flow reactors
Transitioning NO3RR from bench-scale H-type cells to practical applications necessitates overcoming severe mass transport limitations. In static or batch systems, sluggish transport kinetics often mask the intrinsic activity of electrocatalysts at high current densities. Continuous flow reactors address this by employing hydrodynamics to actively manipulate the mass transport of reactants to the catalyst interface, serving as a vital engineering strategy to achieve industrial-level current densities [39-41].
4.1 The flow-by configuration
In a flow-by reactor architecture, the liquid electrolyte is pumped in a cell that runs strictly parallel to the planar surface of the solid electrode [42,43]. This configuration is remarkably effective at rapidly renewing the bulk electrolyte composition and decisively mitigating the accumulation of insulating gaseous phases. The continuous, parallel shear force stabilizes the macroscopic pH of the chamber and prevents the localized buildup of highly concentrated nitrite intermediates that can poison the catalyst surface.
Sun et al. developed a packed flow cathode reactor with serpentine flow channels and Ir-Ta mesh current collectors separated by a cation exchange membrane (Figure 2a-b). A Cu@AC particle suspension circulated through the cathode channel at 9 mL min-1 (HRT: 0.56 min), with ammonia recovered via a hydrophobic gas membrane into an acid chamber. At 20 mA cm-2, the system achieved ~100% nitrate removal, 97% ammonia selectivity, and 70% ammonia recovery. The flow cathode showed excellent stability with negligible Cu leaching and maintained performance over five reuse cycles [40]. Zhang et al. developed a flow reactor using CuCo branched nanowires on Cu foam as the cathode and Ni foam as the anode, separated by an anion exchange membrane (Figure 2c-d). Electrolytes were fed via peristaltic pumps, with NO3- diffusing through the porous Cu foam to the nanowire surface for reduction. The system achieved 90.3% Faradaic efficiency at 1.0 A cm-2 and maintained stable operation for 200 h at 100-200 mA cm-2. A scaling-up prototype with 20 cm2 active area delivered a gram-level ammonia yield rate of 1474 mg h-1 at 20 A, demonstrating practical potential for continuous NO3RR [44].
Despite these advantages, the physical interaction between the dissolved nitrate ions and the solid catalyst surface remains fundamentally constrained by boundary layer diffusion at the immediate, two-dimensional solid-liquid interface. For practical environmental applications treating highly dilute wastewater streams (where the bulk nitrate concentration often falls below 100 mg-N L-1), the flow-by mode frequently struggles to achieve high single-pass conversion efficiencies. A significant fraction of the fluid momentum physically bypasses the reactive interface entirely without making the requisite intimate contact with the nanoscopic catalyst active sites.
4.2 The flow-through configuration
To reach favorable mass transport and maximize catalyst utilization, the flow-through configuration forces the pressurized electrolyte stream to perpendicularly penetrate the macro- and mesoporous void networks of a truly three-dimensional electrode structure. Advanced materials such as highly conductive carbon paper, woven carbon cloth, metallic copper or nickel foams, and specialized titanium-based hollow fiber electrodes are utilized to construct a highly open-framework pathway for the incoming fluid [45-47].
The flow-through structure ensures extreme, intense localized interfacial mixing. As the nitrate-laden electrolyte is forcefully extruded through the tortuous void spaces of the 3D matrix, the effective diffusion distance between the bulk fluid and the active sites is compressed from the macroscopic scale down to the micrometer or even nanometer scale. This effectively overcomes classical mass transfer limitations, enabling rapid reduction kinetics even at extremely depleted nitrate concentrations. Wang et al. developed a nanoporous electro-filtration system employing a free-standing carbon nanotube (CNT) interwoven membrane coated with Fe single-atom catalysts (Fe1/NCBd@CNT-FEM) for nitrate reduction under realistic water conditions (Figure 3a-d) [48]. In this flow-through configuration, the feed solution is driven through the membrane pores under an applied electric field, enabling advective transport of nitrate molecules directly to the highly exposed single-atom active sites within the confined nanopores. This design reduces the diffusion boundary layer thickness to the length scale of the pore radius (~19.3 nm), significantly enhancing mass transport and catalyst utilization compared to conventional flow-by electrodes. At an optimal current density of 6.4 mA cm-2, the membrane achieved 86.1% nitrate removal efficiency and 86.4% Faradaic efficiency for a low-concentration feed (100 mg N L-1) with a residence time of merely 10 seconds, delivering an exceptionally high ammonia turnover frequency of 15.1 g N g-1 metal h-1, up to four orders of magnitude higher than literature values for flow-by systems. The authors further demonstrated the scalability of this approach through sequential electrofiltration using stacked membranes, achieving near-complete nitrate removal (>98.5%) and stable long-term operation over 12 hours with negligible Fe leaching (<0.4%).
Dai et al. constructed an integrated flow-through electrolyzer featuring a monolithic spin-polarized Fe1-Ti pair cathode (SP-Fe1-Ti) coupled with a membrane-based ammonia recovery unit for continuous nitrate reduction and in-situ product separation (Figure 3e-f) [49]. The SP-Fe1-Ti electrode was fabricated by anchoring Fe single atoms onto the inherent surface oxide layer of titanium foam with oxygen vacancy-mediated spin polarization. In the flow-through cell, NO3--containing synthetic effluent (250 mL) was continuously circulated through the cathode chamber at 60 mL min-1, where NO3- was electrochemically reduced to NH3 at the cathode surface. The ammonia-rich effluent was then directed through hydrophobic hollow polypropylene fiber membranes into a recovery chamber containing 1 M HCl, enabling selective gas-permeable extraction and absorption of NH3. This integrated system robustly operated at an industrial-level current density of ~200 mA cm-2 for 200 hours with a high Faradaic efficiency of ~95%, achieving nearly 100% ammonia selectivity and ~90% recovery efficiency. The produced ammonia was subsequently recovered as high-purity NH4Cl crystals through rotary evaporation, demonstrating a sustainable and decentralized approach for upgrading nitrate wastewater to fertilizer products. Zhou et al. integrated the I1Cu4 single-atom electrode into a flow-through device for continuous nitrate conversion and in situ ammonia recovery [50]. NO3--containing wastewater was circulated through the cathode chamber at 60 mL min-1, with NH3 subsequently stripped through hollow polypropylene fibers into HCl solution. At 50 mA cm-2, the device achieved 99.5% nitrate conversion and 98.7% ammonia recovery over 100 h. For high-concentration feeds (1 M KNO3), an industrial-level current density of 1 A cm-2 delivered an NH3 yield rate of 69.4 mg h-1 cm-2 with 88.0% Faradaic efficiency over 110 h. The system reduced total nitrogen in real industrial wastewater from 105 to 3.8 mg N L-1, meeting discharge standards.
5. Membrane Electrode Assemblies (MEAs)
Beyond overcoming mass transport limitations, the structural evolution of NO3RR reactors is fundamentally driven by the need to minimize overall electrical energy consumption. In conventional H-cells, the macroscopic inter-electrode distance induces severe bulk ohmic drops, requiring high cell voltages that result in massive, unviable energy penalties per kilogram of synthesized NH3. Continuous flow reactors partially mitigate this by lowering concentration overpotentials through enhanced hydrodynamics; however, the persistent bulk liquid gap between electrodes still imposes measurable energy losses. In stark contrast, ZG MEA architectures compress the inter-electrode distance to the microscopic thickness of the polymer membrane itself, which profoundly reduces the overall cell voltage. By minimizing these ohmic losses and operating at low applied voltages, highly optimized MEA systems can couple with in-situ product recovery with low energy input, demonstrating a decisive energetic advantage necessary for industrial scale-up [51-53].
The MEA represents the absolute zenith of modern electrochemical reactor engineering, transitioning the field from passive reaction containment to highly active, precision optimization of molecular pathways. In a quintessential ZG MEA configuration, the porous cathodic catalyst layer and the complementary anodic catalyst layer are physically hot-pressed, sprayed, or otherwise intimately bound directly against the opposing faces of a solid polymer ion-exchange membrane (Figure 4a) [54]. By completely obliterating the bulk liquid electrolyte chamber that traditionally separates the anode and cathode, the total inter-electrode distance is compressed to merely the microscopic thickness of the central polymer membrane itself. This ZG spatial architecture drastically decreases the ohmic resistance of the electrochemical cell, allowing massive electrical currents to flow at remarkably low applied operational voltages. For instance, highly optimized, state-of-the-art ZG flow-through systems utilizing advanced amorphous copper domain catalysts have achieved astonishing partial current densities exceeding 3.33 A cm-2 at an incredibly low overall cell voltage of just 2.6 V [55].
Furthermore, by continuously feeding highly concentrated liquid reactants or gaseous precursors directly to the rear surfaces of engineered Gas Diffusion Electrodes (GDEs) or specialized porous transport layers, the MEA establishes an exquisitely optimized triple-phase microenvironment [56-59]. This approach sustains high-flux nitrate delivery directly to the reactive metallic centers while allowing for the precise, localized regulation of proton availability. However, the aggressive transition to ZG MEAs introduces a host of novel, deeply complex chemical engineering challenges. Operating continuously at intense current densities heavily exacerbates fluidic issues such as catastrophic electrode flooding, where the nanoscopic porous networks become terminally saturated with liquid, physically blocking mass transport pathways, as well as the induction of severe salt precipitation and scaling within the GDEs and active catalyst layers. Consequently, the rational selection, physical integration, and atomic-level engineering of the central solid polymer ion-exchange membrane become the absolute most critical determinants governing overall system performance, product selectivity, and device longevity.
Within the stringent confines of a ZG MEA, the polymer membrane transcends its rudimentary role as a mere physical separator; it acts as a highly dynamic, active participant in the complex reaction chemistry, dictating the selective transport of diverse ionic species, rigorously managing the localized pH gradients, and serving as a critical physical barrier against detrimental product crossover. Modern NO3RR MEA research predominantly leverages three distinct, highly specialized classes of polymeric membranes: Cation Exchange Membranes (CEMs), Anion Exchange Membranes (AEMs), and highly advanced Bipolar Membranes (BPMs) (Figure 4b).
5.1 Cation exchange membranes (CEMs)
CEMs, typified most prominently by the industry-standard perfluorosulfonic acid polymers such as the Nafion series (e.g., Nafion-117, Nafion-115, and Nafion 211), are synthetically engineered to selectively facilitate the transport of positively charged cations (such as H+ or Na+) from the anodic chamber directly to the cathodic interface while violently repelling negatively charged anions via the electrostatic principles of Donnan exclusion.
The primary structural advantage of CEMs lies in their exceptional chemical robustness and profound mechanical stability, traits that are particularly crucial when operating in highly aggressive, acidic environments [61-63]. They provide extraordinary proton conductivity, functioning flawlessly as solid electrolytes in humidified environments, which is an absolute necessity for rapidly supplying the massive flux of 9H+ protons required for the complete reduction of a single nitrate anion to a molecule of ammonia. Furthermore, the dense network of negatively charged sulfonate functional groups permanently affixed within the polymer matrix effectively repels the incoming, negatively charged primary NO3- and transient nitrite NO2- anions. This electrostatic shielding prevents these critical species from crossing over to the highly oxidizing anode where they would be uselessly re-oxidized, thereby rigorously preserving the Faradaic efficiency of the cathodic conversion. Li et al. develops a continuous-flow PEM membrane electrode assembly (MEA) electrolyzer based on Nafion 115 cation exchange membrane (CEM) for electrochemical nitrate-to-ammonia conversion (Figure 5a-b) [64]. Unlike batch H-cells, the ZG CEM-MEA structure minimizes ohmic resistance and enables long-term steady operation with pure nitrate wastewater as feedstock, eliminating corrosive alkaline media required by alkaline flow reactors. The pretreated Nafion CEM separates cathode and anode chambers: dilute KNO3 solution flows through the cathode side, while Milli-Q water circulates at the anode with IrO2 oxygen evolution catalyst. Under 80 °C operation, the CEM-MEA delivers stable ammonia production over a 10 h chronoamperometry test, achieving high single-pass nitrate conversion and ammonia Faradaic efficiency.
Despite these strengths, the utilization of CEMs in NO3RR MEAs shows significant, sometimes fatal, drawbacks. The primary vulnerability stems from their inherent, indiscriminate permeability to positively charged species, which tragically facilitates the rapid crossover of the newly synthesized target product NH4+ [65]. As ammonia is successfully synthesized at the cathodic active sites, it inevitably protonates in the locally acidic or neutral aqueous environment to form stable ammonium [32]. This valuable NH4+ product is then inevitably drawn across the CEM toward the anode by the powerful electric field spanning the gap, resulting in product loss, decreased yield rates, and the potential for parasitic anodic oxidation back to nitrogen gas [32]. Wilder et al. quantified severe NH4+ crossover across Nafion CEMs in NO3RR MEAs (Figure 5c) [66] Without applied voltage, massive NH3/NH4+ permeation continuously occurs inside the membrane, severely lowering detected ammonia yields. While -1.6 V operating voltage partially restricts product transport via electroosmotic drag, this suppression fails during intermittent open-circuit stages typical for renewable power supply. Adjusting Nafion thickness only delays crossover onset instead of eliminating inherent cation permeability. This work proves unmodified perfluorosulfonic CEMs possess an intrinsic ammonium leakage flaw, necessitating targeted membrane modification to block cation penetration for high-efficiency NO3RR MEA systems.
5.2 Anion exchange membranes (AEMs)
In stark contrast to CEMs, AEMs, encompassing advanced commercial polymers such as FAA-3-50 and PK-130, feature dense arrays of positively charged functional groups grafted onto the polymer backbone [67]. These groups selectively conduct anion away from the cathode and toward the anode, effectively running the ionic circuit in reverse relative to a PEM system [68-70].
AEMs demonstrate exceptional suitability and superiority for MEA systems engineered to operate within highly alkaline liquid media. In an alkaline NO3RR environment, the surrounding water solvent acts as the primary proton donor, generating copious, potentially paralyzing amounts of hydroxide at the immediate cathode surface according to the modified half-reaction: NO3-+6H2O+8e-→NH3+9OH-. The AEM rapidly and efficiently transports these locally generated, concentrated hydroxide ions away from the delicate cathodic interface directly to the anode to be consumed. This mass transfer successfully completes the electrical circuit while strictly preventing the localized cathodic microenvironment from becoming excessively, destructively caustic.
Zhang et al. adopted FAA-3 AEMs to assemble a ZG NO3RR MEAs reactor under alkaline media. The AEM selectively blocks cationic NH4+ to avoid product crossover, a key advantage over CEM devices (Figure 6a-c) [71]. Equipped with NiCo LDH/Cu NW tandem cathodes, this AEM-MEA maintained >90% NH3 FE at industrial current density. Stable performance lasting 20 h at 2.0 V cell voltage was realized, along with high ammonia production rates. This work validates the great scalability of AEM ZG membrane electrode systems for industrial nitrate wastewater treatment and green ammonia electrosynthesis. Luo et al. fabricated a scalable membrane electrode flow reactor assembled with commercial AlkymerW-25 AEM for alkaline NO3RR [72]. The MEA uses CoCu2O@CoCuHHTP as cathode and Ir-black for anodic oxygen evolution (Figure 6d-e). Under galvanostatic 100 mA cm-2 operation, the AEM-MEA retained over 80% ammonia Faradaic efficiency across 1800 h long-term cycling. Benefiting from the catalyst’s hydrogen-bonding modulation, the AEM flow cell delivered stable industrial-grade performance with mild degradation of active sites, demonstrating the practicability of AEM-equipped MEAs for continuous nitrate wastewater treatment and long-duration green ammonia electrosynthesis.
However, the weakness of the AEM architecture lies in its very function: the transport of anions. AEMs are inherently, severely susceptible to massive NO3- crossover. Unreacted NO3- anions present in the catholyte can easily permeate through the highly conductive AEM to the anodic chamber, particularly when driven by the intense electric fields present under high current densities [73]. This crossover leads to a direct, systemic loss of the primary reactant, drastically decreasing the overall conversion efficiency of the electrolyzer and potentially contaminating the anodic output streams.
5.3 Bipolar Membranes and interfacial engineering
To systematically integrate the respective operational merits of CEMs and AEMs, and alleviate their severe crossover drawbacks, BPMs have gained growing attention for NO3RR reactor engineering [74-76]. A true BPM is a composite structure consisting of a discrete CEM layer physically laminated directly to an AEM layer, creating a distinct, highly active 3D interfacial junction between the opposing polymers. When the BPM is operated under reverse electrical bias, wherein the AEM face is oriented toward the anode and the CEM face is oriented toward the cathode, the intense electrical field localized at this central polymeric junction drives the continuous, rapid dissociation of ambient water molecules into individual protons (H+) and hydroxide ions (OH-). The newly birthed H+ rapidly migrates outward through the CEM layer toward the cathode, while the OH- migrates in the opposite direction through the AEM layer to feed the anode.
The deployment of BPMs confers advantages upon the MEA system, most notably providing a unique, unparalleled dual-pH operational functionality. They permit the anodic and cathodic chambers to operate simultaneously at entirely divergent pH extremes without any bulk mixing of the incompatible liquid electrolytes. Consequently, the anode can be bathed in a highly favorable, strongly alkaline environment to drastically lower the overpotential required for the OER, while the cathode simultaneously receives a steady, precise flux of protons necessary to drive NO3RR [77]. Xu et al. developed a mortise-tenon structured bipolar membrane (MBM) paired with flow MEA for industrial-grade nitrate electroreduction to ammonia (Figure 7a-b) [74]. Unlike CEM and AEM separators that suffer severe ion crossover, the bipolar membrane spontaneously dissociates water to supply cathode H+ and anode OH-, realizing internal ionic balance without extra acid/alkali additives. The 3D interlocked interface greatly boosts water dissociation kinetics and alleviates membrane blistering under high current. Assembled with Co 3D nanoarray cathode, this BPM-MEA reactor stably operates at 1000 mA cm-2 over 100 h, delivering 86.2% ammonia Faradaic efficiency and a yield of 68.4 mg h-1 cm-2 with dilute nitrate wastewater. Boppella et al. systematically contrasts AEM and BPM MEAs for nitrate electroreduction using CuRu alloy cathodes (Figure 7c-f) [78]. Distinct from AEMs that only transport anions, BPMs spontaneously split water to self-generate internal H+ and OH-, eliminating external acid or alkali additives while blocking NO3- crossover to the anode. Such built-in proton supply stabilizes the nitrate-to-ammonia hydrogenation pathway, delivering reliable high ammonia Faradaic efficiency (>90%) even in dilute wastewater. Though BPM incurs mild extra water-splitting overpotential, its self-balanced ionic microenvironment renders it a uniquely self-sustained membrane reactor for decentralized nitrate remediation.
Despite these capabilities, raw BPM performance must be heavily modulated to optimize ammonia yield. While BPMs supply necessary protons to the cathode, an unmitigated, high-velocity proton flux at elevated industrial current densities inevitably saturates the active sites, triggering aggressive, dominant hydrogen evolution. To master this behavior, advanced MEA cell designs incorporate highly sophisticated interface engineering, primarily through the insertion of an interposer layer (IPL) [75]. The IPL frequently comprised of a highly engineered, mixed cellulose ester membrane filter, is carefully sandwiched directly between the outward face of the BPM and the active cathodic catalyst layer. The IPL functions mechanically as a highly tunable proton fence, physically throttling, diffusing, and regulating the rate of absolute proton delivery from the membrane to the catalyst active sites. By limiting excessive proton supply to the catalyst while retaining sufficient proton availability to support the intricate PCET steps of NO3RR, the IPL achieves drastic and selective suppression of the competing HER. Experimental implementations of this advanced architecture, utilizing a BPM outfitted with a precision IPL and a carbonate buffering electrolyte to manage localized pH spikes, have demonstrated spectacular results. Such designs have successfully forced a reduction in HER Faradaic efficiency from a dominant 80% down to a mere 40% at demanding current densities of 200 mA cm-2, culminating in an 576% increase in overall ammonia yield when compared to standard, unmodified MEA setups lacking an IPL.
5.4 Optimization of flow fields and bipolar plate design
Within the dense, highly pressurized confines of a multi-cell stack, bipolar plates are the foundational structural components, serving a critical dual purpose [79]. First, they physically and fluidically separate adjacent electrochemical cells, ensuring absolute isolation of distinct reactant streams to prevent catastrophic cross-contamination. Second, they establish an integrated, highly efficient electrical series connection across the entire stack. In this configuration, one face of the highly conductive plate operates as the anode for a given cell, while the direct opposite face simultaneously operates as the cathode for the immediately adjacent, succeeding unit. This ingenious bipolar configuration maximizes volumetric spatial utilization and drastically minimizes internal electrical resistance by entirely eliminating the need for bulky, resistive external metallic wiring between individual cell units.
The physical machining, routing, and geometry of these bipolar plates are absolutely critical determinants of stack-wide fluid dynamics and mass transport. Traditionally, small-scale laboratory flow cells rely heavily on a serpentine runner configuration [75-80]. In this design, the liquid reactant is forcibly pumped through a single, continuous, highly winding, snake-like channel deeply machined into a conductive current collector, such as a thick titanium or graphite plate. While serpentine flow fields are exceptionally adept at ensuring high localized reactant velocity and forcing the fluid to penetrate the adjacent porous transport layers, they generate massive, unavoidable hydrostatic pressure drops across their lengthy travel path. In a scaled-up, multi-cell system processing massive volumetric flows, attempting to pump viscous fluids through extensive serpentine networks requires immense, economically prohibitive mechanical pumping energy, rendering the design functionally unviable for commercial nitrate treatment.
To circumvent this severe mechanical limitation, industrial-scale NO3RR stacks are rapidly migrating toward advanced full runner or parallel flow field structures. In a full runner design, the high-pressure fluid enters a central distribution manifold and is systematically divided, flowing simultaneously across multiple, short parallel channels spanning the entire active geometric area of the bipolar plate. This structural reconfiguration radically reduces the cumulative pressure drop across the cell while simultaneously ensuring highly uniform, homogenous distribution of the nitrate-rich electrolyte over much larger catalyst arrays. A recent, landmark technological demonstration highlighted the incredible efficacy of this approach by successfully constructing and deploying a scaled-up, full runner MEA stack utilizing multiple 4ⅹ25 cm2 active area plates (Figure 8) [81]. Under continuous, industrial-level operational loads, this massive architecture delivered an extraordinary 40.8 A of total electrical current while maintaining a highly efficient stack voltage of merely 10 V. Crucially, the system retained exceptional catalytic fidelity, achieving a high, stable ammonia Faradaic efficiency of 92.6% over extended operational durations, thereby unequivocally proving the commercial viability of full runner scale-up.
6. Scalable precision manufacturing of reactor components
To fully realize the promise of rational manufacturing, the transition from bench-scale testing to industrial NO3RR deployment requires abandoning rudimentary manual fabrication in favor of scalable, high-precision manufacturing techniques. The electrochemical performance of flow reactors and MEAs is profoundly dictated by a direct tri-relation: manufacturing parameters dictate the component microstructure, which in turn governs macroscopic mass transport and catalytic efficiency.
6.1 Ultrasonic spray coating and roll-to-roll slot-die technologies
For zero-gap membrane electrode assemblies, ultrasonic spray coating prevents nanoparticle agglomeration to deposit highly uniform, pinhole-free catalytic layers. By strictly controlling manufacturing parameters such as ultrasonic nozzle frequency and substrate heating temperature, engineers can manipulate initial droplet size and solvent evaporation kinetics [82]. This induces micron-scale Marangoni flows that create heavily textured catalytic layers populated with large macroscopic pores. This specific microstructural architecture significantly decreases liquid mass transport resistance and alleviates electrode flooding, sustaining high Faradaic efficiencies. For massive industrial scale-up, roll-to-roll slot-die and gravure coating strictly define the coating gap and fluid shear stress, ensuring structural homogeneity over large-area assemblies without catastrophic reactant crossover [83].
6.2 Precision electrodeposition for binder-free hierarchical cathodes
Electrodeposition offers a bottom-up manufacturing approach to fabricate binder-free, highly porous cathodes directly onto macroscopic conductive substrates [84.85]. Applying intentionally high galvanostatic current densities induces vigorous simultaneous HER during metal deposition. These rapidly evolving gas bubbles act as dynamic templates, forcing the depositing metal to grow into highly porous, interconnected nanowire arrays [86]. This engineered hierarchical porosity exponentially increases the electrochemically active surface area and creates micro-channels that accelerate the diffusion of dilute nitrate to inner active sites, decisively suppressing extreme localized alkalinity and mineral scaling.
6.3 Hot-pressing assembly and interfacial engineering
The physical lamination of the porous cathode, solid polymer membrane, and anode under carefully controlled thermal and mechanical conditions is critical for minimizing the entire device’s interfacial contact resistance [87]. Applying optimal compressive pressure at temperatures above the ionomer’s glass transition temperature forces polymeric chains to flow and interlock intimately with the catalyst layer’s nanopores [88]. Furthermore, precisely spraying an ultra-thin ionomer overlayer directly onto the gas diffusion electrode prior to hot-pressing neutralizes the substrate’s inherent micro-roughness. This interfacial engineering modification guarantees a seamless triple-phase boundary, preventing catastrophic membrane delamination under the intense hydrodynamic shear of continuous flow operations [89].
6.4 Three-dimensional additive manufacturing for hydrodynamic control
Traditional porous metal foams possess stochastic, randomly distributed pore networks that trap gaseous byproducts and create unpredictable fluid pathways [90]. Conversely, 3D printing enables the deterministic design of macroscopic topologies and internal pore geometries. By programming 3D printing parameters such as laser power and scanning speed, techniques like selective laser melting can construct customized, geometrically periodic lattice structures. These ordered architectures are explicitly designed to minimize the three-phase contact length and accelerate rapid bubble detachment, ensuring that active catalytic sites remain continuously exposed to the liquid electrolyte even during high-rate NO3RR electrolysis (Figure 9a-d) [91-93].
6.5 Laser micromachining and subtractive surface ablation
Laser micromachining is deployed as a highly precise subtractive manufacturing tool to modify the surface morphology of finished electrodes or carbon-based gas diffusion layers (Figure 9e) [94-96]. Adjusting the laser fluence and geometric scanning pattern selectively ablates the material to etch controlled micro-grooves and via-holes directly into the substrate. These ablated micro-channels act as super-highways for multi-phase fluid transport, efficiently distributing aqueous electrolytes to eradicate localized concentration polarization while simultaneously providing direct, low-resistance venting routes for the rapid extraction of generated gases (Figure 9e).
7. Integration of reactive separations and ammonia recovery
A highly critical, yet historically overlooked and under-engineered facet of overall NO3RR reactor design is the downstream capture, isolation, and purification of the electrochemically synthesized ammonia [65,97,98]. The deployment of the highest-performing single-atom catalyst embedded within the most structurally flawless MEA stack is rendered functionally and economically useless if the generated ammonia cannot be efficiently and cleanly separated from the massive volume of the complex, unreacted, multi-component wastewater matrix [16]. Because ammonia is exceptionally soluble in aqueous media and exists in a delicate, highly pH-dependent equilibrium with the stable ammonium ion ($NH_{3}+H_{2}O\rightleftharpoons NH_{4}^{+}+OH^{-}$, pKa ≈ 9.25), selectively extracting it without precipitating other dissolved solids poses a monumental thermodynamic and chemical engineering hurdle.
7.1 Advanced membrane stripping and in situ distillation
Recent, some advances in reactor architecture have pioneered the seamless integration of highly specialized GDEs coupled with intensely hydrophobic, gas-permeable membranes directly into the primary flow reactor body to achieve continuous, in situ reactive separation. In these highly sophisticated, multi-chamber electrified membrane systems, the actual NO3RR electrocatalysis takes place precisely at a porous cathodic catalyst layer that is intimately, physically bound to a robust, highly hydrophobic polytetrafluoroethylene (PTFE) membrane [45,99].
As the localized pH at the immediate cathode surface aggressively spikes due to the massive consumption of protons and the concurrent generation of hydroxide ions during NO3RR, the local chemical equilibrium is violently shifted, driving the conversion of aqueous ammonium entirely into highly volatile, free ammonia gas (NH3) [65]. The intensely hydrophobic nature of the PTFE membrane acts as an absolute physical barrier, strictly preventing liquid water or dissolved salts from passing through [100,101]. However, it readily allows the highly volatile NH3 gas to rapidly permeate across the boundary, driven by the concentration gradient, directly into an isolated, dedicated trap chamber situated immediately on the reverse side of the membrane. Within this isolated trap chamber, the gaseous ammonia is immediately and aggressively scrubbed by a circulating, highly acidic receiving solution (such as concentrated sulfuric acid). Upon contact, the ammonia undergoes an instantaneous, irreversible acid-base reaction, precipitating directly as a high-purity, highly stable, and commercially valuable liquid fertilizer product, such as ammonium sulfate ((NH4)2SO4). Gao et al. constructed a three-chamber electrified membrane flow cell for synchronous nitrate reduction and in-situ ammonia recovery free of extra acid or alkali (Figure 10) [102]. CuO@Cu foam cathode produces local high pH to convert NH4+ into gaseous NH3, which permeates hydrophobic PTFE membrane and is captured as (NH4)2SO4 in acidic anolyte. Anodic hydrogen oxidation reduces full-cell energy loss. The system achieves 99.9% nitrate removal and 99.5% ammonia recovery, with only 21.8 kWh kg-1 energy input for (NH4)2SO4 production. Stable performance on real industrial wastewater confirms the integrated reaction-separation design for circular nitrogen resource recovery.
7.2 Challenge in real-world aquatic environments
Despite the monumental, undeniable strides achieved in nanoscale catalyst design, macroscopic reactor geometry optimization, advanced membrane science, and integrated in situ product recovery techniques, the physical deployment of massive NO3RR reactors into actual industrial facilities or agricultural runoff sites introduces severe operational friction. This friction stems fundamentally from highly complex, unpredictable matrix effects inherent to raw wastewater and the unforgiving reality of long-term energy economics [103-105].
The vast majority of academic, laboratory-scale studies predominantly evaluate novel reactors and catalysts utilizing highly purified, synthetically prepared electrolytes consisting solely of clean water, specific nitrate salts, and perhaps benign supporting buffers. However, the reality of real-world chemical feedstocks, such as raw agricultural runoff, heavy industrial effluents, and untreated natural groundwater, presents incredibly complex, chemically aggressive matrices heavily laden with high concentrations of co-existing ionic species (most notably Ca2+, Mg2+, SO42-, and Cl-), complex natural organic matter (NOM), and abrasive particulate foulants [106].
The most insidious, fatal threat to long-term reactor stability and continuous operation in these harsh environments is aggressive mineral scaling localized at the cathodic interface. Because the NO3RR process fundamentally and heavily consumes massive quantities of protons, it inherently produces extreme localized alkalinity directly at the electrode-electrolyte interface. As ambient, ubiquitous hardness ions like calcium and magnesium are swept into this highly alkaline microenvironment by the reactor’s flow systems, they rapidly and unavoidably precipitate out of solution, forming dense, rock-like, highly insoluble carbonate and hydroxide scales (predominantly CaCO3 and Mg(OH)2) [107]. Continuous flow reactor evaluations physically processing actual nitrate-polluted natural groundwater have routinely demonstrated catastrophic, complete system failure within a mere matter of days. The rampant formation of these insulating scaling layers physically occludes and blinds the atomically engineered active catalytic sites, severely restricting critical mass transport pathways through the delicate porous flow-through matrices, and consequently induces a rapid, fatal decline in both ammonia Faradaic efficiency and overall nitrate removal rates.
Mitigating this paralyzing phenomenon necessitates profound, further advancements in dynamic reactor engineering, most notably through the mandatory implementation of pulsed electrolysis protocols. Fan et al. identifies severe catalyst fouling by Mg2+, Ca2+ and trace metals in real nitrate wastewater (Figure 11) [108]. Finite element and DFT simulations visualize interfacial ion distribution differences between static and pulsed electrolysis. Static operation accumulates high-concentration Mg2+ near electrodes, while pulsed fields repel Mg2+ and enrich K+ at the surface. Adsorption energy calculations reveal elevated Mg2+ binding barriers with increased K+ coverage via electrostatic repulsion and steric hindrance. Three synergistic anti-scaling mechanisms are validated: suppressed local alkalization, electric-field cation repulsion, and interfacial K+ enrichment to block scale-forming ions. However, the implementation of pulsed electrolysis intrinsically introduces unavoidable engineering trade-offs regarding energy consumption and overall production rates. Because the system operates under a non-continuous duty cycle, the time-averaged absolute yield of ammonia is inherently reduced compared to continuous direct current (DC) operation at the same peak current density. Furthermore, the continuous, rapid fluctuation of the applied electric field requires repeatedly charging and discharging the electrochemical double layer at the electrode-electrolyte interface. This capacitive charging acts as a parasitic energy sink, dissipating electrical energy without driving the Faradaic reduction of NO3-, thereby imposing a measurable energy penalty on the overall system. Therefore, optimizing the pulse frequency and duty cycle is a delicate engineering balancing act: maximizing the electrostatic repulsion of multivalent cations like Ca2+ and Mg2+ while minimizing capacitive energy losses and yield reductions. Furthermore, integrating robust upfront, non-electrochemical anti-scaling pretreatments, coupling the reactor with robust electrodialysis pre-modules, or utilizing highly specific electrified nanoporous membranes explicitly designed to repel multivalent cations via strict size exclusion or intense Donnan exclusion mechanisms will be absolutely mandatory for the viable, long-term deployment of massive NO3RR arrays in real-world settings.
8. Conclusion
NO3RR has emerged as a highly compelling, carbon-neutral strategy to simultaneously remediate nitrate-contaminated wastewater and synthesize high-value ammonia. While the past decade has witnessed tremendous breakthroughs in fundamental electrocatalyst design, the practical industrialization of NO3RR remains severely bottlenecked by macroscopic engineering challenges, primarily sluggish mass transport, concentration polarization, and extreme localized pH fluctuations.
To overcome the divide between laboratory screening and commercial deployment, this review emphasizes the necessary shift from material science to chemical engineering. Conventional batch reactors, e.g., H-cells, are inherently restricted by thick diffusion boundary layers and massive ohmic losses, making them incapable of sustaining industrial-level current densities. In contrast, continuous flow-by and flow-through architectures actively manipulate fluid dynamics to obliterate mass transport limits, while zero-gap MEAs drastically reduce inter-electrode resistance. The judicious selection of ion-exchange membranes profoundly dictates system performance by managing ionic crossover and preserving the delicate interfacial microenvironment. Furthermore, coupling these advanced reactors with downstream hydrophobic membrane stripping units and utilizing pulsed electrolysis represents a vital step toward practical application.
Despite these research progresses, the widespread commercialization of NO3RR technology necessitates further systematic innovation. Future research should prioritize the following key engineering frontiers:
(1) Advanced membrane and interfacial engineering in MEAs. While zero-gap MEAs are essential for high-current-density operation, current commercial membranes suffer from severe vulnerabilities, such as NH4+ crossover in CEMs and NO3- crossover in AEMs. Future efforts must focus on the atomic-level engineering of BPMs coupled with precision interposer layers or proton fences. These structures can selectively throttle proton delivery to suppress parasitic hydrogen evolution while perfectly balancing the dual-pH microenvironments of the anode and cathode.
(2) Scale-up of flow fields and multi-cell stacks. Scaling NO3RR to process massive volumetric flows of wastewater requires a departure from traditional serpentine flow fields, which incur economically prohibitive mechanical pumping energy losses. Future macroscopic reactor designs must optimize full-runner or parallel flow field structures integrated into highly conductive bipolar plates. This will ensure uniform reactant distribution, minimize hydrostatic pressure drops, and sustain high stack-voltage efficiency at industrial current.
(3) Integrated reactive separation and product recovery. The ultimate value of NO3RR relies on extracting pure ammonia from a complex matrix. Research must further refine in situ reactive separation technologies, utilizing robust, highly hydrophobic PTFE gas-permeable membranes integrated directly into the flow reactor body. Advancing these electrified membrane systems will enable the continuous, single-step extraction of volatile NH3 gas into acidic traps to form stable, commercial-grade liquid fertilizer.
(4) Standardization of scalable precision manufacturing. Future research should prioritize scalable precision manufacturing technologies, such as roll-to-roll slot-die coating for large-area MEAs and high-resolution 3D additive manufacturing for mathematically optimized flow-through electrodes. A critical frontier will be establishing strict, quantitative correlations between specific manufacturing parameters (e.g., laser power, ink rheology, deposition current), the resulting hierarchical microstructures, and long-term NO3RR stability, which is essential to realize the true vision of rational manufacturing.
(5) Combating matrix effects in real-world aquatic environments. The most insidious threat to long-term NO3RR operation is catastrophic mineral scaling induced by extreme localized alkalinity when processing raw, hardness-laden wastewater. To ensure long-term resilience, future reactor systems must seamlessly integrate advanced operational protocols, such as pulsed electrolysis, to electrostatically repel multivalent scale-forming cations and disrupt interfacial accumulation. Additionally, integrating robust non-electrochemical pretreatments or electrodialysis pre-modules based on strict size will be mandatory for translating NO3RR into viable, decentralized environmental remediation facilities.
Y.Ma, J. W.Bae, S. H. Kim,M. Jovičević‐Klug, K.Li, D.Vogel, D.Ponge, M.Rohwerder, B.Gault and D.Raabe, Advanced Science, 2023, 10, 2300111.
[6]
B.Lin, F. H.Nowrin, J. J.Rosenthal, A. S.Bhown and M.Malmali, ACS Sustainable Chemistry & Engineering, 2023, 11, 9880-9899.
[7]
L.Fang, Y.Wang, S.Lu, Y.Zhang, Y.Li, F.Yin and H.Liu, Applied Catalysis B: Environment and Energy, 2026, 381, 125816.
[8]
J. M.McEnaney, S. J.Blair, A. C.Nielander, J. A.Schwalbe, D. M.Koshy, M.Cargnello and T. F.Jaramillo, ACS sustainable chemistry & engineering, 2020, 8, 2672-2681.
[9]
E.Murphy, Y.Liu, I.Matanovic, M.Rüscher, Y.Huang, A.Ly, S.Guo, W.Zang, X.Yan, A.Martini, J.Timoshenko, B. R.Cuenya, I. V.Zenyuk, X.Pan, E. D.Spoerke and P.Atanassov, Nature Communications, 2023, 14, 4554.
L.Fang, S.Wang, C.Song, X.Yang, Y.Li and H.Liu, J. Hazard. Mater., 2022, 421, 126628.
[16]
Z.-Y.Wu, M.Karamad, X.Yong, Q.Huang, D. A.Cullen, P.Zhu, C.Xia, Q.Xiao, M.Shakouri and F.-Y.Chen, Nature communications, 2021, 12, 2870.
[17]
K.-H.Kim and J.Lim, Catalysts, 2025, 15, 172.
[18]
Z.Mo, W.Guo, L.Xu, Y.Chen, S.Zhou and B.Liu, Green Chem., 2025, 27, 14478-14506.
[19]
X.Wang, X.Wu, W.Ma, X.Zhou, S.Zhang, D.Huang, L. R.Winter, J.-H.Kim and M.Elimelech, Proceedings of the National Academy of Sciences, 2023, 120, e2217703120.
[20]
T.Zhao, K.Wang, S.Zhang, R.Wang, Y.Chen and S.-H.Ho, Applied Catalysis B: Environment and Energy, 2025, 361, 124693.
F.-Y.Chen, Z.-Y.Wu, S.Gupta, D. J.Rivera, S. V.Lambeets, S.Pecaut, J. Y. T.Kim, P.Zhu, Y. Z.Finfrock, D. M.Meira, G.King, G.Gao, W.Xu, D. A.Cullen, H.Zhou, Y.Han, D. E.Perea, C. L.Muhich and H.Wang, Nature Nanotechnology, 2022, 17, 759-767.
[32]
Y.Wu, K.-K.Lu and L.-H.Xu, Journal of Materials Chemistry A, 2023, 11, 17392-17417.
H.Li, H.Fu, B.Zhang, C.Sun and S.Yin, PCCP, 2026, 28, 6148-6170.
[57]
J.Gao, Q.Ma, J.Young, J. C.Crittenden and W.Zhang, Advanced Energy Materials, 2023, 13, 2203891.
[58]
S.Halarnkar, A.Ruba and N.Singh, ACS Catalysis, 2026, 16, 10842-10850.
[59]
N.Pirrone, S.Garcia-Ballesteros, S.Hernandez and F.Bella, Electrochim. Acta, 2024, 493, 144415.
[60]
J. W.Sun, H. Q.Fu, P. F.Liu, A.Chen, P.Liu, H. G.Yang and H.Zhao, EES Catalysis, 2023, 1, 934-949.
[61]
C. Mehenktaş and Ö. Arar, Current Chinese Science, 2023, 3, 42-56.
[62]
S.Li, D.Han, G.Jiang, Z.Han, H.Lu, J.Gao, X.Wang, Y.Wang, C.Geng and Z.Weng, ACS Applied Energy Materials, 2023, 6, 5067-5073.
[63]
O.Kozaderova, O.Kozaderov and S.Niftaliev, Membranes, 2022, 12, 1144.
[64]
M.Li, J.Zhao, V. ThazheVeettil, S.Fu, A. MohsenIsmail, R.Kortlever and A.Urakawa, ACS Catalysis, 2026, 16, 3430-3441.
[65]
L.Mi, Q.Huo, J.Cao, X.Chen, H.Yang, Q.Hu and C.He, Advanced Energy Materials, 2022, 12, 2202247.
[66]
L. M.Wilder, K.Wyatt, C. A.Skangos, W. E.Klein, M. R.Parimuha, J. L.Katsirubas, J. L.Young and E. M.Miller, ACS applied energy materials, 2024, 7, 536-545.
J.Guo, M. J.Liu, C.Laguna, D. M.Miller, K. S.Williams, B. D.Clark, C.Munoz, S. J.Blair, A. C.Nielander and T. F.Jaramillo, Energy Environ. Sci., 2024, 17.
J.Fan, Y.Wu, L. K.Arrazolo, K.Yin, S.Jin, Y.Gao, X.Wu, Z.Wu and J.-H.Kim, Environmental Science & Technology, 2026, 60, 4393.
RIGHTS & PERMISSIONS
Just Accepted
This article has successfully passed peer review and final editorial review, and will soon enter typesetting, proofreading and other publishing processes. The currently displayed version is the accepted final manuscript. The officially published version will be updated with format, DOI and citation information upon launch. We recommend that you pay attention to subsequent journal notifications and preferentially cite the officially published version. Thank you for your support and cooperation.
PDF
(2338KB)
